UNIVERSIDAD NACIONAL AUTÓNOMA DE MÉXICO POSGRADO EN CIENCIAS BIOLÓGICAS FACULTAD DE MEDICINA BIOLOGÍA EXPERIMENTAL CARACTERIZACIÓN ESTRUCTURAL Y FUNCIONAL DE LA MUTANTE delta-g DE LA F1FO-ATP SINTASA DE Ustilago maydis TESIS QUE PARA OPTAR POR EL GRADO DE: DOCTORA EN CIENCIAS PRESENTA: M. EN C. ESPARZA PERUSQUÍA MARÍA DE LAS MERCEDES TUTOR PRINCIPAL DE TESIS: DR. OSCAR FLORES HERRERA FACULTAD DE MEDICINA, UNAM COMITÉ TUTOR: DRA. MARIETTA TUENA SANGRI INSTITUTO DE FISIOLOGIA CELULAR, UNAM DR. SALVADOR URIBE CARVAJAL INSTITUTO DE FISIOLOGIA CELULAR, UNAM Ciudad Universitaria, CD. MX., Septiembre 2020 UNAM – Dirección General de Bibliotecas Tesis Digitales Restricciones de uso DERECHOS RESERVADOS © PROHIBIDA SU REPRODUCCIÓN TOTAL O PARCIAL Todo el material contenido en esta tesis esta protegido por la Ley Federal del Derecho de Autor (LFDA) de los Estados Unidos Mexicanos (México). El uso de imágenes, fragmentos de videos, y demás material que sea objeto de protección de los derechos de autor, será exclusivamente para fines educativos e informativos y deberá citar la fuente donde la obtuvo mencionando el autor o autores. Cualquier uso distinto como el lucro, reproducción, edición o modificación, será perseguido y sancionado por el respectivo titular de los Derechos de Autor. COORDINACIÓN DEL POSGRADO EN CIENCIAS BIOLÓGICAS FACULTAD DE MEDICINA OFICIO CPCB/517/2020 ASUNTO: Oficio de Jurado M. en C. Ivonne Ramírez Wence Directora General de Administración Escolar, UNAM P r e s e n t e Me permito informar a usted que en la reunión virtual del Subcomité de Biología Experimental y Biomedicina del Posgrado en Ciencias Biológicas, celebrada el día 30 de marzo de 20 se aprobó el siguiente jurado para el examen de grado de DOCTORA EN CIENCIAS de la estudiante ESPARZA PERUSQUÍA MARÍA DE LAS MERCEDES con número de cuenta 511021118 con la tesis titulada “CARACTERIZACIÓN ESTRUCTURAL Y FUNCIONAL DE LA MUTANTE delta-g DE LA F1F0-ATP SINTASA DE Ustilago maydis”, realizada bajo la dirección del DR. OSCAR HERRERA FLORES, quedando integrado de la siguiente manera: Presidente: DRA. MINA KONIGSBERG FAINSTEIN Vocal: DR. RICARDO JASSO CHÁVEZ Secretario: DR. SALVADOR URIBE CARVAJAL Suplente: DR. DIEGO GONZÁLEZ HALPHEN Suplente: DR. MANUEL GUTIÉRREZ AGUILAR Sin otro particular, me es grato enviarle un cordial saludo. A T E N T A M E N T E “POR MI RAZA HABLARÁ EL ESPÍRITU” Cd. Universitaria, Cd. Mx., a 28 de agosto de 2020 COORDINADOR DEL PROGRAMA DR. ADOLFO GERARDO NAVARRO SIGÜENZA AGRADECIMIENTOS A la Universidad Nacional Autónoma de México y al Posgrado en Ciencias Biológicas, UNAM por todo el apoyo brindado durante la realización del presente trabajo. Durante el desarrollo de esta tesis recibí beca de para estudios de Doctorado otorgada por el Consejo Nacional de Ciencia y Tecnología (CONACyT) con número de becario 254400. Este proyecto fue financiado por el Programa de Apoyo a Proyectos de Investigación e Innovación Tecnológica (PAPIIT IN214914-OFH, IN222617-OFH, IN215518-FMM, IN222117-JPP), así como por el CONACyT con el proyecto 168025. Esta tesis se realizó bajo la dirección del Dr. Oscar Flores Herrera en el laboratorio 4Bis del Departamento de Bioquímica de la Facultad de Medicina de la Universidad Nacional Autónoma de México. Especialmente deseo agradecer a los miembros del Comité Tutoral: Dra. Marietta Tuena de Gómez-Puyou y Dr. Salvador Uribe Carvajal, por las valiosas aportaciones a este trabajo. La sección de Biología Molecular se realizó en el laboratorio del Instituto de Microbiología de la Universidad Heinrich-Heine en Düsseldorf, Alemania, bajo la asesoría de los doctores Michael Feldbrügge y Thorsten Langner. Agradezco también a la M. en C. Sara Teresa Méndez del Instituto Nacional de Pediatría, torre de investigación “Dr. Joaquín Cravioto” por la ayuda en las técnicas de Biología Molecular. Un especial agradecimiento a la Dra. Guadalupe Zavala del Instituto de Biotecnología, UNAM por su apoyo en microscopía de transmisión electrónica. Agradezco por sus valiosas aportaciones a este trabajo al jurado del Examen Doctoral que estuvo conformado por: Presidente Dra. Mina Konigsberg Fainstein Universidad Autónoma Metropolitana Vocal Dr. Ricardo Jasso Chávez Instituto Nacional de Cardiología Secretario Dr. Salvador Uribe Carvajal Instituto de Fisiología Celular, UNAM Suplente Dr. Diego González Halphen Instituto de Fisiología Celular, UNAM Suplente Dr. Manuel Gutiérrez Aguilar Facultad de Química, UNAM A Oscar, por tu compromiso y dedicación, por tu apoyo incondicional, por tu confianza, por nunca perder la fe en mí, por permitirme ser parte de este gran grupo de trabajo que comenzó con nosotros dos y que fue creciendo con el tiempo, pero sobre todo por tu gran paciencia. A lo largo de este camino me enseñaste que los datos se dan con emoción, que cuando te aprendes todas las respuestas te cambian todas las preguntas, que ser investigador era el mejor trabajo del mundo porque te pagaban por aprender y que sin importar lo que hiciera en la vida tenía que apasionarme. Y aquí estoy esperando estar siempre a tu altura, agradeciendo a la vida haberte puesto en mi camino, por tomar mi mano cuando más lo necesitaba, por enseñarme a usar la densidad y por levantarme los ánimos con cafecito. Durante estos 11 años no solo tuve al mejor tutor, sino también al mejor amigo. “Gracias Tlanepantla”. A mis compañeros del laboratorio Giovanni, Jaime, Ixchel, Paola, Luis, Toño, y todos aquellos con quién he compartido, ideas, experimentos, enseñanzas y experiencias de vida. Mil gracias a todos porque sin ustedes este trayecto no hubiera sido tan divertido. A ti mamá por hacerme una mujer de valores y que sabe luchar para alcanzar sus sueños. Por darme siempre ese ejemplo de lucha, de tenacidad, de amor y comprensión. Por escucharme cada que contaba de mis experimentos, ahora se que lo hice bien porque ya me hablas de los “catiónicos”. Porque a pesar de lo largo de los años y lo pesado nunca has soltado mi mano, tus brazos siempre serán mi lugar favorito Todo lo que soy te lo debo a ti. Te amo mami. A mi papá por todo su amor. ¡Gracias pá! A la memoria de Manuel Perusquía, mi abuelo que me enseñó a cuestionarme y a pensar más allá de lo lógico, por fomentar mi inquietud y curiosidad y transformarla en ese pensamiento crítico y científico que desconocía que tenía. Gracias por creer en mí Pipito, ojalá estuvieras aquí. Besos hasta el cielo. A la familia Perusquía González por siempre ser la mejor porra, porque siempre dan los mejores abrazos y por tener ese lugar de amor que hacen mi segunda casa, donde siempre me siento tan feliz. Los amo! Marlén y Aránzazu a quienes les pertenece la mitad del miocardio, porque sin importar lo difícil de los años, día con día mi alma se siente tranquila de saberse segura entre sus manos. ¡Male y Gogoro gracias por este par! Abuela de Batman gracias por cuidarme tanto, por ayudarme con las tareas, por estudiar conmigo y por ser la mejor abuela del mundo, se que me miras desde donde estés. Te llevo como una chispita en el corazón. A mi hermosa familia Perusquía, que me vio crecer, física, emocional y académicamente, que nunca nos abandono aún en los momentos mas difíciles, a todos ustedes muchas gracias. A mis segundas mamás Patita y Conchitas no tengo palabras para describir cuan emocionada estoy de llegar hasta aquí, pero no hubiera sido posible de no ser por ustedes, que nos arroparon, que nos mimaron, que nos cuidaron y muchas veces nos llenaron la pancita. Gracias por confiar en mi e impulsarme siempre a seguir mis sueños. ¡Las amo! Mensa eres mi pilar y mi fortaleza, gracias por estar ahí siempre, por compartir conmigo no sólo mis momentos felices, sino también los más desastrosos, vergonzosos y tristes. Quiero dar gracias por todas las veces que nos hemos reído, por las que nos hemos peleado también y porque siempre sacas lo mejor de mí. Por hacer que estos 17 años en los que hemos sido amigas parezcan mucho más de los que son. Porque durante estos años hemos creado recuerdos para toda una vida. Gracias por recordarme que no tengo que preocuparme por los demás, sino que sólo por aquellos que me quieren y por ser mi fiel escudero. Pero sobre todo, gracias por darme todas estas razones y por ser mi persona. Te adoro hermana! A Alejandra Sánchez por ser mi hermana de corazón. Porque eres parte fundamental de mi ser, por tomar mi mano y consolarme, por enseñarme a luchar, eres la persona más amorosa y valiente que conozco. Eres ese pequeño rayo de luz cálida que ilumina mi corazón, eres mi mayor ejemplo de amor. Gracias por adoptarme como hermanita y por enseñarme que no hace falta ser la misma sangre para ser familia. A Blanca, por compartirme a su hermosa y valiosa hermana, besos hasta el cielo. ¡Arriba la laguna!... literal. A Fernando Garrido “el hombre de la casa” por cuidar no solo de mi hermana si no también de nosotras. Gracias por compartirnos tu maravillosa familia, te amamos y lo hacemos extensivo a Mariel y mamá Flor. A Sofía por tu amistad, confianza y por estar cerca de mí siempre. Tu ayuda ha sido fundamental, has estado conmigo hasta en los momentos mas turbulentos. Este proyecto no fue nada fácil y estuviste apoyándome, motivándome y soportándome, creo que lo ultimo fue lo más difícil. Aunque la vida nos ha hecho diferentes en actitud y comportamiento, el cariño siempre va a ser más fuerte. ¡Te quiero 4ever! A Montse y sus alocados chinos, eres de las mejores personas que conozco, amo tu locura, tu pasión por la vida, tu entusiasmo, tu gran sentido del humor y sobre todo tu fuerza de voluntad para no darte por vencida, ese es mi mejor ejemplo y motor. Gracias por siempre tomar mi mano, por ser ese hombro que seca mis lagrimas y por siempre siempre devolverme la sonrisa. Te mega adoro amiga. Amore mío, compañero de aventuras, este trabajo también te lo dedico a ti. Eres esa parte de amor, locura y sensatez que me hacen el día a día. A tu lado siento que nada me falta y que todo se torna más ligero. Apúrate que las vegas no esperan. Te amo nene! A Thorsten Langner for German Classes and Molecular Biology coaching. Thank you Guapo. A Roselia porque contigo aprendí la responsabilidad de dirigir alumnos, por ahí dicen que “echando a perder se aprende”, pero sobre todo a tener paciencia a las necedades, muuucha paciencia (Oscar: aquí donde pague todas las que te hice pasar jajajaja). Nunca te detengas, que eso me encanta de ti, eres una persona muy lista, loca, siempre valiente, entusiasta, decidida y una gran amiga. Te quiero mucho güera! A Héctor Miranda por aguantar mi estrés, siempre darle una explicación lógica a mi alocado pensamiento, por las revisiones críticas y severas, por las ideas de madrugada, por consolarme cuando se me caía el mundo, por los abrazos que me reiniciaban el alma y sobre todo por sacarme tantas sonrisas. Nos debemos esas duvel. A Norma Silvia, por ese consejo siempre tan atinado acompañado de un buen vino. Eres mi ejemplo a seguir, eres una gran maestra y una excelente amiga, no miento cuando te digo que cuando sea grande quiero ser como tú, o cuando menos la mitad de feliz, de inteligente, de tenaz, de comprensiva y cariñosa. Te admiro mucho y te quiero más. A Sara Teresa Méndez por enseñarme la magia de la biología molecular y por trabajar hombro a hombro conmigo, aunque nos dieran las 11pm, pero sobre todo por ser esa mamá cariñosa y atenta que me cuidaba no solo los experimentos. Mil gracias por todo, te adoro. A Juan Pablo Pardo por su aportación a este trabajo, por la revisión del artículo y por el p(ashe) osho osho. Te quiere la pinshe Meshe. A Federico Martínez por permitirnos vivir a su lado tantos años formando el fabuloso 5Bis, gracias por aceptarme como su alumna ese octubre de 2006, por confiar en mi pequeño talento de investigador que fue creciendo a lo largo de los años y ahora todo se concentra en esta tesis. A TODOS USTEDES MIL GRACIAS. DEDICATORIAS A Viridiana, porque gracias a ti comencé a subir la escalera… ¡y ya llegamos hasta arriba! En ti tengo ese espejo de fortaleza en donde algún día me quiero reflejar, pues tus virtudes infinitas y tu gran corazón me llevan a admirarte cada día más. Gracias por no dejarme perder la fe en la humanidad, por curarme el estrés con cacahuates enchilados y coca-cola, por despertarme con un beso a media noche, por ser la mejor amiga y sobetodo por ser la persona más buena que conozco, haces del mundo un lugar mejor o por lo menos mi mundo. Hay días que cuando me duele mirar hacia atrás y me da miedo mirar hacia enfrente tengo la seguridad de que al voltear a la derecha o izquierda tu siempre estas a mi lado. Gracias Dios por concederme a la mejor de las hermanas, Eres mi luz al final del túnel. Te amo chiquitina. “La ATP sintasa” Dr. Oscar Flores-Herrera, 2018. Índice ÍNDICE página LISTA DE FIGURAS LISTA DE TABLAS ABREVIATURAS RESUMEN 1 ABSTRACT 2 1. INTRODUCCIÓN 4 1.1. Generalidades. 4 1.2. Arquitectura y función de la F1FO-ATP sintasa. 10 1.3. Las crestas mitocondriales. 17 1.4. La F1FO-ATP sintasa como dímero y su papel en el 20 plegamiento de las crestas mitocondriales, 1.5. Otros factores que participan en la formación de las 33 crestas mitocondriales. 1.6. Efecto de la eliminación de las subunidades dimerizantes 35 y su repercusión en la arquitectura mitocondrial 1.7. Ustilago maydis como modelo experimental. 37 2. JUSTIFICACIÓN 41 3. OBJETIVOS 43 3.1. General 3.2. Particulares 4. MATERIALES Y MÉTODOS 45 5. RESULTADOS 64 6. DISCUSIÓN 100 7. CONCLUSIONES 107 8. PERSPECTIVAS 109 9. REFERECIAS BIBLIOGRÁFICAS 111 10. ANEXOS 129 10.1. Oligonucleótidos usados para la formación del 129 knock-out para la transformación de U. maydis. 10.2. Transformación por biobalística. 129 10.3. Solución de sales. 130 10.4. Elementos traza. 130 10.5. Cuantificación de la cantidad de F1FO-ATP sintasa 130 en las muestras V1WT, V2WT, VDg. 10.6. Cuantificación de la cantidad de F1FO-ATP sintasa 139 en las mitocondrias de las cepas WT y Dg. 11. ARTÍCULOS 145 11.1. Artículo de requisito. • STRUCTURAL AND KINETICS CHARACTERIZATION OF THE F1F0-ATP SYNTHASE DIMER. NEW REPERCUSSION OF MONOMER-MONOMER CONTACT (2017) Mercedes Esparza- Perusquía, Sofía Olvera-Sánchez, Juan Pablo Pardo, Guillermo Mendoza-Hernández, Federico Martínez, and Oscar Flores- Herrera. BBA-Bioenergetics. 1858(12):975-981. doi: 10.1016/j.bbabio.2017.09.002. 148 11.2. Publicaciones directas. • DELETION OF SUBUNIT G FROM F1F0-ATP SYNTHASE AFFECTS THE STABILITY OF ITS DIMERIC STATE AND THE MITOCHONDRIAL ATP SYNTHESIS. Esparza-Perusquía M, Langner T, Feldbrügge M, Pardo JP, Martínez F, and Flores- Herrera O. En preparación. • DELETION OF THE NATURAL INHIBITORY PROTEIN INH1 FROM USTILAGO MAYDIS ATP SYNTHASE DOES NOT INCREASE THE ACTIVITY OF THE DIMERIC STATE OF F1FO- ATP SYNTHASE. Romero-Aguilar Lucero*, Esparza-Perusquía Mercedes*, Langner Thorsten, García Giovanni, Feldbrügge Michael, Pardo Juan Pablo, Martínez Federico, and Flores-Herrera Oscar. En preparación. 155 178 11.3. Otras publicaciones. • MITOCHONDRIAL PROTEASES ACT ON STARD3 TO ACTIVATE PROGESTERONE SYNTHESIS IN HUMAN SYNCYTIOTROPHOBLAST (2015) Mercedes Esparza- Perusquía, Sofía Olvera-Sánchez, Oscar Flores-Herrera, Héctor Flores-Herrera, Alberto Guevara-Flores, Juan P Pardo, María T Espinosa-García y Federico Martínez. BBA-General Subjects. 1850(1):107-17. doi: 10.1016/j.bbagen.2014.10.009. • MEMBRANE POTENTIAL REGULATES THE MITOCHONDRIAL ATP-DIPHOSPHOHYDROLASE ACTIVITY BUT IT IS NOT INVOLVED IN THE PROGESTERONE BIOSYNTHESIS IN HUMAN SYNCYTIOTROPHOBLAST CELL (2015) Oscar Flores- Herrera, Sofía Olvera-Sánchez, Mercedes Esparza-Perusquía, Juan Pablo Pardo, Juan Luis Rendón, Guillermo Mendoza- 204 215 Hernández, and Federico Martínez. BBA-Bioenergetics. 1847(2):143-52. doi: 10.1016/j.bbabio.2014.10.002. • MULTIPLE FUNCTIONS OF SYNCYTIOTROPHOBLAST MITOCHONDRIA IN PREGNANCY (2015) Federico Martínez, Sofía Olvera-Sánchez, Mercedes Esparza-Perusquía, Erika Gómez-Chang, Oscar Flores-Herrera. Steroids, 103:11-22. doi: 10.1016/j.steroids.2015.09.006. • STREPTOZOTOCIN INDUCED ADAPTIVE MODIFICATION OF THE MITOCHONDRIAL SUPERCOMPLEXES IN LIVER OF WISTAR RATS AND THE PROTECTIVE EFFECT OF MORINGA OLEIFERA LAM (2018) Alejandra Sánchez-Muñoz, Mónica A Valdez-Solana, Mara I Campos-Almazán, Oscar Flores-Herrera, Mercedes Esparza-Perusquía, Sofía Olvera-Sánchez, Guadalupe García-Arenas, Claudia Avitia-Domínguez, Alfredo Téllez-Valencia and Erick Sierra-Campos. Biochemistry Research International. 1-15. ID 5681081. doi:10.1155/2018/5681081. • CARDIOPROTECTIVE STRATEGIES PRESERVE THE STABILITY OF RESPIRATORY CHAIN SUPERCOMPLEXES AND REDUCE OXIDATIVE STRESS IN REPERFUSED ISCHEMIC HEARTS (2018) Ixchel Ramírez-Camacho, Francisco Correa, Mohammed El-Hafidi, Alejandro Silva-Palacios, Marcos Ostolga-Chavarría, Mercedes Esparza-Perusquía, Sofía Olvera- Sánchez, Oscar Flores-Herrera and Cecilia Zazueta. Free Radical Biology and Medicine. 129:407-417. doi: 10.1016/j.freeradbiomed.2018.09.047. • MITOCHONDRIAL RESPIRASOME WORKS AS A SINGLE UNIT AND THE CROSSTALK BETWEEN COMPLEXES I, III2 AND IV STIMULATES NADH DEHYDROGENASE ACTIVITY (2019) Meztli Reyes-Galindo, Roselia Suarez, Mercedes Esparza- 225 237 254 265 Perusquía, Jaime de Lira-Sánchez, Juan Pablo Pardo, Federico Martínez, and Oscar Flores-Herrera. BBA-Bioenergetics. 1860(8):618-627. doi: 10.1016/j.bbabio.2019.06.017. • ASPECTOS GENERALES DEL TRANSPORTE DE COLESTEROL EN LA ESTEROIDOGÉNESIS DE LA PLACENTA HUMANA (2019) Sofía Olvera-Sánchez, Mercedes Esparza- Perusquía, Oscar Flores-Herrera, Viviana A. Urban-Sosa y Federico Martínez. TIP Revista Especializada en Ciencias Químico-Biológicas, 22:1-9. DOI: 10.22201/fesz.23958723e. • STEADY-STATE PERSISTENCE OF RESPIRATORY SYNCYTIAL VIRUS IN A MACROPHAGE-LIKE CELL LINE AND IDENTIFICATION OF NON-SYNONYMOUS MUTATIONS THROUGH SEQUENCING OF THE PERSISTENT VIRAL GENOME. Ximena Ruiz-Gómez, Joel Armando Vázquez-Pérez, Oscar Flores-Herrera, Mercedes Esparza-Perusquía, Carlos Santiago-Olivares, Jorge Gaona, Beatriz Gómez, Carmen Méndez, Evelyn Rivera-Toledo. MDPI-viruses, Manuscript ID: 842162. 275 284 12. CAPITULOS DE LIBRO 306 • “Proteomics of lignocellulosic substrates bioconversion in anaerobic digesters to increase the carbon recovery as methane” Alicia Guadalupe Talavera-Caro, María Alejandra Sánchez-Muñoz, Inty Omar Hernández-De Lira, Lilia Ernestina Montañez- Hernández, Jesús Antonio Morlett-Chávez, María de las Mercedes Esparza-Perusquía, Nagamani Balagurusamy. Biological Approaches for the Management of Agro-industrial Residues Applied Environmental Science and Engineering for a Sustainable Future. (2020) Editor: Zainul Akmar Zakaria, Ramaraj 307 Boopathy, Julian Rafael Dib, Reeta Rani Singhania. Springer, Cham. ISBN 978-3-030-39137-9. LISTA DE FIGURAS página Figura 1. Morfología mitocondrial. 4 Figura 2. Cadena de trasporte de electrones. 8 Figura 3. La F1FO-ATP sintasa. 11 Figura 4. Mecanismo para la síntesis de ATP. 13 Figura 5. Mecanismo de los sitios catalíticos de la porción F1 de 15 la ATP sintasa durante la síntesis de ATP. Figura 6. Estructura de la subunidad a y la translocación de protones. 16 Figura 7. Representación esquemática de la fosforilación oxidativa 18 en una cresta plegada. Figura 8. Mecanismos propuestos para la formación de las crestas. 19 Figura 9. Actividad de ATPasa en gel de los oligómeros de la 20 F1FO-ATP sintasa. Figura 10. Densidad de protones al interior de la cresta. 21 Figura 11. Estructura del dímero de la F1FO-ATP sintasa de 24 Polytomella sp. Figura 12. Dímero de la F1FO-ATP sintasa porcina y sus sitios 27 de unión. Figura 13. Ultraestructura mitocondrial de las células de placenta 28 Humana. Figura 14. Estructura de la F1FO-ATP sintasa mitocondrial de 31 S. cerevisiae a 25 Å de resolución. Figura 15. Reconstrucción tomográfica de una cresta tubular 32 mitocondrial de hígado de rata. Figura 16. Dímeros aislados observados por microscopía 32 de trasmisión electrónica. Figura 17. Elementos proteicos involucrados en la formación 34 de las crestas mitocondriales. Figura 18. Diferencias estructurales en las mitocondrias de 35 Saccharomyces cerevisiae. Figura 19. Ultraestructura mitocondrial de las células Hela. 36 Figura 20. Ciclo de Vida de U. maydis. 38 Figura 21. Cadena de transporte en electrones de U. maydis. 39 Figura 22. Amplificación de los fragmentos UF y DF para la 65 construcción del plásmido KO. Figura 23. Transformación de U. maydis con el DNA exógeno. 66 Figura 24. Comprobación de la cepa mutante Dg. 67 Figura 25. Southern Blot. 68 Figura 26. Curva de crecimiento en YPD. 69 Figura 27. Curva de crecimiento en MM-EtOH. 70 Figura 28. Curva de crecimiento en MM-Glucosa. 71 Figura 29. Microscopia de las células WT y Dg (fase logarítmica). 72 Figura 30. Microscopia de las células WT y Dg (fase estacionaria). 73 Figura 31. Diluciones seriadas (1:10) de las cepas WT y Dg 74 de U. maydis. Figura 32. Ultraestructura de las células de U. maydis. 75 Figura 33. Consumo de oxígeno de las células cultivadas en 76 medio rico YPD. Figura 34. Consumo de oxígeno de las células cultivadas en 77 MM-EtOH. Figura 35. Consumo de oxígeno de las células cultivadas en 78 MM-Glucosa. Figura 36. Consumo de glucosa. 81 Figura 37. Células permeabilizadas de U. maydis. 82 Figura 38. Potencial de membrana (Dym) generado por las células 82 de U. maydis. Figura 39. Síntesis de ATP mitocondrial en las células WT y Dg. 83 Figura 40. Producción de radicales libres en las mitocondrias de 84 U. maydis. Figura 41. Actividad en gel de los supercomplejos y complejos 86 respiratorios. Figura 42. 2D-Tricina-SDS-PAGE. 87 Figura 43. Purificación de los oligómeros de la F1FO-ATP sintasa 89 mitocondrial de la cepa WT. Figura 44. Purificación del monómero de la F1FO-ATP sintasa 90 mitocondrial de la cepa Dg. Figura 45. Caracterización cinética o de la F1FO-ATP sintasa 92 Mitocondrial aislada. Figura 46. Titulación de la actividad de ATPasa del V2 con DDM. 93 Figura 47. Actividad de la F1FO-ATP sintasa mitocondrial en 94 presencia de DDM. Figura 48. Actividad de ATPasa de los oligómeros de la F1FO-ATP 95 sintasa mitocondrial aislados de las cepas WT y Dg. Figura 49. Efecto de la oligomicina en la actividad de ATPasa. 97 Figura 50. Efecto de la temperatura en la actividad de ATPasa. 98 Figura S1. SDS-Tricina-PAGE de V1 y V2. 132 Figura S2. SDS-Tricina-PAGE de las mitocondrias WT y Dg. 139 LISTA DE TABLAS página Tabla 1. Composición de la F1FO-ATP sintasa mitocondrial de S. 12 cerevisiae, B. taurus y Ustilago maydis. Tabla 2. Diferencias metabólicas de las cepas WT y Dg. 79 Tabla 3. Parámetros cinéticos para la actividad de hidrólisis de ATP 105 para el dímero y el monómero de la F1FO-ATP sintasa de Ustilago maydis. Tabla 4. Oligonucleótidos utilizados para generar el plásmido knock-out 129 y para verificar la cepa mutante Dg. Tabla 5. Cuantificación del contenido de las subunidades a y b 133 del V1WT de la F1FO-ATP sintasa (4 diferentes muestras). Tabla 6. Cálculo del contenido de F1FO-ATP sintasa en diferentes 134 muestras de V1WT aislados. Tabla 7. Cuantificación del contenido de las subunidades a y b del 135 V2WT de la F1FO-ATP sintasa (4 diferentes muestras). Tabla 8. Cálculo del contenido de F1FO-ATP sintasa en diferentes 136 muestras de V2WT aislados. Tabla 9. Cuantificación del contenido de las subunidades a y b del 137 V1Dg de la F1FO-ATP sintasa (4 diferentes muestras). Tabla 10. Cálculo del contenido de F1FO-ATP sintasa en diferentes 138 muestras de V1Dg aislados. Tabla 11. Cuantificación del contenido de las subunidades a y b 140 de la F1FO-ATP sintasa total en las mitocondrias de la cepa WT. Tabla 12. Cálculo del contenido de F1FO-ATP sintasa total en las 141 mitocondrias de la cepa WT. Tabla 13. Cuantificación del contenido de las subunidades a y b del 142 Dg de la F1FO-ATP sintasa total en las mitocondrias de la cepa Dg. Tabla 14. Cálculo del contenido de F1FO-ATP sintasa total en las 143 mitocondrias de la cepa Dg. ABREVIATURAS ADP Adenosín 5′-difosfato. AMPc Adenosín 3´,5´- monofosfato cíclico. anti-DIG-AB Anticuerpo contra digonexina marcada. AOX Oxidasa alterna mitocondrial. ASA del inglés “ATP Synthase-Associated protein” ATP Adenosín 5′-trifosfato. Bis-Tris 2,2-Bis(hydroxymethyl)-2,2′,2″-nitrilotriethanol, 2-Bis(2- hydroxyethyl)amino-2-(hydroxymethyl)-1,3-propanediol, Bis(2- hydroxyethyl)amino-tris(hydroxymethyl)methane. BN-PAGE Electroforesis en condiciones nativas en geles azules. BrEt Bromuro de 3,8-diamino-5-etil-6-fenilfenantridinio. C. reinhartii Chlamydomonas reinhartii CCCP Carbonyl cyanide 3-chlorophenylhydrazone. CJ Uniones de Cresta. DAB 3, 3’-Diaminobenzidine. DDM n-Dodecil β-D-maltósido. DF Downstream flank (secuencia río abajo del gen). DMSO Dimetilsulfóxido. dNTP´s Desoxinucleótidos trifosfato. DNA Ácido desoxirribonucleico . DOC Ácido 3a, 12a-Dihidroxi-5b-colan-24-oico. EDTA Ácido 2-({2[bis(carboximetil)amino]etil}(carboximetil) amino)acético. FADH2 Flavín adenín dinucleótido reducido. Fcj1 Proteína formadora de la unión de cresta. G-6P-DH Glucosa-6-fosfato deshidrogenasa. gDNA DNA genómico. GOD Glucosa oxidasa. EGTA Ácido etilenglicol-bis(2-aminoetiléter)-N,N,N'N'-tetraacético H3BO3 Ácido trioxobórico (III). HEPES Ácido 4-(2-hidroxietil)-1-piperazina-etanosulfónico. HK Hexocinasa. hrCN-PAGE Electroforesis en condiciones nativas en geles claros en presencia de DDM y DOC. HygR Cassette de resistencia a la higromicina. Km Constante de Michaelis-Menten. KO Knock-out. Kpb Kilo pares de bases. LB Luria Bertani medio. LBAmp Luria Bertani medio suplementado con ampicilina. LDH Lactato deshidrogenasa. Mapp Masa molecular aparente. MEM Membrana Externa Mitocondrial. MIM Membrana Interna Mitocondrial. MM-EtOH Medio Mínimo-Etanol. MM-Glucosa Medio Mínimo-Glucosa. MOPS 3-(N-morpholino)propanesulfonic acid. MTT 1.2 mM de bromuro de 3-[4,5-dimetiltiazol-2-yi]-2,5- difeniltetrazolio. NADH Nicotinamida adenina dinucleótido reducido. NADP Nicotinamida adenina dinucleótido fosfato. NADPH Nicotinamida adenina dinucleótido fosfato reducido. natgO2 Nanoatomo gramo de Oxígeno. nOg n-octil-galato (inhibidor de la oxidasa alterna). OPA1 del inglés “Optic atrophy type 1”. OSCP del inglés “Oligomycin - Sensitivity Conferring Protein”. pb Pares de bases. PBS Amortiguador de fosfato salino. PCR Reacción en cadena de la polimerasa. PEP Fosfoenolpiruvato. PHU Pushion polimerasa. PK Piruvato cinasa. PMS 5-Methylphenazinium methyl sulfate, N-Methylphenazonium methyl sulfate. PMSF Fluoruro de fenilmetilsulfonilo. POD Peroxidasa. pUMa Plásmido de Ustilago maydis. RNA Ácido ribonucleico. ROS del inglés “Radical Oxidative Species”. S. cerevisiae Saccharomyces cerevisiae SC Supercomplejos. SCS Amortiguador para resuspender células de U. maydis. STC Amortiguador para resuspender protoplastos de U. maydis. STC/PEG Amortiguador para resuspender protoplastos de U. maydis con polietilenglicol. STET Medio para resuspender células (Obtención del DNA plasmídico). TAE Amortiguador para electroforesis en geles de agarosa, formada por Tris, acetato y EDTA. TBE Amortiguador para electroforesis en geles de agarosa, formada por Tris, borato y EDTA. TOPO pCR2.1-TOPO vector. U. maydis Ustilago maydis UF Upstream flank (secuencia río arriba del gen). V1 Monómero de la F1F0-ATP sintasa. V1Dg Monómero de la F1F0-ATP sintasa de la mutante Dg. V2 Dímero de la F1F0-ATP sintasa. V2Dg Dímero de la F1F0-ATP sintasa de la mutante Dg. Vmax Velocidad enzimática máxima. Y. lipolitica Yarrowia lipolitica (u) Velocidad de la reacción. DYm Potencial de membrana mitocondrial. Dym Potencial eléctrico de membrana. DµH+ Gradiente Electroquímico de Protones. Dg Cepa mutante de U. maydis (eliminación del gen ATP20). DpH Diferencial de pH . 4-AF 4–aminofenazona. “La cadena respiratoria mitocondrial y la ATP sintasa: los años 1970 y 1980 debaten si los protones viajan en el circuito quimiosmótico deslocalizado o por vías más localizadas” Bioenergetics3, Nicholls, D. G. et al., 2002 1 RESUMEN La espectrometría de masas demuestra que el dímero y el monómero de la F1FO-ATP sintasa de Ustilago maydis contienen las 14 subunidades canónicas, pero además, el dímero contiene en el sector F0-estator las subunidades e y g, las cuales se sugiere son proteínas dimerizantes. El dímero del complejo V tiene un papel importante en la formación de las crestas de la membrana interna mitocondrial y al eliminar la subunidad g la proporción de dímero disminuye y se modifica la arquitectura mitocondrial. En U. maydis la eliminación genética de la subunidad g no impide la dimerización del complejo V manteniéndose la misma proporción entre dímero y monómero. Por otro lado, observamos que la actividad de ATPasa del dímero es lenta, demostrando que la subunidad g no es esencial para la dimerización pero que podría tener un papel importante para de la actividad de la enzima. Aunado a ésto, en la cepa Dg la AOX aparece a partir de la primera fase de crecimiento, lo que sugiere que la célula podría entrar en un estrés oxidativo. El consumo de oxígeno durante la fase logarítmica de crecimiento fue alrededor de 108 natgO/mg/min para la cepa silvestre y 120 natgO/mg/min para la cepa Dg. Por otro lado, en las células recuperadas de la fase estacionaria el consumo de oxígeno para la cepa WT fue de 348 natgO/mg/min y de 168 natgO/mg/min para la ∆g. La síntesis de ATP en ambas cepas es similar, para la cepa WT es de 0.34 µmolas de ATP/g peso húmedo/min y para la Dg es de 0.31 µmolas de ATP/g peso húmedo/min. Aunado a esto, el análisis de la arquitectura mitocondrial de la cepa Dg de U. maydis mostró crestas tubulares y lamelares, y no los anillos de cebolla descritos para S. cerevisiae. Asimismo, la síntesis de ATP y DYm, en la cepa Dg y WT fueron similares y lo que sugiere que su estado bioenergético no fue afectado. 2 ABSTRACT Mass spectrometry reveals that the dimer and monomer of F1FO-ATP synthase from Ustilago maydis contain the 14 canonic subunits, but in addition the dimer contains the subunits e and g at FO-membrane-stator which are suggested to play a role in dimerization. F1FO-ATP synthase plays a central role in the creation of internal membrane cristae, and therefore in mitochondrial architecture. In U. maydis genetical removal of the g subunit does not prevent dimerization of the V complex maintaining the same ratio of dimer to monomer. On the other hand, we observed that the ATPase activity of the dimer is slow, demonstrating that the subunit is not essential for dimerization but might play an important role in the enzymatic activity. In addition, in the Dg strain the AOX is expressed in the early stages of growth, which suggest that the cell oxidative stress is present since premature phases in contrast to the WT strain. Oxygen consumption in the logarithmic phase was about 108 natgO/mg/min for the wild-type strain and 120 natgO/mg/min for the Dg strain. On the other hand, in cells recovered from the steady state oxygen consumption for the WT strain was 348 natgO/mg/min and 168 natgO/mg/min for the Dg. The synthesis of ATP in both strains was similar; for the WT strain 0.34 µmoles ATP/g wet weight/min and for Dg is 0.3134 µmoles ATP/g wet weight/min. Mitochondrial architecture from the of U. maydis Dg-strain showed a tubular and lamellar crista, and not the onion ring morphology previously described in S. cerevisiae. Also, ATP synthesis, and DYm, in the Dg and WT-strain were similar; suggesting that, although the ATPase activity of dimer was inhibited, its role in cristae structure and bioenergetics was not affected. 3 “Mitochondrial Membranes II” Odra Noel Introducción 4 1. INTRODUCCIÓN 1.1. Generalidades Las mitocondrias son organelos que adaptan su arquitectura y metabolismo a los requerimientos celulares, consiguiendo así producir más del 90% del ATP necesario para alejar a la célula del equilibrio termodinámico. Las mitocondrias están constituidas por dos membranas cuya composición proteica y lipídica varía enormemente: la membrana externa mitocondrial (MEM) es relativamente permeable para la mayoría de las moléculas, mientras que la membrana interna mitocondrial (MIM), es impermeable a los iones y a muchas moléculas orgánicas. La MIM incrementa su área de superficie al proyectarse hacia el interior de la matriz mitocondrial en forma de invaginaciones que se denominan crestas mitocondriales. La biogénesis y el mantenimiento de las crestas dependen de los procesos de fisión y fusión, así como de los cambios en la composición proteica y lipídica (Figura 1) (Hackenbrock et al., 1986; Frey et al., 2002). Figura 1. Morfología mitocondrial. Mitocondrias aisladas de Saccharomyces cerevisiae. (A) Se muestra la reconstrucción tridimensional. (B) Se muestra un corte histológico por criotomografía, en el cual se muestran las membranas mitocondriales, la matriz, las crestas, las uniones de cresta y el espacio intermembranal (Modificado de Kühlbrandt, 2015). 5 Los cambios morfológicos de las crestas son reflejo del nivel energético de la mitocondria, los cuales se clasifican en dos estados: el ortodoxo (alta concentración de ATP y baja de ADP), donde el número de crestas es bajo y las uniones de cresta son mínimas; y el condensado (baja concentración de ATP y alta de ADP), en el que hay un incremento en el número de las crestas al igual que de las uniones de cresta, provocando la contracción de la matriz mitocondrial (Mannella, 2006). En el estado condensado donde el consumo de oxígeno aumenta, se acopla con el bombeo de protones y la síntesis de ATP. La respiración celular aeróbica, es el proceso mediante el cual las células transfieren poder reductor generado a partir de la degradación de azúcares, grasas y proteínas hasta el oxígeno molecular (O2), es fundamental para el metabolismo energético de todas las células eucariontes y algunas procariontes. En las eucariontes las etapas finales de la respiración aeróbica se llevan a cabo en las mitocondrias, donde ocurren diversas vías metabólicas relacionadas con la obtención de energía, entre las cuales podemos mencionar a la β-oxidación, el ciclo de Krebs, la generación del gradiente electroquímico y la síntesis de ATP. El poder reductor (representado por el NADH y FADH2) que se genera en la β-oxidación, la descarboxilación del piruvato y el ciclo de Krebs, pueden ser utilizados por la cadena de transporte de electrones para la generación del gradiente electroquímico de protones (DµH+), el cual dirige la síntesis del ATP por la F1FO-ATP sintasa (Cardol et al., 2009; Letts et al., 2019). Los complejos de la cadena de transporte de electrones se encuentran embebidos en la membrana interna mitocondrial y catalizan las reacciones terminales de transferencia de electrones. Ésta se compone de cuatro grandes complejos proteicos: (1) la bomba de protones NADH-CoQ oxidorreductasa (complejo CI), (2) la succinato- CoQ oxidorreductasa (complejo CII), (3) el dímero obligado de la bomba de protones CoQH2-citocromo c-oxidorreductasa (complejo CIII2) y (4) la bomba de protones citocromo c oxidasa (complejo CIV) responsable 6 de la reducción del oxígeno en agua (Figura 2) (Cardol et al., 2009; Letts et al., 2019). El complejo I (NADH-CoQ oxidorreductasa) es el primero de los complejos de la cadena respiratoria, tiene una masa molecular alrededor de ~1 MDa y está formado por 45 subunidades de proteína dispuestas en dos "brazos": un brazo extrínseco que se extiende hacia la matriz mitocondrial y un brazo de membrana que está incrustado en el MIM. La transferencia de electrones es estimulada por K+ e inhibida por rotenona. El transporte de electrones comienza en el complejo I, el cual acepta los electrones cuando se pierde el ion hidronio del NADH y los transfiere hacia el flavín mononucleótido (FMN), y varios centros fierro-azufre hasta la ubiquinona (Q) convirtiéndola en ubiquinol (CoQH2). Durante este proceso, se obtiene la energía necesaria para el bombeo de 4 H+ al espacio intermembranal o interior de la cresta (lado P). El sitio de oxidación del NADH se encuentra expuesto hacia la cara interna de la MIM, así como el sitio para la reducción de la CoQH2 (lado N). Si tomamos en cuenta que la MIM es impermeable al NADH, el complejo I debe oxidar a los equivalentes reductores que provienen de las deshidrogenaciones de los sustratos del ciclo de Krebs, la descarboxilación oxidativa del piruvato y de la b-oxidación de los ácidos grasos, todas localizadas en la matriz mitocondrial; además de consumir el poder reductor producido en la glucólisis, empleando las lanzaderas de aspartato-malato (Hirst, 2013; Sazanov, 2015). El complejo II (succinato- CoQ oxidorreductasa) es un tetrámero de 140 kDa, que tiene un FAD como grupo prostético y además tiene tres centros fierro-azufre. Esta enzima forma parte del ciclo de Krebs, oxidando al succinato en fumarato. El sitio activo de la succinato deshidrogenasa se encuentra situado en el lado N de la MIM. Por lo tanto, en mitocondrias íntegras, el succinato debe ser generado en la matriz, o bien ser transportado hacia ese EIM para su oxidación. La succinato deshidrogenasa, que reduce a la CoQ, es inhibida por el oxaloacetato y malonato (Chávez et al., 2013). 7 La coenzima Q es una benzoquinona liposoluble que contiene una cadena lateral constituida por 11 unidades de isoprenos (Q10). Esta quinona es reducida por los complejos CI y CII a ubiquinol (CoQH2) y es oxidada por el CIII. Durante su reducción en el CI se realiza el bombeo de H+ al espacio intermembranal, mientras que en el CIII se realiza el ciclo-Q, donde ocurre la oxidación de dos moléculas de CoQH2 y la reducción de una Q, con la formación de una semiquinona (HQ●). Además, durante el ciclo-Q, el CIII acopla la energía libre disponible de la reacción de transferencia de electrones con el bombeo de H+ al espacio intermembranal. Los electrones transferidos de la CoQH2 al citocromo c1 del CIII son translocados al citocromo c, el cual actúa como agente reductor del CIV (Figura 2) (Chávez et al., 2013). La CoQH2-citocromo c-oxidorreductasa (CIII2) es un homodímero, cuyo monómero tiene un peso molecular de 500 kDa y está formado por 10 subunidades, en donde se encuentran 3 centros de óxido-reducción formados por un centro [2Fe- 2S] y los grupos hemo b562, b566 y c1 (Chávez et al., 2013). El complejo III acepta los electrones del CoQH2 y el poder reductor es transferido al citocromo c, el cual es una proteína pequeña de 13 kDa, que se encuentra situada hacia el lado P de la MIM; este citocromo transporta los electrones del complejo III al complejo IV y se libera de la membrana a alta fuerza iónica, utilizando soluciones con altas concentraciones de sales como el KCl. Este complejo es el tercer sitio de acoplamiento, bombea 4 protones y puede inhibirse con Antimicina A y Mixotiazol (Chávez et al., 2013). La citocromo c oxidasa (CIV) tiene una masa molecular alrededor de 240 KDa, se reconoce como la oxidasa terminal y comprende a los citocromos a y a3, está formado por 13 subunidades y contiene tres átomos de cobre iónico (2 CuA y 1 CuB) y los grupos hemo necesarios para la función redox de la enzima. La citocromo oxidasa oxida al citocromo c, reduce al oxígeno molecular (O2) produciendo 2 H2O y bombea 4 H+ al espacio intermembranal. La citocromo oxidasa es inhibida por cianuro, azida de sodio o monóxido de carbono (Chávez et al., 2013). 8 Figura 2. Cadena de trasporte de electrones. Esquema general de la cadena de transporte de electrones, la teoría quimiosmótica propone que los protones se conducen desde la matriz mitocondrial a través de la membrana interna y dentro del espacio intermembranal por el mecanismo de trasporte electrónico para generar un gradiente, el cual es aprovechado para la síntesis de ATP por la ATP sintasa. Dp= fuerza protón motriz. EIM = Espacio InterMembranal (Modificado de Letts et al., 2017). Estos complejos se asocian entre si para formar supercomplejos que pueden ser aislados de la membrana cuando se solubilizan con un detergente suave, como la digitonina. Gracias a esto, se ha propuesto un modelo de estado sólido para la transferencia de los electrones, el bombeo de H+ y el consumo de O2; sin embargo, algunos estudios han demostrado que los supercomplejos no son la entidad funcional esencial de la cadena respiratoria, ya que los complejos respiratorios pueden encontrarse en su forma monomérica (modelo de plasticidad) (Acín-Pérez et al., 2008) y llevar así la transferencia de los electrones y formar el DµH + necesario para la síntesis de ATP por medio de la F1FO-ATP sintasa, proceso que se conoce como fosforilación oxidativa (Schägger, 2000). La teoría quimiosmótica, que explica la fosforilación oxidativa, fue descrita por Peter Mitchell en 1961. En esta se describe que el transporte de electrones de una serie de reacciones de oxido-reducción entre los grupos prostéticos (hemos, flavinas, FeS y Cu2+) de los diferentes complejos respiratorios, que convierten la energía de las coenzimas NADH y FADH2 en energía electroquímica representada por el DµH+. Manipulando la ecuación de Nernst, Mitchell (1961) definió el término conocido como fuerza protón-motriz (Dp), la cual está compuesta por un componente químico (DpH) y un componente eléctrico (DY) de los protones separados por la membrana interna mitocondrial de las células eucariotas o la 9 membrana plasmática de las bacterias. De esta manera, se reconoce como lado positivo (P) a la cara de la membrana interna hacia donde se bombean de forma vectorial los protones (H+), mientras que el lado negativo (N) es la cara de la membrana donde la [H+] disminuye. Posteriormente, el DµH+ se traduce en energía química en forma de ATP a partir de ADP y fosfato (Pi), gracias al flujo de H+ desde el lado P al lado N de la membrana a través del nanomotor conocido como F1FO- ATP sintasa, lo cual se interpreta como la conversión de la energía libre almacenada en el DµH+ (Mitchell, 1961; Boyer, 2000; Nicholls et al., 2013). 10 1.2. Arquitectura y función de la F1FO-ATP sintasa La F1FO-ATP sintasa o complejo V desempeña un papel central al acoplar su actividad al DµH+ para llevar a cabo la síntesis de ATP necesaria para realizar todos los procesos metabólicos incluyendo fusión y fisión (Alfonzo et al., 1981). Esta enzima es considerada un nanomotor y la podemos describir estructuralmente por un sector hidrofílico denominado F1 y un sector hidrofóbico llamado FO; mientras que funcionalmente la podemos dividir en un rotor y un estator. Al considerar ambas clasificaciones se ha definido que el dominio catalítico hidrofílico F1-estator está compuesto por las subunidades solubles a y b; mientras que el dominio F1-rotor contiene las subunidades g, d, y e, con una estequiometría general para la F1 de 3:3:1:1:1 (a, b, g, d, y e, respectivamente). En la interfase de cada par de subunidades a y b se forma un sitio activo, por lo que se tiene un total de 3 (Habersetzer et al., 2013). La estructura de los segmentos catalíticos muestra un alto grado de conservación evolutiva; por ejemplo, se ha determinado que las subunidades de mayor masa molecular de la enzima de las mitocondrias de bovino y de Escherichia coli, comparten un 72% de identidad en sus secuencias. A pesar de su alto grado de complejidad estructural, la F1FO-ATP sintasa mitocondrial es la enzima más estudiada (Yoshida et al., 2001). El dominio F1 está unido al dominio hidrofóbico FO por el FO-estator-tallo periférico el cual está constituido por las subunidades ATP4/b, OSCP, h(F6), d y f, mientras que la porción FO-estator-membranal contiene a las subunidades ATP6(a), ATP8(A6L), i, j y entre 9-11 subunidades ATP9/c componen el FO-rotor (Figura 3). Las subunidades de dimerización son las e, ATP20 (g) y k (Tabla 1) (Walker et al., 2006; Devenish et al., 2008; Watt et al., 2010; Couoh-Cardel et al., 2010). Se han reportado otras proteínas asociadas al sector FO cuya función aún se desconoce, pero se sugiere que podrían estar relacionadas con mecanismos de regulación o que tiene un papel estructural. Adicionalmente, existe una proteína 11 cuyo papel es el de inhibir y/o regular la actividad de hidrólisis de ATP, la cual en Saccharomyces cerevisiae es denominada Inh1 e IF1 en mamíferos. Mientras que en Paracoccus denitrificans, miembro de las a-protobacterias, se encontró el inhibidor denominado subunidad z (Morales-Ríos et al., 2010). La F1FO-ATP sintasa de los mamíferos y las levaduras cuentan con la subunidad OSCP (del inglés “Oligomycin - Sensitivity Conferring Protein”) que le confiere la susceptibilidad al inhibidor oligomicina (Racker, 1693; Pullman et al., 1963; Ebner et al., 1977; Arselin et al., 1996; Minauro-Sanmiguel et al., 2005; Bisetto et al., 2007; Strauss et al., 2008; Vonk et al., 2009;). La F1FO-ATP sintasa es una enzima altamente conservada, sobre todo en aquellas subunidades involucradas en la catálisis y el bombeo de protones (Tabla 1) sin embargo, existen subunidades especie-específicas (Strauss et al., 2008); por ejemplo, las subunidades i/j y k en levadura (Racker, 1693). Figura 3. La F1FO-ATP sintasa. Es la enzima encargada de la síntesis de ATP a partir de ADP y Pi, a través de un mecanismo rotatorio acoplado al gradiente de protones generado por el proceso oxidativo. El flujo de protones desde el EIM hacia la matriz mitocondrial impulsa la rotación de las subunidades c del dominio FO y el tallo del dominio F1 formado por las subunidades g y e, alrededor de la cual se encuentran localizadas las tres subunidades a y las tres subunidades b que tienen una distinta conformación y diferentes afinidades para los nucleótidos, impuestas por la asimetría del tallo central. 12 Tabla 1. Composición de la F1FO-ATP sintasa mitocondrial de S. cerevisiae, B. taurus y U. maydis. Los números corresponden al número de aminoácidos de la proteína madura y, entre paréntesis la inmadura (Modificado de Habersetzer et al., 2013; Esparza-Perusquía et al., 2017). El ID fue recopilado de la base de datos de Uniprot (http://www.uniprot.org) y KEGG (https://www.genome.jp/kegg-bin/show_organism?org=uma). 13 La función de la F1FO-ATP sintasa es producir el ATP a través de un sistema acoplado entre el sector F1 y FO. Esto es posible debido a que la energía almacenada en el DµH+ impulsa el flujo de los H+ desde el interior de la cresta hacia la matriz mitocondrial, a través del sector FO-rotor (Figura 4) (Junge et al.,1997; Allegretti et al., 2015; Guo et al., 2017). El DµH+ tiene dos componentes: un diferencial de pH (DpH) y un potencial eléctrico de membrana (Dym) ambos componentes forman la fuerza protón-motriz (Mitchell, 1961; Boyer, 2000). Cuando los protones atraviesan la membrana acoplados al dominio FO, provocan el giro de un anillo formado por 9 - 15 subunidades c (Cox et al., 1984; Wittig et al., 2008. Kühlbrandt et al., 2016), esta rotación hace girar al tallo central (subunidades d y e del rotor F1) (Boyer et al., 1981) en movimientos de 120°, provocando cambios conformacionales consecutivos en las subunidades catalíticas (subunidades a y b del estator F1) e induciendo la unión de sustratos (ADP + Pi), la síntesis de ATP y la liberación de éste (Figura 5) (Walker, 1998; Itoh et al., 2004). Figura 4. Mecanismo para la síntesis de ATP. Los protones atraviesan la membrana del lado P al lado N a favor del DµH+ a través de un hemicanal formado entre la subunidad a y el anillo de subunidades c generando una rotación en la dirección indicada (flechas). En el modelo aceptado en la actualidad, cada subunidad c contiene un grupo carboxilo de un glutámico que contacta el hemicanal a través del cual pasan los protones. Una vez aceptado el protón por la subunidad c, esta gira dentro de la membrana lipídica y la rotación se completa por una atracción entre el residuo glutámico, con carga negativa, y un residuo de arginina, con carga positiva, de la subunidad a (Modificado de Yoshida et al., 2001). Las subunidades g, d, e y c están involucradas en el movimiento del rotor de la enzima durante la catálisis, mientras que el estator lo forman las subunidades del tallo periférico (Tabla 1), las cuales se mantienen estáticas durante la síntesis de ATP y a su vez mantienen fijas a las subunidades catalíticas a y b (Weber et al., 2003). 14 Para la síntesis de ATP, el sector F1 (heterohexámero a/b) es donde se lleva a cabo la catálisis alternando cambios estructurales (Cabezón et al., 2003; Glendhill et al., 2005). Mediante cristalografía de rayos X se resolvió que el sitio activo de la enzima se encontraba principalmente en la subunidad b en cada uno de los sitios (uno en cada b) y que estos muestran cooperatividad. Esto significa que uno de estos sitios está en la conformación b3 ó sitio T (cerrado que une al ATP fuertemente) y es considerado el sitio catalítico activo, un segundo está en la conformación b1 o sitio L (relajado y con el ADP y Pi unidos) y un tercero está en la conformación b2 o sito O (abierto-vacío de unión muy débil por el ATP, pero afín por el ADP y Pi) (Abrahams et al., 1994; Orriss et al., 1998). La fuerza protón-motriz hace que el eje central rote (subunidad g), el cual se pone en contacto con cada par de subunidades a/b de forma sucesiva. Esto produce un cambio conformacional cooperativo en el que el sitio b3 se convierte en la conformación b2 y el ATP se disocia; simultáneamente el sitio b1 se convierte en la conformación b3, que promueve la condensación del ADP-Pi unidos para formar ATP; y el sitio b2 se convierte en un sitio b1, donde se lleva a cabo la unión del ADP y Pi. Por lo que el DµH+ sirve para el giro de la enzima (cambio conformacional) pero no para la unión del Pi al ADP. La estequiometría H+/ATP puede variar dependiendo del número de subunidades c que forman el semicanal de protones en la membrana, aquí se muestra la estequiometría 3H+/ATP para la F1FO-ATP sintasa de corazón de bovino (9 subunidades c). La flecha central indica la rotación de la subunidad g impulsada por el DµH+ (Figura 5) (Walker, 1998). Este mecanismo que describe la síntesis de ATP también puede ocurrir en sentido contrario durante la hidrólisis del ATP (Adachi et al., 2007). 15 Figura 5. Mecanismo de los sitios catalíticos de la porción F1 de la ATP sintasa durante la síntesis de ATP. Las tres subunidades catalíticas b adquieren diferentes estados conformacionales durante la síntesis de ATP: Abierta (O), semiabierta (L) y cerrada (T). Este ciclo se repite alternadamente en cada una de las subunidades b que componen el dominio catalítico de la enzima (Modificado de Nicholls et al., 2013) Utilizando técnicas de crio-microscopía, se obtuvo un mapa tridimensional de las subunidades involucradas en la translocación de protones, en el cual se pudo observar que la subunidad a tiene una interacción particular con la membrana y con el anillo de subunidades c, ya que presenta un grupo de hélices integrales de membrana, las cuales forman dos horquillas, éstas se acomodan casi de forma paralela a la membrana y parecen abrazar al anillo c (Figura 6A) (Allegretti et al., 2015; Guo et al., 2017; Guo et al., 2018 ). 16 En el modelo propuesto se observa que la disposición de las subunidades a y c da lugar a dos semicanales, cuyas cavidades son hidrofílicas y que permiten la entrada y salida de protones, respectivamente (Figura 6B). Los residuos involucrados en la asociación entre las subunidades a y c están altamente conservados por lo que se considera que el evento ocurre en todas las ATPasas de este tipo (Allegretti et al., 2015; Guo et al., 2018). Figura 6. Estructura de la subunidad a y la translocación de protones. (A) Durante la translocación de protones estos siguen el camino indicado por la línea amarilla (izquierda), ingresando al semicanal desde el espacio intermembranal por las hélices 5 y 6 de la subunidad a y saliendo a través del semicanal de la matriz entre las hélices 5 y 6 del anillo c (B). (Modificado de Guo et al., 2017). (C) Mapa tridimensional propuesto para la subunidad a de Polytomella sp. En este mapa se representa a la subunidad a en color azul y el anillo de subunidades c en color amarillo (Modificado de Allegretti et al., 2015). 17 1.3. Las crestas mitocondriales Las crestas mitocondriales pueden ser estructuras tubulares, lamelares o discoides, las cuales tienen un diámetro de 12 a 40 nm y una longitud de 30 a 50 nm y se conectan con la membrana interna por medio de estructuras tubulares denominadas uniones de cresta. Dichas estructuras limitan el compartimento que corresponde a la cresta con la membrana interna, de esta manera, las uniones de cresta podrían representar barreras para la difusión de moléculas entre el espacio de la cresta y el espacio intermembranal, lo que tendría un impacto en la fosforilación oxidativa (Mannella, 2000; Mannella et al., 2001; Frey et al., 2002; Zick et al., 2009; Mannella et al., 2013). Es en la membrana de las crestas donde se encuentran integrados los complejos de la fosforilación oxidativa, los cuales son responsables de consumir oxígeno y de generar la energía para la síntesis de ATP a través de la F1FO-ATP sintasa. La distribución de los complejos respiratorios y de la F1FO-ATP sintasa también sugiere la presencia de dos compartimientos delimitados por las uniones de crestas, el interior de la cresta y el espacio intermembranal (Figura 7) (Zick et al., 2009; Cogliati et al., 2016). Así, la membrana interna mitocondrial se encuentra dividida en dos secciones, aquella que forma la crestas, donde se encuentran los componentes de la fosforilación oxidativa así como las proteínas encargadas de la síntesis de los centros fierro-azufre y el resto de la membrana interna, donde se encuentran los translocadores que participan en el transporte de proteínas (Wurm et al., 2006). Al igual que con el retículo mitocondrial, la morfología de las crestas varía de una célula a otra e inclusive dentro de la misma célula; sin embargo, los factores que determinan esta morfología aún están por definirse (Zick et al., 2009; Mannella et al., 2013). 18 Figura 7. Representación esquemática de la fosforilación oxidativa en una cresta plegada. Los complejos se ensamblan en diferentes supercomplejos de composiciones muy dinámicas a lo largo de la membrana (en las crestas). Los dímeros de la ATP sintasa están organizados en largas filas en el canto de las crestas y el citocromo c se mantiene dentro de ellas. Así, la cadena de transporte de electrones funciona eficientemente y el gradiente de protones se forma a través de las membranas de las crestas, para ser aprovechado por la ATP sintasa. El espacio delimitado para la cresta se forma con ayuda del complejo MICOS, la proteina OPA, entre otras. MEM = Membrana Externa Mitocondrial; EIM = Espacio InterMembranal; MIM = Membrana Interna Mitocondrial; CM = Cresta Mitocondrial; CJ = Uniones de cresta (Crista Junction) (Modificado de McArthur et al., 2020). Diversos modelos explican la formación de las crestas, como el modelo de invaginación y el modelo de globo. El modelo de invaginación propone que, debido a la mayor producción de membrana interna, hay una tendencia a su plegamiento termodinámicamente favorable; una vez que se inicia este plegamiento, se reclutan las proteínas involucradas en la formación de las uniones de cresta (Figura 8A) (Zick et al., 2009; Song et al., 2014; Milenkovic et al., 2015). En el modelo del globo, hay un agrupamiento inicial de los componentes proteicos de las uniones de cresta (CJ) y a partir de éstos se inicia la formación de la cresta (Figura 8B) (Legros et al., 2002). Asimismo, la topología de las crestas cambia con el estado energético de la mitocondria. En el estado III de la respiración mitocondrial (Chance et al., 1956), cuando la concentración de ADP es alta y la mitocondria entra en el modo de 19 síntesis de ATP, este organelo adopta una conformación o estado condensado, con un espacio muy grande entre las crestas (Mannella, 2006). Por otro lado, cuando la concentración de ADP baja (estado IV de la respiración), la mitocondria cambia al estado ortodoxo, en el cual el volumen entre las crestas disminuye (Mannella, 2006). Esta transición entre el estado ortodoxo y condensado se relaciona con un balance dinámico entre la fusión y la fisión de las crestas. Distintas proteínas “esenciales” generan las crestas, tales como las prohibitinas (PHB1 y PHB2), la proteína OPA1 (Optic atrophy 1), la proteína formadora de la unión de cresta (Fcj1), el complejo MICOS, entre otros componentes que aún no están del todo identificados. Aunado a esto, se reconoce al dímero de la F1FO-ATP sintasa como la principal entidad responsable de la formación de los bordes y las puntas de las crestas (Figura 7) (Zick et al., 2009; Song et al., 2014; Milenkovic et al., 2015). Figura 8. Mecanismos propuestos para la formación de las crestas. A) El modelo de invaginación supone el inicio de la cresta previo a la incorporación de los elementos proteicos de las uniones de cresta (CJ); en contraste, en el modelo del globo (B) el reclutamiento de los elementos proteicos de la CJ es previo a la formación de la cresta. M = Matriz mitocondrial EIM = Espacio InterMembranal; MIM = Membrana Interna Mitocondrial (Modificado de Zick et al., 2009). 20 1.4. La F1FO-ATP sintasa como dímero y su papel en el plegamiento de las crestas mitocondriales La F1FO-ATP sintasa no sólo tiene la función bioenergética de producir ATP, sino que también puede desempeñar un papel morfo-funcional dentro de la arquitectura mitocondrial. En este sentido, se ha reportado que la F1FO-ATP sintasa puede formar homo-oligómeros (Zick et al., 2009), siendo los estados monoméricos y diméricos los más estudiados. Se ha sugerido que los dímeros se asocian formando largas filas, las cuales forzan a la membrana a asumir una curvatura positiva, conservando así la morfología de las crestas mitocondriales (Strauss et al., 2008). La manera por la cual se ha estudiado a los oligómeros de la F1FO-ATP sintasa es solubilizándolos con un detergente suave como la digitonina y separándolos por medio de electroforesis en condiciones nativas en presencia de azul brillante de Coomassie® G250 (BN-PAGE) o de n-dodecil β-D-maltósido (DDM) y desoxicolato de sodio (DOC) (hrCN-PAGE). La ubicación del complejo V en el gel se determina por medio de una tinción que revela la actividad de ATPasa (Figura 9) (Schägger et al., 1991; Wittig et al., 2006). Figura 9. Actividad de ATPasa en gel de los oligómeros de la F1FO-ATP sintasa. Mitocondrias solubilizadas con digitonina donde se muestra la formación de distintos oligómeros de la F1FO-ATP sintasa en A) Corazón de Bovino (Modificado de Strauss et al., 2008), B) Chlamydomonas reinhardtii (Modificado de van Lis et al., 2003), C) Yarrowia lipolítica (Modificado de Wittig et al., 2008), D) Saccharomyces cerevisiae (Modificado de Guerrero-Castillo et al., 2012), E) Ustilago maydis (Modificado de Esparza-Perusquía et al., 2017). La figura muestra la actividad de ATPasa de los Hexámeros (H), Tetrámeros (T), Dímeros (D) y Monómeros (M) que fueron resueltos en geles nativos azules (C) o claros (A, B, D y E). 21 Aunque aún se desconoce el papel bioenergético de los diferentes estados oligoméricos de la F1FO-ATP sintasa, se ha demostrado que el dímero tiene un papel importante en la arquitectura de las crestas de la membrana interna mitocondrial (Arnold et al., 1998). Debido a la asociación angular de dos monómeros, la dimerización conduce a la flexión de la membrana interna mitocondrial, es decir, al plegamiento de las crestas mitocondriales. La presencia de la F1FO-ATP sintasa dimérica en el vértice de las crestas crea una fuerte curvatura positiva que genera una trampa de protones; esto es, la curvatura de la membrana permite que incremente el número de H+ translocados por unidad de área de la MIM, lo que produce el aumento de la [H+] local y por consiguiente una mayor densidad de cargas positivas (Figura 10). En consecuencia, la contribución de la fuerza protón-motriz aumenta en las regiones curvadas (punta y borde de la cresta); suponiendo que la fuerza protón-motriz a lo largo de la membrana interna es constante, todos los protones en las crestas mitocondriales tendrían el mismo potencial químico, pero la contribución del DpH a la fuerza protón-motriz será más grande para los protones que se encuentran en el ápice y el borde, ya que al plegarse la membrana interna el espacio donde se encuentran almacenados los protones se reduce, lo que incrementa la concentración local de los protones que se traduce en una mayor disponibilidad de energía para la síntesis de ATP (Strauss et al., 2008, Rabl et al., 2009). Figura 10. Densidad de protones al interior de la cresta. El esquema representa la densidad de protones dentro de la cresta, (A) la presencia del dímero de la F1FO-ATP sintasa crea una fuerte curvatura positiva que genera una trampa de protones; lo que produce el aumento aparente de la [H+] dentro de la cresta y por consiguiente una mayor densidad de cargas positivas (B) (Modificado de Strauss et al., 2008, Rabl et al., 2009). 22 Aunque la estructura de la F1FO-ATP sintasa es altamente conservada y el plegamiento de las crestas es fundamental para la bioenergética mitocondrial, se ha demostrado que las proteínas involucradas en la interacción monómero-monómero y en la estabilidad del dímero son diferentes en cada sistema biológico. A continuación, se describe la composición proteica del dímero del complejo V en diferentes reinos biológicos. El dímero de la F1FO-ATP sintasa en las algas clorofíceas. Chlamydomonas reinhartii y Polytomella sp son algas clorofíceas o algas verdes. El análisis de la composición peptídica de la F1FO-ATP sintasa en estas especies mostró ocho péptidos canónicos: a, b, g, e, d, OSCP, a (ATP6) y c (ATP9), las subunidades en paréntesis corresponden a las subunidades en mamífero, pero no se encontraron homólogos para las subunidades b, d, e, f, g, IF1, A6L, y F6, que se han reportado como parte del estator de la enzima en el resto de los eucariontes. Tampoco se encontraron aquellas involucradas en la dimerización del complejo; sin embargo se identificaron 10 subunidades que conforman al brazo periférico y alguna de ellas posiblemente son responsables de la dimerización (ASA10). Todas estas subunidades son diferentes a otras proteínas, y han sido denominadas subunidades ASA (del inglés “ATP Synthase-Associated protein”) (Figura 11) (Funes et al., 2002; Cardol et al., 2005; Vázquez-Acevedo et al., 2006; van Lis et al., 2007, Colina- Tenorio et al., 2018). Las F1FO-ATP sintasas de las algas clorofíceas (Figura 11A) tienen un tallo periférico muy robusto formado principalmente por las subunidades ASA, donde éstas tienen diferentes interacciones; esto se ha comprobado mediante experimentos llevados a cabo con subunidades recombinantes. De esta forma se encontró que las subunidades ASA4 y ASA7 interaccionan por medio de sus extremos carboxilos y su vez interaccionan con ASA2 (Miranda-Astudillo et al., 2014). Por otro lado, se ha reportado que ASA6, ASA8 y ASA9 pueden tener cruces 23 transmembranales y podrían estar encargadas de la dimerización (Sánchez- Vázquez et al., 2017). Existen interacciones típicas de una ATP sintasa mitocondrial como OSCP con g y d las cuales se asocian extrínsecamente con ASA2, ASA4 y ASA 7 (Cano- Estrada et al., 2010). Colina-Tenorio et al., (2016) demostró que la subunidad ASA1 es la subunidad del tallo periférico más grande uniéndose desde la subunidad OSCP hasta ASA3, que a su vez mantiene interacción con las subunidades ASA2, ASA7 y ASA8. A pesar de que las subunidades catalíticas presentan alta similitud con el resto de las ATP sintasas, éstas presentan diferentes extensiones en la secuencia de aminoácidos; por ejemplo, la subunidad a cuenta con aproximadamente 20 aminoácidos en el extremo N-terminal y la subunidad b contiene 60 aminoácidos extras en el extremo C-terminal. Aunque aún no es clara la función de esta última extensión, se ha propuesto que podría desempeñar un papel regulador de la actividad del complejo, ya que aún no se ha identificado el homólogo de la proteína inhibidora IF1 (Villavicencio-Queijeiro et al., 2009). El dímero de la ATP sintasa de estas especies pesa aproximadamente 1600 kDa y es altamente estable (van Lis et al., 2003; Cano-Estrada et al., 2010), siendo las subunidades Asa2/Asa4/Asa5/Asa8 las que estabilizan la unión entre lo sectores FO y F1 (Villavicencio-Queijeiro et al., 2015). Cuando el dímero se disocia es sus respectivos monómeros, se ha observado una disminución en la proporción de las subunidades ASA6, ASA8 y ASA9, que probablemente están involucradas en la dimerización (van Lis et al., 2007). Todas estas nuevas subunidades están involucradas en la formación del estator de la enzima, aunque todavía no es clara la función de cada una de ellas dentro del complejo enzimático. Se ha sugerido que podrían ser esenciales para el 24 buen ensamble del complejo, estabilizando una forma dimérica necesaria para mantener la morfología de las crestas mitocondriales (van Lis et al., 2007). Figura 11. Estructura del dímero de la F1FO-ATP sintasa de Polytomella sp. Las proteínas involucradas en la asociación entre los monómeros son altamente conservadas en las algas clorofíceas, éstas cuentan con nueve subunidades ASA, las cuales sustituyen a las subunidades clásicas que forman el estator de la enzima. (A) Estructura tridimensional generada a partir de criomicroscopía electrónica (Modificado de Kühlbrandt et al., 2016; Allegretti et al., 2015). (B) Mapa de densidad electrónica de la F1FO-ATP sintasa dimérica de Polytomella sp. El mapa de densidad electrónica corresponde al generado por Allegretti et al., 2015 (EMD-2852) (Colina-Tenorio et al., 2018). El dímero de la F1FO-ATP sintasa en los mamíferos. La F1FO-ATP sintasa del corazón de bovino presenta las subunidades catalíticas homólogas a E.coli, así como las subunidades que conforman el rotor de la enzima, el cual está compuesto por las subunidades b, d, F6, OSCP y como proteína reguladora a la proteína inhibidora IF1. Además, cuenta con las subunidades supernumerarias membranales e, g, A6L y f. Se ha reportado que las subunidades a y A6L (ATP8) también tienen un papel importante dentro de la estabilización del homo-dímero (Wittig et al., 2010). Se ha reportado que la subunidad IF1 promueve la dimerización de la F1FO- ATP sintasa (Cabezón et al., 2000; García et al., 2006; Couoh-Cardel et al., 2010; De los Rios-Castillo et al., 2011). Experimentos de entrecruzamiento demostraron que la IF1 interactúa con el dominio C-terminal de una subunidad b. Durante años, esto ha hecho suponer que el mecanismo de inhibición de la IF1 es el de interferir con los cambios conformacionales de las interfases catalíticas necesarios para la síntesis y la hidrólisis de ATP (Minauro-Sanmiguel et al., 2002). Se ha logrado 25 resolver la estructura de la proteína inhibidora aislada, la cual resultó ser una larga a hélice extendida con un dominio inhibitorio localizado en la región N-terminal. En el extremo C-terminal la IF1 contiene un dominio de dimerización con la cual se auto- asocia, y que podría inducir la dimerización de la F1-ATPasa soluble (Cabezón et al., 2000; Cabezón et al., 2003). Aunque la presencia de la IF1 no es esencial para la dimerización (Tomasetig et al., 2002), podría inducir la formación de un puente dimérico que conecta las dos F1 solubles (Cabezón et al., 2000; Cabezón et al., 2003). Por otra parte, la sobreexpresión y/o reconstitución de la IF1 modifica la relación monómero/dímero a favor de los complejos diméricos. Lo anterior sugiere que la IF1 podría cumplir un papel estabilizador en las mitocondrias de corazón de bovino mediante la formación de un puente ≤ 12Å que la conecta con las subunidades g del rotor central de la F1 (Minauro-Sanmiguel et al., 2002; Bravo et al., 2004). Mediante microscopía electrónica, Minauro-Sanmiguel et al., (2005) resolvierón la estructura de la F1FO-ATP sintasa dimérica aislada de las mitocondrias de corazón de bovino y encontró que la interfase del dímero está formada por el contacto tanto entre los dominios F1 como entre los dominios FO, que además pueden conectarse a través de la proteína IF1. Por otro lado, se encontró que el dímero de bovino presenta dos puentes: uno que une a los sectores F1 y otro a los sectores FO; se propuso que éstos corresponden a la dimerización de la IF1 y la subunidad e, respectivamente (Minauro-Sanmiguel et al., 2005). Dado que la proteína inhibidora promueve y estabiliza a la estructura dimérica de la F1FO-ATP sintasa en corazón de bovino y en el hígado de rata, se sugiere que la IF1 forma una estructura cruzada en la interfase del dímero que explica tanto su papel inhibitorio como su participación en la dimerización (García et al., 2006). Otros estudios han demostrado que al eliminar la subunidad IF1 se pierden los dímeros y simultáneamente la arquitectura mitocondrial se ve afectada, particularmente en la formación de crestas (Campanella et al., 2008). Lo anterior apoya la hipótesis de la IF1 como proteína dimerizante (Campanella et al., 2008). 26 Recientemente Gu et al., (2019) aislaron la ATP sintasa tetramérica porcina (Sus scrofa domestica) y resolvieron su estructura a 6.2 Å (Figura 12A) utilizando un método de criomicroscopía crioelectrónica (Cryo-EM). Demostraron que dos dímeros de la F1FO-ATP sintasa se encuentran antiparalelos entre sí para formar un tetrámero unido por un dímero de IF1. Los dos dímeros están unidos entre sí a través de seis sitios (Figura 12B). Para los sitios de unión 1 y 6, dos proteínas IF1 forman un dímero antiparalelo que interactúa con los sectores F1 dentro de la matriz mitocondrial (Figura 12B). Los sitios 2 y 4 están por encima de la membrana, incluyen el bucle N-terminal de la subunidad k y la hélice N-terminal de la subunidad b de dos dímeros adyacentes. El sitio 3 incluye dos hélices N-terminales de subunidades g de dímeros opuestos dentro del tetrámero, colocadas en paralelo en el centro de la estructura (Figura 12B). El sitio 5 está dentro de la membrana interna, donde dos subunidades e que interactúan entre sí, mantienen al tetrámero unido (Figura 12C). 27 Figura 12. Dímero de la F1FO-ATP sintasa porcina y sus sitios de unión. (A) Modelo cristalográfico del dímero de la F1FO- ATP sintasa obtenido a 6.2 Å. (B) Sitios de interacción de los tetrámetros vistos desde el lado de la matriz. Los modelos de alta resolución se ajustaron a los mapas de tetrámero con un coeficiente de correlación de 0.85. (C) Vista desde el espacio intermembranal (Modificado de Gu et al., 2019). El dímero de la F1FO-ATP sintasa en un tejido humano. Adicionalmente, se ha demostrado que la expresión de la IF1 y su repercusión en la dimerización del complejo V está asociada a procesos de diferenciación celular; por ejemplo, la transformación del citotrofoblasto en sinciciotrofoblasto y la adquisición del fenotipo esteroidogénico (Figura 13) (De los Rios-Castillo et al., 2011). Las micrografías electrónicas de las vellosidades de la placenta humana a 28 término, muestran que la arquitectura mitocondrial del citotrofoblasto obedece a la estructura tradicional, tienen forma redondeada y presenta crestas lamelares en una configuración ortodoxa (Figura 13B) y que el dímero del complejo V se encuentra en alta concentración (Figura 13C). Por otra parte, el sinciciotrofoblasto contiene mitocondrias pequeñas de forma irregular con protuberancias de la membrana interna y externa, con una matriz condensada (Figura 13B) y las crestas en forma de vesículas, debido a que la F1FO-ATP sintasa se encuentra principalmente como monómero (Figura 13C) (De los Ríos-Castillo et al., 2011). Se determinó que la cantidad de la IF1 es mayor en las mitocondrias de citotrofoblasto, lo que está directamente relacionado con la formación del dímero y la arquitectura mitocondrial observada. Figura 13. Ultraestructura mitocondrial de las células de placenta humana. Las micrografías muestran que la arquitectura de las mitocondrias es diferente aun tratándose del mismo tejido. (A) Corte de la placenta humana donde se muestra el cito y el sinciciotrofoblasto, las flechas hacen referencia las mitocondrias. (B) Mitocondrias aisladas del cito y del sinciciotrofoblasto. (C) La actividad de ATPasa de los complejos solubilizados de los dos tipos mitocondriales, donde se muestra que el dímero solo está presente en las mitocondrias del citotrofoblasto (Modificado de De los Ríos-Castillo et al., 2011). De los Ríos-Castillo et al., (2011) reportó que los valores de control respiratorio se encuentran entre 2.85 y 12 natgO/mg/min, y la síntesis de ATP por la F1FO-ATP sintasa fue de 151 ± 16 y 153 ± 13 nmol/mg/min para las mitocondrias del citotrofoblasto y sinciciotrofoblasto respectivamente, lo que demuestra un acoplamiento entre la respiración mitocondrial y la síntesis de ATP. Esto indica que las mitocondrias esteroidogénicas y no esteroidogénicas de la placenta humana son capaces de sintetizar ATP aún sin mostrar la misma arquitectura de sus crestas. 29 Esto sugiere que aunque el dímero tiene un papel estructural importante, el monómero puede cubrir el papel bioenergético (De los Ríos-Castillo et al., 2011). El dímero de la F1FO-ATP sintasa en las levaduras. La F1FO-ATP sintasa de las levaduras presentan algunas subunidades conservadas entre bacterias, algas y mamíferos y también pueden adoptar estructuras supramoleculares. A diferencia de los dímeros de mamífero, el dominio de interacción entre los monómeros de S. cerevisiae es el sector FO. Se ha establecido que las subunidades e, g y k están asociadas únicamente al dímero de la F1FO-ATP sintasa de S. cerevisiae y no a su forma monomérica (Arnold et al., 1998). También se ha sugerido que la subunidad a tiene un papel importante formando parte de la base para la dimerización ya que se le ha predicho un alto número de hélices transmembranales (Wittig et al., 2008). Junto con la subunidad a, las subunidades del tallo estator y subunidades accesorias (e, g, b, i, k y 8) estabilizan la interfaz del homo-dímero (Figuras 14A) (Wittig et al., 2008; Wittig et al., 2010, Couoh-Cardel et al., 2010). Se ha demostrado que una alteración en los componentes arriba descritos desestabiliza las estructuras oligoméricas, lo que conduce a la aparición de especies monoméricas y morfologías anómalas en las crestas mitocondriales, que se describen como aros de cebolla (Velours et al., 2009). Sin embargo, en las mutantes de las subunidades e, g y k no se eliminan del todo las crestas mitocondriales, pero hacen sensible al dímero, lo cual sugiere que no son esenciales para la dimerización, pero sí para su estabilidad (Figura 14B). En contraste, se ha podido demostrar que la subunidad Inh1 (proteína inhibidora de la F1FO-ATP sintasa de levadura) no es necesaria para la oligomerización (Arnold et al., 1998; Dienhart et al., 2002; Wittig et al., 2008; Wagner et al., 2010). 30 Por su parte, la subunidad g tiene dos a hélices en el extremo N-terminal que se encuentran expuestas en la superficie del sector FO y una única a hélice transmembrana que interactúa con la subunidad e, probablemente a través de los motivos conservados Gly-XXX-Gly de las dos proteínas (Figura 14C) (Bustos et al., 2005; Saddar et al., 2005). La estructura curva del dominio formado por las subunidades e, g y b sugiere cómo las subunidades e y g, con el apoyo adicional de la subunidad k, crean una estructura que refuerza la curvatura de la bicapa lipídica (Figura 14D) (Guo et al., 2017) y porqué la supresión de los genes que codifican a estas subunidades en la levadura conduce a defectos en la formación de crestas (Paumard et al., 2002). Como se puede observar, cada reino biológico ha desarrollado una estrategia diferente para formar y/o estabilizar los dímeros del complejo V y plegar así la membrana interna mitocondrial en crestas, lo que podría ser un evento de convergencia evolutiva. Por lo tanto, es importante ampliar el conocimiento sobre la composición del dímero de la F1FO-ATP sintasa en otros organismos y determinar cuáles son sus subunidades dimerizantes. Por otro lado, mediante el empleo de la microscopía de fuerza atómica se ha analizado la formación de los oligómeros de la F1FO-ATP sintasa de S. cerevisiae desde la membrana interna-cresta, lo cual permitió detallar la posición espacial de los canales de protones y de 3 densidades circulares que corresponden a los dominios transmembranales de las otras subunidades que conforman los segundos dominios de los sectores FO. Así mismo, se describió la existencia de dos tipos de dímeros cuyos puntos más distantes de los anillos de sub c miden 15 y 10 nm; con esto se propuso que el dímero con una mayor distancia corresponde a un estado activo que sintetiza ATP y que es capaz de unirse a otros dímeros para oligomerizarse, mientras que el dímero más cerrado se forma durante la inhibición de la hidrólisis de ATP y es incapaz de oligomerizarse (Buzhynskyy et al., 2007). 31 Figura 14. Estructura de la F1FO-ATP sintasa mitocondrial de S. cerevisiae a 25 Å de resolución. (A) Reconstrucción tridimensional de la F1FO-ATP sintasa dimérica por microscopía electrónica, donde se muestra la disposición de las subunidades propuesta según su densidad electrónica (Modificado de Couoh-Cardel et al., 2010). (B) Identificación de las subunidades que componen al monómero (Modificado de Couoh-Cardel et al., 2010). (C) Vista superior del dímero FO, que revela la disposición de las subunidades e, g y b. El monómero se perfila con una línea punteada. Resolución a una escala de 25 Å (Modificado de Guo et al., 2017). (D) Vista lateral del dímero FO muestran que las subunidades b, e, g y k crean la estructura que dobla la bicapa lipídica casi 90°. La línea naranja discontinua indica la longitud completa de la subunidad k. Resolución a una escala de 25Å (Modificado de Guo et al., 2017). Papel del dímero en las crestas. En la actualidad, se cuenta con la primera reconstrucción tomográfica de una cresta tubular mitocondrial de hígado de rata con un diámetro de 34 nm, a partir de la cual se ha determinado que el dímero de la F1FO-ATP sintasa tiene un ángulo de 70º, con una distancia entre F1 - F1 de 28 nm y entre los centros FO-FO de 13 nm (Figura 15) (Strauss et al., 2008). En corazón de bovino el ángulo oscila entre 55º y 95º, aunque la microscopía electrónica permitió definir un ángulo de 40º entre los cuellos centrales de los monómeros (Minauro-Sanmiguel et al., 2005; Strauss et al., 2008). En S. cerevisiae, se observaron dos familias de ángulos, una de 35º y otra de 90º; y en Polytomella sp, la formación dimérica tiene un ángulo de 70º. Estudios recientes han definido la existencia de al menos 6 tipos diméricos en S. cerevisiae (Figura 16) (Thomas et al., 2008). 32 Figura 15. Reconstrucción tomográfica de una cresta tubular mitocondrial de hígado de rata. (A) Tomografía de membranas mitocondriales de hígado de rata. Las flechas grises indican los mismos pares de cabezas de F1FO-ATP sintasas ordenadas linealmente. Barra de escala en panel A, 50 nm. (B) Reconstrucción tridimensional de una cresta tubular. Las cabezas de F1 son amarillas, mientras que la membrana es gris. Las partículas no asignadas a las cintas de dímero se muestran en gris más claro. La longitud del tubo en el panel B es de 280 nm, y el diámetro de las vesículas es 51 nm (Modificado de Strauss et al., 2008). Se ha sugerido que los diferentes ángulos dependen de la forma en la que el detergente solubiliza a los oligómeros, los cuales se definen como dímeros verdaderos (estructuras abiertas) y pseudo-dímeros (estructuras cerradas). Los cuellos periféricos de éstos últimos no son observables debido a que, probablemente, se localizan detrás de los cuellos centrales mientras que en el dímero verdadero son visibles en una vista lateral (Dudkina et al., 2006). Figura 16. Dímeros aislados observados por microscopía de trasmisión electrónica. Los dímeros de diferentes especies han sido aislados a través de gradientes de glicerol o sacarosa. (A) Polytomella sp. presentó un dímero con una amplitud de 70° (Modificado de Dudkina et al., 2006). (B) Bos Taurus tiene una amplitud de 40°, se propone que el puente generado por las FO´s (flecha negra) es debido a la dimerización de la subunidad e mientras que el formado por las F1´s, la dimerización de IF1 (Modificado de Minauro-Sanmiguel et al., 2005). (C) S. cerevisiae presenta un ángulo de 60° dado por las interacciones entre las subunidades e y g (Modificado de Couho-Cardel et al., 2010). 33 1.5. Otros factores que participan en la formación de las crestas mitocondriales Como ya se mencionó, las crestas están formadas por dos hojas de la membrana interna, que están paralelas y muy próximas, dejando entre sí un espacio estrecho llamado lumen de la cresta. Estas hojas de membrana interna están unidas en sus bordes formando puntas o cantos, en donde la bicapa lipídica se dobla con una curvatura positiva. En su base la cresta tiene forma tubular y se conecta a la membrana interna en una región conocida como frontera de la membrana interna. Estas conexiones están limitadas por las uniones de cresta y en ellas la bicapa lipídica tiene una curvatura negativa provocada por una proteína denominada Fcj1 (proteína formadora de la unión de cresta); estas uniones de cresta controlan la apertura de la cresta hacia el espacio intermembranal (Rabl et al., 2009). Se ha propuesto que la arquitectura de la cresta depende del equilibrio entre la concentración del dímero de la F1FO-ATP sintasa y la proteína Fcj1. Así pues, las puntas de las crestas y los cantos están formados por los dímeros de la F1FO-ATP sintasa, los cuales imponen una curvatura positiva a la membrana (Rabl et al., 2009) y donde la proteína Fcj1 está ausente; por el contrario, en las uniones de cresta los dímeros están virtualmente ausentes, lo que permite una curvatura negativa de la membrana provocada por la presencia de la proteína Fcj1. La cara de la cresta contiene tanto a la proteína Fcj1 como al dímero de la F1FO-ATP sintasa, lo que puede determinar su forma plana (Figura 17). Aunque esta hipótesis trata de explicar la distribución dinámica del dímero de la F1FO-ATP sintasa y la proteína Fcj1, no se puede excluir que otros componentes tales como las prohibitinas, la proteína OPA1 u otros componentes que aún no se identifican, participen en la formación de las uniones o las puntas y cantos de las crestas (Paumard et al., 2002; Rabl et al., 2009). 34 Figura 17. Elementos proteicos involucrados en la formación de las crestas mitocondriales. Representación esquemática de la curvatura de la membrana en las diferentes regiones de la cresta y la localización submitocondrial de la proteína Fcj1 y las subunidades e y g. La curvatura positiva de la membrana está indicada en azul, la curvatura negativa en rojo, y las regiones con ambas curvaturas o ninguna en especial está en purpura. El esquema representativo de una cresta con una sola CJ se muestra en su vista lateral (izquierda) y su sección transversal después de un giro de 90º (en el medio). Las ampliaciones del área enmarcada muestran el arreglo propuesto de Fcj1, Sub e/Sub g, y la F1FO-ATP sintasa y su influencia en la curvatura de la membrana. El recuadro de la izquierda muestra la estructura tubular propuesta para la unión de las crestas (CJ) a la membrana interna mitocondrial. (Modificado de Rabl et al., 2009). 35 1.6. Efecto de la eliminación de las subunidades dimerizantes y su repercusión en la arquitectura mitocondrial Como se mencionó, dentro de las proteínas necesarias para la formación de las crestas mitocondriales se encuentran las formas oligoméricas de la F1FO-ATP sintasa. Al eliminar los genes que codifican para las subunidades e y g en S. cerevisiae se producen solamente monómeros de la F1FO-ATP sintasa, se altera la arquitectura de las crestas y las mitocondrias adquieren morfologías o formas inusuales (Paumard et al., 2002; Rabl et al., 2009; Velours et al., 2009). En estas condiciones S. cerevisiae crece lentamente en medios no fermentables y las crestas mitocondriales se observan como discos concéntricos en forma de “aros de cebolla” (Figura 18) (Arselin et al., 2004). Sin embargo, no se ha realizado un estudio sistemático de la síntesis e hidrólisis de ATP en estas cepas mutantes. También la formación de los supercomplejos III2IV se ve afectada (Saddar et al., 2008) y el potencial de membrana disminuye, motivo por el cual cerca de 40% del ADN mitocondrial se pierde (Bornhövd et al., 2006; Duvezin-Caubet et al., 2006). Figura 18. Diferencias estructurales en las mitocondrias de Saccharomyces cerevisiae. Micrografía de las mitocondrias de Saccharomyces cerevisiae (A) cepa WT y (B) cepa mutante De/Dg, en la cual se observan las estructuras anómalas de las mitocondrias, crestas en forma de aros de cebolla. m=mitocondria. La barra indica 500nm (Modificado de Velours et al., 2009). 36 Aunado a esto, se ha propuesto que la oligomerización de la F1FO-ATP sintasa puede estar involucrada de manera directa en la actividad catalítica de la enzima y de manera indirecta también puede afectar el estado metabólico de las células, ya que cuando se impide la dimerización de la F1FO-ATP sintasa eliminando genéticamente a las subunidades e y g, las mitocondrias presentan una morfología alterada en las crestas. Así pues, el rearreglo de las crestas durante la regulación bioenergética de la mitocondria supone el cambio en la proporción dímero/monómero del complejo V (Velours et al., 2009). En el caso de las mitocondrias de las células HeLa, Campanella et al.,(2008) demostraron que al eliminar la subunidad IF1 se pierden los dímeros y simultáneamente la arquitectura mitocondrial se ve afectada; en particular disminuye el número de crestas por mitocondria y por μm2 y la muerte celular inducida incrementa hasta en un 40% (Figura 19). Cuando la respiración celular se ve comprometida, la célula permite la despolarización mitocondrial para promover la liberación de inductores proapoptóticos que desencadenan las vías de muerte celular. Sin embargo, cuando se sobreexpresa a la subunidad IF1, se observa un aumento en los niveles del dímero del complejo V y en la actividad de síntesis de ATP (Campanella et al., 2008). Figura 19. Ultraestructura mitocondrial de las células Hela. (A) Mitocondrias control y (B) cepa mutante -IF1, donde se puede observar que el interior de la mitocondria está desorganizado y con disminución de las crestas mitocondriales. Las flechas rojas señalan las mitocondrias por campo. Simultáneamente se reportó la pérdida del DY y la síntesis de ATP (Modificado de Campanella et al., 2008). 37 1.7. Ustilago maydis como modelo experimental. U. maydis es un hongo biotrófico que causa el carbón del maíz, comúnmente llamado cuitlacoche o huitlacoche. Se presenta como agallas negras en el maíz y se ha utilizado como un ingrediente tradicional en la gastronomía mexicana. U. maydis ha sido un modelo ejemplar en los estudios de fitopatogenicidad, dimorfismo, eventos de recombinación homóloga e interacciones planta-patógeno (Vollmeister et al., 2012). U. maydis durante su ciclo de vida transita por tres diferentes etapas morfológicas: saprofítica, en la que crece en forma de levadura o basidioespora haploide no patógena; el micelio dicariótico que aparece cuando dos basidiosporas compatibles se fusionan, que es la forma infecciosa del organismo, y la teliospora o esporas binucleadas de latencia, de gran resistencia a cambios ambientales adversos y que constituye el mecanismo de diseminación del hongo (Herrera et al., 1990; Feldbrugge et al., 2004). El ciclo de vida se inicia cuando las teliosporas germinan sobre la superficie del tejido del hospedero produciendo el promicelio, a partir del cual se liberan las esporidias haploides o levaduras. Cuando éstas son compatibles (A1B1-A2B2), se fusionan y se forma el micelio dicariótico que invade a la planta y que finalmente dará origen a las teliosporas. Estas permanecen viables en el ambiente hasta que se presentan las condiciones favorables para su germinación (Figura 20) (Herrera et al., 1990; Bölker et al.,1992; Bölker, 2001). Los cambios en la patogenicidad, la morfología y la invasión a la planta se regulan por las vías de la MAPK y del AMPc (Mendoza-Mendoza et al., 2009; Kahmann et al., 1999; Doehlemann et al., 2008; Feldbrugge et al., 2004; Lengeler, 2000; Reifschneider et al., 2005; Rexroth et al., 2004; Marques et al., 2007). 38 Figura 20. Ciclo de vida de U. maydis. El área sombreada en verde claro indica los procesos que son absolutamente dependientes de la presencia de la planta. Los núcleos azules y rojos indican los diferentes tipos de apareamiento en el locus A y B y se utiliza para visualizar las fases haploide y diploide dicarióticas durante el ciclo de vida. En el centro del diagrama se muestra una mazorca de maíz infectada con síntomas de la enfermedad típica. En la parte inferior, el tumor se rompe y revela cantidades masivas de teliosporas negras. Cuando las esporas diploides germinan sufren meiosis y producen esporidios haploides. En este trabajo se emplearon a las levaduras haploides (Modificado de Feldbrügge et al., 2004). U. maydis ofrece grandes ventajas como modelo biológico, por ejemplo, aunque requiere a la planta para completar su ciclo de vida, en el laboratorio es posible obtener levaduras o micelio en función de las condiciones de crecimiento. Así mismo, se pueden producir cepas mutantes nulas carentes de genes de interés (knock-out); este proceso es relativamente sencillo, debido a la alta eficiencia con que la recombinación homóloga se lleva a cabo en este hongo. Por otra parte, debido a que U. maydis presenta un metabolismo aerobio estricto, de tipo no fermentativo y que depende totalmente de la fosforilación oxidativa para el suministro de ATP, la mitocondria es parte esencial para el crecimiento de la levadura. La cadena de transporte de electrones y fosforilación oxidativa de U. maydis está compuesta por los 5 componentes clásicos, el complejo I (NADH: ubiquinona oxidorreductasa), el complejo II (succinato: ubiquinona oxidorreductasa), el complejo III (ubiquinol: citocromo c oxidorreductasa), el 39 complejo IV (citocromo c oxidasa) y el complejo V (F1FO-ATP sintasa). Además de éstos, cuenta con otros componentes alternos como la oxidasa alterna (AOX) y las deshidrogenasas alternas externas (Nde1 y 2) e interna (Ndi1). La AOX es una enzima respiratoria que cataliza una reacción de oxidación terminal a nivel de la poza de quinonas, de manera similar a lo que harían en conjunto los complejos III y IV en la vía citocrómica, pero sin generar una diferencia de potencial de membrana, por lo que dicha enzima no participa en la síntesis de ATP (Figura 21). Figura 21. Cadena de transporte de electrones en U. maydis. Esquema general de la cadena de transporte de electrones, en el cual podemos localizar los cuatro complejos respiratorios clásicos, la F1FO- ATP sintasa, un para de deshidrogenasas alternas (externa e interna) y la oxidasa alterna (AOX) (Modificado de Rogov et al., 2014). 40 “Mitonulceus” Odra Noel Justificación “Lo importante es que la ciencia avance” Dr. Edmundo Chávez 41 2. JUSTIFICACIÓN U. maydis es un basidiomiceto fitopatógeno del maíz que provoca grandes pérdidas económicas a nivel mundial. Otros ustilaginales infectan plantas como el arroz (Oryza sativa) y el sorgo (Sorghum bicolor). Es importante incrementar el conocimiento sobre esta familia, a fin de contar con más herramientas para ayudar a su control biológico. Se ha reportado que el dímero de la F1FO-ATP sintasa tiene un papel importante en la formación de las crestas de la membrana interna mitocondrial, y su ausencia provoca crestas anormales, y pérdida del potencial de membrana (DYm) y de la síntesis de ATP, es importante mencionar que debido a que U. maydis es un organismo aerobio, de tipo no fermentativo, que depende totalmente de la fosforilación oxidativa para el suministro de ATP, U. maydis es un organismo aerobio, de tipo no fermentativo, que depende totalmente de la fosforilación oxidativa para el suministro de ATP. La mitocondria es la encargada de oxidar los equivalentes reductores provenientes de la glucólisis, el ciclo de Krebs y la b- oxidación y así alimentará la cadena de transporte de electrones y convertir el poder reductor en ATP. El dímero de la F1FO-ATP sintasa de U. maydis tiene 17 subunidades homólogas a las reportadas para S. cerevisiae, incluyendo a las subunidades dimerizantes g y e (Arnold et al., 1998; Wittig et al., 2008; Esparza-Perusquía et al., 2017). Con base en lo anterior, se decidió construir la cepa mutante de la subunidad g en U. maydis y determinar la repercusión de la eliminación de dicha subunidad (Dg) sobre la formación del dímero de la ATP sintasa, el plegamiento de las crestas, el retículo mitocondrial y el estado bioenergético (DYm, consumo de oxígeno y síntesis de ATP). Asimismo, se pretende caracterizar la actividad de ATPasa de los dímeros (V2Dg) y en los monómeros (V1Dg) aislados de la cepa mutante. 42 Odra Noel Objetivos 43 3. OBJETIVOS 3.1. OBJETIVO GENERAL Determinar la repercusión de la eliminación de la subunidad g en la oligomerización y la actividad enzimática de la F1FO-ATP sintasa de Ustilago maydis, así como su papel en la formación de crestas mitocondriales y en el estado bioenergético mitocondrial. 3.2. OBJETIVOS PARTICULARES • Generar la cepa mutante Dg de la F1FO-ATP sintasa de Ustilago maydis. • Determinar la presencia del V2Dg y V1Dg. • Aislar los oligómeros (V2Dg y V1Dg) de F1FO-ATP sintasa. • Caracterizar la actividad de hidrólisis de ATP de los oligómeros aislados de la F1FO-ATP sintasa. • Determinar la morfología de crestas mitocondriales de la cepa Dg en comparación con la cepa silvestre (WT). • Cuantificar el potencial de membrana, la respiración y síntesis mitocondrial de ATP en las cepas WT y Dg de Ustilago maydis. 44 “Mitchell’s equation I” Odra Noel Materiales y métodos “Si modifica el protocolo le puede costar sangre” Dr. Armando Gómez-Puyou 45 4. MATERIALES Y METODOS Cultivo celular para biología molecular. La cepa silvestre FB2 (a2b2-WT; ATCC 201384) de Ustilago maydis se sembró en un medio sólido de YPD (0.5% (p/v) extracto de levadura, 0.25% (p/v) bactopeptona, 0.5% (p/v) de glucosa, 2% (p/v) de agar), se incubó por 18 horas a 28°C. A partir de ese cultivo se inoculó una azada de células en 3 mL de YPD y se incubaron durante 12 h a 28°C en agitación orbital constante. Obtención del DNA genómico (gDNA) de la cepa WT de U. maydis. El cultivo anterior se centrifugó a 15,700 g durante 2 min para recuperar las células. Éstas se rompieron con perlas de vidrio (0.5 mm) en 500 µL de amortiguador de lisis (10 mM Tris-HCl pH 8.0, 10 mM NaCl, 1% (p/v) SDS, 2% (v/v) Tritón X100, 1 mM EDTA) y 500 µL de la solución de fenol/cloroformo en una relación 1:1, en agitación constante durante 10 min. El DNA se recuperó de la fase acuosa (400 µL), el cual se transfirió a un tubo que contenía 1 mL de etanol al 100% para lavar el DNA; dicho volumen se centrifugó para compactar el DNA. Finalmente, se incubó en agitación orbital constante (400 rpm) durante 10 min a 50°C en presencia de 50 µL de TE/Rnase (1.31 mM Tris-Base, 8.69 mM Tris-HCl, 10 mM Na2-EDTA y 10 mg/mL de Rnase-A para remover el RNA). Para corroborar la integridad del DNA genómico (gDNA), las muestras se resolvieron en un gel de agarosa al 0.8% (p/v), 1x-TBE (89 mM Tris-base, 89 mM ácido bórico, 2 mM EDTA, pH 8.0), 25 µg/mL bromuro de etidio (BrEt) y la concentración se determinó en un NanoDrop 2000c (Thermo Scientific). Diseño de los oligonucleótidos. Para la construcción del knock-out (KO) se amplifican los fragmentos UF (upstream flank) y DF (downstream flank) de aproximadamente 1000 pares de bases flanqueando al gen que codifica para la subunidad g (ATP20), el diseño de los oligonucleótidos se realizó en el programa SnapGene versión 2.0.1 y se sintetizaron en Ludwig-Maximilians Universiät München, Alemania. 46 Para el caso del fragmento UF los oligonucleótidos correspondientes para las diferentes PCR son denominados U1-Fw (aproximadamente 20 pb antes de UF) 5’- GCGTAGTCGATGTCCTTGG-3’, U2-Fw (aproximadamente 1000 pb antes del gen ATP20) 5’-atttaaatGCTTCCTTGTATTCGGC-3’ (las letras en minúsculas indican el sitio de corte para la enzima SwaI), U3-Rv (aproximadamente 20 pb antes del gen ATP20) 5’-ggccatctaggccGATGACCGTATTACCCGAAAGAC -3’ (las letras en minúsculas indican el sitio de corte para la enzima SfiI) y P1-Fw (aproximadamente 20 pb al final del UF) 5’- TCCTCACACCATCCCCTT-3’. Para el fragmento DF se diseñaron los oligonucleótidos D1-Fw (aproximadamente 20 pb después del gen ATP20) 5’- ggcctgagtggccACGCTCGACAATTGAATTCG -3’ (las letras en minúsculas indican el sitio de corte para la enzima SfiI), D2-Rv (aproximadamente 1000 pb después del gen ATP20) 5’- atttaaatCTGGCATGTGCTCACC -3’ (las letras en minúsculas indican el sitio de corte para la enzima SwaI) y D3-Rv (aproximadamente 1020 pb después del gen ATP20) 5’- GGCCTGCCGTATCAAGTC -3’. Los oligonucleótidos U2, U3, D1, D2 fueron diseñados con sitios de corte para las enzimas de restricción específicas, ya que el cassette de resistencia a la higromicina (HygR) cuenta con esos sitios para la ligación y formación del plásmido KO. P2-Rv (aproximadamente 20 pb al inicio del gen ATP20) 5’- GGGAGCAGCAGCTCGGC -3’, MF167-Rv (aproximadamente 20 pb al final del HygR) 5’- AACTCGCTGGTAGTTACCAC -3’ y MF168-Fw (aproximadamente 20 pb al inicio del HygR) 5’- ACTAGATCCGATGATAAGCTG -3’. Amplificación de los fragmentos UF y DF. Para la construcción del KO se amplifican los fragmentos UF y DF mediante PCR punto final en un termociclador MAXYGENE II (THERM-1000), siguiendo los protocolos de temperatura variable y tamaño del producto para la polimerasa Phusion High-Fidelity (New England Biolabs, Cat. E0553L). El medio de reacción contiene gDNA (aproximadamente 1 µg), 10 mM dNTP´s, 1X Phusion HF buffer, 5% (v/v) DMSO y 10 mM MgCl2. 47 Preparación de células competentes TOP10 de Escherichia coli. Para la preparación de las células competentes de E. coli se siguío el método Kushner utilizando cloruro de rubidio (RbCl) (Kushner, 1978). Se sembró una colonia de TOP10 de E. coli en 25 mL de medio LB (1.6% (p/v) bacto-triptona, 1% (p/v) extracto de levadura, 0.5% (p/v) NaCl) y se incubó durante 16 horas a 37ºC con agitación orbital constante. Las células se cosecharon a 1,690 g por 20 min a 4°C, el precipitado se resuspendió en 1 mL de LB sin antibióticos. Se inocularon 30 mL de LB con la suspensión anterior hasta una densidad óptica (D.O.600nm) de 0.1 y se crecieron por 2 horas a 37°C en agitación orbital constante hasta alcanzar una D.O.600nm de 0.6. Las células se recuperaron a 1,690 g por 30 min a temperatura ambiente. El botón se resuspendió en 20 mL de solución TFB1 frío (30 mM acetato de potasio, 50 mM MnCl2, 100 mM RbCl, 10 mM CaCl2, 15% (v/v) glicerol, pH 5.8, esterilizado por filtración), y se incubó por 2 h a 4°C. Las células se recuperan a 1,690 g por 30 min a 4°C y se resuspendieron cuidadosamente en 4 mL de solución TBF2 frío (10 mM MOPS, 75 mM CaCl2, 10 mM RbCl, 15% (v/v) glicerol, pH 6.8, estéril por filtración) y se incubarón por 1 h a 4°C. Finalmente, las células se conservaron en alícuotas de 100 µL a -70ºC hasta su uso. Formación y clonación de los plásmidos TOPO (TA cloning) para UF y DF. Ya que los amplificados UF y DF provienen de una PCR donde se utiliza la polimerasa Phusion, a éstos se les debe añadir extensiones de adenina para que sean reconocidos por el plásmido TOPO para poder integrarse. En la reacción se añadió 1 µL de Taq polimerasa, 0.5 µL dNTP´s y 5 µL del amortiguador 10x y se incubaron por 15 min a 72°C. Una vez finalizada la proyección, los fragmentos UF y DF se ligaron a un vector pCT2.1-TOPO (pCR2.1-TOPO plásmido. Invitrogen Cat. 45-0641) en presencia de 3.5 µL del PCR modificado, 1 µL de solución salina (incluida en el Topo-kit), 0.2 µL del Topo-vector y se ajusta a un volumen final de 5 48 µL con H2O, se incubaron por 5 min a temperatura ambiente y se clonaron en TOP10 E. coli. La transformación de las células TOP10 con los plásmidos UF y DF se llevó a cabo por choque térmico siguiendo el método descrito en Sambrook et al., (2001) con algunas modificaciones. Las células se incubaron por 30 min a 4°C, seguido de un choque térmico a 37°C durante 5 min; éstas se recuperan en 1 mL de LB a 37°C durante 1 hora; al término de la incubación las células se centrifugan a 13,800 g por 1 minuto y se resuspenden en 100 µL de LB. Posteriormente, se sembraron en medio LB sólido, suplementado con ampicilina (LBAmp contiene: 1.6% (p/v) bacto- triptona, 1% (p/v) extracto de levadura, 0.5% (p/v) NaCl y 0.1 mg ampicilina/mL) y se incubaron por 16 h a 37°C. Obtención del DNA plasmídico. Una colonia de la cepa TOP10 transformada se sembró en 3 mL de LBAmp y se incubó por 12 h a 37°C con agitación orbital constante. Para obtener el DNA plasmídico, 1 mL de la muestra se centrifugó a 15,700 g durante 2 min, el sobrenadante se descartó y las células fueron resuspendidas en 200 µL del amortiguador STET (100 mM NaCl, 10mM Tris-HCl pH 8.0, 1 mM EDTA, 5% (v/v) Tritón X100) y 0.1% (p/v) lisozima; se incubaron por 60 seg a 95°C y se centrifugó a 15,700 g durante 10 min. Al sobrenadante, que contenía al RNA y al DNA plasmídico, se le añadieron 20 µL de solución Mini III (3 M NaOAc) y 500 µL de 2-propanol para precipitar a los ácidos nucléicos. Las muestras se centrifugaron a 15,700 g durante 10 min, el precipitado que contenía los ácidos nucleicos se lavó con 200 µL de etanol al 70% (v/v). Finalmente, se centrifugó a 15,700 g durante 2 min y se eliminó el sobrenadante, al precipitado se le agregaron 100 µL TE-RNase para obtener una muestra libre de RNA. La concentración de DNA se determinó en un NanoDrop 2000c (Thermo Scientific). 49 Purificación del DNA. El DNA se purificó con un kit JET Quick Spin Column Technique de Genomed. Al DNA plasmídico se le adicionó 400 µL del amortiguador M1 (hidroxicloruro de guanidina e isopropanol), se mezcló por inversión y se transfirió a una columna de sílica de purificación, la cual se centrifugó a 15,700 g por 1 min y se descartó el filtrado. La columna se lavó dos veces con 400 µL del amortiguador M2 (etanol, NaCl, EDTA y Tris-HCl) y se centrifugó a 15,700 g por 1 min eliminando el filtrado. Para eluir el plásmido de la columna se le adicionó 50 µL de TE (10 mM Tris-HCl, pH 8.0) y se centrifugó a 15,700 g por 2 min. Se determinó concentración de DNA obtenido. Secuenciación del DNA. La mezcla de reacción fue compuesta por: 250 ng de la muestra, 3.2 pmoles del cebador MF963/MF964 y 10 mM Tris-HCl, pH 8.0, en un volumen final de 7 µL; la secuenciación fue realizada y registrada en Ludwig- Maximilians Universiät München, Alemania (Sequencing LMU http://www.gi.bio.lmu.de/sequencing/index_html); se verificaron las secuencias obtenidas y los sitios de corte. Construcción del plásmido knock-out. Las mutaciones se realizaron como lo describió Brachmann et al., (2004). La digestión del plásmido que contiene el cassette de resistencia a higromicina (HygR) (pUMa_198, donado por el Dr. Michael Feldbrügge del Instituto de Microbiología, Universidad Heinrich-Heine, Düesseldorf, Alemania) así como aquellos que contienen los fragmentos UF y DF fueron digeridos con la enzima de restricción SfiI. El DNA fue purificado del gel mediante un Zymoclean Gel DNA Recovery kit de Zymo Research (USA). El plásmido KO (pUMa_1704) se obtuvo a partir de la ligación de los fragmentos de DNA empleando una relación 1:1:3 de UF:DF:HygR, durante 3 h a temperatura ambiente en presencia de 200 unidades de DNA ligasa (Invitrogen). Generación de la cepa mutante (∆g) de U. maydis. Para la obtención de la cepa ∆g se realizó por medio de transformación química como se describe a continuación. 50 Formación de protoplastos. Se inoculan 3 mL de medio YPD con la cepa WT y se incubaron 24 h a 28°C en agitación orbital constante. Posteriormente, el preinóculo se diluyó hasta 0.4 unidades de D.O.600nm y se cultivó hasta alcanzar las 0.8 unidades de D.O. 600nm. Se comprobó la pureza celular por microscopía, se centrifugó un volumen de 50 mL de las células a 1,700 g durante 5 min; el paquete celular se resuspendió con 25 mL del amortiguador SCS (20 mM ácido cítrico, 1 M sorbitol, pH 5.8) y se centrifugó a 1,700 g durante 5 min. El precipitado se recuperó y resuspendió con una solución de Novoenzima (7 mg/2 mL SCS); posteriormente se incubó de 10 a 20 min a temperatura ambiente; la permeabilización celular se observó con el microscopio como la aparición de burbujas en el ápice celular; cuando el 40% de las células fueron permeabilizadas, se continuó con la transformación genética de U. maydis. Se añadió 10 mL de SCS y se centrifugó a 1,100 g durante 5 min a 4°C; el precipitado se resuspendió con 10 mL de SCS para eliminar las enzimas y nuevamente se centrifugó a 1,100 g durante 5 min; el precipitado se resuspendió en 10 mL del amortiguador STC (1 M sorbitol, 10 mM Tris-HCl, 100 nM CaCl2, pH 7.5) y se centrifugó a 1,100 g durante 5 min (este paso se repitió dos veces). Posteriormente se desechó el sobrenadante y el precipitado se resuspendió en 1 mL de STC. Finalmente, las muestras se almacenaron a -70°C en alícuotas de 100 µL. Del mismo modo se intentó realizar la transformación por biobalística para para eliminar el gel ATP21 que codifica para la la subunidad e (mutane De). No se obtuvieron resultados positivos (ver anexo 10.1). Incorporación del genoma Dg exógeno a los protoplastos de U. maydis. Para la transformación de la levadura se utilizaron cajas de Petri con YPD agar (suplementado con 18.22% (p/v) sorbitol y 50 µg higromicina B/mL de PBS). Se adicionó 10 mL de YPD agar con 80 µL de higromicina (aproximadamente 4 mg) y finalmente a los 15 min se agregó 10 mL de YPD agar (sin higromicina). A los 51 protoplastos se les añadió 15 µg de heparina (15 mg/mL) y 5 µL del plásmido KO (pUMa_1704), y se incubaron por 10 min a 4°C; posteriormente se añadieron 500 µL del amortiguador STC/PEG (60% (v/v) STC-amortiguador, 40% (p/v) polietilenglicol, Sigma P-3640) y se incubaron por 15 min a 4°C. Finalmente, las células transformadas se dispersaron muy suavemente en la placa y se incubaron por 7 días a 28°C. Comprobación por PCR de la cepa Dg (eliminación del gen ATP20). Se sembró una azada de la cepa transformada en medio CM-Hyg agar (0.25% (p/v) casaminoácidos, 0.1% (p/v) extracto de levadura, 1% solución Holliday de vitaminas, 6.25% solución Holliday de sales, 0.05% DNA (sigma D-3159), 0.15% (p/v) NH4NO3, 18.22% (p/v) sorbitol, 1.5% (p/v) bacto-agar, 5 µg/mL Hyg, pH 7.0) en forma de estría para obtener cepas independientes y se incubó por 24 h a 28°C. Una vez multiplicadas las células, se tomó una azada de cada una de las colonias (aproximadamente 10) y fueron sembradas en medio YPD líquido e incubadas por 24 h a 28°C en agitación orbital constante; posteriormente se tomaron 5 µL del cultivo y se sembraron en una caja con CM-Agar para mantener la línea celular, el cultivo restante se empleó para aislar gDNA (ver arriba el procedimiento descrito para aislar el gDNA de la cepa WT). Para comprobar que el gen ha sido sustituido por el HygR se realizó una PCR con el gDNA ∆g de las colonias aisladas en el paso anterior, utilizando los oligonucleótidos P1 y P2; la migración del producto de PCR se realizó en un gel de agarosa al 0.8% (p/v), 1x-TBE, 25 µg/mL BrEt. Para comprobar que el gen ATP20 ya no estaba presente se incluyó como control negativo el gDNA de la cepa WT, las clonas que no amplificaron fueron usadas como positivas para una segunda PCR donde se comprobó la presencia del cassette HygR. 52 Para comprobar que el cassette se insertó en el lugar correcto después de la recombinación homóloga, se realizó una PCR utilizando un par oligonucleótidos que flanquearon las regiones UF (U1 y el MF167) y DF (MF168 y D3), usando como control negativo el gDNA de la cepa WT. Las muestras se resolvieron en un gel al 0.8% de agarosa, 1x-TBE, 25 µg/mL BrEt/mL. Southern Blot. Para llevar a cabo las réplicas tipo Southern se digirió el gDNA con una enzima de restricción, cuidando de que la enzima reconociera un sitio de corte dentro del cassette HygR sin cortar en la secuencia del gen de interés. En este caso la reacción contenía 10 unidades de Nco-I, 1 µg DNA, 1X NEBuffer, 39 µL dH2O y se incubó por 1 h a 37°C. Posteriormente, las muestras se corrieron en un gel de agarosa al 0.8% (p/v) con 25 µg BrEt/mL en 1x de TAE (40 mM Tris-Base, 40 mM ácido acético, 1 mM EDTA pH 8.0), a 90V/2 h. Al finalizar la corrida, el gel se lavó con 250 mM HCl durante 20 min, posteriormente se lavó 2 veces con dH2O por 5 min y se incubó en la solución DENAT (1.5 M NaCl, 0.4 M NaOH) por 20 min; se lavó con dH2O y se incubó con la solución RENAT (1.5 M NaCl, 282 mM Tris- HCl, 218 mM Tris-Base) por 20 min. El gel se colocó sobre papel filtro Whatman en una base plana, y se humedeció con una solución 20x de SSC (3 M NaCl, 0.3 M citrato de sodio) cuidando que el papel siempre estuviera en contacto con la solución. Se colocó la membrana Hybond-N+Nylon (GE Healthcare) sobre el gel y se sellaron los bordes con una película plástica flexible (parafilm). Se colocaron papeles Whatman sobre la membrana, luego una pila de 5 cm de altura de papel absorbente y encima una placa de aproximadamente 500 gr para producir presión sobre el gel y la membrana para permitir que el DNA se movilizara del gel a la membrana por capilaridad. Al finalizar la transferencia después de 48 h, la membrana se humectó con 2x de SSC y se dejó secar por 15 min; el DNA se fijó a la membrana irradiando con UV- crosslinker directamente, cuidando siempre de no exponer la cara de la membrana que estuvo en contacto directo con el gel para no dañar el DNA. 53 Se prehibridizó la membrana en un tubo para Southern blot con el amortiguador de hibridación 26% (v/v) SSPE (3 M NaCl, 200 mM fosfato de sodio, 20 mM EDTA, pH 7.4), 20% (v/v) solución Denhardt (1% (v/v) ficoll, 1% (p/v) polivinilpirrolidona, 1% (p/v) BSA, 5% (p/v) SDS) durante 15 min a 65°C (Southern, 1975; Rybicki et al., 1996). Para la detección del fragmento de interés, la membrana se incubó toda la noche con una sonda de DNA complementaria al sector de DNA buscado y marcada con digonexina (Feinberg et al., 1983); el marcaje, la hibridación y el revelado se realizaron siguiendo las instrucciones del fabricante (New England Biolabs). Posteriormente, se retiró la sonda y la membrana se lavó con diferentes concentraciones de la solución SSPE. Se realizó un primer lavado con 10 mL del amortiguador DIG1 (100 mM ácido málico, 150 mM NaCl, pH 7.5) por 5 minutos; y el segundo con 10 mL del amortiguador DIG2 (DIG1, 10% leche en polvo desgrasada) durante 30 min para bloquear la membrana, posteriormente se incubó en 10 mL de DIG2 que contenía el anticuerpo anti-DIG-AB en una relación 1:10,000, por 3 h a 60°C. La membrana se lavó dos veces por 15 min con el amortiguador DIG-wash (DIG1, 0.3% (v/v) Tween-20); se incubó con 20 mL del amortiguador DIG3 (100 mM Tris-HCl, 100 mM NaCl), y finalmente se incubó en presencia de la solución CDP- Star® Reagent de NEB labs (CDP-Star 1:100 en DIG3) por 5 min. Se retiró la solución y se detectó la señal de quimioluminiscencia con una cámara ImageQuant LAS4000 de GE Healthcare. Curva de crecimiento. Las cepas WT y Dg de U. maydis se sembraron en un medio sólido de YPD y se cultivaron por 48 h a 28°C. A partir de este cultivo se preparó un inóculo en 100 mL de YPD y se incubó durante 24 h a 28°C, en agitación orbital constante. Las células se lavaron y se resuspendieron en 5 mL de agua destilada estéril. Se inoculó el equivalente a 40 unidades de absorbancia (600 nm) de esta suspensión a un litro de YPD, Medio Mínimo-Etanol (0.3% (p/v) sulfato de 54 amonio, 1% (v/v) etanol absoluto, 1x solución de sales (ver anexo 10.2), 1x elementos traza (ver anexo 10.3) y Medio Mínimo-Glucosa (0.3% (p/v) sulfato de amonio, 1% (p/v) glucosa, 1x solución de sales, 1x elementos traza). Las levaduras fueron cultivadas a 28°C, en agitación orbital constante y se tomaron alícuotas de 1 mL cada 2 h para determinar la D.O.600nm, el pH y la concentración de glucosa. Curva de crecimiento en placa (diluciones seriadas). Las cepas WT y Dg de U. maydis fueron sembradas en 100 mL YPD y cultivadas por 24 h a 28°C. Se tomó 1 mL de U. maydis, se centrifugó a 12,000 g por 5 min y se descartó el sobrenadante. Las células fueron resuspendidas en 1 mL de agua destilada estéril, se tomaron 10 mg (peso húmedo) de las mismas y se realizaron 5 diluciones seriadas (1/10). Determinación cuantitativa de glucosa (método enzimático colorimétrico para la cuantificación de glucosa en medios de levadura). La concentración de glucosa se determinó mediante un “Glucose-LQ kit”, siguiendo las instrucciones del fabricante. La glucosa oxidasa (GOD) cataliza la oxidación de glucosa a ácido glucónico. El peróxido de hidrógeno (H2O2) producido se detecta mediante un aceptor cromogénico de oxígeno, fenol, 4–aminofenazona (4-AF), en presencia de la peroxidasa (POD): La intensidad del color formado es proporcional a la concentración de glucosa presente en la muestra ensayada. Para la determinación se tomó 1 mL de células y se centrifugaron a 16,100 g por 5 min. Se recuperaron 500 µl del sobrenadante y se continuó con la técnica como la reportó el fabricante, realizando las lecturas a 505 nm. 55 Microscopía de transmisión electrónica. La microscopía electrónica se realizó en el Instituto de Biotecnología de la UNAM, mediante la técnica LR White, usando 2.5% (v/v) glutaraldehído (GTA) y 4% (v/v) paraformaldehído (PFA) como fijador, las muestras se incubaron en 160 mM de cacodilato de Na/K, pH 7.4 por 1 hora a temperatura ambiente, una vez transcurrido el tiempo se lavaron dos veces con el amortiguador de cacodilato y se incubaron toda la noche a 4ºC. Las muestras fueron deshidratadas gradualmente en etanol al 70, 80 y 96 %. Se usó una mezcla 50:50 de LR White-EtOH absoluto y se incubó 1 hora. Se realizaron cortes ultrafinos (60-80 nm) en un ultramicrotomo (UltraCut-R, Leica), los cortes se recogieron en rejillas Formvar® recubiertas de cobre (EMS) y se tiñeron con 2% (p/v) de acetato de uranilo y 0.3% (p/v) de citrato de plomo. Las muestras se analizaron en un microscopio de transmisión electrónica (ZEISS Libra 120 Oberkochen Alemania) equipado con un dispositivo de carga acoplada (CCD) (300 W de doble visión; Gatan, Pleasanton, CA, EE. UU.) a la cámara. Las imágenes fueron capturadas a una tensión de aceleración de 80 kV. Consumo celular de oxígeno. Se realizó la determinación del consumo de oxígeno en las células WT y Dg cosechadas en la fase logarítmica de crecimiento (8 hrs) y en la fase estacionaria (24 hrs); las células se recuperaron de 50 mL de cada cultivo por centrifugación a 2,300 g por 5 min, y se resuspendieron en 3 mL de 10 mM K2PO4, pH 7.4. La concentración de oxígeno se determinó polarograficamente usando un electrodo tipo Clark acoplado a un oxímetro YSI- 5300 y una cámara de incubación manteniendo la temperatura constante a 30°C (Estambrook, 1967, Guerrero-Castillo et al., 2012, Affourtit et al., 2012). La respiración se inhibió con 5 mM de KCN (inhibidor del complejo IV) y/o 10 mM de n- octil-galato (nOg - inhibidor de la oxidasa alterna) (Juárez et al., 2006). Permeabilización de las células para ensayos bioenergéticos. Se centrifugaron 50 mL de células (fase estacionaria), las cuales fueron resuspendidas en 3 mL del amortiguador de permeabilización (300 mM sorbitol, 10 mM HEPES, 1 56 mM EGTA, 10 mM KH2PO4, 10 mM MgSO4, 150 mM KCl, pH 7.4) y se incubaron por 20 min en presencia de 0.02% (p/v) de digitonina. Se verificó en el oxímetro la permeabilización de las células y se estimuló la respiración adicionando 10 mM de succinato (Esparza-Perusquía et al., en preparación). La respiración se inhibió con KCN o nOg. Monitoreo del potencial de membrana mitocondrial (DYm). Para determinar el potencial de membrana mitocondrial se utilizó 10 µM safranina O (Akerman et al., 1976), el cual es un colorante lipofílico catiónico que se transporta (internaliza) a la matriz mitocondrial cuando la membrana interna está energizada, aumentando su absorbancia. Cuando se abate el DYm, el colorante sale de la matriz y la señal disminuye. Estos cambios se determinan en un espectrofotómetro de doble longitud de onda con agitación orbital constante, en modo dual (AMINCO DW 2000, Olis, Inc., Bogart, GA, EUA) a 511 nm, utilizando de referencia el punto isosbéstico 533 nm, para determinar el cambio en la absorción de la safranina O. Para la cuantificación del potencial de membrana mitocondrial se utilizaron células permeabilizadas (ver permeabilización de las células para ensayos bioenergéticos). La reacción se llevó a cabo en un amortiguador que contenía 300 mM sorbitol, 10 mM HEPES, 1 mM EGTA, 10 mM MgSO4, 150 mM KCl y 10 mM KH2PO4 a pH 7.4. El potencial de membrana se generó mediante la adición de 10 mM de succinato en presencia de 10 µM safranina O. El DYm se abatió con 5 µM de CCCP. Síntesis de ATP mitocondrial. Para la cuantificación de la producción de ATP mitocondrial se utilizaron las células permeabilizadas con digitonina (0.02% p/v) y se incubaron con agitación con 10 µM de P1, P5-Di(adenosine-5´)pentaphosphate penta-ammonium (Inhibidor de la miocinasa de ADP) por 5 minutos en el amortiguador de permeabilización. Las mitocondrias se energizaron con 10 mM succinato. 57 Una vez transcurrida la incubación se tomó una muestra para determinar el T0; posteriormente, se les adicionó 5 mM de ADP y se tomaron muestras diferentes tiempos (10, 20, 30, 40, 50 y 60 seg). Para detener la reacción se adicionó ácido perclórico (0.7 M) y se centrifugaron a 16,100 g por 1 min. Se recuperaron 900 µL del sobrenadante y se amortiguaron con 0.5 M KOH para neutralizar al ácido perclórico. El KClO4 es una sal que se precipita, la que puede ser removida al centrifugar a 16,100 g por 1 min. Se recuperaron nuevamente 900 µL del sobrenadante y se les ajustó el pH a 7.0 La concentración de ATP sintetizado se determinó por medio de un sistema enzimático acoplado a la reducción de NADP+. La mezcla de reacción contenía 0.5 mM NADP+, 5 mM ADP, 6 unidades/mL de la glucosa-6-fosfato deshidrogenasa (G- 6P-DH), 16 unidades/mL de hexocinasa (HK) y 10 mM de glucosa. En este proceso enzimático la hexocinasa produce glucosa-6-fosfato a partir de glucosa y ATP, la glucosa-6-fosfato deshidrogenasa oxida a la glucosa-6-fosfato y genera d- gluconolactona y NADPH. Los resultados se analizaron determinando el cambio de absorbancia tomando en cuenta el coeficiente de extinción molar del NADP+ (eNADPH = 6.22 mM- 1cm-1) en función del tiempo, para obtener la velocidad de síntesis de ATP (Trautschold et al., 1995; De los Ríos-Castillo et al., 2011). El ajuste de los datos para determinar los parámetros cinéticos se realizó en el programa Sigma Plot versión 10.0. Aislamiento de las mitocondrias. Las mitocondrias de U. maydis fueron aisladas como lo describió Juárez et al., (2004) con algunas modificaciones (Guérin et al., 1979; Díaz-Ruiz, et al., 2008, Esparza-Perusquía, et al., 2017). Las células fueron cosechadas por centrifugación a 5,500 g durante 5 min a 4°C; se resuspendieron y se lavaron dos veces con agua destilada. Posteriormente las levaduras se resuspendieron en 0.6 M de sulfato de amonio, 20 mM KH2PO4, en una relación de 100 mL por cada 8 g de células; para degradar la pared celular se 58 adicionaron las enzimas líticas de Trichoderma harzianum (0.016 g/g de peso húmedo de células) (Sigma-Aldrich L1412) y se incubaron durante 60 min a 30°C. Para eliminar las enzimas líticas, los protoplastos se centrifugaron a 5,500 g durante 10 min a 4°C; se resuspendieron y lavaron con una solución compuesta por 0.8 M sacarosa, 10 mM Tris, 2 mM EDTA, 20 mM KH2PO4, 0.3% (p/v) albúmina sérica de bovino desgrasada, pH 7.0 y se recuperaron por centrifugación a 5,500 g durante 10 min a 4°C. Los protoplastos se resuspendieron en 40 mL de una solución de 0.4 M sacarosa, 10 mM Tris, 2 mM EDTA, 20 mM KH2PO4, 0.3% (p/v) albúmina sérica de bovino desgrasada, pH 7.0 y se homogenizaron con 15 ± 30 golpes con un homogeneizador Potter, en presencia de 1 mM PMSF y coctel de inhibidores de proteasas (Sigma-Aldrich P-8340). Al finalizar, el homogenado se diluyó a un volumen final de 130 mL y se centrifugó a 5,500 g durante 10 min a 4°C. Para recuperar a las mitocondrias, el sobrenadante se centrifugó a 17,300 g durante 10 min a 4°C; finalmente las mitocondrias se resuspendieron en 0.4 M sacarosa, 10 mM Tris, 2 mM EDTA, 20 mM KH2PO4, 0.3% (p/v) albúmina desgrasada, pH 7.0, con el menor volumen posible. Determinación de la concentración de proteína. La concentración de concentración de proteína se determinó por medio del método de Lowry (Lowry, 1951) con algunas modificaciones (Mahuran et al., 1983). Las muestras fueron tratadas con 0.4% (p/v) desoxicolato de sodio (DOC) y se continuó con el método tradicional. La absorbancia de las muestras se determinó a 660 nm usando albúmina sérica de bovino (BSA) como estándar. Para conocer la concentración de proteína en mg/mL, se utilizó el método de regresión lineal o mínimos cuadrados (Waterborg et al., 1984). Producción mitocondrial de peróxido de hidrogeno. La cuantificación de la producción de H2O2 se realizó con el kit Amplex® Red (Invitrogen, Molecular Probes, EE. UU.), siguiendo las instrucciones del fabricante. Se utilizaron 30 µg de 59 proteína mitocondrial en presencia de 600 µM de 2, 6-Dimethoxy-1, 4-benzoquinona (DBQ), 10 mM de Citocromo c de corazón de caballo (SIGMA C2037) y 3 mM NADH. La reacción se inició 10 min después de haber añadido el NADH y se incubó por 30 min. La fluorescencia se determinó en un lector Varioskan LUX (Thermo Scientific) a 530-590nm. Solubilización de los supercomplejos y complejos mitocondriales. Para solubilizar las membranas mitocondriales liberando así los complejos y supercomplejos respiratorios embebidos en la membrana interna, las mitocondrias se incubaron en presencia de diferentes concentraciones del detergente digitonina (0.5, 1, 2, 3, 5 g/g proteína), en una solución de 50 mM Bis-Tris, 500 mM de ácido aminocaproico, pH 7.0, 10 mM succinato y 10 mM de ATP; el detergente (solucion stock 500 mg/mL en DMSO) se añadió mientras la solución se agita lentamente. La solución se incubó durante 30 min a 4°C, posteriormente se ultracentrifugó a 100,000 g durante 35 min a 4°C. Los complejos y supercomplejos respiratorios se recuperaron en el sobrenadante (Schägger et al., 1991; Esparza-Perusquía et al., en preparación). Electroforesis en condiciones nativas (BN-PAGE y hrCN-PAGE). Los complejos respiratorios de las mitocondrias de U. maydis se resolvieron por medio de la electroforesis azul en condiciones nativas (BN-PAGE) y la electroforesis clara en condiciones nativas (hrCN-PAGE) en geles en gradiente de poliacrilamida (4 – 10%) (Schägger et al., 1991; Wittig et al., 2006). Para la BN-PAGE el amortiguador del ánodo contenía 50 mM Bis-Tris/HCl, pH 7.0; el amortiguador del cátodo contenía 50 mM tricina, 15 mM Bis-Tris, pH 7.0 y el colorante aniónico Coomassie® G250 (0.02%). Para la hrCN-PAGE el amortiguador del ánodo contenía 25 mM imidazol- HCl, pH 7.0; el amortiguador del cátodo contenía 50 mM tricina, 7.5 mM imidazol, pH 7.0, 0.05% (p/v) desoxicolato de sodio (DOC) y 0.01% (p/v) n-dodecil β-D- maltósido (DDM) (Wittig et al., 2007). Se utilizó como marcador del frente de corrida al colorante rojo de Ponceau. Se aplicaron 100 µg de proteína por carril y las condiciones de la corrida electroforética fueron 30 V durante 16 horas a 4°C. 60 El peso molecular de los complejos respiratorios y supercomplejos fue determinado a partir de su movilidad electroforética y el revelado de su actividad catalítica en gel, utilizando como estándares de peso molecular los complejos mitocondriales de corazón de bovino solubilizados con digitonina en las mismas condiciones que los complejos de U. maydis. Actividad en gel de los complejos y supercomplejos respiratorios. La actividad de los complejos respiratorios se realizó como los reportó Wittig et al., (2007). *Actividad de NADH deshidrogenasa. Al concluir la electroforesis nativa, los geles se incubaron en una solución de 10 mM tris-HCl pH 7.4, 0.5 mM NADH como sustrato y 1.2 mM bromuro de 3-[4,5-dimetiltiazol-2-yi]-2,5-difeniltetrazolio (MTT) como agente oxidante. En esta reacción el NADH es oxidado por la flavina unida a las deshidrogenasas y ésta última reduce directamente al tetrazolio. Mediante esta tinción es posible detectar la actividad del complejo I y de las deshidrogenasas alternas. La presencia de actividad de observó como un depósito violeta y la reacción se detuvo con una solución de 40% de metanol y 10% de ácido acético (Wittig et al., 2007). *Actividad de succinato deshidrogenasa. Los geles se incubaron en una solución de 10 mM KH2PO4, pH 7.4, 10 mM succinato, 0.19 mM 5-metilfenazinium metil sulfato (PMS), 5 mM MTT y 4.5 mM EDTA. La presencia de actividad se observó como un precipitado violeta y se detuvo la reacción con una solución de 40% de metanol y 10% de ácido acético (Wittig et al., 2007). *Actividad de citocromo c oxidasa. La actividad en gel del complejo IV se determinó utilizando 3, 3’-diaminobenzidine (DAB) como agente reductor y citocromo c como donador de electrones (Zerbetto et al., 1997). El gel se incubó en una solución de 10 mM KH2PO4 pH 7.4, 4.66 mM DAB, 4 mg citocromo c y 1.8 KU catalasa, ésta última se añade para evitar que peróxidos contaminantes oxiden a la diaminobencidina, evitando así que la reacción sea inespecífica. La presencia de 61 actividad se observó como un depósito marrón. La reacción se detuvo con una solución de 40% (v/v) metanol y 10% (v/v) ácido acético (Wittig et al., 2007). *Actividad de ATPasa. La ubicación en el gel de los oligómeros del complejo V fue determinada por medio de su actividad de hidrólisis de ATP. El gel se incubó en una solución de 50 mM de glicina, 10 mM MgCl2, 0.2% (p/v) Pb(NO3)2, 5 mM ATP, pH 8.0; ajustando el pH con trietanolamina para evitar la precipitación inespecífica del plomo. La presencia de actividad de hidrólisis de ATP se observó como un depósito blanco del fosfato de plomo, sobre un fondo oscuro (Jung et al., 2000). La reacción se detuvo con una solución de 40% (v/v) metanol y 10% (v/v) ácido acético (Wittig et al., 2007). Purificación del dímero y del monómero de la F1FO-ATP sintasa de U. maydis. Los supercomplejos y complejos respiratorios solubilizados con digitonina (2:1 digitonina/proteína) se aislaron por medio de un gradiente continuo de sacarosa de 0.5 - 1.5 M, en presencia de 20 mM KCl, 15 mM tris-base, pH 7.4 y 0.2% (p/v) digitonina; se añadieron aproximadamente 48 mg de proteína (3.5 mL del solubilizado) a 24 mL del gradiente y se centrifugaron a 131,000 g durante 16 h a 4°C. Se colectaron fracciones de 500 µL; la densidad de cada fracción se determinó por medio de un densitómetro ATAGO N1 (Brix 0~32%). La presencia de los supercomplejos se determinó por medio de una electroforesis nativa (hrCN-PAGE). Las fracciones de interés se equilibraron con 30 mM HEPES, 5% (v/v) glicerol, pH 8.0 y se concentraron en filtros Amicon de 10 K (Esparza-Perusquía et. al., 2017). Ensayo enzimático de la actividad de ATPasa. La actividad de ATPasa del complejo V se ensayó en presencia de un sistema regenerador de ATP acoplado a la oxidación del NADH; el medio de reacción contenía 30 mM HEPES, pH 8.0, 5 mM fosfoenolpiruvato (PEP), 1 mM Mg2+ (MgSO4) libre, 50 unidades de la piruvato cinasa (PK), 30 unidades de la lactato deshidrogenasa (LDH), 90 mM KCl, 0.01% (p/v) DDM, 150 µM NADH y 100 µg de muestra; se empleó una solución equimolar de Mg-ATP. 62 La reacción de hidrolisis de ATP se inició agregando el ATP a una celda de cuarzo la cual contenía todo el sistema regenerador y al V1 o V2 del complejo V. Se realizaron lecturas cada 0.5 seg por 15 min a 340 nm. Los resultados se analizaron tomando en cuenta el coeficiente de extinción del NADH (e = 6.22 mM-1cm-1) y la región lineal del trazo espectrofotométrico (Penefsky et al., 1960; Andrianaivomananjaona et al., 2011; Esparza-Perusquía et al., 2017). Sensibilidad a la oligomicina. En los ensayos enzimáticos también se determinó la sensibilidad a la oligomicina de la F1FO-ATP sintasa. Debido a que la oligomicina es un inhibidor lento pero fuertemente unido, los oligómeros de la F1FO- ATP sintasa fueron incubados en el medio de reacción durante 5 min en presencia de diferentes concentraciones del inhibidor (0.1, 0.25, 0.5, 1, 5, 10, 20 µg/mg proteína). Al término de este tiempo la reacción fue iniciada como se describió arriba (Esparza-Perusquía et al., 2017). Estabilidad térmica. Para determinar la termoestabilidad del V1 y V2 la F1FO- ATP sintasa se incubó la enzima por 15 min a diferentes temperaturas, al término de este tiempo la reacción de hidrólisis de Mg-ATP se realizó a 25°C como se describió en el apartado ensayo enzimático (Esparza-Perusquía et al., 2017). Análisis Estadísticos. Los datos de la actividad enzimática se procesaron con un análisis de regresión no lineal, robusto, ponderado, usando el programa Systat Software, Inc., versión 10.0, a partir de un promedio de 3 réplicas de 5 a 7 preparaciones independientes. Las imágenes derivadas de la tinción de las proteínas con azul de Coomassie® R250 de los geles de Tricina-SDS-PAGE para la cuantificación de la F1FO-ATP sintasa fueron escaneadas y la intensidad de las bandas se determinó utilizando el programa My Image Analysis 2.0 (ThermoScientific). El ajuste de todos los datos cinéticos se realizó con el programa Sigma Plot versión 10.0. 63 “Mitochondrial Membranes I” Odra Noel Resultados “Que Dios la lleve y la virgen la acompañe” Dr. Antonio Peña 64 5. RESULTADOS La construcción de los plásmidos y la transformación para generar la cepa mutante Dg se realizó en el Instituto de Microbiología de la Universidad Heinrich- Heine en Düsseldorf, Alemania, bajo la asesoría del Dr. Michael Feldbrügge. Construcción y clonación de los plásmidos UF y DF. Se amplificaron los fragmentos UF (1.06 kpb) y DF (1.36 kpb), flanqueando al gen ATP20, el cual codifica para la subunidad g, mediante un PCR punto final acoplado a la polimerasa Phusion (PHU), tomando como molde al gDNA de la cepa WT (Figura 22A). Dichas secuencias fueron insertadas en el plásmido TOPO (pCR2.1-TOPO vector. Invitrogen Cat. 45-0641) usando un Topo TA Cloning Kit. Las células competentes TOP10 de E. coli fueron transformadas con los plásmidos UF (pUMa_1706) y DF (pUMa_1705) para obtener múltiples copias idénticas y aislar el DNA. Una vez obtenido el DNA plasmídico se realizó una digestión controlada mediante enzimas de restricción, donde se verificó el estado del plásmido y la orientación de la secuencia; finalmente, el DNA se secuenció para comprobar que no tenía mutaciones inespecíficas y posteriormente se continuó con el proceso de construción del plásmido knock-out (KO). Construcción del plásmido Knock-out (KO). La mutación se realizó como lo describió Brachmann et al., (2004), digiriendo los plásmidos UF y DF con 20 unidades de las enzimas de restricción SfiI y ScaI-HF, los productos de la restricción para UF y DF que nos interesan corresponden a 3.1 y 3.0 kpb, los sitios para cortar el pUMa_198 y liberar completamente a HygR son reconocidos por la enzima SfiI liberando un fragmento de 1.8 kpb (Figura 22B). Al finalizar la digestión, las bandas correspondientes a UF, DF e Hyg se purificaron del gel. Los fragmentos obtenidos se colocaron en una reacción de ligación manteniendo la relación 1:1:3 (UF:DF:HygR). Al finalizar la ligación se llevó a cabo la transformación y clonación en E.coli. 65 Figura 22. Amplificación de los fragmentos UF y DF para la construcción del plásmido KO. (A) (A) Los amplificados UF (1.06 kpb) y DF (1.36 kpb) se resolvieron en un gel de agarosa al 1% en presencia de bromuro de etidio. El control negativo (*) no contiene oligonucleótidos. (B) Los plásmidos fueron digeridos con las enzimas SfiI-ScaI-HF. Se obtuvieron los fragmentos de 3.1 y 3.0 kpb para UF y DF respectivamente y 1.88 kpb para el HygR. Verificación del plásmido Knock-out (KO). Para verificar la correcta ligación en el plásmido KO, el DNA bacteriano fue extraído y digerido con la enzima SfiI, la cual produce dos bandas que corresponden al HygR con un tamaño de 1.88 kpb y al resto del plásmido con un tamaño de 6.37 kpb (Figura 23A). Los plásmidos se analizaron en la unidad de Secuenciación de DNA Ludwig- Maximilians Universiät München, Alemania (Sequencing LMU http://www.gi.bio.lmu.de/sequencing) utilizando los oligonucleótidos MF167 y MF168, los cuales amplifican parte del HygR y los oligonucleótidos universales M13R y M13F, que reconocen una parte de la secuencia del vector TOPO; se obtuvieron aproximadamente 1000 pb de cada uno sin mutaciones inespecíficas. Transformación química de Ustilago maydis. Una vez que se secuenció el plásmido KO para verificar su correcta construcción, se linearizó mediante una digestión controlada con la enzima SwaI. Este plásmido lineal es usado para la transformación química de U. maydis como DNA exógeno (Figura 23B). La transformación significa que el DNA digerido se incorporó a los protoplastos y se llevó a cabo la recombinación homóloga, donde el DNA endógeno se intercambia por el DNA exógeno, conservando las regiones UF y DF del DNA original. 66 Cabe mencionar que U. maydis requiere que los fragmentos UF y DF sean de 1000 pb aproximadamente para que se lleve a cabo el reconocimiento eficiente de las secuencias durante la recombinación homóloga. Para comprobar la transformación de los protoplastos, estos son esparcidos en una caja de YPD agar suplementado y se incubaron a 30ºC por 7 días hasta observar el crecimiento de la cepa (Figura 23C). Figura 23. Transformación de U. maydis con el DNA exógeno. (A) El plásmido se digirió con SfiI y se obtuvieron dos fragmentos, uno de 1.88 kpb correspondiente al HygR y otro de 6.4 kpb correspondiente al resto del plásmido. (B) El plásmido fue digerido con SwaI obteniendo el fragmento UF-Hyg-DF (3.98 kpb). P= plásmido no linealizado y SwaI= plásmido digerido. (C) Las células transformadas después de 7 días de crecimiento sembradas en medio YPD agar suplementado con higromicina para la selección de la cepa mutante. El crecimiento sugiere la incorporación del gen de resistencia a higromicina y la pérdida de la subunidad g. Comprobación de la cepa mutante Dg. Se tomaron diferentes colonias de las cepas transformadas, se resembraron en medio YPD a 28°C por 48 horas. Se aisló el gDNA de las cepas seleccionadas para comprobar por PCR que el gen ATP20 fue sustituido por el HygR usando los oligonucleótidos P1 y P2 y como control negativo el gDNA de la cepa WT. Si las bandas tienen el mismo tamaño que derivado de la cepa WT, se descartan, ya que el oligonucleótido P2 está hecho diseñado directamente sobre una secuencia del gen, lo que demuestra que en esas muestras el gen no fue sustituido por el HygR. Los productos de PCR separaron en un gel al 0.8% agarosa, 1x-TBE, BrEt (Figura 24A). 67 Figura 24. Comprobación de la cepa mutante Dg. (A) Las muestras 1 y 3 presentaron diferente patrón electroforético a la cepa WT, lo que se comprueba que el gen ATP20 fue sustituido por el gen HygR. (B) Se comprobó por PCR la integridad de los flancos UF y DF, ambos unidos al gen HygR. Se utilizó como control negativo el gDNA WT. En la figura se muestran dos amplificados de 1.5 y 1.7 kpb para UF y DF respectivamente de las cepas positivas 1 y 3. Las muestras 1 y 3 presentaron un patrón electroforético diferente al del gDNA WT, lo que sugiere la incorporación del gen HygR. Dichas muestras se usaron para realizar otro ciclo de PCR con los oligonucleótidos U1 y D3 (ambos cuentan con aproximadamente 20 pb dentro de gDNA pero fuera de las construcciones UF y DF, respectivamente), así como los oligonucleótidos MF167 y MF168, los cuales se alinean 20 pb al inicio y 20 pb al final del gen HygR, respectivamente. Se observó un amplificado de aproximadamente 1.5 kpb para el UF y 1.7 kpb para el DF. El DNA de las muestras se resolvió en un gel de agarosa al 8% (Figura 24B), el gDNA mutante fue secuenciado para demostrar que fue sustituido por HygR y que no había otras mutaciones inespecíficas. Southern Blot. Para el análisis por Southern Blot, se realizó un PCR con dNTP´s marcados con digoxigenina, la cual permite su detección por medio de anticuerpos específicos marcados con fosfatasa alcalina y detectados con CDP- Star. Se identificaron bandas correspondientes a 8.3 kpb para la cepa WT, 5.07 y 4.1 kpb para la cepa mutante Dg (Figura 25). Se obtuvieron 5 muestras positivas para la mutante, las cuales fueron cultivadas nuevamente en medio YPD y conservadas en glicerol esteril (25%) a -70°C. Finalmente después de 24 horas se sembraron en un medio sólido y se cultivaron a 28°C durante 24 horas. 68 Figura 25. Southern Blot. Detección de los fragmentos de 5.07 y 4.1 kpb para la cepa mutante y se utilizó como control negativo el gDNA WT observando una banda de 8.3 kpb. En el primer carril se encuentra la cepa WT y 5 clonas positivas para la Dg. Curva de crecimiento. Se analizó la curva de crecimiento de la cepa WT y Dg a partir del inóculo de 40 U D.O. de células/L de medio de cultivo y se monitoreó su crecimiento cada 2 horas en medio liquido YPD, medio mínimo con etanol (MM- EtOH) y medio mínimo con glucosa (MM-glucosa). - Crecimiento en medio YPD. Se observó que durante la fase logarítmica de crecimiento (8 horas) las cepas WT y Dg presentaron un comportamiento similar; se tomaron lecturas a 600 nm cada 2 horas y se calculó el tiempo de duplicación celular, observando para la WT un valor de 2 horas con 7 min y para la Dg de 2 horas con 45 minutos. Sin embargo, a partir de las 20 horas podemos ver un desfasamiento en el crecimiento de la cepa Dg, obteniendo una absorbancia para la cepa WT de 1.81 D.O. (Figura 26, círculos llenos) y de 1.31 para la cepa Dg (Figura 26, círculos vacíos). Las células se fijaron con formaldehído y se observaron al microscopio para verificar su estructura. El tamaño de la WT en la fase logarítmica fue de 18.8 ± 1.8 µm mientras que la cepa Dg fue más grande (28.1 ± 6.5 µm) (Figuras 29A y 29B); en la fase estacionaria fue de 22.6 ± 3.9 y 30.0 ± 5.9 µm 69 respectivamente (Figuras 30A y 30B). Otro parámetro útil para la caracterización de las cepas mutantes es determinar su capacidad de acidificar el medio. En este caso, ambas cepas producen un cambio en el pH de 6.63 (fase logarítmica) a 6.0 (fase estacionaria) (Figura 26). Sin embargo, la cepa Dg (Figura 65, triángulos vacíos) inicia la acidificación del medio inmediatamente, mientras que la WT (Figura 26, triángulos llenos) muestra un periodo de lenta acidificación (0 - 5 hrs). Esto sugiere que la cepa Dg es más grande y es capaz de acidificar el medio ligeramente más rápido que la WT. Aunque no es clara la relación entre el tamaño de la célula con la velocidad de acidificación del medio, ambos fenómenos pueden estar relacionados con una posible aceleración del catabolismo. Figura 26. Curva de crecimiento en YPD. Las cepas WT (●) y Dg (○) se cultivaron en un medio de YPD y se determinó su crecimiento a 600nm. Se observó un comportamiento similar durante la fase logarítmica y un ligero decremento del crecimiento para la Dg en la fase estacionaria. También se determinó el pH durante las diferentes fases del crecimiento para las cepas WT (▲) y Dg (Δ), donde se observó una acidificación del medio de 6.63 a 6.0 respectivamente. Los datos son el promedio de tres repeticiones de cuatro preparaciones independientes. Las barras de error representan S.D. 70 Para tener una aproximación al efecto de la mutante sobre su adaptación metabólica, se decidió estudiar el efecto de la fuente de carbono sobre el crecimiento. -Crecimiento en MM-EtOH. Se observó que el crecimiento de ambas cepas durante la fase logarítmica es similar (Figura 27); también se observó que el crecimiento en este medio es más lento para ambas cepas siendo el tiempo de duplicación celular de la WT de 2 horas 43 minutos (Figura 27, círculos llenos) y para la Dg de 3 horas 18 minutos (Figura 27, círculos vacíos). Las células se fijaron con formaldehído y se observaron al microscopio para verificar su estructura. Figura 27. Curva de crecimiento en MM-EtOH. Las cepas WT (●) y Dg (○) se cultivaron en un medio mínimo con 1% de etanol y se determinó su crecimiento a 600 nm. Se observó un comportamiento similar durante el crecimiento en ambas cepas y se determinó el pH de las cepas WT (▲) y Dg (Δ) en los tiempos señalados. Los datos son el promedio de tres repeticiones de cuatro preparaciones independientes. Las barras de error representan S.D. El tamaño de la cepa WT fue de 19.9 ± 1.9 µm y para la cepa Dg de 29.3 ± 3.7 µm en la fase logarítmica (Figuras 29C y 29D) y en la fase estacionaria de 26.1 ± 7.2 y 39.2 ± 10.4 µm, respectivamente (Figuras 30C y 30D). Ambas cepas son 71 capaces de producir un cambio en pH del medio de 6.63 (fase logarítmica) a 2.68 (fase estacionaria) (Figura 27). Debido a que la cepa Dg es más activa en la acidificación del medio, nuevamente se sugiere un metabolismo más activo respecto a la WT. -Crecimiento en MM-Glucosa. El crecimiento de ambas cepas durante la fase logarítmica es similar (Figura 28). La WT tiene un tiempo de duplicación de 3 horas 8 minutos (Figura 28, círculos llenos) y la Dg de 3 horas 17 minutos (Figura 28, círculos vacíos). El tamaño de la cepa WT fue de 26.40 ± 7.96 µm mientras que para la cepa Dg fue de 37.24 ± 12.07 µm en la fase logarítmica (Figuras 29E y 29F) y en la fase estacionaria fue de 41.37 ± 5.25 y 46.25 ± 8.41 µm, respectivamente (Figuras 30E y 30F). Figura 28. Curva de crecimiento en MM-Glucosa. Las cepas WT (●) y Dg (○) se cultivaron en un medio mínimo con 1% de Glucosa y se determinó su crecimiento a 600nm. Se observó un comportamiento similar durante el crecimiento en ambas cepas y se determinó el pH de las cepas WT (▲) y Dg (Δ) durante la curva de crecimiento. Los datos son el promedio de tres repeticiones de cuatro preparaciones independientes. Las barras de error representan S.D. 72 Ambas cepas producen un cambio en el pH de 6.35 (fase logarítmica) a 2.48 (fase estacionaria) (Figura 28). En este caso, la velocidad de acidificación es similar en ambas cepas, lo que sugiere que el aumento del tamaño de la cepa Dg no es el único factor que acelera el catabolismo con la subsecuente acidificación. Figura 29. Microscopia de las células WT y Dg (fase logarítmica). Microscopia de las células WT y Dg en la fase logarítmica de crecimiento con un objetivo 40x, dilución 1:100. A) WT en YPD. B) Dg en en YPD. C) WT en MM-EtOH. D) Dg en en MM- EtOH. E) WT en MM-Glucosa. F) Dg en en MM-Glucosa. 73 Figura 30. Microscopia de las células WT y Dg (fase estacionaria). Microscopia de las células WT y Dg en la fase Estacionaria de crecimiento con un objetivo 40x, dilución 1:100. A) WT en YPD. B) Dg en en YPD. C) WT en MM-EtOH. D) Dg en en MM-EtOH. E) WT en MM-Glucosa. F) Dg en en MM-Glucosa. 74 Aunado a esto se determinó la resistencia de la cepa Dg a la dilución. La figura 31 muestra el crecimiento a las 48 horas en medio sólido. Se observa que el fenotipo de crecimiento en medios respiratorios, como el MM-EtOH, es similar aún con la eliminación del gen ATP20. Esto sugiere que la mutación no afecta el crecimiento, apoyando lo que se observó en las figuras 25-27, tampoco repercute en la viabilidad de las células para duplicarse y formar colonias. Figura 31. Diluciones seriadas (1:10) de las cepas WT y Dg de U. maydis. Se realizaron diluciones seriadas de las cepas WT y Dg en diferentes medios de crecimiento (A) YPD, (B) MM-EtOH y (C) MM-Glucosa. Las células se cultivaron por 48 horas a 28°C. Análisis de la arquitectura mitocondrial. Como ya se mencionó, se sabe que el dímero de la F1FO-ATP sintasa (V2) está involucrado en el plegamiento de las crestas mitocondriales. Dentro de los objetivos se planteó observar por microscopía de transmisión electrónica la ultraestructura de las mitocondrias de U. maydis. La 75 microscopía electrónica se realizó en el Instituto de Biotecnología de la UNAM, mediante la técnica LR White. Las mitocondrias de la cepa WT de U. maydis (Figura 32A) así como la Dg (Figura 32B) presentan crestas lamelares con la misma morfología, por lo que la eliminación de la subunidad g no modifica la ultraestructura mitocondrial. También podemos sugerir que la eliminación de la subunidad g no afecta la formación del dímero. Por otro lado las micrografías mostraron que la cepa Dg presenta mitocondrias más grandes, lo que podría ser respuesta a la modificación genética. Figura 32. Ultraestructura de las células de U. maydis. (A) Imágenes de las mitocondrias de U. maydis de la WT y (B) Dg. Ambas cepas muestran crestas de tipo lamelares y sin morfologías anómalas. 76 Consumo de oxígeno. Las células WT y ∆g se cultivaron en diferentes medios de carbono como YPD, MM-EtOH y MM-glucosa, y se tomaron muestras a las 8 horas (fase logarítmica) y a las 24 horas (fase estacionaria) de crecimiento. Para las células cultivadas en YPD, durante la fase logarítmica de crecimiento se determinó que el consumo de oxígeno por las cepas WT y ∆g, fue alrededor de 108 natgO/mg/min y 120 natgO/mg/min respectivamente. Al bloquear el transporte de electrones a nivel del complejo IV con 1 mM de KCN, se eliminó la respiración en las células WT al 100% (Figura 33A); por el contrario, en las células ∆g el KCN sólo disminuyó en un 10% la velocidad del consumo de oxígeno (108 natgO/min) y fue totalmente inhibido con 10 µM de n-octil-galato (nOg) (Figura 33B), lo que nos indica la actividad de la oxidasa alterna en las cepas ∆g. Figura 33. Consumo de oxígeno de las células cultivadas en medio rico YPD. Las células fueron obtenidas de la fase logarítmica de crecimiento WT (A) y ∆g (B) y de la fase estacionaria WT (C) y ∆g (D) y se incubaron en 10 mM de KH2PO4 pH 7.4 y se determinó el consumo de oxígeno con un electrodo tipo Clark como se describió en métodos, donde se señala fue añadido 1 mM KCN o 10 µM de nOg. Los datos son el promedio de tres repeticiones de cuatro preparaciones independientes. Por otro lado, en las células cosechadas en la fase estacionaria, el consumo de oxígeno para la WT fue de 348 natgO/mg/min (Figura 33C) y de 168 natgO/mg/min para la ∆g (Figura 33D). En ambos casos el KCN no inhibió totalmente la respiración; para el caso de la WT se observó una inhibición parcial 77 de la velocidad del consumo de oxígeno (288 natgO/mg/min) y para la cepa incluso ∆g un aumento de 252 ngatO/mg/min, confirmando la actividad de la AOX sensible al nOg en ambas cepas en la fase estacionaria. En MM-EtOH, el consumo de oxígeno en la fase logarítmica de la WT y de la mutante ∆g fue de 72 natgO/mg/min (Figura 34A y B). Al igual que para las células recuperadas del cultivo en YPD, el transporte de electrones fue bloqueado a nivel del complejo IV con 1 mM de KCN, eliminando la respiración en las células de la cepa WT, mientras que en las células ∆g el KCN sólo disminuyó la velocidad del consumo de oxígeno a 24 natgO2/mg/min (66%) y fue totalmente inhibido con 10 µM de nOg. Lo anterior también sugiere la presencia de la oxidasa alterna en la cepa ∆g durante la fase logarítmica. Para las células recuperadas en la fase estacionaria el consumo de oxígeno para la WT fue de 276 natgO/mg/min (Figura 34C) y 324 natgO/mg/min (Figura 34D) para la Dg, esta fue sensible tanto a KCN como a nOg. Figura 34. Consumo de oxígeno de las células cultivadas en MM-EtOH. Las células fueron obtenidas de la fase logarítmica de crecimiento de la WT (A) y ∆g (B) y de la fase estacionaria de la WT (C) y ∆g (D), se incubaron en 10 mM de KH2PO4 pH 7.4 y se determinó el consumo de oxígeno con un electrodo tipo Clark como se describió en métodos, donde se señala fue añadido 1 mM KCN o 10 µM de nOg. Los datos son el promedio de tres repeticiones de cuatro preparaciones independientes. El comportamiento de las células cultivadas en MM-glucosa fue similar a los cultivos anteriores obteniendo un consumo de oxígeno dentro de la fase logarítmica para la WT y ∆g de 60 natgO/mg/min (Figura 35A y B). Para las células recuperadas en la fase estacionaria el consumo de oxígeno para la WT y Dg fue de 348 78 natgO/mg/min (Figura 35C) y 228 natgO/mg/min (Figura 35D) respectivamente, y fue necesario añadir KCN y nOg para inhibir al 100% la respiración. Figura 35. Consumo de oxígeno de las células cultivadas en MM-Glucosa. Las células fueron obtenidas de la fase logarítmica de crecimiento de la WT (A) y ∆g (B) y de la fase estacionaria de la WT (C) y ∆g (D), se incubaron en 10 mM de KH2PO4 pH 7.4 y se determinó el consumo de oxígeno con un electrodo tipo Clark como se describió en métodos, donde se señala 1 mM KCN o 10 µM de nOg fue añadido. Los datos son el promedio de tres repeticiones de cuatro preparaciones independientes. En las levaduras, el requerimiento de ATP es alto en la fase logarítmica de crecimiento y disminuye cuando las células entran en la fase estacionaria. Como consecuencia, en la fase logarítmica, el consumo de oxígeno debe ser elevado y acoplado a la generación del DµH+ para producir el ATP necesario para la célula. En cambio, durante la fase estacionaria, la forma de mantener un consumo de oxígeno alto sería promoviendo un desacoplamiento controlado. 79 Tabla 2. Diferencias metabólicas de las cepas WT y Dg. WT Dg YPD Tiempo de duplicación (horas) 2.1 ± 0.5 2.75 ± 0.7 Peso seco (mg/mL) 1.9 ± 0.15 1.51 ± 0.26 Tamaño celular (µm) 18.8 ± 1.8log 22.6 ± 3.9stat 28.1 ± 1.5log 30.1 ± 2.9stat Número de células (1L) 4.12 x 109 log 3.37 x 109 log Consumo de oxigeno celular (natgO2/mg/min) 108log Se inhibe con KCN 348stat 288stat+KCN 120log 108log+KCN 168stat 252stat+KCN MM-EtOH Tiempo de duplicación (horas) 2.7 ± 0.6 3.3 ± 0.7 Peso seco (mg/mL) 0.9 ± 0.07 0.82 ± 0.16 Tamaño celular (µm) 19.9 ± 1.9log 26.1 ± 7.2stat 29.3 ± 2.7log 39.2 ± 10.4stat Número de células (1L) 3.61 x 109 log 2.96 x 109 log Consumo de oxigeno celular (natgO2/mg/min) 72log Se inhibe con KCN 276stat Se inhibe con KCN 72log 24log+KCN 324stat 24stat+KCN MM-Glucosa Tiempo de duplicación (horas) 3.13 ± 0.2 3.28 ± 0.5 Peso seco (mg/mL) 1.5 ± 0.17 1.39 ± 0.14 Tamaño celular (µm) 26.4 ± 7.96log 41.37 ± 5.25stat 37.24 ± 12.07log 46.25 ± 8.41stat Número de células (1L) 3.53 x 109 log 3.01 x 109 log Consumo de oxigeno celular (natgO2/mg/min) 60log 36log+KCN 234stat 12stat+KCN 60log 48log+KCN 228stat 204stat+KCN 80 Se ha reportado en U. maydis que existen dos rutas de transporte de electrones, la vía citocrómica y la vía de los elementos alternos, la cual está constituida por las NADH deshidrogenasas tipo 2 y por la oxidasa alterna (AOX). La AOX es una oxidasa terminal que cataliza la transferencia de electrones del ubiquinol hasta el oxígeno molecular. En contraste con el complejo IV de la vía citocrómica, la AOX no bombea protones y es insensible al cianuro y a la antimicina, pero se inhibe con n-octil-galato (nOg) (Juárez et al., 2006). Por consiguiente, al añadir KCN bloqueamos la respiración a nivel del complejo IV pero el consumo de oxígeno no se ve afectado debido a la presencia de la AOX y ésta sólo se inhibe una vez que se agrega nOg. Estudios de nuestro grupo de trabajo indican que la AOX desempeña dos papeles fundamentales en U. maydis: 1) forma parte de los mecanismos que previenen la producción de ROS y 2) le permite a la célula adaptarse a condiciones generadas por factores externos que limitan o inhiben la actividad de la vía citocrómica (Juárez et al., 2006). Determinación del consumo de glucosa. En experimentos alternos se determinó el consumo de glucosa durante el crecimiento de las cepas WT y Dg. En la figura 36 se muestra que la glucosa está en condiciones saturantes (26 mM) y que las levaduras sólo han consumido el 40% (» 15 mM) de la glucosa total a las 24 h de crecimiento. Esto podría sugerir que la vía de la degradación de glucosa no incrementa su velocidad para generar más sustratos oxidables que pueda consumir la mitocondria. 81 Figura 36. Consumo de glucosa. Las células se cultivaron por 24 h, se inocularón 40u/L en medio YPD. Se incubaron a 28°C en agitación orbital constante a 200rpm. Se tomaron muestras a diferentes tiempos durante el crecimiento de las cepas WT (●) y Dg (○) y se determinó el consumo de glucosa por el método de la glucosa oxidasa a 510 nm. Los datos son el promedio de tres repeticiones de cuatro preparaciones independientes. Las barras de error representan S.D. Permeabilización de las células para ensayos bioenergéticos. Para determinar la funcionalidad de las células permeabilizadas se determinó el consumo mitocondrial de oxígeno. La figura 37 muestra que la digitonina permeabiliza las células, favoreciendo la salida de los metabolitos y produciendo la disminución de la velocidad del consumo de oxígeno. Posteriormente la adición de succinato estimula la respiración, lo cual indica que este sustrato es utilizado por la mitocondria y acelera el consumo de oxígeno. En estas condiciones experimentales es posible controlar los sustratos y metabolitos que utiliza la mitocondria para el consumo de O2. Después de la permeabilización, la velocidad del consumo de oxígeno para la WT es de 32.2 natgO2/mg/min (Figura 37A) y para la ∆g de 50.6 natgO2/mg/min (Figura 37B). 82 Figura 37. Células permeabilizadas de U. maydis. Se permeabilizaron las células WT (A) y ∆g (B) con digitonina para realizar los ensayos bioenergéticos. Para el consumo de oxígeno se ocupó como sustrato oxidable 10 mM de succinato. Los datos son el promedio de tres repeticiones de cinco preparaciones independientes. Determinación del potencial de membrana mitocondrial (DYm). Se determinó el potencial de membrana mitocondrial en las células permeabilizadas y energizadas con succinato (10 mM). A continuación, se adicionó el ADP, el cual es utilizado para la síntesis de ATP acoplada a la despolarización de la membrana interna mitocondrial, lo que se puede interpretar como el estado 3 de la respiración mitocondrial y en presencia de Pi (Figura 38). La adición de CCCP abate el DYm y confirma la integridad mitocondrial. Se observó que la magnitud del DYm en ambas cepas es similar (Figura 38), lo que sugiere que la mutante Dg presenta mitocondrias integras y acopladas y responden a la estimulación por ADP. Figura 38. Potencial de membrana (Dym) generado por las células de U. maydis. Se emplearon células WT (A) y ∆g (B) permeabilizadas con digitonina. Las longitudes de onda para la safranina O son de 533-511 nm. Las células se energizaron con 10 mM de succinato, se despolarizó la membrana con 1 mM ADP y se abatió el potencial con 5 µM de CCCP. Los datos son el promedio de tres repeticiones de cuatro preparaciones independientes. 83 Cuantificación de la síntesis de ATP. Para la cuantificación de la producción de ATP mitocondrial se utilizaron células permeabilizadas con digitonina. La cuantificación de la síntesis de ATP se realizó por medio de una reacción acoplada siguiendo la reducción del NADP+ (Figura 39). Los resultados muestran que la síntesis de ATP en la cepa WT fue de 19.49 ± 0.84 µmolas de ATP (g seco·min-1) y para la Dg es de 14.95 ± 0.57 µmolas de ATP (g seco·min-1) (Figura 39). Figura 39. Síntesis de ATP mitocondrial en las células WT y Dg. Cursos temporales para cuantificar la síntesis de ATP mitocondrial en las células de las cepas (A) WT y (B) Dg. (C) El gráfico muestra la producción de ATP contra tiempo por las células permeabilizadas, el cual se determinó mediante un ensayo acoplado a la reducción del NADP+. La síntesis de ATP para la cepa WT (•) fue de 19.49 µmoles de ATP (g·min-1) y para la cepa Dg (○) fue de 14.95 µmoles de ATP (g·min-1). Los datos son el promedio de tres repeticiones de cuatro preparaciones independientes. Esto indica que la mutante (Dg) tiene una disminución del 20% en la velocidad de síntesis con respecto a la cepa silvestre, lo que sugiere que el incremento en la acidificación del medio extracelular puede deberse a la aceleración del metabolismo con el fin compensar la baja producción de ATP mitocondrial. Cuantificación de la producción mitocondrial de peróxido de hidrogeno. La producción de H2O2 se realizó cuantificando la fluorescencia de la resorufina e interpolando los datos en una curva estándar de peróxido de hidrógeno. Los resultados muestran que las mitocondrias frescas de la cepa WT produjeron 13.45 pmoles de H2O2·mg de proteína mitocondrial, mientras que las de la Dg es de 16.95 pmoles de H2O2·mg de proteína mitocondrial en presencia de NADH (Figura 40). 84 Figura 40. Producción de radicales libres en las mitocondrias de U. maydis. En la gráfica se muestra la producción de H2O2 en mitocondrias aisladas de la cepa WT (barras negras) y Dg (barras grises) donde se muestra que la cepa Dg tiene una disminución de la cantidad de especies reactivas de oxígeno generadas en comparación con la WT. Los datos son el promedio de tres repeticiones de cuatro preparaciones independientes. Las barras de error representan S.D. En el ensayo de H2O2 se adicionó NADH y los inhibidores clásicos de los complejos respiratorios. Los datos muestran que en presencia de cianuro la cepa WT aumenta 31 veces y la Dg aumentó 15 veces la producción de especies reactivas de oxígeno, quizá a través de los complejos I y III. Por el contrario, cuando usamos el n-octil galato, inhibidor de la AOX, la producción de H2O2 disminuyó un 50%, lo que nos sugiere que la toma de los electrones provenientes del ubiquinol por la vía alterna es más lenta en comparación la vía citocrómica. Solubilización de los supercomplejos y complejos mitocondriales de U. maydis. Los complejos respiratorios de las cepas WT y ∆g fueron solubilizados con digitonina, empleando una relación de 0.5, 1.0, 2.0, 3.0 y 5.0 g de digitonina/g de proteína mitocondrial, y se resolvieron en una electroforesis nativa. La ubicación de los complejos y supercomplejos se determinó por la actividad para cada uno de los 85 complejos respiratorios (Figura 41A-F). La actividad del complejo I en la cepa WT se asoció a los supercomplejos; sin embargo, en la cepa Dg el complejo I se encontró principalmente en forma monomérica. Es importante recordar que la velocidad de síntesis de ATP en la cepa mutante está disminuida, lo que provocaría la acumulación de los protones en el lado P, aumentando el DYm y promoviendo el flujo reverso de los electrones en el complejo I, con lo que se estimularía la producción de especies reactivas de oxígeno. En este sentido, el desensamble de los supercomplejos permite la reducción de la poza de quinol y el subsecuente drenado de electrones por la AOX, en un proceso que no involucra el bombeo de protones. 86 Figura 41. Actividad en gel de los supercomplejos y complejos respiratorios. Los complejos respiratorios solubilizados con digitonina se resolvieron en una BN-PAGE y hrCN-PAGE (para el complejo V) en gel del 4 al 10% de poliacrilamida, usando como estándares los complejos respiratorios de corazón de bovino (Bt). Se hicieron tinciones de actividad en gel para identificar a los complejos respiratorios; (A) tinción con Coomassie® R250, (B) actividad en gel del complejo I (CI), (C) actividad en gel del complejo II (CII), (D) actividad en gel del complejo IV (CIV), (E) actividad en gel del complejo V (Actividad de ATPasa), (F) actividad en gel del complejo V (Actividad de ATPasa) después de 25 horas de incubación y varios cambios del amortiguador de actividad. 87 Aunado a esto, en lo que se refiere al análisis de la actividad en gel de ATPasa del complejo V mostró la presencia del V2 y V1 en las muestras correspondientes a la cepa WT y para el caso de la cepa Dg la actividad de ATPasa solo se asoció con el V1Dg durante los primeros 10 min (Figura 41E). No obstante después de 25 horas de incubación y varios cambios de la solución de actividad de ATPasa observó la presencia del V2Dg (Figura 41F). Al teñir las proteínas con azul de Coomassie se observó una banda que corresponde al V2Dg (Figura 41A), la cual se sometió a una electroforesis en segunda dimensión para comprobar que el V2Dg realmente estaba presente aunque en muy baja concentración (Figura 42B). Nuestro grupo de trabajo reportó que el V2wt es estable a los detergentes gracias a la interfase entre los monómeros, por lo tanto, al eliminar la subunidad g, éste pierde la estabilidad y presenta sensibilidad a la digitonina, aun en condiciones bajas del detergente (Figura 42B). Figura 42. 2D-Tricina-SDS-PAGE. Los complejos respiratorios solubilizados con digitonina se resolvieron en una primera dimensión nativa (BN-PAGE) en gel del 4 al 10% de poliacrilamida. La segunda dimensión se realizó cortando el carril de la primera dimensión nativa y sometiéndolo a una electroforésis desnaturalizante. Las proteínas de los complejos respiratorios se resolvieron en un gel de tricina-SDS-PAGE al 10%. A) Identificación de las subunidades a y b en el V2 y V1 WT. B Identificación de las subunidades a y b en el V1 de la cepa Dg. 88 Aislamiento del dímero y del monómero de la F1FO-ATP sintasa. Los oligómeros del complejo V de las cepas WT y Dg fueron aislados en un gradiente continuo de sacarosa y se analizaron en una hrCN-PAGE (Figura 43A y 44A). El análisis de la actividad en gel de ATPasa del complejo V mostró que el V2WT se localiza en las fracciones del gradiente de sacarosa que tienen una densidad entre 1.14 y 0.99 g/mL, mientras que el V1WT se ubica entre 0.84 y 0.77 g/mL; la mezcla de ambos oligómeros se encontró entre 0.99 y 0.84 g/mL (Figura 43A). Las diferentes fracciones se reúnen en tres lotes independientes, el V2wt, el V1wt, y la mezcla de ambos (V2/V1). Se determinó su concentración de proteína y se almacenaron a -70°C. Por medio de una hrCN-PAGE se realizó el análisis de las muestras concentradas (Figura 43B), con lo que se determinó que los oligómeros mantienen su estructura y actividad de ATPasa y que éstos son las únicas enzimas en esta preparación, que llevan a cabo la hidrólisis del Mg-ATP, además no se observó la presencia del sector F1 libre. Para corroborar la integridad del complejo V se determinó su sensibilidad a oligomicina (Figura 43C). Las proteínas fueron teñidas con Coomassie, lo que mostró la presencia de otras proteínas distintas al CV (Figura 43D). Debido a la presencia de proteínas contaminantes se realizó la cuantificación del CV en cada preparación para corregir la actividad de ATPasa (ver anexo 10.4 Figura S1). 89 Figura 43. Purificación de los oligómeros de la F1FO-ATP sintasa mitocondrial de la cepa WT. La presencia de los oligómeros de la F1FO-ATP sintasa en el gradiente de sacarosa se determinó en geles hrCN-PAGE (A) La actividad de ATPasa en gel muestra las formas dimérica y monomérica de la ATP sintasa de U. maydis. (B) Los oligómeros de la F1FO-ATP sintasa se reunieron y concentraron en filtros de Amicon, el corte de la membrana fue de 100K. El análisis de las muestras concentradas se realizó mediante hrCN-PAGE con lo que se determinó que los oligómeros mantienen su estructura y actividad de ATPasa aún después de almacenarlos a -70oC. (C) Inhibición por oligomicina de la actividad ATPasa en gel de V2 y V1 aislado. (D) hrCN-PAGE teñido con azul Coomassie donde se observa al V2 y V1 aislados y concentrados sin la presencia de la F1 libre. (E) La identidad de las subunidades del complejo V se realizó por espectrometría de masas a partir de una electroforesis 2D-Tricina-SDS en geles al 16 % de poliacrilamida. Los oligómeros del complejo V se obtuvieron de una electroforesis BN-PAGE y sus subunidades se resolvieron en una 2D-PAGE. Se muestra la identidad y la masa molecular de cada una de las bandas. PHB1, PHB2 y ANT corresponden a las prohibitinas 1 y 2 y al translocador de adenín nucleótidos respectivamente (En el gradiente de sacarosa 33%=1.1383 g/mL y 25%=.0.7706 g/mL). 90 El análisis de la actividad de ATPasa en gel de los oligómeros de la cepa Dg, mostró que el V1Dg se localiza en las fracciones del gradiente de sacarosa que tienen una densidad entre 0.84 y 0.77 g/mL (Figura 44A). El V1Dg aislado (ver materiales y métodos) mantiene su estructura y actividad de ATPasa (Figura 44B) sensible a oligomicina (Figura 44C). La tinción con Coomassie del V1Dg (Figura 44D) demostró la presencia de otras proteínas diferentes a la F1FO-ATP sintasa, por lo que nuevamente fue necesario cuantificar la concentración del complejo V para realizar la corrección de la actividad de ATPasa (ver anexo 10.4 Figura S1). Es importante hacer notar que el V2Dg es poco estable y que, aunque está presente en el solubilizado con digitonina, no fue posible aislarlo bajo estas condiciones experimentales (gradiente de sacarosa). Figura 44. Purificación del monómero de la F1FO-ATP sintasa mitocondrial de la cepa Dg. La presencia de los oligómeros de la F1FO-ATP sintasa en del gradiente de sacarosa se determinó en geles hrCN-PAGE (A) La actividad de ATPasa en gel muestra la actividad del V1Dg de la ATP sintasa de U. maydis. Los geles se tiñeron con azul Coomassie y las fracciones que contienen al V1Dg se reunieron y concentraron en filtros de Amicon, el corte de la membrana fue de 10K. (B) El análisis de las muestras concentradas se realizó mediante hrCN-PAGE con lo que se determinó que el V1Dg mantiene su estructura y actividad de ATPasa aún después de almacenado a -70oC. (C) Inhibición por oligomicina de la actividad ATPasa en gel de V1Dg aislado. (D) hrCN-PAGE teñido con azul Coomassie® 250 donde se observa al V1Dg aislado y concentrado sin la presencia de F1 libre. 91 Identificación de las subunidades del dímero y monómero de la F1FO-ATP sintasa. Las subunidades del V2 y del V1 de la F1FO-ATP sintasa se resolvieron mediante su movilidad electroforética en la 2D-Tricina-PAGE (Figura 43E). Las proteínas se tiñeron con azul brillante de Coomassie® R250 y se identificaron mediante espectrometría de masas. Se identificaron 15 subunidades asociadas al dímero y monómero de la F1FO- ATP sintasa de la cepa WT. Así mismo, se puede observar que el dímero muestra a las subunidades e y g, responsables de la dimerización del complejo V en S. cerevisiae (Arnold et al., 1998; Couoh-Cardel et al., 2010). Dentro de este análisis se encontró a la subunidad ATP12, la cual es una chaperona que participa en el ensamblaje de la enzima. También se identificaron a las prohibitinas 1 y 2 (PBH1 y PBH2) que han sido relacionadas con la formación de las crestas mitocondriales. También se identificó al translocador de adenín nucleótidos (ANT) y el transportador de fosfatos (PC) los cuales son muy importantes para la síntesis de ATP. El encontrar el ANT y el PC asociados al V2 y V1 de la ATP sintasa sugiere la presencia del sintasoma de ATP (Ko et al., 2003), lo cual abre la posibilidad de estudiar dicho supercomplejo en este organismo (Esparza-Perusquía et al., 2017). Determinación de los parámetros cinéticos de la F1FO-ATP sintasa. La caracterización de la actividad de ATPasa de los oligómeros del complejo V, se determinó mediante la oxidación del NADH en presencia de un sistema regenerador de ATP (ver métodos). La presencia de un sistema regenerador permite eliminar el ADP producido por la actividad de ATPasa, ya que altas concentraciones de este nucleótido pueden inhibir la actividad de ATPasa de la enzima (Fiske et al., 1925; Penefsky, 1974). A partir de la pendiente de cada registro espectrofotométrico del curso temporal se obtiene la actividad específica, que aumentó conforme incrementó la concentración de Mg-ATP. Los datos obtenidos se ajustaron al modelo cinético propuesto por Michaelis-Menten. Si la velocidad de una reacción se calcula a una 92 determinada concentración de sustrato [S] << Km, la velocidad de la reacción (u) aumenta linealmente con respecto a [S]. Sin embargo, cuando la [S] >> Km, la enzima se satura y alcanza su velocidad máxima (Vmax), la cual no sobrepasará en ningún caso, independientemente del aumento de [S]. La constante de Michaelis (Km) se define como la afinidad de la enzima por el sustrato, cuando [S] = Km y Vmax/2 (Segel, 1993). Asimismo, los cursos temporales del V2WT, V1WT y V1Dg mostraron que la actividad es lineal con respecto al tiempo y que aumenta conforme se incrementa la concentración de Mg-ATP (Figura 45A-C). Para corroborar la integridad del complejo V se inhibió con oligomicina (vide infra). Figura 45. Caracterización cinética de la F1FO-ATP sintasa mitocondrial aislada. La actividad del V2, V1 y V1Dg se determinó como se describe en la sección de materiales y métodos. Cursos temporales de la actividad de ATPasa del complejo V a diferentes concentraciones de Mg-ATP. (A) V1WT. (B) V2WT. (C) V1Dg. 93 Sin embargo, se ha reportado que la actividad de ATPasa del complejo V se estimula por el detergente n-dodecil β-D-maltósido (DDM) (Villavicencio-Queijeiro A, et al., 2009). Por tal motivo, se realizó la titulación de la actividad de ATPasa del V2 de U. maydis con el DDM (Figura 46). Como se puede observar, la actividad de hidrólisis de ATP del V2 incrementó en presencia del DDM mostrando una AC50 de 4x10-4%; la máxima estimulación del V2 de U. maydis se alcanza a una concentración del detergente de 5 X 10-3%. Figura 46. Titulación de la actividad de ATPasa del V2 con DDM. La actividad de ATPasa del V2 aislado se estimuló por el DDM (A). A partir de los cursos temporales se calculó una AC50 = 4X10-4%. La máxima activación se alcanzó con 0.005% de DDM (B). Con esta información, se determinó la actividad de ATPasa de V2WT, V1WT y V1Dg partir de los cursos temporales en presencia y ausencia de 0.005% de DDM. La figura 47 muestra que la velocidad de ATPasa aumenta en presencia de DDM en cada uno de los oligómeros. 94 Figura 47. Actividad de la F1FO-ATP sintasa mitocondrial en presencia de DDM. Actividad de ATPasa del V2, V1 y V1Dg en presencia de 0.005% de DDM. Donde se indica, se realizó la adición de la muestra. Cursos temporales de la actividad de ATPasa del V1 (A), V2 (B) y V1Dg (C). Los cursos temporales mostrados en las Figuras 45 y 47 se ajustaron al modelo de Michaelis-Menten (Figura 48) a partir del cual se determinaron las constantes cinéticas. El V1WT presenta una KM de 308 ± 90 µM para el Mg-ATP y Vmax de 0.83 ± 0.05 µmol ATP hidrolizado/mg F1FO-ATP sintasa·min−1 (Figura 48A, círculos negros). Para el caso del V2WT la actividad de ATPasa presenta una KM de 884 ± 100 µM para el Mg-ATP y Vmax de 0.54 ± 0.08 µmol ATP hidrolizado/mg F1FO- ATP sintasa·min−1 (Figura 48B, círculos negros). Para el V1Dg se determinó una KM de 155 ± 0.05 µM para el Mg-ATP y Vmax de 0.58 ± 0.03 µmol ATP hidrolizado/mg F1FO-ATP sintasa·min−1 (Figura 48C, círculos negros). Por otra parte, analizando los cursos temporales de los oligómeros activados con 0.005% de DMM (Figura 48) observamos que también mostraron un comportamiento cinético que se ajusta al modelo de Michaelis-Menten. El V1WT 95 activado presenta una KM de 207 ± 30 µM para el Mg-ATP y una Vmax de 1.43 ± 0.04 µmol ATP hidrolizado/mg F1FO-ATP sintasa·min−1 (Figura 48A, círculos blancos). Para el caso del V2WT activado tiene una actividad de ATPasa con una KM de 488 ± 80 µM para el Mg-ATP y una Vmax de 9.60 ± 0.26 µmol ATP hidrolizado/mg F1FO- ATP sintasa·min−1 (Figura 48B, círculos blancos). El V1Dg tiene una KM de 920 ± 200 µM para el Mg-ATP y una Vmax de 2.3 ± 0.15 µmol ATP hidrolizado/mg F1FO-ATP sintasa·min−1 (Figura 48C, círculos blancos). Figura 48. Actividad de ATPasa de los oligomeros de la F1FO-ATP sintasa mitocondrial aislados de las cepas WT y Dg. La hidrólisis de diferentes concentraciones de Mg-ATP por los oligómeros de la F1FO-ATP sintasa aislados se realizó a pH 8.0 y se ajusto a la ecuación de Michaelis-Menten, en presencia (○) o ausencia (●) de 0.005% de DDM. (A) V1WT. (B) V2WT. Para mayor claridad, el recuadro en el panel representa la actividad sin la activación por DDM. (C) V1Dg. Los datos son el promedio de cuatro réplicas de cinco preparaciones independientes ±. Las barras de error representan S. D. La concentración de F1FO-ATP sintasa se determinó mediante un análisis densitométrico de las subunidades a y b. Las proteínas de las muestras V1WT, V2WT y V1Dg se resolvieron en geles de Tricina-SDS-PAGE al 10%, utilizando BSA como estándar. Las proteínas fueron teñidas con azul brillante de Coomassie® R250 (Ver 96 anexo 10.4, Fig. S1). El análisis de densitometría se realizó con el software MyImage Analysis 2.0 (Thermo Scientific) (ver anexo 10.4 Figura S1). Efecto de la oligomicina en la actividad de ATPasa del V1 y V2. La oligomicina es un inhibidor específico de la actividad de ATPasa del complejo V que solo actúa cuando se mantiene la interacción del sector FO con el F1. Debido a que la oligomicina es un inhibidor de asociación lenta pero fuertemente unido, se decidió explorar la velocidad de inhibición de los oligómeros activados por DDM. Se puede observar que el tiempo de inhibición del V2WT, al parecer, es menor que para el V1WT y el V1Dg (Figura 49A y D), lo que va acompañado de una mayor sensibilidad del V2WT a la oligomicina (Figura 49B). La inhibición de V1wt fue superior a 90% a 3.8 µM de oligomicina (Figura 49C, círculos blancos) mientras que para el V2wt (Figura 49C, círculos negros) y el V1Dg fue de 1.3 µM (Figura 49E). El análisis de la inhibición de ATPasa por oligomicina mostró que V2wt tenía un Ki = 24 ± 3 nM, la Ki para V1WT fue de 169 ± 10 nM y una Ki = 52.7 ± 4.4 nM para el V1Dg. Dado que la cantidad de complejo V en ambas preparaciones se cuantificó, fue posible calcular la relación de oligomicina/F1FO- ATP (Figura 49C y E). La relación fue de 0.85 ± 0.12 para V2WT, de 6 ± 0.43 para V1WT y 5.03 ± 0.57 para el V1Dg. En conjunto, nuestros resultados sugieren que los monómeros en el dímero tienen una comunicación interfuncional y que al inhibir a uno de los monómeros se puede alterar la interfase y modificar la Ki para la oligomicina. Cuando se elimina a la subunidad g, el dímero parece ser más sensible a los detergentes. U. maydis es un basidiomiceto aerobio estricto que depende principalmente de la fosforilación oxidativa para obtener la energía necesaria para llevar a cabo los procesos metabólicos, si consideramos que la cepa Dg tarda más tiempo en duplicarse y a su vez tiene una síntesis de ATP menor que la cepa WT, podemos suponer que la eliminación de la subunidad g no sólo modifica la interfase de la F1FO-ATP sintasa, sino que también su capacidad catalítica. 97 Figura 49. Efecto de la oligomicina sobre la actividad de ATPasa. La actividad del complejo V se determinó en presencia de 0.005% de DDM y 5 mM ATP-Mg. El curso temporal para la oxidación del NADH acoplado a la hidrólisis de ATP se monitoreo durante 25-30 minutos para la máxima inhibición de ATPasa de V1 (A) o V2 (B) y se analizó el registro espectrofotométrico final como se describe en la sección de materiales y métodos. Las flechas indican las adiciones de enzima y oligomicina, a la derecha se muestra la concentración de inhibidor. (C) El gráfico muestra 1-u/u0 en el intervalo de menor concentración de la oligomicina, para mostrar la sensibilidad diferencial de los oligómeros, la máxima inhibición de V1 (○) o V2 (●) en función de la concentración de oligomicina. (D) El curso temporal para la oxidación del NADH acoplado a la hidrólisis de ATP para la máxima inhibición de ATPasa de V1Dg. Las flechas indican las adiciones de enzima y oligomicina, a la derecha se muestra la concentración de inhibidor. (E) La máxima inhibición de V1Dg en función de la concentración de oligomicina. V representa la actividad residual de la inhibición con oligomicina y V0 representa la actividad ATPasa antes de la adición de oligomicina. Los datos son el promedio de tres repeticiones de cuatro preparaciones independientes. Las barras de error representan S.D. Termoestabilidad del V2 y V1 de la F1FO-ATP sintasa. La termoestabilidad del V1 y V2 se realizó incubando a la enzima durante 15 min a diferentes temperaturas en un medio que contenía 30 mM de HEPES, 5% glicerol, pH 8.0. Al término de la incubación se determinó la actividad a 25°C empleando el sistema regenerador de ATP acoplado a la oxidación del NADH (ver materiales y métodos). La estabilidad térmica del V2 y V1 se evaluó determinando la actividad residual de ATPasa. En la figura 50 se puede observar que la actividad residual 98 (u/u0) del V1 y V2 se mantuvo constante entre los 20 y los 40°C, con un ligero incremento a los 45°C; posteriormente se observó un descenso en la actividad con respecto a la temperatura. En el intervalo de 46 a 59°C ocurre la inactivación de la enzima con un comportamiento que se ajusta a una curva monotónica, de la cual se puede calcular una temperatura de transición aparente (Tm) de aproximadamente 52°C. Esto sugiere que la inactivación del V1 y V2 ocurre en un solo paso. Para verificar que funcionalmente el V1 y V2 mantienen unidos a los sectores F1 y FO, se determinó la inhibición de la actividad de ATPasa por la oligomicina (2 µg/mL); con esto se demostró que, a cualquier temperatura ensayada, la F1FO-ATP sintasa permanece ensamblada. Adicionalmente, en el análisis hrCN-PAGE no se observó la liberación de sector F1 para el V1WT y V2WT en el intervalo de temperaturas evaluadas (Figuras 50B y C). Esto sugiere que la inactivación de V2WT podría ocurrir a través de su disociación en V1 y finalmente la inactivación de V1WT; por lo tanto, la interfase monómero-monómero aparentemente no contribuye a la estabilidad de V2WT. Figura 50. Efecto de la temperatura en la actividad de ATPasa. (A) Curvas de inactivación térmica de V1 y V2. V1 (○) o V2 (●) se incubaron durante 15 minutos a las temperaturas indicadas y luego se enfriaron a temperatura ambiente. La actividad residual (V/V0) se calculó usando V0 como actividad ATPasa a 25°C antes de calentarse. hrCN-PAGE de V1 (B) y V2 (C) se incubaron como se describe en (A). La Tm se calculó en 52.3°C. Los datos son el promedio de tres repeticiones de cuatro preparaciones independientes. 99 “Networking energies” Odra Noel Discusión “Si haces las cosas bien desde el principio jamás tendrás problemas” Dr. Armando Gómez-Puyou 100 6. DISCUSIÓN El objetivo del presente trabajo fue caracterizar la actividad y a la estructura de la F1FO-ATP sintasa de U. maydis cuando se elimina el gen que codifica para la subunidad g dimerizante y su repercusión en le estado bioenergético mitocondrial. El dímero de la F1FO-ATP sintasa (V2) tiene un papel sobresaliente en el plegamiento de la membrana interna mitocondrial para la formación de las crestas. Es importante señalar que en las crestas se localiza la mayor concentración de los complejos y supercomplejos respiratorios, por lo que es ahí donde se realiza principalmente la oxidación de las coenzimas (NADH y FADH2) y la síntesis de más del 90% del ATP celular. Debido a la estrecha relación que hay entre los V2, el plegamiento de las crestas y la síntesis de ATP, el estudio del V2 ha cobrado gran relevancia. Se ha determinado que cada reino biológico emplea diferentes subunidades para conformar y estabilizar al V2: la IF1 en los mamíferos como Bos taurus, las subunidades ASA en las algas como Chlamydomonas reinhartii, y las subunidades e y g en los hongos como Saccharomyces cerevisiae y Ustilago maydis. Esta observación indica claramente que la formación del dímero es un evento de convergencia evolutiva. Las F1FO-ATP sintasas son enzimas altamente conservadas que juegan un papel fundamental tanto en la síntesis del ATP como en el mantenimiento de la arquitectura mitocondrial. Tanto el sector F1 como el FO presentan todas las subunidades conservadas involucradas en la catálisis (Giraud et al., 2002; Wittig et al., 2008; Couoh-Cardel et al., 2010). Debido a la interacción de las diferentes subunidades, hemos clasificado el rearreglo de las crestas relacionado con dos tipos de asociaciones en la interfase del dímero, un central y otra periférica; la primera constituida por la interacción de las subunidades a y i/j y la segunda por la interacción de las subunidades g, e y k. 101 En realidad, se ha descrito que la interfaz monómero-monómero del dímero de levadura y mamífero está constituida por las subunidades a, i/j, k, g y e (Couoh- Cardel et al., 2010; Guo et al., 2017). La interfaz se puede dividir en una zona central y una zona periférica. En la zona central, las subunidades i/j interactúan a través de dos tramos cortos de ~10 residuos; mientras que la interacción entre las subunidades a se produce a través de dos hebras que constituyen una estructura plana de cuatro hebras con una superficie hidrófoba y una superficie hidrofílica (Guo et al., 2017). La zona periférica de la interfaz monómero-monómero está constituida por las subunidades k y e. Estas subunidades poseen una estructura similar: una hélice a N-terminal con dos dominios, uno transmembrana y uno soluble que se extiende (alrededor de 40 Å para la subunidad e) hacia el espacio intermembranal para interactuar con las subunidades equivalentes del otro monómero (Guo et al., 2017). La subunidad g se une la subunidad e a través de una sola a hélice transmembranal, probablemente a través del motivo conservado Gly-X-X-X-Gly de las dos proteínas (Bustos et al., 2005). Simultáneamente, el N-terminal de la subunidad g interactúa con ~ 50 residuos del N-terminal de la subunidad b. Así pues el dominio de la zona periférica de la interfaz monómero-monómero constituido por las subunidades e, g y b, con el apoyo de la subunidad k, ayuda a plegar la membrana mitocondrial interna (Baker et al., 2012; Guo et al., 2017). Para el caso de S. cerevisiae, se cree que la oligomerización de la F1FO-ATP sintasa puede estar involucrada en la actividad catalítica de la enzima y que de manera indirecta afecta el estado metabólico de las células. Cuando se impide la dimerización de la F1F0-ATP sintasa, eliminando genéticamente las subunidades e y g simultáneamente, las mitocondrias presentan una morfología alterada en las crestas. Así pues, el rearreglo de las crestas durante la regulación bioenergética de la mitocondria supone el cambio en la proporción dímero/monómero del complejo V (Velours et al., 2009). 102 Gracias a la estrecha relación que hay entre el V2, el plegamiento de las crestas y la síntesis de ATP, se han realizado muchos esfuerzos para determinar que ocurre cuando alguno de estos componentes se altera. Por ejemplo, ¿Qué ocurre con la síntesis de ATP o la arquitectura de las crestas si se pierde el V2? Campanella et al., (2008) haciendo un knock out condicional de la subunidad IF1 en las células HeLa, determinó que al eliminarse el V2, se abatía el DYm, la producción de ATP y el número y tamaño de las crestas disminuía. Sin embargo, el papel de la subunidad IF1 en la formación del V2 aún es controvertido. Recientemente, se ha creado un ratón knock out para la subunidad IF1 sin que presente alteraciones metabólicas significativas o cambios en la arquitectura mitocondrial (Campanella et al., 2008). Adicionalmente, la eliminación simultánea de las subunidades dimerizantes g y e en S. cerevisiae induce la pérdida de V2 y cambios en la morfología de las crestas mitocondriales, presentando estructuras conocidas como aros de cebolla (Velours et al., 2009). Sin embargo, como S. cerevisiae es un organismo facultativo, la producción mitocondrial de ATP puede ser suplida por el metabolismo fermentativo, por lo que la participación mitocondrial en la bioenergética de estos mutantes no está clara. Como U. maydis es un basidiomiceto no fermentativo, su metabolismo bioenergético está en manos de las mitocondrias; así, es un buen modelo para determinar la repercusión que tiene la eliminación de la subunidad g sobre la formación del V2 y la síntesis de ATP. Como se demostró en este trabajo, la eliminación de la subunidad g no cambia significativamente el metabolismo celular, ya que no se detectaron cambios en el tiempo de duplicación celular, en el consumo de glucosa y en la producción de peso seco. Esto sugiere que la asimilación de carbono y su incorporación en biomasa es similar tanto en la cepa mutante como en la silvestre. Sin embargo, hay que hacer notar que el número de células y el tamaño de las mismas son diferentes entre la cepa WT y la mutante, sin que 103 tengamos hasta ahora una explicación. No obstante, es claro que la estructura clásica de mitocondrias se conserva en la cepa mutante. El análisis bioenergético mostró que las mitocondrias de ambas cepas producen un DYm similar cuando consumen oxígeno por la vía citocrómica clásica utilizando succinato como sustrato (Tabla 2); sin embargo, la cepa Dg mostró una disminución de la síntesis de ATP asociada con la expresión de la AOX desde las primeras etapas del crecimiento. Se ha reportado que condiciones experimentales similares (es decir, DYm alto, succinato como suministro de electrones para la cadena respiratoria y una disminución de la actividad del complejo V) puede producir el transporte inverso de electrones (RET) en el complejo I y conducir a un aumento en la producción de radicales libres (Robb et al., 2018). En este sentido, la AOX transfiere los electrones de la QH2 directamente a O2, sin pasar por el complejo IV y, por lo tanto, actúa como una válvula de seguridad para evitar la reducción excesiva de la poza de quinonas (Robb et al., 2018; El-Khoury et al., 2013; El- Khoury et al., 2014). Es posible sugerir que la eliminación de la subunidad g produce una disminución de la síntesis de ATP debido a que el dímero de la F1FO-ATP sintasa fue modificado, lo que resulta en un bajo consumo de DYm, una acumulación de NADH y un aumento de producción de ROS (es decir, a través de RET en el complejo I). En este sentido, la producción de ROS durante la oxidación de succinato se redujo por la expresión de AOX en la cepa Dg. Aunado a esto, la cepa Dg mostró una reducción del 20% en la síntesis de ATP. Como se ha descrito, la subunidad g desempeña un papel importante en la zona periférica de la interfaz monómero-monómero, y su eliminación conduce a que solo la subunidad e de un monómero se una a la subunidad k del otro monómero; produciendo una interfaz débil. Es importante recordar que la comunicación monómero-monómero es importante para estimular la actividad de ATPasa del V2 (Esparza-Perusquía et al., 2017); si consideramos que la F1FO-ATP sintasa es 104 termodinámicamente reversible a expensas de consumir o generar el DYm, entonces podemos suponer que la capacidad de síntesis de ATP es mayor por el V2 que por el V1. Esta hipótesis es respaldada por el hecho de que la cepa Dg presenta una baja síntesis de ATP. Además de la participación de la interfaz monómero-monómero en la membrana interna, se ha propuesto que la subunidad g podría desempeñar un papel importante en la actividad de V2 (Esparza-Perusquía et al., 2017). La actividad ATPasa de V2WT fue 7 veces mayor que V1WT, además de presentar una alta sensibilidad a la oligomicina (Esparza-Perusquía et al., 2017). Por otro lado pudimos caracterizar el V1Dg (Tabla 3), sin embargo no se logró aislar el V2Dg ya que este se disocia cuando pasa por el gradiente. Por lo tanto, la eliminación de la subunidad g no compromete la formación del V2Dg pero si la estabilidad de éste, así como su resistencia a los detergentes. Aunque se tiene que verificar su capacidad para sintetizar ATP, podemos proponer que la velocidad de consumo del gradiente electroquímico de protones disminuirá y la proporción de complejos respiratorios reducidos aumentará. Por último, se observó que en la cepa ∆g los respirasomas I1:(III2)1-2:(IV)1-4 son lábiles y los complejos respiratorios se encuentran principalmente como unidades independientes. Se ha reportado que el ensamble de los respirasomas puede limitar la generación de superóxido durante el flujo de electrones y el consumo de oxígeno (Panov et al., 2007; Reyes-Galindo et al., 2019). La inestabilidad de los respirasomas de la cepa ∆g nos permite suponer que durante el transporte de electrones aumenta la probabilidad de producir especies reactivas de oxígeno a nivel de los complejos I y III (Panov et al., 2007; Reyes-Galindo et al., 2019). La relación que existe entre la estabilidad del V2 (y la presencia de la subunidad g) sobre los respirasomas es una hipótesis que debe comprobarse experimentalmente. 105 Tabla 3. Parámetros cinéticos para la actividad de hidrólisis de ATP para el dímero y el monómero de la F1FO-ATP sintasa de U. maydis. Monómero (V1WT) a Dímero (V2WT) a Monómero (V1Dg) a F1FO-ATP sintasa no activada Vmax (µmol ATP hidrolizado/mg F1FO-ATP sintasa·min-1) 0.83 ± 0.05 0.54 ± 0.08 0.58 ± 0.15 Km (µM) 308 ± 90 884 ± 100 155 ± 0.047 kcat (s -1) 8.4 ± 0.06 10.6 ± 0.15 5.86 ± 0.05 kcat/Km (M-1s-1) 2.7 x 104 ± 0.3 x 104 1.2 x 104 ± 0.1 x 104 3.78 x 104 ± 0.3 x 104 F1FO-ATP sintasa activada con DDM Vmax (µmol ATP hidrolizado/mg F1FO-ATP sintasa·min-1) 1.43 ± 0.04 9.60 ± 0.26 2.3 ± 0.15 Km (µM) 207 ± 30 488 ± 80 920 ± 200 kcat (s -1) 14.5 ± 0.03 207.7 ± 0.03 23.26 ± 0.1 kcat/Km (M-1s-1) 7.0 x 104 ± 0.2 X 104 4.2 x 105 ± 0.2 X 104 2.53 x 104 ± 0.08 X 104 Ki (nM) para la oligomicina 169 ± 10 160 ± 17b 24 ± 3 21 ± 4b 52.7 ± 4.4 Relación oligomicina / F1FO-ATP sintasa (mol/mol) para alcanzar el 50% de la inhibición de ATPasa 6.0 ± 0.43 0.85 ± 0.12 5.03 ± 0.57 a = Los moles de F1FO-ATP sintasa en las muestras V1 y V2 se determinaron como se describe en la sección de procedimientos, y los parámetros cinéticos se mostraron como mg de F1FO-ATP sintasa. b = valores Ki calculados por el software DynaFit (ver sección Material y métodos). 106 “Mitchell’s dream” Odra Noel Science is beautiful: it has truth, it has drama, it is full of wonder. Conclusiones 107 7. CONCLUSIONES - La cepa Dg conserva el fenotipo respiratorio mitocondrial y la capacidad de crecimiento empleando diferentes fuentes de carbono. - El tiempo de duplicación celular y producción de biomasa fue similar para ambas cepas (WT y Dg) en las tres fuentes de carbono (1% etanol, 1% glucosa y 1% lactato). - Las mitocondrias de la cepa Dg son más grandes que las de la cepa WT. - La eliminación de la subunidad g no compromete el estado bioenergético de la mitocondria ya que la velocidad de consumo mitocondrial de oxígeno y el potencial de membrana mitocondrial no se afectaron. - La síntesis de ATP mitocondrial disminuyó un 20% en la cepa Dg. - En la cepa Dg la AOX se expresa a partir de la fase logarítmica de crecimiento y se mantiene durante la fase estacionaria. - La presencia de la AOX en la cepa ∆g está asociada a una menor producción de espacies reactivas de oxígeno. - La cepa ∆g no produce una gran cantidad de especies reactivas de oxígeno comparado con la WT gracias a la presencia de la AOX desde las primeras fases del crecimiento. - En la cepa ∆g los supercomplejos son más sensibles a la digitonina y los complejos respiratorios se encuentran principalmente como supercomplejos menores y unidades independientes. - La eliminación de la subunidad g en U. maydis no compromete la dimerización del complejo V pero modifica su estabilidad, a juzgar por su disociación en presencia de bajas concentraciones de digitonina. 108 “Mitochondrial dawn” Odra Noel Perspectivas 109 8. PERSPECTIVAS - Construir la mutante De y la doble mutante De/Dg. - Realizar experimentos de carbonilación de proteínas como indicador de estrés oxidativo. - Realizar experimentos de envejecimiento en la cepa Dg. - Realizar experimentos con diferentes sustratos por ejemplo alimentando el CI con NADH en células permeabilizadas. - Determinar la formación de crestas mitocondriales de la cepa Dg en ayuno. - Determinar el potencial de membrana, la respiración y síntesis mitocondrial de ATP en células en ayuno de la cepa Dg. 110 “Mitchell’s equation II” Odra Noel Referencias Bibliográficas “Aquí no somos envidiosos” Dr. Antonio Peña 111 9. REFERENCIAS BIBLIOGRÁFICAS 1. Acín-Pérez, R., Fernández-Silva, P., Peleato, M. L., Pérez-Martos, A. and Enriquez, J. A. (2008) Respiratory active mitochondrial supercomplexes. Molecular cell, 32(4):529-39. 2. Abrahams, J. P., Leslie, A. G., Lutter, R. and Walker, J. E. (1994) Structure at 2.8 A resolution of F1-ATPase from bovine heart mitochondria. Nature, 370(6491):621-8. 3. Adachi, K., Oiwa, K., Nishizaka, T., Furuike, S., Noji, H., Itoh, H., Yoshida, M. and Kinosita, K. Jr. (2007) Coupling of rotation and catalysis in F(1)-ATPase revealed by single-molecule imaging and manipulation. Cell, 130:309–321. 4. Affourtit, C., Quinlan, C. L. and Brand. M. D. (2012) Measurement of Proton Leak and Electron Leakin Isolated Mitochondria. Methods in molecular biology, 11:165-182. 5. Akerman, K. E. and Wikström, M. K. (1976) Safranine as a probe of the mitochondrial membrane potential. FEBS Letters, 68(2):191-7. 6. Alfonzo, M., Kandrach, M. A. and Racker, E. (1981) Isolation, characterization, and reconstitution of a solubilized fraction containing the hydrophobic sector of the mitochondrial proton pump. Journal of Bioenergetics and Biomembranes, 13:375-391. 7. Allegretti, M., Klusch, N., Mills, D. J., Vonck, J., Kühlbrandt, W. and Davies, K. M. (2015) Horizontal membrane-intrinsic α-helices in the stator a-subunit of an F-type ATP synthase. Nature, 521(7551):237-40. 8. Andrianaivomananjaona, T., Moune-Dimala, M., Herga, S., David, V. and Haraux, F. (2011) How the N-terminal extremity of Saccharomyces cerevisiae IF1 interacts with ATP synthase: a kinetic approach. Biochimica et Biophysica Acta, 1807(2):197-204. 9. Arnold, S., Pfeiffer, K., Neupert, W., Stuart, R. A. and Schägger, H. (1998) Yeast mitochondrial F1F0-ATP synthase exists as a dimer: identification of three dimer-specific subunits. The EMBO Journal, 17:7170-7178. 112 10. Arselin, G., Vaillier, J., Graves, P. V. and Velours, J. (1996) ATP synthase of yeast mitochondria. Isolation of the subunit h and disruption of the ATP14 gene. The Journal of Biological Chemistry, 271:20284-20290. 11. Arselin, G., Vaillier, J., Salin, B., Schaeffer, J., Giraud, M. F., Dautant, A., Brèthes, D. and Velours, J. (2004) The modulation in subunits e and g amounts of yeast ATP synthase modifies mitochondrial cristae morphology. The Journal of Biological Chemistry, 279:40392 – 40399. 12. Bisetto, E., Di Pancrazio, F., Simula, M. P., Mavelli, I. and Lippe, G. (2007) Mammalian ATPsynthase monomer versus dimer profiled by blue native PAGE and activity stain. Electrophoresis, 28:3178-3185. 13. Bölker, M., Urban, M. and Kahmann, R. (1992) The a mating type locus of Ustilago maydis specifies cell signaling components. Cell, 68(3):441-50. 14. Bölker, M. Ustilago maydis – a valuable model system for the study of fungal dimorphism and virulence (2001) Microbiology, 147: 1395- 1401 15. Bonnefoy, N. and Fox, T. D. (2002) Genetic transformation of Saccharomyces cerevisiae mitochondria. Methods and Enzymology, 350:97-111. 16. Bornhövd, C., Vogel, F., Neupert, W. and Reichert, A. S. (2006) Mitochondrial membrane potential is dependent on the oligomeric state of F1F0-ATP synthase supracomplexes. The Journal of biological chemistry, 281(20):13990- 8. 17. Bottin, A., Kamper, J. and Kahmann, R. (1996) Isolation of a carbon source- regulated gene from Ustilago maydis. Molecular & General Genetics, 253: 342- 352. 18. Boyer, P. D. and Kohlbrenner, W. E. (1981) The present status of the binding change mechanism and its relation to ATP formation by chloroplasts. In: Selman BR, Selman-Peimer S (eds) Energy coupling in photosynthesis. Elsevier, Amsterdam, pp 231-240. 19. Boyer, P. D. (2000) Catalytic site forms and controls in ATP synthase catalysis. Biochimica et Biophysica Acta, 1458:252-262. 113 20. Bornhövd, C., Vogel, F., Neupert, W. and Reichert, A. S. (2006) Mitochondrial membrane potential is dependent of the oligomeric state of F1F0-ATP synthase supracomplexes. The Journal of biological chemistry, 281:13990-13998. 21. Brachmann, A., König, J., Julius, C. and Feldbrügge, M. (2004) A reverse genetic approach for generating gene replacement mutants in Ustilago maydis. Molecular Genetics and Genomic, 272:261-226. 22. Bravo, C., Minauro-Sanmiguel, F., Morales-Rios, E., Rodríguez-Zabala J. S. and Garcia, J. J. (2004) Overexpression of the inhibitor protein IF(1) in AS-30D hepatoma produces a higher association with mitochondrial F(1)F(0) ATP synthase compared to normal rat liver: functional and cross-linking studies. Journal of Bioenergetics and Biomembranes, 36:257-64. 23. Bustos, M. D. and Velours, J. (2005) The modification of the conserved GXXXG motif of the membrane-spanning segment of subunit g destabilizes the supramolecular species of yeast ATP synthase. The Journal of biological chemistry, 280(32):29004-10. 24. Buzhynskyy, N., Sens, P., Prima, V., Sturgis, J. N. and Scheuring, S. (2007) Rows of ATP synthase dimers in native mitochondrial inner membranes. Biophysical Journal, 93:2870-2876. 25. Cabezón, E., Arechaga, I., Jonathan, P., Butler, G. and Walker, J. E. (2000) Dimerization of bovine F1-ATPase by binding the inhibitor protein, IF1. The Journal of Biological Chemistry, 275:28353-28355. 26. Cabezón, E., Montgomery, M., G., Leslie, A., G. and Walker, J.E. (2003) The structure of bovine F1-ATPase in complex with its regulatory protein IF1. PNAS, 10:744-750. 27. Campanella, M., Casswell, E., Chong, S., Farah, Z., Wieckowski, M. R., Abramov, A. Y., Tinker, A. and Duchen, M. R. (2008) Regulation of Mitochondrial Structure and Function by the F1F0-ATPase Inhibitor Protein, IF1. Cell Metabolism, 8(1):13-25. 28. Cano-Estrada, A., Vázquez-Acevedo, M., Villavicencio-Queijeiro, A., Figueroa- Martínez, F., Miranda-Astudillo, H., Cordeiro, Y., Mignaco, J. A., Foguel, D., Cardol, P., Lapaille, M., Remacle, C., Wilkens, S. and González-Halphen, D. 114 (2010) Subunit-subunit interactions and overall topology of the dimeric mitochondrial ATP synthase of Polytomella sp. Biochimica et Biophysica Acta, 1797:1439-1448. 29. Cárdenas-Monroy, C. A., Pohlmann, T., Piñón-Zárate, G., Matus-Ortega, G., Guerra, G., Feldbrügge, M. and Pardo, J. P. (2017) The mitochondrial alternative oxidase Aox1 is needed to cope with respiratory stress but dispensable for pathogenic development in Ustilago maydis. PLoS One, 12(3):e0173389. doi: 10.1371/journal.pone.0173389. 30. Cardol, P., González-Halphen, D., Reyes-Prieto, A., Baurain, D., Matagne, R. F. and Remacle, C. (2005) The mitochondrial oxidative phosphorylation proteome of Chlamydomonas reinhardtii deduced from the Genome Sequencing Project. Plant Physiology, 137:447-489. 31. Cardol, P., Figueroa, F., Remacle, C., Franzén, L. G. and González-Halphen, D. Oxidative phosphorylation: building blocks and related components. David B. Stern (Ed.), The Chlamydomonas Sourcebook, Organellar and Metabolic Processes, Volume 2, Elsevier Inc., San Diego (2009), pp. 469-502. 32. Chance, B. and Williams, G. R. (1957) The respiratory chain and oxidative phosphorylation. Advances in enzymology and related subjects of biochemistry, 17:65-134. 33. Chávez, E., Flores-Herrera, O. y Martínez, F. Capítulo 22, Ciclo de los ácidos tricarboxílicos, Bioquímica de Laguna, 7ª edición, Laguna, J., Piña, E., Martínez, F., Pardo, J. P. y Riveros, H. (Editores), Manual Moderno, 2013. 34. Chávez, E., Flores-Herrera, O. y Martínez, F. Capítulo 23, Oxidaciones Biológicas y Bioenergética, Bioquímica de Laguna, 7ª edición, Laguna, J., Piña, E., Martínez, F., Pardo, J. P. y Riveros, H. (Editores), Manual Moderno, 2013. 35. Cogliati, S., Enriquez, J. A. and Scorrano, L. (2016) Mitochondrial Cristae: Where Beauty Meets Functionality. Trends in biochemical sciences, 41(3):261- 273. 36. Colina-Tenorio, L., Miranda-Astudillo, H., Cano-Estrada, A., Vázquez- Acevedo, M., Cardol, P., Remacle, C. and González-Halphen, D. (2016) 115 Subunit Asa1 spans all the peripheral stalk of the mitochondrial ATP synthase of the chlorophycean alga Polytomella sp. Biochimica et Biophysica Acta, 1857(4):359-369. 37. Colina-Tenorio, L., Dautant, A., Miranda-Astudillo, H., Giraud, M. F. and González-Halphen, D. (2018) The Peripheral stalk of rotary ATPases. Frontiers in Physiology, doi: 10.3389/fphys.2018.01243. 38. Couoh-Cardel, S., Uribe-Carvajal, S., Wilkens, S. and García-Trejo, J. J. (2010) Structure of dimeric F1F0-ATP synthase. The Journal of biological chemistry, 285(47):36447-36455. 39. Cox, G.B., Jans, D. A., Fimmel, A.L., Gibson, F. and Hatch, L. (1984) Hypothesis. The mechanism of ATP synthase. Conformational change by rotation of the beta-subunit. Biochimica et Biophysica Acta, 768:201–208. 40. De los Rios-Castillo, D., Zarco-Zavala, M., Olvera-Sanchez, S., Pardo, J. P., Juarez, O., Martinez, F., Mendoza-Hernandez, G., García-Trejo, J. J. and Flores-Herrera, O. (2011) Atypical cristae morphology of human syncytiotrophoblast mitochondria: role for complex V. he Journal of biological chemistry, 286(27):23911-23919. 41. Devenish, R. J., Prescott, M. and Rodgers, A. J. (2008) The structure and function of mitochondrial F1F0-ATP synthases. International review of cell and molecular biology, 267:1-58. 42. Díaz-Ruiz, R., Avéret, N., Araiza, D., Pinson, B., Uribe-Carvajal, S., Devin, A. and Rigoulet, M. (2008) Mitochondrial oxidative phosphorylation is regulated by fructose 1,6-bisphosphate. A possible role in Crabtree effect induction? The Journal of biological chemistry, 283(40):26948-26955. 43. Dienhart, M., Pfeiffer, K., Schagger, H. and Stuart, R. A. (2002) Formation of the yeast F1F0-ATP synthase dimeric complex does not require the ATPase inhibitor protein, Inh1. The Journal of Biological Chemistry, 277:39289-39295. 44. Doehlemann, G., Wahl, R., Horst, R. J., Voll, L. M., Usadel, B., Poree, F., Stitt, M., Pons-Kühnemann, J., Sonnewald, U., Kahmann, R. and Kämper, J. (2008) Reprogramming a maize plant: transcriptional and metabolic changes induced 116 by the fungal biotroph Ustilago maydis. The Plant journal: for cell and molecular biology, 56(2):181-195. 45. Dudkina, N. V., Sunderhaus, S., Braun, H. P. and Boekema, E. J. (2006) Characterization of dimeric ATP synthase and cristae membrane ultrastructure from Saccharomyces and Polytomella mitochondria. FEBS Letters, 580:3427- 3432. 46. Duvezin-Caubet, S., Rak, M., Lefebvre-Legendre, L., Tetaud, E., Bonnefoy, N. and di Rago, J. P. (2006) A ‘petite-obligate’ mutant of Saccharomyces cerevisiae: Functional mtDNA is lethal in cells lacking the δ subunit of mitochondrial F1-ATPase. The Journal of Biological Chemistry, 281:16305- 16313. 47. Ebner, E. and Maier, K. L. (1977) A protein inhibitor of mitochondrial adenosine triphosphatase (F1) from Saccharomyces cerevisiae. The Journal of biological chemistry, 252(2):671-676. 48. El-Khoury, R., Dufour, E., Rak, M., Ramanantsoa, N., Grandchamp, N., Csaba, Z., Duvillié, B., Bénit, P., Gallego, J., Gressens, P., Sarkis, C., Jacobs, H. T. and Rustin, P. (2013) Alternative Oxidase Expression in the Mouse Enables Bypassing Cytochrome C Oxidase Blockade and Limits Mitochondrial ROS Overproduction. PLoS genetics, 9(1):e1003182. 49. El-Khoury, R., Kemppainen, K. K., Dufour, E., Szibor, M., Jacobs, H. T. and Rustin, P. (2014) Engineering the Alternative Oxidase Gene to Better Understand and Counteract Mitochondrial Defects: State of the Art and Perspectives. PLoS genetics, 171(8):2243-9. 50. Esparza-Perusquía, M., Olvera-Sánchez, S., Pardo, J. P., Mendoza- Hernández, G., Martínez, F. and Flores-Herrera O. (2017) Structural and kinetics characterization of the F1F0-ATP synthase dimer. New repercussion of monomer-monomer contact. Biochimica et Biophysica Acta -Bioenergetics. 1858(12):975-981. 51. Esparza-Perusquía, M., Langner, T., García-Cruz, G., Feldbrügge, M., Zavala, G., Pardo, J. P., Martínez, F. and Flores-Herrera, O. Deletion of subunit g 117 affects the stability of the dimeric state of the F1FO-ATP synthase and decreases the mitochondrial ATP synthesis. En preparación. 52. Estambrook, R. W. (1967) Mitochondrial respiratory control and the polarographic measurement of ADP: O ratios. Methods and Enzymology, 10:41-47. 53. Feinberg, A. P. and Vogelstein, B. (1983) A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Analytical biochemistry, 132(1):6-13. 54. Feldbrügge, M., Kämper, J., Steinberg, G. and Kahmann, R. (2004) Regulation of mating and pathogenic development in Ustilago maydis. Current opinion in microbiology, 7(6):666-672. 55. Fiske, C. H. and Subbarow, Y. (1925) The Colorimetric Determination Of Phosphorus. The Journal of Biological Chemistry, 66:375-400. 56. Frey, T. G., Renken, C. W. and Perkins, G. A. (2002) Insight into mitochondrial structure and function from electron tomography. Biochimica et Biophysica Acta, 1555(1-3):196-203. 57. Funes, S., Davidson, E., Claros, M. G., van Lis, R., Pérez-Martínez, X., Vázquez-Acevedo, M., King, M. P. and González-Halphen, D. (2002) The typically mitochondrial DNA-encoded ATP6 subunit of F1F0-ATPase is ecoded by a nuclear gene in Chlamydomonas reinhaedtii. The Journal of Biological Chemistry, 277:6051-6058. 58. García, J. J., Morales-Ríos, E., Cortés-Hernández, P. and Rodríguez-Zavala, J. S. (2006) The inhibitor protein (IF1) promotes dimerization of the mitochondrial F1F0-ATP synthase. Biochemistry, 45:12695-12703. 59. Gledhill, J. R. and Walker, J. E. (2005) Inhibition sites in F1-ATPase from bovine heart mitochondria. The Biochemical journal, 386(Pt 3):591-598. 60. Giraud, M. F., Paumard, P., Soubannier, V., Vaillier, J., Arselin, G., Salin, B., Schaeffer, J., Brèthes, D., di Rago, J. P. and Velours, J. (2002) Is there a resationship between the supramolecular organization of the mitochondrial ATP synthase and the formation of cristae? Biochimica et Biophysica Acta, 1555:174-180. 118 61. Guérin, B., Labbe, P. and Somlo, M. (1979). Preparation of yeast mitochondria (Saccharomyces cerevisiae) with good P/O and respiratory control ratios. Methods Enzymology, 55:149-159. 62. Guerrero-Castillo, S., Cabrera-Orefice, A., Vázquez-Acevedo, M., González- Halphen, D. and Uribe-Carvajal, S. (2012) During the stationary groWTh phase, Yarrowia lipolytica prevents the overproduction of reactive oxygen species by activating an uncoupled mitochondrial respiratory pathway. Biochimica et Biophysica Acta, 1817:353–362. 63. Gu, J., Zhang, L., Zong, S., Guo, R., Liu, T., Yi, J., Wang, P., Zhuo, W. and Yang, M. (2019) Cryo-EM structure of the mammalian ATP synthase tetramer bound with inhibitory protein IF1. Science, 364(6445):1068-1075. 64. Guo, H., Bueler, S. A. and Rubinstein, J. L. (2017) Atomic model for the dimeric FO region of mitochondrial ATP synthase. Science, 358(6365):936-940. 65. Guo, H. and Rubinstein, J. L. (2018) Cryo-EM of ATP Synthases. Current opinion in structural biology, 52:71-79. 66. Habersetzer, J., Ziani, W., Larrieu, I., Stines-Chaumeil, C., Giraud, M. F., Brèthes, D., Dautant, A. and Paumard, P. (2013) ATP synthase oligomerization: from the enzyme models to the mitochondrial morphology. The international journal of biochemistry & cell biology, 1:99-105. 67. Hackenbrock, C. R., Chazotte, B. and Gupte, S. S. (1986) The random collision model and a critical assessment of diffusion and collision in mitochondrial electron transport. Journal of Bioenergetics and Biomembranes, 18:331–368. 68. Herrera, T., Ulloa, M. and Ruiz-Oronoz, M. (1990) El reino de los hongos. Micología básica y aplicada. Fondo de la Cultura Económica/IBT-UNAM, Mexico, D.F. pp 294-301. 69. Hirst, J. (2013). Mitochondrial complex I. Annual review of biochemistry, 82, 551–575. 70. Horst, R. J., Zeh, C., Saur, A., Sonnewald, S., Sonnewald, U. and Voll, L. M. (2012) The Ustilago maydis Nit2 homolog regulates nitrogen utilization and is required for efficient induction of filamentous groWTh. Eukaryot Cel,l 11: 368- 380. 119 71. Itoh, H., Takahashi, A., Adachi, K., Noji, H., Yasuda, R., Yoshida, M. and Kinosita, K. (2004) Mechanically driven ATP synthesis by F1-ATPase. Nature, 427:465-468. 72. Juárez, O., Guerra, G., Martínez, F. and Pardo, J. P. (2004) The mitochondrial respiratory chain of Ustilago maydis. Biochimica et Biophysica Acta, 1658(3):244-51. 73. Juárez, O., Guerra, G., Velázquez, I., Flores-Herrera, O., Rivera-Pérez, R. E. and Pardo, J. P. (2006) The physiologic role of alternative oxidase in Ustilago maydis. The FEBS Journal, 273:4603-4615. 74. Jung, C., Higgins, C. M. and Xu, Z. (2000) Measuring the quantity and activity of mitochondrial electron transport chain complexes in tissues of central nervous system using blue native polyacrylamide gel electrophoresis. Analitycal Biochemistry, 286:214-223. 75. Junge, W., Lill, H. and Engelbrecht, S. (1997) ATP synthase: an electrochemical transducer with rotatory mechanics. Trends in biochemical sciences, 22(11):420-3. 76. Kahmann, R., Basse, C. and Feldbrügge, M. (1999) Fungal-plant signalling in the Ustilago maydis-maize pathosystem. Current opinion in microbiology, 2(6):647-650. 77. Ko, Y. H., Delannoy, M., Hullihen, J., Chiu, W. and Pedersen, P. L. (2003) Mitochondrial ATP Synthasome. Cristae-enriched Membranes and a Multiwell Detergent Screening Assay Yield Dispersed Single Complexes Containing the ATP Synthase and Carriers for Pi and ADP/ATP. The Journal of Biological Chemistry, 278(14):12305-9. 78. Kühlbrandt, W. (2015) Structure and function of mitochondrial membrane protein complexes. BMC Biology, 13(1):89-99. 79. Kühlbrandt, W. and Davies, K. M. (2016) Rotary ATPases: A New Twist to an Ancient Machine. Trends in biochemical sciences, 41(1):106-116. 80. Kushner, S. R. (1978) An improved method for transformation of E. coli with ColE1 derived plasmids. In Genetic Engineering, Boyer H. B. and Nicosia S. (Amsterdam: Elsavier/North-Holland), 17–23. 120 81. Legros, F., Lombes, A., Frachon, P. and Rojo, M. (2002) Mitochondrial fusion in human cells is efficient, requires the inner membrane potential, and is mediated by mitofusins. Molecular Biology of the Cell, 13:4343-4354. 82. Lengeler, J. W. (2000) Metabolic networks: a signal-oriented approach to cellular models. Biological Chemistry, 381(9-10):911-920. 83. Letts, J. A. and Sazanov, L. A. (2017) Clarifying the supercomplex: the higher- order organization of the mitochondrial electron transport chain. Nature structural and molecular biology, 24(10):800-808 84. Letts, J. A., Fiedorczuk, K., Degliesposti, G., Skehel, M. and Sazanov, L. A. (2019) Structures of respiratory supercomplex I+III2 reveal functional and conformational crosstalk. Molecular cell, 75(6):1131-1146. 85. Levine, R. L., Garland, D., Oliver, C. N., Amici, A., Climent, I., Lenz, A. G., Ahn, B. W., Shaltiel, A. and Stadtman, E. (1990) Determination of carbonyl contents oxidatively modified proteins. Methods in enzymology, 186:464-478. 86. Liu, Y. B., Koh, C. M. J. and Ji, L. H. (2011) Bioconversion of crude glycerol to glycolipids in Ustilago maydis. Bioresource Technology, 102:3927-3933. 87. Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J. (1951) Protein measurement with the Folin phenol reagent. The Journal of biological chemistry, 193:265-275. 88. Mahuran, D., Clements, P., Carrella, M. and Strasberg, P. M. (1983) A High Recovery Method for Concentrating Microgram Quantities of Protein from Large Volumes of Solution. Analytical biochemistry, 129:513-516. 89. Mannella, C. A. (2000) Introduction: Our changing view of mitochondria. Journal of Bioenergetics and Biomembranes, 32:1-4. 90. Mannella, C. A., Pfeiffer, D. R., Bradshaw, P. C., Moraru, I. I., Slepchenko, B., Loew, L. M., Hsieh, C., Buttle, K. and Marko, M. (2001) Topology of the mitochondrial inner membrane: Dynamics and bioenergetic implications. International Union of Biochemistry and Molecular Biology, 52:93-100. 91. Mannella C. A. (2006) The relevance of mitochondrial membrane topology to mitochondrial function. Biochimica et Biophysica Acta, 1762:140-147. 121 92. Mannella, C. A., Lederer, W. J. and Jafri, M. S. (2013) The connection between inner membrane topology and mitochondrial function. International Society for Heart Research, 62:51-57. 93. Marques, I., Dencher, N.A., Videira, A. and Krause, F. (2007) Supramolecular organization of the respiratory chain in Neurospora crassa mitochondria. American Society for Microbiology, 6:2391-2405. 94. McArthur, K. and Ryan, M. T. (2020) Resolving mitochondrial cristae: introducing a new model the fold. The EMBO journal, 39(14):e105714. 95. McKee, T. and McKee J. R. Capítulo 2, Las células vivas, Bioquímica, Las bases moleculares de la vida, 4a edición, Mc Graw Hill, 2009. 96. Mendoza-Mendoza, A., Berndt, P., Djamei, A., Weise, C., Linne, U., Marahiel, M., Vranes, M., Kämper, J. and Kahmann, R. (2009) Physical-chemical plant- derived signals induce differentiation in Ustilago maydis. Molecular Microbiology, 71(4):895-911. 97. Milenkovic, D. Larsson, N. G. (2015) Mic10 Oligomerization Pinches off Mitochondrial Cristae. Cell Metabolism, 21(5):660-1. 98. Minauro-Sanmiguel, F., Bravo, C. and García, J. J. (2002) Cross-linking of the endogenous inhibitor protein (IF1) with rotor (gamma, epsilon) and stator (alpha) subunits of the mitochondrial ATP synthase. Journal of Bioenergetics and Biomembranes, 34(6):433-443. 99. Minauro-Sanmiguel, F., Wilkens, S. and García, J. J. (2005) Structure of dimeric mitochondrial ATP synthase: novel F0 bridging features and the structural basis of mitochondrial cristae biogenesis. PNAS, 102:12356-12358. 100. Miranda-Astudillo, H., Cano-Estrada, A., Vázquez-Acevedo, M., Colina- Tenorio, L., Downie-Velasco, A., Cardol, P., Remacle, C., Domínguez- Ramírez, L. and González-Halphen, D. (2014) Interactions of subunits Asa2, Asa4 and Asa7 in the peripheral stalk of the mitochondrial ATP synthase of the chlorophycean alga Polytomella sp. Biochimica et Biophysica Acta - Bioenergetics, 1837(1):1-13. 101. Mitchell, P. (1961) Coupling of phosphorylation to electron and hydrogen transfer by a chemi-osmotic type of mechanism. Nature, 191:155-148. 122 102. Morales-Ríos, E., De la Rosa-Morales, F., Mendoza-Hernández, G., Rodríguez-Zavala, J. S., Celis, H., Zarco-Zavala, M. and García-Trejo, J. J. (2010) A novel 11-kDa inhibitory subunit in the F1F0 ATP synthase of Paracoccus denitrificans and related alpha-proteobacteria. FASEB journal, 24(2):599-608. 103. Nicholls, D. G. and Ferguson, S. T. (2013) Bioenergetics 4, Academic Press, 207. 104. Orriss, G. L., Leslie, A. G., Braig, K. and Walker, J. E. (1998) Bovine F1- ATPase covalently inhibited with 4-chloro-7-nitrobenzofurazan: the structure provides further support for a rotary catalytic mechanism. Structure, 6(7):831- 7. 105. Panov, A., Dikalov, S., Shalbuyeva, N., Hemendinger, R., Greenamyre, J. T. and Rosenfeld, J. (2007) Species- and tissue-specific relationships between mitochondrial permeability transition and generation of ROS in brain and liver mitochondria of rats and mice. Am. J. Physiol. Cell. Physiol. 292, C708–C718 106. Paumard, P., Vaillier, J., Coulary, B., Schaeffer, J., Soubannier, V., Mueller, D. M., Brèthes, D., Di Rago, J. P. and Velours, J. (2002) The ATP synthase is involved in generating mitochondrial cristae morphology. The EMBO Journal, 21:221-230. 107. Penefsky, H. S., Pullman, M. E., Datta, A. and Racker, E. (1960) Partial resolution of the enzymes catalyzing oxidative phosphorylation. II. Purification and properties of soluble dinitrophenol-stimulated adenosine triphosphatase. The Journal of Biological Chemistry, 235:3322–3329. 108. Penefsky, H. S. (1974) Mitochondrial and chloroplast ATPases. The Enzymes, 10:375-429. 109. Pullman, M. E. and Monroy, G.C. (1963) A naturally occurring inhibitor of mitochondrial adenosine triphosphate. The Journal of Biological Chemistry, 238:3762-3769. 110. Rabl ,R., Soubannier, V., Scholz, R., Vogel, F., Mendl, N., Vasiljev-Neumeyer, A., Körner, C., Jagasia, R., Keil, T., Baumeister, W., Cyrklaff, M., Neupert, W. and Reichert, A. S. (2009) Formation of cristae and crista junction in 123 mitochondria depends on antagonism between Fcj and su e/g. The Journal of cell biology, 185:1047-1063. 111. Racker, E. (1963) A mitochondrial factor conferring oligomycin sensitivity on soluble mitochondrial ATPase. Biochemical and biophysical research communications, 10:435-439. 112. Reifschneider, N., Goto, S., Nakamoto, H., Takahashi, R., Sugawa, M., Dencher, N. A. and Krause, F. (2005) Defining the mitochondrial proteomes from five rat organs in a physiologically significant context using 2D blue- native/SDS-PAGE. Journal of proteome research, 5:1117-1132. 113. Reyes-Galindo, M., Suarez, R., Esparza-Perusquía, M., de Lira-Sánchez, J., Pardo, J. P., Martínez, F., and Flores-Herrera O. (2019) Mitochondrial respirasome works as a single unit and the crosstalk between complexes I, III2 and IV stimulates NADH dehydrogenase activity. Biochimica et Biophysica Acta -Bioenergetics. 1860(8):618-627. 114. Rexroth, S, Tittingdorf, J. M. W. M. Z., Schwassmann, H. J., Krause, F., Seelert, H. and Dencher, N. A. (2004) Dimeric H+-ATPsynthase in the chloroplast of Chlamydomonas reinhardtii. Biochimica et Biophysica Acta, 1658:202-211. 115. Robb, E. L., Hall, A. R., Prime, T. A., Eaton, S., Szibor, M., Viscomi, C., James, A. M. and Murphy, M. P. (2018) Control of Mitochondrial Superoxide Production by Reverse Electron Transport at Complex I. The Journal of biological chemistry, 293(25):9869-9879. 116. Rogov, A. G., Sukhanova, E. I., Uralskaya, L. A., Aliverdieva, D. A. and Zvyagilskaya, R. A. (2014) Alternative oxidase: distribution, induction, properties, structure, regulation, and functions. Biochemistry (Moscow), 79(13):1615-34. 117. Rybicki, E. P. and Purves, M. (1996) Western or enzyme-assisted immunoelectro-blotting (IEB). 118. Saddar. S and Stuart, R. A. (2005) The yeast F(1)F(0)-ATP synthase: analysis of the molecular organization of subunit g and the importance of a conserved GXXXG motif. The Journal of biological chemistry, 280(26):24435-42. 124 119. Saddar, S., Dienhart, M. K. and Stuart, R. A. (2008) The F1F0-ATP synthase complex influences the assembly state of the cytochrome bc1-cytochrome oxidase supercomplexes and its association with TIM23 machinery. The Journal of biological chemistry, 283:6677-6686. 120. Sambrook, J. and D. W. Russell, 2001 Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratories, New York. 121. Sánchez-Vásquez, L., Vázquez-Acevedo, M., de la Mora, J., Vega-de Luna, F., Cardol, P., Remacle, C., Dreyfus, G. and González-Halphen, D. (2017) Near-neighbor interactions of the membrane-embedded subunits of the mitochondrial ATP synthase of a chlorophycean alga. Biochimica et Biophysica Acta, 1858(7):497-509. 122. Sazanov, L. A. (2015). A giant molecular proton pump: structure and mechanism of respiratory complex I. Nature reviews. Molecular cell biology, 16(6):375-88. 123. Schägger, H. and von Jagow, G (1991) Blue native electrophoresis for isolation of membrane protein complexes in enzymatically active form. Analytical biochemistry, 199(2):223-231. 124. Schägger, H. (2000) Supercomplexes in the respiratory chains of yeast and mammalian mitochondria. The EMBO journal, 19(8):1777-83. 125. Segel, I. H. (1993) Enzyme kinetics: behavior and analysis of rapid equilibrium and steady-state enzyme systems. New York: Wiley-Interscience. 126. Song, D. H., Park, J., Philbert, M. A., Sastry, A. M. and Lu, W. (2014) Effects of local pH on the formation and regulation of cristae morphologies. Physical review. E, Statistical, nonlinear, and soft matter physics, 90(2):022702. 127. Southern, E. M. (1975) Detection of specific sequences among DNA fragments separated by gel electrophoresis. Journal of molecular biology, 98(3):503-517. 128. Strauss, M., Hofhaus, G., Schröder, R. R. and Kühlbrandt, W. (2008) Dimer ribbons of ATP synthase shape the inner mitochondrial membrane. The EMBO Journal, 27:1154-1160. 125 129. Thomas, D., Bron, P., Weimann, T., Dautant, A., Giraud, M.F., Paumard, P., Salin, B., Cavalier, A., Velours, J. and Brèthes, D. (2008) Supramolecular organization of the yeast F1F0-ATP synthase. Biology of the Cell, 100:591-601. 130. Tomasetig, L., Di Pancrazio, F., Harris, D. A., Mavelli., I., Lippe, G. (2002) Dimerization of F0F1ATP synthase from bovine heart is independent from the binding of the inhibitor protein IF1. Biochimica Biophysica Acta, 1556, 133-141. 131. Trautschold, I., Lamprecht, W. and Schweitzer, G. (1995) UV method with hexokinase and glucose-6-phosphate dehydrogenase. Method of enzymatic analysis, Bergmeyer, 7:346-357. 132. van Lis, R., Atteia, A., Mendoza-Hernández, G. and González-Halphen, D. (2003) Identification of novel mitochondrial protein components of Chlamydomonas reinhardtii. A proteomic approach. Plant Physiology, 132:318-330. 133. van Lis, R., Mendoza-Hernández, G., Groth, G. and Atteia, A. (2007) New insights into the unique structure of the F0F1-ATP synthase from the chlamydomonad algae Polytomella sp. and Chlamydomonas reinhardtii. Plant Physiology, 144:1190-1199. 134. Vázquez-Acevedo, M., Cardol, P., Cano-Estrada, A., Lapaille, M., Remacle, C. and González-Halphen, D. (2006) The mitochondrial ATP synthase of chlorophycean algae contains eight subunits of unknown origin involved in the formation of an atypical stator-stalk and in the dimerization of the complex. Journal of bioenergetics and biomembranes, 38:271-282. 135. Velours, J., Dautant, A., Salin, B., Sagot, I. and Brèthes, D. (2009) Mitochondrial F1F0-ATP synthase and organellar internal architecture. The Internal Journal of Biochemistry and Cell Biology, 41:1783-1789. 136. Villavicencio-Queijeiro, A., Vázquez-Acevedo, M., Cano-Estrada, A., Zarco- Zavala, M., Tuena de Gómez, M., Mignaco, J. A., Freire, M. M., Scofano, H. M., Foguel, D., Cardol, P., Remacle, C. and González-Halphen, D. (2009) The fully-active and structurally stable form of the mitochondrial ATP synthase of Polytomella sp. is dimeric. Journal of bioenergetics and biomembranes, 41:1- 13. 126 137. Villavicencio-Queijeiro, A., Pardo, J. P. and González-Halphen, D. (2015) Kinetic and hysteretic behavior of ATP hydrolysis of the highly stable dimeric ATP synthase of Polytomella sp. Archives of biochemistry and biophysics, 575:30-37. 138. Vollmeister, E., Schipper, K., Baumann, S., Haag, C., Pohlmann, T., Stock, J. and Feldbrügge, M. (2012) Fungal development of the plant pathogen Ustilago maydis. FEMS microbiology reviews, 36(1):59-77. 139. Vonk, J. and Schäfer, E. (2009) Supramolecular organization of protein complexes in the mitochondrial inner membrane. Biochimica et Biophysica Acta, 1793:117-124. 140. Wagner, K., Perschil, I., Fichter, C. D. and van der Laan, M. (2010) Stepwise assembly of dimeric F1F0-ATP synthase in mitochondria involves the small F0- subunits k and i. Molecular biology of the cell, 9:1494-1504. 141. Walker, J. E. (1998) ATP Synthesis by Rotary Catalysis (Nobel lecture). Angewandte Chemie International Edition, 37:2308-2319. 142. Walker, J. E. and Dickson, V.K. (2006) The peripheral stalk of the mitochondrial ATP synthase. Biochimica et Biophysica Acta, 1757:286–296. 143. Waterborg, J. and Matthews, H. (1984) The Lowry method for protein quantitation. The Protein Protocols Handbook, DOI: 10.1385/1-59259-169-8:7. 144. Watt, I. N., Montgomery, M. G., Runswick, M. J., Leslie, A. G. and Walker, J. E. (2010) Bioenergetic cost of making an adenosine triphosphate molecule in animal mitochondria. PNAS, 107:16823-16827. 145. Weber, J. and Senior, A. E. (2003) ATP synthesis driven by proton transport in F1F0-ATP synthase. FEBS Letters, 545:61-70. 146. Wittig, I., Braun, H. P. and Schägger, H. (2006) Blue native PAGE. Nature protocols, 1(1):418-428. 147. Wittig, I., Karas, M. and Schägger, H. (2007) High resolution clear native electrophoresis for in-gel functional assays and fluorescence studies of membrane protein complexes. Molecular & cellular proteomics, 6:1215-1225. 127 148. Wittig, I., Velours, J., Stuart, R. and Schägger, H. (2008) Characterization of domain interfases in monomeric and dimeric ATP synthase. Molecular and cellular proteomics, 5:995-1004. 149. Wittig. I. and Schägger, H. (2008) Structural organization of mitochondrial ATP synthase. Biochimica et Biophysica Acta, 1777:592-598. 150. Wittig, I., Meyer, B., Heide, H., Steger, M., Bleier. L., Wumaier. Z., Karas, M. And Schägger, H. (2010) Assembly and oligomerization of human ATP synthase lacking mitochondrial subunits a and A6L. Biochimica et Biophysica Acta, 1797:1004-1011. 151. Wurm, C. A. and Jakobs, S. Differential protein distributions define two sub- compartments of the mitochondrial inner membrane in yeast. FEBS letters, 580(24):5628-34. 152. Yoshida, M., Muneyuki, E. and Isabori, T. (2001) ATP synthase — a marvellous rotatory engine of the cell. Nature reviews. Molecular cell biology, 2:669-677. 153. Zerbetto, E., Vergani, L. and Dabbeni-Sala, F. (1997) Quantification of muscle mitochondrial oxidative phosphorylation enzymes via histochemical staining of blue native polyacrylamide gels. Electrophoresis, 18(11):2059-2064. 154. Zick, M., Rabl, R. and Reichert, A.S. (2009) Cristae formation-linking ultrastructure and function of mitochondria. Biochimica et Biophysica Acta, 1793:5-19. 128 “The Classical Organell I- Mitochondria” Odra Noel Anexos 129 10. ANEXOS 10.1. Oligonucleótidos usados para la formación del Knock-out para la transformación de Ustilago maydis. Tabla 4. Oligonucleótidos utilizados para generar el plásmido knock-out y para verificar la cepa mutante Dg. Gen Secuencia del oligonucleótido ATP20 (sub g) U1 U2 U3 D1 D2 D3 P1 P2 5ʹ- GCGTAGTCGATGTCCTTGG -3ʹ 5ʹ- ATTTAAATGCTTCCTTGTATTCGGC -3ʹ 5ʹ- GGCCATCTAGGCCGATGACCGTATTACCCGAAAGAC -3ʹ 5ʹ- GGCCTGAGTGGCCACGCTCGACAATTGAATTCG -3ʹ 5ʹ- ATTTAAATCTGGCATGTGCTCACC -3ʹ 5ʹ- GGCCTGCCGTATCAAGTC -3ʹ 5ʹ- TCCTCACACCATCCCCTT -3ʹ 5ʹ- GGGAGCAGCAGCTCGGC -3ʹ Secuenciar M13F M13R MF167 MF168 5ʹ- GTTTTCCCAGTCACGAC -3ʹ 5ʹ- CAGGAAACAGCTATGAC -3ʹ 5ʹ- AACTCGCTGGTAGTTACCAC -3ʹ 5ʹ- ACTAGATCCGATGATAAGCTG -3ʹ Las letras en negrita indican el sitio de corte para la enzima SwaI o SfiI. 10.2. Transformación por biobalística. La trasformación para generar la cepa ∆g se realizó como lo reporta Bonnefoy et al., (2002). Se cultiva la cepa WT en 30 mL de medio YPD durante 24 horas a 28°C, se toman 500 µL del cultivo para inocular 50 mL de YPD y se cultivan durante 24 horas a 28°C en agitación orbital constante. Las células se centrifugan a 2300 g por 5 min y el paquete celular se resuspende en 600 µL de medio YPD. Las células se extienden en 6 cajas de Petri con YPD-Agar suplementado con higromicina (50 µg higromicina B/mL de PBS) y se dejan secar por 3 horas a temperatura ambiente en esterilidad. 130 Preparación de los microproyectiles y precipitación del DNA. Se colocan 50 mg de micropartículas de tungsteno en un tubo de 1.5 mL y se esterilizan incubando por 10 min con metanol al 70% (p/v). Se centrifugan a 13,200 g por 15 min y se elimina cuidadosamente el sobrenadante. Las partículas se lavan con agua estéril y se resuspenden hasta una concentración de 50 mg/mL en 50% (v/v) glicerol y se mantienen en hielo. Posteriormente se mezclan 5 µg del pUMa_1704 con 100 µg de la suspensión de partículas de tungsteno, 4 µL de espermidina 1 M (Cf= 26.66 mM) y 100 µL de CaCl2 2.5 M (Cf= 1.66 M). Se agitan en vortex inmediatamente después de la adición de cada reactivo. Se adiciona 20 µL de la suspensión en cada uno de los microproyectiles. Bombardeo de células. El bombardeo se realiza en un Biolistic Gun (PDS- 1000He BioRad) esterilizado con metanol al 70% (v/v) y siguiendo las instrucciones del fabricante. Se bombardean las partículas a 1100 psi, considerando un vacío de 29 a 29.5 pulgadas de mercurio. Una vez realizado el bombardeo, las placas se incuban de 5 a 7 días a 28°C hasta que se observe el crecimiento de las colonias. 10.3. Solución de Sales. El buffer 1x de solución de sales está compuesto por 117 mM de KH2PO4, 28.2 mM de NaSO4, 107 mM de KCl, 16.6 mM de MgSO4, 9 mM de CaCl2. 10.4. Elementos traza. El buffer 1x de elementos traza está compuesto por 0.97 mM de H3BO3, 0.7 mM de MnCl2-4H2O, 2.9 mM de ZnCl2, 0.18 mM de NaMoO4-2H2O, 0.37 mM de FeCl3-6H2O y 1.6 mM CuSO4-5H2O, pH 7.0. 10.5. Cuantificación de la cantidad de F1FO-ATP sintasa en las muestras V2WT, V1WT y V1Dg. Para cuantificar el contenido de F1FO-ATP sintasa presente en las muestras V2WT, V1WT y V1Dg (Figura S1), se resolvieron 10 µg de proteína de cada muestra mediante SDS-Tricina-PAGE y se tiñeron con azul brillante de Coomassie® R250. Se utilizó BSA como estándar en el mismo gel y su concentración se determinó 131 mediante espectroscopía utilizando su coeficiente de extinción molar (e = 6.58 mM- 1cm-1 para el 1% (p/v) BSA). La intensidad de la tinción con azul de Coomassie® R250 de la curva estándar de BSA fue lineal en el intervalo de concentración de proteína utilizada. El gel fue escaneado y la intensidad de la señal de las subunidades a y b del complejo V fue determinada por el software MyImageAnalysis 2.0 (ThermoScientific). El mol (o la cantidad de moles) de cada subunidad se determinó usando el peso molecular de la proteína madura (https://www.genome.jp/kegg/kegg2.html) y la estequiometría reportada para el complejo V de S. cerevisiae. 132 Figura S1. SDS-Tricina-PAGE de V1 y V2. Las proteínas V1 (A), V2 (B) y V2Dg (C) (10 µg por carril) se resolvieron mediante geles de Tricina-SDS-PAGE y se tiñeron con Coomassie® Blue. La intensidad de la subunidad a y b del complejo V se usó para la cuantificación de la concentración de F1FO-ATP sintasa usando una curva de BSA. i, ii, iii, y iv fueron muestras de V1 y V2, de diferentes preparaciones. 133 Tabla 5. Cuantificación del contenido de las subunidades a y b del V1WT de la F1FO-ATP sintasa (4 diferentes muestras). BSA (curva estándar) Señal total 1.0 µg 5.474 Regresión lineal: r2 = 0.979; b = 0.185; m = 5.04 0.5 µg 2.140 0.3 µg 1.728 0.1 µg 0.981 V1, muestra i Subunidad Señal total µg mol proporción a/b a 2.210 0.4 7.27 x 10-12 0.8 b 2.490 0.458 9.08 x 10-12 Promedioa = 8.18 x 10-12 V1, muestra ii a 1.830 0.326 5.92 x 10-12 0.8 b 2.110 0.382 7.57 x 10-12 Promedioa = 6.75 x 10-12 V1, muestra iii a 1.690 0.298 5.41 x 10-12 0.7 b 2.040 0.368 7.29 x 10-12 Promedioa = 6.35 x 10-12 Los datos se obtuvieron del análisis de las proteínas teñidas en gel mostradas en la Figura S1 = Dado que a/b tienen una estequiometría 1:1 en la F1FO-ATP sintasa nativa, el valor para a y b se enfocó como el promedio de sus valores independientes. 134 Tabla 6. Cálculo del contenido de F1FO-ATP sintasa en diferentes muestras de V1WT aislados. Subunidad MW subunidad madura (Da) Muestra i Muestra ii Muestra iii pmol de subunidad pmol total µga pmol de subunidad pmol total µga pmol de subunidad pmol total µga a (x3) 55047 2.72 8.18 0.450 2.25 6.75 0.372 2.12 6.35 0.350 b (x3) 50465 2.72 8.18 0.412 2.25 6.75 0.341 2.12 6.35 0.320 b (x2) 22281 2.72 5.45 0.121 2.25 4.50 0.100 2.12 4.24 0.010 g (x1) 33157 2.72 2.72 0.090 2.25 2.25 0.075 2.12 2.12 0.070 a (x1) 27742 2.72 2.72 0.075 2.25 2.25 0.062 2.12 2.12 0.059 OSCP (x1) 21001 2.72 2.72 0.057 2.25 2.25 0.047 2.12 2.12 0.045 d (x1) 15552 2.72 2.72 0.042 2.25 2.25 0.035 2.12 2.12 0.033 f (x1) 7371 2.72 2.72 0.020 2.25 2.25 0.017 2.12 2.12 0.016 d (x1) 14615 2.72 2.72 0.039 2.25 2.25 0.033 2.12 2.12 0.031 h (x1) 10419 2.72 2.72 0.028 2.25 2.25 0.023 2.12 2.12 0.022 e (x1) 7538 2.72 2.72 0.021 2.25 2.25 0.017 2.12 2.12 0.016 Inh1 (x1) 7281 2.72 2.72 0.019 2.25 2.25 0.016 2.12 2.12 0.015 c (x12) 7394 2.72 32.6 0.244 2.25 27.0 0.200 2.12 25.4 0.188 i/j (x1) 6454 2.72 2.72 0.018 2.25 2.25 0.015 2.12 2.12 0.014 8 (x1) 5713 2.72 2.72 0.016 2.25 2.25 0.013 2.12 2.12 0.012 g (x0) 17300 ― ― ― ― ― ― ― ― ― e (x0) 9240 ― ― ― ― ― ― ― ― ― Total 606 669 87.8 1.652 69.8 1.366 65.7 1.201 a = El valor en µg fue calculado como (MW de la subunidad madura) x (moles totales). 135 Tabla 7. Cuantificación del contenido de las subunidades a y b del V2WT de la F1FO-ATP sintasa (4 diferentes muestras). BSA (curva estándar) Señal total 1.0 µg 6.076 Regresión lineal: r2 = 0.999; b = 0.242; m = 5.78 0.5 µg 3.038 0.3 µg 1.930 0.1 µg 0.909 V2, muestra i Subunidad Señal total µg mol Proporción a/b a 1.569 0.2295 4.169 x 10-12 0.9 b 1.635 0.2409 4.77 x 10-12 Promedioa = 4.47 x 10-12 V2, muestra ii a 1.143 0.1558 2.83 x 10-12 0.8 b 1.320 0.1865 3.7 x 10-12 Promedioa = 3.27 x 10-12 V2, muestra iii a 2.192 0.337 6.12 x 10-12 1.4 b 1.524 0.2217 4.39 x 10-12 Promedioa = 5.26 x 10-12 Los datos se obtuvieron del análisis de las proteínas teñidas en gel mostradas en la Figura S1 = Dado que a/b tienen una estequiometría 1:1 en la F1FO-ATP sintasa nativa, el valor para a y b se enfocó como el promedio de sus valores independientes. 136 Tabla 8. Cálculo del contenido de F1FO-ATP sintasa en diferentes muestras de V2WT aislados. Subunidad MW subunidad madura (Da) Muestra i Muestra ii Muestra iii pmol de subunidad pmol total µga pmol de subunidad pmol total µga pmol de subunidad pmol total µga a (x3) 55047 1.49 4.47 0.246 1.09 3.27 0.180 1.75 5.26 0.290 b (x3) 50465 1.49 4.47 0.226 1.09 3.27 0.165 1.75 5.26 0.265 b (x2) 22281 1.49 2.98 0.066 1.09 2.18 0.049 1.75 3.50 0.078 g (x1) 33157 1.49 1.49 0.049 1.09 1.09 0.036 1.75 1.75 0.058 a (x1) 27742 1.49 1.49 0.041 1.09 1.09 0.030 1.75 1.75 0.049 OSCP (x1) 21001 1.49 1.49 0.031 1.09 1.09 0.023 1.75 1.75 0.037 d (x1) 15552 1.49 1.49 0.023 1.09 1.09 0.017 1.75 1.75 0.027 f (x1) 7371 1.49 1.49 0.011 1.09 1.09 0.008 1.75 1.75 0.013 d (x1) 14615 1.49 1.49 0.022 1.09 1.09 0.016 1.75 1.75 0.026 h (x1) 10419 1.49 1.49 0.016 1.09 1.09 0.011 1.75 1.75 0.018 e (x1) 7538 1.49 1.49 0.011 1.09 1.09 0.008 1.75 1.75 0.013 Inh1 (x1) 7281 1.49 1.49 0.011 1.09 1.09 0.008 1.75 1.75 0.013 c (x12) 7394 1.49 17.9 0.132 1.09 13.1 0.097 1.75 21.0 0.155 i/j (x1) 6454 1.49 1.49 0.010 1.09 1.09 0.007 1.75 1.75 0.011 8 (x1) 5713 1.49 1.49 0.009 1.09 1.09 0.006 1.75 1.75 0.010 g (x1) 17300 1.49 1.49 0.026 1.09 1.09 0.019 1.75 1.75 0.030 e (x1) 9240 1.49 1.49 0.014 1.09 1.09 0.010 1.75 1.75 0.016 Total 606 669 25.33 49.2 0.944 18.53 36.0 0.690 29.75 57.8 0.476 a = El valor en µg fue calculado como (MW de la subunidad madura) x (moles totales). 137 Tabla 9. Cuantificación del contenido de las subunidades a y b del V1Dg de la F1FO-ATP sintasa (4 diferentes muestras). BSA (curva estándar) Señal total 1.0 µg 6.076 Regresión lineal: r2 = 0.999; b = 0.242; m = 5.78 0.5 µg 3.038 0.3 µg 1.930 0.1 µg 0.909 V1Dg, muestra i Subunidad Señal total µg mol Proporción a/b a 1.569 0.4788 8.7 x 10-12 0.92 b 1.635 0.4788 9.49 x 10-12 Promedioa = 9.09 x 10-12 V1Dg, muestra ii a 1.143 0.5331 9.68 x 10-12 0.86 b 1.320 0.571 11.31 x 10-12 Promedioa = 10.50 x 10-12 V1Dg, muestra iii a 3.990 0.4475 8.13 x 10-12 0.82 b 3.290 0.4998 9.90 x 10-12 Promedioa = 9.02 x 10-12 V1Dg, muestra iv a 2.192 0.3615 6.57 x 10-12 0.85 b 1.524 0.3919 7.77 x 10-12 Promedioa = 7.17 x 10-12 Los datos se obtuvieron del análisis de las proteínas teñidas en gel mostradas en la Figura S1 = Dado que a/b tienen una estequiometría 1:1 en la F1FO-ATP sintasa nativa, el valor para a y b se enfocó como el promedio de sus valores independientes. 138 Tabla 10. Cálculo del contenido de F1FO-ATP sintasa en diferentes muestras de V1Dg aislados. Subunidad MW subunidad madura (Da) Muestra i Muestra ii Muestra iii Muestra iv pmol de subunidad pmol total µga pmol de subunidad pmol total µga pmol de subunidad pmol total µga pmol de subunidad pmol total µga a (x3) 55047 3.0310 9.0929 0.5005 3.4999 10.4996 0.5780 3.0056 9.0167 0.4963 3.7887 11.3661 0.6257 b (x3) 50465 3.0310 9.0929 0.4589 3.4999 10.4996 0.5299 3.0056 9.0167 0.4550 3.7887 11.3661 0.5736 b (x2) 22281 3.0310 6.0619 0.1351 3.4999 6.9997 0.1560 3.0056 6.0111 0.1339 3.7887 7.5774 0.1688 g (x1) 33157 3.0310 3.0310 0.1005 3.4999 3.4999 0.1160 3.0056 3.0056 0.0997 3.7887 3.7887 0.1256 a (x1) 27742 3.0310 3.0310 0.0841 3.4999 3.4999 0.0971 3.0056 3.0056 0.0834 3.7887 3.7887 0.1051 OSCP (x1) 21001 3.0310 3.0310 0.0637 3.4999 3.4999 0.0735 3.0056 3.0056 0.0631 3.7887 3.7887 0.0796 d (x1) 15552 3.0310 3.0310 0.0471 3.4999 3.4999 0.0544 3.0056 3.0056 0.0467 3.7887 3.7887 0.0589 f (x1) 7371 3.0310 3.0310 0.0223 3.4999 3.4999 0.0258 3.0056 3.0056 0.0222 3.7887 3.7887 0.0279 d (x1) 14615 3.0310 3.0310 0.0443 3.4999 3.4999 0.0512 3.0056 3.0056 0.0439 3.7887 3.7887 0.0554 h (x1) 10419 3.0310 3.0310 0.0316 3.4999 3.4999 0.0365 3.0056 3.0056 0.0313 3.7887 3.7887 0.0395 e (x1) 7538 3.0310 3.0310 0.0228 3.4999 3.4999 0.0264 3.0056 3.0056 0.0227 3.7887 3.7887 0.0286 Inh1 (x1) 7281 3.0310 3.0310 0.0221 3.4999 3.4999 0.0255 3.0056 3.0056 0.0219 3.7887 3.7887 0.0276 c (x12) 7394 3.0310 36.3716 0.2689 3.4999 41.9984 0.3105 3.0056 36.0666 0.2667 3.7887 45.4643 0.3362 i/j (x1) 6454 3.0310 3.0310 0.0196 3.4999 3.4999 0.0226 3.0056 3.0056 0.0194 3.7887 3.7887 0.0245 8 (x1) 5713 3.0310 3.0310 0.0173 3.4999 3.4999 0.0200 3.0056 3.0056 0.0172 3.7887 3.7887 0.0216 g (x1) 17300 - - - - - - - - - - - e (x1) 9240 - - - - - - - - - - - Total 606 669 45.4645 93.9599 1.8388 52.4981 108.4960 2.1233 45.0833 93.1721 1.8234 56.8304 117.4495 2.2985 a = El valor en µg fue calculado como (MW de la subunidad madura) x (moles totales). 139 10.6. Cuantificación de la cantidad de F1FO-ATP sintasa en las mitocondrias de las cepas WT y Dg. La cuantificación del contenido de F1FO-ATP sintasa presente en las WT y Dg (Figura S2), se resolvió como se describió en el punto 10.4 (anexo). Figura S2. SDS-Tricina-PAGE de las mitocondrias WT y Dg. Las proteínas WT (A) y Dg (B) se resolvieron mediante geles de Tricina-SDS-PAGE y se tiñeron con Coomassie® Blue. La intensidad de la subunidad a y b del complejo V se usó para la cuantificación de la concentración de F1FO-ATP sintasa usando una curva de BSA. 140 Tabla 11. Cuantificación del contenido de las subunidades a y b de la F1FO-ATP sintasa total en las mitocondrias de la cepa WT. BSA (curva estándar) Señal total 0.1 µg 21588 R = 0.9992; b = 19301.42; m = 20080.02 0.3 µg 25186 0.5 µg 29812 0.8 µg 35980 WT Muestra i (10µg) Subunidad Señal total µg mol Proporción a/b a 25443 0.295 5.36 0.73 b 26985 0.369 7.31 Promedioa 6.34 WT Muestra ii (15µg) a 27756 0.406 7.38 0.72 b 30069 0.518 10.26 Promedioa 8.82 WT Muestra iii (20µg) a 32125 0.616 11.19 0.75 b 34952 0.752 14.90 Promedioa 13.05 Los datos se obtuvieron del análisis de las proteínas teñidas en gel mostradas en la Figura S2 = Dado que a/b tienen una estequiometría 1:1 en la F1FO-ATP sintasa nativa, el valor para a y b se enfocó como el promedio de sus valores independientes. 141 Tabla 12. Cálculo del contenido de F1FO-ATP sintasa total en las mitocondrias de la cepa WT. Subunidad MW subunidad madura (Da) Muestra i Muestra ii Muestra iii pmol de subunidad pmol total µga pmol de subunidad pmol total µga pmol de subunidad pmol total µga a (x3) 55047 2.1118 6.3355 0.3488 2.9400 8.8200 0.4855 4.3486 13.0459 0.7181 b (x3) 50465 2.1118 6.3355 0.3197 2.9400 8.8200 0.4451 4.3486 13.0459 0.6584 b (x2) 22281 2.1118 4.2237 0.0941 2.9400 5.8800 0.1310 4.3486 8.6973 0.1938 g (x1) 33157 2.1118 2.1118 0.0700 2.9400 2.9400 0.0975 4.3486 4.3486 0.1442 a (x1) 27742 2.1118 2.1118 0.0586 2.9400 2.9400 0.0816 4.3486 4.3486 0.1206 OSCP (x1) 21001 2.1118 2.1118 0.0444 2.9400 2.9400 0.0617 4.3486 4.3486 0.0913 d (x1) 15552 2.1118 2.1118 0.0328 2.9400 2.9400 0.0457 4.3486 4.3486 0.0676 f (x1) 7371 2.1118 2.1118 0.0156 2.9400 2.9400 0.0217 4.3486 4.3486 0.0321 d (x1) 14615 2.1118 2.1118 0.0309 2.9400 2.9400 0.0430 4.3486 4.3486 0.0636 h (x1) 10419 2.1118 2.1118 0.0220 2.9400 2.9400 0.0306 4.3486 4.3486 0.0453 e (x1) 7538 2.1118 2.1118 0.0159 2.9400 2.9400 0.0222 4.3486 4.3486 0.0328 Inh1 (x1) 7281 2.1118 2.1118 0.0154 2.9400 2.9400 0.0214 4.3486 4.3486 0.0317 c (x12) 7394 2.1118 25.3421 0.1874 2.9400 35.2801 0.2609 4.3486 52.1837 0.3858 i/j (x1) 6454 2.1118 2.1118 0.0136 2.9400 2.9400 0.0190 4.3486 4.3486 0.0281 8 (x1) 5713 2.1118 2.1118 0.0121 2.9400 2.9400 0.0168 4.3486 4.3486 0.0248 g (x1) 17300 2.1118 2.1118 0.0365 2.9400 2.9400 0.0509 4.3486 4.3486 0.0752 e (x1) 9240 2.1118 2.1118 0.0195 2.9400 2.9400 0.0272 4.3486 4.3486 0.0402 Total 606 669 35.9013 69.6908 1.3372 49.9802 97.0203 1.8616 73.9269 143.5052 2.7536 a = El valor en µg fue calculado como (MW de la subunidad madura) x (moles totales). 142 Tabla 13. Cuantificación del contenido de las subunidades a y b de la F1FO-ATP sintasa total en las mitocondrias de la cepa Dg. BSA (curva estándar) Señal total 0.1 µg 17219 R = 0.9986; b = 15030.90; m = 21929.1 0.3 µg 21331 0.5 µg 26471 0.8 µg 32382 Dg Muestra i (10µg) Subunidad Señal total µg mol Proporción a/b a 19275 0.193 3.51 0.57 b 21845 0.311 6.16 Promedioa 4.83 Dg Muestra ii (15µg) a 23387 0.381 6.92 0.74 b 25443 0.475 9.41 Promedioa 8.17 Dg Muestra iii (20µg) a 26728 0.533 9.68 0.68 b 30840 0.721 14.29 Promedioa 11.98 Los datos se obtuvieron del análisis de las proteínas teñidas en gel mostradas en la Figura S2 = Dado que a/b tienen una estequiometría 1:1 en la F1FO-ATP sintasa nativa, el valor para a y b se enfocó como el promedio de sus valores independientes. 143 Tabla 14. Cálculo del contenido de F1FO-ATP sintasa total en las mitocondrias de la cepa Dg. Subunidad MW subunidad madura (Da) Muestra i Muestra ii Muestra iii pmol de subunidad pmol total µga pmol de subunidad pmol total µga pmol de subunidad pmol total µga a (x3) 55047 1.6115 4.8344 0.2661 2.7223 8.1669 0.4496 3.9950 11.9849 0.6597 b (x3) 50465 1.6115 4.8344 0.2440 2.7223 8.1669 0.4121 3.9950 11.9849 0.6048 b (x2) 22281 1.6115 3.2229 0.0718 2.7223 5.4446 0.1213 3.9950 7.9899 0.1780 g (x1) 33157 1.6115 1.6115 0.0534 2.7223 2.7223 0.0903 3.9950 3.9950 0.1325 a (x1) 27742 1.6115 1.6115 0.0447 2.7223 2.7223 0.0755 3.9950 3.9950 0.1108 OSCP (x1) 21001 1.6115 1.6115 0.0338 2.7223 2.7223 0.0572 3.9950 3.9950 0.0839 d (x1) 15552 1.6115 1.6115 0.0251 2.7223 2.7223 0.0423 3.9950 3.9950 0.0621 f (x1) 7371 1.6115 1.6115 0.0119 2.7223 2.7223 0.0201 3.9950 3.9950 0.0294 d (x1) 14615 1.6115 1.6115 0.0236 2.7223 2.7223 0.0398 3.9950 3.9950 0.0584 h (x1) 10419 1.6115 1.6115 0.0168 2.7223 2.7223 0.0284 3.9950 3.9950 0.0416 e (x1) 7538 1.6115 1.6115 0.0121 2.7223 2.7223 0.0205 3.9950 3.9950 0.0301 Inh1 (x1) 7281 1.6115 1.6115 0.0117 2.7223 2.7223 0.0198 3.9950 3.9950 0.0291 c (x12) 7394 1.6115 19.3376 0.1430 2.7223 32.6676 0.2415 3.9950 47.9395 0.3545 i/j (x1) 6454 1.6115 1.6115 0.0104 2.7223 2.7223 0.0176 3.9950 3.9950 0.0258 8 (x1) 5713 1.6115 1.6115 0.0092 2.7223 2.7223 0.0156 3.9950 3.9950 0.0228 g (x1) 17300 1.6115 1.6115 0.0000 2.7223 2.7223 0.0000 3.9950 3.9950 0.0000 e (x1) 9240 1.6115 1.6115 0.0149 2.7223 2.7223 0.0252 3.9950 3.9950 0.0369 Total 606 669 27.3949 53.1783 0.9925 46.2792 89.8360 1.6767 67.9143 131.8337 2.4605 a = El valor en µg fue calculado como (MW de la subunidad madura) x (moles totales). 144 “Mitochondrial Oroboros” Odra Noel Artículos 145 11. ARTÍCULOS 11.1. Artículo de requisito • STRUCTURAL AND KINETICS CHARACTERIZATION OF THE F1F0-ATP SYNTHASE DIMER. NEW REPERCUSSION OF MONOMER-MONOMER CONTACT (2017) Mercedes Esparza-Perusquía, Sofía Olvera-Sánchez, Juan Pablo Pardo, Guillermo Mendoza-Hernández, Federico Martínez, and Oscar Flores-Herrera. BBA-Bioenergetics. 1858(12):975-981. doi: 10.1016/j.bbabio.2017.09.002. 11.2. Publicaciones directas. • DELETION OF SUBUNIT G FROM F1F0-ATP SYNTHASE AFFECTS THE STABILITY OF ITS DIMERIC STATE AND THE MITOCHONDRIAL ATP SYNTHESIS. Esparza-Perusquía M, Langner T, Feldbrügge M, Pardo JP, Martínez F, and Flores-Herrera O. En preparación. • DELETION OF THE NATURAL INHIBITORY PROTEIN INH1 FROM USTILAGO MAYDIS ATP SYNTHASE DOES NOT INCREASE THE ACTIVITY OF THE DIMERIC STATE OF F1FO-ATP SYNTHASE. Romero- Aguilar Lucero*, Esparza-Perusquía Mercedes*, Langner Thorsten, García Giovanni, Feldbrügge Michael, Pardo Juan Pablo, Martínez Federico, and Flores-Herrera Oscar. En preparación. 11.3. Otras publicaciones. • MITOCHONDRIAL PROTEASES ACT ON STARD3 TO ACTIVATE PROGESTERONE SYNTHESIS IN HUMAN SYNCYTIOTROPHOBLAST (2015) Mercedes Esparza-Perusquía, Sofía Olvera-Sánchez, Oscar Flores- Herrera, Héctor Flores-Herrera, Alberto Guevara-Flores, Juan P Pardo, María T Espinosa-García y Federico Martínez. BBA-General Subjects. 1850(1):107- 17. doi: 10.1016/j.bbagen.2014.10.009. • MEMBRANE POTENTIAL REGULATES THE MITOCHONDRIAL ATP- DIPHOSPHOHYDROLASE ACTIVITY BUT IT IS NOT INVOLVED IN THE 146 PROGESTERONE BIOSYNTHESIS IN HUMAN SYNCYTIOTROPHOBLAST CELL (2015) Oscar Flores-Herrera, Sofía Olvera-Sánchez, Mercedes Esparza-Perusquía, Juan Pablo Pardo, Juan Luis Rendón, Guillermo Mendoza-Hernández, and Federico Martínez. BBA- Bioenergetics. 1847(2):143-52. doi: 10.1016/j.bbabio.2014.10.002. • MULTIPLE FUNCTIONS OF SYNCYTIOTROPHOBLAST MITOCHONDRIA IN PREGNANCY (2015) Federico Martínez, Sofía Olvera-Sánchez, Mercedes Esparza-Perusquía, Erika Gómez-Chang, Oscar Flores-Herrera. Steroids, 103:11-22. doi: 10.1016/j.steroids.2015.09.006. • STREPTOZOTOCIN INDUCED ADAPTIVE MODIFICATION OF THE MITOCHONDRIAL SUPERCOMPLEXES IN LIVER OF WISTAR RATS AND THE PROTECTIVE EFFECT OF MORINGA OLEIFERA LAM (2018) Alejandra Sánchez-Muñoz, Mónica A Valdez-Solana, Mara I Campos- Almazán, Oscar Flores-Herrera, Mercedes Esparza-Perusquía, Sofía Olvera-Sánchez, Guadalupe García-Arenas, Claudia Avitia-Domínguez, Alfredo Téllez-Valencia and Erick Sierra-Campos. Biochemistry Research International. 1-15. ID 5681081. doi:10.1155/2018/5681081. • CARDIOPROTECTIVE STRATEGIES PRESERVE THE STABILITY OF RESPIRATORY CHAIN SUPERCOMPLEXES AND REDUCE OXIDATIVE STRESS IN REPERFUSED ISCHEMIC HEARTS (2018) Ixchel Ramírez- Camacho, Francisco Correa, Mohammed El-Hafidi, Alejandro Silva-Palacios, Marcos Ostolga-Chavarría, Mercedes Esparza-Perusquía, Sofía Olvera- Sánchez, Oscar Flores-Herrera and Cecilia Zazueta. Free Radical Biology and Medicine. 129:407-417. doi: 10.1016/j.freeradbiomed.2018.09.047. • MITOCHONDRIAL RESPIRASOME WORKS AS A SINGLE UNIT AND THE CROSSTALK BETWEEN COMPLEXES I, III2 AND IV STIMULATES NADH DEHYDROGENASE ACTIVITY (2019) Meztli Reyes-Galindo, Roselia Suarez, Mercedes Esparza-Perusquía, Jaime de Lira-Sánchez, Juan Pablo Pardo, Federico Martínez, and Oscar Flores-Herrera. BBA-Bioenergetics. 1860(8):618-627. doi: 10.1016/j.bbabio.2019.06.017. 147 • ASPECTOS GENERALES DEL TRANSPORTE DE COLESTEROL EN LA ESTEROIDOGÉNESIS DE LA PLACENTA HUMANA (2019) Sofía Olvera- Sánchez, Mercedes Esparza-Perusquía, Oscar Flores-Herrera, Viviana A. Urban-Sosa y Federico Martínez. TIP Revista Especializada en Ciencias Químico-Biológicas, 22:1-9. DOI: 10.22201/fesz.23958723e. • STEADY-STATE PERSISTENCE OF RESPIRATORY SYNCYTIAL VIRUS IN A MACROPHAGE-LIKE CELL LINE AND IDENTIFICATION OF NON- SYNONYMOUS MUTATIONS THROUGH SEQUENCING OF THE PERSISTENT VIRAL GENOME. Ximena Ruiz-Gómez, Joel Armando Vázquez-Pérez, Oscar Flores-Herrera, Mercedes Esparza-Perusquía, Carlos Santiago-Olivares, Jorge Gaona, Beatriz Gómez, Carmen Méndez, Evelyn Rivera-Toledo. MDPI-viruses, Manuscript ID: 842162. Enviado. Contents lists available at ScienceDirect BBA - Bioenergetics journal homepage: www.elsevier.com/locate/bbabio Structural and kinetics characterization of the F1F0-ATP synthase dimer. New repercussion of monomer-monomer contact Mercedes Esparza-Perusquía, Sofía Olvera-Sánchez, Juan Pablo Pardo, Guillermo Mendoza-Hernández, Federico Martínez, Oscar Flores-Herrera⁎ Departamento de Bioquímica, Facultad de Medicina, Universidad Nacional Autónoma de México, 04510 México City, México A R T I C L E I N F O Keywords: F1F0-ATP synthase dimer Supercomplexes ATPase activity Ustilago maydis mitochondria A B S T R A C T Ustilago maydis is an aerobic basidiomycete that fully depends on oxidative phosphorylation for its supply of ATP, pointing to mitochondria as a key player in the energy metabolism of this organism. Mitochondrial F1F0- ATP synthase occurs in supramolecular structures. In this work, we isolated the monomer (640 kDa) and the dimer (1280 kDa) and characterized their subunit composition and kinetics of ATP hydrolysis. Mass spectro- metry revealed that dimerizing subunits e and g were present in the dimer but not in the monomer. Analysis of the ATPase activity showed that both oligomers had Michaelis-Menten kinetics, but the dimer was 7 times more active than the monomer, while affinities were similar. The dimer was more sensitive to oligomycin inhibition, with a Ki of 24 nM, while the monomer had a Ki of 169 nM. The results suggest that the interphase between the monomers in the dimer state affects the catalytic efficiency of the enzyme and its sensitivity to inhibitors. 1. Introduction The F1F0-ATP synthase is the principal source of cellular ATP. The yeast mitochondrial F1F0-ATP synthase is a 600 kDa complex that contains at least 17 distinct subunits [1] organized into a catalytic so- luble-component called F1, with three catalytic sites, and a membrane- spanning component called F0, composed by hydrophobic subunits forming a specific proton channel. F0 and F1 are linked by central and peripheral stalks [2]. The enzyme uses the proton electrochemical gradient generated by the respiratory chain to produce ATP from ADP and inorganic phosphate. This enzyme is a nano-motor where the proton translocation through F0 induces the rotation of a ring of 9–14 hydrophobic c-subunits [3], which drives the rotation of a central stalk inside the catalytic head which results in the synthesis of ATP [4]. The mitochondrial ATP synthase acquires supramolecular structures that have been found in many organisms by native gel electrophoresis, mainly in the dimeric form [5–7] but also in higher molecular forms (i.e. tetramers, hexamers) according to the electrophoresis technique used [7–10]. Depending on the detergent used during the extraction and the purification procedures, the isolated complexes were monomers and/or dimers as shown by electron microscopy single particle analysis [11–16], although higher structures such as tetramers have also been reported [16]. Dimers of mitochondrial ATP synthase have been observed in situ by atomic force microscopy [17], transmission electron microscopy [16] and electron cryo-tomography [18]. Higher supramolecular structures (tetramers, hexamers) have been reported both in blue native gel and cross-linking experiments showing that subunits e and g are involved in dimer stabilization [19] and the oligomerization process [7,20–22]. The ATP synthase dimer from Saccharomyces cerevisiae adopts a V- like structure with the two F0 parts linked together [23]. Several F0 subunits, as e, g, i, k and 8, are essential for dimerization [5,23] and oligomerization [19], but are not involved in the ATPase or ATP syn- thase activity. Even though the dimer's structure and its role in cristae architecture have been defined with high precision, only a few attempts have been made to understand the effect of oligomerization on ATPase activity. Because the ATP synthase oligomers remain in the native state, BN- PAGE had been combined with in-gel activity staining to estimate ATPase activity [7,21,24–26]. However, the results are not conclusive due to differences in assay conditions, optimization of the in-gel ATPase assay, and/or ADP accumulation in the assay mixture. Therefore, it is important to study spectrophotometrically the ATPase activity to fa- cilitate the comparison of the specific activities of monomer and dimer of the F1F0-ATP synthase. In this work, the dimer and monomer of the F1F0-ATP synthase from Ustilago maydis, an aerobic basidiomycete, were solubilized with digi- tonin, a soft detergent, and isolated by sucrose gradient centrifugation. http://dx.doi.org/10.1016/j.bbabio.2017.09.002 Received 11 May 2017; Received in revised form 24 August 2017; Accepted 12 September 2017 ⁎ Corresponding author at: Departamento de Bioquímica, Facultad de Medicina, Universidad Nacional Autónoma de México, Apdo. Postal 70-159, Coyoacán 04510, México, Cd. Mx., México. E-mail address: oflores@bq.unam.mx (O. Flores-Herrera). BBA - Bioenergetics 1858 (2017) 975–981 Available online 14 September 2017 0005-2728/ © 2017 Elsevier B.V. All rights reserved. MARK It was determined by MS/MS analysis that the dimer showed the sub- units e and g as dimerizing components; additionally, the dimer was 7 times more active than the monomer and more sensitive to oligomycin. Thermostability kinetics showed that the dimer dissociates into monomers previous to their inactivation. The apparent Tm was about 52 °C for both oligomeric states. This is the first report on the kinetic characterization of the isolated dimer and monomer of the F1F0-ATP synthase from a single organism, and it shows that these oligomeric states have different catalytic properties besides their subunit compo- sition. In this sense, a new role of the interphase between individual monomers is discussed. 2. Materials and methods 2.1. Cell culture and mitochondria isolation Ustilago maydis cells (strain FB2) were prepared as previously de- scribed [27]. U. maydis mitochondria were isolated using the method described by Waterfield and Sisler [28]. 2.2. Solubilisation and isolation of dimer and monomer of F1F0-ATP synthase The dimer (V2) and monomer (V1) of F1F0-ATP synthase from U. maydis mitochondria (35 mg) were solubilized as described [29–31], by incubation at 4 °C for 30 min with digitonin (detergent/protein ratio of 2:1), and centrifuged at 100,000 g for 30 min at 4 °C. The supernatant was recovered (16 mg protein) and immediately loaded on 24 ml of a continuous sucrose gradient (16–42% sucrose, 15 mM Tris, pH 7.4, 20 mM KCl, and 0.2% digitonin). This mixture was centrifuged at 131,000 g for 18 h at 4 °C. Fractions (0.5 ml) were collected from the bottom of the gradient. Fractions containing V1 and V2 were identified by hrCN-PAGE [29], pooled separately and diluted 10 folds with 30 mM HEPES, pH 8.0 and 5% glycerol and concentrated using a Centrifugal Filters Units (100 K, Millipore Amicon Ultra) to a final volume of 100 μl, and stored at −70 °C until used. The concentration of F1F0-ATP synthase in each V1 and V2 sample was determined by a densitometry analysis of Coomassie© Brilliant Blue R-125 stained α- and β-subunits from an SDS-Tricine-PAGE, using Coomassie stained BSA as a standard (see Supplemental material section, Fig. S1). Densitometry analysis was performed with ImageJ software [32]. For hrCN-PAGE, 50 μg protein (per lane) were applied to 4–10% native-PAGE gels and run at 4 °C, 35 V for 10 h. The molecular weight of complexes and supercomplexes was determined by their electro- phoretic mobilities and in-gel catalytic activities, using the complexes of digitonin-solubilized bovine heart mitochondria as standards. The in-gel activity of complex V was performed as described by Jung [30] in 50 mM glycine (adjusted to pH 8.0 with triethanolamine), 10 mM MgCl2, 0.15% Pb(ClO4)2 and 5 mM ATP. The presence of re- spiratory complexes and supercomplexes was determined by their in- gel activity, and by MS/MS analysis (see Supplemental material section, Fig. S1). 2D-Tricine-SDS-polyacrylamide gel electrophoresis (2D-SDS-PAGE) was performed according to Schägger [31] on a 16% polyacrylamide gel under denaturing conditions, and proteins were visualized with Coomassie and used for subunit identification by tandem mass spec- trometry (LC/ESI-MS/MS) [33]. Database searching and protein iden- tification were performed with the MS/MS spectra data sets using the MASCOT search algorithm (version 1.6b9, Matrix Science, London, U.K., available at http://www.matrixscience.com). Mass tolerances of 0.5 Da for the precursor and 0.3 Da for the fragment ion masses were used. Carbamidomethyl-cysteine was the fixed modification, and one missed cleavage for trypsin was allowed. Searches were conducted using the Fungi subset of the NCBI nr database (http://www.ncbi.nih. gov). Protein identifications were accepted when at least two MS/MS spectra matched at 95% confidence level (p < 0.05). 2.3. Measurements of ATP hydrolysis by F1F0-ATP synthase dimer and monomer ATP hydrolysis by isolated V1 or V2 was measured spectro- photometrically at 25 °C using an assay coupled to the oxidation of NADH (ε340nm = 6.2 mM−1 cm−1) in an Agilent 8453 UV–visible spectrophotometer (Agilent Technologies, USA). The assay medium contained 50 mM HEPES, pH 8.0, 1 mM MgSO4, 90 mM KCl; the ATP- regenerating system was 5 mM phosphoenolpyruvate (PEP), 0.2 mM NADH, 50 units/ml of pyruvate kinase (PK), and 30 units/ml of lactate dehydrogenase (LDH). The protein concentration was 10–15 μg F1F0- ATP synthase/ml, and the ATPase reaction was started by the addition of the enzyme and NADH absorbance was continuously monitored (see Supplemental material section). The time response, checked by ADP additions, was< 1 s. Initials velocities were further obtained from the slope of the linear region in each spectrophotometric recording, and the linear region of the traces was corroborated with the plot of the first- derivative against time (see Supplemental material section). Since a non-specific NADH oxidation was observed prior to the enzyme addi- tion, the actual value of ATPase activity was calculated as the Fig. 1. Isolation, in-gel activity, and subunit composition of the dimer (V2) and monomer (V1) of the F1F0-ATP synthase. V2 and V1 from Ustilago maydis mitochondria were solu- bilized with digitonin and subsequently isolated by sucrose-gradient ultracentrifugation and their in-gel activity analyzed by hrCN-PAGE (A). Fractions from the bottom [12–13] and the middle [19–22] of the sucrose gradient were used to obtain isolated V2 and V1, respectively (B). Inhibition by oligomycin of the in-gel ATPase activity of isolated V2 or V1 (C). Note that no F1 activity was observed. Coomassie stained of V1 and V2 proteins from CN-PAGE (D). The in-gel activity of V2 and V1 were excised from CN-PAGE gel and proteins were resolved by 2D-Tricine-SDS-PAGE and subunits identities elucidated by MS/MS (E, and Table 1). M. Esparza-Perusquía et al. BBA - Bioenergetics 1858 (2017) 975–981 976 subtraction of the slope of this region from the slope that is obtained after adding the enzyme (see Supplemental material section). Oligomycin (3 μM) was added to inhibit ATPase activity and verify F1F0-ATP synthase integrity. Only samples with ATPase activity 100% inhibited by oligomycin were used in this study. Data for the hydrolysis of ATP were analyzed by robust, weighted, non-linear regression ana- lysis using the SigmaPlot software (Systat Software, Inc., Version 10.0), and represent the average of three replicates from five independent preparations. 2.4. Thermal inactivation of F1F0-ATP synthase dimer and monomer V2 or V1 (500 μg/ml) were suspended in a medium containing 50 mM HEPES, pH 8.0, 1 mM MgSO4, 90 mM KCl. 250 μl was placed into a glass tube and heated at indicated temperatures for 15 min. The suspension was cooled to room temperature, and a 100 μl aliquot was drawn and immediately assayed at 25 °C for Mg-ATP hydrolysis in the presence of an ATP-regenerating system as described in previously section. Triplicates were performed at each temperature, as well as a control sample which was kept at room temperature for an equal length of time before essay. 2.5. Oligomycin inhibition of F1F0-ATP synthase dimer and monomer Continuous monitoring of ATP hydrolysis by F1F0-ATP synthase V2 or V1 coupled to NADH oxidation was carried out as described in the previous section. ATP hydrolysis was initiated by Mg-ATP addition; after 2–4 min oligomycin was added and ATPase activity decayed. Because oligomycin is a slow-binding inhibitor the total recording time for maximal ATPase activity inhibition, at different oligomycin con- centrations, was about 25–35 min. The spectrophotometric recording was used to calculate the residual activity of ATPase as V(I) / V(0). V(0) is the constant rate of absorbance variation before oligomycin addition in absorbance units per second (proportional to the initial ATPase ac- tivity), and V(I) the final rate of absorbance variation after oligomycin addition (proportional to the final ATPase activity). To calculate the Ki the relation between the 1 − (V(I) / V(0)) ratio and the inhibitor concentration ([I]) were fitted to the following function: − = − + − +V V v v1 (I) (0) 1 ( (1 ) (1 [I] K ))r r i (1) where vr is the inhibitor-insensitive fraction of V(0) (always lower than 3%). Initially, it was assumed that the [I] was much higher than the enzyme concentration. Therefore, the total and free concentrations of the [I] could be considered identical and constant during the kinetics of inhibition; however, when the concentration of F1F0-ATP synthase was quantified, it was observed a low oligomycin/F1F0-ATP synthase ratio when the oligomycin concentration was lower than 1 μM (Fig. 3C). With this information, a new statistical analysis of data was performed with the DynaFit software [34], in which no assumption about a higher concentration of oligomycin over the concentration of the enzyme is required. The Ki values obtained by both methods were similar and were inserted in Table 2. Oligomycin/F1F0-ATP synthase ratio was determined from F1F0-ATP synthase concentration and the oligomycin used in each spectrophotometric assay. 3. Results 3.1. Isolation of dimeric and monomeric F1F0-ATP synthase and subunits identification Dimeric and monomeric ATP synthase were extracted from U. maydis mitochondria with digitonin and isolated by sucrose density gradient centrifugation. The presence of isolated dimeric and mono- meric F1F0-ATP synthase at the bottom and the middle of the sucrose density gradient, respectively, was confirmed by ATPase activity staining in hrCN-PAGE (Fig. 1A). The V2 and V1 samples were pooled independently and concentrated until F1F0-ATP synthase content was similar as monitored by Coomassie Blue staining of the α and β subunits (Fig. 1E); their integrity was analyzed by hrCN-PAGE (Fig. 1B). These samples contained oligomycin-sensitive ATPase activity corresponding to V2 or V1 without free F1 sector (Fig. 1C). As expected, the Coomassie® stain (Fig. 1D) showed the presence of supercomplexes (I-III2-IV) and complex I in the V2 sample, while individual complexes III2 and IV were observed in V1 sample (see Supplemental material section). Therefore, Table 1 Molecular mass and subunit identity of complex V1 and V2 Ustilago maydis mitochondria. V2 (1200 kDa) V1 (660 kDa) Molecular mass (kDa) Exclusive unique peptides Unique exclusive spectra/total spectra Coverage (%) Access number NCBI Subunit identity α α 55.05 16 20/41 32 XP_011392137 β β 50.47 21 39/99 50 XP_011389783 γ γ 31.16 9 10/18 39 XP_011388164 a a 27.74 5 6/10 40 Q0H8Y6 b b 22.28 7 7/9 21 XP_011392148 OSCP OSCP 21.00 7 8/14 38 XP_011388314 d d 15.55 3 4/9 41 XP_011392758 δ δ 14.62 4 4/4 39 XP_011387060 h h 10.42 5 7/9 43 XP_011388666 ε ε 7.54 7 9/22 74 XP_011391935 c c 7.39 7 10/15 62 Q0H8W9 f f 7.37 3 3/5 32 XP_011388343 Inh1 Inh1 7.28 5 6/11 41 XP_011388667 i/j i/j 6.45 5 5/9 36 XP_011391953 8 8 5.71 5 6/11 28 Q0H8Y5 g ― 17.3 5 6/14 22 XP_011386964 e ― 9.24 6 8/14 31 XP_011389516 Accessory proteins ANT ANT 33.75 11 12/23 32 XP_011386926 PC PC 34 12 13/25 30 XP_011388175 PHB2 PBH2 35.56 13 17/75 43 XP_011391320 PHB1 PBH1 29.76 11 15/49 32 XP_011390099 The molecular weight from V2 and V1 was determined by BN-PAGE using mitochondrial respiratory complexes from heart bovine as standard. The subunit molecular weight was determined by 2D-Tricine-SDS-PAGE. ANT = adenine nucleotide translocase; PC = phosphate carrier; PHB1 = prohibitin 1; PHB2 = prohibitin 2; ANT = ADP/ATP carrier protein. The identity of each protein was determined by mass spectrometry. The accession number was obtaining from NCBI. M. Esparza-Perusquía et al. BBA - Bioenergetics 1858 (2017) 975–981 977 to compare the kinetics of V1 and V2, the concentration of F1F0-ATP synthase was determined in each V2 and V1 preparations, and the A- TPase activities were corrected by the actual amount of complex V present (see Supplemental material section). The molecular weight calculated for V1 and V2 was 660 and 1200 kDa respectively. The V2 and V1 samples were analyzed by MS/MS tandem spectro- metry. The monomer showed the complete set of F1F0-ATP synthase subunits (Fig. 1E and Table 1); the dimer showed the supernumerary subunits e and g (Table 1), which have been related to the dimerization process of F1F0-ATP synthase [5–7]. Interestingly, the ADP/ATP translocase (ANT) and phosphate carrier (PC) were joined to both oli- gomeric states (Fig. 1E), suggesting the presence of the ATP-syntha- some. Also, prohibitin 1 (PBH1) and 2 (PBH2) were associated to both oligomers. Then, it was possible to obtain both V2 and V1 separated from each other with oligomycin-sensitive ATPase activity. 3.2. Kinetics of ATP hydrolysis by the dimer and monomer of F1F0-ATP synthase ATPase activity of V1 and V2 was dependent on Mg-ATP concentration (Fig. 2A and B) and showed a Michaelis-Menten kinetics for the hydrolysis of ATP. Vmax value was 0.83 ± 0.05 μmol ATP hydrolyzed(mg F1F0-A- TP synthase·min)−1 and a Km =308 ± 90 μM for V1 (Fig. 2G, close circles); while V2 exhibited a low ATPase activity, with a Vmax =0.54 ± 0.08 μmol ATP hydrolyzed(mg F1F0- ATP synthase·min)−1 and a Km =884 ± 100 μM (Fig. 2H inset). It has been reported that ATPase activity of the F1F0-ATP synthase increased in the presence of a nonionic detergent, i.e. DDM [35]. Titration curves with increasing concentration of DDM showed a significant increase in the A- TPase activity of the V2 (Fig. 2C and D). Maximal activation was observed at a concentration of 0.005% (w/v) of DDM, and the enzyme was fully sensitive to oligomycin (Figs. 1C and 3). Activation of V2 by DDM was not due to dissociation of the oligomer, since increasing the detergent con- centration to 0.5% in the hrCN-PAGE did not release the monomer (Fig. 1). In the presence of DDM, V1 showed a Vmax and Km values of 1.431 ± 0.04 μmol ATP hydrolyzed(mg F1F0-ATP synthase·min)−1 and 207 ± 30 μM, respectively (Fig. 2G, open circles). V2 showed a Vmax =9.604 ± 0.26 μmol ATP hydrolyzed(mg F1F0-A- TP synthase·min)−1 and a Km =490 ± 80 μM (Fig. 2H, open circles, for comparison the activity of V2 without DDM activation was included, closed circles). The kcat for DDM activated oligomers were 14.5 and 207.7 s−1 for V1 and V2, respectively (Table 2). In this sense, the kcat/Km values from the V1 and the V2 were 7× 104 and 4.2 × 105 M−1 s−1, respectively, indicating a higher specificity of V2. 3.3. Inhibition of V1 and V2 F1F0-ATP synthase by oligomycin Fig. 3 shows typical kinetics of V1 (A) and V2 (B) inhibition by oligomycin. The inhibition by oligomycin was time dependent, and the steady state ATPase activity decreased with oligomycin concentration. Fig. 2. Kinetic characterization of V2 and V1. Time course of ATP hydrolysis by V1 (A) and V2 (B) mon- itored by NADH absorbance at 340 nm. Stimulation of V2 activity by different additions of DDM (C); maximal activation was observed at 0.005% DDM (D). ATPase activity of V1 (E) and V2 (F) in the presence of 0.005% DDM at different Mg-ATP concentrations. Enzyme or DDM additions indicated by vertical arrows and Mg- ATP concentration at the left side. The dependence of V1 (G) and V2 (H) activity on Mg-ATP concentration at pH 8 was fitted to the Michaelis-Menten equation, in the presence (open circles) or absence (closed circles) of DDM. For clarity, inset in panel H represent the activity of V2 without DDM activation. The data are the average of three replicates from five independent preparations. Error bars represent S.D. M. Esparza-Perusquía et al. BBA - Bioenergetics 1858 (2017) 975–981 978 The time course of the inhibition did not follow monoexponential decay. Inhibition of V1 was over 90% at 3.8 μM (Fig. 3C, open circles) while for V2 it was at 1.3 μM (Fig. 3C, closed circles). Analysis of A- TPase inhibition by oligomycin showed that V2 had a Ki = 24 ± 3 nM, while the Ki for V1 was 169 ± 10 nM. Since the amount of complex V in both preparations was quantified (see Supplemental material), it was possible to calculate the oligo- mycin/F1F0-ATP synthase ratio that resulted in 50% of ATPase inhibi- tion (Fig. 3C, bottom offset axis and Table 2). The ratio was 0.85 ± 0.12 for V2 and 6 ± 0.43 for V1. Taken together, our results suggest that the interphase between monomers in the dimer changed the oligomycin sensitivity. 3.4. Thermal inactivation of isolated V1 and V2 from F1F0-ATP synthase Thermal stability of V1 and V2 was monitored by measuring their residual ATPase activity. Fig. 4A shows typical transition curves for Mg- ATP hydrolysis rate of V1 and V2 plotted against the temperature of incubation. The activity (V/V0) of V1 and V2 was not affected by in- cubation below 40 °C. However, a dramatic reduction in activity oc- curred for samples preincubated above 40 °C. The midpoint of the ir- reversible inactivation (Tm) was 52.3 °C. No loss of oligomycin sensitivity for V1 and V2 was detected in the range of temperatures tested, and no release of F1 upon heat denaturation was observed in the hrCN-PAGE analysis (Fig. 4B and C). This suggested that V2 inactivation might occur through its dissociation into V1 and finally the inactivation of V1; therefore, the monomer-monomer interphase apparently did not contribute to V2 stability. 4. Discussion Subunit composition of the dimer and monomer of mitochondrial complex V have been determined in many organisms, including mam- mals [15], the colorless alga Polytomella sp. [36,37] and S. cerevisiae [7,8,38]. Structurally, V2 and V1 from the basidiomycete U. maydis contains fifteen conserved subunits, while V2 exhibited the e and g dimerizing subunits (Table 1). Interestingly, the presence of the ANT and the phosphate carrier suggests that the ATP-synthasome could be formed in both ATP-synthase oligomers, highlighting their role in mi- tochondrial bioenergetics. Although mitochondrial architecture is closely related to the pre- sence of V2 [39], studies focused on its kinetic activity are scarce. In the past decade, there have been some efforts to determine the kinetic constants of the dimer [26]. In that report, the activity rate V2/V1 was equal to 1; and a slight increase was observed at 37 °C. However, the results were based on in-gel activity analysis of complex V using a long time span. In these conditions, ADP produced by ATPase activity will accumulate in the medium, which in turn would inhibit the enzyme. Additionally, densitometry quantification of the in-gel activity could be inaccurate due to gel oversaturation by calcium phosphate precipita- tion. Also, this method does not allow the calculation of enzyme affinity nor its inhibition by oligomycin. These drawbacks can be prevented by isolating the dimer. In this sense, Gonzalez-Halphen's group isolated Fig. 3. ATPase inhibition by oligomycin of V1 and V2. Conditions as in Fig. 2E and F except that Mg-ATP concentration was held at 5 mM. Time course of NADH oxidation coupled to ATP hydrolysis was monitored during 25–30 min for maximal ATPase in- hibition of V1 (A) or V2 (B) and final spectrophotometric recording was analyzed as de- scribed in material and methods section. Enzyme or oligomycin additions indicated by vertical arrows and inhibitor concentration at the right side. (C) Maximal extent of in- hibition of isolated V1 (open circles) or V2 (closed circles) as a function of oligomycin concentration. V0 was the ATPase activity before oligomycin addition. The data are the average of three replicates from four independent preparations. Error bars represent S.D. Table 2 Kinetics parameters of ATPase activity of the monomer and the dimer of F1F0-ATP syn- thase from Ustilago maydis mitochondria. Monomer (V1) a Dimer (V2) a Non-activated F1F0-ATP synthase Vmax (μmol ATP hydrolyzed/mg F1F0- ATP synthase·min−1) 0.83 ± 0.05 0.54 ± 0.08 Km (μM) 308 ± 90 884 ± 100 kcat (s −1) 8.4 ± 0.06 10.6 ± 0.15 kcat/Km (M−1 s−1) 2.7 × 104 ± 0.3 1.2 × 104 ± 0.1 DDM activated F1F0-ATP synthase Vmax (μmol ATP hydrolyzed/mg F1F0- ATP synthase·min−1) 1.43 ± 0.04 9.60 ± 0.26 Km (μM) 207 ± 30 488 ± 80 kcat (s −1) 14.5 ± 0.03 207.7 ± 0.03 kcat/Km (M−1 s−1) 7.0 × 104 ± 0.2 4.2 × 105 ± 0.2 Ki (nM) for oligomycin 169 ± 10 24 ± 3 160 ± 17b 21 ± 4b Ratio oligomycin/F1F0-ATP synthase (mol/mol) to reach 50% of ATPase inhibition 6.0 ± 0.43 0.85 ± 0.12 a F1F0-ATP synthase mol in V1 and V2 samples was determined as described in pro- cedures section, and kinetics parameters were showed as mg of F1F0-ATP synthase. b Ki values calculated by DynaFit software (see Materials and methods section). M. Esparza-Perusquía et al. BBA - Bioenergetics 1858 (2017) 975–981 979 and characterized kinetically a highly stable V2 from Polytomella sp. [35]; however, because the high stability of V2, no data about V1 ac- tivity was provided for comparison. In this study, we isolated both the V2 and V1 from U. maydis to determine their subunit composition by MS/MS, and to study their kinetics by spectrophotometric analysis in the presence of an ATP-regenerating system to eliminate the inhibition by ADP. Characterization of the ATPase activity of V2 and V1 from U. maydis showed that the monomer is 1.5 times more active than the dimer; the low ATPase activity of V2 was also observed in the Polytomella sp. dimer of complex V [35]. However, activation by DDM increased 18-fold the activity of V2 while the affinity was not modified. Activation of V1 by DDM increased only a 1.7-fold the Vmax with no change of Km. Speci- ficity constant (kcat/Km) was 6 times higher for the V2 than for V1, il- lustrating the differences between both oligomers again. It is important to mention that no dissociation of V2 into V1 or F1 sector was observed in these experiments (Fig. 1), and ATPase activity was fully inhibited by oligomycin (Fig. 3). Since the CMC of DDM is 120–160 μM and it was used at 98 μM in the assay, activation of V2 by DDM could be attributed to the interaction between the enzyme and the detergent. Interestingly, both oligomers were activated by DDM, and their maximum ATPase activities were achieved. The activation of V1 and V2 by DDM could also be due to the Inh1 release (the regulatory subunit); however, it has been shown in mam- malian cells that this subunit is removed from the F1F0-ATP synthase at high pH (i.e. pH = 8.0), promoting full ATPase activity [40]. In this work the ATPase activity of V1 and V2 was assayed at pH = 8, but it is important further to explore the effect of pH on the activation of V1 and V2 by DDM. An important correction used for the comparison of the activities of V1 and V2 was the quantification of the actual concentra- tion of F1F0-ATP synthase in each V2 and V1 samples. Thus for the spectrophotometric assay, approximately the same amount of F1F0-ATP synthase was added either in the dimer or monomer state. Therefore, the V2 activity was much more than the sum of two V1 activities, even when both enzymes were activated by DDM. These results indicate that the monomer-monomer interphase affected the function of the F1F0- ATP synthase, increasing the Vmax of the dimer by 7 times and its sensitivity to oligomycin. Importantly, kinetic parameters for DDM- activated V1 agreed with those reported for the F1F0-ATP synthase from Escherichia coli [41] or Bos taurus [42–44]; while parameters of V2 from U. maydis resembles those of Polytomella sp. dimer [35]. Regarding to the role of the subunit interphase in the enzyme ac- tivity, it has been described that intersubunit contacts play a significant role in the catalytic properties of Escherichia coli glucosamine-6-P dea- minase, particularly in the allosteric equilibrium between the R and T- state [45]. Similarly, in the 3-Deoxy-D-manno-octulosonate-8-P syn- thase the subunit interphase is important for substrate selectivity and binding; the mutation of amino acid residues in this region produces a reduction in protein stability [46]. The deletion of the dimerizing subunit g in U. maydis doesn't prevent complex V dimerization, but isolated V2 from the mutant strain showed a large decrease in ATPase activity comparing to the V2 from WT, and mutant V2 was not activated by DDM (Esparza-Perusquia, manuscript in preparation). Since subunit g is located at the monomers interphase [23], its deletion may promote a change in the structure of the interphase that affects the ATPase ac- tivity. Interestingly, ATPase activity of V1 was similar in both U. maydis strains (Esparza-Perusquia, manuscript in preparation). Thermal inactivation of V2 and V1 showed a Tm = 52 °C for both oligomers, and a single inactivation transition was observed. Analysis of the inactivation by hrCN-PAGE showed the dissociation of V2 into V1 as an early step in the inactivation; later, V1 was denatured without the release of an active F1, contrary to what has been observed for the Polytomella sp. dimer [35]. In this sense, the interaction between monomers didn't increase the stability of the dimer. Under physiological conditions, the monomer-monomer interface may have two possible roles: 1) a well-supported function of the dimer in the cristae folding, and 2) optimization of V2 activity and the in- crease of the specificity constant of the F1F0-ATP synthase. If the cel- lular requirement of ATP increases and there is an optimal ΔμH+, V2 could be activated 7-times in the ATP synthesis direction, compared with the monomer; contrary, if ATP synthesis is not required, V2 shows a lower ATPase activity than the monomer. In this sense, it would be interesting to study the synthesis of ATP by V1 and V2 incorporated into liposomes, but the presence of high detergent concentrations during liposomes preparation process will disturb the monomer-monomer in- terface. Transparency document The Transparency document associated with this article can be found, in online version. Acknowledgments We dedicate this work as a memorial to Dr. Guillermo Mendoza- Hernandez, an exceptional friend and colleague, and coauthor of this paper, who passed away suddenly on July 13, 2012. This work was supported by Dirección General de Asuntos del Personal Académico (DGAPA) (IN214914, IN209614 and IN211715) from Universidad Nacional Autónoma de México (UNAM) and Consejo Nacional de Ciencia y Tecnología (CONACyT) (168025). Mercedes Esparza- Perusquía is a PhD student of the Programa de Doctorado en Ciencias Biológicas (511021118), UNAM and supported by the CONACyT through a doctoral scholarship (254400). The authors thank to the Fig. 4. Thermal inactivation curves of V1 and V2. V1 (open cir- cles) or V2 (closed circles) were incubated for 15 min at in- dicated temperatures and then cooled to room temperature. Activity assay was as described in Fig. 2E and F with 5 mM of Mg-ATP. A) Residual activity (V/V0) was calculated using V0 as the ATPase activity at 25 °C before heated. Tm = 52.3 °C. BN- PAGE of V1 (B) and V2 (C) incubated as described in (A). The data are the average of three replicates from four independent preparations. Error bars represent S.D. M. Esparza-Perusquía et al. BBA - Bioenergetics 1858 (2017) 975–981 980 Posgrado en Ciencias Biológicas, UNAM for the academic support. This manuscript partially fulfils the requirements of MEP to obtain her Ph.D. degree. Conflict of interest The authors declare no conflict of interest. Author contribution Esparza-Perusquía M and Flores-Herrera O designed and performed principal experiments; Olvera-Sánchez S, technical assistance; Pardo JP and Martínez F analyzed data; Mendoza-Hernández G carried out mass spectrometric analyses, Flores-Herrera O supervised project and wrote the paper with contributions from all authors. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.bbabio.2017.09.002. References [1] J. Velours, G. Arselin, The Saccharomyces cerevisiae ATP synthase, J. Bioenerg. Biomembr. 4 (2000) 383–390. [2] J.E. Walker, V.K. Dickson, The peripheral stalk of the mitochondrial ATP synthase, Biochim. Biophys. Acta 1757 (2006) 286–296. [3] J. Symersky, V. Pagadala, D. Osowski, A. Krah, T. Meier, J.D. Faraldo-Gómez, D.M. Mueller, Structure of the c10 ring of the yeast mitochondrial ATP synthase in the open conformation, Nat. Struct. Mol. Biol. 19 (2012) 485–491. [4] R.J. Devenish, M. Prescott, A.J. Rodgers, The structure and function of mitochon- drial F1F0-ATP synthases, Int. Rev. Cell Mol. Biol. 267 (2008) 1–58. [5] I. Arnold, K. Pfeiffer, W. Neupert, R.A. Stuart, H. Schägger, Yeast mitochondrial F1F0-ATP synthase exists as a dimer: identification of three dimer-specific subunits, EMBO J. 17 (1998) 7170–7178. [6] H. Eubel, L. Jänsch, H.P. Braun, New insights into the respiratory chain of plant mitochondria. Supercomplexes and a unique composition of complex II, Plant Physiol. 133 (2003) 274–286. [7] F. Krause, N.H. Reifschneider, S. Goto, N.A. Dencher, Active oligomeric ATP syn- thases in mammalian mitochondria, Biochem. Biophys. Res. Commun. 329 (2005) 583–590. [8] P. Paumard, J. Vaillier, B. Coulary, J. Schaeffer, V. Soubannier, D.M. Mueller, D. Brèthes, J.P. di Rago, J. Velours, The ATP synthase is involved in generating mitochondrial cristae morphology, EMBO J. 21 (2002) 221–230. [9] M.F. Giraud, P. Paumard, V. Soubannier, J. Vaillier, G. Arselin, B. Salin, J. Schaeffer, D. Brèthes, J.P. di Rago, J. Velours, Is there a relationship between the supramolecular organization of the mitochondrial ATP synthase and the formation of cristae? Biochim. Biophys. Acta 1555 (2002) 174–180. [10] I. Wittig, H. Schägger, Structural organization of mitochondrial ATP synthase, Biochim. Biophys. Acta 1777 (2008) 592–598. [11] J.L. Rubinstein, J.E. Walker, R. Henderson, Structure of the mitochondrial ATP synthase by electron cryomicroscopy, EMBO J. 22 (2003) 6182–6192. [12] W.C. Lau, L.A. Baker, J.L. Rubinstein, Cryo-EM structure of the yeast ATP synthase, J. Mol. Biol. 382 (2008) 1256–1264. [13] N.V. Dudkina, J. Heinemeyer, W. Keegstra, E.J. Boekema, H.P. Braun, Structure of dimeric ATP synthase from mitochondria: an angular association of monomers in- duces the strong curvature of the inner membrane, FEBS Lett. 579 (2005) 5769–5772. [14] N.V. Dudkina, S. Sunderhaus, H.P. Braun, E.J. Boekema, Characterization of di- meric ATP synthase and cristae membrane ultrastructure from Saccharomyces and Polytomella mitochondria, FEBS Lett. 580 (2006) 3427–3432. [15] F. Minauro-Sanmiguel, S. Wilkens, J.J. García, Structure of dimeric mitochondrial ATP synthase: novel F0 bridging features and the structural basis of mitochondrial cristae biogenesis, Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 12356–12358. [16] D. Thomas, P. Bron, T. Weimann, A. Dautant, M.F. Giraud, P. Paumard, B. Salin, A. Cavalier, J. Velours, D. Brèthes, Supramolecular organization of the yeast F1F0- ATP synthase, Biol. Cell. 100 (2008) 591–601. [17] N. Buzhynskyy, P. Sens, V. Prima, J.N. Sturgis, S. Scheuring, Rows of ATP synthase dimers in native mitochondrial inner membranes, Biophys. J. 93 (2007) 2870–2876. [18] M. Strauss, G. Hofhaus, R.R. Schröder, W. Kühlbrandt, Dimer ribbons of ATP syn- thase shape the inner mitochondrial membrane, EMBO J. 27 (2008) 1154–1160. [19] R. Fronzes, T. Weimann, J. Vaillier, J. Velours, D. Brèthes, The peripheral stalk participates in the yeast ATP synthase dimerization independently of e and g sub- units, Biochemistry 45 (2006) 6715–6723. [20] G. Arselin, M.F. Giraud, A. Dautant, J. Vaillier, D. Brèthes, B. Coulary-Salin, J. Schaeffer, J. Velours, The GxxxG motif of the transmembrane domain of subunit e is involved in the dimerization/oligomerization of the yeast ATP synthase complex in the mitochondrial membrane, Eur. J. Biochem. 270 (2003) 1875–1884. [21] I. Wittig, H. Schägger, Advantages and limitations of clear-native PAGE, Proteomics (2005) 4338–4346. [22] D.M. Bustos, J. Velours, The modification of the conserved GXXXG motif of the membrane-spanning segment of subunit g destabilizes the supramolecular species of yeast ATP synthase, J. Biol. Chem. 280 (2005) 29004–29010. [23] S.J. Couoh-Cardel, S. Uribe-Carvajal, S. Wilkens, J.J. García-Trejo, Structure of dimeric F1F0-ATP synthase, J. Biol. Chem. 285 (2010) 36447–36455. [24] L. Tomasetig, F. Di Pancrazio, D.A. Harris, I. Mavelli, G. Lippe, Dimerization of F0F1-ATP synthase from bovine heart is independent from the binding of the in- hibitor protein IF1, Biochim. Biophys. Acta 1556 (2002) 133–141. [25] E. Zerbetto, L. Vergani, F. Dabbeni-Sala, Quantification of muscle mitochondrial oxidative phosphorylation enzymes via histochemical staining of blue native polyacrylamide gels, Electrophoresis 18 (1997) 2059–2064. [26] E. Bisetto, F. Di Pancrazio, M.P. Simula, I. Mavelli, G. Lippe, Mammalian ATPsynthase monomer versus dimer profiled by blue native PAGE and activity stain, Electrophoresis 28 (2007) 3178–3185. [27] O. Juárez, G. Guerra, F. Martínez, J.P. Pardo, The mitochondrial respiratory chain of Ustilago maydis, Biochim. Biophys. Acta 1658 (2004) 244–251. [28] W.F. Waterfield, H.D. Sisler, A convenient procedure for rapid release of protoplasts from Ustilago maydis, BioTechniques 6 (1988) 832–834. [29] I. Wittig, M. Karas, H. Schägger, High resolution clear native electrophoresis for in- gel functional assays and fluorescence studies of membrane protein complexes, Mol. Cell. Proteomics 6 (2007) 1215–1225. [30] C. Jung, C.M. Higgins, Z. Xu, Measuring the quantity and activity of mitochondrial electron transport chain complexes in tissues of central nervous system using blue native polyacrylamide gel electrophoresis, Anal. Biochem. 286 (2000) 214–223. [31] H. Schägger, W.A. Cramer, G. von Jagow, Analysis of molecular masses and oli- gomeric states of protein complexes by blue native electrophoresis and isolation of membrane protein complexes by two-dimensional native electrophoresis, Anal. Biochem. 217 (1994) 220–230. [32] W.S. Rasband, ImageJ, U.S. National Institutes of Health, Bethesda, Maryland, USA, 1997–2016http://imagej.nih.gov/ij/. [33] M. González-Zamorano, G. Mendoza-Hernández, W. Xolalpa, C. Parada, A.J. Vallecillo, F. Bigi, C. Espitia, Mycobacterium tuberculosis glycoproteomics based on ConA-lectin affinity capture of mannosylated proteins, J. Proteome Res. 8 (2009) 721–733. [34] P. Kuzmic, DynaFit–a software package for enzymology, Methods Enzymol. 467 (2009) 247–280. [35] A. Villavicencio-Queijeiro, J.P. Pardo, D. González-Halphen, Kinetic and hysteretic behavior of ATP hydrolysis of the highly stable dimeric ATP synthase of Polytomella sp, Arch. Biochem. Biophys. 575 (2015) 30–37. [36] P. Cardol, D. González-Halphen, A. Reyes-Prieto, D. Baurain, R.F. Matagne, C. Remacle, The mitochondrial oxidative phosphorylation proteome of Chlamydomonas reinhardtii deduced from the Genome Sequencing, Project, Plant Physiol. 137 (2005) 447–459. [37] M. Vázquez-Acevedo, P. Cardol, A. Cano-Estrada, M. Lapaille, C. Remacle, D. González-Halphen, The mitochondrial ATP synthase of chlorophycean algae contains eight subunits of unknown origin involved in the formation of an atypical stator-stalk and in the dimerization of the complex, J. Bioenerg. Biomembr. 38 (2006) 271–282. [38] J. Velours, A. Dautant, B. Salin, I. Sagot, D. Brèthes, Mitochondrial F1F0-ATP syn- thase and organellar internal architecture, Int. J. Biochem. Cell Biol. 41 (2009) 1783–1789. [39] D. De los Ríos-Castillo, M. Zarco-Zavala, S. Olvera-Sanchez, J.P. Pardo, O. Juarez, F. Martinez, G. Mendoza-Hernandez, J.J. García-Trejo, O. Flores-Herrera, Atypical cristae morphology of human syncytiotrophoblast mitochondria: role for complex V, J. Biol. Chem. 286 (2011) 23911–23919. [40] D.L. Feinstein, E.N. Moudrianakis, Response of the adenosine triphosphatase ac- tivity of the soluble latent F1 enzyme from beef heart mitochondria to changes in Mg2+ and H+ concentrations, J. Biol. Chem. 259 (1984) 4230–4236. [41] D.L. Foster, M.E. Mosher, M. Futai, R.H. Fillingame, Subunits of the H+-ATPase of Escherichia coli. Overproduction of an eight-subunit F1F0-ATPase following induc- tion of a lambda-transducing phage carrying the unc operon, J. Biol. Chem. 255 (1980) 12037–12041. [42] P. Swanljung, L. Frigeri, K. Ohlson, L. Ernster, Studies on the activation of purified mitochondrial ATPase by phospholipids, Biochim. Biophys. Acta 305 (1973) 519–533. [43] F. Zanotti, A. Gnoni, R. Mangiullo, S. Papa, Effect of the ATPase inhibitor protein IF1 on H+ translocation in the mitochondrial ATP synthase complex, Biochem. Biophys. Res. Commun. 384 (2009) 43–48. [44] H.S. Penefsky, Differential effects of adenylyl imidodiphosphate on adenosine tri- phosphate synthesis and the partial reactions of oxidative phosphorylation, J. Biol. Chem. 249 (1974) 3579–3585. [45] D.A. Cisnerosa, G.M. Montero-Moran, S. Lara-Gonzalez, M.L. Calcagno, Inversion of the allosteric response of Escherichia coli glucosamine-6-P deaminase to N-acet- ylglucosamine 6-P, by single amino acid replacements, Arch. Biochem. Biophys. 421 (2004) 77–84. [46] T.M. Allison, F.C. Cochrane, G.B. Jameson, E.J. Parker, Examining the role of in- tersubunit contacts in catalysis by 3-deoxy-D-manno-octulosonate 8-phosphate synthase, Biochemistry 52 (2013) 4676–4686. M. Esparza-Perusquía et al. BBA - Bioenergetics 1858 (2017) 975–981 981 Deletion of subunit g affects the stability of the dimeric state of the F1FO-ATP synthase and decreases the mitochondrial ATP synthesis. F1FO-ATP synthase complex interactions in vivo can occur in the absence of the dimer specific subunit g Esparza-Perusquía Ma, Langner Tb, Feldbrügge Mc, Pardo JPa, Martínez Fa, and Flores-Herrera Oa1 a Departamento de Bioquímica, Facultad de Medicina, Universidad Nacional Autónoma de México, Apartado Postal 70-159, Coyoacán 04510, México, D. F., México. b The Sainsbury Laboratory, Norwich Research Park, Norwich NR4 7UH, United Kingdom. c Institute for Microbiology, Cluster of Excellence on Plant Sciences, Department of Biology, Heinrich-Heine University Düsseldorf, Düsseldorf, Germany. To whom correspondence should be addressed: Oscar Flores-Herrera. Departamento de Bioquímica, Facultad de Medicina, Universidad Nacional Autónoma de México, Apdo. Postal 70- 159, Coyoacán 04510, México, Cd. Mx., México; Phone: 55-56232510; Fax: 55-56162419; E-mail: oflores@bq.unam.mx Keywords: Abstract F1FO-ATP synthase is the transductor of electrochemical proton gradient to ATP and occurs in the inner membrane mitochondria as a monomer and dimer. The dimer shows a higher ATPase activity than monomer, and is involved with the cristae folding. The interface monomer-monomer is constituted by subunits a, i/j, e, g, and k. The role of the subunit g in a respiratory strict organism is unknown. A gene knockout was generated in order to verify subunit g role in the mitochondrial metabolism and cristae architecture of Ustilago maydis. Deletion of subunit g does not affect cell growth, glucose consumption, and biomass production. Ultrastructure analyzes showed that mitochondrial size and cristae shape was similar to wild strain. The membrane potential from wild type and mutant had a similar magnitude, but oxygen consumption was higher in mutant strain. ATP synthesis was 20% lower in mutant strain. Additionally, mutant strain expresses the AOX since exponential face of growth, and a lower ROS production was observed. Dimer from mutant strain was unstable during digitonin solubilization avoiding its isolation and kinetics characterization. Isolate monomeric state showed similar kinetics constant to monomer from WT strain. The ATP synthesis decrease and the AOX presence suggests that deletion of subunit g induces a ROS stress. 1. Introduction The F1FO-ATP synthase (Complex V) of mitochondria, chloroplast and eubacteria use the energy of the proton electrochemical gradient for ATP synthesis from ADP and Pi. This enzyme is constituted by a catalytic sector (called F1), a membrane sector (called F0) and two connecting stalks (central and peripheral). The F1 sector is a water-soluble entity composed of subunits c, d, i, f, and g. The F0 sector is embedded in the membrane and is composed by the hydrophobic subunits a, b, c, f, and 8. The central stalk is constituted by i, f, and g subunits; while the peripheral stalk by subunits 4, h, d, f, and OSCP. The functional enzyme occurs when F1 and F0 sectors are coupled by connecting stalks and the whole complex could be considered as an H+-pumping ATP synthase or ATPase (1). The F1FO-ATP synthase is a nano-motor which uses the proton translocation through F0 sector to promote the rotation of the ring of hydrophobic subunit c and the central stalk inside the F1 sector. Additionally, regulation of catalytic sector is carried out by the natural inhibitor peptide called IF1 in mammalian or Inh1 in yeast. Supernumerary subunits i/j, e, g, and k have been identify associated to F0 sector (2, 3) and are involved in dimerization of F1FO-ATP synthase (3- 8). Dimer of complex V (V2) folds the inner membrane into mitochondrial cristae (9) and its dissociation in monomer (V1) is related to loss of cristae architecture. Particularly, deletion of subunits g and e in Saccharomyces cerevisiae produces mitochondria with onion-like structures (10, 11). It has been observed that these mutants growth in non-fermentable media (3) suggesting that dimerizing subunits g and e could not be involved in the activity of complex V. Nevertheless, it is widely accepted that the mitochondrial cristae architecture is important to generate the membrane potential (F[m) used for the ATP synthesis. Then, are the mitochondria with onion-like cristae able to synthesize ATP? In this work deletion of dimerizing subunit g (Fg) was performed in Ustilago maydis, a strict aerobic basidiomycete, and the analysis of mitochondrial ultrastructure, membrane potential (F[m), oxygen uptake, ATP synthesis, and H2O2 production were performed. Quantification of mitochondrial complex V content was performed in both strains, and the presence of V2 and V1 was determined as reported previously by our group (12). Results show that mutant strain preserves their mitochondrial network and cristae ultrastructure, but have a lower F1FO-ATP synthase amount. An unstable dimer of complex V (V2Fg) was observer over digitonine solubilization of OXPHOS complexes. Kinetics characterization demonstrate that the monomeric state of complex V from WT and Fg strains have similar kinetics parameters. Bioenergetics analysis shows that ATP synthesis decrease in mutant mitochondria while F[m remains constant. Interestingly, Fg strain expresses the alternative oxidase (AOX) since log-phase of growth. The role of subunit g in the stability and function of V2 is discussed. 2. Materials and methods 2.1 Standard molecular biology techniques and Fg strain generation To generate a gene replacement construct, the module of choice is fused to PCR-amplified flanks of about 1 kb using the restriction enzymes SfiI and SwaI and gene was replaced for the hygR cassette. Oligos RL615_um00975 U2 (atttAAATGCTTCCTTGTATTCGGC- small letters indicated the restriction site for SwaI) and RL616_um00975 U3 (ggccatctAGGCCGATGACCGTATTACCCGAAAGAC- small letters indicated the restriction site for SfiI) were used to UF (upstream flank PRC-amplification) and RL617_um00975 D1 (ggcctgagTGGCCACGCTCGACAATTGAATTCG- small letters indicated the restriction site for SfiI) and RL618_um00975 D2 (atttAAATCTGGCATGTGCTCACC - small letters indicated the restriction site for SwaI). Strain was constructed by transformation of FB2 strain with linearized plasmid pUMa1704. After transformation the cells were streaked on solid 1% glucose, 1% yeast extract, 0.25% bactopeptone, 2% agar, 25 og Hygromycin. All homologous integration events were verified by Southern blot analysis. 2.2 Cell culture Cultivation was performed using standard techniques. Growth conditions for U. maydis strain and source of antibiotics were described elsewhere (13). The Fg strain was growth at 28ºC in complete medium (1% glucose, 1% yeast extract and 0.25% bactopeptone) (14). From this solid culture, an inoculum was used to grow cells in 100 ml of YPD (0.5% yeast extract, 0.25% bactopeptone, 0.5% glucose) for 20-24 h at 28oC and then transfer 20-30 absorbance units (600 nm) of this suspension to 1 liter of YPD medium, to continue the growth of cells for 8 or 24 h at 28oC in a gyratory shaker at 200 rpm. Aliquots were withdrawn and their absorbance at 600 nm was determinate for cell growth curve evaluation. Alternatively, glucose consumption was evaluated using a kit (Glucose- LQ, Spinreact) following the manufacturer instructions. The strain was preserve at -70ºC with 30% glycerol. 2.3 Mitochondrial ultrastructure analyzes Cell pellets were fixed in 4% paraformaldehyde/2.5% glutaraldehyde for 1 h at room temperature; then they were washed two times (10 min each) in PBS and stained with 1% osmium tetraoxide for 1 h at 4°C, followed by two washes with PBS and water. Total dehydration was made in ethanol (graded 50 Î 100%) and propylene oxide. Pellets were embedded in epoxy resin and cut into 70 nm sections. For immunoelectron microscopy, cell pellets were fixed with 4% paraformaldehyde/0.1% glutaraldehyde for 30 min at 4°C. After ethanol dehydration (graded 70 Î 100%) pellets without osmication were embedded in LR White. Ultra-thin sections were mounted on nickel grids, blocked with 5% fat free milk in TBST for 15 min and probed with a 1∶10 dilution of anti LC3 (Cell Signaling Technology Inc., Danvers, MA, USA) at 4°C, overnight. A 1∶10 dilution of colloidal gold-conjugated secondary antibody (GAR Auroprobe, Amersham) was incubated for 2 h at room temperature. Sections were washed with TBST and PBS, post-fixed with 1% glutaraldehyde in PBS, thoroughly washed and stained with 2% aqueous uranyl acetate. A control without first antibody was included. Electron microscopy was performed at 80 kV on a Zeiss EM900 Transmission Electron Microscope. Images were recorded with a Gatan Dual Vision CCD 300W camera (Gatan, Pleasanton, CA). 2.4 Oxygen consumption measurement Oxygen uptake was estimated polarographically using a Clark type electrode in 10 mM KH2PO4, pH 7.4, and 1 mg of wet weight of wild type strain or mutant strain cells per ml was added (15). Temperature was set at 30°C. Oxygen consumption was supported by endogenous substrates. Were indicates, 1 mM cyanide (KCN) or 50 oM n-octil-galate was added to inhibit respiration. Alternatively, the oxygen uptake was assayed using isolated mitochondria in a solution composed by 500 mM sucrose, 20 mM MgCl2, 20 mM KH2PO4, 2 mM EGTA, 0.2% BSA, pH 7.4. Maximum activity of complex IV was assayed with 4 mM ascorbate and 6 mM 2,3,5,6-tetramethyl-p- phenylendiamine (TMPD) to reduce the horse heart cytochrome c. 2.5 Cell permeabilization U. maydis cells (125 mg wet weight/ml) from WT and Fg strains were permeabilized with 0.02% digitonin in 300 mM sorbitol, 10 mM HEPES, 1 mM EGTA, 7 mM MgSO4, 150 mM KCl, 10 mM KH2PO4, pH 7.4, at 25oC during 20 min with constant stirring. 2.6 Membrane potential determination The membrane potential (F[m) of permeabilized U. maydis cells (125 mg wet weight/ml) was determined in the medium described for cell permeabilization and supplemented with 10 oM Safranine O. Succinate (10 mM, pH 7.4) was added to F[m generation (16). Where indicated, 1 mM ADP was added to verify the membrane potential depolarization coupled to oxygen uptake and ATP synthesis stimulation. CCCP (5 oM) was added to abolish the membrane potential. The membrane potential was evaluated in a double beam spectrophotometer (Aminco DW 2000, Olis, Inc.) by using the difference of wavelengths between 533Î511 nm. The recording was performed in a 3.0 ml cuvette with constant stirring and the temperature was held at 25oC. 2.7 Mitochondrial ATP synthesis ATP synthesis was performed at 25oC using permeabilized U. maydis cells (125 mg wet weight/ml) in the medium described for cell permeabilization supplemented with 10 mM succinate, 100 oM P1,P5-Di(adenosine-5Ó)renvarhourhave ammoniwm ualv. ATP u{nvheuiu yau uvatved b{ vhe addivion of 5 mM ADP and aliquots were withdrawn at different times and mixed with perchloric acid (7% final concentration) for stop the reaction. The samples were spin-out in a refrigerate Eppendorf microfuge at 13,000 rpm during 5 min. Supernatant was recovered and pH adjusted to 7.0 with KOH. To remove the potassium perchlorate the samples were spin-out again, and supernatant recovered and pH adjusted to 7.4. ATP content in each sample was quantify with an assay coupled to the reduction of NADP+ (g340 nm = 6.2 mM-1 ·cm-1). The reaction mixture contained 0.5 mM NADP+, 6 units/ml glucose-6-phosphate dehydrogenase, 16 units/ml hexokinase, 10 mM glucose, 5 mM MgCl2, 10 mM KH2PO4, pH 7.5. 2.8 Mitochondria isolation The method described by Waterfield and Sisler (17) was used with minor modifications. Cells were grown in YPD medium for 24 h, collected by centrifugation (3,800 g for 10 min), washed twice with distilled water, and the pellet suspended with 0.6 M ammonium sulfate (11.25 ml/g wet weight). To produce protoplasts, 1 g of wet weight of U. maydis cells suspension were incubated with 0.016 g of Trichoderma harzianum lytic enzymes, during 60 min at 30oC. The protoplasts were recovered by centrifugation at 3,800 g for 10 min, and suspended in 0.8 M sucrose, 10 mM Tris, 2 mM EDTA, 3% BSA, and 20 mM KH2PO4, pH 7.0; this suspension was centrifuged in similar condition described above to wash out the lytic enzymes. The pelleted protoplasts were suspended in 0.4 M sucrose, 10 mM Tris, 2 mM EDTA, 3% BSA, and 20 mM KH2PO4, pH 7.0 (buffer A). PMSF (1 mM) and 25 ol/g wet weight of the protease inhibitors cocktail (fungal and yeast protease inhibitor cocktail, Sigma p8215) were added to the suspension and homogenized 15-20 times in a Teflon potter homogenizer. The homogenization extract was 3-fold diluted with buffer A and centrifuged at 3,800 g for 10 min at 4oC. The supernatant was centrifuged at 17,000 g for 10 min and the mitochondrial pellet suspended in buffer A to a protein concentration of 20-30 mg/ml. 2.9 Solubilisation of OXPHOS complexes The OXPHOS complexes from U. maydis mitochondria were solubilized with digitonin, a very mild detergent, as described (12, 18-20). Briefly, two mg of mitochondrial protein were suspended in 0.2 ml of 50 mM Bis-Tris and 500 mM 6-aminocaproic acid, pH 7.0, supplemented with 10 mM succinate and 10 mM ATP and solubilized with increasing digitonin concentration in a ratio of 0.5, 1, 2, 3, and 5 mg of digitonin per mg of protein. The digitonin was added drop by drop and the mixtures were incubated for 30 min at 4oC with gentle stirring, and then centrifuged at 100,000 g for 30 min at 4oC. The supercomplexes and complexes were recovered from supernatant and analyzed by blue native PAGE (BN-PAGE) and clear native PAGE high resolution (hrCN-PAGE) as described (18-20). 2.10 F1FO-ATP synthase complexes isolation Mitochondrial supercomplexes and complexes from U. maydis were solubilized with a digitonin/protein ratio of 2:1 and processed as described by Esparza-Perusquia et. al (2017). Digitonin-solubilized mitochondrial supercomplexes and complexes (16 mg of protein) were loaded on 24 ml of a continuous sucrose gradient (16 Î 42% sucrose, 15 mM Tris, pH 7.4, 20 mM KCl, and 0.2% digitonin) and centrifuged at 131,000 g for 16 h at 4°C. Fractions (0.5 ml) were collected from the bottom of the gradient. The presence of V2 and V1 in each gradient-fraction was determined by in-gel activity of ATPase from BN-PAGE and hrCN-PAGE. If applicable, the fraction containing each oligomer were pooled separately, and diluted 7 folds with 30 mM HEPES, pH 8.0 and 5% glycerol; then concentrated using a Centrifugal Filters Units (100K, Millipore Amicon Ultra) to a final volume of 100 ol, and stored at -70oC until used. The V2Fg was so unstable and scarce that it was used immediately as gradient-fractionation was done and its presence detected. The anode buffer for BN-PAGE was 50 mM Bis-Tris/HCl, pH 7.0; the cathode buffer contained 50 mM tricine, 15 mM Bis-Tris, pH 7.0 and the anionic Coomassie dye (0.02%). For hrCN-PAGE the anode buffer contained 25 mM imidazole/HCl, pH 7.0; the cathode buffer was 50 mM Tricine, 7.5 mM imidazole, pH 7.0, 0.01% dodecyl-く-D-maltopyranoside (DDM) and 0.05% deoxycholate (DOC). The gel front was visualized with Ponceau Red (21). The electrophoresis (BN or hrCN) were performed at 4oC and the voltage was set to 35 V for 10 h and stopped when the sharp line of the dye approached the gel front. Molecular weight of the respiratory complexes and supercomplexes was determined by their electrophoretic mobility and in-gel catalytic activity, using the complexes of digitonine-solubilized bovine heart mitochondria as standards. The concentration of F1FO-ATP synthase in each V1 and V2 sample was determined by a densitometry analysis of Coomassie© Brilliant Blue R-125 stained c- and d-subunits from an SDS- Tricine-PAGE, using Coomassie stained BSA as a standard (See Supplemental material section, Figure S1). Densitometry analysis was performed with the My Image Software Thermo Fisher Scientific Inc, 2014. 2.11 In-gel catalytic activity assays The in-gel assays of digitonine-solubilized supercomplexes and complexes from U. maydis mitochondria were performed as described by Jung (19) using gel strips loaded with 150 µg of protein. NADH dehydrogenase activity (NADH:methylthiazolyldiphenyl tetrazolium bromide (MTT) oxidoreductase) was assayed in a buffer containing 1.2 mM MTT and 1.0 mM NADH in 10 mM Tris/HCl, pH 7.4, at 20 Î 25oC. For succinate dehydrogenase activity (Succinate:MTT oxidoreductase) NADH was replaced by 10 mM succinate, in a buffer of 10 mM K2HPO4, pH 7.4, 5 mM EDTA and 0.2 mM phenazine methosulfate (PMS). NADH or succinate dehydrogenase activity was correlated with the development of purple precipitates on the gel. Activity was monitoring as purple-staining appear (10 Î 20 min) and then the reaction was stopped with fixing solution (50% methanol, 10% acetic acid). The in-gel activity of complex IV was assayed in 50 mM K2HPO4, rH 7.2, 4.7 mM 3,3Ódiaminoben¦idine vevtah{dtochlotide (DAB) and 16 oM horse heart cytochrome c, during 30 Î 40 min of incubation at 20 Î 25oC. The activity was observed as a brown precipitate and the reaction was stopped with the fixing solution. Activity of complex V was assayed in 50 mM glycine (adjusted to pH 8.0 with triethanolamine), 10 mM MgCl2, 0.15% Pb(ClO4)2 and 5 mM ATP. ATP hydrolysis correlated with the appearance of a white lead phosphate precipitates. The reaction was stopped using 50% methanol, and subsequently the gel was transferred to water and scanned against a dark background as described previously (18-19). 2.12 2D-Tricine-SDS gel electrophoresis 2D-Tricine-SDS-polyacrylamide gel electrophoresis (2D-SDS-PAGE) was performed according to Schägger (20). Proteins from a native PAGE gel strip loaded with 150 og of digitonine-solubilized supercomplexes and complexes were separated by 2D-Tricine-SDS-PAGE on a 16% polyacrylamide gel under denaturing conditions. After the run, the proteins were stained with Coomassie© Brillant Blue R-125. 2.13 Measurements of ATP hydrolysis by F1FO-ATP synthase complexes ATP hydrolysis by F1FO-ATP synthase complexes was measured spectrophotometrically in an Agilent 8453 UV-visible spectrophotometer (Agilent Technologies, USA) as described by (12), using an assay coupled to the oxidation of NADH (g340nm = 6.2 mM-1 ·cm-1). The assay medium contained 50 mM HEPES, pH 8.0, 1 mM MgSO4, 90 mM KCl, and the temperature was hold at 25°C. The ATP-regenerating system was 5 mM phosphoenolpyruvate (PEP), 1 mM NADH, 50 units/ml pyruvate kinase (PK), and 30 units of lactate dehydrogenase (LDH)/ml. The ATPase reaction was started by the addition of an equimolar mixture of Mg-ATP complex. The NADH absorbance was continuously monitored and the time response, checked by ADP additions, was less than 1 s. The slope of the linear region of each spectrophotometric recording was used to obtain the initials velocities. The linear region of the traces was corroborated with plotting the first-derivative against time. Since a non-specific NADH oxidation was observed with no enzyme addition, the actual value of ATPase activity was calculated as the subtraction of the slope of this region from the slope that is obtained after adding the enzyme. Oligomycin (3 oM) was added to inhibit ATPase activity to verify F1FO-ATP synthase integrity. Only samples with ATPase activity 100% inhibited by oligomycin were used in this study. Data for the ATP hydrolysis were analyzed by robust, weighted, non-linear regression analysis using the SigmaPlot software (Systat Software, Inc., version 10.0), and represent the average of three replicates from seven independent experiments. 2.14 Oligomycin inhibition of F1FO-ATP synthase Continues monitoring of ATP hydrolysis by F1FO-ATP synthase dimer or monomer coupled to NADH oxidation was carried out as described by Esparza-Perusquia et al (12). ATP hydrolysis was initiated by Mg-ATP addition; after 2-4 min oligomycin was added and ATPase activity decayed. Because oligomycin is a slow-union inhibitor the total recording time for maximal ATPase activity inhibition, at different oligomycin concentrations, was about 25-35 min. The final spectrophotometric recording was fitted to a monoexponential decay of the ATPase activity as described by (12). 2.15 Determination of protein concentration. Samples were treated with 0.4% deoxycholate and the protein content was determined as described by Lowry et al. ]56_. Bovine serum albumin (BSA) was used as standard. 3. RESULTS 3.1 Deletion of the g subunit from the Ustilago maydis genome The g subunit gene was deleted by homologous recombination as reported by Brachmann et. al., (13) using a hygromycin resistant gene (Hyg). The correct construction of knock-out plasmid (pUMa1704) was verify by SfiI enzyme restriction activity (Fig. 1A), which produced one band of 1.88 Kb (Hyg gene resistant) and other of 6.37 Kb (the UF-DF vector). The U. maydis cells transformed with the plasmid pUMa1704 were selected by hygromycin resistant (Fig. 1B). The deletion of the g subunit gene by the correct insertion of the hygromycin resistant gene was determined by PCR using the UF and DF flanks described in materials and methods section. The presence of the 1.5 and 1.7 amplified confirm the correct insertion of the Hyg-gene (Fig. 1C). Finally, the genomic DNA (gDNA) from WT and Fg mutant strains was processed by the restriction enzyme NcoI and the product analyzed by Southern Blot (Fig. 1D). As the Hyg gene has a cutoff point for the NcoI, two PCR product were identified from Fg mutant gDNA while only one from WT strain (Fig. 1D). These results showed that g subunit was successfully deleted from U. maydis genome and is not present in the Fg strain. 3.1 Growth of Fg mutant strain The elimination of subunit g from F1FO-ATP synthase in Ustilago maydis was obtained as described in methods section, and growth cells characterization of mutant strain included the doubling time, number of cell, glucose consumption, dry weight, and cell length. The doubling time was similar for both strain (2.9 ‒ 0.7 h and 2.4 ‒ 0.5 h for mutant and WT, respectively). However, the number of cells was quiet different; mutant strain produce 82% (3.37 X 109 cell·L-1) of the total number of WT cells (4.12 X 109 cell·L-1); however, the mutant cells were 1.5-times longer in the log-phase (28 ‒ 2 and 19 ‒ 2 µm, for the mutant and WT strains, respectively) and 1.3-times in the stationary-phase (30 ‒ 3 and 23 ‒ 4 µm, for the mutant and WT strains, respectively) than WT cells. This indicates that the mutant cells were longer that WT, but its cell production was lower. In this line of thought, the glucose consumption was similar for both strains in the log-phase (25 ‒ 3 and 24 ‒ 2 mmol by mutant and WT strain, respectively) or in the stationary-phase (15 ‒ 0.7 and 15 ‒ 0.9 mmol for mutant and WT strain, respectively). Interestingly, both strain reached a similar dry weight at stationary phase of growth (2.0 ‒ 0.3 and 2.0 ‒ 0.2 mg/ml, for mutant and WT strain, respectively). This suggested that carbon incorporation in both strains was similar, regardless that the cells in the mutant strain were longer but fewer. 3.2 Mitochondrial ultrastructure and bioenergetics of Fg mutant strain The mitochondrial cristae from Fg strain showed a tubular and lamellar ultrastructure, similar to WT (Fig. 2). Additionally, the F[m from Fg mutant was stimulated by succinate in a similar magnitude to WT mitochondria (Fig. 2). Simultaneously, the activity of the NADH:DBQ oxido- reductase (Complex I) and succinate:DBQ oxido-reductase (Complex II) was determined. WT strain showed a complex I activity of 0.18 ± 0.06 omol NADH oxidized·(mg·min)-1, while Fg mutant of 0.16 ± 0.03 omol NADH oxidized·(mg·min)-1. The complex II has an activity of 0.05 ± 0.01 omol DCPIP reduced·(mg·min)-1 from WT and of 0.04 ± 0.01 omol DCPIP reduced·(mg·min)- 1 from mutant strain. In this condition the ATP synthesis was 20 ± 1 omol ATP·(mg·min)-1 and 15 ± 0.9 omol ATP·(mg·min)-1 for the WT and Fg strains, respectively. This difference (25%) in the ATP synthesis suggested that 1) respiratory oxygen uptake and ATP synthesis could be uncoupled, or 2) the F1FO-ATP synthase activity is decreased. As a first step to explore these hypotheses, mitochondrial oxygen uptake at log- and stationary-phase of growth (i.e. 8 and 24 hours, respectively) was determined (Fig. 2). At 8 hours of growth, the respiration was similar in both strains (112 ± 15 ng atoms O and 130 ± 17 ng atoms O, for WT and Fg strains, respectively); however, the cyanide inhibits only the WT strain respiration (Fig. 2). Maximum activity of complex IV was stimulated with ascorbate, TMPD, and cytochrome c as described in methods section. Activity of complex IV was higher in WT mitochondria (i.e. 1048 ‒ 21 ng atoms O·(mg·min)-1) than in Fg strains mitochondria (i.e. 827 ‒ 18 ng atoms O·(mg·min)-1) (Table 1). Total inhibition of oxygen uptake was reached with n-octyl- galate addition, indicating the presence of the alternative oxidase (AOX). AOX is a quinol:oxygen oxido-reductase but it is not a proton pump. At stationary state of growth, both strains express the AOX but their activity was sharp different, 124 ‒ 23 ng atoms O·(mg·min)-1 and 262 ‒ 4 ng atoms O·(mg·min)-1 for WT and Fg strains, respectively (Table 1). The decrease of the activity of complex IV (i.e. 20%) and the 2-time increase of AOX activity in the Fg strain, suggests that ROS stress could occur in the mutant strain. The AOX presence has been associated with ROS production prevention. To explore this possibility, the H2O2 production was determinate at 8 and 24 hour of growth (Table 1). Succinate or NADH was used as oxidizable substrate and the H2O2 production was 470 ‒ 80 nmol H2O2·(mg·min)-1 and 280 ‒ 30 nmol H2O2·(mg·min)-1 in WT and Fg strains, respectively; suggesting that in the Fg strain the AOX could avoid ROS production. Although mitochondrial ultrastructure and F[m was similar in both strains, the ATP synthesis was affect in the Fg mutant simultaneously to AOX expression since early state of growth. Deletion of the dimerizing g-subunit from complex V could be involved with the decrease of ATP synthesis, while AOX presence with the stress resistance. 3.3 Supercomplexes and complexes analyzes The respiratory supercomplexes and complexes from WT and Fg strains were solubilized with different ratio of digitonine/protein and analyzed by BN-PAGE (Fig. 3). The activity of NADH:MTT oxidoreductase from WT strain was associated to supercomplexes of high molecular weight but principally to a main band of 1600 kDa, and in lower proportion to free-complex I (i.e. 990 kDa) (Fig. 3B). Activity of succinate:MTT oxidoreductase from complex II was associated with a single band of 130 kDa (Fig. 3C). Free-complex IV activity was located at 200 kDa and supercomplexes also exhibits activity of complex IV (Fig. 3D), similarly to previously reported by our laboratory (22). The in-gel activity of the F1FO-ATP synthase from WT occurs as monomeric (V1) and dimeric (V2) state, with 660 and 1200 kDa, respectively (Fig. 3E). It is important to note that amount of V1 and V2 from the WT strain was very similar, determined as described in materials and methods section (vide infra) (Fig. 3A). The analysis of supercomplexes and complexes from Fg strain showed that in-gel activity of NADH:MTT oxidoreductase was mainly associated to free-complex I (990 kDa) and supercomplexes of low molecular weight (i.e. 1200 Î 1300 kDa) (Fig. 3B). Interestingly, monomeric complex II was clearly observed from a ratio digitonine/protein = 3 (Fig. 3C). The complex IV activity was associated to supercomplexes and a free-complex IV (Fig. 3D). The ATPase activity of F1FO-ATP synthase from Fg strain was associated to a single band of 660 kDa, corresponding to the V1 (Fig. 3E). The activity of the V2 was observed only after a 24 h of incubation in the in-gel ATPase activity medium (Fig. 3F), indicating that this oligomers showed a very low activity; indeed, in some replicates was not observed at all. A 2D-SDS-PAGE from the BN-PAGE lane showed that the amount of complex V from V2 was substantially diminished respect to the V1 (data no shown). To explore the kinetics of V2 from Fg strain, many attempts to isolate it were performed (Fig. 4A); however, due that V2 was very unstable it was impossible to isolate it in sufficient quantity to perform kinetics studies. Interestingly, the total amount of complex V in the mitochondria was 132 ± 6 og of complex V·(mg mitochondrial protein)-1 and 111 ± 7 og of complex V·(mg mitochondrial protein)-1 for WT and Fg, respectively. This suggests that deletion of subunit g decrease only a 16% the F1FO-ATP synthase amount in the Fg mitochondria but V2 was unstable and dissociate straightaway into single subunits during digitonine solubilization, without V1 accumulation. The kinetics characterization of the V1Fg was performed and it has had a KM of 155 ‒ 47 oM and a Vmax of 0.58 ‒ 0.03 omol hydrolyzed ATP·(mg F1FO-ATP synthase·min)-1. The kcat was 5.86 ‒ 0.05 s-1 and the kcat/KM of 3.78 X 104 ‒ 0.3 X 104 (M-1s-1). The V1Fg activated with DDM increases its Vmax to 2.3 ‒ 0.2 omol hydrolyzed ATP·(mg F1FO-ATP synthase·min)-1 with a KM value of 920 ‒ 20 oM. The values for kcat and specificity constant were 23.26 ‒ 0.1 s-1 and 2.53 X 104 ‒ 0.08 X 104, respectively. Additionally, the inhibition of V1Fg ATPase activity by oligomycin had a Ki value of 53 ‒ 4 nM. The kinetics parameters of V1Fg, its activation by DDM and oligomycin inhibition were similar to V1WT previously reported (12), suggesting that ATPase activity of V1Fg was not modified by deletion of subunit g. Dimerizing g-subunit deletion doeunÓv avoid V2Fg state of complex V (i.e. determined as mitochondrial cristae folding, generation of the F[m, and oxygen uptake) but could deteriorate the interphase monomer-monomer inducing an instable dimeric state, which probably could be related to lower ATP synthesis. 4. DISCUSSION The role of the dimeric state of F1FO-ATP synthase in the mitochondrial crista folding has been widely accepted (6-9). Actually, it has been described that the interface monomer-monomer of the yeast dimer is constituted by subunits a, i/j, k, g, and e (23). The interface can be divided into a central zone and a peripheral zone. In the central zone, the subunits i/j interact through two short stretches of ~10 residues; while interaction between subunits a occurs through two strands that constituted a four stranded planar structure with one hydrophobic surface and one hydrophilic surface (23). The peripheral zone of the monomer-monomer interface is constituted by subunits k and e. These subunits possess a similar structure: an N-terminal c-helix with two domains, one transmembrane, and one soluble which extent (~40 Å for subunit e) into the intermembrane space to interact between them (23). Subunit g is holding the subunit e through a single transmembrane c-helix, probably via the conserved Gly-X-X-X-Gly motif of the two proteins (24). Simultaneously, the N- terminal c-helix of subunit g interact with the N-terminal ~50 residues of subunit b. The domain of the peripheral zone of interface monomer-monomer constituted by subunits e, g, and b with support from subunit k bend the mitochondrial inner membrane (23, 25). In Saccharomyces cerevisiae simultaneous deletion of dimerizing subunits g and e induces the loss of V2 and the change of mitochondrial cristae morphology into structures called onion rings. However, as S. cerevisiae is a facultative organism, mitochondrial participation in the bioenergetics of these mutants is not clear. As Ustilago maydis is a non-fermentative basidiomycete, which bioenergetics metabolism is hold by mitochondria, the study of the effect of deletion of dimerizing subunits in mitochondrial bioenergetics could be very illustrative. Deletion of subunit g in U. maydis doeunÓv changeu vhe cellwlat dowbling vime, glwcoue conuwmrvion and dry weight production, suggesting a similar carbon uptake and it incorporation into biomass. Hoyexet, wnvil noy, ye haxenÓv an ezrlanavion abowv yh{ vhe cellu of Fg strain were longest than WT, and why the total amount of mutant cells was few. Additionally, presence of classic mitochondria structure from mutant strain was conserved. Bioenergetics analysis showed that mitochondria from both strains used succinate to produce a similar F[m associated to oxygen uptake by the classic cytochrome chain (Table 1); however, Fg strain showed a decrease of ATP synthesis associated with expression, since early stage of growth, of the AOX. It has been reported that similar experimental conditions (i.e. high F[m, succinate as electron supply of respiratory chain, and a decrease of complex V activity) could produce reverse electron transport (RET) at complex I and lead to an increase in superoxide production (26). In this sense, AOX transfers electrons from QH2 directly to O2, bypassing complex IV, and thus act as a safety valve to prevent the excessive reduction of the Q pool (26-28). It is plausible to hypothesize that deletion of subunit g produce a decrease of the ATP synthesis by the dimer of F1FO-ATP synthase, which results in a low consumption of the F[m, an accumulation of NAD(P)H, and an increases of H2O2 production (i.e. through RET at complex I). In this sense, H2O2 production during succinate oxidation was decreased by AOX expression by Fg strain, similarly as shown previously (26, 29). Alternatively the role of deletion of subunit g in F1FO-ATP synthase could be analyzing by ATP synthesis. In this sense, the ATP production showed 20% reduction in the mutant strain. As subunit g is involved in the peripheral zone of monomer-monomer interface the conformation of dimer cowld be rteuetxed bwv vhe ATP u{nvhaue acvixiv{ donÓv. Additionally to the participation of the interface monomer-monomer in folding inner membrane, it has been proposed that it could play an important role in the activity of V2, highlighting the role of the dimerizing subunits (12). Indeed, the ATPase activity of V2WT was 7 times higher than V1WT and exhibits a high sensitivity to oligomycin (12). In this sense, modification of the interface monomer-monomer could provide evidences about the role of the dimerizing subunit in the activity of the V2. Unfortunately, the mutant V2Fg was so unstable and scarce to be isolated and characterized as WT dimer was; and only a few determination of its activity showed that it was 8-times lower than mutant V1Fg. As has been describe above, subunit g play an important role in the peripheral zone of the interface monomer-monomer, and its deletion lead only subunit e from one monomer interacting with subunit k from the other monomer; producing a weak interface. This leak interface monomer-monomer could evoke a low ATP synthesis, contrary to observed in V2WT, which show a high activity compared with V1WT (12). Acknowledgment This work was supported by Direccion General de Asuntos del Personal Academico (DGAPA) (IN222617) from Universidad Nacional Autonoma de Mexico (UNAM). ME-P is a PhD student of the Programa de Doctorado en Ciencias Biologicas (511021118) from UNAM and supported by CONACyT through a doctoral scholarship (254400). The authors thank to the Posgrado en Ciencias Biologicas from UNAM for the academic support. References 1) Pedersen PL. (1996) J. Bioenerg. Biomembr. 28, 389Î395. 2) VaillierJ, Arselin G, Graves PV, Camougrand N, Velours J. (1999) J. Biol. Chem. 274, 543Î 548 3) Arnold I, Pfieffer K, Neupert W, Stuart RA, Scha¨gger H. (1998) EMBO J. 17, 7170Î 7178 4) Rubinstein JL,Walker JE, Henderson R. (2003) EMBO J. 22, 6182Î92 5) LauWC, Baker LA, Rubinstein JL. (2008) J Mol Biol. 382,1256Î64 6) Minauro-Sanmiguel F, Wilkens S, García JJ. (2005) Proc Natl Acad Sci USA.102, 12356Î8 7) Dudkina NV, Heinemeyer J, KeegstraW, Boekema EJ, Braun HP. (2005) FEBS Lett. 579, 5769Î 72 8) Thomas D, Bron P,Weimann T, Dautant A, Giraud MF, Paumard P, et al. (2008) Biol Cell. 100, 591Î601 9) Dudkina NV, Sunderhaus S, Braun HP, Boekema EJ. (2006) FEBS Lett. 580, 3427Î32 10) Paumard P, Vaillier J, Coulary B, Schaeffer J, Soubannier V, Mueller DM, et al. (2002) EMBO J. 21, 221Î30 11) Soubannier V, Vaillier J, Paumard P, Coulary B, Schaeffer J, Velours J. (2002) J Biol Chem. 277, 10739Î45 12) Esparza-Perusquía M, Olvera-Sánchez S, Pardo JP, Mendoza-Hernández G, Martínez F, Flores- Herrera O. (2017) Biochim Biophys Acta Bioenerg. 1858, 975-981. 13) Brachmann A1, König J, Julius C, Feldbrügge M. (2004) Mol Genet Genomics. 272, 216-26. 14) Holliday R. Ustilago maydis. In: King R, editor. Handbook of Genetics. New York: Plenum; 1974. p. 575Î95 15) De los Rios Castillo D, Zarco-Zavala M, Olvera-Sanchez S, Pardo JP, Juarez O, Martinez F, Mendoza-Hernandez G, García-Trejo JJ, Flores-Herrera O. (2011) J Biol Chem. 286, 23911-9. 16) Flores-Herrera O, Olvera-Sánchez S, Esparza-Perusquía M, Pardo JP, Rendón JL, Mendoza- Hernández G, Martínez F. (2015) Biochim Biophys Acta. 1847, 143-152. 17) Waterfield WF & Sisler HD (1988) Biotechniques 6, 832-834 18) Wittig I, Karas M, Schägger H. (20070 Mol Cell Proteomics. 6, 1215-25. 19) Jung C, Higgins CM, Xu Z. (2000) Anal Biochem. 286, 214-223. 20) Schägger H, Cramer WA, von Jagow G. (1994) Anal Biochem. 217, 220-230. 21) Waterfield WF, Sisler HD. (1988) Biotechniques. 6, 832-834. 22) Reyes-Galindo M, Suarez R, Esparza-Perusquía M, de Lira-Sánchez J, Pardo JP, Martínez F, Flores-Herrera O. (2019) Biochim Biophys Acta Bioenerg. 1860, 618-627. 23) Guo H, Bueler SA, Rubinstein JL. (2017) Science. 358, 936-940. 24) Bustos DM, Velours J. (2005) J. Biol. Chem. 280, 29004Î29010 (2005) 25) Baker LA, Watt IN, Runswick MJ, Walker JE, Rubinstein JL. (2012) Proc. Natl. Acad. Sci. U.S.A. 109, 11675Î11680 26) Ellen L. Robb, Andrew R. Hall, Tracy A. Prime, Simon Eaton, Marten Szibor, Carlo Viscomi, Andrew M. James, and Michael P. Murphy. (2018) J. Biol. Chem. 293, 9869Î9879 27) El-Khoury R, Dufour E, Rak M, Ramanantsoa N, Grandchamp N, Csaba Z, Duvillie B, Benit P, Gallego J, Gressens P, Sarkis C, Jacobs HT, Rustin P. (2013) PLoS Genet. 9, e1003182 28) El-Khoury R, Kemppainen KK, Dufour E, Szibor M, Jacobs HT, Rustin P. (2014) Br. J. Pharmacol. 171, 2243Î2249 29) Szibor M, Dhandapani PK, Dufour E, Holmstrom KM, Zhuang Y, Salwig I, Wittig I, Heidler J, Gizatullina Z, Gainutdinov T, Fuchs H, Gailus-Durner V, de Angelis MH, Nandania J, Velagapudi V, (2017) Dis. Model. Mech. 10, 163Î171 Table 1. Structural and bioenergetics parameters of the wild type and Fg strains of Ustilago maydis. WT Fg Mitochondrial structure Cristae folding Yes Yes Bioenergetics Membrane potential generation Yes Yes, 100% of WT Mitochondrial oxygen consumption with succinate (ng at O/mg·min -1 ) cyanide sensitive 397 ‒ 31 441 ‒ 12 n-octyl gallate sensitive 112 ‒ 24 278 ‒ 18 Maximum mitochondrial oxygen consumption by complex IV or AOX (ng at O/mg·min -1 ) Total 1210 ‒ 35 (Fp›0) 1150 ‒ 20 (Fp›0) complex IV in presence of ascorbate and TMPD, cyanide sensitive 1048 ‒ 21 827 ‒ 18 AOX in presence of ascorbate and TMPD, n-octyl-gal sensitive 124 ‒ 23 262 ‒ 4 Mitochondrial NADH dehydrogenase activity (omol NADH oxidized/mg·min -1 ) Total 0.21 ‒ 0.03 (Fp›0) 0.20 ‒ 0.02 (Fp›0) Rotenone sensitive 0.18 ‒ 0.06 0.16 ‒ 0.03 Flavone sensitive 0.04 ‒ 0.015 0.03 ‒ 0.010 Succinate dehydrogenase activity (omol DCPIP reduced/mg·min -1 ) Total 0.05 ‒ 0.01 (Fp›0) 0.04 ‒ 0.01 (Fp›0) Mitochondrial F1FO-ATP synthase activity (omol ATP·(mg·min) -1 ) ATP synthesis (omol ATP·(mg·min)-1) 20 ‒ 0.9 15 ‒ 0.9 ATP hydrolysis (omol ATP·(mg·min)-1) 45 ‒ 6 31 ‒ 5 Table 2. Kinetics parameters of the ATPase activity of the monomer of the F1FO-ATP synthase from Ustilago maydis Fg strain. Monomer (V1Fg) a Non-activated F1FO-ATP synthase Vmax (omol ATP hydrolyzed/mg F1FO-ATP synthase·min-1) 0.58 ‒ 0.03 KM (oM) 155 ‒ 47 kcat (s -1) 5.86 ‒ 0.05 kcat/KM (M-1s-1) 3.78 X 104 ‒ 0.3 X 104 DDM activated F1FO-ATP synthase Vmax (omol ATP hydrolyzed/mg F1FO-ATP synthase·min-1) 2.3 ‒ 0.2 KM (oM) 920 ‒ 20 kcat (s -1) 23.26 ‒ 0.1 kcat/KM (M-1s-1) 2.53 X 104 ‒ 0.08 X 104 Ki (nM) for oligomycin 53 ‒ 4 Ratio oligomycin/F1FO-ATP synthase (mol/mol) to reach 50% of ATPase inhibition 5.03 ± 0.57 a = F1FO-ATP synthase mol in V1 sample was determined as described in procedures section, and kinetics parameters were showed as mg of F1FO-ATP synthase (See Material and methods section). Figure 1. Confirmation of replacement by homologous recombination of subunit g gene by hygromicine resistant gene. A) PCR was used to amplify sequences of DNA that includes the sequences of Inh1 or HygR and the flanking regions on each side of the Inh1 gene. B) Hygromicine resistant phenotype of the FInh1 strain. C) The products were analyzed on 1% agarose gel (see supplementary material for the primers used). The expected lengths of the product from the WT and FInh1 are shown at the right. D) WB analysis of the DNA sequence. Figures 2. Mitochondrial ultrastructure and bioenergetics analysis of the WT and Fg strains. The Fg strain display typical mitochondrial morphologies (D) similar to WT (A). Cells were harvested at 24 h and permeabilized with digitonine as described in material and methods section and the generation of F[m was assayed. The F[m was measurement with Safranine O. Arrows indicate sequential additions of permeabilized cells from WT (B) or mutant (E) strains. Where showed 10 mM succinate (Succ); ADP (1 mM); CCCP (5-10 oM). Oxygen uptake by WT cells (C) or Fg cells (F) was inhibited by cyanide (CN) and n-octyl-galate (nOg). Cells were harvested at 8 or 24 h. Where indicated 5 mM CN or 2 mM nOg was added. Numbers below each recording represent the velocity of oxygen consumption. Figure shows representative experiments of at least four different and independent cell culture preparations Figure 3. Digitonin Solubilization and in-gel activity of the respiratory complex and supercomplexes from Ustilago maydis mitochondria. Mitochondria from WT or Fg strains were isolated as described in the material and method section. Respiratory complex and supercomplexes from WT or Fg mitochondria were solubilized with digitonin and their electrophoretic profile determined in a BN-PAGE. Proteins were stained with brilliant blue Coomassie (A) and the in-gel activities of NADH:MTT oxidoreductase (B), succinate:MTT oxidoreductase (C), cytochrome c oxidase (D) and ATPase (E and F) were assayed. CI = complex I activity; CII = complex II activity; CIV = complex IV activity; CV = complex V activity. In-gel activity of the V2Fg and V1Fg was incubated from 24 h (F) Respiratory complexes and supercomplexes from bovine heart mitochondria were solubilized with digitonine and used as molecular weight standart. Figure 4. Isolation of V1 from mutant strain and its kinetics characterization. Mitochondrial supercomplexes were digitonin solubilized and isolated in a sucrose gradient as described in material and methods section. In-gel activity of the ATPase from V1 was observed from fraction 18 to 29 from sucrose gradient (A). No dimeric state was observed. These fractions were pooled and concentrated, and ATPase activity was observed at monomer position (B). Oligomicin inhibited ATPase in-gel activity (C). Coomassie stained showed a main protein band at the position of monomer and no dimer was observed (D). Kinetics characterization of the monomer with (open circles) or without (close circles) DDM addition (E). Kinetics ATPase inhibiton by oligomicin (F). Deletion of the natural inhibitory protein Inh1 from Ustilago maydis does not modify the oligomeric states of the F1FO-ATP synthase but increase their ATPase activity Romero-Aguilar Lucero a *, Esparza-Perusquía Mercedes a *, Langner Thorsten b , García-Cruz Giovanni a , Feldbrügge Michael c , Pardo Juan Pablo a , Martínez Federico a , and Flores-Herrera Oscar a1 . a Departamento de Bioquímica, Facultad de Medicina, Universidad Nacional Autónoma de México, Apartado Postal 70-159, Coyoacán 04510, México, D. F., México. b The Sainsbury Laboratory, Norwich Research Park, Norwich NR4 7UH, United Kingdom. c Institute for Microbiology, Cluster of Excellence on Plant Sciences, Department of Biology, Heinrich-Heine University Düsseldorf, Düsseldorf, Germany. To whom correspondence should be addressed: Oscar Flores-Herrera. Departamento de Bioquímica, Facultad de Medicina, Universidad Nacional Autónoma de México, Apdo. Postal 70- 159, Coyoacán 04510, México, Cd. Mx., México; Phone: 55-56232510; Fax: 55-56162419; E-mail: oflores@bq.unam.mx *L. Romero-Aguilar and M. Esparza-Perusquia contributed equally to this paper. Keywords: dimer of complex V, bioenergetics, ATPase activity, oxidative phosphorylation, Inh1 subunit. Abstract Transduction of electrochemical proton gradient to ATP synthesis is performed by F1FO-ATP synthase. The reverse reaction is avoided by regulatory subunit Inh1 in fungi. A gene knockout was generated in order to assign a function to Inh1 in the mitochondrial metabolism and cristae architecture. Deletion of Inh1 protein does not affect cell growth, glucose consumption, and biomass production. Ultrastructure and fluorescence analyzes showed that mitochondrial size and cristae shape, network, and distribution was similar to wild strain. The membrane potential, ATP synthesis, and oxygen consumption from wild type and mutant had a similar magnitude. Kinetics analysis of ATPase activity of complex V in permeabilized mitochondria showed similar values of Vmax and KM for both strains, and no effect of pH was observed. Isolated monomer and dimer from mutant mitochondria have a Vmax values 5-times higher than WT strain suggesting a Inh1 regulatory role; however no effect of pH was observed. ATPase activity of WT oligomers was stimulated several times by dodecyl-maltoside (DDM); however DDM induces an inactive form of the mutant oligomers, probably by dissociation of F1 sector suggesting a structural role for Inh1. The structural and kinetics role of the Inh1 subunit is discussed. 1 Introduction The energy stored in the electrochemical proton-motive force (p) across the membrane of almost all eubacteria, thylakoids or mitochondria is used by the F1FO-ATP synthase, an energy- transducing enzyme, to ATP synthesis from ADP and inorganic phosphate. F1FO-ATP synthase consist of two interconnected sectors, F1 and FO. FO is an integral membrane complex which acts as a proton-driven turbine, spinning the  subunit that drives sequential conformational changes in the three / subunits pairs (principally in the  subunits) from F1 sector. These sequential conformational changes are related to binding ADP + Pi (DP), to synthetize tightly bound ATP (TP), and to release the bond ATP (E) to the N-side of the membrane. Then, ATP production occurs in the membrane-extrinsic F1 domain, by a rotatory mechanism where the  chain within 33 and hydrophobic c ring domains, rotate in one direction with 3/120 o steps. The F1FO-ATP synthase could be considered as a reversible ATP-hydrolyzing protons pump whose direction depends on the thermodynamics balance between p and free energy for ATP synthesis (Gp). Under conditions of high p and for most energy-conserving membranes, the balance is in favor of ATP synthesis; however, damage to the electron transport chain, increase proton leakage, or severe hypoxia can lower p such that the F1FO-ATP synthase reverses and start to hydrolyze ATP over ATP synthesis. To avoid ATP hydrolysis by F1FO-ATP synthase, different strategies have been development: in thylakoids, in the dark, an intermolecular disulfide bond occurs in the  subunit near the interface with FO, and F1FO-ATP synthase relaxes back into an inactive state; in bacteria such as Escherichia coli and Bacillius PS3, the subunit  confer the unidirectional behavior of F1FO-ATP synthase (i.e. ATP synthesis) by acting like a ratchet. This subunit might take up two different conformations, in one of these the C-terminal domain extends toward the F1 sector and allows the enzyme only to synthetize ATP. In mitochondria from mammals (i.e. Bos taurus) and yeast (i.e. Saccharomyces cerevisiae) the ratchet subunit is the IF1 and Inh1, respectively. Interaction of the IF1 or Inh1 at a catalytic interface between  and  subunits and the  subunit in F1 is pH dependent, binding under conditions of lowered pH generally associated with a decrease of p. Actually, Hirst’s group has been determined that deletion of  subunit from Paracoccus denitrificans F1FO-ATP synthase does not activate ATP hydrolysis (1), suggesting that  subunit could adopt an inhibitory conformation, similar to observed in E. coli enzyme (2, 3), and blocks ATP hydrolysis. Additionally to the bioenergetics role of the F1FO-ATP synthase, it plays an important role, as a homodimer, in the mitochondrial cristae folding. In yeast subunit g and e maintain the dimer structure. Characterization of the ATPase activity of the F1FO-ATP synthase dimer showed that it was 7-times more active than monomer, was more sensitive to oligomicyn inhibition, and its specificity constant was 12-times higher than monomer (4). These features clearly show that dimer and monomer are kinetically different and point out the role of the monomer-monomer interphase. In mammals the IF1 was, controversially, implicated in dimerization of the F1FO-ATP synthase. Deletion of IF1 subunit from mouse F1FO-ATP synthase has not effect on the growth, in both male and female cases (5). There were no changes in the expression levels of subunits α and β of F1 in the brain, thymus, heart, lung, stomach, liver and testicle. In addition, changes were not found in the crista architecture and in the mitochondrial network (5). Similar results were observed in the zebra fish erythroblasts, Caenorhahbditis elegans, in the HeLa cells (6), and in S. cerevisiae. Although deletion of IF1 from mitochondria has been performed to assess its role in the dimerization of F1FO-ATP synthase, no studies on dimer ATPase activity have been done. In this work, the subunit Inh1 was deleted from the genome of Ustilago maydis, a strict respiratory basidiomycete, and F1FO-ATP synthase dimer and monomer were isolated and their activity characterized. We found that deletion of Inh1 does not activate ATPase activity of the dimer and monomer, but ATP synthesis was lowered in the intact mitochondria. The role of Inh1 on the ATP synthesis is discussed. 2 Materials and methods 2.1 Molecular biology techniques, strain generation and growth conditions E. coli top 10 (Life Technologies, Carlsbad, CA, USA) were used for cloning purpose. U. maydis strain FB2 (a2b2 genotype ATCC 201384) were described previously by (7). Cell transformation was performed using standard molecular techniques and strain generation methods. Plasmid pCRII-Topo (Invitrogen) was as a cloning vehicles. The plasmid pUMa1737 was constructed by replacing the 468-bp (ATPI gene, UMAG_02361) by hygromycin resistance cassette (HygR). Plasmid for generating deletion mutant Inh1_HygR (Inh1), resistance cassette is flanked 1161 pb upstream and 925 pb downstream region of Inh1 gene. Flanking regions were amplified by PRC using RL682/RL683 and RL686/RL687 and UM518 (FB2) wild-type DNA template; then plasmid linearized was inserted into the protoplasts for transformation. All homologous interaction events were verified by Southern blot analysis (7). The Inh1 R strain recovered to CM (0.25% Casaminoacids (Difco), 0.1% Yeast-Extract (Difco), 1.0% vitamin solution from Holliday '74, 6.25% salt solution from Holliday '74, 0.05% DNA degraded. Free Acid (Sigma, D-3159), 0.15% NH4NO3 (Sigma, A9642), pH 7.0 and 2% agar) supplemented whit 1% glucose. The cells were resuspended in 100 ml YPD (1% glucose, 1% yeast extract, 0.25% bactopeptona), growth for about 24 hr at 28°C, and harvested for make aliquots (1 ml cell in 25% glycerol) and stored at -70 o C stock. 2.2 Growth conditions of WT and Inh1 strains The WT (FB2) and Inh1 cells were sowed on 2% agar, supplemented with 1% glucose, 1% yeast extract, and 0.25% bactopeptone, and growth at 28 o C for 72 h. An inoculum was added to 100 ml of YPD (0.5% yeast extract, 0.25% bactopeptone, 0.5% glucose) and cultivated during 24 h at 28 o C and then, 30 absorbance units (600 nm) of this suspension were transfer to 1 liter of YPD medium, to continue the growth for 24 h at 28 o C at 200 rpm. 2.3 Oxygen consumption measurement Oxygen uptake was determined using a Clark-type electrode (YSI Model 5300). The respiration was started with the addition of 1 mg of cells (dry weigth) in 10 mM KH2PO4, pH 7.3 (final volume of 1.5 mL of air-saturated medium; 100% = 660 ng atoms O2); 1.7 mM potassium cyanide (KCN) was added for the inhibition of cytochromic pathway and 13 M of n-octyl-gallate (nOg) for alternative oxidase inhibition (8-10). 2.4 Mitochondrial membrane potential (m) measurement The cells were permeabilized with 0.02% digitonin in 300 mM sorbitol, 10 mM HEPES, 1 mM EGTA, 10 mM KH2PO4, 10 mM MgSO4, 150 mM KCl, pH 7.4, at 4 o C during 20 min with constant stirring. Generation of membrane potential (m) was determined using 30 M safranine- O and initiated with 10 mM succinate, pH 7.4; 5 M carbonyl cyanide m-chlorophenyl hydrazone (CCCP) was added to collapse m. The m was recorded in an AMINCO DW 2000 double beam spectrophotometer (Olis, Inc., Bogart, GA, EUA) at 533 and 511 nm. The final volume was 3.0 ml, at 25°C with constant stirring (11). 2.5 Microscopy Analysis Cell pellets were fixed with a mixture of 4% paraformaldehyde/2.5% glutaraldehyde for 1 hr at room temperature; then it was washed two times in PBS and stained with 1% osmium tetraoxide for 1 hr at 4°C, and washed with PBS. The gradual dehydration was made in ethanol (graded 50–100%) and propylene oxide. Pellets were embedded in epoxy resin and cut into 70 nm sections. For immunoelectron microscopy cell pellets were fixed with 4% paraformaldehyde/0.1% glutaraldehyde for 30 min at 4°C. After ethanol dehydration (graded 70–100%) pellets without osmication were embedded in LR White. Electron microscopy was performed at 80 kV on a Zeiss EM900 Transmission Electron Microscope. Images were recorded with a Gatan Dual Vision CCD 300W camera (Gatan, Pleasanton, CA). 2.6 Mitotracker staining Mitochondrial staining was done in cells grown in YPD for 24 h, harvested and washed twice with 0.9% NaCl. Then, they were suspended at a final optical density of 20 unit/L in a previously warmed (28ºC) media containing 5% of glucose and 30 nM of MitoTracker® Deep Red FM (Thermo Fisher Scientific). Cells were microscopically examined with a confocal microscope (Zeiss LSM5 Pascal, Carl Zeiss GmbH, Göttingen, Germany) with a water immersion 160x N.A. 1.3 objective. 2.7 Mitochondria isolation The U. maydis mitochondria isolation was performed as described by Júarez et al., (12) with a minor modifications (4). Cells were grown in complete medium (YPD) for 24 h, harvested by centrifugation at 3,800 g for 10 min and washed with distilled water (twice). The cells were resuspended with 12.5 ml of 0.6 M ammonium sulfate, 20 mM KH2PO4 pH 5.5, per g wet weight, supplemented with Trichoderma harzianum lytic enzyme (0.016 g/g wet weight) and incubated during 60 min at 30°C for protoplast production. The protoplasts were recovered by centrifugation at 3,800 g for 10 min, suspended in 0.8 M sucrose, 10 mM Tris, 2 mM EDTA, 20 mM KH2PO4, and 3% bovine albumin serum fatty acid free, pH 7.0 and centrifuged at 3,800 g for 10 min to wash out the lytic enzymes. The protoplast was resuspended in 0.4 M sucrose, 10 mM Tris, 2 mM EDTA, 20 mM KH2PO4, and 3% bovine albumin serum fatty acid free, pH 7.0 (buffer A) and homogenizer in a Teflon potter for 20 times with 1 mM PMSF and 25 l of the protease inhibitor cocktail (Sigma P8340). The samples were diluted to 130 ml per 16 g wet weight and centrifuged at 3,800 g for 10 min. Finally, the supernatant was centrifuged at 17,000 g for 10 min and mitochondrial fraction was suspended in buffer A. 2.8 Determination of protein concentration The protein content was determined as described by Lowry et. al., (13) with minor modifications (14). Crystalline bovine serum albumin (BSA) was used as standard and its concentration was determined at 278 nm using a molar extinction coefficient () of 6.58 (1%). 2.9 Solubilization of the respiratory complex and supercomplexes The mitochondrial respiratory complexes and supercomplexes were solubilized with different digitonin:protein (g/g) ratios as described (15-17). Briefly, mitochondrial protein (2 mg) were suspended in 200 L of 50 mM Bis-Tris and 500 mM 6-aminocaproic acid, pH 7.0, supplemented with 10 mM succinate and 10 mM ATP (4). The digitonin was added drop by drop until reach the ratio required and the mixture was incubated for 30 min at 4 o C with gentle stirring, and then centrifuged at 100,000 g for 30 min at 4 o C. The supercomplexes and complexes were recovered from supernatant and analyzed by blue native PAGE (BN-PAGE) and clear native PAGE high resolution (hrCN-PAGE) as described (15-17). 2.10 In-gel catalytic activity assay of respiratory complexes and supercomplexes. The in-gel activity assays were performed as described by Jung et al., (16) and Wittig et al., (15), using a gel strip loaded with 150 µg of protein. The NADH dehydrogenase activity of complex I (NADH:methylthiazolyldiphenyl tetrazolium bromide (MTT) oxidoreductase) was assayed in 10 mM Tris/HCl pH 7.4, 1.2 mM MTT, and 1 mM NADH, at 25 o C. The succinate dehydrogenase activity (Succinate:MTT oxidoreductase) was assayed in 10 mM K2HPO4, pH 7.4, 5 mM EDTA, 0.2 mM phenazine methosulfate (PMS), and 10 mM succinate. NADH and succinate activity was correlated with the development of purple precipitates on the gel. After purple-staining appear (10 – 20 min) the reaction was stopped with fixing solution (50% methanol, 10% acetic acid). The in-gel activity of complex IV was assayed in 50 mM K2HPO4, pH 7.2, 4.7 mM 3,3’diaminobenzidine tetrahydrochloride (DAB) and 16 M horse heart cytochrome c, during 30 – 40 min of incubation at 25 o C.The activity was observed as a brown precipitate and the reaction was stopped with the fixing solution. Activity of complex V was assayed in 50 mM glycine (adjusted to pH 8.0 with triethanolamine), 10 mM MgCl2, 0.15% Pb(ClO4)2 and 5 mM ATP. ATP hydrolysis correlated with the appearance of a white lead phosphate precipitates. The reaction was stopped using 50% methanol, and subsequently the gel was transferred to water and scanned against a dark background as described previously (4). 2.11 Isolation of submitochondrial particles (SMP). Mitochondria were diluted 1:1 (vol/vol) in buffer A (see mitochondria isolation section), and sonicated at full power (14 microns with ½ inch diameter tip) in a Soniprep 150 MSE (USA) during 30 sec with 90 sec of rest (5 cycles). Sonication was performed in an ice-cooled bath to hold temperature at 4 o C. After sonication the mixture was centrifuged at 17,300 g for 10 min, the supernatant recovered and ultracentrifuged at 100,000 g for 30 min at 4 o C. SMP were recovered from the pellet and resuspended in buffer A (See mitochondrial isolation section) and stored at - 17 o C until used (18). 2.12 Isolation of dimer and monomer F1FO-ATP synthase The monomer (V1) and dimer (V2) of F1FO-ATP synthase were isolated as reported by (4). The digitonin-solubilized mitochondrial complexes and supercomplexes (16 mg) were loaded on 24 mL of a continuous sucrose gradient (16 - 42%) and ultracentrifuged at 13,000 g during 16 h at 4 o C. Fractions (500 L) were collected and the presence of the F1FO-ATP synthase oligomers were identified by hrCN-PAGE (15). The fractions with the V1 or V2 were pooled separately, diluted in a ratio of 1:5 with 30 mm HEPES, pH 8.0 and 5% glycerol and then concentrated in 100K Millipore Amicon Ultra, the final samples (100 - 200 L) were stored at -70°C until use (4). The concentration of F1F0-ATP synthase in V1WT, V2WT, V1Inh1, and V2Inh1, samples was determined by a densitometry analysis of Commassie® Brilliant Blue R-125 stained α- (ID XP_011392137) and β- (ID XP_011389783) subunits from an SDS-Tricine-PAGE, using Coomassie stained BSA as a standard. The gel was scanned and the stain-intensity of - and - subunits, and BSA was determined by the Image Analysis software version 1.0 (Thermo Fisher Scientific Inc.), and their intensities were measured by peak integration after densitometry analyses (4). The mol of - and -subunits were determined using the molecular weight of the mature protein (55.05 and 50.47 kDa, respectively). The amount of F1FO-ATP synthase in V1WT and V2WT samples was 3.3 ± 0.7 g/10 g total protein and 2.8 ± 0.5 g/10 g total protein, respectively; while in V1Inh1, and V2Inh1 samples was 3.3 ± 0.7 g/10 g total protein and 2.8 ± 0.5 g/10 g total protein. These amounts of complex V were used to kinetics parameters estimation. 2.13 Measurements of ATP hydrolysis by F1FO-ATP synthase by V1 or V2 The ATPase activity was determined spectrophotometrically as described by (4). Briefly, ATP hydrolysis was measured using an assay coupled to NADH oxidation (ε340nm = 6.2 mM −1 ·cm −1 ) in an Agilent 8453 UV–visible spectrophotometer (Agilent Technologies, USA). NADH absorbance was continuously monitored and the time response was less than 1 s. The reaction mixture was 1 mM MgSO4, 90 mM KCl, 50 units/mL of pyruvate kinase (PK), 30 units/mL of lactate dehydrogenase (LDH), and 5mM of phosphoenol pyruvate (PEP). The pH buffer was HEPES (50 mM) for pH = 8.0 or MES (50 mM) for pH = 6.0. The pH change doesn’t modified perceptibly the activity of the coupling enzymes. The reaction was started by F1FO-ATP synthase (10-15 g) addition and its kinetic analysis of ATPase activity (initial velocity) was carried out using the direct spectrophotometric recording. Initial velocities were further obtained from the slope of the linear region in each spectrophotometric recording, and the linear region of the traces was corroborated with the plot of the first derivative against time. Data were analyzed by robust, weighted, non-linear regression analysis using the SigmaPlot software (Systat Software, Inc., version 10.0). The data represent the average of five independent experiments. The data were corrected to F1-FO-ATP synthase content in each V1 or V2 samples as described by (4). 2.14 Activity assay of mitochondrial NADH and succinate dehydrogenase and ATPase Isolated mitochondria were permeabilized with 0.01% Triton X-100 and the reaction mixture was 30 mM KH2PO4, 5 mM MgCl2, 1 mM EGTA, 120 mM KCl, pH 7.4, at 25 o C. Spectrophotometric recorder of the NADH:2,6-dimethoxy-1,4-benzoquinone (DBQ) oxidoreductase activity was determined at 340 nm by following the oxidation of NADH. NADH addition was 100 M plus 600 M of DBQ, and NADH dehydrogenase activity was inhibited by 10 M rotenone (19-21) and 50 M flavone. Activity of succinate:DCPIP oxidoreductase was stimulated with 10 mM succinate and determined at 600 nm by following the reduction of the artificial electron acceptor 2,6-dichlorophenol-indophenol (DCPIP; 50 M; εDCPIP = 21 mM -1 cm -1 ), and was inhibited by 20 mM malonate. ATPase activity in permeabilized mitochondria was performed in 200 mM KCl, 50 mM HEPES pH 8.0 (or MES pH 6.0), 3 mM MgSO4, and using 500 g mitochondrial protein/mL. The reaction was initiated by adding different concentrations of Mg-ATP. Aliquots of 0.15 mg of mitochondria were withdrawn at intervals of 30 seconds, placed in 6% of previously cooled trichloroacetic acid. Samples were centrifuged at 10,000 rpm at 4ºC, in a tabletop centrifuge. Supernatant was analyzed according to the Fiske & Subbarow, (1925) (22) method with minor modifications (23). 2.15 Structure prediction of Inh1 protein The Inh1 sequence was obtained from the database of the U. maydis genome using the sequence of the bovine IF1. Only one coding sequence was found for the Inh1 protein, UMAG- 02361 (XP_011388667). This sequence was analyzed with MITOPROT (https://ihg.gsf.de/ihg/mitoprot.html) (24) to eliminate the signal peptide. In order to compare the amino acids sequence of the Inh1 protein against Saccharomyces cerevisiae and Bos taurus subunits, the sequences of the inhibitor proteins of these organisms were obtained from the protein database UniProtKB (https://www.uniprot.org/help/uniprotkb). The mature protein from U. maydis consists of 64 amino acids with a molecular weight of 7.3 kDa, while the S. cerevisiae protein consists of 36 amino acids (MW of 4.0 kDa), and the B. taurus of 65 amino acids (MW of 7.7 kDa). These proteins were aligned with Clustal W (25) and the score of the predicted alignment with T- coffee (http://tcoffee.crg.cat/) (26) was 65%. The protein 3D structure was predicted using Modeller 9.2 (https://salilab.org/modeller/) (27). 2.16 Statistic analysis The statistical analysis of data was performed in the GraphPad Prism V.6b (Trial). 3 Results 3.1 Inh1 of Ustilago maydis has an alpha-helix type structure The S. cerevisiae, B. taurus and U. maydis sequence conserve: two lysine residues, two glutamic residues, one arginine-rich and one alanine-rich, as well as five residues of amino acids with similar properties (> 0.5 score) and one residue with similar low properties (< 0.5 score) (Fig. 1). Interestingly, U. maydis subunit is the only one which contain tryptophan (1.6%). The 3D structure prediction shows that Inh1 from U. maydis has an alpha-helix similar to predicted for S. cerevisiae and B. taurus (Fig. 1). There is a disordered structure that stands out in the NH2-terminus which interacts with the F1 portion of the ATP synthase (Fig. 1). The PYMOL software alignment showed that the alpha helix of IF1 is larger than Inh1 from U. maydis and S. cerevisiae; while the S. cerevisiae is the smallest one. Modeling of Inh1 from U. maydis using the crystalized structure of S. cerevisiae as template shows that both have a disordered NH2-terminus which could interacts with the F1 sector (Fig. 1), particularly at the / interphase. The 3D structure prediction for Inh1 subunit from U. maydis allows us to hypothesize that its role as regulatory ATPase activity of the F1FO-ATP synthase could be similar to reported in S. cerevisiae and bovine (i.e. avoid ATP hydrolysis if p0). Then, deletion of Inh1 from U. maydis genome could promote an increase in the ATPase activity of F1FO- ATP synthase. 3.2 Deletion of the Inh1 subunit from the Ustilago maydis genome The Inh1 gene was deleted by homologous recombination as reported by Brachmann et. al., (7) using a hygromycin resistant gene (Hyg). The mutant strain of U. maydis with the Inh1 gene deleted (Inh1) was confirmed by PCR analyses using different primers (see supplemental material). The first pair of primer sequences is internal to the Inh1 gene and to the flanking regions on each side of the Inh1 gene, so the WT strain gave a products of length of 1431 and 1227 bp while the Inh1 strain gave no product (Fig. 2A). The other pair of primers have an internal sequence to the Hyg gene and to the flanking regions of the Inh1 gene, so the Inh1 strain gave two products of 2270 and 1988 bp (Fig. 2A) indicating the gene substitution and its correctly incorporation. Finally, the mutant strain was hygromicine resistant while WT doesn’t (Fig. 2B). These results showed that Inh1 subunit was deleted from U. maydis genome and is not present in the Inh1 strain. 3.3 Inh1 deletion doesn’t modified strain growth In order to determine the possible participation of Inh1 in the U. maidys (FB2:a2b2) growth, the ∆Inh1 and the WT strains were cultivated in YPD. The cell duplication time was of 2.4  0.1 and 2.7  0.1 hours for WT and ∆Inh1 strains respectively. Glucose consumption was of 1.62  0.1 g·h -1 for WT and 1.44  0.1 g·h -1 for ∆Inh, and no statistically significant difference was found using a non-parametric t-student. The biomass produced at 24 h of cultivating was 1.9 ± 0.2 g·L -1 and 1.4 ± 0.04 g·L -1 for WT and Inh1, respectively. Biomass generation matches the glucose consumption. WT strain showed a cell amount of 4.1 X 10 9 cells·L -1 while ∆Inh1 strain had 3.4 X 10 9 cells·L -1 at the end of culture. The cellular length was of 23  4 m for WT while for Inh1 was 28  3 m at stationary phase of growth. These observations indicate that deletion of Inh1 doesn’t modify the growth, time of duplication, and biomass generation by U. maydis mutant. 3.4 The ∆Inh1 strain preserves the cristae and the mitochondrial network ATP-synthase forms dimers in the inner mitochondrial membrane of yeast and mammals folding the inner membrane in cristae; it has been suggested that IF1 could take part in a dimerization process; however, the role of Inh1 is not conclusive (28-30). To determine whether Inh1 deletion modified mitochondrial cristae, a transmission electron microscopy of the WT and ∆Inh1 strain cells was conducted. Large mitochondria with similar morphology were observed in both strains and no alterations were found in the mitochondrial cristae density in ∆Inh1 strain (Fig. 3A and E); indeed, both mitochondria contain lamellar (and presumably tubular) cristae in an orthodox configuration. The mitochondrial network analysis showed that a tubular mitochondrial network was found across the mutant cell similarly as in WT strain (Fig. 3B and F). Both observations indicated that mitochondrial network and cristae architecture were intact in the Inh1 strain. 3.5 Bioenergetics in WT and Inh1 strains To verify mitochondrial functions, some bioenergetics parameters from WT and Inh1 strain were determined. For instance, cellular respiration recording showed a similar value for both strains: 348 ng atom O·(mg·min) -1 for WT (Fig. 3C) and 322 ng atom O·(mg·min) -1 for Inh1 (Fig. 3G). Cyanide addition reduced respiration to 288 and 225 ng atom O·(mg·min) -1 for WT and Inh1, respectively. Total inhibition of respiration was observed with cyanide plus n-octyl-gallate addition; suggesting that alternative oxidase (AOX) was expressed as well as classical electron transport chain complexes. Mitochondrial membrane potential (m) from WT and Inh1 strains was determined in digitonin-permeabilized cells in order to allow Safranine O and substrates reach mitochondria. Succinate addition induces m generation in both strains with similar magnitude (Fig. 3D and H), and ADP produces a transitory membrane depolarization (related to ATP synthesis, vide infra). CCCP addition increased the proton conductance (m0). These observations showed that mitochondria from both strains are functional. Additionally, the activity of NADH dehydrogenase was 0.22  0.03 and 0.28  0.04 mol NADH oxidized·(mg·min -1 ) for the WT and Inh1 strains, respectively. Triton X-100 or CCCP was used to p0 and the NADH activity was increased to 0.38  0.04 and 0.36  0.02 mol NADH oxidized·(mg·min -1 ) (Table 1). This suggested that complex I and alternative NADH dehydrogenases (i.e. Ndei, Nde2, and Ndi1) were presents in both strains. The succinate dehydrogenase showed an activity of 0.035  0.01 and 0.049  0.01 mol DCPIP reduced·(mg·min -1 ) for WT and mutant strains, respectively. Dissipation of the p with Triton X-100 or CCCP reduce the succinate activity to 0.017  0.007 and 0.017  0.006 mol DCPIP reduced·(mg·min -1 ) for the WT and Inh1 strains, respectively (Table 1). ATP synthesis promoted by the p is the central role of F1FO-ATP synthase, and elimination of the Inh1 subunit could affect its correct functioning. ATP synthesis was determined using digitonin-permeabilized cells and was stimulated by ADP addition (Fig. 3D and H). The WT and Inh1 showed an ATP synthesis very similar (i.e. 18  0.9 µmol·(g of dry weight·min) -1 , and 20  0.9 µmol·(g of dry weight·min) -1 , respectively); suggesting that p promotes the ATP synthesis in the absence of the Inh1 subunit (Table 1). This results indicate that deletion of Inh1 subunit don’t affect the principal mitochondrial bioenergetics parameters. 3.6 ATPase activity by mitochondrial F1FO-ATP synthase from the WT and ∆Inh1 strains The Inh1 is the regulatory subunit of ATPase activity of the complex V under unfavorable conditions; generally, a pH increase at the mitochondrial matrix triggers hydrolase activity. The ATPase activity was determined in both strains (WT and ∆Inh1) at pH values of 8.0 and 6.0 (Fig. 4) using Triton-X100 to permeabilize mitochondria in order to dissipate the p and allow that ATP reach matrix. The activity of the complex V in the mitochondrial membranes was similar for both strains assayed at pH of 8.0 (Fig. 4A), with a Vmax values of 130 ± 3 nmoles/min·mg mitochondrial protein, and KM = 1.5 ± 0.1 mM for the ∆Inh1 strain, while the value of Vmax for the WT strain was 125 ± 4 nmoles/min·mg mitochondrial protein and KM of 1.3 ± 0.1 mM (Table 2). This suggested that at pH of 8 the Inh1 subunit could be removed from complex V from WT strain and ATPase activity of both strain reach similar values of Vmax. However, at pH = 6.0 (Fig. 4B) the ATP hydrolysis by complex V from WT strain showed no change in magnitude (Vmax = 140 ± 3 nmoles/min·mg mitochondrial protein and KM = 1.4 ± 0.1 mM); moreover, the ∆Inh1 strain showed a slight increase (i.e. the Vmax = 190 ± 7 nmoles/min·mg mitochondria and KM = 1.5 ± 0.2 mM) (Table 2). There was no statistically significant difference using a non-parametric t-student, suggesting that deletion of Inh1 subunit has a worthless stimulatory effect on ATPase activity. These results suggested that pH shift could not be involved with the displacement of the regulatory subunit of F1FO-ATP synthase. 3.7 The ∆Inh1 assemble an active F1FO-ATP synthase dimer Supercomplexes and complexes were efficiently solubilized using a digitonin:protein ratio from 2 to 5, and their activities analyzed by BN-PAGE (Fig. 5). The staining of protein with the Coomassie brilliant blue shows that the complexes and supercomplexes profile was similar in both strains (Fig. 5A). The ATPase activity of the complex V was associated to two main protein bands with a molecular weight of 1200 kDa, which correspond to the V2, and 660 kDa which is the V1 (Fig. 5B), similar to previous report (4). Occasionally, a third ATPase activity band appear with a molecular weight of 1100 kDa, which contained the 17 subunits of the dimeric state of complex V and no kinetics differences has been observed against the dimer of 1200 kDa (4). The NADH dehydrogenase activity (Fig. 5C) was associated to a protein band of 1000 kDa (individual complex I), and a protein set of different molecular weight (i.e. from 1200 to 1700 kDa) which correspond to the supercomplexes (i.e. I:III2:IV, vide infra). Interestingly, the presence and activity of the individual complex I was enriched in the Inh1 strain, while supercomplexes were diminished, suggesting that supercomplexes could be rearranged or could be labile to digitonin solubilization. Complex II showed a molecular weight of 130 kDa and no supercomplexes were associated with this activity (Fig. 5D). Finally, the activity of complex IV was distributed into a protein band of 240 kDa and the supercomplexes of higher molecular weight (Fig. 5 E) similar to previous report from our group (31). To get inside in the effect of Inh1 deletion on the ATPase activity of the F1FO-ATP synthase, its dimeric (V2) and monomeric (V1) state from WT (V2WT, V1WT) and Inh1 (V2Inh1, V1Inh1) mitochondria were digitonin-solubilized with a ratio 2:1 and isolated as reported previously by our group (4). 3.8 Kinetics characterization of the V1 and V2 at different pH values The monomeric and dimeric states of the F1FO-ATP synthase from the WT and Inh1 strains were efficiently isolated (Fig. 6) and their activity was assayed at pH 8.0 and 6.0 using an ATP-regenerating system (4). The activity of the V1WT assayed at pH = 8.0 (Fig. 6B, open circles) showed an Michaelis- Menten kinetics with a Vmax of 0.93 ± 0.02 mol of ATP hydrolyzed·(mg of complex V·min) -1 and a KM of 1.4 ± 0.1 mM. The V2WT showed a very low activity and affinity (Fig. 6E, open circles), with a Vmax and KM values of 0.5 ± 0.01 mol of ATP hydrolyzed·(mg of complex V·min) -1 and 0.35 ± 0.05 mM, respectively. It has been reported that DDM (0.05%) could stimulate the activity of the V1WT and the V2WT (4). As showed in figure 6E (close circles) the V2WT increase several times its ATPase activity while V1WT have a 1.6-time increase (Fig. 6B, close circles, and Table 3). Interestingly, when pH was shifted to 6.0, the V1WT shows a 3.6-time increase in the Vmax, with or without DDM addition respect to control conditions (Table 2). For the V2WT there no changes in the kinetics parameters neither in the DDM ATPase activity stimulation were observed (Fig. 6C and F, and Table 3) suggesting that Inh1 role in the ATPase activity regulation couldn’t be associated to proton concentration. To verify this hypothesis, the kinetics characterization of V1Inh1 and V2Inh1 was performed. The ATPase activity of the V1Inh1 assayed at pH = 8.0 (Fig. 6H, open circles) showed a Michaelis-Menten kinetics and was 4.4-times higher than V1WT, showing a Vmax values of 4.1 ± 0.1 moles of ATP hidrolized·(mg of F1FO-ATP synthase·min) -1 and a KM = 1.8 ± 0.2 mM (Table 2). Contrary to observed with the V2WT, the V2Inh1 showed a Vmax = 3.4 ± 0.16 moles·(mg of F1FO- ATP synthase·min) -1 and a KM = 1.4 ± 0.23 mM (Table 2), a similar ATPase activity to V1Inh1 (Fig. 6K, open circles). These results clearly showed that deletion of Inh1 increased the ATPase activity of complex V oligomers but the pH effect was not observed. Interestingly, the ATPase activation of the V1WT and V2WT by DDM addition doesn’t occurs with the V1Inh1 and V2Inh1; indeed, DDM induces the inhibition of the ATPase activity of the V1Inh1 and V2Inh1 (Fig. 6H and K, close circles, respectively). The shift of pH to 6.0 doesn’t modify the activity of the V1Inh1 and V2Inh1 (Table 2), neither DDM inhibition (Fig. 6I and L). These results indicate that pH shift (from 8 to 6) doesn’t modify the interaction between the Inh1 subunit and the F1 sector in the WT strain; but deletion of Inh1 increases the ATPase activity independently of the pH shift. The effect of DDM suggests that Inh1 could play an alternative role as a stabilizer of the F1FO-ATP synthase, since its deletion induces the inactivation of the ATPase activity by DDM. 4. Discussion In healthy respiratory mitochondria the complexes from respiratory chain generate the p which drives the F1FO-ATP synthase to synthetizes ATP. Under unfavorable conditions where the oxygen or substrate availability is limited such as isquemia, Parkinson´s and motor neuron diseases the F1FO-ATP synthase hydrolyze the ATP (30). Indeed, the ATPase activity of F1FO-ATP synthase increases as well as matrix pH decrease, which is related with the oxygen limitation (32). In these conditions, the inhibitory protein (IF1 in mammals and Inh1 in fungi) is the responsible to inhibit the ATPase activity; this binds to the soluble F1 sector, interfering with rotation of the central stalk and the conformational changes in the catalytic / interfaces (33). If p increases, the inhibitor protein is expelled and ATP synthesis resumes (34). As the Inh1 subunit of U. maydis adopts a similar conformation to the IF1 inhibitor protein of mammalian and Inh1 of S. cerevisiae (Fig. 1), a similar regulatory mechanism could be proposed for it during unfavorable conditions. In this research, deletion of the Inh1 subunit in the eukaryotic model U. maydis has no observable effect on the growth of cells, biomass production, glucose consumption, neither on the cristae structure and mitochondrial network (see Fig. 3), nor on the mitochondrial oxygen consumption, membrane potential generation, and complexes I, II, and IV activities (see Table 1). Additionally, the functionality of the F1FO-ATP synthase was corroborated as its ability to synthesize ATP in both WT and Inh1 U. maydis cells (Table 1). Additionally, activity of the complexes I, II, and IV was similar (Table 1), however their activity and distribution into individual complexes or supercomplexes was slightly different between WT and Inh1 strains. The amount of individual complex I in the Inh1 was higher than WT, and its activity associated with supercomplexes was lower (Fig. 5). Similarly, single complex IV activity was higher in the WT than Inh1 and its presence into supercomplexes was greater in WT strain. This suggest that supercomplexes with the composition I:III2:IV (i.e. the respirasoma as reported by (31) is most stable in the WT strain than in the Inh1. Finally, complex II was observed as a single state in both strains; however, its activity was higher in the WT strain. However, new studies must be performed to determine the effect of Inh1 deletion into the composition and amount of the supercomplexes. The role of the regulatory subunit, Inh1, is to inhibits the ATPase activity of the F1FO-ATP synthase when p0 and the mitochondrial matrix pH decrease; in this sense, an acidic condition (i.e. pH = 6.0) does not modify ATP hydrolysis in permeabilized mitochondria from WT or Inh1 strains (see Table 2 and Fig. 4), suggesting that deletion of Inh1 subunit was not sufficient to activate ATP hydrolysis (i.e. pH = 8.0). As the F1FO-ATP synthase in the inner membrane mitochondria occurs principally as a monomer (V1) and a dimer (V2), a further analysis of ATPase activity at different pH-values was performed using the isolated oligomers (see Table 2 and Fig. 6). The ATPase activity of the V1 from WT was higher than the dimeric state assayed at pH 8.0. Similarly to previous report, both states were stimulated by DDM (0.005%); in this conditions, V2 showed an increasing of Vmax by 9-times than V1 (Fig. 4). Similar effect was observed at pH = 6.0. The oligomers from Inh1 strain showed a slightly higher activity than WT (Fig. 4) at both pHs assayed; however, DDM addition decreased their activity, contrary to observed for the WT (Fig. 4). This result suggests that Inh1 could play a main stabilizing role rather than as a regulatory subunit of F1FO-ATP synthase activity. Interestingly, activity of isolated oligomers from mutant mitochondria showed Vmax 3-times higher than WT strain, suggesting that deletion of Inh1 could Acknowledgment This work was supported by Direccion General de Asuntos del Personal Academico (DGAPA) (IN222617) from Universidad Nacional Autonoma de Mexico (UNAM). ME-P is a PhD student of the Programa de Doctorado en Ciencias Biologicas (511021118) from UNAM and supported by CONACyT through a doctoral scholarship (254400). The authors thank to the Posgrado en Ciencias Biologicas from UNAM for the academic support and to DGAPA-UNAM for the fellowship provided to Lucero Romero-Aguilar. Author contributions OFH: He conceptualized and coordinated the research, suggested experiments, analyzed data, wrote and edited the manuscript and provided funding to support the research. MF: He conceptualized and coordinated the mutant construction, revised and edited the manuscript. LT: He produced the mutant strain and revised the manuscript. ME: She produced the mutant strain, suggested experiments, analyzed the data, performed experiments wrote the original draft and reviewed and edited the manuscript. LRA: She performed experiments and analyzed the data. JPP: He suggested experiments, analyzed the data, revised and edited the manuscript. References 1) Varghese F, Blaza JN, Jones AJY, Jarman OD, Hirst J. (2018) Open Biol. 8: 170206 2) Cingolani G, Duncan TM. (2011) Nat. Struct. Mol. Biol. 18, 701–707 3) Sobti M, Smits C, Wong ASW, Ishmukhametov R, Stock D, Sandin S, Stewart AG. (2016). eLife. 5, e21598 4) Esparza-Perusquia M, Olvera-Sanchez S, Pardo JP, Mendoza-Hernandez G, Martinez F, Flores- Herrera O. (2017) Biochimica et biophysica acta Bioenergetics. 1858, 975-981 5) Nakamura J, Fujikawa M, Yoshida M. (2013) Bioscience reports. 33 6) Sanchez-Cenizo L, Formentini L, Aldea M, Ortega AD, Garcia-Huerta P, Sanchez-Arago M, Cuezva JM. (2010) J. Biol Chem. 285, 25308-13 7) Brachmann A, König J, Julius C, Feldbrügge M. (2004) Molecular Genetics and Genomic, 272:261-226 8) Estambrook RW. (1967) Methods and Enzymology, 10:41-47 9) Guerrero-Castillo S, Cabrera-Orefice A, Vázquez-Acevedo M, González-Halphen D, Uribe- Carvajal S. (2012) Biochimica et Biophysica Acta, 1817:353–362 10) Affourtit C, Quinlan CL, Brand MD. (2012) Methods in molecular biology, 11:165-182 11) Flores-Herrera O, Olvera-Sánchez S, Esparza-Perusquía M, Pardo JP, Rendón JL, Mendoza- Hernández G, Martínez F. (2015) Biochim Biophys Acta. 1847, 143-152. 12) Juárez O, Guerra G, Velázquez I, Flores-Herrera O, Rivera-Pérez RE, Pardo JP. (2006) FEBS Journal, 273:4603-4615 13) Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. (1951) J Biol Chem. 193:265-275 14) Mahuran D, Clements P, Carrella M, Strasberg PM. (1983) Analytical biochemistry, 129: 513- 516 15) Wittig I, Karas M, Schägger H. (20070 Mol Cell Proteomics. 6, 1215-25. 16) Jung C, Higgins CM, Xu Z. (2000) Anal Biochem. 286, 214-223. 17) Schägger H, Cramer WA, von Jagow G. (1994) Anal Biochem. 217, 220-230. 18) Harmon HJ. (1982) Journal of Bioenergetics and Biomembranes. 14, 377-386 19) Nakashima Y, Shinzawa-Itoh K, Watanabe K, Naoki K, Hano N, Yoshikawa S. (2002) J. Bioenerg. Biomembr. 34:11-9 20) Vinogradov AD, Grivennikova VG. (2005) Biochemistry (Mosc). 70:120-127 21) Dröse S, Galkin A, Brandt U. (2005) Biochim. Biophys. Acta. 1710:87-95 22) Fiske CH, Subbarow Y. (1925) Journal of Biological Chemistry. 66, 375-400 23) Rule CS, Patrick M, Sandkvist M. (2016) Journal of visualized experiments : JoVE 24) Claros MG, Vincens P. (1996) European journal of biochemistry. 241, 779-86 25) Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, Valentin F, Wallace IM, Wilm A, Lopez R, Thompson JD, Gibson TJ, Higgins DG. (2007) Bioinformatics. 23, 2947-8 26) Di Tommaso P, Moretti S, Xenarios I, Orobitg M, Montanyola A, Chang JM, Taly JF, Notredame C. (2011) Nucleic acids research. 39, W13-7 27) Webb B, Sali A. (2014) Current protocols in bioinformatics. 47, 5 6 1-32 28) Campanella M, Casswell E, Chong S, Farah Z, Wieckowski MR, Abramov AY, Tinker A, Duchen MR. (2008) Cell metabolism. 8, 13-25 29) Garcia JJ, Morales-Rios E, Cortes-Hernandez P, Rodriguez-Zavala JS. (2006) Biochemistry. 45, 12695-703 30) Campanella M, Parker N, Tan CH, Hall AM, Duchen MR. (2009) Trends in biochemical sciences. 34, 343-50 31) Reyes-Galindo M, Suarez R, Esparza-Perusquía M, de Lira-Sánchez J, Pardo JP, Martínez F, Flores-Herrera O. (2019) Biochim Biophys Acta Bioenerg. 1860, 618-627. 32) Khodjaev E, Komarnitsky FB, Capozza G, Dukhovich VF, Chernyak BV, Papa S. (1990) FEBS letters. 272, 145-8 33) Garcia JJ, Morales-Rios E, Cortes-Hernandez P, Rodriguez-Zavala JS. (2006) Biochemistry. 45, 12695-703 34) Scwerzmann K, Pedersen PL. 1981. Biochemistry 20, 6305-6311. Table 1. Structural and bioenergetics parameters of the wild type and Inh1strains of Ustilago maydis. WT Inh1 Presence of mitochondrial network Yes Yes Mitochondrial cristae folding Yes Yes Membrane potential generation Yes Yes, 100% of WT Oxygen consumption (ng at O/mg·min -1 ) 348  32 322  35 Oxygen consumption cyanide resistant (ng at O/mg·min -1 ) 288  30 225  22 NADH dehydrogenase activity (mol NADH oxidized/mg·min -1 ) 0.22  0.03 0.38  0.04 (p0) 0.28  0.04 0.36  0.02 (p0) Succinate dehydrogenase activity (mol DCPIP reduced/mg·min -1 ) 0.035  0.009 0.017  0.007 (p0) 0.049  0.01 0.017  0.006 (p0) ATP synthesis (mol ATP·(mg·min) -1 18  0.9 20  0.9 Table 2. Kinetics parameters of ATPase activity of the F1FO-ATP synthase from WT and Inh1 strains from Ustilago maydis mitochondria. WT Inh1 pH = 8.0 pH = 6.0 pH = 8.0 pH = 6.0 Permeabilized mitochondria Vmax (nmol ATP hydrolyzed/mg protein·min - 1 ) 125  4 136  3 128  3 189  7 KM (mM) 1.3  0.1 1.4  0.1 1.5  0.1 1.5  0.2 Non-activated monomer (V1) ATPase activity Vmax (mol ATP hydrolyzed/mg CV·min -1 ) 0.93  0.02 3.4  0.1 4.1  0.11 3.6  0.07 KM (mM) 1.4  0.1 0.6  0.1 1.8  0.2 1.5  0.1 DDM-activated V1 ATPase activity Vmax (mol ATP hydrolyzed/mg CV·min -1 ) 1.5  0.02 5.4  0.1 0.12  0.01 0.038  0.003 KM (mM) 0.8  0.06 0.4  0.05 0.5  0.6 2.2  0.5 Non-activated dimer (V2) ATPase activity Vmax (mol ATP hydrolyzed/mg CV·min -1 ) 0.5  0.01 0.7  0.03 3.4  0.16 3.3  0.1 KM (mM) 0.35  0.05 0.31  0.09 1.4  0.23 0.64  0.1 DDM-activated V2 ATPase activity Vmax (mol ATP hydrolyzed/mg CV·min -1 ) 11  0.3 12.7  0.3 0.0098  0.003 0.02  0.001 KM (mM) 0.9  0.1 0.35  0.05 0.79  0.1 0.73  0.1 Figure 1. Multiple alignment and molecular modeling of Inh1 from Ustilago maydis. A) Clustal W alignment of the amino acid sequence of the inhibitor proteins of U. maydis, S. cerevisiae and B. taurus. Identical (*) and similar (:) amino acid preserved across all sequences. B) 3D model of the F1 domain of ATP synthase-bovine with inhibitor protein (IF1); in white is showed the Inh1 which is embedded between subunits α and β, at the NH2-terminal. C and D) 3D model obtained with Modeller 9.2, using the S. cerevisiae protein as a template (PDB). The structural alignment was made with PYMOL. The green color shows the S. cerevisiae Inh1 and the blue color shows the forecast for U. maydis. Figure 2. Confirmation of replacement by homologous recombination of Inh1 gene by hygromicine resistant gene. A) PCR was used to amplify sequences of DNA that includes the sequences of Inh1 or HygR and the flanking regions on each side of the Inh1 gene. The products were analyzed on 1% agarose gel (see supplementary material for the primers used). The expected lengths of the product from the WT and Inh1 are shown at the right. B) Hygromicine resistant phenotype of the Inh1 strain. Figure 3. Structural and bioenergetics analysis of the mitochondria from WT and Inh1 strains. WT is showed at the upper panel and Inh1 at the bottom panel. Transmission electron microscopy of the WT (A) and Inh1 (E) shown round shape mitochondria containing lamellar cristae. MitoTracker® Deep Red FM stain and light field microscopy of the mitochondrial network of the WT (B) and Inh1 (F). Oxygen uptake by the WT (C) and Inh1 (G) cells. Where is indicated cells, cyanide (to inhibit the complex IV) or n-octyl-gallate (to inhibit the alternative oxidase) were added. Mitochondrial membrane potential (m) of the WT (D) and Inh1 (H) cells was monitoring in presence of safranine-O. Where is indicated cells, succinate (to produce the m), ADP (to stimulate ATP synthesis), and CCCP (to increase the proton conductance) were added. Figure 4. Dependence on ATP concentration of rate of ATP hydrolysis in permeabilized mitochondria. Mitochondria from WT (●) or Inh1 (○) were permeabilized with digitonine and ATPase activity of complex V was assayed at pH = 8.0 (A) or pH = 6.0 (B). Data were adjusted with a Michaelis-Menten equation and were analyzed with the GraphPad software. The results were the average ± SD of 3 different preparations assayed by triplicates. Figure 5. In-gel activity of the mitochondrial OXPHOS complexes and supercomplexes. OXPHOS complex and supercomplexes from U. maydis WT and Inh1 strains were solubilized with different digitonin/protein ratio and analyzed by BN-PAGE. A) Coomassie brilliant blue stained native gel. In-gel activities assay of the complexes V (B), I (C), II (D), and IV (E). Bos taurus respiratory complex and supercomplexes were solubilized with digitonin (ratio of 2:1) and used as standard. Where indicate CI = complex I, CII = complex II, CIII2 = dimer of complex III2, CIV = complex IV, CV1 = monomeric state of complex V, and CV2 = dimeric state of complex V. Figure 6. Kinetics characterization of the monomeric and dimeric state of the F1FO- ATP synthase from WT and Inh1 strains. V1 and V2 from WT (V1WT and V2WT) and Inh1 (V1Inh1 and V2Inh1) were isolated and its in-gel ATPase activity analyzed (A, D, G, and J, respectively). Kinetics analysis was performed at pH = 8.0 (B, E, H, and K) or pH = 6.0 (C, F, I, and L). Close symbols = +DDM addition (0.005%); open symbols = no DDM added. Data were the average ± SD of three independent assays from 4 different preparations. Mitochondrial proteases act on STARD3 to activate progesterone synthesis in human syncytiotrophoblast Mercedes Esparza-Perusquía a,1, Sofía Olvera-Sánchez a,1, Oscar Flores-Herrera a, Héctor Flores-Herrera b, Alberto Guevara-Flores a, Juan Pablo Pardo a, María Teresa Espinosa-García a, Federico Martínez a,⁎ a Departamento de Bioquímica, Facultad de Medicina, Universidad Nacional Autónoma de México, Mexico b Departamento de Bioquímica y Biología Molecular, Instituto Nacional de Perinatología “Isidro Espinosa de los Reyes”, Mexico a b s t r a c ta r t i c l e i n f o Article history: Received 20 February 2014 Received in revised form 6 October 2014 Accepted 10 October 2014 Available online 18 October 2014 Keywords: Human syncytiotrophoblast mitochondria Progesterone synthesis STARD3 protein Mitochondrial metalloprotease Background: STARD1 transports cholesterol into mitochondria of acutely regulated steroidogenic tissue. It has been suggested that STARD3 transports cholesterol in the human placenta, which does not express STARD1. STARD1 is proteolytically activated into a 30-kDa protein. However, the role of proteases in STARD3modification in the human placenta has not been studied. Methods: Progesterone determination andWestern blot using anti-STARD3 antibodies showed thatmitochondri- al proteases cleave STARD3 into a 28-kDa fragment that stimulates progesterone synthesis in isolated syncytiotrophoblast mitochondria. Protease inhibitors decrease STARD3 transformation and steroidogenesis. Results: STARD3 remained tightly bound to isolated syncytiotrophoblastmitochondria. Simultaneous to the increase in progesterone synthesis, STARD3was proteolytically processed into four proteins, of which a 28-kDa protein was themost abundant. This protein stimulatedmitochondrial progesterone production similarly to truncated-STARD3. Maximum levels of protease activity were observed at pH 7.5 and were sensitive to 1,10-phenanthroline, which inhibited steroidogenesis and STARD3proteolytic cleavage. Addition of 22(R)-hydroxycholesterol increased proges- terone synthesis, even in the presence of 1,10-phenanthroline, suggesting that proteolytic products might be involved in mitochondrial cholesterol transport. Conclusion:Metalloproteases from human placental mitochondria are involved in steroidogenesis through the pro- teolytic activation of STARD3. 1,10-Phenanthroline inhibits STARD3 proteolytic cleavage. The 28-kDa protein and the amino terminal truncated-STARD3 stimulate steroidogenesis in a comparable rate, suggesting that both proteins share similar properties, probably the START domain that is involved in cholesterol binding. General significance: Mitochondrial proteases are involved in syncytiotrophoblast-cell steroidogenesis regulation. Understanding STARD3 activation and its role in progesterone synthesis is crucial to getting insight into its action mechanism in healthy and diseased syncytiotrophoblast cells. © 2014 Elsevier B.V. All rights reserved. 1. Introduction Mitochondria carry out cellular respiration and ATP synthesis to sup- ply the energy requirements of aerobic cells. Mitochondria are essential for the synthesis of a number of important biological compounds such as lipids, heme, amino acids, nucleotides and steroid hormones. The phys- iological function and homeostasis ofmitochondria entail selective prote- olysis in which various specific mitochondrial proteases, including processing peptidases, ATP-dependent proteases, and oligopeptidases are involved [1]. Steroid hormones are synthesized from cholesterol, a substrate for mitochondria of specialized cells of the adrenal cortex, gonads and pla- centa. The steroidogenic acute regulatory protein (StAR; STARD1) [2–6], a nuclear-encoded mitochondrial protein expressed upon stimulation of steroidogenic tissues by their respective trophic hormones [7–9], promotes cholesterol supply to mitochondria from acutely regulated steroidogenic tissue. It has been suggested that STARD3 (or MLN64), a member of the START domain family, is the protein responsible for transporting cholesterol in the human placenta [10], a steroidogenic tissue which does not show acute regulation of steroidogenesis nor expresses STARD1 [11]. The amino acid sequence of the STARD3 carboxy-terminal region sequence is similar to that of STARD1 [12]. While full-length STARD3 has minimal STARD1-like activity, the 234 amino-terminal residue deletion (N-218 STARD3) results in a protein with substantial STARD1-like activity in transfected cells [10]. Like N-62 STARD1 (a StAR protein with a deletion of 62 amino-acids in its amino- Biochimica et Biophysica Acta 1850 (2015) 107–117 ⁎ Corresponding author at: Departamento de Bioquímica, Facultad de Medicina, Universidad Nacional Autónoma de México, Apdo. Postal 70-159, Coyoacán 04510, México, D. F., México. Tel.: +52 55 56232168; fax: +52 55 56162419. E-mail address: fedem@bq.unam.mx (F. Martínez). 1 M. Esparza-Perusquía and S. Olvera-Sánchez contributed equally to this paper. http://dx.doi.org/10.1016/j.bbagen.2014.10.009 0304-4165/© 2014 Elsevier B.V. All rights reserved. Contents lists available at ScienceDirect Biochimica et Biophysica Acta j ourna l homepage: www.e lsev ie r .com/ locate /bbagen terminal region), N-218 STARD3 lacks a mitochondrial leader se- quence that prevents it from entering themitochondrion, and appar- ently exerts its function in the outer mitochondrial membrane. Alpy et al. [13] identified the full-length STARD3 to be associated with late endosomes, which constitute its sole cellular location reported to date. STARD1 and STARD3 transfer cholesterol from the outer to the inner mitochondrial membrane, where the cholesterol side chain cleavage enzyme, cytochrome P450 (P450scc, CYP11A1; EC 1.14.15.6), converts it to the first product in steroid hormone synthesis: pregnenolone [14–16]. Hypothetical mechanisms suggest that STARD1 activates cholesterol transfer by virtue of its association with a macromolecular complex that consists of outer membrane proteins such as the mitochondrial membrane translocator protein (TSPO), and the TSPO- associated protein PAP7, that bind and lead the regulatory subunit RI- α of the cAMP-dependent protein kinase (PKARIα) towards mitochon- dria [17]. However, it has been recently described that, in knockout mice with a specific TSPO deletion, gametogenesis, reproduction, histo- logical structure, and steroidogenesis of Leydig cells are not affected. This suggests that the presence of TSPO is not crucial for steroidogenesis [18]. Previous data from our laboratory have also demonstrated that TSPO (earlier named PBR) is absent from the human placenta [19]. In human placentamitochondria, STARD3 is associatedwith steroidogenic contact sites and HSP60, resulting in an increase in progesterone syn- thesis [20,21]. STARD1 is synthesized in the cytosol as a 37-kDa pre-protein car- rying an amino terminal targeting sequence that directs its import into mitochondria, where it is proteolytically processed to a mature 30-kDa protein [2,22–24]. Although the important role of STARD1 proteolysis during steroid hormone synthesis by acutely regulated steroidogenic tissue has been described, the role of proteases in the modification of STARD3 in the human placenta has not been studied. In this work, the participation of mitochondrial proteases in the modification of STARD3 in progesterone synthesis in human syncytiotrophoblast mitochondria was determined. Although the role of STARD3 in the human placenta remains to be elucidated, its proteol- ysis from a 55-kDa protein into lower molecular weight proteins ap- pears to be essential for placental steroidogenesis. 2. Materials and methods 2.1. Isolation of human syncytiotrophoblast mitochondria Full term human placentas were collected immediately after normal delivery. Mitochondria were prepared as previously described [25]. Briefly, placental cotyledons were placed in ice-cold 250 mM sucrose, 1 mMEDTA, and 10mMTris, pH 7.4. The suspension was homogenized with a Polytron (Brinkmann Instruments, Westbury, NY, USA) at 3000 rev/min for 1 min for two cycles with a one minute interval. The whole process was carried out at 4 °C. The pH of the homogenate was adjusted to pH 7.4 with Tris and centrifuged at 1500 g for 15 min. The supernatant was recovered and centrifuged at 4000 g to obtain a pellet of cytotrophoblast mitochondria (heavy mitochondria). The superna- tantwas centrifuged again at 16,000 g for 15min and the pellet contain- ing the syncytiotrophoblast mitochondria (light mitochondria) was suspended in the same solution and centrifuged at 1500 g for 10 min to remove the remaining erythrocytes. Then, the mitochondrial pellet was obtained by centrifugation of the last supernatant at 12,000 g for 10min. To purifymitochondria of syncytiotrophoblast, the enrichedmi- tochondrial suspension was loaded onto a 35% sucrose solution (25 ml) and centrifuged at 15,000 g for 45 min at 4 °C. The mitochondrial frac- tion was collected, suspended in 250 mM sucrose, 1 mM EDTA, and 10 mM Tris (pH 7.4) and centrifuged at 16,000 g for 15 min at 4 °C. The resulting mitochondrial pellet was suspended in this buffer and stored at 4 °C. Protein concentration was measured as reported by Refs. [26,27]. 2.2. Mitochondrial oxygen consumption Oxygen uptake was estimated polarographically using a Clark type electrode in a mixture containing 250 mM sucrose, 10 mM HEPES, pH 7.4, 1 mM EGTA, 1 mM EDTA, 10 mM succinate, 10 mM KH2PO4, 5 mM MgCl2, 0.2% bovine serum albumin and 1 mg/ml of syncytio- trophoblast mitochondrial protein [28]. Temperature was maintained at 37 °C and oxygen consumption was stimulated by the addition of 300–500 nmol ADP (state 3 of respiration). Respiratory control was de- fined as oxygen uptake rate of state 3/oxygen uptake rate of state 4 (state 4 started when all ADP was converted into ATP, and respiration slowed down) [29]. 2.3. Mitochondrial enzyme activity determination Activities of complex I (NADH:DCPIP oxidoreductase) and complex II (succinate:DCPIP oxidoreductase) were determined spectrophoto- metrically at 600 nmby following the reduction of the artificial electron acceptor 2,6-dichlorophenol-indophenol (DCPIP; 50 μM; DCPIP = 21 mM−1 cm−1). Mitochondria were permeabilized with 0.01% Triton X-100 and incubated in 30 mM KH2PO4, 5 mM MgCl2, 1 mM EGTA, 120mMKCl, pH 7.4, and either 500 μMNADH (complex I) or 2mMsuc- cinate (complex II). Complex II was activated by pre-incubation in the presence of 0.2 mM phenazinemethosulfonate (PMS) for 10 min at 25 °C [30,31]. The protein concentration of syncytiotrophoblast mito- chondria was 50 μg/ml and the reaction was started by the addition of NADH or succinate. ATP synthesis by complex V was measured at 37 °C using an assay coupled to the reduction of NADP+ ( 340 nm = 6.2 mM−1 cm−1). The reaction mixture contained 0.5 mM NADP+, 1 mM ADP, 6 units/ml glucose-6-phosphate dehydrogenase, 16 units/ ml hexokinase, 10 mM succinate, 100 μM P1,P5-di(adenosine-5′) pentaphosphate-ammonium, 10 mM glucose, 150 mM sucrose, 5 mM MgCl2, 20 mM Tris/HCl, and 20 mM KH2PO4, at pH 7.5. ATP synthesis was started by the addition of syncytiotrophoblast mitochondria (50 μg/ml). The values reported were obtained by subtracting the rate of ATP synthesis in the presence of oligomycin (5 μg/mgmitochon- drial protein) from the amount of ATP synthesis under control condi- tions [32]. 2.4. Mitochondrial progesterone synthesis Progesterone synthesis was determined at 37 °C as previously reported [28] in a medium (P4M) containing 120 mM KCl, 10 mM MOPS, 0.5 mM EGTA, 10 mM isocitrate, and 5 mM KH2PO4, pH 7.4 in a final volume of 500 μl with 1 mg/ml of syncytiotrophoblast mitochondrial protein. Where indicated, 25 μM 22(R)-hydro- xycholesterol was added to verify cytochrome P450scc, adrenodoxin, adrenodoxin reductase and 3β-hydroxysteroid dehydrogenase activi- ties [33]. After incubation, the reactionwas stoppedwith 75 μl methanol and progesteronewas determined using a radioimmunoassay kit (Diag- nostic Systems Laboratories, Inc. Webster, Texas, USA), following the manufacturer's instructions. The concentration of progesterone at time zero was subtracted from the amount of progesterone at different times and this net progesterone synthesis was reported. Alternatively, syncytiotrophoblast mitochondria were incubated for 20 min at 37 °C in P4M and centrifuged at 14,000 g in an Eppendorf 5415R refrigerated centrifuge for 15min at 4 °C. Themitochondrial pellet and the superna- tant were separated. The supernatant was concentrated in an Amicon Ultra Centrifugal Filter system (10K) and either processed for SDS- PAGE and Western blot analysis using anti-MLN64 antibodies (vide infra) or used to stimulate mitochondrial progesterone synthesis. 108 M. Esparza-Perusquía et al. / Biochimica et Biophysica Acta 1850 (2015) 107–117 2.5. Effect of protease inhibitors in proteolytic STARD3 cleavage and mitochondrial progesterone synthesis The effect of protease inhibitors in STARD3 cleavagewas assessed by incubatingmitochondria in PM4mediumwith either one of the follow- ing: the inhibitor mixture from Sigma (cat. P8215), PMSF (1mM), EGTA (5 mM), EDTA (5 mM), or 1,10-phenanthroline (9 mM), for 20 min at 37 °C. Afterwards, mitochondrial proteins were processed for SDS- PAGE and Western blot analysis against the STARD3 protein (vide infra). To verify the effects of STARD3 on progesterone synthesis, 2 μM of purified N-218 STARD3 was added to mitochondria incubated in the P4M medium. 2.6. In-gel protease activity assays SDS-polyacrylamide (8%) gels were co-polymerized with porcine gelatin (1 mg/ml) and loaded with syncytiotrophoblast mitochondria (50 μg per well) in non-denaturing loading buffer with or without β- mercaptoethanol. Electrophoresis was performed under constant cur- rent (10 mA per gel) for 6 h at 4 °C. Gels were washed in 2.5% Triton X-100 for 30 min to eliminate SDS remnants and incubated overnight at 37 °C in an activation buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 20 mM CaCl2, and 0.02% NaN3). The buffers used to determine pH- dependence activity were: sodium acetate (from pH 4 to 5.3), MES (from pH 5.3 to 7), Tris (from pH 7 to 8.8) and AMPSO (pH 8.8). Gels were stained with 0.2% Coomassie® brilliant blue R-250 (Sigma). MMP2 and MMP9, constitutively secreted by U937 (ATCC; Rockville, MD) (a promyelocyte cell line) were used as activity standard markers. Protease activity correlated with the unstained band and the densito- metric analysis was performed with the Image Analysis software ver- sion 1.0 (Thermo Fisher Scientific Inc.). The intensities of activity were measured by peak integration after densitometric analysis. 2.7. SDS gel electrophoresis and Western blot analysis Syncytiotrophoblast mitochondrial proteins (50 μg per well) were separated by SDS-PAGE according to Laemmli [34] in a 10% polyacryl- amide gel under denaturing conditions. After the run, proteins were stained with either Coomassie® brilliant blue R-250 or silver, using a commercial kit (Bio-Rad) (see Fig. 2D). Alternatively, proteins were electrotransferred to a PVDF membrane (Immobilon P; Millipore, Bed- ford, MA) in a semi-dry electroblotting system (Bio-Rad) at 25 V for 50 min. Membranes were blocked in 500 mM NaCl, 0.05% Tween-20, and 20 mM Tris–base, pH 7.5 (TTBS buffer), containing 5% blotting grade blocker non-fat drymilk (Bio-Rad). Then, membranes were incu- bated with anti-MLN64 polyclonal antibodies (1:1000). Immunoreac- tive bands were visualized with the Enhanced ChemiLuminescence assay (Amersham Life Science, Inc.), according to the manufacturer's instructions, using horseradish peroxidase-conjugated goat antimouse IgG (Pierce) at a dilution ratio of 1:35,000, and densitometric analyses were performed with the Image Analysis software version 1.0 (Thermo Fisher Scientific Inc.). The intensities of proteinsweremeasured by peak integration after densitometric analysis. The presence of endosome marker proteins, Niemann–Pick type C1 protein (NPC1) and Rab5 was assayed [35,36] by Western blot, as de- scribed above. The dilution ratio of either anti-NPC1 or anti-Rab5 anti- bodies was 1:2000. After being washed, the blots were incubated with the corresponding secondary antibodies. Protein–antibody complexes were visualized as described above. 2.8. Production of the N-218 STARD3 protein The recombinant N-218 STARD3 protein was produced in BL21 Escherichia coli expressing human STARD3-START (amino acids 218–445; N-218 STARD3) [10] as previously described [37]. The expressed protein contained a His6-tag at the C-terminus. Bacteria were cultivated in LBmedium containing 25 μg/ml ampicillin. For protein expression, 400 ml of growth medium (with antibiotic) was inoculated with 1 ml of BL21 overnight culture. The culture was incubated at 37 °C with constant shaking until an optical density of 0.5–1.0 at 600 nm was reached. Expression was induced by the addition of 0.5 M isopropyl-β- D-thiogalactopyranoside. After 4.5 h bacteria were pelleted. The resulting pellet was suspended in ice in 10 ml of the following buffer: 300 mM NaCl, 50 mM NaH2PO4, 20 mM Tris–HCl (pH 7.4), and 10 mM β-mercaptoethanol. Bacteria were sonicated in ice (15 pulses of 1 s, three times at maximum output level), using a MSE Soniprep (UK) model 150. The suspension was centrifuged at 4 °C for 30 min at 20,000 g. The supernatant was incubated with 500 μl of Ni2+– nitrilotriacetic acid–agarose matrix (Qiagen, Hilden, Germany). The mixture was incubated with constant rotation at 4 °C overnight. The matrix was placed in a column and washed with 20 ml of the following buffer: 300mMNaCl, 50 mMNaH2PO4 (pH 8.0), and 20mM imidazole. To avoid aggregation of N-218 MLN64, the elution buffer was supple- mented with 40% (w/v) glycerol. The eluted proteins were dialyzed (molecular mass cutoff: 12-kDa; Sigma) against the following buffer: 150 mM NaCl, 50 mM KCl, 50 mM Tris (pH 7.4), 10 mM dithiothreitol, and 40% (w/v) glycerol [37]. 2.9. Tandem mass spectrometry (LC/ESI–MS/MS) The protein band (indicated as 28-kDa in Fig. 2B) was cut off from the Coomassie® brilliant blue R-250-stained SDS-PAGE gels and sent to the Proteomics Core Facility at the University of Arizona, USA, to de- termine its identity. Outer mitochondrial membranes were isolated as reported by Uribe et al. [20]. Briefly, 20–25 mg of mitochondrial protein was incubated at 4 °C with 10 mM H3PO4 and adjusted to pH 7.3 with Tris base in the presence of protease inhibitor cocktail (Complete, Roche). After incuba- tion, sucrose was added to attain a final concentration of 0.38 M. The resulting mixture was incubated for 20 more minutes at 4 °C and then centrifuged at 12,500 g for 10 min. The supernatant containing the outer mitochondrial membranes was centrifuged at 137,000 g for 1 h to recover the membrane fraction. The resulting mitochondrial outer membrane was incubated in 100 mM ammonium bicarbonate (pH = 7.8) for 30 min, centrifuged at 100,000 g at 4 °C and sent to the Proteo- mics Core Facility at the University of Arizona, USA. 2.10. Statistical analyses Statistical analyses (one- and two-way analyses of variance, ANOVA) of the data were performed using Sigma Stat software, version 3.5. When necessary, nonlinear regression of the data to a single exponen- tial decay equation was performed in Sigma Plot software, version 10.0. 2.11. Materials Analytical grade reagents were purchased from Sigma Chemical Co. (St. Louis, MO, USA), E. Merck (Darmstadt, Germany), and Bio-Rad (Hercules, CA, USA). 3. Results 3.1. Functional state of syncytiotrophoblast mitochondria To determine the functional integrity of isolated syncytiotrophoblast mitochondria, respiratory rates and respiratory controls were calculated from oxygen uptake traces using succinate as substrate (Table 1). Oxygen uptake in state 3 and state 4 was 135 ± 28 ng atom of oxygen/min·mg protein, and 21 ± 8 ng atom of oxygen/min·mg protein, respectively. The respiratory control value was 6.7 ± 2 while ATP synthesis rate of complex V was 160 ± 12 nmol/min·mg protein. Addition of 2,4-dinitro- phenol to energized mitochondria increased the permeability of the 109M. Esparza-Perusquía et al. / Biochimica et Biophysica Acta 1850 (2015) 107–117 coupling membrane to protons, and induced maximum respiratory rate (205 ± 33 ng atom of oxygen/min·mg protein); no oligomycin- sensitive-ATP synthesis occurred. These data indicate functional coupling ofmitochondrial respiration andATP synthesis in syncytiotrophoblastmi- tochondrial preparations. Activities of 113±20 μmol/min·mgprotein for NADH:DCPIP oxidoreductase (complex I), and 14±4 μmol/min·mg pro- tein for succinate:DCPIP oxidoreductase (complex II) were also assessed (Table 1). Endosomemarker proteins, NPC1 and Rab5 [35,36], were not identi- fied by Western blot in isolated syncytiotrophoblast mitochondria. Taken together, these results indicate the presence of functional mito- chondria isolated from syncytiotrophoblast cells, capable of increasing oxygen consumption and of synthesizing ATP upon the addition of ADP. Human syncytiotrophoblast mitochondria are steroidogenic organ- elles that synthesize progesterone due to the presence of 3β- hydroxysteroid dehydrogenase in their inner membrane [25,38]. The rate of progesterone synthesis was 30.6 ± 1.3 ng progesterone/min·mg protein and was not modified when exogenous cholesterol was added (data not shown). Furthermore, previously reported data indicated that placental mitochondria have enough cholesterol [39] so addition of exog- enous cholesterol is not necessary to stimulate progesterone synthesis. The addition of 22(R)-hydroxycholesterol — a soluble substrate used to verify cytochrome P450scc, adrenodoxin, adrenodoxin reductase and 3β-hydroxysteroid dehydrogenase activities [33], increased steroidogenic activity to 85 ± 6.0 ng progesterone/min·mg protein (Table 1 and Fig. 5A). These results are in agreement with the specialized role of syncytiotrophoblastic tissue [25] and evidence that isolated mitochon- dria, as used in this work, retain their physiological function. 3.2. Syncytiotrophoblast mitochondrial protease activity The experiments performed in the present study were designed to assay the role of syncytiotrophoblast mitochondrial proteases in pro- gesterone synthesis. The first approachwas to detect the in-gel protease activities and its pH dependence (Fig. 1A). Densitometric analysis of in- gel protease activity showed that it was null at low pH (pH = 4.3), while an increase was observed starting at pH = 5.3 (Fig. 1A). Since the identity of the proteases in syncytiotrophoblast mitochondria is un- known (but an effort to elucidate it is in progress in our lab), the inten- sity of each activity band was pooled to obtain the total activity of proteases from each pH value (Fig. 1B). Total protease activity wasmax- imum at pH 7.5 (Fig. 1B) and five different protease activity bands were observed (Fig. 1A). Interestingly, the band with the lowest molecular weight (marked as E) was only detected at pH = 7.5 (Fig. 1A). Mito- chondrial protease activitywas comparedwithMMP9 andMMP2 activ- ity constitutively secreted by U937, used as protease activity standards (Fig. 1C). The proteolytic activity was sensitive to 1,10-phenanthroline (Fig. 1D) or β-mercaptoethanol (data not shown), suggesting that syncytiotrophoblast mitochondria contain metalloproteases. Table 1 Bioenergetics and steroidogenic parameters of syncytiotrophoblast mitochondria. Complexes activitiesa Complex I 113 ± 20 μmol/min·mg Complex II 14 ± 4 μmol/min·mg Complex V 160 ± 12 nmol/min·mg Oxygen uptake State 3b 135 ± 28 ng atom of oxygen/min·mg protein State 4c 21 ± 8 ng atom of oxygen/min·mg protein Respiratory controld 6.7 ± 2 Progesterone synthesise Control 30.6 ± 1.3 ng progesterone/min·mg +22(R)-hydroxycholesterol 85 ± 6.0 ng progesterone/min·mg a Specific activities from complexes I and II were measured spectrophotometrically in sonicated mitochondria: complex I, NADH:DCPIP oxidoreductase and complex II, succi- nate:DCPIP oxidoreductase. Complex II activity was stimulated as described in Materials and methods section. Specific complex V activity was determined in intact mitochondria as ATP synthesis. Values shows are the mean ± S.D. (n = 7 independent determinations from different placental tissue). b Defined as oxygen consumption stimulated by ADP added in presence of succinate as substrate. Values are the mean ± S.D. (n = 25 independent determinations from differ- ent placental tissue). c Defined as oxygen consumption reduction due to all ADP addedwas converted toATP. Values are themean ± S.D. (n = 25 independent determinations fromdifferent placental tissue). d Respiratory control = oxygen uptake rate of state 3/oxygen uptake rate of state 4. Values are themean ± S.D. (n = 25 independent determinations fromdifferent placental tissue). e Progesterone synthesis was determined as described inMaterials andmethods section. 22(R)-hydroxycholesterol was used to verify cytochrome P450scc, adrenodoxin, adrenodoxin reductase and 3β-hydroxysteroid dehydrogenase activities [30]. Values here are the mean ± S.D. from eight determinations from eight different placental tissues. Fig. 1. In-gel protease activity from syncytiotrophoblast mitochondria. A) pH dependence of protease activity. Buffer used: sodium acetate (frompH4 to 5.3),MES (frompH5.3 to 7), Tris (from pH 7 to 8.8) and AMPSO (pH 8.8), and continued with the protocol described in the Syncytiotrophoblast mitochondrial protease activity section. B) Densitometric analysis from in-gel protease activity shown in A. The density of each band of protease activitywas defined as band-intensity/area and the total protease activity shown is the sumof protease activity in each band. A significant increase in the intensity of total protease activity was observed at pH = 7.5, as compared to protease-activity at alkaline or acid pH values. The one-way ANOVA test indicates that these differences are statistically significant (indicated as a) (p b 0.001, n = 4, from four different placental tissues) (all Pairwise Multiple Comparison Procedures were performed with the Tukey test). Results are presented as the mean ± S.D.; a.u. = arbitrary units. C) Mitochondrial protease activity was determined at pH = 7.4 and compared toMMP2 andMMP9 proteases constitutively secreted byU937 used as prote- ase activity standard (ATCC; Rockville, MD). S means protease activity from syncytiotrophoblast mitochondria. D) Inhibition of protease activity by 1,10-phenanthroline (9 mM) added to protease incubation mixture (+1,10-PHEN). Gels were stained with Coomassie® brilliant blue R-250 and protease activity correlated with the unstained band. 110 M. Esparza-Perusquía et al. / Biochimica et Biophysica Acta 1850 (2015) 107–117 3.3. Syncytiotrophoblast mitochondrial protease and progesterone synthesis In steroidogenic tissues acutely regulated, STARD1 promotes choles- terol transfer from the outer to the inner mitochondrial membrane. Once in the mitochondrion, steroidogenesis increases and STARD1 is proteolytically cleaved. Since the human placenta expresses STARD3 in- stead of STARD1 [10,21] we investigated the possible relationship among protease activity, STARD3 proteolytic cleavage, and proges- terone synthesis in syncytiotrophoblast mitochondria (Fig. 2). It is important to mention that STARD3 remains tightly bound to isolat- ed syncytiotrophoblast mitochondria (Fig. 2B and [21]), although it was described as an endosome protein [13]. The endosome markers Rab5 and NPC1 were not detected in the syncytiotrophoblast mito- chondrial fraction (data not shown), in accordance with previous reports [21]. Since protease activity was dependent on pH, progesterone synthesis and STARD3 proteolysis were analyzed at different pH values. The syn- thesis of progesterone increased frompH=5 to 7.4 (Fig. 2A, black bars), but a decrease was observed at pH = 9. This decrease might be due to the steroidogenic machinery being affected by the alkaline conditions. The effect of pH on cytochrome P450scc and 3β-hydroxysteroid dehydrogenase activities was verified by the addition of 22(R)- hydroxycholesterol to mitochondria incubated under different pH con- ditions (Fig. 2A, gray bars). As anticipated, progesterone productionwas increased by 22(R)-hydroxycholesterol at different pH values, with maximum levels of production detected at pH = 7.4, confirming the inhibitory effect of acid or alkaline pH on cytochrome P450scc and 3β- hydroxysteroid dehydrogenase. Concomitant with the increase in pro- gesterone synthesis at pH = 7.4, Western blot assays for STARD3 revealed that the 55-kDa protein was proteolytically processed into four proteins released from mitochondria with molecular weights of 27, 28, 31, and 33-kDa (Fig. 2B). Densitometric analysis showed that the relative signal of the STARD3 55-kDa protein was similar at every pH value tested. However, the relative intensity of the proteolytic prod- ucts increased as pH became alkaline (Fig. 2B and C). The initial proteo- lytic product was the 27-kDa protein, although the signal of the 28-kDa protein appeared at pH= 6.0. The signal increased and the 28-kDa pro- tein was shown to be the main proteolytic product throughout the pH range tested (Fig. 2C). STARD3 cleavagematched the increase of proges- terone synthesis in the physiologic pH range. Although the actual 28- kDa/STARD3 stoichiometry could not be estimated by densitometric analysis, the results are consistent with the hypothesis that STARD3 must be proteolytically processed into a smaller protein that promotes progesterone synthesis in the human placenta [40]. Since Western blot results suggested that STARD3 was processed into a protein belonging to a family of low molecular weight proteins (Fig. 2B), it was important to define its identity. MS/MS analysis of the 28-kDa protein showed that it contains the START-domain sequence (Fig. 3). Moreover, the 28-kDa protein shared 68% of protein identity and 72% of protein similitude to the human STARD3 (Fig. 3). These results confirm the identity of the 28-kDa protein as STARD3, and there- forewill be indicated as STARD3-28 kDa from now on. Interestingly, the size of STARD3-28 kDa is approximate to that of the domain of N-218 Fig. 2. Proteolytic cleavage of STARD3 and progesterone synthesis in syncytiotrophoblast mitochondria. A) pH dependence of progesterone synthesis in the presence (dashed bars) or absence (black bars) of 9 mM 1,10-phenanthroline. The progesterone synthesis medium described in theMaterials andmethods sectionwas supplementedwith the following buffer: sodium acetate (from pH 4 to 5.3), MES (from pH 5.3 to 7), Tris (from pH 7 to 8.8) and AMPSO (pH 8.8), and the results are expressed as the mean ± S.D. of at least five separate experiments, with five different placental tissues. The one-way ANOVA analysis of control conditions (black bars) showed a significant increase in progesterone (P4) synthesis at pH = 7.4 when compared to P4 production at lower or higher values of pH (indicated by a). The difference is greater than expected by chance and there is a statistically significant dif- ference (p ≤ 0.001, n = 8) (all Pairwise Multiple Comparison Procedures were performed with the Tukey test). 100% of progesterone synthesis taken at pH 7.4 was considered as the maximum with a value of 30 ± 1.3 ng P4/mg/min. The two way ANOVA analysis showed a statistically significant difference between the treatment groups when 1,10-phenantroline was present (dashed bars) or absent (black bars) (indicated as b) (p ≤ 0.005, n = 16) (all Pairwise Multiple Comparison Procedures was performed with the Holm–Sidak method). The addition of 22(R)-hydroxycholesterol to mitochondria increased P4 synthesis (gray bars). A two-way ANOVA analysis showed a statistically significant difference between con- trol conditions and 22(R)-hydroxycholesterol addition (indicated as c) (p ≤ 0.001, n = 5) (all Pairwise Multiple Comparison Procedures were performed with the Holm–Sidak meth- od). B)Western blot against STARD3 protein. Mitochondria were incubated in P4Mmedium asdescribed in theMaterials andmethods section, and after 20minof incubation at 37 °Cmi- tochondrial proteins were resolved in SDS-PAGE, and processed for Western blot or stained with Coomassie® brilliant blue R-125 (D). C) Densitometric analysis from Western blot shown in B. The one-way ANOVA of each band intensity at different pH values showed a sig- nificant increase in the 28-kDa protein band intensity at pH=6.5 (indicated by a), 7.0 (indi- cated by b), 7.4 (indicated by c), and 8.0 (indicated by d), when compared to the other STARD3 proteolytic products (i.e. 27, 31 and 33-kDa proteins) (p ≤ 0.001, n = 4, from four different placental tissues) (all Pairwise Multiple Comparison Procedures were performed with the Tukey test), whereas the 55-kDa protein (STARD3 full length protein) intensity was similar at all pH values tested. The density of each band was defined as band-intensi- ty/area. a.u. = arbitrary units. 111M. Esparza-Perusquía et al. / Biochimica et Biophysica Acta 1850 (2015) 107–117 STARD3 (Fig. 3) reported by Ref. [40], whichmight have a role in choles- terol flux from the outer to the innermitochondrial membrane. Howev- er, no mitochondrial targeting presequence was observed. To verify the participation of proteases in STARD3 cleavage during progesterone synthesis, the effect of 1,10-phenanthroline, a metalloprotease inhibitor, on steroidogenesis was studied (Fig. 4). The time course of progesterone synthesis showed that 1,10-phenanthroline inhibits steroidogenesis (Fig. 4A). Indeed, progesterone synthesis was inhibited by 1,10-phenanthroline depending on its concentration and was totally abolished at 9 mM (Fig. 4B). Additionally, the 1,10- phenanthroline effect on progesterone synthesis at different pH values was explored (Fig. 2A, dashed bars). Once again, the protease inhibitor abolished mitochondrial steroidogenesis at every pH value, suggesting protease participation inprogesterone synthesis. Simultaneously, the pro- teolytic cleavage of STARD3 was abolished by 1,10-phenanthroline (Fig. 4C). Other protease inhibitors like the protease inhibitor cocktail (Sigma), PMSF, EGTA, or EDTA (data not shown) prevented STARD3 cleavage into low molecular weight proteins (Fig. 4C). 3.4. STARD3 cleavage participates in the placental steroidogenesis Although the relationship between protease activity, STARD3 cleav- age, and progesterone synthesis had been demonstrated, it was necessary to determine the potential role of proteolytically-cleaved STARD3. 22(R)- hydroxycholesterol – a cholesterol analog that reaches P450scc indepen- dently of the mitochondrial transport system used by cholesterol – was added to syncytiotrophoblast mitochondria to stimulate progesterone synthesis when the proteolytic cleavage of STARD3 into the STARD3- 28 kDa protein was inhibited by 1,10-phenanthroline (Fig. 5A). 22(R)- hydroxycholesterol produced a three-fold increase in progesterone syn- thesis, even in the presence of 1,10-phenanthroline (Fig. 5A), which inhibited the formation of the STARD3-28 kDa protein (Fig. 4C), sug- gesting the following implications: 1) P450scc and all the enzymes involved in progesterone synthesis are functional during protease inhi- bition, and 2) protease activity might be involved in the transport of cholesterol, i.e. the STARD3-28 kDa protein is obtained from STARD3 proteolytic cleavage. When purified the N218-STARD3 protein was added to syncytiotrophoblast mitochondria (Fig. 4C, last lane) and an increase in progesterone synthesis was observed (Fig. 5B). This result suggests that STARD3 proteolysis is an important step in the humanpla- centa progesterone synthesis. In an attempt to determine the sub-mitochondrial site where the pro- teolytic cleavage of STARD3 occurs, syncytiotrophoblast mitochondria were centrifuged during progesterone synthesis. The supernatant, where proteins not bound tomitochondriawere released, was recovered. This procedure allowed isolating the low molecular weight proteins derived from STARD3 proteolysis (Fig. 6A). It is important to mention that mitochondria showed a respiratory control of 3.2 ± 1.1 after centri- fugation, which confirms that the inner membrane remained intact. This suggests that the proteolytic cleavage of STARD3 could take place in the intermembrane space and that proteins are released through the outer membrane as a consequence of centrifugation (Fig. 6A). The subsequent hypothesis is that STARD3 is cleaved in the intermembrane space where its proteolytic products (mainly STARD3-28 kDa) are involved in Fig. 3. Identification of the 28-kDa protein produced during steroidogenesis from human syncytiotrophoblast mitochondria as STARD3. The MS/MS analysis of the 28-kDa protein pro- duced three peptides whose sequences are indicated in the boxes. The coverage of the 28-kDa protein was 16.3% and its identity was defined as STARD3 protein (ID J3QLM1_HUMAN fromUNIPROTKB/TrEMBL). Sequences fromwhole humanplacenta STARD3 (indicated as STARD3-28 kDa) and STARD3 (NP_001159410 fromNCBI)were aligned and showed an identity of 68% and 72% similitude (Clustal W program). START domain is underlined with a solid line; STARD3-28 kDa sequence is underlined with a dashed line. The arrow indicates the first amino acid in the N218-STARD3 reported by Ref. [40]. 112 M. Esparza-Perusquía et al. / Biochimica et Biophysica Acta 1850 (2015) 107–117 cholesterol efflux between mitochondrial membranes. The high concen- tration of progesterone bound to a protein fraction that was released from the mitochondria (Fig. 6B, white bars) evidences the interaction of these peptides with steroid molecules. Denaturation of these lowmolec- ular weight proteins with cold methanol releases progesterone (Fig. 6B, black bars and Ref. [21]). To verify the effect of the STARD3-28 kDa protein in progesterone synthesis,mitochondrial protein supernatantwas collected and concen- trated (as described in theMaterials andmethods section) and added to fresh and intact syncytiotrophoblast mitochondria (Fig. 6C, close cir- cles), and progesterone synthesis was determined. Simultaneously, another set of fresh mitochondria was incubated with isolated N-218 STARD3 as control (Fig. 6C, open circles). The addition of the released protein fraction (i.e. STARD3-28 kDa protein) to fresh mitochondria in- duced an increase in progesterone synthesis just as N-218 STARD3does, suggesting a similar role for both proteins. 4. Discussion The physiological functions of human syncytiotrophoblast are cru- cial for the maintenance of pregnancy. Our group is interested in the study of themolecular mechanisms involved in the synthesis of proges- terone by the human placenta. The study described here sought to reveal the role of mitochondrial proteases regarding STARD3 during progesterone synthesis in mitochondria from syncytiotrophoblast cells. It has been put forward that the STARD3 protein, similarly to STARD1, transfers cholesterol from the outer to the innermitochondrial membrane [40]. Although themolecularmechanism of STARD1 activity is still unknown, two models have been proposed. In the first one, STARD1 transfers cholesterol during its import into mitochondria. In the matrix, proteases degrade STARD1 to prevent its accumulation and subsequent mitochondrial damage [41–43]. Thus, it has been pro- posed that STARD1 import into the mitochondrial matrix serves as an off-switch for STARD1 activity. In the second model, mitochondrial im- port is not required for STARD1 activity, since N-terminally truncated STARD1 proteins retain full activity and it was not imported into mito- chondria [4,6,22,44]. Despite the important role of STARD1, some steroidogenic tissues do not express it, as the human placenta. In this regard, elucidating the function of STARD3 is of considerable interest since it might promote steroidogenesis in tissues that do not express STARD1. Because the STARD1 role in steroidogenesis is associated with mitochondrial prote- ase activity, in the present work we determined both, protease activity in syncytiotrophoblast mitochondria and proteolytic activation of STARD3. Protease activity in syncytiotrophoblast mitochondria was associated with five different bands with an optimal pH = 7.5, and was sensitive to 1,10-phenanthroline, EGTA, EDTA, PMSF and β-mercaptoethanol, which suggests the presence of metalloproteases (Figs. 1, 2B and 4C). Western blotting of isolatedhuman syncytiotrophoblastmitochondria revealed the presence of apparently full-length (55-kDa) STARD3 and various proteolytic products (27, 28, 31, and 33-kDa) (Fig. 2B). Although STARD3 has been reported to be associated with endosomes [13], STARD3-antibodies recognized a protein of approximately 54 kDa in the isolated syncytiotrophoblast mitochondria, where Rab5 and NPC1 were not detected. This result is in accordance with Ref. [21]. A possible expla- nation is that some endosomes could still be present in the pellet that comprises isolated mitochondria. Also, this result could indicate a close Fig. 4. Effect of 1,10-phenanthroline on mitochondrial progesterone synthesis and STARD3 cleavage. A) Time course of mitochondrial progesterone synthesis in the presence (○) or absence (●) of 9 mM 1,10-phenanthroline. Mitochondria were incubated in P4Mmedium at 37 °C and at the indicated times, an aliquot was removed and progesterone determined as described in the Materials and methods section. Results are presented as the mean ± S.D. of three separate experiments performed with three different placental tissues. B) Inhibition of progesterone synthesis by 1,10-phenanthroline (1,10-PHEN). The inhibition follows an exponential decay and the percentage of inhibitionwas calculated as the ratio be- tween progesterone synthesis at the indicated 1,10-phenanthroline concentration against control conditions. The data were fitted to the equation f = y0 + a · exp(−b · x) using the Sigma Plot software, where y0 = −32.6 ± 7.5; a = 143.72 ± 16.4; b = 0.118 ± 0.025; R = 0.9998. C) Immunodetection of STARD3 protein during mitochondrial progesterone synthesis. Mitochondria were incubated in P4M for 20 min at 37 °C in the presence of an inhibitor cocktail from Sigma (cat. P8215), or PMSF (1 mM), or EGTA (5 mM), or 1,10- phenanthroline (9 mM) and then processed for Western blot analysis. Additionally, purified N-218 STARD3 (85 μM) was added to mitochondria. Fig. 5. Effect of 22(R)-hydroxycholesterol and purified N-218 STARD3 protein on proges- terone synthesis. A) Syncytiotrophoblastmitochondriawere incubated in P4Mmixture for progesterone (P4) synthesis and 25 μM 22(R)-hydroxycholesterol was added. P4 was de- termined as described in the Materials and methods section. 1,10-Phenanthroline (1,10- PHEN) concentrationwas 9mM. Results are presented as themean± S.D. of five separate experiments performedwith five different placental tissues. The one-wayANOVA analysis showed a significant increase of P4 synthesis in the presence of 22(R)-hydroxycholesterol against control conditions (indicated as a) (p≤ 0.0001, n= 4, from four different placen- tal tissues), or against 1,10-PHEN addition (indicated as b) (p ≤ 0.0001, n = 4, from four different placental tissues) (all PairwiseMultiple Comparison Procedures were performed with the Tukey test). B) Time course of mitochondrial progesterone synthesis with (□) or without (■) N-218 STARD3 protein (85 μM). N-218 STARD3 protein was overexpressed and purified as described in the Materials and methods section. Results are presented as the mean ± S.D. of seven separate experiments performed with seven different placental tissues. The one-way ANOVA analysis showed a statistically significant difference in P4 production, which increased in the presence of N-218 STARD3 protein as compared to control conditions (indicated as a for 20 min; b for 40 min; c for 60 min) (p ≤ 0.001, n= 4, from four different placental tissues) (all PairwiseMultiple Comparison Procedures were performed with the Tukey test). 113M. Esparza-Perusquía et al. / Biochimica et Biophysica Acta 1850 (2015) 107–117 relationship betweenmitochondria and endosomes during placental ste- roidogenesis. Nevertheless this hypothesis should be further investigated. The fact that inner-membrane-impermeable 1,10-phenanthroline, EDTA and EGTA inhibited STARD3 proteolytic cleavage (Fig. 4C) suggests that metalloproteases could be located in the intermembrane space (IMS), i.e. the YME1L, an i-AAA protease that exerts its activity on the IMS side of the inner membrane of mitochondria (Fig. 7) [45, 46], or in the cytoplasmic side of the outer membrane. Proteolytic Fig. 6. Release of proteolytically cleaved STARD3 from syncytiotrophoblast mitochondria during progesterone synthesis. Syncytiotrophoblast mitochondria were incubated in P4Mmedi- um for 20 min at 37 °C as described in the Materials and methods section, and then mitochondria were centrifuged and the mitochondrial pellet and supernatant were separated and analyzed byWestern blot against STARD3 (A); progesterone content was also determined (B). Where indicated, (■) corresponds to mitochondrial pellet and (□) corresponds to protein released to supernatant. A one-wayANOVA test indicated that the differences in progesterone content betweenmitochondrial pellet and supernatant are statistically significant (statistical significance indicated as a for 20min; b for 40min; c for 60min) (p b 0.001, n= 7, from seven different placental tissues) (all Pairwise Multiple Comparison Procedures were performed with the Tukey test). Results are presented as themean, and error bars indicate S.D. of seven separate experiments. The effect of isolated N-218 STARD3 (○) ormitochondrial supernatant (●) on progesterone synthesis is shown in (C). Protein released frommitochondria during progesterone synthesis was collected, pooled and concentrated in an Amicon Ultra Centrifugal Filter (10K) and added to fresh syncytiotrophoblast mitochondria to determine its effect on progesterone synthesis. Simultaneously, another set of mitochondria were incubated with N- 218 STARD3 protein (85 μM), and progesterone synthesis was determined. No significant differences were observed between protein released frommitochondria and N-218 STARD3 in progesterone synthesis stimulation. The total progesterone content of the mitochondrial supernatant protein was 900 ng of progesterone/mg of protein/60 min, and it was subtracted from the values showed in the graph. Fig. 7.Model proposed for the role ofmitochondrial proteasomes in STARD3 cleavage and progesterone synthesis in human syncytiotrophoblast cells. Protease activities in the activation of MLN64 and the subsequent progesterone synthesis increase have been divided into three steps. Step 1: Proteolytic transformation of STARD3 from a 55-kDa into a 28-kDa protein by a protease that exerts its activity on the IMS side of the inner membrane of mitochondria or at the outer membrane (see text for details). This protease is sensitive to 1,10-phenanthroline, EGTA, or EDTA (shown as an empty arrow). Step 2: The STARD3-28 kDa protein, which has been shown to contain the cholesterol binding domain ([40] and the present work), would increase cholesterol flux from the outer into the inner mitochondrial membrane to reach the cytochrome P450scc machinery and increase progesterone production (Step 3). If protease activity is inhibited and no STARD3-28 kDa protein is produced, 22(R)-hydroxycholesterol might promote progesterone synthesis. STARD3 incorporation to mitochondria could occur without a classical mitochondrial targeting presequence (A) or through the association between mitochondria and lipid droplets via the SNARE complex (B). OM= outer mitochondrial membrane; IMS = intermembrane space; IM= inner mitochondrial membrane; steroidogenic contact site = marked with a dashed line; Chol = cholesterol; P450scc = cytochrome P450scc; 3βHSDH = 3β-hydroxysteroid dehydrogenase; 22-OH-Chol = 22(R)-hydroxycholesterol. 114 M. Esparza-Perusquía et al. / Biochimica et Biophysica Acta 1850 (2015) 107–117 activation of STARD3 in the intermembrane space or in the outer mem- brane is also supported by the fact that STARD3contains nomitochondrial leader sequence and hence does not enter the mitochondrial matrix. Instead its proteolytic products (i.e. STARD3-28 kDa protein) could be re- leased from mitochondria by centrifugation without inner membrane damage (Fig. 6A). Although STARD3 lacks a classical mitochondrial targeting presequence, it is tightly joined to syncytiotrophoblast mito- chondria and its role in steroidogenesis is demonstrated in the present work. It has been reported that many mitochondrial hydrophobic mem- brane proteins are synthesized without cleavable extensions [47]. These proteins typically contain several targeting signals that are distributed throughout the length of the protein [48]. This could be the case for STARD3. Once incorporated into mitochondrial membranes, STARD3 might be proteolytically activated and then associatedwith steroidogenic contact sites to promote cholesterol transport (Fig. 7A). Although this hy- pothesis needs to be elucidated, STARD3 intermembrane space location is important due to the possible role that has been suggested for the STARD3-28 kDa protein in cholesterol transfer between mitochondrial membranes [40]. Additionally, a model for the incorporation of STARD3 into mito- chondria and its activation was put forward, based on the results from mass spectrometric analysis of mitochondrial outer membrane proteins (Fig. 7B). It has been reported that lipid droplets contain constituent proteins of the SNARE complexes [49–53], which include α-SNAP, Syntaxins and VAMP. Human syncytiotrophoblast mito- chondria contained SNAP (P54920_SNAA_HUMAN), Syntaxin-3 (F8W9Y0_STX3_HUMAN), Syntaxin-7 (O15400-2_STX7_HUMAN), Syntaxin-12 (Q86Y82_STX12_HUMAN), Syntaxin-binding Protein-2 (STXBP-2, Q15833_E7EQD5_HUMAN), Syntaxin-binding Protein-3 (STXBP-3, O00186_STXB3_HUMAN), and VAMP-8 proteins (Q9BV40_VAMP8_HUMAN). It has been shown that the SNAP pro- tein promotes interaction between lipid droplets and mitochondria [54], and that steroidogenic cells express SNARE proteins such as Syntaxin-17, SNAP-23, and SNAP-25 [55–59]. These observations strongly suggest that SNARE proteins might mediate cholesterol transport from lipid droplets to steroidogenic mitochondria, most likely by promoting the functional interaction between lipid droplets and mitochondria. In syncytiotrophoblast cells, the STARD3 protein might be incorporated into mitochondria from lipid droplets through SNARE complexes and proteolytically activated by mito- chondrial proteases. Once STARD3-28 kDa has been produced, it might then be incorporated to steroidogenic contact sites [20] and promote cholesterol transport for progesterone synthesis (Fig. 7). Steroidogenic contact sites are multiprotein complexes associated with cholesterol transport and steroidogenesis [20]. Several proteins like HSP60 might be involved, as suggested by Ref. [21]. Although TPSO is a protein apparently essential for several mitochondrial processes, its func- tion does not seem to be crucial to the permeability of the transition pore [60], the steroidogenesis of Leydig cells [18] and the human placenta [19]. Nevertheless, it has been reported that the rate-determining step of placental progesterone synthesis is the electron supply to cytochrome P450scc from adrenodoxin reductase [61,62]. It has been demonstrated that in purified human syncytiotrophoblast mitochondria [63] or in the isolated steroidogenic contact sites [20], the addition of 22(R)- hydroxycholesterol increases progesterone synthesis compared with the control condition (i.e.mitochondria were incubated with isocitrate as oxidable substrate in a medium that promotes progesterone synthe- sis; see Refs. [20,21,28,32] and theMaterials andmethods section). This suggests that electron supply to P450scc might not be the limiting-step. The addition of 22(R)-hydroxycholesterol, a cholesterol analog that freely reaches cytochrome P450scc [33], rendered three-fold increases in progesterone production, even if proteases were inhibited with 1,10-phenanthroline (Fig. 5A) and no STARD3 proteolytic cleavage oc- curred (Fig. 4C). This evidences a specific effect of the protease inhibitor, as well as that the proteolytic products of STARD3 might be involved in cholesterol transport. Mitochondrial metalloprotease activity, relative to alkaline pH; simul- taneous proteolytic activation of STARD3; and an increase in progesterone synthesis, were observed (see Figs. 1, 2 and 4). Moreover, the proteolysis of 55-kDa STARD3 into a STARD3-28 kDa proteinwas associatedwith the maximal rate of progesterone synthesis observed (Fig. 2). It has been shown that the N-218 STARD3 protein (with similar size to STARD3- 28 kDa) has substantial StAR-like activity in transfected cells [10]. In addition, Bose et al. [40] used a total protein homogenate frommidterm human placenta to show that the presence of the full-length STARD3 protein and various proteolytic products, among which the 28-kDa peptidewas predominant. However, purified syncytiotrophoblastmito- chondria were used in this work to demonstrate that mitochondrial metalloproteases could be responsible for the proteolytic transforma- tion of STARD3 into the STARD3-28 kDa protein, which could be in- volved in cholesterol transfer between mitochondrial membranes. Furthermore, the addition of the protein N-218 STARD3 or the STARD3-proteolytic-products (i.e. the STARD3-28 kDa protein which is released from mitochondria while progesterone synthesis occurs) to isolated syncytiotrophoblastmitochondria produced a similar enhance- ment of steroidogenesis (Fig. 6C), highlighting the role of STARD3 and its proteolytic product, the STARD3-28 kDa protein. Currently, transfec- tion of HEK-293 cells with thewhole progesterone synthesismachinery and the N-218 STARD3 protein is being performed in our laboratory to determine if STARD3 cleavage by proteases is the key step in progester- one synthesis. However, with all the results described so far, we pro- pose that STARD3 is processed in vivo by metalloproteases from human syncytiotrophoblast mitochondria to the STARD3-28 kDa pro- tein, a product similar to N-218 STARD3 that might promote steroido- genesis in a similar way to that described in the proposed model for the STARD1 protein (Fig. 7). Although less is known about the roles ofmitochondrial proteases in mammalian cells, it has been shown that loss-of-function mutations in human genes encoding mitochondrial proteases are often associated with clinical disorders [64–67]. Therefore, molecular and biochemical characterization of such proteolytic activities is of the utmost impor- tance, as suggested by Ref. [68]. Finally, future research using STARD3 knockoutmice [69]will be relevant to determine the specific cellular lo- cation of STARD3 and its role in reproduction. The present study suggests that the STARD3 protein can be used as an authentic natural substrate to explore multiple mitochondrial prote- ases, to provide new insights into their mode of action in healthy and diseased steroidogenic cells, and to allow understanding of the different ways in which steroids are produced. Acknowledgements This work was supported by grants IN211912, IN217609 and IN214914 from Dirección General de Apoyo al Personal Académico de la Universidad Nacional Autónoma de México, and the grant 168025 from Consejo Nacional de Ciencia y Tecnología (CONACYT). Mercedes Esparza-Perusquía is a PhD student of the Biological Science Program of Universidad Nacional Autónoma de México (511021118) and fellow to CONACYT (254400). Héctor Flores-Herrera is a PhD student of the Biomedical Science Program of Universidad Nacional Autónoma de México (513025057). We thank Dr. Jerome Strauss (Virginia Common- wealth University) for the STARD3 (MLN64)-Ab. We also thank Dr. José Luis Pérez-García (Facultad de Medicina, UNAM) and Dra. Elizabeth Rodríguez Salinas for reviewing the correct usage of English in thisman- uscript andDr. Esther Urrutia for her support in the statistical analysis of data. References [1] M. Koppen, T. Langer, Protein degradation within mitochondria: versatile activities of AAA proteases and other peptidases, Crit. Rev. Biochem. Mol. Biol. 42 (2007) 221–242. 115M. Esparza-Perusquía et al. / Biochimica et Biophysica Acta 1850 (2015) 107–117 [2] B.J. Clark, J. Wells, S.R. King, D.M. Stocco, The purification, cloning, and expression of a novel luteinizing hormone-inducedmitochondrial protein inMA-10mouse Leydig tumor cells. Characterization of the steroidogenic acute regulatory protein (StAR), J. Biol. Chem. 269 (1994) 28314–28322. [3] D. Lin, T. Sugawara, J.F. Strauss III, B.J. Clark, D.M. Stocco, P. Saenger, A. Rogol, W.L. Miller, Role of steroidogenic acute regulatory protein in adrenal and gonadal ste- roidogenesis, Science 267 (1995) 1828–1831. [4] C.B. Kallen, J.T. Billheimer, S.A. Summers, S.E. Stayrook, M. Lewis, J.F. Strauss III, Ste- roidogenic acute regulatory protein (StAR) is a sterol transfer protein, J. Biol. Chem. 273 (1998) 26285–26288. [5] D.M. Stocco, The steroidogenic acute regulatory (StAR) protein two years later. An update, Endocrine 6 (1997) 99–109. [6] X. Wang, Z. Liu, S. Eimerl, R. Timberg, A.M. Weiss, J. Orly, D.M. Stocco, Effect of trun- cated forms of the steroidogenic acute regulatory protein on intramitochondrial cholesterol transfer, Endocrinology 139 (1998) 3903–3912. [7] D.M. Stocco, B.J. Clark, Regulation of the acute production of steroids in steroidogen- ic cells, Endocr. Rev. 17 (1996) 221–244. [8] J. Orly, D.M. Stocco, The role of the steroidogenic acute regulatory (StAR) protein in female reproductive tissues, Horm. Metab. Res. 31 (1999) 389–398. [9] N. Cherradi, M.F. Rossier, M.B. Vallotton, R. Timberg, I. Friedberg, J. Orly, X.J. Wang, D.M. Stocco, A.M. Capponi, Submitochondrial distribution of three key steroidogenic proteins (steroidogenic acute regulatory protein and cytochrome P450scc and 3β- hydroxysteroid dehydrogenase isomerase enzymes) upon stimulation by intracellu- lar calcium in adrenal glomerulosa cells, J. Biol. Chem. 272 (1997) 7899–7907. [10] H. Watari, F. Arakane, C. Moog-Lutz, C.B. Kallen, C. Tomasetto, G.L. Gerton, M.C. Rio, M.E. Baker, J.F. Strauss III, MLN64 contains a domain with homology to the steroido- genic acute regulatory protein (StAR) that stimulates steroidogenesis, Proc. Natl. Acad. Sci. U. S. A. 94 (1997) 8462–8467. [11] T. Sugawara, J.A. Holt, D. Driscoll, J.F. Straus III, D. Lin, W.L. Miller, D. Patterson, K.P. Clancy, I.M. Hart, B.J. Clark, D.M. Stocco, Human steroidogenic acute regulatory pro- tein: functional activity in COS-1 cells, tissue-specific expression, and mapping of the structural gene to 8p11.2 and a pseudogene to chromosome 13, Proc. Natl. Acad. Sci. U. S. A. 92 (1995) 4778–4782. [12] C. Moog-Lutz, C. Tomasetto, C.H. Regnier, C. Wendling, Y. Lutz, D. Muller, M.P. Chenard, P. Basset, M.C. Rio, MLN64 exhibits homology with the steroidogenic acute regulatory protein (STAR) and is over-expressed in human breast carcinomas, Int. J. Cancer 71 (1997) 183–191. [13] F. Alpy, M.-E. Stoeckel, A. Dierich, J.-M. Escola, C. Wendling, M.-P. Chenardi, M.T. Vanier, J. Gruenberg, C. Tomasetto, M.-C. Rio, The steroidogenic acute regulatory protein homolog MLN64, a late endosomal cholesterol-binding protein, J. Biol. Chem. 276 (2001) 4261–4269. [14] P.F. Hall, Cytochromes P-450 and the regulation of steroid synthesis, Steroids 48 (1986) 131–196. [15] E.R. Simpson, M.R. Waterman, Regulation of the synthesis of steroidogenic enzymes in adrenal cortical cells by ACTH, Annu. Rev. Physiol. 50 (1988) 427–440. [16] W.L. Miller, Mitochondrial specificity of the early steps in steroidogenesis, J. Steroid Biochem. Mol. Biol. 55 (1995) 607–616. [17] J. Liu, M.B. Rone, V. Papadopoulos, Protein–protein interactions mediate mitochon- drial cholesterol transport and steroid biosynthesis, J. Biol. Chem. 281 (2006) 38879–38893. [18] K. Morohaku, S.H. Pelton, D.J. Daugherty, W.R. Butler, W. Deng, V. Selvaraj, Translocator protein/peripheral benzodiazepine receptor is not required for steroid hormone biosynthesis, Endocrinology 155 (2014) 89–97. [19] M.G. Maldonado-Mercado, M.T. Espinosa-Garcia, C. Gomez-Concha, J. Monreal- Flores, F. Martinez, Steroidogenesis in BeWo cells: role of protein kinase A and ben- zodiazepines, Int. J. Biochem. Cell Biol. 40 (2008) 901–908. [20] A. Uribe, J.F. Strauss III, F. Martinez, Contact sites from human placental mitochon- dria: characterization and role in progesterone synthesis, Arch. Biochem. Biophys. 413 (2003) 172–181. [21] S. Olvera-Sanchez, M.T. Espinosa-Garcia, J. Monreal, O. Flores-Herrera, F. Martinez, Mitochondrial heat shock protein participates in placental steroidogenesis, Placenta 32 (2011) 222–229. [22] F. Arakane, T. Sugawara, H. Nishino, Z. Liu, J.A. Holt, D. Pain, D.M. Stocco, W.L. Miller, J.F. Strauss III, Steroidogenic acute regulatory protein (StAR) retains activity in the absence of its mitochondrial import sequence: implications for the mechanism of StAR action, Proc. Natl. Acad. Sci. U. S. A. 93 (1996) 13731–13736. [23] S.R. King, T. Ronen-Fuhrmann, R. Timberg, B.J. Clark, J. Orly, D.M. Stocco, Steroid pro- duction after in vitro transcription, translation, andmitochondrial processing of pro- tein products of complementary deoxyribonucleic acid for steroidogenic acute regulatory protein, Endocrinology 136 (1995) 5165–5176. [24] T. Ronen-Fuhrmann, R. Timberg, S.R. King, K.H. Hales, D.B. Hales, D.M. Stocco, J. Orly, Spatio-temporal expression patterns of steroidogenic acute regulatory protein (StAR) during follicular development in the rat ovary, Endocrinology 139 (1998) 303–315. [25] F. Martinez, M. Kiriakidou, J.F. Strauss III, Structural and functional changes in mito- chondria associated with trophoblast differentiation: methods to isolate enriched preparations of syncytiotrophoblast mitochondria, Endocrinology 138 (1997) 2172–2183. [26] A. Bensadoun, D. Weinstein, Assay of proteins in the presence of interfering mate- rials, Anal. Biochem. 70 (1976) 241–250. [27] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein measurement with the Folin phenol reagent, J. Biol. Chem. 193 (1951) 265–275. [28] O. Flores-Herrera, A. Uribe, C. Garcia-Perez, R. Milan, F. Martinez, 5′-p- Fluorosulfonylbenzoyl adenosine inhibits progesterone synthesis in human placen- tal mitochondria, Biochim. Biophys. Acta 1585 (2002) 11–18. [29] B. Chance, G.R. Williams, The respiratory chain and oxidative phosphorylation, Adv. Enzymol. 17 (1956) 65–134. [30] J.C. Fischer, W. Ruitenbeek, J.A. Berden, J.M.F. Trijbels, J.H. Veerkamp, A.M. Stadhouders, R.C.A. Sengers, J.M. Janssen, Differential investigation of the capacity of succinate oxidation in human skeletal muscle, Clin. Chim. Acta 153 (1985) 23–36. [31] K. Kita, C.R.T. Vibat, S. Meinhardt, J.R. Guest, R.B. Gennis, One-step purification from Escherichia coli of complex II (succinate:ubiquinoneoxidoreductase) associated with succinate-reducible cytochrome b556, J. Biol. Chem. 264 (1989) 2672–2677. [32] D. De Los Rios-Castillo, M. Zarco-Zavala, S. Olvera-Sanchez, J.P. Pardo, O. Juarez, F. Martinez, G. Mendoza-Hernandez, J.J. García-Trejo, O. Flores-Herrera, Atypical cris- tae morphology of human syncytiotrophoblast mitochondria role for complex V, J. Biol. Chem. 286 (2011) 23911–23919. [33] R.C. Tuckey, Cholesterol side-chain cleavage by mitochondria from the human pla- centa. Studies using hydroxycholesterols as substrates, J. Steroid Biochem. Mol. Biol. 42 (1992) 883–890. [34] U.K. Laemmli, Cleavage of structural proteins during the assembly of the heat of bac- teriophage T4, Nature 227 (1970) 680–685. [35] J. Huotari, A. Helenius, Endosome maturation, EMBO J. 30 (2011) 3481–3500. [36] K. Subramanian, W.E. Balch, NPC1/NPC2 function as a tag team duo tomobilize cho- lesterol, Proc. Natl. Acad. Sci. U. S. A. 105 (2008) 15223–15224. [37] J. Reitz, K. Gehrig-Burger, J.F. Strauss III, G. Gimp, Cholesterol interaction with the related steroidogenic acute regulatory lipid-transfer (START) domains of StAR (STARD1) and MLN64 (STARD3), FEBS J. 275 (2008) 1790–1802. [38] J.L. Thomas, R.P. Myers, R.C. Strickler, Human placental 3 beta-hydroxy-5-ene-ste- roid dehydrogenase and steroid 5Δ4-ene-isomerase: purification from mitochon- dria and kinetic profiles, biophysical characterization of the purified mitochondrial and microsomal enzymes, J. Steroid Biochem. 33 (1989) 209–217. [39] J. Navarrete, O. Flores-Herrera, A. Uribe, F. Martinez, Differences in cholesterol incor- poration into mitochondria from hepatoma AS-30D and human term placenta, Pla- centa 20 (1999) 285–291. [40] H.S. Bose, R.M. Whittal, M.C. Huang, M.A. Baldwin, W.L. Miller, N-218 MLN64, a pro- tein with StAR-like steroidogenic activity, is folded and cleaved similarly to StAR, Biochemistry 39 (2000) 11722–11731. [41] Z. Granot, R. Geiss-Friedlander, N. Melamed-Book, S. Eimerl, R. Timberg, A.M. Weiss, K.H. Hales, D.B. Hales, D.M. Stocco, J. Orly, Proteolysis of normal and mutated ste- roidogenic acute regulatory proteins in themitochondria: the fate of unwanted pro- teins, Mol. Endocrinol. 17 (2003) 2461–2476. [42] Z. Granot, E. Silverman, R. Friedlander, N. Melamed-Book, S. Eimerl, R. Timberg, K.H. Hales, D.B. Hales, D.M. Stocco, J. Orly, The life cycle of the steroidogenic acute regu- latory (StAR) protein: from transcription through proteolysis, Endocr. Res. 28 (2002) 375–386. [43] T. Minamikawa, D.A. Williams, D.N. Bowser, P. Nagley, Mitochondrial permeability transition and swelling can occur reversibly without inducing cell death in intact human cells, Exp. Cell Res. 246 (1999) 26–37. [44] H.S. Bose, V.R. Lingappa, W.L. Miller, Rapid regulation of steroidogenesis by mito- chondrial protein import, Nature 417 (2002) 87–91. [45] K. Leonhard, A. Stiegler, W. Neupert, T. Langer, Chaperone-like activity of the AAA domain of the yeast Yme1 AAA protease, Nature 398 (1999) 348–351. [46] R.N. Rainey, J.D. Glavin, H.W. Chen, S.W. French, M.A. Teitell, C.M. Koehler, A new function in translocation for the i-AAA protease Yme1: import of polynucleotide phosphorylase into the intermembrane space, Mol. Cell. Biol. 26 (2006) 8488–8497. [47] A. Chacinska, N. Pfanner, C. Meisinger, How mitochondria import hydrophilic and hydrophobic proteins, Trends Cell Biol. 12 (2002) 299–303. [48] T. Stan, J. Brix, J. Schneider-Mergener, N. Pfanner, W. Neupert, D. Rapaport, Mito- chondrial protein import: recognition of internal import signals of BCS1 by the TOM complex, Mol. Cel. Biol. 23 (2003) 2239–2250. [49] J. Hu, Z. Zhang,W.J. Shen, S. Azhar, Cellular cholesterol delivery, intracellular processing and utilization for biosynthesis of steroid hormones, Nutr. Metab. 7 (2010) 47–72. [50] P. Boström, L. Andersson, M. Rutberg, J. Perman, U. Lidberg, B.R. Johansson, J. Fernandez-Rodriguez, J. Ericson, T. Nilsson, J. Borén, S.O. Olofsson, SNARE proteins mediate fusion between cytosolic lipid droplets and are implicated in insulin sensi- tivity, Nat. Cell Biol. 9 (2007) 1286–1293. [51] R. Jahn, R.H. Scheller, SNAREs—engines for membrane fusion, Nat. Rev. Mol. Cell Biol. 7 (2006) 631–643. [52] S. Martens, H.T. McMahon, Mechanisms of membrane fusion: disparate players and common principals, Nat. Rev. Mol. Cell Biol. 9 (2009) 543–556. [53] T.C. Sühof, J.E. Rothman, Membrane fusion: grappling with SNARE and SM proteins, Science 323 (2009) 474–477. [54] M.S. Jägerström, S. Polesie, Y. Wickström, B.R. Johansson, H.D. Schroder, K. Højlund, P. Boström, Lipid droplets interact withmitochondria using SNAP23, Cell Biol. Int. 33 (2009) 934–940. [55] M. Steegmaier, V. Oorschot, J. Klumperman, R.H. Scheller, Syntaxin 17 is abundant in steroidogenic cells and implicated in smooth endoplasmic reticulum membrane dynamics, Mol. Biol. Cell 11 (2000) 2719–2731. [56] N.J. Grant, R. Hepp,W. Krause, D. Aunis, P. Oehme, K. Langley, Differential expression of SNAP-25 isoforms and SNAP-23 in the adrenal gland, J. Neurochem. 72 (1999) 363–372. [57] M. Jo, M.C. Gieske, C.E. Payne, S.E. Wheeler-Price, J.B. Gieske, I.V. Ignatius, T.E. Curry, C. Ko, Development and application of a rat ovarian gene expression database, Endo- crinology 145 (2004) 5384–5396. [58] J. Grosse, A. Bulling, C. Brucker, U. Berg, A. Amsterdam, A. Mayerhofer, M. Gratzl, Synaptosome-associated protein of 25 kilodaltons in oocytes and steroid- producing cells of rat and human ovary: molecular analysis and regulation by go- nadotropins, Biol. Reprod. 63 (2000) 643–650. [59] M. Shimada, Y. Yanai, T. Okazaki, Y. Yamashita, V. Sriraman, W.C. Wilson, J.C. Richards, Synaptosomal-associated protein 25 gene expression is hormonally regu- lated during ovulation and is involved in cytokine/chemokine exocytosis from gran- ulose cells, Mol. Endocrinol. 21 (2007) 2487–2502. 116 M. Esparza-Perusquía et al. / Biochimica et Biophysica Acta 1850 (2015) 107–117 [60] J. Sileikyte, E. Blachly-Dyson, R. Sewell, A. Carpi, R. Menabò, F. Di Lisa, F. Ricchelli, P. Bernardi, M. Forte, Regulation of the mitochondrial permeability transition pore by the outer membrane does not involve the peripheral benzodiazepine receptor (Translocator Protein of 18 kDa (TSPO)), J. Biol. Chem. 289 (2014) 13769–13781. [61] R.C. Tuckey, J. Sadleir, The concentration of adrenodoxin reductase limits cytochrome p450scc activity in the human placenta, Eur. J. Biochem. 263 (1999) 319–325. [62] R.C. Tuckey, A.J. McKinley, M.J. Headlam, Oxidized adrenodoxin acts as a competitive inhibitor of cytochrome P450scc in mitochondria from the human placenta, Eur. J. Biochem. 268 (2001) 2338–2343. [63] M. Zhang, P. Liu, N.K. Dwyer, L.K. Christenson, T. Fujimoto, F. Martinez, M. Comly, J.A. Hanover, E.J. Blanchette-Mackie, J.F. Strauss III, MLN64 mediates mobilization of ly- sosomal cholesterol to steroidogenic mitochondria, J. Biol. Chem. 277 (2002) 33300–33310. [64] N. Wang, S. Gottesman, M.C. Willingham, M.M. Gottesman, M.R. Maurizi, A human mitochondrial ATP-dependent protease that is highly homologous to bacterial LON protease, Proc. Natl. Acad. Sci. U. S. A. 90 (1993) 11247–11251. [65] O. Hori, F. Ichinoda, T. Tamatani, A. Yamaguchi, N. Sato, K. Ozawa, Y. Kitao, M. Miyazaki, H.P. Harding, D. Ron, M. Tohyama, D.M. Stern, S. Ogawa, Transmission of cell stress from endoplasmic reticulum to mitochondria: enhanced expression of Lon protease, J. Cell Biol. 157 (2002) 1151–1160. [66] P. Bross, T.J. Corydon, B.S. Andresen, M.M. Jorgensen, L. Bolund, N. Gregersen, Protein misfolding and degradation in genetic diseases, Hum. Mutat. 14 (1999) 186–198. [67] G. Casari, M. De Fusco, S. Ciarmatori, M. Zeviani, M. Mora, P. Fernandez, G. De Michele, A. Filla, S. Cocozza, R.Marconi, A. Durr, B. Fontaine, A. Ballabio, Spastic para- plegia and OXPHOS impairment caused by mutations in paraplegin, a nuclear- encoded mitochondrial metalloprotease, Cell 93 (1998) 973–983. [68] L. Issop, M.B. Rone, V. Papadopoulos, Organelle plasticity and interactions in choles- terol transport and steroid biosynthesis, Mol. Cell. Endocrinol. 371 (2013) 34–46. [69] T. Kishida, I. Kostetskii, Z. Zhang, F. Martinez, P. Liu, S.U. Walkley, N.K. Dwyer, E.J. Blanchette-Mackie, G.L. Radice, J.F. 3er Strauss, Targeted mutation of the MLN64 START domain causes only modest alterations in cellular sterol metabolism, J. Biol. Chem. 279 (2004) (2004) 19276–19285. 117M. Esparza-Perusquía et al. / Biochimica et Biophysica Acta 1850 (2015) 107–117 Membrane potential regulates mitochondrial ATP-diphosphohydrolase activity but is not involved in progesterone biosynthesis in human syncytiotrophoblast cells Oscar Flores-Herrera ⁎, SofiaOlvera-Sánchez,Mercedes Esparza-Perusquía, Juan Pablo Pardo, Juan Luis Rendón, Guillermo Mendoza-Hernández, Federico Martínez Universidad Nacional Autónoma de México, Facultad de Medicina, Departamento de Bioquímica y Biología Molecular, México City, Mexico a b s t r a c ta r t i c l e i n f o Article history: Received 16 May 2014 Received in revised form 17 September 2014 Accepted 7 October 2014 Available online 14 October 2014 Keywords: ATP-diphosphohydrolase Placental mitochondria Mitochondrial bioenergetics ATP hydrolysis Progesterone synthesis ATP-diphosphohydrolase is associated with human syncytiotrophoblast mitochondria. The activity of this enzyme is implicated in the stimulation of oxygen uptake and progesterone synthesis. We reported previously that: (1) the detergent-solubilized ATP-diphosphohydrolase has low substrate specificity, and (2) purine and pyrimidine nucleosides, tri- or diphosphates, are fully dephosphorylated in the presence of calcium or magne- sium (Flores-Herrera 1999, 2002). In this study we show that ATP-diphosphohydrolase hydrolyzes first the nu- cleoside triphosphate to nucleoside diphosphate, and then to nucleotidemonophosphate, in the case of all tested nucleotides. The activation energies (Ea) for ATP, GTP, UTP, and CTP were 6.06, 4.10, 6.25, and 5.26 kcal/mol, respectively; for ADP, GDP, UDP, and CDP, they were 4.67, 5.42, 5.43, and 6.22 kcal/mol, respectively. The corre- sponding Arrhenius plots indicated a single rate-limiting step for each hydrolyzed nucleoside, either tri- or diphosphate. In intact mitochondria, the ADP produced by ATP-diphosphohydrolase activity depolarized the membrane potential (ΔΨm) and stimulated oxygen uptake. Mitochondrial respiration showed the state-3/ state-4 transition when ATP was added, suggesting that ATP-diphosphohydrolase and the F1F0-ATP synthase work in conjunction to avoid a futile cycle. Substrate selectivity of the ATP-diphosphohydrolase was modified by ΔΨm (i.e. ATP was preferred over GTP when the inner mitochondrial membrane was energized). In contrast, dissipation ofΔΨm by CCCP produced a loss of substrate specificity and so the ATP-diphosphohydrolase was able to hydrolyze ATP and GTP at the same rate. In intact mitochondria, ATP hydrolysis increased progesterone synthesis as compared with GTP. Although dissipation of ΔΨm by CCCP decreased progesterone synthesis, NADPH production restores steroidogenesis. Overall, our results suggest a novel physiological role for ΔΨm in steroidogenesis. © 2014 Elsevier B.V. All rights reserved. 1. Introduction One of the main functions of the placenta is the synthesis of proges- terone (P4) to maintain pregnancy. Mitochondria from human syncytiotrophoblast cells contain the machinery for steroid synthesis. It consists of an electron transport chain (ETC-P450scc) composed by the cytochrome P450scc (CYP11A1; EC 1.14.15.6) that receives elec- trons from NADPH + H+ through two proteins: adrenodoxine and adrenodoxine reductase. These proteins are located in the inner mito- chondrial membrane and transform cholesterol into pregnenolone (P5) [3–5]. An additional enzyme, type II 3β-hydroxysteroid- dehydrogenase-Δ4-5 isomerase (3βHSD) also embedded in the inner mitochondrial membrane of syncytiotrophoblast cells, transforms pregnenolone into progesterone [4,5]. Mitochondria are best known as the major source of ATP in aerobic cells. Oxidative phosphorylation provides the main source of ATP. This metabolic pathway relies on the activity of two components: the oxida- tive and the phosphorylation systems. The oxidative system (i.e. respi- ratory chain) couples redox reactions to the production of a proton electrochemical gradient that drives the synthesis of ATP by the phosphorylation system (i.e. F0F1-ATP synthase and the ADP/ATP and phosphate carriers). Studies conducted in primary and MA-10 tumor Leydig cells suggest an interrelation between steroidogenesis and oxi- dative phosphorylation. Steroidogenic mitochondria perform a double role: synthesize ATP and produce hormones. Steroidogenesis is affected when the classic mitochondrial electron-transport chain (ETC), mem- brane potential (ΔΨm), or ATP synthesis is disrupted [6,7], suggesting a close relationship between both metabolic pathways. The relative amount of ETC-P450scc and classic ETC components present in cells depends on the type of steroidogenic tissue. In acute- Biochimica et Biophysica Acta 1847 (2015) 143–152 ⁎ Corresponding author at: Departamento de Bioquímica, Facultad de Medicina, Universidad Nacional Autónoma de México, Apdo. Postal 70-159, Coyoacán 04510, México, D. F., Mexico. Tel.: +52 1 55 56232510; fax: +52 1 55 56162419. E-mail address: oflores@bq.unam.mx (O. Flores-Herrera). http://dx.doi.org/10.1016/j.bbabio.2014.10.002 0005-2728/© 2014 Elsevier B.V. All rights reserved. Contents lists available at ScienceDirect Biochimica et Biophysica Acta j ourna l homepage: www.e lsev ie r .com/ locate /bbab io regulated steroidogenic tissues (i.e. the adrenal gland and gonads) the content of ETC-P450scc is several times higher than in the classic ETC. In the syncytiotrophoblast, a constitutive steroidogenic tissue, the amounts of both electron transfer chain components is similar [8]. This suggests that the activity of both pathways will generate enough ATP for the cell to function properly and enough P4 to maintain pregnancy. Both processes are equally important for the physiological role of human placenta, and their activity must be tightly regulated. Human syncytiotrophoblast mitochondria contain accessory enzymes, like the ATP-diphosphohydrolase. This enzyme is tightly bound to mitochondrial membranes and is involved in progesterone synthesis, mainly in cholesterol transport [1]. Studies have related cholesterol transport across the membranes of syncytiotrophoblast mi- tochondria with the activity of mitochondrial ATP-diphosphohydrolase [1,2]. Probably, ATP-diphosphohydrolase provides the required energy to drive cholesterol transport between mitochondrial membranes in an analogous way to that of the mitochondrial GDPase in the adrenal gland [9]. The underlying molecular mechanism involved remains unknown. Simultaneously, in the intact and energized mitochondria, the ATP-diphosphohydrolase hydrolyzes ATP to ADP. The latter promotes oxygen uptake and ATP synthesis by the F1F0-ATP synthase [10]. The activities of the ATP-diphosphohydrolase (ATP hydrolysis) and the F1F0-ATP synthase (ATP synthesis)must be coordinated to avoid a futile cycle and energy dissipation. Nevertheless, the physiological role of the ATP-diphosphohydrolase must be examined in intact syncytiotrophoblast mitochondria. Since ATP-diphosphohydrolase activity is involved in progesterone synthesis [2] and mitochondrial bioenergetics [10], regulatory mechanisms must be involved to keep trophoblast cells alive and functional. In the present work we evaluated the relationship between ΔΨm, ATP-diphosphohydrolase activity, and progesterone synthesis in syncytiotrophoblast mitochondria. Results suggest that ATP- diphosphohydrolase activity is modified by ΔΨm, but an increase in NADPH content and ATP hydrolysis supports progesterone synthesis when ΔΨm decreases. This study puts forward a novel physiological role for the ΔΨm in human placenta steroidogenesis. 2. Experimental procedures 2.1. Isolation of human syncytiotrophoblast mitochondria Full term human placenta was collected immediately after normal delivery at the IMSS Hospital No. 4, approval under the Ethical Commit- tee regulations. Mitochondria were prepared as previously described [3]. Briefly, placental cotyledons were placed in ice-cold 250 mM sucrose and 1 mM EDTA, 10 mM Tris, pH 7.4. The suspension was homogenized by means of a Polytron (Brinkmann Instruments, Westbury, NY, USA), at 3000 rev/min for 1min for two cycles separated by an interval of 1 min. The whole process was carried out at 4 °C. The pH of the homogenate was adjusted to pH 7.4 with Tris and centrifuged at 1500 g for 15min. The supernatant was recovered and centrifuged at 4000 g to pellet the cytotrophoblast mitochondria (i.e. heavymitochon- dria). The supernatantwas centrifuged again at 16,000 g for 15min and the pellet containing the syncytiotrophoblast mitochondria (i.e. light mitochondria) was resuspended in the same solution and then centri- fuged at 1500 g for 10 min to remove any remaining erythrocytes. Mitochondria were pelleted by centrifugation at 12,000 g for 10 min. The resulting syncytiotrophoblast mitochondria were loaded on a 35% sucrose solution (25 ml) and centrifuged at 15,000 g for 45 min at 4 °C. The mitochondrial fraction was collected, suspended in 250 mM of sucrose, 1 mM of EDTA, and 10 mM of Tris (pH 7.4) and centrifuged at 16,000 g for 15 min at 4 °C; the mitochondrial pellet was suspended in this buffer and stored at 4 °C. Protein concentrationwasmeasured as reported by [11,12]. 2.2. Mitochondrial oxygen consumption Oxygen uptake was estimated polarographically using a Clark type electrode in a mixture containing 250 mM of sucrose, 10 mM of HEPES pH 7.4, 1 mM of EGTA, 1 mM of EDTA, 10 mM of succinate, 10 mM of KH2PO4, 5 mM of MgCl2, 0.2% bovine serum albumin and 1 mg/ml of syncytiotrophoblast mitochondrial protein [2]. Tempera- ture was set at 37 °C and oxygen consumption was stimulated by the addition of 300–500 nmol of ATP or ADP (state 3 of respiration). Respiratory controlwas defined as oxygen uptake rate of state 3/oxygen uptake rate of state 4 (state 4 of respiration started when all ADPwas converted to ATP and respiration slowed down) [13]. Where indicat- ed, mitochondria were incubated with 5 μM carboxyatractyloside (CAT) to inhibit the translocation of adenine nucleotides by blocking the ADP/ATP carrier. Simultaneously, 10 μM of carbonyl cyanide m- chlorophenyl hydrazine (CCCP) was added to depolarize the inner membrane and stimulate maximal oxygen uptake (vide infra). At the indicated times (see Figs. 3–5) an aliquot was withdrawn and used to determine the nucleotide concentration by HPLC (vide infra). 2.3. Mitochondrial membrane potential (ΔΨm) The following media was used to determine the ΔΨm of syncytiotrophoblast mitochondria: 125 mM of KCl; 5 mM of MgCl2; 10 mMof acetate–Tris, pH 7.4; 10mMof Tris–HCl, pH 7.4; 1 μMof rote- none; 3.3 mM of H3PO4, pH 7.4; 9.6 mM of Safranine O, and 1 mg of mitochondrial protein/ml. Generation of the membrane potential was initiated by adding 10 mM of succinate–Tris, pH 7.4 [14] to the solution containing syncytiotrophoblast mitochondria. Where indicated, mito- chondria were incubated with 5 μM CAT to inhibit the ADP/ATP carrier, and 10 μMof CCCPwas added to abolishΔΨm. Themembrane potential was evaluated in a double beam spectrophotometer by using the differ- ence of wavelengths between 533 and 511 nm. The final volume was 1.5 ml and was kept at 25 °C. 2.4. Activity determinations of mitochondrial enzymes Activities of complex I (NADH:DCPIP oxidoreductase) and complex II (succinate:DCPIP oxidoreductase) were determined spectrophoto- metrically at 600 nmby following the reduction of the artificial electron acceptor 2,6-dichlorophenol-indophenol (DCPIP; 50 μM; εDCPIP = 21 mM−1 cm−1). Mitochondria were permeabilized with 0.01% Triton X100, incubated in 120 mM of KCl, 5 mM of MgCl2, 1 mM of EGTA, 30 mM of KH2PO4, pH 7.4, and either 500 μM of NADH (complex I) or 2 mM of succinate (complex II). Complex II was activated by pre- incubation in the presence of 0.2 mM of phenazine methosulfonate (PMS) for 10 min at 25 °C [15,16]. Protein concentration of syncytiotrophoblast mitochondria was 50 μg/ml and the reaction was started by the addition of NADH or succinate. Mitochondrial ATP-diphosphohydrolase activity was determined either by nucleotide separation by HPLC (vide infra) or by measuring the release of inorganic phosphate (Pi) as described by Flores-Herrera et al. [1], using an ATP-diphosphohydrolase enriched fraction [1], or isolated syncytiotrophoblast mitochondria. Briefly, proteins (50 μg) were incubated in a final volume of 0.5 ml at 30 °C in 30 mM Tris–HCl (pH 8.5), and the ATP-diphosphohydrolase reaction was started by the addition of the substrate (Mg-nucleotide complex) plus 1 mM of free Mg2+. Aliquots were withdrawn at one minute intervals and used for nucleotide separation by HPLC (vide infra), or mixed with the malachite–molybdate–Triton X-100 mixture for Pi determination, as described by Lanzetta et al. [17]. The nucleotides used were ATP, ADP, GTP, GDP, UTP, UDP, CTP, CDP or TTP. The experiments were performed at least four times in duplicate. 144 O. Flores-Herrera et al. / Biochimica et Biophysica Acta 1847 (2015) 143–152 2.5. Nucleotide separation by HPLC Quantification of nucleotides was performed from either mitochon- drial oxygen consumption experiments or a fraction enriched with a detergent-solubilized ATP-diphosphohydrolase. At the indicated times aliquots from ATP-diphosphohydrolase nucleotide hydrolysis activity or oxygen uptake determination were withdrawn and mixed with trichloroacetic acid (6% final) to stop the reaction. Nucleotides were separated by anion exchange HPLC on a Hypersil SAX column (120 Å, 5 μm, 250 × 4.6 mm) from Alltech International. The low concentration buffer (A) was 5 mM NH4H2PO4 (pH 2.8) and the high concentration buffer (B) was 750 mM NH4H2PO4 (pH 3.7). The sample was loaded on the column equilibrated with buffer (A). Then, a gradient of buffer (B) (30 min, 0–100%) was used for elution. The flow rate was 1 ml/min and detection was performed at 254 nm [18]. 2.6. Mitochondrial progesterone synthesis Progesterone synthesis was determined at 37 °C as reported previ- ously [2] in 120 mM of KCl, 10 mM of MOPS, 0.5 mM of EGTA, 10 mM of isocitrate, 4 μg of aprotinin/ml, 1 μM of leupeptin, and 5 mM of KH2PO4, pH 7.4, in a final volume of 500 μl with 1 mg/ml of syncytiotrophoblast mitochondrial protein. Where indicated, 25 μM 22-(R)-hydroxy-cholesterol was added to verify cytochrome P450scc, adrenedoxin, adrenedoxin reductase, and 3β-hydroxysteroid dehydro- genase activities [19]. After 20 min of incubation the reaction was arrested with 75 μl methanol and progesterone concentration was determined using a radioimmunoassay kit (Diagnostic Systems Labora- tories, Inc. Webster, Texas, USA), according to the manufacturer's instructions. The concentration of progesterone at time zero was subtracted from the amount of progesterone quantified at 20 min and the resulting net progesterone synthesis was reported. 2.7. Sample preparation for native electrophoresis The ATP-diphosphohydrolase from syncytiotrophoblast mitochon- dria was resolved by native PAGE following the general procedures reported previously [20,21], with minor modifications [22]. Briefly, syncytiotrophoblast (2 mg) mitochondria were suspended in 50 mM Bis-Tris and 500 mM 6-aminocaproic acid, pH 7.0, and solubilized by adding digitonin, at a detergent/protein ratio of 2 (g/g) in a final volume of 200 μl. The mixture was incubated for 30 min at 4 °C and centrifuged at 100,000 g for 30 min at 4 °C. The supernatants were recovered and immediately loaded on a linear polyacrylamide gradient gel (5–10%) for Blue Native PAGE (BN-PAGE) or Clear Native PAGE (CN-PAGE) [21]. The molecular weight of ATP-diphosphohydrolase activity was estimated by using the digitonin-solubilized bovine mitochondrial complexes as standard. 2.8. In-gel catalytic activity assays The in-gel activity assayswere performed as described by [18]. Brief- ly, gel strips were preincubated in 30 mM Tris–HCl, pH 8.5, 5% glycerol, 15mMCaCl2 for 30 min at 37 °C in the presence or absence of the com- plex V inhibitor oligomycin (5 μg/ml), or 5′-p-fluorosulfonylbenzoyl adenosine (FSBA, 1 mM) the ATP-diphosphohydrolase inhibitor. The equilibration solution was discarded and the gel strips were then added to the assay buffer containing 30 mM of Tris–HCl, pH 8.5, 5% glycerol, 15 mM of CaCl2 and 5 mM of ATP, ADP, GTP or GDP, with or without oligomycin (5 μg/ml) or FSBA (1 mM). After incubation at 37 °C for approximately 2 h, nucleotide hydrolysis correlated with the development of white calcium phosphate precipitates. The reaction was stopped using 50% methanol, and subsequently, the gel was transferred to water and scanned against a dark background as described previously [22]. 2.9. Isolation of steroidogenic contact sites Steroidogenic contact sites were isolated as reported by Uribe et al. [23]. Briefly, 20–25 mg of mitochondrial protein was incubated at 4 °C with 10 mM H3PO4, adjusted to pH 7.3 with Tris base in the presence of 10 μg aprotinin/ml, 1 mM phenylmethylsulfonyl fluoride, and 10 μg leupeptin/ml. After incubation, sucrosewas added to attain a concentra- tion of 0.38 M. The resulting mixture was incubated for another 20 min at 4 °C and then centrifuged at 12,500 g for 10 min. The pellet contain- ingmitoplastswas recovered and incubated in 1mMH3PO4, adjusted to pH 7.3 with Tris base, for 20 min at 4 °C. Sucrose was added to reach a concentration of 0.31 M and the mixture was incubated for another 20min at 4 °C and centrifuged at 102,000 g for 1 h. The pellet containing the inner membrane fractionwas sonicated four times in an ice bath for five seconds in aMSE Soniprep 150 atmaximal output. The fraction con- taining the inner membranes was layered over a discontinuous sucrose gradient (densities of 1.06 to 1.29 g/ml) and centrifuged at 96,000 g for 20 h at 4 °C. The steroidogenic contact sites were recovered at sucrose densities of 1.20–1.22 g/ml and washed three times with 0.25 M su- crose, 1mMEDTA,with pHadjustedwith Tris base to 7.4, and recovered by centrifugation at 137,000 g for 30 min at 4 °C. Protein content was determined as described above and the obtained samples were stored at−70 °C. 2.10. Tandem mass spectrometry (LC/ESI–MS/MS) Themitochondrial inner membrane obtained during the isolation of steroidogenic contact sites (see previous section) was incubated in 100mMammoniumbicarbonate (pH=7.8) for 30min and centrifuged at 100 000 ×g at 4 °C. The pellet containing the mitochondrial inner membranes was sent to the Proteomics Core Facility at the University of Arizona, USA. 2.11. Statistical analyses Statistical analyses (one- and two-way analysis of variance, ANOVA) of the data were performed using Sigma Stat software, version 3.5. When necessary, nonlinear regression of the data to a single exponen- tial decay equation was performed in Sigma Plot software, version 10.0. 2.12. Materials Analytical grade reagents were purchased from Sigma Chemical Co. (St. Louis, MO, USA), E. Merck (Darmstadt, Germany), and BioRad (Hercules, CA, USA). 3. Results 3.1. Functional state of syncytiotrophoblast mitochondria We calculated respiratory controls from oxygen uptake traces, using succinate as a substrate, to determine the functional integrity of isolated syncytiotrophoblast mitochondria (Table 1). Oxygen uptakes in state 3 and state 4 were 110 ± 18 ng atom of oxygen/min·mg protein, and 19 ± 6 ng atom of oxygen/min·mg protein, respectively. The value of the respiratory control was 5.5 ± 1.2. Adding CCCP to energized mitochondria increased the permeability of the membrane to protons and induced maximum respiration rate (200 ± 35 n atom g of oxygen/min·mg protein) and dissipation of ΔΨm (Fig. 4 later in the paper), inhibiting oligomycin-sensitive-ATP synthesis. These data indicated functional coupling of respiration and ATP synthesis in syncytiotrophoblast mitochondria. In addition, activities of 110 ± 27 μmol/min·mg protein for the NADH:DCPIP oxidoreductase (complex I), and 7 ± 1.5 μmol/min·mg protein for the succinate: DCPIP oxidereductase (complex II) were obtained (Table 1). These results indicated the presence of functional mitochondria that retained 145O. Flores-Herrera et al. / Biochimica et Biophysica Acta 1847 (2015) 143–152 the ability to increase the consumption of oxygen and the synthesis of ATP upon the addition of ADP. Human placental mitochondria are steroidogenic organelles that synthesize progesterone, due to the presence of the type II 3-β- hydroxy steroid dehydrogenase in their inner membrane [3–5]. We determined steroidogenic activity of syncytiotrophoblast mitochondria to verify their physiological function. As observed in Table 1 synthesis of progesterone by syncytiotrophoblast mitochondria was 143 ± 12ng progesterone/min·mg protein (Fig. 6 later in the paper), reaching a maximum of 606 ± 52 ng progesterone/min·mg protein in the pres- ence of 22-(R)-hydroxy-cholesterol, which is a soluble substrate used to verify cytochrome P450scc, adrenedoxin, adrenedoxin reductase and 3β-hydroxysteroid dehydrogenase activities [19]. These results agree with the specialized role of syncytiotrophoblast tissue [3] and support the functional integrity of the isolated syncytiotrophoblast mitochon- dria used in this work. 3.2. Nucleotide hydrolysis by mitochondrial ATP-diphosphohydrolase We designed the experiments in the present work to investigate the possible involvement of ΔΨm in mitochondrial ATP- diphosphohydrolase activity and progesterone synthesis in human syncytiotrophoblast cells. We first obtained a detergent-solubilized ATP-diphosphohydrolase fraction to determine their nucleotide hydro- lysis activity [1].Wemonitored the time course of nucleotide hydrolysis by HPLC (Figs. 1 and 1S). Hydrolysis of nucleoside diphosphates (NDP) by mitochondrial ATP-diphosphohydrolase was associated with the accumulation of the corresponding nucleoside monophosphate (NMP), whether a purine or pyrimidine nucleotide was involved (Figs. 1A and 2S). A transient accumulation of NDP was observed when a nucleoside triphosphate (NTP) was hydrolyzed. NDP were dephosphorylated to NMP (Figs. 1B and 2S). Since this hydrolyzing activity is exerted by the mitochondrial ATP-diphosphohydrolase, it can be inhibited by 5′-p-fluorosulfonyl benzoyl adenosine [1]. The kinetics of ATP-diphosphohydrolase was similar regardless of the sub- strate of choice (Figs. 1 and 2S), confirming its low substrate specificity [1]. Additionally, we determined the activation energy (Ea) for the solubilized ATP-diphosphohydrolase activity bymeasuring the reaction rate constant (Vmax) at different temperatures and by plotting ln(Vmax) versus 1/T (Fig. 3S). Data were adjusted to the integrated form of the Arrhenius equation: Ea=((RT2T1) / (T2− T1))·ln(Vmax). The Arrhenius plot was linear in the temperature range spanning 10–55 °C (Fig. 3S), suggesting a single rate-limiting step. A sudden drop in the Arrhenius plot at low1/T (high temperature, 60–70 °C) indicated protein denatur- ation (Fig. 3S). The Ea values for nucleoside triphosphates such as ATP, GTP, UTP, and CTPwere 6.06, 4.10, 6.25, and 5.26 kcal/mol, respectively. For ADP, GDP, UDP, and CDP the Ea values were 4.67, 5.42, 5.43, and 6.22 kcal/mol, respectively. These results suggest that solubilized ATP-diphosphohydrolase had a similar rate-limiting step for either tri- or diphosphates nucleoside hydrolysis. 3.3. BN-PAGE analysis To support the hypothesis that a single enzyme hydrolyzes ATP, ADP, GTP, or GDP, we conducted blue native PAGE of syncytiotrophoblast mitochondria (Fig. 2). Since the calcium–nucleotide complex can be used as a substrate by the mitochondrial ATP- diphosphohydrolase, but not by the F1F0-ATP synthase, the in-gel activ- ity was determined in the presence of CaCl2 with ATP, ADP, GTP or GDP as substrate, in the absence or presence of oligomycin (not shown). Phosphohydrolytic activity produced a single band of calcium phos- phate precipitate (Fig. 2), which displayed no oligomycin inhibition but identical electrophoretic mobility with an apparent molecular weight of 167 kDa. FSBA inhibited ATP-diphosphohydrolase activity with any of the tested substrates (Fig. 2). Results indicated that solubi- lized ATP-diphosphohydrolase from syncytiotrophoblast mitochondria was the only enzyme capable of hydrolyzing calcium–nucleotide complexes. These results confirmed the wide spectrum of substrates Table 1 Bioenergetics and steroidogenic parameters of syncytiotrophoblast mitochondria. Oxygen uptake State 3a 110 ± 18 ng atom of oxygen/min·mg protein State 4b 19 ± 6 ng atom of oxygen/min·mg protein Respiratory controlc 5.5 ± 1.2 Complexes activitiesd Complex I 110 ± 27 μmol/min·mg Complex II 7 ± 1.5 μmol/min·mg Progesterone synthesise Control 143 ± 1.5 ng progesterone/min·mg +22(R)-hydroxy-cholesterol 606 ± 52 ng progesterone/min·mg a Defined as oxygen consumption stimulated by ADP added in presence of succinate as substrate. Values are the mean ± S.D. (n = 20 independent determinations from different placental tissue). b Defined as oxygen consumption reduction because all ADP addedwas converted to ATP. Values are the mean ± S.D. (n = 20 independent determinations from different placental tissue). c Respiratory control = oxygen uptake rate of state 3/oxygen uptake rate of state 4. Values are themean ± S.D. (n = 20 independent determinations fromdifferent placental tissue). d Specific activities from complexes I and II were measured spectrophotometrically in sonicated mitochondria: complex I, NADH:DCPIP oxide reductase; and complex II, succinate:DCPIP oxide reductase. Complex II activity was stimulated as described in the Experimental procedures section. Values shows are themean ± S.D. (n = 9 independent determinations from different placental tissue). e Progesterone synthesis was determined as described in the Experimental procedures section. The 22(R)-hydroxy-cholesterol was used to verify cytochrome P450scc, adrenedoxin, adrenedoxin reductase and 3β-hydroxysteroid dehydrogenase activities [19]. Values here are the mean ± S.D. from four determinations, from four different placental tissues. Fig. 1. Nucleotide hydrolysis of isolated syncytiotrophoblast mitochondrial ATP- diphosphohydrolase. ATP-diphosphohydrolase from syncytiotrophoblast mitochondria was isolated and its hydrolytic nucleotide activity was determined as described in the Experimental procedures section. Nucleotide concentration was quantified at different times by HPLC. A) Hydrolysis of Mg-ADP (3 mM) or Mg-GDP (1 mM) by mitochondrial ATP-diphosphohydrolase: (○) = nucleoside diphosphate (NDP); (▲) = nucleoside monophosphate (NMP); (∆) = nucleoside. B) Hydrolysis of Mg-ATP (3 mM) or Mg-GTP (3 mM) by mitochondrial ATP-diphosphohydrolase: (•) = nucleoside triphosphate (NTP); (○) = NDP; (▲) = NMP; (∆) = nucleoside. The figure shows representative experiments. 146 O. Flores-Herrera et al. / Biochimica et Biophysica Acta 1847 (2015) 143–152 that ATP-diphosphohydrolase exhibits. In the second approach, we analyzed the ATP-diphosphohydrolase activity associated to intact mitochondria undergoing oxygen uptake or progesterone synthesis. The tandem mass spectrometry (LC/ESI–MS/MS) analysis showed the isoform 2 of the ATP-diphosphoydrolase (P49961-2|ENTP1_HUMAN), which belong to the ectonucleoside triphosphate diphosphohydrolase group, with a molecular weight of 59 kDa. The hydropathy analysis displayed four transmembrane segments (see Fig. 4S). However, no mitochondrial targeting presequence was observed. 3.4. ATP hydrolysis by ATP-diphosphohydrolase stimulates mitochondrial respiration and depolarizes the inner membrane To evaluate the ATP-diphosphohydrolase activity during mitochon- drial oxygen uptake, the time course of the hydrolysis of nucleotides was analyzed by HPLC (Fig. 3). Syncytiotrophoblast mitochondria were energized by succinate, and oxygen uptake was stimulated by ATP (Fig. 3A). Results show that ATP-diphosphohydrolase hydrolyzed ATP and produced ADP (Fig. 3B), which in turn was translocated into the mitochondrial matrix, where it was transformed to ATP by the F1F0-ATP synthase at the expense of the proton electrochemical gradi- ent (ΔμH+). Simultaneously, the inner membrane was depolarized (Fig. 3C). This series of events is defined as state 3 of mitochondrial respiration [13], and continues until the ATP-diphosphohydrolase hydrolyses ATP to ADP and ADP to AMP. In this situationΔΨm increased (Fig. 3C) and mitochondrial respiration decreased to a minimum, a condition that is known as state 4 of mitochondrial respiration [13] (Fig. 3). A new cycle of mitochondrial oxygen uptake stimulation, depolarization of ΔΨm, ATP synthesis and ATP-diphosphohydrolase activitywas observedwhen ADPwas added (Fig. 3). It is crucial to high- light that ATP-diphosphohydrolase activity is closely related to an increase in mitochondrial respiration and inner membrane depolariza- tion in thepresence of ATP. Additionally, syncytiotrophoblastmitochon- dria contain a phosphatase [1] that is responsible for adenosine production from AMP. However, inhibiting it with phenylalanine or sodium molibdate [1] did not modify the results described (data no shown). 3.5. Substrate selectivity and catalytic rate by ATP-diphosphohydrolase during mitochondrial respiration We added a different nucleoside triphosphate to support the notion that the ATP-diphosphohydrolase activity could be modified by mitochondrial bioenergetics during oxygen uptake (Fig. 4). After the transition from state 3 to 4 of mitochondrial respiration, addition of GTP did not stimulate oxygen uptake (Fig. 4A), nor depolarized the inner membrane (Fig. 4B). Although GTP was hydrolyzed by ATP- Fig. 2. In-gel activity of digitonin-solubilized mitochondrial ATP-diphosphohydrolase from syncytiotrophoblast in native gels. Mitochondria were solubilized using digitonin (2 g/g of protein), and ATP-diphosphohydrolase was separated by BN-PAGE. Native-PAGE was performed in linear polyacrylamide gradient gels from 5 to 10% as described in the Experimental procedures section. Electrophoresis was conducted at 30 V for 12 h at 4 °C. In the spots containing ATP-diphosphohydrolase activity, white precipitates of calcium phosphate appeared within 2 h after the addition of Ca–ATP, Ca–ADP, Ca–GTP or Ca–GDP complex. The presence of 1 mM of FSBA inhibited ATP hydrolysis while oligomycin (5 μg/mg) did not modify the hydrolytic activity (data not shown). The molecular weight of ATP-diphosphohydrolase activity was estimated by using the digitonin-solubilized bovine mitochondrial complexes as standard. The figure shows representative experiments from four different mitochondrial preparations. Fig. 3.Mitochondrial ATP-diphosphohydrolase induces mitochondrial oxygen consump- tion and ΔΨm depolarization. Syncytiotrophoblast mitochondria were isolated as described in the Experimental procedures section. A) Mitochondria were incubated in oxygen uptake medium at 37 °C and mitochondrial respiration was stimulated by ATP or ADP addition. Arrows indicate sequential additions of (M) mitochondria (1 mg/ml); ATP (130 μM); or ADP (130 μM). Simultaneous tomitochondrial oxygen uptake recording, an aliquot was withdrawn at the indicated time (bold arrows) and used to quantify nucleotide concentration by HPLC (B). Concentrations of ATP, ADP, AMP or Adenosine (Ade) at different times after addition of ATP or ADP during the time course of oxygen uptake showed in (A). C) ΔΨm measurement with Safranine O as described in the Experimental procedures section. Arrows indicate sequential additions of mitochondria (M); 10 mM succinate (Succ); ATP (130 μM); ADP (130 μM); CCCP (10 μM). In all cases curves show representative experiments of at least four different mitochondrial preparations. 147O. Flores-Herrera et al. / Biochimica et Biophysica Acta 1847 (2015) 143–152 diphosphohydrolase (855 ± 137 μmol/mg·min), its catalytic rate was lower than that observed with ATP (1910 ± 265 μmol/mg·min) (Figs. 4C and 6B). Indeed, addition of ADP, in the presence of GDP produced from GTP hydrolysis, induced oxygen consumption and decreased ΔΨm, while the rate of GDP hydrolysis was very small (Fig. 4C). This result contrasts with the one of the detergent- solubilized ATP-diphosphohydrolase, which displays a similar hydroly- sis rate for ATP, ADP, GTP, and GDP (Fig. 1). To examine substrate selectivity of the mitochondrial ATP- diphosphohydrolase, we added simultaneously ATP and GTP (Fig. 4D). The enzyme consistently hydrolyzed ATP instead of GTP (Fig. 4F), and the produced ADP induced oxygen consumption and membrane depolarization (Fig. 4D and E, respectively). 3.6. ΔΨm determines ATP-diphosphohydrolase substrate selectivity and catalytic rate ΔΨm is a central component of mitochondrial metabolism that provides the driving force for oxidative phosphorylation, for the import of proteins andmetabolites, and for regulating the activity ofmembrane proteins like the adenine nucleotide translocase (ANT) [24–26]. We examined the effect of the mitochondrial protonophore and respiration uncoupler CCCP on ΔΨm to assess whether the ATP- diphosphohydrolase substrate selectivity was mediated by ΔΨm (Fig. 5). Syncytiotrophoblast mitochondria were incubated with CAT to inhibit the adenine nucleotide translocase and avoid ATP internaliza- tion into the mitochondrial matrix in the presence of CCCP (see the Experimental procedures section).When CCCPwas added tomitochon- dria, oxygen uptake was stimulated and ΔΨm collapsed (Fig. 5A and B, respectively). Further addition of ATP did not modify oxygen consump- tion norΔΨm. Importantly, ATP-diphosphohydrolase activity was lower (812± 30 μmol/mg·min)when compared to control conditions, (i.e. in the absence of CCCP). Also, a transient accumulation of ADP was observed (Figs. 5C and 6B). In identical experimental conditions, GTP addition to CCCP- uncoupled mitochondria rendered a lower rate of GTP hydrolysis (642 ± 37 μmol/mg·min), and a transient accumulation of GDP (Figs. 5D–F and 6B), similar to the results obtained with ATP. To compare ATP and GTP hydrolysis in CCCP-uncoupled mitochondria, both nucleotides were added at the same time (Fig. 5G–I). The ATP-diphosphohydrolase catalyzed simultaneously the hydrolysis of both nucleotides and displayed similar velocities (Fig. 5I), without any Fig. 4.Mitochondrial ATP-diphosphohydrolase selectively hydrolyzesATP in energized and coupledmitochondria. A)Mitochondriawere incubated in oxygen uptakemediumat 37 °C and mitochondrial respiration was stimulated by ATP or ADP addition. The arrows indicate sequential additions of (M) mitochondria (1 mg/ml); ATP (130 μM); GTP (130 μM); or ADP (130 μM). Duringmitochondrial oxygen uptake an aliquot was withdrawn at the indicated time (bold arrows) and used to quantify nucleotide concentration by HPLC. B) ΔΨm measure- ment with Safranine O as described in Fig. 3. Arrows indicate the sequential additions of mitochondria (M); 10 mM succinate (Succ); ATP (130 μM); GTP (130 μM); ADP (130 μM); CCCP (10 μM). C) Concentrations of ATP, ADP, AMP, Ade, GTP, GDP, GMP, or guanosine (Gno) after addition of ATP, GTP orADP during oxygen uptake. D)Mitochondriawere incubated in oxygen uptake medium and amixture of ATP (130 μM) and GTP (130 μM)was added to stimulate respiration. Oxygen uptake (D), ΔΨm measurement (E) and nucleotide concentration (F) were determined as described in the Experimental procedures section. Arrows indicate sequential additions of mitochondria (M); 10 mM succinate (Succ); a mixture of ATP + GTP (130 μM each one); ADP (130 μM); CCCP (10 μM). In all cases curves show representative experiments of at least four different and independent mitochondrial preparations. 148 O. Flores-Herrera et al. / Biochimica et Biophysica Acta 1847 (2015) 143–152 selectivity for ATP as observed in Fig. 4. This observation suggests that substrate selectivity of ATP-diphosphohydrolase (i.e. ATP versus GTP preference) is regulated by ΔΨm. 3.7. Progesterone synthesis by syncytiotrophoblast mitochondria In intact syncytiotrophoblast mitochondria, ATP, ADP, GTP, and GDP hydrolysis by the ATP-diphosphohydrolase has been associated with progesterone synthesis, particularly with cholesterol flux between mitochondrial membranes [2]. As ATP-diphosphohydrolase inhibition with FSBA decreased progesterone production [2] we explored the control of ΔΨm on progesterone synthesis (Fig. 6A). Syncytiotrophoblast mitochondria were incubated as described in the Experimental procedures section in the presence of isocitrate to maintain a high NADPH/NADP+ ratio to supply energy to P450scc [27]. Under these conditions mitochondrial progesterone (P4) synthesis was 143 ± 12 ng P4/min·mg protein; it increased to 391 ± 10 ng P4/min·mg protein when ATP was added, and to 606 ± 52 ng P4/min·mg protein if 22-(R)-hydroxy cholesterol was present (Fig. 6A). The concomitant addition of ATP and 22-(R)-hydroxy cholesterol augmented progesterone production to 594 ± 82 ng P4/min·mg protein. This increase in progesterone synthesis was observed even in the presence of CCCP (427 ± 15 and 536 ± 12 ng/mg·min respectively) (Fig. 6A). The addition of ADP, GTP, or GDP slightly increased progesterone synthesis (340 ± 82; 206 ± 17, and 173 ± 19 ng P4/min·mg protein, respectively). However, when mitochondria were incubated without isocitrate, with the consequent suppression of NADPH synthesis, progesterone production decreased even in the presence of ATP, ADP, GTP or GDP (100 ± 10; 75 ± 12; 53 ± 9, and 44 ± 11 ng/mg·min, respectively) (Fig. 6A). ATP- diphosphohydrolase activity was evaluated simultaneously for proges- terone synthesis (Fig. 6B). In the presence of ΔΨm adenine nucleosides, tri- and diphosphates were preferentially hydrolyzed over guanosine nucleotides, i.e. GTP or GDP (Fig. 6B, black bars). In the presence of CCCP (which collapsed the ΔΨm), ATP-diphosphohydrolase activity was similar with all nucleotides tested (ATP, ADP, GTP, and GDP) (Fig. 6B, white bars). Although CAT was added prior toΔΨm dissipation with CCCP, it did not have any significant effect on ATP- diphosphohydrolase activity (Fig. 6B, gray bars). This suggested that ΔΨm might be involved in cholesterol flow by regulating ATP-diphosphohydrolase activity, but for progesterone synthesis the NADPH/NADP+ ratio is important. To verify this possibility, we used the mitochondrial steroidogenic contact sites [23] from human placenta as an alternative experimental approach. These contact sites can synthesize progesterone [23] since they contain the whole steroidogenic machinery, including the cytochrome Fig. 5.ΔΨm regulates syncytiotrophoblastmitochondrial ATP-diphosphohydrolase substrate selectively.Mitochondriawere incubated in oxygen uptakemedium (described in the Exper- imental procedures section) plus 5 μMcarboxyatractyloside (CAT) to inhibit the translocation of adenine nucleotides by blocking the ADP/ATP carrier. CCCP (10 μM)was added to collapse ΔΨm and obtain the maximum oxygen uptake rate. After CCCP, the addition of 130 μM ATP (A), 130 μM GTP (D) or an ATP + GTP mixture (130 μM each one, G) was performed. Mitochondria were incubated with CAT (5 μM), CCCP (10 μM), and 10 mM of succinate (Succ), and then 130 μM of ATP (B), 130 μM of GTP (E), or an ATP + GTP mixture (130 μM of each, H) was added. At the times indicated in the oxygen uptake recording, an aliquot was withdrawn to quantify the nucleotide concentration by HPLC when ATP (C), GTP (F), or ATP + GTP (I) were added. The figure shows representative experiments of at least four different and independent mitochondrial preparations. 149O. Flores-Herrera et al. / Biochimica et Biophysica Acta 1847 (2015) 143–152 P450scc, the 3-β-hydroxy-steroid dehydrogenase, the adrenodoxin and adrenodoxin reductase, the ATP-diphosphohydrolase [23], and a NADP- dependent isocitrate dehydrogenase [27]. It is important to mention that these contact sites do not generate ΔΨm, so that the participation of this parameter was excluded. Contact sites were isolated as described in the Experimental procedures section, incubated in either the absence or presence of isocitrate to stimulate NADPH production, with or with- out CCCP, and the amount of progesterone synthesis was determined (Fig. 6C). NADPH is the substrate of the adrenodoxin reductase that sup- ports cytochrome P450scc activity, and isocitrate dehydrogenase activ- ity regenerates NADPH [27]. Progesterone synthesis reached a value of 330 ± 50 ng/mg·h in the presence of NADPH production, while in the absence of NADPH there was no progesterone synthesis (Fig. 6C, black bars). CCCP had no effect on progesterone synthesis in the presence of NADPH (Fig. 6C, white bars). ATP-diphosphohydrolase activity associat- ed with the contact sites was unaffected by the presence or absence of NADPH (Fig. 6D). 4. Discussion Human placenta is essential to maintain pregnancy. Mitochondria of syncytiotrophoblast cells, besides generating ATP, synthetize progesterone using cholesterol as a substrate [3]. Therefore, syncytiotrophoblast mitochondria must reconcile ATP synthesis with hormone production. Isolated syncytiotrophoblast mitochondria retain their ability to couple oxygen uptake to ATP synthesis as well as their capacity to synthesize progesterone (Table 1). The specific hormone(s) or substance(s) that modulate P4 synthesis and ATP production during pregnancy are currently unknown [3,28]. Themost striking observation of this work shows that the ATP-diphosphohydrolase, an accessory enzyme involved in cholesterol flux between outer and inner syncytiotrophoblast mitochondrial membranes [1,2], might be regulat- ed by ΔΨm. Kinetic characterization of the detergent-solubilized ATP- diphosphohydrolase from mitochondria showed that this enzyme had low substrate selectivity (Figs. 1 and 2). It was capable of hydrolyzing purine or pyrimidine, tri or diphosphate nucleotides with similar affin- ities [1]. Its activation energy showed values between 4 and 6 kcal/mol with a single rate-limiting step during catalysis. In sharp contrast, ATP-diphosphohydrolase is regulated by ΔΨm in energized mitochondria. If mitochondria were energized (i.e. with succinate), the ATP-diphosphohydrolase preferentially hydrolyzed ATP, even if other substrates such as GTP, were present (Fig. 4), and Fig. 6. Steroidogenesis and nucleotide hydrolysis in syncytiotrophoblastmitochondria or steroidogenic contact sites. A) Progesterone synthesis by intactmitochondria. Mitochondriawere incubated in the presence (black bars) or absence (white bars) of isocitrate to maintain NADPH production (see text for details). ATP (1 mM), GTP (1mM), ADP (1mM), GDP (1mM) or 22-hydroxy cholesterol (22OH, 25 μM) was added to stimulate progesterone synthesis. Alternatively, CCCP (10 μM) was added in the presence (gray bars) or absence (dashed bars) of isocitrate. The results are the mean ± S.D. of n = 4 different and independent mitochondrial preparations. B) ATP-diphosphohydrolase activity in intact mitochondria incubated in the presence (white bars) or absence (black bars) of CCCP. Results are the mean ± S.D. with n = 4. C) Progesterone production by steroidogenic contact sites in the presence (+NADPH) or absence (−NADPH) of 2 mM of NADPH. CCCPwas added (white bars) to test its effect on progesterone synthesis. Results are themean± S.D. with n= 3. D) ATP-diphosphohydrolase activity by steroidogenic contact sites in the presence (+NADPH, black bars) or absence (−NADPH, white bars) of 2 mM NADPH. Results are the mean ± S.D. with n = 3. The one-way ANOVA analysis showed a statistical significant difference (p≤ 0.001) between different groups of datamarkedwith: a, g, h, l, *, and **. The difference is greater than it would be expected by chance (all pairwise multiple comparison procedures were performed with the Tukey test). The comparison marked with § shows no significant difference (p≤ 0.001). The two way ANOVA analysis showed a statistically significant difference (p≤ 0.005) between groupsmarkedwith: b, c, d, e, f, I, j, k, m, n, o, ¥, £, ¶, and ¤ (all pairwisemultiple comparison procedures were performed with the Holm–Sidak method). 150 O. Flores-Herrera et al. / Biochimica et Biophysica Acta 1847 (2015) 143–152 promoted mitochondrial oxygen uptake coupled to ATP synthesis (Fig. 3). In energized mitochondria from all organisms tested, ADP but not ATP increased the respiration rate. This is controlled by the adenine nucleotide translocase (ANT), which does not recognize outer mitochondrial ATP if ΔΨm is generated by ETC. This is a key control mechanism that prevents a futile-cycle and energy waste as heat. In syncytiotrophoblast mitochondria energized with succinate, ATP addition induced oxygen uptake through the ATP-diphosphohydrolase activity, which hydrolyzed ATP to ADP and inorganic phosphate (Fig. 3 and [1,10]). The activities of the ATP-diphosphohydrolase (ATPhydroly- sis) and F1F0-ATP synthase (ATP synthesis) do not produce permanent oxygen uptake stimulation ([10] and the present study) as has been observed with the brain hexokinase associated with the outer mitochondrial membrane [29]. This observation is consistentwith theproposed role for this enzyme [2]; it was suggested that the ATP-diphosphohydrolase could perform sequential hydrolysis of ATP to ADP, andADP to AMPand simultaneous- ly stimulate cholesterol transport between mitochondrial membranes for progesterone synthesis [2] (Fig. 7). Although cholesterol transfer during steroidogenesis in the adrenals glands occurs through a macro- molecular complex that consists of outer membrane proteins such as the mitochondrial membrane translocator protein (TSPO), and the TSPO-associated protein PAP7 that bind and lead the regulatory subunit RI-α of the cAMP-dependent protein kinase (PKARIα) towards mito- chondria [30], the TSPO protein in human syncytiotrophoblast mito- chondria remains unidentified [31]. A potential candidate exists. A multiprotein complex, the steroidogenic contact sites, has been associated with cholesterol transport and steroidogenesis in human syncytiotrophoblast mitochondria [23]. These contact sites contain proteins such as HSP60 [32], ANT, VDAC, cytochrome P450scc, adrenodoxin reductase, adrenodoxin, NADP-dependent isocitrate dehydrogenase, 3-β-hydroxysteroid dehydrogenase, STARD3 protein and ATP-diphosphohydrolase [3,23]. The molecular weight of the ATP-diphosphohydrolase calculated from native-gel was 163 kDa, in contrast with the 59 kDa determined from tandem mass spectrometry. The reported molecular weight from SDS-PAGE, radiation-inactivation or gel filtration goes from 64 to 70 kDa [33–35]. The molecular weights in native conditions can be explained by one of the following hypotheses: A) that the native state of mitochondrial ATP-diphosphohydrolase is a homo-oligomer (i.e. a dimer or trimer), as has been reported for other organisms [36,37], or B) that ATP-diphosphohydrolase interacts with other mitochondrial membrane protein(s) (i.e. steroidogenic contact site proteins). We can hypothesize that ΔΨm regulates ATP-diphosphohydrolase activity through the close interactions between proteins from contact sites. Although ATP-diphosphohydrolase lacks a classic mitochondrial targeting presequence, it has been reported that many mitochondrial hydrophobic membrane proteins are synthesized without cleavable extensions [38]. These proteins typically contain several targeting signals that are distributed over the entire length of the protein [39]. However, this hypothesis needs to be elucidated. Even though ΔΨm reflects efficientmitochondrial oxygen consump- tion and ATP synthesis, it is not crucial for progesterone synthesis in syncytiotrophoblast cells. When intact mitochondria were incubated in the presence of CCCP to collapse ΔΨm, and isocitrate to maintain NADPH, cholesterol was efficiently transformed into progesterone (Fig. 6A). Although in the presence of CCCP ATP-diphosphohydrolase did not show substrate selectivity and its activity was decreased, ATP hydrolysis in the presence or absence of CCCP increased cholesterol transport and its transformation into progesterone, in the presence of NADPH. This suggested that the remaining activity of the ATP- diphosphohydrolase (around 50%) fully supports the synthesis of progesterone. As ΔΨm, respiration and ATP synthesis are solid indicators of functional mitochondria, they are believed to be crucial for Leydig cell steroidogenesis [6,7]. However, using steroidogenic contact sites from syncytiotrophoblast mitochondria in this study allowed us to focus on the role of ATP (ATP-diphosphohydrolase activity) and NADPH (NADP-dependent isocitrate dehydrogenase activity) in the synthesis of progesterone; it is important to mention that steroidogenic contact sites do not generate ΔΨm, but transform cholesterol into progesterone (Fig. 6C). With this experimental approach, we were able to establish that progesterone synthesis is sensitive to the presence of NADPH and nucleotide hydrolysis but insensitive to ΔΨm (the present study and [23]). In syncytiotrophoblast cells ΔΨm regulates the specificity of ATP- diphosphohydrolase allowing most of the ATP available to be used for cholesterol transport during progesterone synthesis (Fig. 7), leaving GTP, GDP and other nucleotides available for other important metabolic reactions (i.e. proteins synthesis). However, in some patho- logical events such as preeclampsia, calcium accumulation in mito- chondria of trophoblast cells may collapse ΔΨm and interfere with ATP synthesis. In an attempt to sustain progesterone synthesis, ATP- diphosphohydrolase may use any nucleotide available (i.e. GTP), to maintain the required cholesterol flux for progesterone synthesis. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.bbabio.2014.10.002. Acknowledgements We dedicate this work as a memorial to Dr. Guillermo Mendoza- Hernandez, an exceptional friend and colleague, and coauthor of this paper, who passed away suddenly on July 13, 2012. This work was partially supported by research grants IN214914 (OFH), IN209614 (JPP) and IN211912 (FM) from Dirección General de Asuntos del Personal Académico (DGAPA) from Universidad Nacional Autónoma de México (UNAM), as well as Grant 168025 from CONACYT (FM). Mercedes Esparza-Perusquía is a Ph.D. student of the Biological Science Program of Universidad Nacional Autónoma de México (511021118) and fellow of CONACYT (254400). Fig. 7. Model for the regulation of mitochondrial ATP-diphosphohydrolase by ΔΨm in syncytiotrophoblast cells. ΔΨm induces ATP hydrolysis by ATP-diphosphohydrolase (1) during cholesterol transport (2) at the contact sites. ADP can be hydrolyzed to AMP by ATP-diphosphohydrolase (1) or can enter the mitochondrial matrix for ATP synthesis by F1F0-ATP synthase (3), stimulating oxygen uptake and the proton pump by CTE (4). Cholesterol is transformed into pregnenolone (P5) by cytochrome P450scc and then into progesterone (P4) by 3β-hydroxysteroid dehydrogenase (5). Steroidogenic contact sites are constituted by different proteins including adenine nucleotide translocase (ANT), ATP-diphosphohydrolase (ATP-D), NADP-dependent isocitrate dehydrogenase (ICD), cytochrome P450scc (P450) and 3β-hydroxysteroid dehydrogenase (3βHSD). OM = outer mitochondrial membrane; IM = inner mitochondrial membrane; IMS = inter membrane space; Chol = cholesterol. 151O. Flores-Herrera et al. / Biochimica et Biophysica Acta 1847 (2015) 143–152 References [1] O. Flores-Herrera, A. Uribe, J.P. Pardo, J.L. Rendon, F. Martinez, A novel ATPdiphosphohydrolase from human term placental mitochondria, Placenta 20 (1999) 475–484. [2] O. Flores-Herrera, A. Uribe, C. Garcıa-Perez, R. Milan, F. Martinez, 5′-p- Fluorosulfonylbenzoyl adenosine inhibits progesterone synthesis in human placental mitocondria, Biochim. Biophys. Acta 1585 (2002) 11–18. [3] F. Martinez, M. Kiriakidou, J.F. Strauss III, Structural and functional changes in mitochondria associated with trophoblast differentiation: methods to isolate enriched preparations of syncytiotrophoblast mitochondria, Endocrinology 138 (1997) 2172–2183. [4] N. Cherradi, G. Defaye, E.M. Chambaz, Characterization of the 3 betahydroxysteroid dehydrogenase activity associated with bovine adrenocortical mitochondria, Endocrinology 134 (1994) 1358–1364. [5] C. Brand, N. Cherradi, G. Defaye, A. Chinn, E.M. Chambaz, J.J. Feige, S. Bailly, Transforming growth factor beta1 decreases cholesterol supply to mitochondria via repression of steroidogenic acute regulatory protein expression, J. Biol. Chem. 273 (1998) 6410–6416. [6] J.A. Allen, T. Shankara, P. Janus, S. Buck, T. Diemer, K.H. Hales, D.B. Hales, Energized, polarized, and actively respiring mitochondria are required for acute Leydig cell steroidogenesis, Endocrinology 147 (2006) 3924–3935. [7] A.S. Midzak, H. Chen, M.A. Aon, V. Papadopoulos, B.R. Zirkin, ATP synthesis, mitochondrial function, and steroid biosynthesis in rodent primary and tumor Leydig cells, Biol. Rep. 84 (2011) 976–985. [8] F. Martinez, R. Milan, O. Flores-Herrera, S. Olvera-Sanchez, E. Gomez-Chang, M.T. Espinosa-Garcia, The role of mitochondria in syncytiotrophoblast cells: bioenergetics and steroidogenesisChapter 19 in: J. Zheng (Ed.), Recent Advances in: Research on the Human Placenta, Croatia, 2012, pp. 397–428. [9] X. Xu, T. Xu, D.G. Robertson, J.D. Lamberth, GTP stimulates pregnenolone generation in isolated rat adrenal mitochondria, J. Mol. Biol. 264 (1989) 17674–17680. [10] F. Martinez, M.T. Espinosa-Garcıa, O. Flores-Herrera, J.P. Pardo, Respiratory control induced by ATP in human term placental mitochondria, Placenta 14 (1993) 321–331. [11] A. Bensadoun, D. Weinstein, Assay of proteins in the presence of interfering materials, Anal. Biochem. 70 (1976) 241–250. [12] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein measurement with the Folin phenol reagent, J. Biol. Chem. 193 (1951) 265–275. [13] B. Chance, G.R. Williams, The respiratory chain and oxidative phosphorylation, Adv. Enzymol. 17 (1956) 65–134. [14] F.Martinez, J.P. Pardo, O. Flores-Herrera,M.T. Espinosa-Garcia, Theeffect of osmolarity on human placental mitochondria function, Int. J. Biochem. Cell Biol. 27 (1995) 795–803. [15] J.C. Fischer,W. Ruitenbeek, J.A. Berden, J.M.F. Trijbels, J.H. Veerkamp, A.M. Stadhouders, R.C.A. Sengers, J.M. Janssen, Differential investigation of the capacity of succinate oxidation in human skeletal muscle, Clin. Chim. Acta 153 (1985) 23–36. [16] K. Kita, C.R.T. Vibat, S. Meinhardt, J.R. Guest, R.B. Gennis, One-step purification from Escherichia coli of complex II (succinate:ubiquinoneoxidoreductase) associated with succinate-reducible cytochrome b556, J. Biol. Chem. 264 (1989) 2672–2677. [17] P.A. Lanzetta, L.J. Alvarez, P.S. Reinach, O.A. Candia, An improved assay for nanomole amounts of inorganic phosphate, Anal. Biochem. 100 (1979) 95–97. [18] A. Guevara-Flores, S. Olvera-Sanchez, C. Gómez-Concha, O. Juárez, M. Esparza- Perusquia, J.P. Pardo, G. Mendoza-Hernández, F. Martinez, O. Flores-Herrera, 5′-p- Fluorosulfonyl benzoyl adenosine inhibits an ecto-ATP-diphosphohydrolase in the tegument surface of Taenia crassiceps cysticerci, Mol. Biochem. Parasitol. 162 (2008) 123–133. [19] R.C. Tuckey, Cholesterol side-chain cleavage by mitochondria from the human placenta. Studies using hydroxycholesterols as substrates, J. Steroid Biochem. Mol. Biol. 42 (1992) 883–890. [20] H. Schagger,W.A. Cramer, G. von Jagow,Analysis ofmolecularmasses and oligomeric states of protein complexes by blue native electrophoresis and isolation of mem- brane protein complexes by two-dimensional native electrophoresis, Anal. Biochem. 217 (1994) 220–230. [21] I. Wittig, H. Schagger, Electrophoretic methods to isolate protein complexes from mitochondria, Methods Cell Biol. 80 (2007) 723–741. [22] D. de Los Rios-Castillo, M. Zarco-Zavala, S. Olvera-Sanchez, J.P. Pardo, O. Juarez, F. Martinez, G. Mendoza-Hernandez, J.J. García-Trejo, O. Flores-Herrera, Atypical cristae morphology of human syncytiotrophoblast mitochondria. Role for complex V, J. Biol. Chem. 286 (2011) 23911–23919. [23] A. Uribe, J.F. Strauss III, F. Martinez, Contact sites from human placental mitochon- dria: characterization and role in progesterone synthesis, Arch. Biochem. Biophys. 413 (2003) 172–181. [24] B. Kadenbach, R. Ramzan, L. Wen, S. Vogt, New extension of the Mitchell Theory for oxidative phosphorylation in mitochondria of living organisms, Biochim. Biophys. Acta 1800 (2010) 205–212. [25] G. Daum, J.E. Vance, Import of lipids into mitochondria, Prog. Lipid Res. 36 (1997) 103–130. [26] D. Mokranjac, W. Neupert, Energetics of protein translocation into mitochondria, Biochim. Biophys. Acta 1777 (2008) 758–762. [27] C. Garcia-Perez, J.P. Pardo, F. Martinez, Ca(2+) modulates respiratory and steroidogenic activities of human term placental mitochondria, Arch. Biochem. Biophys. 405 (2002) 104–111. [28] J.F. Strauss III, F. Martinez, M. Kiriakidou, Placental steroid hormone synthesis: unique features and unanswered questions, Biol. Reprod. 54 (1996) 303–311. [29] W.S. da-Silva, A. Gómez-Puyou, M.T. de Gómez-Puyou, R. Moreno-Sanchez, F.G. de Felice, L. de Meis, M.F. Oliveira, A. Galina, Mitochondrial bound hexokinase activity as a preventive antioxidant defense: steady-state ADP formation as a regulatory mechanism of membrane potential and reactive oxygen species generation in mitochondria, J. Biol. Chem. 279 (2004) 39846–39855. [30] J. Liu, M.B. Rone, V. Papadopoulos, Protein–protein interactions mediate mitochon- drial cholesterol transport and steroid biosynthesis, J. Biol. Chem. 281 (2006) 38879–38893. [31] M.G. Maldonado-Mercado, M.T. Espinosa-Garcia, C. Gomez-Concha, J. Monreal- Flores, F. Martinez, Steroidogenesis in BeWo cells: role of protein kinase A and benzodiazepines, Int. J. Biochem. Cell Biol. 40 (2008) 901–908. [32] S. Olvera-Sanchez, M.T. Espinosa-Garcia, J. Monreal, O. Flores-Herrera, F. Martinez, Mitochondrial heat shock protein participates in placental steroidogenesis, Placenta 32 (2011) 222–229. [33] M.A. Valenzuela, A.M. Kettlun, S. Sandoval, L. García, M. Mancilla, G. Neckelmann, L. Chayet, A. Alvarez, F. Cuevas, L. Collados, V. Espinosa, A. Traverso-Cori, I. Bravo, C.G. Acevedo, E. Aranda, Comparison of the biochemical properties, regulation and function of ATP-diphosphohydrolase from human placenta and rat kidney, Braz. J. Med. Biol. Res. 29 (1996) 589–597. [34] A. Biederbick, C. Kosan, J. Kunz, H.P. Elsässer, First apyrase splice variants have different enzymatic properties, J. Biol. Chem. 275 (2000) 19018–19024. [35] A.M. Kettlun, A. Alvarez, R. Quintar, M.A. Valenzuela, L. Collados, E. Aranda, A. Banda, L. Chayet, M. Chiong, M. Mancilla, Human placental ATP-diphosphohydrolase: biochemical characterization, regulation and function, Int. J. Biochem. 26 (1994) 437–448. [36] E. Faudry, J.M. Santana, C. Ebel, T. Vernet, A.R. Teixeira, Salivary apyrases of Triatoma infestans are assembled into homo-oligomers, Biochem. J. 396 (2006) 509–515. [37] B.U. Failer, A. Aschrafi, G. Schmalzing, H. Zimmermann, Determination of native oligomeric state and substrate specificity of rat NTPDase1 and NTPDase2 after heterologous expression in Xenopus oocytes, Eur. J. Biochem. 270 (2003) 1802–1809. [38] A. Chacinska, N. Pfanner, C. Meisinger, How mitochondria import hydrophilic and hydrophobic proteins, Trends Cell Biol. 12 (2002) 299–303. [39] T. Stan, J. Brix, J. Schneider-Mergener, N. Pfanner, W. Neupert, D. Rapaport, Mitochondrial protein import: recognition of internal import signals of BCS1 by the TOM complex, Mol. Cel. Biol. 23 (2003) 2239–2250. 152 O. Flores-Herrera et al. / Biochimica et Biophysica Acta 1847 (2015) 143–152 Review Multiple functions of syncytiotrophoblast mitochondria Federico Martinez ⇑, Sofia Olvera-Sanchez, Mercedes Esparza-Perusquia, Erika Gomez-Chang, Oscar Flores-Herrera Departamento de Bioquímica, Facultad de Medicina, Universidad Nacional Autónoma de México, Apdo. Postal 70-159, Coyoacan 04510, México, D.F., Mexico a r t i c l e i n f o Article history: Received 2 February 2015 Received in revised form 16 September 2015 Accepted 27 September 2015 Available online 3 October 2015 Keywords: Syncytiotrophoblast mitochondria MLN64 Mitochondrial kinases Mitochondrial structure Protein phosphorylation Cholesterol transport a b s t r a c t The human placenta plays a central role in pregnancy, and the syncytiotrophoblast cells are the main components of the placenta that support the relationship between the mother and fetus, in apart through the production of progesterone. In this review, the metabolic processes performed by syncytiotrophoblast mitochondria associated with placental steroidogenesis are described. The metabolism of cholesterol, specifically how this steroid hormone precursor reaches the mitochondria, and its transformation into progesterone are reviewed. The role of nucleotides in steroidogenesis, as well as the mechanisms associ- ated with signal transduction through protein phosphorylation and dephosphorylation of proteins is dis- cussed. Finally, topics that require further research are identified, including the need for new techniques to study the syncytiotrophoblast in situ using non-invasive methods.  2015 Elsevier Inc. All rights reserved. Contents 1. General aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2. Cholesterol synthesis in the human placenta. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 3. Lipoprotein receptors for cholesterol supply to placenta . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 4. Involvement of organelle-mitochondria interactions in intracellular lipid flux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 5. StAR in adrenal glands and gonads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 6. MLN64 in human placenta . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 7. Mitochondrial contact sites associated with steroidogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 8. Steroidogenic contact sites in human syncytiotrophoblast mitochondria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 9. Architecture and physiological characteristics of syncytiotrophoblast mitochondria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 10. Role of Dwm in placental steroidogenesis: regulation of mitochondrial ATP-diphosphohydrolase. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 11. Placental signal transduction machinery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 12. cAMP-mediated signal transduction cascade in progesterone synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 13. Conclusion and perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 1. General aspects The human placenta plays a major role during pregnancy. Through progesterone synthesis, it establishes central communication mechanisms between mother and fetus. Proges- terone is essential for blastocyst implantation, extracellular matrix remodeling, and promoting trophoblast migration. The human pla- centa safeguards the fetus and contributes to its maturation [1]. The human placenta undertakes metabolic processes related to hormone synthesis as well as transport of nutrients, gases and http://dx.doi.org/10.1016/j.steroids.2015.09.006 0039-128X/ 2015 Elsevier Inc. All rights reserved. ⇑ Corresponding author. E-mail address: fedem@bq.unam.mx (F. Martinez). Steroids 103 (2015) 11–22 Contents lists available at ScienceDirect Steroids journal homepage: www.elsevier .com/locate /s teroids disposal of waste products. In this context, the placenta exhibits functions that can be compared to those of the pituitary gland, ovaries, intestines, lungs, and kidneys. During human placentation, cell fusion of mononuclear cytotro- phoblast cells occurs to produce a multinucleated cellular layer named syncytiotrophoblast, which forms an interface between the maternal and fetal circulation [2], and is also the site of proges- terone synthesis [3,4]. Approximately, two weeks after conception, the villus structures appear containing cytotrophoblasts and syn- cytiotrophoblast. The syncytiotrophoblast is maintained during pregnancy by continuous fusion of cytotrophoblasts. Cytotrophoblast differentiation into syncytiotrophoblast is accompanied by changes in cell morphology, mitochondrial cristae remodeling, and the acquisition of the steroidogenic machinery. The steroidogenic machinery of syncytiotrophoblast cells consists of an electron transport chain (ETC-P450scc) composed of cyto- chrome P450scc (CYP11A1), which receives electrons from NADPH + H+ through adrenodoxin and adrenodoxin reductase. These proteins are located in the inner mitochondrial membrane and transform cholesterol into pregnenolone [5–7]. An additional enzyme, the 3b-hydroxysteroid-dehydrogenase-D4–5 isomerase type I (3b HSDI), also embedded in the inner mitochondrial mem- brane of syncytiotrophoblast cells [6], transforms pregnenolone into progesterone [5,7]. The human placenta is a transient tissue exclusively present during pregnancy. Its multiple functions must be strictly regulated to ensure the successful completion of pregnancy, and hence, from an evolutionary perspective, the maintenance of the human race. It has been suggested that the placenta functions under autonomous control through autocrine, paracrine, or intracrine signals which render it independent from other tissues and organs. It has been demonstrated that human placental explants resist non-physiological physical conditions (i.e., hypotonic media or temperatures higher than 45 C) unlike other tissues, and that tro- phoblasts can synthesize progesterone under stress conditions [8]. These data suggest that the human placenta has self-preserving mechanisms that are independent of external stimuli and serve as physiological, metabolic, and functional control strategies to protect the fetus [8]. The regulation of the human placenta and how it modifies maternal functions, remain to be elucidated. Recently, an editorial comment was issued about how little attention is given to placen- tal studies. It focused on the lack of non-invasive methods to treat illnesses of both the mother and the fetus, which could prove life- saving or allow in utero treatment to prevent birth defects [9]. The human placenta produces high levels of progesterone from the cholesterol provided by the mother, a steroid hormone synthe- sized by syncytiotrophoblast mitochondria that regulates the immunological processes that prevent fetal rejection. The absence or a significant decrease in progesterone levels, regardless of its etiology, results in the lack of communication between mother and fetus, and usually in spontaneous abortion or miscarriage. Hence, a central role of human placental steroidogenesis is to maintain the pregnancy. However, pregnancies where the fetus has a CYP11A1 mutation can end in a viable neonate despite very low progesterone levels [10,11]. It has been suggested that intracelluar cholesterol transport takes place through a strictly regulated mechanism performed by multi-protein complexes [12]. Biological membranes and some cellular processes require specific concentrations of this sterol, since slight modifications in its concentration may produce signif- icant functional alterations. The lipid composition of membranes contributes to the cellular distribution of cholesterol. For instance, the concentration of cholesterol in the endoplasmic reticulum of mammalian cells is low, 5 mol% of phospholipids. In contrast, in the Golgi apparatus, the concentration is P30 mol% [12]. However, the largest concentrations of cholesterol in a cell are found in the plasma membrane, which has been reported to contain up to 60% of the sterol [12]. The different cholesterol concentrations found among cellular compartments suggest the existence of systems that distribute cholesterol to the required sites, as is the case of steroidogenic mitochondria [13]. Cholesterol transport pathways involve vesicles and specific proteins. There are proteins that con- trol the synthesis of lipids, those that function as inducers or mod- ulators of gene expression, and the ones that transport lipids themselves. Regardless of the mechanism, it is essential to know what determines the movement of cholesterol, because the con- centration of cholesterol in membranes can modify their biological functions as well as that of associated proteins. When preg- nenolone is synthesized, particularly in the placenta, cholesterol transport toward the mitochondria is vectorial, efficient, and in adequate concentration to elicit a response according to cellular needs without disturbing other cell functions. Among sterol carrier proteins involved in the intracellular flux of cholesterol in steroidogenic tissues, the START family of proteins are most relevant [14], particularly the StAR (STARD1) and MLN64 (STARD3) proteins, which play a central role in cholesterol move- ment between mitochondrial membranes supplying substrate for hormones synthesis. The START proteins are associated, either with multiprotein complexes belonging to vesicles, or with specific organelles (i.e., mitochondria). This allows for the efficient and tar- geted distribution of cholesterol. Diverse contact sites between organelles have been described (see Organelles and mitochondria interaction for intracellular lipid flux section). These consist of transporters, enzymes and anchor pro- teins, among others, that generate protein membrane domains [15]. Among the different types of contact sites that have been described are the Mitochondria Associated Membrane (MAMs) between the endoplasmic reticulum and mitochondria, which con- tribute to the control of metabolism. However, although the mito- chondria play a pivotal role in steroidogenic tissues [16], it should be noted that mitochondria also participate in other cellular pro- cesses [17], such as ATP production, intracrine signaling [18], apop- tosis, innate immunity, autophagy, or the stress response [19], suggesting that mitochondria by themselves, or associated with other structures, are important to maintaining cellular functions. This is a topic that requires a further investigation in the human placenta. 2. Cholesterol synthesis in the human placenta Distinct from other steroidogenic tissues, such as adrenal glands [20], the human placenta does not produce significant amounts of cholesterol de novo. Studies using acetate-14C have shown that this precursor is not the main substrate for the sterol synthesis path- way in human placenta [21–24]. However, cholesterol is needed for the substantial placental progesterone synthesis during preg- nancy. In this sense, cholesterol production by human placenta is not sufficient to support steroidogenesis. The modest amount of cholesterol synthesized by the placenta could be used to satisfy its own requirements such as the structural needs, i.e., to maintain the plasma membrane morphology and function. In the syncy- tiotrophoblast, the cholesterol needed for progesterone synthesis comes from LDL [25,26], which are internalized through receptors mediated by endocytosis [27]. 3. Lipoprotein receptors for cholesterol supply to placenta The trophoblast cells have different types of receptors; one of these is for LDL [28]. Also, HDL receptors have been described on trophoblast cells [29], in BeWo cells line [30], and numerous steroidogenic tissues (Fig. 1). Unlike LDL, HDL are not processed 12 F. Martinez et al. / Steroids 103 (2015) 11–22 through the lysosomal degradation pathway [31], suggesting that placenta can obtain cholesterol from different pathways. Progesterone synthesis presumably depends on the capacity of the cells to bind LDL and thus supply mitochondria with choles- terol. The concentration of LDL receptors is regulated by the pres- ence of LDL. In dispersed trophoblast culture cells incubated with LDL-125I for long periods of time, uptake and degradation of LDL decreased by 90%. These results suggest that trophoblast cells reg- ulate the number of LDL receptors expressed in the plasma mem- brane. This effect was observed when trophoblast cells were incubated in the presence of HDL [32]. Recent data suggest that the placenta uses both LDL and HDL as a source of cholesterol for progesterone synthesis [33], which secures a constant cholesterol input for steroidogenesis, even under conditions that may alter cholesterol concentration, safeguarding pregnancy. However, in abetalipoproteinemia, progesterone levels during pregnancy are reduced indicating that HDL cholesterol cannot compensate for diminished LDL levels [34]. Although it is beyond the scope of this review, it is important to mention that the human placenta also transports cholesterol to the fetal bloodstream (Fig. 1). The fetus requires cholesterol for cell growth and development. The following references address this issue [35,36] and highlight the diverse array of proteins that is required [37–40]. 4. Involvement of organelle-mitochondria interactions in intracellular lipid flux Mitochondrial Inner Membrane Organizing System (MINOS) and Mitochondria Associated Membrane (MAMs) have been put forward as potential types of unions between membranes of diverse organelles. Diverse types of unions are associated with transport mechanisms, vesicle mediated transport, and protein mediated translocation. MAMs are multiprotein structures whose existence was identified by cosedimentation of the endoplasmic reticulum and mitochondria. MAMs participate in lipid exchange [41], highlighting their role in the endoplasmic reticulum and mitochondria contact sites [42–48]. Membrane Contact Sites (MCS) result from the proximity between two membranes (<20–30 nm). Proteins associate transiently and undertake a speci- fic function without the fusion of membranes [15]. It has been sug- gested that MCS and MAMs participate in the association between lipid droplets and endosomes, where perilipin and SNARE elements potentially play a crucial role [42,44,45]. The cytoskeleton is also important since colchicine decreases steroidogenesis [49–52]. Par- ticularly for steroidogenic tissues, these observations provide a basis for an integrated model of cholesterol flux from endoplasmic reticulum, late endosomes or lipid droplets, to mitochondria, which could represent an initial step in steroidogenesis [53–55]. However, lipid composition of mitochondrial membranes is pre- cisely controlled. Horvath and Daum [56] describe the reasons why mitochondria are unique organelles regarding lipid transport. Firstly, they pos- sess a double bilayer. Although they are closely positioned, each layer maintains its structural and functional properties. Secondly, each layer and its subcompartments perform specific biochemical functions. The mitochondrial membranes have a particular lipid composi- tion. High concentrations of cholesterol accumulate in the outer membrane but not in the inner membrane [58]. In the presence of a BSA-cholesterol complex, human placental mitochondria increased progesterone synthesis, but notwhenmito- chondria where treated with trypsin [57]. Similar results were observed under different experimental conditions, showing that mitochondria treated with trypsin were unable to accumulate cholesterol in the presence of the BSA-cholesterol complex, suggest- ing that proteins are involved in the incorporationof cholesterol into the human placental mitochondria [58]. Therefore, successful cholesterol transport, its transformation into pregnenolone in acutely regulated steroidogenic tissues, or progesterone in the case of the human placenta (a ‘‘chronically” regulated tissue), requires multiple systems that include: proteins like StAR and MLN64; late endosomes; lipid droplets; the cytoskeleton; MAMs, MINOS, MCS; and contact sites between mitochondrial membranes (see below). Fig. 1. Cholesterol uptake by syncytiotrophoblast and its route through the placenta. Cholesterol is transported by endocytosis via specific receptors for HDL or VLDL and is distributed by endosomes to mitochondria and into the fetal bloodstream. Mitochondria transform cholesterol into progesterone to maintain pregnancy. Syncytiotrophoblast has a low capacity of cholesterol synthesis, which is probably used for membrane microvilli structure. ST = syncytiotrophoblast cell; CT = cytotrophoblast cell; V = vascular endothelium; N = nuclei; LE = late endosome; M = mitochondria; ER = endoplasmic reticulum; P4 = progesterone; Chol = cholesterol; LDL = low density lipoproteins; HDL = high density lipoproteins. F. Martinez et al. / Steroids 103 (2015) 11–22 13 5. StAR in adrenal glands and gonads As it has been mentioned, once cholesterol has been incorpo- rated into the cell (i.e., late endosomes, or lipid droplets), it is directed through a precise and effective transport system to mitochondria, where pregnenolone is synthesized [59] (Fig. 2). This transport system may be mediated by vesicles or sterol carrier proteins; in steroidogenic tissues both systems operate simultaneously. The sterol carrier proteins called START have been broadly stud- ied in steroidogenic tissues. From this family, the StAR is the most important protein in the cholesterol transfer system in mitochon- dria from adrenal glands and gonads [60]; mutations in the STARD1 gene cause Congenital Lipoid Adrenal Hyperplasia [61]. Several studies analyzed its genetic regulation [62], function and its role in steroidogenesis [16,63,64]. The transcription of the STARD1 gene increases significantly in response to adrenocorti- cotropic and luteinizing hormones, and phosphorylation of StAR stimulates cholesterol transport and steroidogenesis [65,66]. The StAR protein does not need to enter into the mitochondria to pro- mote cholesterol transfer since a truncated StAR protein (StAR- N62) lacking the mitochondrial targeting sequence is able to drive cholesterol transport, although it is not incorporated into the mito- chondria [67]. Interaction between cholesterol and StAR under acidic condi- tions induces a conformational change from the native structure to the molten globule state [68], suggesting that this dynamic state is involved in intramitochondrial sterol transfer [68]. Although these studies were performed in vitro, the mitochondrial condi- tions, particularly the acidic conditions of the intermembrane space could support this hypothesis. The mechanism of action of StAR remains unknown. However, it has been suggested that it mediates cholesterol transport from the outer to the inner mito- chondrial membrane. This process regulates steroidogenesis by controlling the conversion rate of cholesterol into pregnenolone by cytochrome P450scc. While acutely regulated steroidogenic tissues display a high, but transitory flux of cholesterol to mitochondria, the amount of cholesterol in mitochondrial membranes is low [56], suggesting a fast transformation of the sterol into pregnenolone. In contrast, the constant flow of cholesterol from cytosol to syncytiotro- phoblast outer mitochondrial membrane impacts the mitochon- drial sterol content, which reaches a value of 13.5 lg of cholesterol/mg mitochondrial protein [58], five times higher than reported for other types of mitochondria and close to the value observed in the plasma membrane. The constant transport of cholesterol to syncytiotrophoblast mitochondria allows the increased production of progesterone during pregnancy [59], i.e., in the last trimester of pregnancy, the human placenta can produce about 250–500 mg of progesterone daily [65,69]. Once located in the outer mitochondrial membrane, cholesterol must continue its transit to the inner mitochondrial membrane, reach cytochrome P450scc, and be transformed into pregnenolone. This step needs a sterol carrier protein, MLN64. 6. MLN64 in human placenta Similar to the gonads and adrenal glands, in human syncytiotro- phoblast there must be a sterol carrier protein involved in mito- chondrial cholesterol transport. Although StAR is the main protein associated with mitochondrial cholesterol flux in different acutely regulated steroidogenic tissues, it is not expressed in the human placenta [61]. In the last decade a new member of the START family, named MLN64, was identified in syncytiotro- phoblast cells [70]. MLN64 was initially described as an endosomal membrane protein in a metastatic lymphatic nodule, and contains the START domain for cholesterol binding [71]. Since the syncy- tiotrophoblast does not express StAR, MLN64 was considered to be the main intracellular protein involved in cholesterol transport in this tissue [56]. Alpy and Tomasetto [64] reported that although StAR and MLN64 contain a START domain, they also have different features that determine their intracellular location and structure. MLN64 is an endosomal protein with four transmembrane segments in its amino terminal end (named the MENTAL domain). Its subcellu- lar localization is relevant because late endosomes and lysosomes contain 50% of total cellular cholesterol, and they might play a role in the distribution of cholesterol in the cell [72]. Also, van der Kant et al. [72] showed that there are two types of late endosomes, one that contains the protein ORP1L and transports oxysterols; and another with MLN64 and ABCA3 (an ATP dependent transport pro- tein that translocate substrates, including lipids). Hence, they sug- gested that MLN64 could be involved in cholesterol flux from late Fig. 2. Flux of cholesterol to syncytiotrophoblast mitochondria. Cholesterol is incorporated by endocytosis. Early endosomes initiate cholesterol transport to mitochondria to stimulate steroid hormone synthesis. Incorporation of STARD3 into early endosomes promotes its maturation into late endosomes. STARD3 facilitates transport of late endosomal cholesterol to mitochondria. STARD3 is a member of the START domain superfamily that possesses cholesterol binding and transport activity. Lipid droplets are an alternative route to cholesterol flux to mitochondria. Late endosomes and lipid droplets might anchor to mitochondria through SNARE complexes. ST = syncytiotrophoblast; LDL = low density lipoproteins; HDL = high density lipoproteins; M = mitochondria; ER = endoplasmic reticulum; SNARE = S- NARE complexes. 14 F. Martinez et al. / Steroids 103 (2015) 11–22 endosomes to mitochondria. It has also been noted that amino acid residues M307 and N311 are essential for cholesterol binding to MLN64 [60,73,74]. We have demonstrated that MLN64 is incorpo- rated into syncytiotrophoblast mitochondria and proteolytically activated by a mitochondrial protease to stimulate steroidogenesis [75]. The proteolytic product (i.e., MLN64-28 kDa protein) is associated with steroidogenic contact sites (see below), probably to facilitate cholesterol transport between mitochondrial membranes. Charman et al. [73] transformed CHO cells into steroidogenic cells by transfecting them with a vector encoding a fusion protein with the complete electron transport chain of P450scc (F2-plas- mid). Although the expression of MLN64 was depleted by 80–90% with siRNA, progesterone synthesis was inhibited only by 30%, suggesting that MLN64 is not the only protein responsible for cholesterol transport for steroidogenesis [73]. Kishida et al. [76] demonstrated that in embryonic fibroblasts, lacking StAR, and obtained from MLN64 mutant mice, transfection with F2 plasmid resulted in reduced but not absent transformation of endogenous cholesterol into steroid hormones compared to wild-type embry- onic fibroblasts. This suggests that the MLN64 START domain is not necessarily the only protein participating in sterol metabolism. This is relevant for human placental steroidogenesis, since it expresses MLN64 and lacks the StAR protein. In this way, we have determined that HSP60 protein shares epi- topes with the START domain of MLN64. Both proteins co-im- munoprecipitated, and HSP60 labeled with fluorescein maleimide inhibited progesterone synthesis [77]. This observation suggests that HSP60 could participate in steroidogenesis in addition to MLN64. At present, we are conducting experiments to elucidate the role of HSP60 in this process. The presence of multiple proteins and steps during proges- terone synthesis in adrenal glands, gonads and placenta, supports the hypothesis, that at least in part, the limiting step in the regula- tion of steroidogenesis is cholesterol transport from the outer to the inner mitochondrial membrane. In this sense, the contact sites between mitochondrial membranes which contains multiple pro- teins, such MLN64 and HSP60, could play a major role for choles- terol transfer to P450scc [78,79]. The interaction between StAR with a mitochondrial membrane multiprotein complex has been proposed to affect cholesterol transport. These complexes could be the transduceosome complex, which consists of membrane and cytoplasmic proteins [80]; or the steroidogenic metabolon, composed of mitochondrial membrane proteins [80,81]. In human placenta, MLN64 interacts with the steroidogenic contact sites [75] (see below). 7. Mitochondrial contact sites associated with steroidogenesis Mitochondrial contact sites are domains where inner and outer membranes are in close proximity, allowing exchange between them. Contact sites are established and maintained in durable or transient states by different proteins and enzymes. Hackenbrock [82] noted that mitochondria exhibit different types of contact sites: multiprotein complexes that perform various mitochondrial functions. Their assembly takes place through a dynamic process that requires the association of specific proteins and enzymes. According to their protein composition, contact sites may partici- pate in multiple mitochondrial processes, including protein import or progesterone synthesis. Proteins that are found in mitochondrial contact sites also par- ticipate in other processes, including oxidative phosphorylation (adenine nucleotide translocase –ANT–, VDAC, creatine kinase, and hexokinase [83,84]; or apoptosis (ANT, VDAC, cyclophilin D, and hexokinase II). The outer membrane contains proteins that are able to interact with actin and establish ATP-dependent junc- tions, and antiapoptotic proteins of the Bcl-2 family that, when overexpressed, block apoptosis [85–87]. Another example is the permeability transition pore [88], a multiprotein complex that forms non-selective pores in the inner membrane and is associated with structural components as the ANT, cyclophilin D and VDAC. TPSO has been identified in association with the outer membrane pore; creatine kinase with the intermembrane space; and hexoki- nase II with VDAC in the outer membrane, and with the proteins Bax/Bcl-2 [55,87]. It has been suggested that in steroidogenesis by the adrenal glands and gonads, the contact sites are constituted by various proteins such as voltage-dependent anion channel (VDAC) [89,90], the r-1 receptor [91], an ATPase-ATD3 [92], TPSO, ANT, IP3R (ER-resident inositol triphosphate receptor), Mfn1 and Mfn2 (mitofusine 1 y 2), among others [55,93]. Outer membranal proteins (i.e., VDAC1 and TSPO) associated with inner membranal proteins (i.e., ATPase-ATAD3a) constitute the core of a complex that regulates mitochondrial cholesterol import in adrenal glands and gonads [51,94]. Recent data indicate that although ANT is part of the contact sites, apparently, it is not essential for cholesterol transport [94]. The association between the proteins Tom22, Tim23, Tim 50, and 3b HSD2 has been shown to affect steroidoge- nesis [95,96]. Alternatively, some elements of this complex could interact with the endoplasmic reticulum, lipid droplets or late endosomes [16,80,97,98]. 8. Steroidogenic contact sites in human syncytiotrophoblast mitochondria Mitochondrial contact sites in syncytiotrophoblast have been isolated and their composition determined [99]. They have been proposed to be the conduit for cholesterol flux from the outer to the inner membrane. In the human syncytiotrophoblast, it has been proposed that cholesterol flux is accomplished through a series of steps that involve various intracellular organelles associ- ated with lipid droplets [75] and proteins from contact sites like HSP60 [77], ATP-diphosphohydrolase [100,101], and MLN64 [75]. ATP-diphosphohydrolase activity supplies energy for cholesterol flux during progesterone synthesis (see below). HSP60 is a mitochondrial chaperone that has been associated with MLN64 in progesterone production [77]. Other proteins or enzymes that are associated with placental steroidogenic contact sites are VDAC, ANT, ETC-P450scc, 3b HSD1, NADP-dependent isocitrate dehydrogenase [102], and the catalytic subunit of PKA [103]. The steroidogenic machinery reported in adrenal glands includes TPSO [16]. Although in human term placental explants specific ligands for peripheral benzodi- azepine binding sites (PBzS) caused an increase of progesterone and estradiol-17b secretion [104], and it has been also reported that maternal obesity during pregnancy negatively regulates mito- chondrial TSPO, which impairs mitochondrial steroidogenesis [105], the role and localization of TSPO remains to be clarified. In fact, the effect of ligands of PBzS on steroidogenesis was observed only in intact placental explants (where plasmamembrane, cytosol and the whole signal transduction machinery were present) [104], and the identification of TSPO in mitochondria was performed on frozen placental tissue where the intactness of cellular and mito- chondrial architecture is not assured [105]. We isolated highly purified syncytiotrophoblast mitochondria from fresh placentas [6]. These syncytiotrophoblast mitochondria retain their steroido- genic and bioenergetics functions, and the TSPO protein was not present, nor did ligands of PBzS have any effect on steroidogenesis [106]. Similarly, in isolated mitochondria from BeWo cells, TSPO protein was absent [77,106]. F. Martinez et al. / Steroids 103 (2015) 11–22 15 Steroidogenic mitochondrial contact sites from syncytiotro- phoblast contain cholesterol and the complete machinery for pro- gesterone synthesis: STARD3, which promotes cholesterol flux; ETC-P450scc, which transforms cholesterol into pregnenolone; 3bHSD1, which transforms pregnenolone into progesterone; and NADP-isocitrate dehydrogenase, which supplies the energy for ETC-P450scc activity (Fig. 3). Thus, the contact sites synthesize progesterone [99,102]. Interestingly, in isolated mitochondrial contact sites, Dwm was not produced, and therefore, its role in steroidogenesis must be re-assessed, leaving NADPH and ATP as the principal energy sources [99]. In addition to the description of proteins from steroidogenic contact sites, we have discovered the presence of diverse SNARE elements associated with mitochondrial outer membranes [75], and put forward a model for MLN64 in human placenta mitochon- dria [75] in which lipid droplets containing SNARE protein complexes [74,107–110] (i.e., a-SNAP, syntaxin 3, 7 y 12, syntaxin-Binding Protein-2, syntaxin-Binding Protein-3, and VAMP-8) probably participate in cholesterol transport. It has been demonstrated that SNAP promotes interactions between lipid dro- plets and mitochondria [111] and that steroidogenic cells express syntaxin-17, SNAP-23 and SNAP-25 [112–116]. These observations suggest that SNARE proteins could participate in cholesterol trans- port by promoting functional interactions among lipid droplets, the endoplasmic reticulum, endosomes and mitochondria. In syncy- tiotrophoblast cells, MLN64 could be incorporated into mitochon- dria through these complexes and then be activated by mitochondrial proteases [75], favoring its uptake into mitochon- dria (i.e., contact sites) [46,79], and promoting cholesterol trans- port [77,102]. 9. Architecture and physiological characteristics of syncytiotrophoblast mitochondria As it has been mentioned earlier, differentiation of the syncy- tiotrophoblast involves mitochondrial cristae remodeling. This mitochondrial remodeling involves turning large round Fig. 3. Model proposed for progesterone synthesis in human syncytiotrophoblast mitochondria. Progesterone synthesis has been divided into seven steps. (1) Anchoring of lipid droplets or late endosomes to mitochondria via SNARE complexes. STARD3 is attached to lipid droplets or late endosomes. (2) Proteolytic transformation of STARD3 from a 55-kDa into a 28-kDa protein by a metalloprotease that exerts its activity on the cytosolic side of the outer membrane of mitochondria. This metalloprotease is sensitive to 1,10-phenantroline, EGTA, or EDTA. (3) The STARD3–28 kDa protein, which contains the cholesterol binding domain (START), is associated with steroidogenic contact sites, and increases cholesterol flux from the outer into the inner mitochondrial membrane, where it reaches the machinery of cytochrome P450scc and increases progesterone production (step 4). If protease activity is inhibited and no STARD3-28 kDa protein is produced, no progesterone synthesis occurs. (5) Dwm induces ATP hydrolysis by ATP- diphosphohydrolase during cholesterol transport throughout the steroidogenic contact sites. ADP can be hydrolyzed to AMP by ATP-diphosphohydrolase or can enter the mitochondrial matrix for ATP synthesis by F1F0-ATP synthase, stimulating oxygen uptake and the proton pump by the ETC. (6) NADP-dependent isocitrate dehydrogenase produces NADPH to supply energy to the ETC associated to P450scc constituted by adrenodoxin and adrenodoxin reductase. (7) PKA activity stimulates progesterone synthesis. Several syncytiotrophoblast mitochondrial proteins are suitable substrates for PKA and perhaps for other kinases. PKA maintains a major pool of inactive phosphorylated phosphatases, leaving few active phosphatases that remove the phosphate from the amino acids that belong to RRXT/S motifs. PKA inhibition by H89 produced a decrease in progesterone synthesis. Steroidogenic contact sites, marked with a dashed line, are constituted by different proteins including adenine nucleotide translocase (ANT), voltage dependent anionic channel (VDAC); ATP-diphosphohydrolase (ATP-D), NADP-dependent isocitrate dehydrogenase (ICD), cytochrome P450scc (P450scc) and type II 3b-hydroxysteroid dehydrogenase (3bHSD2). RRXT/S-specific motifs recognized by PKA; Y/T/S-tyrosine, threonine and serine residues; OM = outer mitochondrial membrane; IM = inner mitochondrial membrane; Chol = cholesterol. 16 F. Martinez et al. / Steroids 103 (2015) 11–22 mitochondria with lamellar cristae with an orthodox configuration from cytotrophoblasts, into small irregular structure exhibiting protuberances in the outer and inner membranes, a condensed matrix, and cristae composed of vesicular regions connected by narrow tubules observed in syncytiotrophoblast [6,59,99,117]. We have demonstrated that the acquisition of steroidogenic activity in syncytiotrophoblast mitochondria involves changes in F1F0-ATP synthase dimerization [117]. The F1F0-ATP synthase is present as a native functional dimer assembled into long rows of oligomers in the mitochondrial membrane [118]. This constitutive self-association promotes membrane curvature and the formation of the classical mitochondrial cristae ultrastructure [118–121]. The dimeric structure of F1F0-ATP synthase was solved by transmission electron microscope [119–125]. This supercomplex had a critical role in maintaining a high transmembranal potential that ensures optimal conditions for efficient ATP synthesis [126]. Syncytiotrophoblast mitochondria contain low levels of dimeric complex V, in contrast to cytotrophoblast mitochondria, which have standard cristae morphology and a higher content of the ATP synthase dimer [117]. The proportion of F1F0-ATP synthase dimer that can be extracted from mitochondria was assessed through Blue Native-PAGE and WB analysis. It was demonstrated that cytotrophoblast cells contain a higher dimer/monomer ratio than syncytiotrophoblast cells [117], and that the ATP synthase supercomplex is the consequence of a higher relative amount of IF1 in the cytotrophoblasts [117]. Syncytiotrophoblast mitochondria exhibit the same morphol- ogy (i.e., tubular, vesicular or tubulovesicular cristae [117]) like those of other steroidogenic tissues such as Leydig cells [127]. In contrast, electron microscopic tomography studies of liver [128,129], neuronal [130,131], brown adipose tissue [132], fungi [133,134], rods and cones [135] mitochondria show typical lamel- lar cristae architecture. It has been suggested that cristae morphology has a direct impact on ATP production in Leydig cells, since the narrow gap between lamellae prevents the correct distribution of the F1 sub- unit of ATP synthase [127]. Allen et al. [136] have shown that mito- chondrial membrane potential (Dwm), mitochondrial ATP synthesis, and mitochondrial respiration are required to support Leydig cell steroidogenesis [136]. In contrast, the rate of ATP syn- thesis in isolated human cytotrophoblast and syncytiotrophoblast mitochondria is similar (151 ± 16 nmol/mg/min and 153 ± 13 nmol/mg/min, respectively), while progesterone synthesis is ten times higher in syncytiotrophoblast than cytotrophoblast mitochondria (35.7 ± 0.90 ng of progesterone/mg/min and 3.6 ± 1.34 ng of progesterone/mg/min, respectively) [117]. This suggests that in human syncytiotrophoblast, ATP and progesterone synthe- sis are quite important to cell physiology. In this sense, the relative amount of ETC-P450scc and classic electron transport chain (ETC) components is different in each steroidogenic tissue. In acutely regulated steroidogenic tissues (i.e., the adrenal glands and gonads) the content of ETC-P450scc is several times higher than classic ETC elements, suggesting that these cells have a quick response to the external signal to activate steroidogenesis. In contrast, in syncy- tiotrophoblast the amount of both electron transfer chain compo- nents is similar [59], indicating that this cell, with its unique mitochondrial architecture, has parity between steroidogenesis and the bioenergetics metabolism. We have assembled a general model for steroidogenic metabo- lism for human syncytiotrophoblast [59,99] (Fig. 3). It has been suggested that the size reduction of mitochondria and the struc- tural changes of their cristae may improve the steroidogenic activ- ity of syncytiotrophoblast cells [6]. If translocation of cholesterol to cytochrome P450scc is the rate-limiting step in steroidogenesis, a larger surface to volume ratio might improve the movement of cholesterol to the inner membrane where cytochrome P450scc is located. A non-orthodox structure may also play a crucial role in maintaining progesterone synthesis [117,137]. 10. Role of Dwm in placental steroidogenesis: regulation of mitochondrial ATP-diphosphohydrolase Human syncytiotrophoblast mitochondria contain proteins and enzymes that participate in steroidogenic metabolism, i.e., an ATP- diphosphohydrolase (P49961-2|ENTP1_HUMAN) in the inner membrane [99], which is involved in progesterone synthesis, mainly in cholesterol transport [100,101]. ATP-diphosphohydro- lase might provide the required energy to drive cholesterol trans- port between the mitochondrial membranes in a similar way to mitochondrial GTPase in the adrenal glands [138]. Detergent solubilized ATP-diphosphohydrolase catalyzes the hydrolysis of tri- (ATP, GTP, UTP, and CTP), and di-phosphonucleosides (ADP, GDP, UDP, and CDP) in a Mg2+-, Ca2+-, or Mn2+-dependent manner with a single rate-limiting step for each hydrolyzed nucleoside [99]. However, in the intact and energized mitochondria, ATP- diphosphohydrolase hydrolyzes mainly ATP [99]. The resulting ADP promotes oxygen uptake and ATP synthesis by F1F0-ATP synthase [99,139]. The activities of ATP-diphosphohydrolase (ATP hydrolysis) and F1F0-ATP synthase (ATP synthesis) are strongly coordinated to avoid a futile cycle and energy dissipation. Mitochondrial inner membrane potential (Dwm) is a central component of mitochondrial metabolism. It provides the driving force for oxidative phosphorylation, for protein or metabolites import, and to regulate the activity of membrane proteins like the ANT [11,140,141]. We have demonstrated that, in syncytiotro- phoblast mitochondria, Dwm regulates ATP-diphosphohydrolase activity [99], particularly its substrate selectivity (i.e., ATP versus GTP preference). Also, ATP-diphosphohydrolase has been associ- ated with cholesterol transport between mitochondrial mem- branes during progesterone synthesis [101]. Dissipation of Dwm by CCCP (carbonyl cyanide m-chlorophenyl-hydrazone) induces ATP-diphosphohydrolase to hydrolyze ATP as well as GTP. Under these conditions, when Dwm decreased, NADPH production by mitochondrial isocitrate dehydrogenase and ATP hydrolysis by ATP-diphosphohydrolase were essential for progesterone synthesis [99]. Studies with mitochondrial steroidogenic contact sites allow the assessment of the role and impact of the ETC function, the electrochemical proton gradient, oxygen consumption, and ATP synthesis on progesterone synthesis [99]. 11. Placental signal transduction machinery Steroid hormones have important functions in mammalian physiology and metabolic regulation through signaling cascades. Protein phosphorylation is the most common regulatory mecha- nism of protein function in cells. These have been widely studied in steroidogenic tissues like the adrenal glands and gonads, whose regulation is associated with changes in cAMP concentrations [142]. The human placenta is dynamically regulated by a wide range of intrinsic and extrinsic factors mediated by several signal trans- duction pathways. The proteins that constitute signaling cascades have been associated with different events, including: myometrial relaxation [143], maintenance of uterine quiescence during preg- nancy [144], myometrium adrenoreceptor coupling to adenylate cyclases (AC) [145], AC isoform expression in human myometrium during pregnancy [146], and high intracellular cAMP levels associ- ated with syncytium formation in human placental tissue [147]. Signaling cascades are perfectly orchestrated and are pivotal in pregnancy. Studying the function and composition of signaling systems is crucial to understanding cellular processes and the F. Martinez et al. / Steroids 103 (2015) 11–22 17 interaction between the placenta, the mother and the fetus. Protein kinases control cell signaling pathways, which consist of sequential steps, initiated by different stimuli that result in the phosphoryla- tion of serine, threonine or tyrosine residues. It is important to study the subcellular locations of the different phosphodiesterase (PDE) families, adenylate cyclases (AC), phosphatases (PP), and kinases associated with protein post-translational modification. The effect of these enzymes on the regulation of intracellular con- centrations of second messengers and on cellular processes is an important subject of study [142,148]. Although mitochondria contain proteins associated with the cell signaling machinery and their own pool of second messengers, the kinetics, targets and effectors remain unknown. Moreover, how proteins can be phosphorylated in the mitochondrial matrix is yet to be elucidated [149]. Second messengers (i.e., cAMP and cGMP) regulate several cellular functions. PDEs (PDE 1-11) regulate intra- cellular levels of cyclic nucleotides, determining the balance between their production and degradation, which leads to the rapid turnoff of the cAMP signal. cAMP cannot permeate the inner mitochondrial membrane, however, a soluble adenylate cyclase (sAC) activated by bicarbonate anion has been described [150,151]. The cellular expression and location of specific PDEs is determined by the local concentration of cyclic nucleotide and the phosphorylation/dephosphorylation mechanism [152]. Acin- Perez et al. [150] have described a mitochondrial signaling cascade where PDE is involved in the phosphorylation/dephosphorylation mechanism, and interacts with sAC, PKA and a cAMPmitochondrial microdomain.cAMP is synthetized by the transmembrane (tmAC) or sAC. The latter is located in the nucleus, mitochondria, micro- tubules, and centrioles. This variable location suggests that cAMP microdomains exist and are generated in response to extracellular or intrinsic signals regulated by PDEs [153,154]. It has been demonstrated that cytosolic sAC can generate second messengers at the site of the complex of A-kinase anchoring proteins (AKAPs)-cAMP-dependent protein kinases is located, removing the membrane-proximal limitation of cAMP generation [155]. Bernatchez et al. [156] demonstrated the differential expression of several adenylyl cyclase isoforms in trophoblast cells from human placenta, according to the stage of differentiation (i.e., cytotrophoblast and syncytiotrophoblast). AC isoforms are subdi- vided into four groups according to the activating or inhibiting involved mechanism: group 1 is stimulated by calcium and calmodulin (I, III and VIII ACs), group 2 is activated by G-protein and protein kinase C (PKC) (II, IV and VII ACs), group 3 is inhibited by low concentrations of calcium (V and VI ACs), and group 4 is inhibited by phosphatase calcineurin (AC IX) [157–159]. ACs exert an important function during differentiation of trophoblast cells, depending on external factors such as PKC phosphorylation, cal- cium/calmodulin concentrations, or the binding of agonists that stimulate cytotrophoblast cells to differentiate into syncytiotro- phoblast. However, the complete signaling pathways, the proteins involved, and upstream activators and downstream effectors, have not been elucidated. Protein kinases play a key role in multiple cellular processes. Akt 1/2/3 (or PKBa/b/c) is a serine/threonine-specific kinase whose activity is to modulate some cellular functions through phosphory- lation of several substrates. Akt1 deficient mice exhibit structural anomalies in the placenta, suggesting fetal growth failure [160]. Another protein kinase that has been identified and expressed abundantly in the human placenta is ERK1/2. ERK2 transgenic mice exhibit defects in trophoblast morphogenesis and development, which induces damage in the vascularization of the fetal labyr- inthine layer [161]. The ERK1/2 cascade is associated with the 11b-HSD2 [162], which is negatively regulated, suggesting an important regulation point for the function of the human placenta and fetal development [163]. MAPK pathways are involved in morphological and functional differentiation of villous trophoblast, and have been implicated in oxidative injury where JNK/p38 and ERK pathways have an impor- tant effect. These data suggest an underlying equilibrium in the oxidative functions that help maintain pregnancy by preventing abortion, and preeclampsia due to oxidative stress [164–166]. Since the system of phosphorylation/dephosphorylation of pro- teins is an important regulatory mechanism in many cellular pro- cesses, the identification of tyrosine and serine/threonine phosphatases is essential. PP2C phosphatase has been described as a negative modulator of MAPK cascades, mainly p38 and JNK, dephosphorylating and inactivating at different levels of the signal- ing pathway [167]. Nevertheless, Daoud’s group [168] demon- strated that ERK1/2 and p38 expression in human trophoblast cells is crucial to mediate the initiation of trophoblast differentia- tion. Other proteins are probably implicated in this process and control trophoblast differentiation and processes that depend on signaling cascades [168]. The identification and subcellular distri- bution of these signaling proteins is essential to prevent damage to the fetus, to develop selective and effective pharmacological methodologies to reduce the incidence of preterm labor and of complications associated with preterm deliveries attributed to pro- gesterone synthesis. It has been hypothesized that several cellular processes are related to alkaline phosphatases. The placenta expresses an alka- line phosphatase named placental alkaline phosphatase (PLAP). It has been described in early and full-term placentae, associated with insulin-like growth factor (IGF) regulation through dephos- phorylation of the IGF binding protein-1 (IGFBP-1) in the mater- nal-fetal interface. Therefore, abnormal regulation of these pathways could lead to complications in fetal growth and preg- nancy [169,170]. The orientation and location of enzymatic systems associated with progesterone synthesis as well as protein phosphorylation/ dephosphorylation, cholesterol transport between mitochondrial membranes, and cholesterol processing by cytochrome P450scc is important to understand mitochondrial function in placental tis- sue, including steroidogenesis. It has been proposed that steroido- genesis is a constitutive process of the human placenta, apparently without a regulatory control mechanism. However, the presence of PKA tightly bound to syncytiotrophoblast mitochondria suggests that a hormonal control mechanism is present [171]. 12. cAMP-mediated signal transduction cascade in progesterone synthesis The most studied transduction pathway in steroidogenic tissues is the one mediated by cAMP-PKA. In Leydig tumor cell line, basal steroidogenesis, independent of cAMP, is lower than 1% compared to the activation through cAMP-PKA dependent pathways induced by LH/hCG [171], however, in isolated primary Leydig cells nor- mally comprises 10–25% of maximally stimulated steroidogenesis [172]; a similar situation is seen in isolated adrenocortical cells [173,174]. The human placenta is considered to be an autonomous/ constitutively active tissue due to the absence of short term fluctuations in maternal progesterone. However, human placental steroidogenesis can be regulated by the luteinizing (LH) and human chorionic gonadotropin (hCG) hormones. hCG is known to bind and activate the G-protein coupled to the luteinizing hormone/hCG receptor, activating the cAMP/PKA pathway, which triggers events related to progesterone synthesis, decidualization and trophoblast differentiation [175]. The role of other key hormones that modulate cAMP levels and the participation of other signal transduction pathways independent of cAMP/PKA 18 F. Martinez et al. / Steroids 103 (2015) 11–22 are still unknown. Insulin, insulin-like growth factor 1, calcitriol, phorbol esters, epidermal growth factor, oestrogens and cytokines (IL-1and TNFa) have been implicated in the stimulation of progesterone synthesis independent of PKA [176–180]. Primary cultures of trophoblast cells, choriocarcinoma derived cell lines and mitochondria isolated from syncytiotrophoblast, have been extensively employed as models of placental functions [6,106,181–183]. In these models steroidogenesis has been assessed through the use of activators of the PKA signaling cas- cade. In BeWo cells, it was observed that progesterone synthesis increases in the presence of the cAMP analog, 8-Br-cAMP. In the presence of H89, a kinase inhibitor widely used to block PKA activity, progesterone synthesis decreases from 70% to 90%, besides increasing phosphorylation of mitochondrial proteins [106]. In order to facilitate phosphorylation of the target substrate, several phosphorylation/dephosphorylation systems have a speci- fic cell distribution that allows kinases to associate with specific organelles. This spatial and temporal modulation of signaling cas- cades can be achieved by the presence of AKAPs, which tether PKA to precise intracellular sites, such as mitochondria [184,185]. A PKA and a phosphotyrosine phosphate D1 (PTPD1) have been reported to be tightly associated with mitochondria isolated from human placenta through AKAP-121. The activity of this kinase is sensitive to H89 and the specific PKA inhibitor, PKI. The reported sub-localization of PKA and AKAP-121 in syncytiotrophoblast is mitochondrial contact sites [186], which are known to participate in progesterone synthesis [102]. Additionally, phosphorylation of proteins in the steroidogenic contact sites was blocked by H89, suggesting that contact sites contain potential PKA substrates. In the same study, it was observed that progesterone synthesis is either inhibited or abolished, depending on the H89 concentration. Also, phosphorylation of intact mitochondrial proteins increases in the presence of this inhibitor, suggesting the participation of other kinases and phosphatases modulated by PKA [187]. The presence of the PKA a catalytic (Ca) subunit has also been detected in the cytoplasm, nucleus and mitochondria of placental cells; PKA Ca subunit is located predominantly in the outer mitochondrial membrane [103]. We have assessed the overall contribution of PKA in proges- terone synthesis and protein phosphorylation in syncytiotro- phoblast mitochondria. We observed by submitochondrial fractionation that the PKA Ca subunit is located in the outer, the inner and the soluble fraction. The Ca subunit of the inner membrane has the highest activity. The bII regulatory (R bII) subunit was identified as the main isoform of PKA distributed in both the outer and inner membranes. This is relevant since the specificity in PKA signaling can be ensured by a differential expression of the R subunits, which have different roles in cell homeostasis. Additionally, the identification of PKA associated with the membrane fraction of placental mitochondria suggests that it could make the signal transduction pathway involved in steroido- genesis more efficient. However, PKA is tightly associated with the mitochondrial membrane fraction, and dissociation of its subunits cannot be induced by holoenzyme activators, such as db-cAMP, as in the classical model of PKA activation. Furthermore, incubation of mitochondria in the presence or absence of cAMP or db-cAMP neither modified progesterone synthesis, nor protein phosphorylation [103]. These data suggest that the human placenta contains a constitutively active PKA as part of a complex signaling cascade where other kinases and phosphatases may be participating to modulate the dynamic phosphorylation and dephosphorylation of mitochondrial proteins to synthesize progesterone during pregnancy. 13. Conclusion and perspectives The human placenta plays a central role during pregnancy because it carries out multiple metabolic functions similar to those performed by the pituitary, ovaries, bowels, liver, and kidneys. The placenta plays an important role in the transport of nutrients, oxy- gen, and waste products from the fetal metabolism; additionally, the placenta synthesizes multiple hormones, progesterone being among the most important because it establishes and maintains the maternal fetal interface required to assure normal pregnancy. Although an effort has been made to understand the metabo- lism of the human placenta, our knowledge of the function of this organ is still not complete. In the future, it will be important to know how placental cells handle cholesterol distribution for its own metabolic needs in the trophoblast cells, and for the fetus. Also, the participation of mitochondria in the distribution of energy to synthesize ATP and/or progesterone should be studied. Finally, the signal transduction with the central participation of mitochon- dria to control cellular processes will be important to elucidate, since this information will inform how the trophoblast maintains the relationship between the mother and fetus leading to mainte- nance of pregnancy to term. Then, a further research should be ori- ented to new methods, including the need for new techniques to study the syncytiotrophoblast in situ using non-invasive methods. Acknowledgements This work was partially supported by Grants IN211912, IN211715 and IN214914 from Dirección General de Asuntos del Personal Académico (DGAPA) from the Universidad Nacional Autó- noma de México and Grant 168025 from CONACYT, México. The authors thanks Dr. Jerome F. Strauss III of Virginia Commonwealth University School of Medicine, for his valuable suggestions and the critical review of the manuscript. References [1] M. Halasz, J. Szekeres-Bartho, The role of progesterone in implantation and trophoblast invasion, J. Reprod. Immunol. 97 (1) (2013) 43–50. [2] K. Benirschke, P. Kaufmann, Pathology of the Human Placenta, Springer- Verlag, New-York, 2000. p. 22–70. [3] B. Eaton, S. Contractor, in: C.W. Redman, I.L. Sargent, P.M. Starkey (Eds.), In vitro Assessment of Trophoblast Receptors and Placental Transport Mechanisms, Blackwell Scientific Publication, London, 1993, pp. 471–503. [4] H. Kliman, J. Nestler, E. Sermasi, J. Sanger, J. Strauss III, Purification, characterization, and in vitro differenciation of cytotrophoblasts from human term placentae, Endocrinology 118 (1986) 1567–1582. [5] N. Cherradi, G. Defaye, E.M. Chambaz, Characterization of the 3 beta- hydroxysteroid dehydrogenase activity associated with bovine adrenocortical mitochondria, Endocrinology 134 (3) (1994) 1358–1364. [6] F. Martinez, M. Kiriakidou, J.F. Strauss III, Structural and functional changes in mitochondria associated with trophoblast differentiation: methods to isolate enriched preparations of syncytiotrophoblast mitochondria, Endocrinology 138 (5) (1997) 2172–2183. [7] C. Brand, N. Cherradi, G. Defaye, A. Chinn, E.M. Chambaz, J.J. Feige, et al., Transforming growth factor beta1 decreases cholesterol supply to mitochondria via repression of steroidogenic acute regulatory protein expression, J. Biol. Chem. 273 (11) (1998) 6410–6416. [8] S. Paul, B.L. Jailkhani, G.P. Talwar, Isolation and functional maintenance in culture of syncytiotrophoblasts from human placenta, Indian J. Exp. Biol. 16 (12) (1978) 1226–1235. [9] J. Kaiser, Gearing up for a closer look at the human placenta, Science 344 (6188) (2014) 1073. [10] C.J. Kim, Congenital lipoid adrenal hyperplasia, Ann. Pediatr. Endocrinol. Metab. 19 (2014) 179–183. [11] D.W. Nebert, K. Wikvall, W.L. Miller, Human cytochromes P450 in health and disease, Philos. Trans. R. Soc. B 368 (2013) 20120431. [12] B.J. Clark, The mammalian START domain protein family in lipid transport in health and disease, J. Endocrinol. 212 (2012) 257–275. [13] G. Daum, J.E. Vance, Import of lipids into mitochondria, Prog. Lipid Res. 36 (2–3) (1997) 103–130. F. Martinez et al. / Steroids 103 (2015) 11–22 19 [14] R.E. Soccio, J.L. Breslow, StAR-related lipid transfer (START) proteins: mediators of intracellular lipid metabolism, J. Biol. Chem. 278 (25) (2003) 22183–22186. [15] S.C.J. Helle, G. Kanfer, K. Kolar, A. Lang, A.H. Michel, B. Kornmann, Organization and function of membrane contact sites, Biochim. Biophys. Acta 1833 (2013) 2526–2541. [16] V. Papadopoulos, W.L. Miller, Role of mitochondria in steroidogenesis, Best Pract. Res. Clin. Endocrinol. Metab. 26 (6) (2012) 771–790. [17] S.W. Tait, D.R. Green, Mitochondria and cell signalling, J. Cell Sci. 125 (2012) 807–815. [18] A. Carlucci, L. Lignitto, A. Feliciello, Control of mitochondria dynamics and oxidative metabolism by cAMP, AKAPs and the proteasome, Trends Cell Biol. 18 (12) (2008) 604–613. [19] L. Galluzzi, O. Kepp, G. Kroemer, Mitochondria: master regulators of danger signaling, Nat. Rev. Mol. Cell Biol. 13 (12) (2012) 780–788. [20] J.T. Gwynne, J.F. Strauss 3rd., The role of lipoproteins in steroidogenesis and cholesterol metabolism in steroidogenic glands, Endocr. Rev. 3 (3) (1982) 299–329. [21] L. Zelewski, C.A. Villee, The biosynthesis of squalene, lanosterol, and cholesterol by minced human placenta, Biochemistry 5 (6) (1966) 1805– 1814. [22] H. van Leusden, C.A. Villee, The de novo synthesis of sterols and steroids-from acetate by preparations of human term placenta, Steroids 6 (1965) 31–45. [23] R.S. Mathur, D.F. Archer, N. Wiqvist, E. Diczfalusy, Quantitative assessment of the de novo sterol and steroid synthesis in the human foeto-and secretion of cholesterol and cholesterol sulphate, Acta Endocrinol. (Copenh) 65 (4) (1970) 663–674. [24] H.A. van Leusden, C. Thomas, G. Telegdy, E. Diczfalusy, Lack of squalene and lanosterol synthesis by the perfused human placenta at term, J. Steroid Biochem. 4 (4) (1973) 349–354. [25] C.A. Winkel, J.M. Snyder, P.C. MacDonald, E.R. Simpson, Regulation of cholesterol and progesterone synthesis in human placental cells in culture by serum lipoproteins, Endocrinology 106 (4) (1980) 1054–1060. [26] C.A. Winkel, P.C. MacDonald, P.G. Hemsell, E.R. Simpson, Regulation of cholesterol metabolism by human trophoblastic cells in primary culture, Endocrinology 109 (4) (1981) 1084–1090. [27] N.M. Gude, C.T. Roberts, B. Kalionis, R.G. King, Growth and function of the normal human placenta, Thromb. Res. 114 (5–6) (2004) 397–407. [28] M. Furuhashi, H. Seo, S. Mizutani, O. Narita, Y. Tomoda, N. Matsui, Expression of low density lipoprotein receptor gene in human placenta during pregnancy, Mol. Endocrinol. 3 (8) (1989) 1252–1256. [29] C. Wadsack, A. Hammer, S. Levak-Frank, G. Desoye, K.F. Kozarsky, B. Hirschmugl, et al., Selective cholesteryl ester uptake from high density lipoprotein by human first trimester and term villous trophoblast cells, Placenta 24 (2–3) (2003) 131–143. [30] T.A. Pagler, S. Golsabahi, M. Doringer, S. Rhode, G.J. Schutz, M. Pavelka, et al., A Chinese hamster ovarian cell line imports cholesterol by high-density lipoprotein degradation, J. Biol. Chem. 281 (50) (2006) 38159–38171. [31] C. Röhrl, H. Stangl, HDL endocytosis and resecretion, Biochim. Biophys. Acta 1831 (11) (2013) 1626–1633. [32] C.A. Winkel, P.C. MacDonald, E.R. Simpson, The role of receptor-mediated low-density lipoprotein uptake and degradation in the regulation of progesterone biosynthesis and cholesterol metabolism by human trophoblasts, Placenta Suppl. 3 (1981) 133–143. [33] J.T. Sanderson, Placental and fetal steroidogenesis, in: J. Lafond, C. Vaillancourt (Eds.), Human Embryogenesis: Methods and Protocols, vol. 550, Tennessee: Humana Press, a part of Springer Science + Business Media, 2009, pp. 127–136. [34] R. Zamel, R. Khan, R.L. Pollex, R.A. Hegele, Abetalipoproteinemia: two case reports and literature review, Orphanet J. Rare Dis. 8 (3) (2008) 19. [35] L.A. Woollett, Review: transport of maternal cholesterol to the fetal circulation, Placenta 32 (Suppl. 2) (2011) S218–S221. [36] J. Bhattacharjee, F. Ietta, R. Romagnoli, N. Bechi, I. Caniggia, L. Paulesu, ABC transporters in human placenta and their role in maternal-fetal cholesterol transfer: ABCA1 candidate target, in: Jing Zheng (Ed.), Recent Advances in Research on the Human Placenta, In Tech, Croatia, 2012, pp. 336–354. [37] T. Plösch, E.M.E. van Straten, F. Kuipers, Cholesterol transport by the placenta: placental liver X receptor activity as a modulator of fetal cholesterol metabolism?, Placenta 28 (7) (2007) 604–610 [38] M.E. Baardman, W.S. Kerstjens-Frederikse, R.M.F. Berger, M.K. Bakker, R.M.W. Hofstra, T. Plösch, The role of maternal-fetal cholesterol transport in early fetal life: current insights, Biol. Reprod. 88 (1) (2013) 24 (1–9). [39] C.D. Klaassen, L.M. Aleksunes, Xenobiotic, bile acid, and cholesterol transporters: function and regulation, Pharmacol. Rev. 62 (1) (2010) 1–96. [40] G. Desoye, M. Gauster, C. Wadsack, Placental transport in pregnancy pathologies, Am. J. Clin. Nutr. 94 (Suppl. 6) (2011) 1896S–1902S. [41] B. Kornmann, The molecular hug between the ER and the mitochondria, Curr. Opin. Cell Biol. 25 (4) (2013) 443–448. [42] A. Toulmay, W.A. Prinz, Lipid transfer and signaling at organelle contact sites: the tip of the iceberg, Curr. Opin. Cell Biol. 23 (4) (2011) 458–463. [43] M. Fujimoto, T. Hayashi, T.P. Su, The role of cholesterol in the association of endoplasmic reticulum membranes with mitochondria, Biochem. Biophys. Res. Commun. 417 (1) (2012) 635–639. [44] V.K. Khora, W.J. Shena, F.B. Kraemer, Lipid droplet metabolism, Curr. Opin. Clin. Nutr. Metab. Care 16 (6) (2013) 632–637. [45] M. Scharwey, T. Tatsuta, T. Langer, Mitochondrial lipid transport at a glance, J. Cell Sci. 126 (Pt 23) (2013) 5317–5323. [46] L. Issop, M.B. Rone, V. Papadopoulos, Organelle plasticity and interactions in cholesterol transport and steroid biosynthesis, Mol. Cell. Endocrinol. 371 (1– 2) (2013) 34–46. [47] F.B. Kraemer, V.K. Khor, W.J. Shen, S. Azhar, Cholesterol ester droplets and steroidogenesis, Mol. Cell. Endocrinol. 371 (1–2) (2013) 15–19. [48] T. Tatsuta, M. Scharwey, T. Langer, Mitochondrial lipid trafficking, Trends Cell Biol. 24 (1) (2014) 44–52. [49] P.F. Hall, G. Almahbobi, The role of the cytoskeleton in the regulation of steroidogenesis, J. Steroid Biochem. Mol. Biol. 43 (8) (1992) 769–777. [50] M.B. Sewer, D. Li, Regulation of steroid hormone biosynthesis by the cytoskeleton, Lipids 43 (12) (2008) 1109–1115. [51] G. Almahbobi, L.J. Williams, P.F. Hall, Attachment of mitochondria to intermediate filaments in adrenal cells: relevance to the regulation of steroid synthesis, Exp. Cell Res. 200 (1992) 361–369. [52] G. Almahbobi, L.J. Williams, P.F. Hall, Attachment of steroidogenic lipid droplets to intermediate filaments in adrenal cells, J. Cell Sci. 101 (1992) 383– 393. [53] F. Alpy, A. Rousseau, Y. Schwab, F. Legueux, I. Stoll, C. Wendling, et al., STARD3/STARD3NL and VAP make a novel molecular tether between late endosomes and the ER, J. Cell Sci. 126 (Pt 23) (2013) 5500–5512. [54] R.C. Tuckey, H.S. Bose, I. Czerwionka, W.L. Miller, Molten globule structure and steroidogenic activity of N-218 MLN64 in human placental mitochondria, Endocrinology 145 (4) (2004) 1700–1707. [55] L.A. Martin, B.E. Kennedy, B. Karten, Mitochondrial cholesterol: mechanisms of import and effects on mitochondrial function, J. Bioenerg. Biomembr. (2014), http://dx.doi.org/10.1007/s10863-014-9592-6. [56] S.E. Horvath, G. Daum, Lipids of mitochondria, Prog. Lipid Res. 52 (4) (2013) 590–614. [57] M.T. Espinosa-Garcia, J.F. Strauss 3rd, F. Martinez, A trypsin-sensitive protein is required for utilization of exogenous cholesterol for pregnenolone synthesis by placental mitochondria, Placenta 21 (2000) 654–660. [58] J. Navarrete, O. Flores-Herrera, A. Uribe, F. Martínez, Differences in cholesterol incorporation into mitochondria from hepatoma AS-30D and human term placenta, Placenta 20 (4) (1999) 285–291. [59] F. Martinez, R. Milan, O. Flores-Herrera, S. Olvera-Sanchez, E. Gomez-Chang, M.T. Espinosa-Garcia, Espinosa-Garcia, The role of mitochondria in syncytiotrophoblast cells: bioenergetics and steroidogenesis, in: J. Zheng (Ed.), Recent Advances in Research on the Human Placenta, In Tech, Croatia, 2012, pp. 397–428. [60] R.C. Tuckey, M.J. Headlam, H.S. Bose, W.L. Miller, Transfer of cholesterol between phospholipid vesicles mediated by the steroidogenic acute regulatory protein (StAR), J. Biol. Chem. 277 (49) (2002) 47123–47128. [61] D. Lin, T. Sugawara, J.F. Strauss 3rd, B.J. Clark, D.M. Stocco, P. Saenger, A. Rogol, W.L. Miller, Role of steroidogenic acute regulatory protein in adrenal and gonadal steroidogenesis, Science 267 (5205) (1995) 1828–1831. [62] P.R. Manna, M.T. Dyson, D.M. Stocco, Regulation of the steroidogenic acute regulatory protein gene expression: present and future perspectives, Mol. Hum. Reprod. 15 (6) (2009) 321–333. [63] H.A. LaVoie, S.R. King, Transcriptional regulation of steroidogenic genes: STARD1, CYP11A1 and HSD3B, Exp. Biol. Med. (Maywood). 234 (8) (2009) 880–907. [64] F. Alpy, C. Tomasetto, START ships lipids across interorganelle space, Biochimie 96 (2014) 85–95. [65] F. Arakane, S.R. King, Y. Du, C.B. Kallen, L.P. Walsh, H. Watari, et al., Phosphorylation of steroidogenic acute regulatory protein (StAR) modulates its steroidogenic activity, J. Biol. Chem. 272 (51) (1997) 32656–32662. [66] G. Sasaki, M. Zubair, T. Ishii, T. Mitsui, T. Hasegawa, R.J. Auchus, The contribution of serine 194 phosphorylation to steroidogenic acute regulatory protein function, Mol. Endocrinol. 28 (7) (2014) 1088–1096. [67] F. Arakane, C.B. Kallen, H. Watari, J.A. Foster, N.B. Sepuri, D. Pain, et al., The mechanism of action of steroidogenic acute regulatory protein (StAR). StAR acts on the outside of mitochondria to stimulate steroidogenesis, J. Biol. Chem. 273 (26) (1998) 16339–16345. [68] M. Rajapaksha, J. Kaur, M. Bose, R.M. Whittal, H.S. Bose, Cholesterol-mediated conformational changes in the steroidogenic acute regulatory protein are essential for steroidogenesis, Biochemistry 52 (41) (2013) 7242–7253. [69] E.R. Simpson, P.C. MacDonald, Endocrine physiology of the placenta, Annu. Rev. Physiol. 43 (1981) 163–188. [70] H.S. Bose, R.M. Whittal, M.C. Huang, M.A. Baldwin, W.L. Miller, N-218MLN64, a protein with StAR-like steroidogenic activity, is folded and cleaved similarly to StAR, Biochemistry 39 (2000) 11722–11731. [71] C. Tomasetto, C. Regnier, C. Moog-Lutz, M.G. Mattei, M.P. Chenard, R. Lidereau, et al., Identification of four novel human genes amplified and overexpressed in breast carcinoma and localized to the q11–q21.3 region of chromosome 17, Genomics 28 (3) (1995) 367–376. [72] R. van der Kant, I. Zondervan, L. Janssen, J. Neefjes, Cholesterol-binding molecules MLN64 and ORP1L mark distinct late endosomes with transporters ABCA3 and NPC1, J. Lipid Res. 54 (8) (2013) 2153–2165. [73] M. Charman, B.E. Kennedy, N. Osborne, B. Karten, MLN64 mediates egress of cholesterol from endosomes to mitochondria in the absence of functional Niemann-Pick Type C1 protein, J. Lipid Res. 51 (5) (2010) 1023–1034. [74] J. Hu, Z. Zhang, W.J. Shen, S. Azhar, Cellular cholesterol delivery, intracellular processing and utilization for biosynthesis of steroid hormones, Nutr. Metab. (Lond) 7 (2010) 47. 20 F. Martinez et al. / Steroids 103 (2015) 11–22 [75] M. Esparza-Perusquía, S. Olvera-Sánchez, O. Flores-Herrera, H. Flores-Herrera, A. Guevara-Flores, J.P. Pardo, et al., Mitochondrial proteases act on STARD3 to activate progesterone synthesis in human syncytiotrophoblast, Biochim. Biophys. Acta 1850 (1) (2015) 107–117. [76] T. Kishida, I. Kostetskii, Z. Zhang, F. Martinez, P. Liu, S.U. Walkley, et al., Targeted mutation of the MLN64 START domain causes only modest alterations in cellular sterol metabolism, J. Boil. Chem. 279 (18) (2004) 19276–19285. [77] S. Olvera-Sanchez, M.T. Espinosa-Garcia, J. Monreal, O. Flores-Herrera, F. Martinez, Mitochondrial heat shock protein participates in placental steroidogenesis, Placenta 32 (3) (2011) 222–229. [78] J.F. Strauss 3rd, T. Kishida, L.K. Christenson, T. Fujimoto, H. Hiroi, StAR domain proteins and the intracellular trafficking of cholesterol in steroidogenic cells, Mol. Cell. Endocrinol. 202 (1) (2003) 59–65. [79] W.L. Miller, H.S. Bose, Early steps in steroidogenesis: intracellular cholesterol trafficking, J. Lipid Res. 52 (12) (2011) 2111–2135. [80] J. Fan, V. Papadopoulos, Evolutionary origin of the mitochondrial cholesterol transport machinery reveals a universal mechanism of steroid hormone biosynthesis in animals, PLoS ONE 8 (10) (2013) e76701. [81] A.T. Slominski, T.K. Kim, J. Chen, M.N. Nguyen, W. Li, C.R. Yates, et al., Cytochrome P450scc-dependent metabolism of 7-dehydrocholesterol in placenta and epidermal keratinocytes, Int. J. Biochem. Cell Biol. 44 (2012) 2003–2018. [82] C.R. Hackenbrock, Chemical and physical fixation of isolated mitochondria in low-energy and high-energy states, Proc. Natl. Acad. Sci. U.S.A. 61 (2) (1968) 598–605. [83] V. Adams, W. Bosch, J. Schelege, T. Wallimann, D. Brdiczka, Further characterization of contact sites from mitochondria of different tissues: topology of peripheral kinases, Biochim. Biophys. Acta 981 (2) (1993) 213– 225. [84] D. Brdiczka, Contact sites between mitochondrial envelope membranes. Structure and function in energy- and protein-transfer, Biochim. Biophys. Acta 1071 (3) (1991) 291–312. [85] J.M. Adams, S. Cory, Life-or-death decisions by the Bcl-2 protein family, Trends Biochem. Sci. 26 (1) (2001) 61–66. [86] N. Zamzami, G. Kroemer, The mitochondrion in apoptosis: how Pandora’s Box opens, Nat. Rev. Mol. Cell Biol. 2 (1) (2001) 67–71. [87] D.F. Suen, K.L. Norris, R.J. Youle, Mitochondrial dynamics and apoptosis, Genes Dev. 22 (12) (2008) 1577–1590. [88] A. Rasola, P. Bernardi, The mitochondrial permeability transition pore and its adaptive responses in tumor cells, Cell Calcium 56 (6) (2014) 437–445. [89] M. Bose, R.M. Whittal, W.L. Miller, H.S. Bose, Steroidogenic activity of StAR requires contact with mitochondrial VDAC1 and phosphate carrier protein, J. Biol. Chem. 283 (14) (2008) 8837–8845. [90] M. Prasad, J. Kaur, K.J. Pawlak, M. Bose, R.M. Whittal, H.S. Bose, Mitochondria associated ER-Membrane (MAM) regulates steroidogenic activity via StAR- VDAC2 interaction, J. Biol. Chem. 290 (5) (2015) 2604–2616. [91] K.S. Marriott, M. Prasad, V. Thapliyal, H.S. Bose, R-1 receptor at the mitochondrial-associated endoplasmic reticulum membrane is responsible for mitochondrial metabolic regulation, J. Pharmacol. Exp. Ther. 343 (3) (2012) 578–586. [92] L. Issop, J. Fan, S. Lee, M.B. Rone, K. Basu, J. Mui, et al., Mitochondria- associated membrane formation in hormone-stimulated Leydig cell steroidogenesis: role of ATAD3, Endocrinology 1 (1) (2015) 334–345. [93] C. Giatzakis, A. Batarseh, L. Dettin, V. Papadopoulos, The role of Ets transcription factors in the basal transcription of the translocator protein (18 kDa), Biochemistry 46 (16) (2007) 4763–4774. [94] M.B. Rone, A.S. Midzak, L. Issop, G. Rammouz, S. Jagannathan, J. Fan, et al., Identification of a dynamic mitochondrial protein complex driving cholesterol import, trafficking and metabolism to steroid hormones, Mol. Endocrinol. 26 (11) (2012) 1868–1882. [95] J. Nicholas, N.J. Hoogenraad, L.A. Ward, M.T. Ryan, Import and assembly of proteins into mitochondria of mammalian cells, Biochim. Biophys. Acta 1592 (1) (2002) 97–105. [96] J. Kevin, K.J. Pawlak, M. Prasad, J.L. Thomas, R.M. Whittal, H.S. Bose, Inner mitochondrial translocase Tim50 interacts with 3b-hydroxysteroid dehydrogenase type 2 to regulate adrenal and gonadal steroidogenesis, J. Biol. Chem. 286 (45) (2011) 39130–39140. [97] M. Beller, K. Thiel, P.J. Thul, H. Jäckle, Lipid droplets: a dynamic organelle moves into focus, FEBS Lett. 584 (11) (2010) 2176–2182. [98] H. Yang, A. Galea, V. Sytnyk, M. Crossley, Controlling the size of lipid droplets: lipid and protein factors, Curr. Opin. Cell Biol. 24 (4) (2012) 509–516. [99] O. Flores-Herrera, S. Olvera-Sánchez, M. Esparza-Perusquía, J.P. Pardo, J.L. Rendón, et al., Membrane potential regulates mitochondrial ATP- diphosphohydrolase activity but is not involved in progesterone biosynthesis in human syncytiotrophoblast cells, Biochim. Biophys. Acta 1847 (2) (2015) 143–152. [100] O. Flores-Herrera, A. Uribe, J.P. Pardo, J.L. Rendón, F. Martínez, A novel ATP- diphosphohydrolase from human term placental mitochondria, Placenta 20 (5–6) (1999) 475–484. [101] O. Flores-Herrera, A. Uribe, C. Garcıa-Perez, R. Milan, F. Martinez, 50-p- fluorosulfonylbenzoyl adenosine inhibits progesterone synthesis in human placental mitocondria, Biochim. Biophys. Acta 1585 (1) (2002) 11–18. [102] A. Uribe, J.F. Strauss 3rd, F. Martínez, Contact sites from human placental mitochondria: characterization and role in progesterone synthesis, Arch. Biochem. Biophys. 413 (2) (2003) 172–181. [103] E. Gomez-Chang, M.T. Espinosa-Garcia, S. Olvera-Sanchez, O. Flores-Herrera, F. Martinez, PKA tightly bound to human placental mitochondria participates in steroidogenesis and is not modified by cAMP, Placenta 35 (9) (2014) 748– 762. [104] E.R. Barnea, F. Fares, M. Gavish, Modulatory action of benzodiazepines on human term placental steroidogenesis in vitro, Mol. Cell. Endocrinol. 64 (2) (1989) 155–159. [105] L Lassance et al., Obesity-induced down-regulation of the mitochondrial translocator protein (TSPO) impairs placental steroid production, J. Clin. Endocrinol. Metab. 100 (2015) E11–E18. [106] M.G. Maldonado-Mercado, M.T. Espinosa-García, C. Gómez-Concha, J. Monreal-Flores, F. Martinez, Steroidogenesis in BeWo cells: role of protein kinase A and benzodiazepines, Int. J. Biochem. Cell Biol. 40 (5) (2008) 901– 908. [107] R. Jahn, R.H. Scheller, SNAREs-engines for membrane fusion, Nat. Rev. Mol. Cell Biol. 7 (9) (2006) 631–643. [108] P. Boström, L. Andersson, M. Rutberg, J. Perman, U. Lidberg, B.R. Johansson, et al., SNARE proteins mediate fusion between cytosolic lipid droplets and are implicated in insulin sensitivity, Nat. Cell Biol. 9 (11) (2007) 1286– 1293. [109] S. Martens, H.T. McMahon, Mechanisms of membrane fusion: disparate players and common principals, Nat. Rev. Mol. Cell Biol. 9 (7) (2009) 543– 556. [110] T.C. Sühof, J.E. Rothman, Membrane fusion: grappling with SNARE and SM proteins, Science 323 (5913) (2009) 474–477. [111] M.S. Jägerström, S. Polesie, Y. Wickström, B.R. Johansson, H.D. Schroder, K. Højlund, et al., Lipid droplets interact with mitochondria using SNAP23, Cell Biol. Int. 33 (9) (2009) 934–940. [112] N.J. Grant, R. Hepp, W. Krause, D. Aunis, P. Oehme, K. Langley, Differential expression of SNAP-25 isoforms and SNAP-23 in the adrenal gland, J. Neurochem. 72 (1) (1999) 363–372. [113] M. Steegmaier, V. Oorschot, J. Klumperman, R.H. Scheller, Syntaxin 17 is abundant in steroidogenic cells and implicated in smooth endoplasmic reticulum membrane dynamics, Mol. Biol. Cell 11 (8) (2000) 2719–2731. [114] J. Grosse, A. Bulling, C. Brucker, U. Berg, A. Amsterdam, A. Mayerhofer, et al., Synaptosome-associated protein of 25 kilodaltons in oocytes and steroid- producing cells of rat and human ovary: molecular analysis and regulation by gonadotropins, Biol. Reprod. 63 (2) (2000) 643–650. [115] M. Jo, M.C. Gieske, C.E. Payne, S.E. Wheeler-Price, J.B. Gieske, I.V. Ignatius, et al., Development and application of a rat ovarian gene expression database, Endocrinology 145 (11) (2004) 5384–5396. [116] M. Shimada, Y. Yanai, T. Okazaki, Y. Yamashita, V. Sriraman, W.C. Wilson, et al., Snaptosomal-associated protein 25 gene expression is hormonally regulated during ovulation and is involved in cytokine/chemokine exocytosis from granulose cells, Mol. Endocrinol. 21 (10) (2007) 2487–2502. [117] D. De los Rios Castillo, M. Zarco-Zavala, S. Olvera-Sanchez, J.P. Pardo, O. Juarez, F. Martinez, et al., Atypical cristae morphology of human syncytiotrophoblast mitochondria: role for complex V, J. Biol. Chem. 286 (27) (2011) 23911–23919. [118] M. Strauss, G. Hofhaus, R.R. Schröder, W. Kühlbrandt, Dimer ribbons of ATP synthase shape the inner mitochondrial membrane, EMBO J. 27 (7) (2008) 1154–1160. [119] R.D. Allen, C.C. Schroeder, A.K. Fok, An investigation of mitochondrial inner membranes by rapid-freeze deep-etch techniques, J. Cell Biol. 108 (6) (1989) 2233–2240. [120] R.D. Allen, Membrane tubulation and proton pumps, Protoplasma 189 (1995) 1–8. [121] P. Paumard, J. Vaillier, B. Coulary, J. Schaeffer, V. Soubannier, D.M. Mueller, et al., The ATP synthase is involved in generating mitochondrial cristae morphology, EMBO J. 21 (3) (2002) 221–230. [122] H. Yao, R.A. Stuart, S. Cai, D.S. Sem, Structural characterization of the transmembrane domain from subunit e of yeast F1Fo-ATP synthase: a helical GXXXG motif located just under the micelle surface, Biochemistry 47 (7) (2008) 1910–1917. [123] J. Velours, A. Dautant, I. Sagot, D. Brethes, Mitochondrial F1F0-ATP synthase and organellar internal architecture, Int. J. Biochem. Cell Biol. 41 (10) (2009) 1783–1789. [124] K.M. Davies, M. Strauss, B. Daum, J.H. Kief, H.D. Osiewacz, A. Rycovska, et al., Macromolecular organization of ATP synthase and complex I in whole mitochondria, Proc. Natl. Acad. Sci. U.S.A. 108 (34) (2011) 14121–14126. [125] F. Minauro-Sanmiguel, S. Wilkens, J.J. Garcia, Structure of dimeric mitochondrial ATP synthase: novel F0 bridging features and the structural basis of mitochondrial cristae biogenesis, Proc. Natl. Acad. Sci. U.S.A. 102 (35) (2005) 12356–12358. [126] C. Bornhövd, F. Vogel, W. Neupert, A.S. Reichert, Mitochondrial membrane potential is dependent on the oligomeric state of F1F0-ATP synthase supracomplexes, J. Biol. Chem. 281 (20) (2006) 13990–13998. [127] F.P. Prince, Lamellar and tubular associations of the mitochondrial cristae: unique forms of the cristae present in steroid-producing cells, Mitochondrion 1 (4) (2002) 381–389. [128] C.A. Mannella, M. Marko, P. Penczek, D. Barnard, J. Frank, The internal compartmentation of rat-liver mitochondria: tomographic study using the high-voltage transmission electron microscope, Microsc. Res. Tech. 27 (4) (1994) 278–283. [129] C.A. Mannella, Minireview: on the structure and gating mechanism of the mitochondrial channel, VDAC, J. Bioenerg. Biomembr. 29 (6) (1997) 525–531. F. Martinez et al. / Steroids 103 (2015) 11–22 21 [130] G.A. Perkins, C.W. Renken, J.Y. Song, T.G. Frey, S.J. Young, S. Lamont, et al., Electron tomography of large, multicomponent biological structures, J. Struct. Biol. 120 (3) (1997) 219–227. [131] G.A. Perkins, C.W. Renken, T.G. Frey, M.H. Ellisman, Membrane architecture of mitochondria in neurons of the central nervous system, J. Neurosci. Res. 66 (5) (2001) 857–865. [132] G.A. Perkins, J.Y. Song, L. Tarsa, T.J. Deerinck, M.H. Ellisman, T.G. Frey, Electron tomography of mitochondria from brown adipocytes reveals crista junctions, J. Bioenerg. Biomembr. 30 (5) (1998) 431–442. [133] D. Nicastro, A.S. Frangakis, D. Typke, W. Baumeister, Cryo-electron tomography of neurospora mitochondria, J. Struct. Biol. 129 (1) (2000) 48–56. [134] G.A. Perkins, C.W. Renken, I.J. van der Klei, M.H. Ellisman, W. Neupert, T.G. Frey, Electron tomography of mitochondria after the arrest of protein import associated with Tom19 depletion, Eur. J. Cell Biol. 80 (2) (2001) 139–150. [135] G.A. Perkins, M.H. Ellisman, D.A. Fox, Three-dimensional analysis of mouse rod and cone mitochondrial cristae architecture: bioenergetic and functional implications, Mol. Vis. 9 (2003) 60–73. [136] J.A. Allen, T. Shankara, P. Janus, S. Buck, T. Diemer, K.H. Hales, D.B. Hales, Energized, polarized, and actively respiring mitochondria are required for acute Leydig cell steroidogenesis, Endocrinology 147 (8) (2006) 3924–3935. [137] W.L. Miller, Steroid hormone biosynthesis and actions in the materno-feto- placental unit, Clin. Perinatol. 25 (4) (1998) 799–817. [138] X. Xu, T. Xu, D.G. Robertson, J.D. Lamberth, GTP stimulates pregnenolone generation in isolated rat adrenal mitochondria, J. Mol. Biol. 264 (30) (1989) 17674–17680. [139] F. Martínez, T. Espinosa-García, O. Flores-Herrera, J.P. Pardo, Respiratory control induced by ATP in human term placental mitochondria, Placenta 14 (3) (1993) 321–331. [140] B. Kadenbach, R. Ramzan, L. Wen, S. Vogt, New extension of the Mitchell Theory for oxidative phosphorylation in mitochondria of living organisms, Biochim. Biophys. Acta 1800 (3) (2010) 205–212. [141] D. Mokranjac, W. Neupert, Energetics of protein translocation into mitochondria, Biochim. Biophys. Acta 1777 (7–8) (2008) 758–762. [142] V.C. Manganiello, T. Murata, M. Taira, P. Belfrage, E. Degerman, Diversity in cyclic nucleotide phosphodiesterase isoenzyme families, Arch. Biochem. Biophys. 322 (1) (1995) 1–13. [143] J. Senior, K. Marshall, R. Sangha, G.S. Baxter, J.K. Clayton, in vitro characterization of prostanoid EP-receptors in the nonpregnant human myometrium, Br. J. Pharmacol. 102 (3) (1991) 747–753. [144] M.L. Casey, J. Smith, G. Alasbrook, P.C. MacDonald, Activation of adenylyl cyclase in human myometrial smooth muscle cells by neuropeptides, J. Clin. Endocrinol. Metab. 82 (9) (1997) 3087–3092. [145] M.H. Litime, G. Pointis, M. Breuiller, D. Cabrol, F. Ferre, Disappearance of b- adrenergic response of human myometrial adenylate cyclase at the end of pregnancy, J. Clin. Endocrinol. Metab. 69 (1) (1989) 1–6. [146] S.A. Price, I. Pochun, S. Phaneuf, Lopez Bernal A. Adenylyl cyclase isoforms in pregnant and non-pregnant human myometrium, J. Endocrinol. 164 (1) (2000) 21–30. [147] L. Kao, G.O. Babalola, G.S. Kopf, C. Coutifaris, J.F. Strauss III, Differentiation of human trophoblasts: structure–function relationships, in: P.K.C. Leung, A.J. W. Hsueh, H.G. Friesen (Eds.), Molecular Basis of Reproductive Endocrinology, Springer-Verlag, New York, 1993, pp. 159–170. [148] M. Conti, G. Nemos, C. Sette, E. Vicini, Recent progress in understanding the hormonal regulation of phosphodiesterases, Endocr. Rev. 16 (3) (1995) 370– 389. [149] K. Lefkimmiatis, D. Leronni, A.M. Hofer, The inner and outer compartments of mitochondria are sites of distinct cAMP/PKA signaling dynamics, J. Cell Biol. 202 (3) (2013) 453–462. [150] R. Acin-Perez, E. Salazar, M. Kamenetsky, J. Buck, L.R. Levin, G. Manfredi, Cyclic AMP produced inside mitochondria regulates oxidative phosphorylation, Cell Metab. 9 (3) (2009) 265–276. [151] R. Acin-Perez, D.L. Gatti, Y. Bai, G. Manfredi, Protein phosphorylation and prevention of cytochrome oxidase inhibition by ATP: coupled mechanisms of energy metabolism regulation, Cell Metab. 13 (6) (2011) 712–719. [152] K. Omori, J. Kotera, Overview of PDEs and their regulation, Circ. Res. 100 (3) (2007) 309–327. [153] J.H. Zippin, Y. Chen, P. Nahirney, M. Kamenetsky, M.S. Wuttke, D.A. Fischman, et al., Compartmentalization of bicarbonate-sensitive adenylyl cyclase in distinct signaling microdomains, FASEB J. 17 (1) (2003) 82–84. [154] D.H. Miles, M. Graeme, Tailoring cAMP-signalling responses through isoform multiplicity, Trends Biochem. Sci. 22 (6) (1997) 217–224. [155] J. Buck, M.L. Sinclair, L. Schapal, M.J. Cann, L.R. Levin, Cytosolic adenylyl cyclase defines a unique signaling molecule in mammals, Proc. Natl. Acad. Sci. U.S.A. 96 (1) (1999) 79–84. [156] R. Bernatchez, L. Belkacemi, E. Rassart, G. Daoud, L. Simoneau, J. Lafond, Differential expression of membrane and soluble adenylyl cyclase isoforms in cytotrophoblast cells and syncytiotrophoblasts of human placenta, Placenta 24 (6) (2003) 648–657. [157] R. Taussig, A.G. Gilman, Mammalian membrane-bound adenylyl cyclases, J. Biol. Chem. 270 (1) (1995) 1–4. [158] J. Hanoune, Y. Pouille, E. Tzavara, T. Shen, L. Lipskaya, N. Miyamoto, et al., Adenylyl cyclases: structure, regulation and function in an enzyme superfamily, Mol. Cell. Endocrinol. 128 (1–2) (1997) 179–194. [159] F.A. Antoni, S.M. Smith, J. Simpson, R. Rosie, G. Fink, J.M. Paterson, Calcium control of adenylyl cyclase: the calcineurin connection, Adv. Second Messenger Phosphoprotein Res. 32 (1998) 153–172. [160] Z.Z. Yang, O. Tschopp, M. Hemmings-Mieszczak, J. Feng, D. Brodbeck, E. Perentes, B.A. Hemmings, Protein kinase B alpha/Akt1 regulates placental development and fetal growth, J. Biol. Chem. 278 (34) (2003) 32124–32131. [161] N. Hatano, Y. Mori, M. Oh-hora, A. Kosugi, T. Fujikawa, N. Nakai, et al., Essential role for ERK2 mitogen-activated protein kinase in placental development, Genes Cells 8 (11) (2003) 847–856. [162] Z. Krozowski, J.A. Maguire, A.N. Stein-Oakley, J. Dowling, R.E. Smith, R.K. Andrews, Immunohistochemical localization of the 11 b-hydroxysteroid dehydrogenase type II enzyme in human kidney and placenta, J. Clin. Endocrinol. Metab. 80 (7) (1995) 2203–2209. [163] H. Guan, K. Sun, K. Yang, The ERK1/2 signaling pathway regulates 11beta- hydroxysteroid dehydrogenase type 2 expression in humantrophoblast cells through a transcriptional mechanism, Biol. Reprod. 89 (4) (2013) 1–7. [164] L Poston, M.T. Raijmakers, Trophoblast oxidative stress, antioxidants and pregnancy outcome – a review, Placenta 25 (Suppl. A) (2004) S72–S78. [165] C. Vaillancourt, D. Lanoix, F. Le Bellego, G. Daoud, J. Lafond, Involvement of MAPK signalling in human villous trophoblast differentiation, Mini Rev. Med. Chem. 9 (8) (2009) 962–973. [166] C. Runchel, A. Matsuzawa, H. Ichijo, Mitogen-activated protein kinases in mammalian oxida-tive stress responses, Antioxid. Redox Signal. 15 (1) (2011) 205–218. [167] M. Takekawa, T. Maeda, H. Saito, Protein phosphatase 2Calpha inhibits the human stress-responsive p38 and JNK MAPK pathways, EMBO J. 17 (16) (1998) 4744–4752. [168] G. Daoud, M. Amyot, E. Rassart, A. Masse, L. Simoneau, J. Lafond, ERK1/2 and p38 regulate trophoblasts differentiation in human term placenta, J. Physiol. 566 (Pt 2) (2005) 409–423. [169] T. Sakiyami, J.C. Robinson, J.Y. Chou, Characterization of alkaline phosphatases from first trimester placentas, J. Biol. Chem. 254 (3) (1979) 935–938. [170] A.L. Solomon, K.W. Siddals, P.N. Baker, J.M. Gibson, J.D. Aplin, M. Westwood, Placental alkaline phosphatase de-phosphorylates insulin-like growth factor (IGF)-binding protein-1, Placenta 35 (7) (2014) 520–522. [171] D.M. Stocco, X. Wang, Y. Jo, P.R. Manna, Multiple signaling pathways regulating steroidogenesis and steroidogenic acute regulatory protein expression: more complicated than we thought, Mol. Endocrinol. 19 (11) (2005) 2647–2659. [172] H. Chen, M.P. Hardy, I. Huhtaniemi, B.R. Zirkin, Age-related decreased Leydig cell testosterone production in the Brown Norway rat, J. Androl. 15 (6) (1994) 551–557. [173] R.V. Carsia, G.J. Macdonald, S. Malamed, Steroid control of steroidogenesis in isolated adrenocortical cells: molecular and species specificity, Steroids 41 (6) (1983) 741–755. [174] M.J. DiBartolomeis, C.R. Jefcoate, Characterization of the acute stimulation of steroidogenesis in primary bovine adrenal cortical cell cultures, J. Biol. Chem. 259 (16) (1984) 10159–10167. [175] M.S. Weedon-Fekjær, K. Taskén, Review: spatiotemporal dynamics of hCG/ cAMP signaling and regulation of placental function, Placenta 33 (2012) S87– S91. [176] O. Ritvos, Modulation of steroidogenesis in choriocarcinoma cells by cholera toxin, phorbol ester, epidermal growth factor and insulin-like growth factor I, Mol. Cell. Endocrinol. 59 (1–2) (1988) 125–133. [177] B.B. Feinberg, D.J. Anderson, M.A. Steller, V. Fulop, R.S. Berkowitz, J.A. Hill, Cytokine regulation of trophoblast steroidogenesis, J. Clin. Endocrinol. Metab. 78 (3) (1994) 586–591. [178] Y.G. Shanker, A.J. Rao, Regulation of progesterone biosynthesis in the human placenta by estradiol 17 beta and progesterone, Biochem. Mol. Biol. Int. 43 (3) (1997) 591–599. [179] R.C. Tuckey, Progesterone synthesis by the human placenta, Placenta 26 (4) (2005) 273–281. [180] D. Barrera, E. Avila, G. Hernández, A. Halhali, B. Biruete, F. Larrea, et al., Estradiol and progesterone synthesis in human placenta is stimulated by calcitriol, J. Steroid Biochem. Mol. Biol. 103 (3–5) (2007) 529–532. [181] G.E. Ringler, J.F. Strauss 3rd., In vitro systems for the study of human placental endocrine function, Endocr. Rev. 11 (1) (1990) 105–123. [182] L. Lin, B. Xu, N.S. Rote, The cellular mechanism by which the human endogenous retrovirus ERV-3 env gene affects proliferation and differentiation in a human placental trophoblast model, BeWo, Placenta 21 (1) (2000) 73–78. [183] C. Wadsack, A. Hrzenjak, A. Hammer, B. Hirscmugl, S. Levak-Frank, G. Desoye, et al., Trophoblast-like human choriocarcinoma cells serve as a suitable in vitro model for selective cholesteryl ester uptake from high density lipoproteins, Eur. J. Biochem. 270 (3) (2003) 451–462. [184] A. Feliciello, M.E. Gottesmas, E.V. Avvedimento, The biological functions of A- kinase anchor proteins, J. Mol. Biol. 308 (2) (2001) 99–114. [185] C.K. Means, B. Lygren, L.K. Langeberg, A. Jain, R.E. Dixon, A.L. Vega, et al., An entirely specific type I A-kinase anchoring protein that can sequester two molecules of protein kinase A at mitochondria, Proc. Natl. Acad. Sci. U.S.A. 108 (48) (2011). E1227-35. [186] C. Gomez-Concha, O. Flores-Herrera, S. Olvera-Sanchez, M.T. Espinosa-Garcia, F. Martinez, Progesterone synthesis by human placental mitochondria is sensitive to PKA inhibition by H89, Int. J. Biochem. Cell Biol. 43 (6) (2011) 1402–1411. [187] M.P. Ma, M. Thomson, Protein kinase A subunit a catalytic and A kinase anchoring protein 79 in human placental mitochondria, Open Biochem. J. 6 (2012) 23–30. 22 F. Martinez et al. / Steroids 103 (2015) 11–22 Hindawi Biochemistry Research International Article ID 5681081 Research Article Streptozotocin-Induced Adaptive Modification of Mitochondrial Supercomplexes in Liver of Wistar Rats and the Protective Effect of Moringa oleifera Lam Marı́a Alejandra Sánchez-Muñoz,1 Mónica Andrea Valdez-Solana,1 Mara Ibeth Campos-Almazán,2 Óscar Flores-Herrera,3 Mercedes Esparza-Perusquı́a,3 Sofia Olvera-Sánchez,3 Guadalupe Garcı́a-Arenas,2 Claudia Avitia-Domı́nguez,2 Alfredo Téllez-Valencia,2 and Erick Sierra-Campos 1 1Facultad de Ciencias Quı́micas, Universidad Juárez del Estado de Durango Campus, Gómez Palacio, DGO, Mexico 2Facultad de Medicina y Nutrición, Universidad Juárez del Estado de Durango Campus, Gómez Palacio, DGO, Mexico 3Departamento de Bioquı́mica, Facultad de Medicina, Universidad Nacional Autónoma de México, Mexico City, Mexico Correspondence should be addressed to Erick Sierra-Campos; ericksier@gmail.com Received 15 November 2017; Accepted 28 December 2017 Academic Editor: Saad Tayyab Copyright © 2018 Maŕıa Alejandra Sánchez-Muñoz et al. his is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in anymedium, provided the original work is properly cited. he increasing prevalence of diabetes continues to be a major health issue worldwide. Alteration of mitochondrial electron transport chain is a recognized hallmark of the diabetic-associated decline in liver bioenergetics; however, the molecular events involved are only poorly understood. Moringa oleifera is used for the treatment of diabetes. However, its role on mitochondrial functionality is not yet established. his study was aimed to evaluate the efect ofM. oleifera extract on supercomplex formation, ATPase activity, ROS production, GSH levels, lipid peroxidation, and protein carbonylation. he levels of lipid peroxidation and protein carbonylation were increased in a diabetic group. However, the levels were decreased inM. oleifera-treated diabetic rats. Analysis of in-gel activity showed an increase in all complex activities in the diabetic group, but spectrophotometric de- terminations of complex II and IV activities were unafected in this treatment. However, we found an oxygen consumption abolition through complex I-III-IV pathway in the diabetic group treated withMoringa. While respiration with succinate feeding into complex II-III-IV was increased in the diabetic group. hese indings suggest that hyperglycemia modiies oxygen con- sumption, supercomplexes formation, and increases ROS marker levels in mitochondria from the liver of STZ-diabetic rats, whereas M. oleifera may have a protective role against some alterations. 1. Introduction Mitochondria, which are mainly composed by proteins and lipids, are considered the most complex and the most im- portant organelles of eukaryotic cells. hey not only play a leading role in the energy metabolism, but also closely involve in many cellular processes [1]. Moreover, mito- chondria are highly dynamic organelles that continuously divide and fuse as well as move within the cell [2]. In ad- dition, it is now well established that the individual re- spiratory complexes can be organized into supercomplexes, but the composition and abundance of these may vary among organisms and tissues depending on the metabolic and physiological conditions [3–5] as well as on the lipid content of the mitochondrial inner membrane [6, 7]. How- ever, mitochondria are a source of reactive oxygen species (ROS) which are involved in many pathological scenarios [8] and often play an essential role in physiological cell death mechanisms [9]. Mitochondrial dysfunction has recently been identiied as a common metabolic defect associated with diabetes, obesity, and its metabolic complications [10, 11]. Previous studies have demonstrated that chronic diabetes induced by streptozotocin (STZ) provoked signiicant alterations in hepatic mitochondrial function which were restored to normality with insulin treatment [12] or with mifepristone (RU 38486) treatment [13]. In addition, it has been postu- lated that STZ-induced cytotoxicity in HepG2 cells is me- diated, at least in part, by the increase in ROS and reactive nitrogen species (RNS) production, oxidative stress, and mitochondrial dysfunction [14]. Moreover, diverse studies suggest that mitochondrial oxidative function was com- promised in diabetic and prediabetic humans as evidenced by reduced levels of fatty acid oxidation, insulin-stimulated ATP synthesis, and expression of genes involved in oxidative phosphorylation (OXPHOS) [15–17]. With respect to OXPHOS, activity was suggested that mitochondrial di- abetes may also afect the complex V [18], and it is in- teresting to mention that, in diabetic patients’ muscle, blue native gel electrophoresis revealed a striking decrease in complex I, III, and IV containing supercomplexes [19]. In addition, impairment of pyruvate dehydrogenase complex on the citric acid cycle and glucokinase activity during di- abetes has been reported [19, 20].1 hese indings can be associated with an increased in ROS production and a de- crease in cellular reduced glutathione (GSH) content in STZ- induced diabetic rats [21] and diabetic patients [22]. Moringa oleifera is commonly used in folk medicine as an antidiabetic agent via its antioxidant property. Yet, its biological activity is not limited to the antioxidant capacity. In fact, other important biological activities such as hypo- lipidaemic, antiatherosclerotic, and anticarcinogenic activ- ities of M. oleifera leaves and seeds have been reported [23–26]. However, phenolic compounds found in M. olei- fera, especially lavonoids, possess both antioxidant and prooxidant properties depending on concentration used. he latter which is exhibited at higher concentrations of phenolic compounds such as quercetin, galangin, taxifolin, catechin, and prenylated lavonoid have been shown to afect mitochondrial energetic processes (see supplementary ma- terials (available here)) [27, 28].2 In addition, it has been shown that mitochondria are a plausible main target of lavonoids mediating preventive actions against stress and mitochondrial dysfunction-associated pathologies [29]. Recent evidence indicates that M. oleifera aqueous leaf ex- tract presents anticancerous efect on A549 cancer cells by afecting mitochondrial membrane potential and ATP levels [30]. More recently, Khan et al. [31] showed that aqueous extract of M. oleifera leaf protects pancreas against ROS- mediated damage by enhancing cellular antioxidant defenses and minimizing hyperglycemia in STZ-induced diabetes, which might be due to the glucose uptake enhancement in skeletal muscle, insulin secretion stimulation, and alpha- amylase and alpha-glucosidase inhibition. hus, the fa- vorable roles of M. oleifera in glucose metabolism and antioxidant system led us to investigate the efects of M. oleifera on diabetes-induced mitochondrial changes in liver. he aim of this study was to investigate the protecting efect of M. oleifera extract upon STZ-induced mitochon- drial dysfunction. To assess the degree of injury of the STZ, both respiratory and enzyme activity parameters were evaluated and compared with the changes in theM. oleifera- treated group. 2. Materials and Methods 2.1.Preparationof theExtract. he extract was prepared using 23 g of dry-ground sample and 260mL of 80% methanolic aqueous solution by successive maceration. he mixture was shaken in a magnetic grid at room temperature for 24 h and then iltered through Whatman ilter paper number 1. he inal extract was concentrated on a rotary evaporator, placed in a deep freezer for 24 h and lyophilized to obtain a pow- dered extract that was kept at −80°C. 2.2. Ethics Statement. All experiments were performed in compliance with the guideline for the welfare of experi- mental animals by the National Institutes of Health and in accordance with the guidelines of Institutional Animal Care. his study was approved by the Institutional Animal Ethics Committee at the Faculty of Health Science, UJED. 2.3. Diabetic Model and Treatment. Streptozotocin (STZ) was dissolved in a citrate bufer (0.1M, pH 4.5) and in- traperitoneally injected (55mg/kg) to induce diabetes in rats. Rats injected only with citrate bufer served as control. Type 1 diabetes was conirmed evaluating fasting plasma glucose levels after 5 days of induction; the inclusion criteria to establish diabetes were 200mg/dL of fasting plasma glucose. Rats were divided in control (C group), diabetic (D group), andM. oleifera-treated diabetic (M group) groups. M group was daily administered with a 200mg/kg dose of extract by gavage during 3 weeks, and remaining groups were ad- ministered with water as vehicle. 2.4. Isolation and Puriication of Mitochondria. he rat liver was collected immediately after euthanasia and homoge- nized in 100mL of a bufer containing 20mM Tris-HCl, 200mM mannitol, 50mM sucrose, 1mM EDTA, 1mM PMSF, 1 protease inhibitor tablet, and 0.1% bovine serum albumin (BSA) (pH 7.4; bufer A). Cellular and nuclear fractions were removed in the pellet by centrifuging at 3,500 rpm for 10min at 4°C. Mitochondria were obtained by centrifuging the supernatant for 10min at 11,000 rpm. hen, mitochondria were washed and resuspended in bufer A without BSA and centrifuged at 11,000 rpm for 10min. Mitochondria were loaded on a Percoll gradient 15, 23, and 40% in bufer A without BSA and centrifuged for 35min at 25,000 rpm at 4°C [32]. 2.5. Oxygen Consumption. Oxygen uptake was estimated polarographically using a Clark-type electrode in a 1.5ml water-jacketed chamber at 37°C. he mixture contained 250mM sucrose, 20mM HEPES, 50mM K2HPO4, 10mM H3PO4, 10mMMgCl2, and 1mM EGTA, 0.1% BSA (pH 7.4) [33]. Oxygen consumption was stimulated by the addition of 0.1mM NADH or 10mM succinate (in the presence of 2 µM rotenone). Otherwise, artiicial substrates such as 2 Biochemistry Research International ascorbate/TMPD (10mM and 100 µM, respectively, in the presence of 2 µM antimycin) were used for complex IV activity, and malonate and KCN were added to inhibit complex IV and complex II (10mM and 5mM, respectively). 2.6. NADH Dehydrogenase and Succinate Dehydrogenase Activities. Activities of complex I (NADH :DCPIP oxido- reductase) and complex II (succinate : DCPIP oxidoreduc- tase) were determined spectrophotometrically at 600 nm by following the reduction of the artiicial electron acceptor 2,6-dichlorophenol-indophenol (DCPIP; 50 μM; εDCPIP� 21mM−1·cm−1). Mitochondria were permeabilized with 0.03% zwittergent and incubated in 10mM KH2PO4, 5mM MgCl2, 1mM EGTA, and 120mM KCl (pH 7.4), either with 0.2mM NADH (complex I) or 2mM succinate (complex II), plus 0.2 mMmethosulfate phenazine (PMS). Mitochondria protein concentration was 1mg/ml, and the reaction was started by the addition of NADH or succinate [34]. 2.7. ATP Synthase Assay. ATP hydrolysis of complex V was measured spectrophotometrically at 25°C using a coupled assay to the oxidation of NADH (ε340 nm� 6.22mM−1·cm−1). he assay contained 100 μg mitochondrial protein, 10mM HEPES (pH 8.0), 100mM NaN3, 100 µM NO4Na, 90mM KCl, 3mM MgSO4; the ATP regenerating system consisted of 5mM phosphoenolpyruvate, 2mM ATP, 0.03% zwit- tergent, 50 units/mL pyruvate kinase, and 30 units/mL lac- tate dehydrogenase. he ATPase reaction was started by the addition of 0.1mM NADH. Oligomycin (6 µg/mL) was added to inhibit ATPase activity and verify F1F0-ATP synthase integrity; mitochondria were incubated with oli- gomycin for 30min [35]. 2.8. Native Electrophoresis. Respiratory complexes and supercomplexes were resolved by native PAGE as reported previously [36]. Puriied liver mitochondria (1mg) were suspended in 50mM Bis-Tris and 500mM 6-aminocaproic acid (pH 7.0) and solubilized by adding digitonin (de- tergent : protein ratio of 1 : 5). he mixtures were incubated for 30min at 4°C and centrifuged at 100,000 g for 30min. he supernatants were recovered and immediately loaded on a linear gradient polyacrylamide gradient gels (4–10%) for Blue Native PAGE (BN-PAGE) or Clear Native PAGE high resolution (hrCN-PAGE). For BN-PAGE, the anode bufer contained 50mM Bis- Tris/HCl (pH 7.0); the cathode bufer contained 50mM tricine and 15mMBis-Tris (pH 7.0), and Coomassie (0.02%). For the hrCN-PAGE, the anode bufer contained 25mM imidazole/HCl (pH 7.0); while the cathode bufer contained 50mM tricine, 7.5mM imidazole, 0.01% β dodecyl D-maltoside, and 0.05% sodium deoxycholate (pH 7.0), supplemented with Ponceau S red [37]. Gels were run at 4°C and 35V for 16 h. he molecular weights of the respiratory complexes or supercomplexes were estimated by using dig- itonin bovine heart mitochondrial complexes as standard: single complex: I� 1,000 kDa, V� 750 kDa, III2� 500 kDa, IV � 230 kDa, II � 130 kDa; supercomplexes: I-III- IV1–4 �1500–2100 kDa, V2 �1500 kDa. 2.9. Complex and Supercomplexes In-Gel Activities. he in- gel activity assays were performed as Wittig and Schägger [38] for complex I activity (NADH :methylthiazolyldiphenyl tetrazolium bromide reductase), complex II activity (succinate : methylthiazolyldiphenyl tetrazolium bromide reductase), and complex IV activity (cytochrome c : dia- minobenzidine reductase). In all cases, the assays were performed at 20–25°C and stopped with 50% methanol and 10% acetic acid, after 10–25min. he in-gel activity of complex V was performed in 50mM glycine (adjusted to pH 8.0 with triethanolamine), 10mM MgCl2, 0.15% Pb(ClO4)2, and 5mM ATP. ATP hydrolysis was correlated with the development of white lead phosphate precipitates. he reaction was stopped using 50% methanol, and subsequently, the gel was transferred to water and scanned against a dark background as described pre- viously [39]. 2.10. SDS-Gel Electrophoresis andWesternBlotAnalysis. Liver mitochondrial proteins (20 μg per well) were separated by SDS-PAGE according to Laemmli [40] in a 10% poly- acrylamide gel under denaturing conditions. Proteins were then transferred from gel to PVDF membrane (Immobilon P; Millipore, Bedford, MA) in a semidry electroblotting system (Bio-Rad) at 25V for 50min. Membranes were blocked in 500mMNaCl, 0.05% Tween-20, and 20mMTris- base (pH 7.5) (TTBS bufer), containing 5% blotting grade blocker nonfat dry milk. hen, membranes were incubated with antitotal OXPHOS antibody cocktail (at 1/500 di- lution). Immunoreactive bands were visualized by en- hanced chemiluminescence (Amersham Life Science, Inc.), according to the manufacturer’s instructions, using horse- radish peroxidase-conjugated antimouse IgG (at 1/10,000 dilution), and densitometric analyses were performed with the software Image Studio Lite version 5.2 (LI-COR Biosciences). 2.11. Protein Determination. he protein levels were esti- mated by the method described by Lowry et al. using BSA as standard [41]. 2.12.MitochondrialGlutathioneReductaseActivity. Glutathione reductase enzymatic activity was recorded by NADPH consumption. Briely, 50 µg of puriied mitochondria was placed in a phosphate bufer (50mM, pH 7.0) containing 1mM GSH and 0.1M NADPH. NADPH reduction was measured at 340 nm (εNADPH� 6.22M −1 ·cm−1). 2.13. Measurement of Glutathione Concentration by HPLC- UV. To quantify GSH and GSSG concentrations, a standard curve of oxidized and reduced glutathione was used as described by Yilmaz et al. [42]. Mitochondrial samples were centrifuged at 500 rpm for 10min and iltered to be injected Biochemistry Research International 3 onto a Kromasil ETERNITY C18 column (4.6×150mm). Mobile phase containing 10mM of monobasic sodium phosphate and 2%methanol (pH 3.0) was used at a low rate of 1mL/min in isocratic run. GSH andGSSG eluted from the column were detected at 210 nm. 2.14. Lipid Peroxidation Assay. Mitochondrial lipid perox- idation was estimated by the thiobarbituric acid reactive substances (TBARS) method consisting of TBA-TCA-HCl reaction as described by Buege and Aust [43]. Samples were boiled at 95°C for 60min, followed by a cooling and cen- trifugation steps at 12,000 rpm for 10min at 4°C. he pink product absorbance (formed when the MDA reacts with TBA) was spectrophotometrically recorded at 532 nm (εMDA� 1.56×10 5M−1·cm−1). MDA-TBA adduct peak was calibrated with tert-butyl hydroperoxide simultaneously processed as samples. 2.15.Mitochondrial H2O2Measurement. H2O2 emission was determined by the luorogenic indicator Amplex Red (Invitrogen) oxidation in presence of horseradish peroxidase as described by Starkov [44]. Fluorescence was recorded in a spectroluorometer (LS 55 PerkinElmer Life Sciences) with excitation and emission wavelengths of 555 and 581, re- spectively. Briely, 300 µg of puriied mitochondria was added to 1mL incubation bufer containing 125mM KCl, 20mM Hepes, 0.2mM EGTA, 2mM KH2PO4, 2% BSA, 1 µMAmplex Red, and 4U horseradish peroxidase (pH 7.2). H2O2 production was initiated after addition of 5mM py- ruvate, 2.5mM malate, and 10mM succinate as substrates and 1 µM rotenone, 0.2 µM antimycin A, and 5mM malo- nate as inhibitors. 2.16. Measurement of Protein Carbonylation. Determination of carbonyl content was followed as Levine et al. [45]. he oxidative damage to proteins was determined by carbonyl groups based on their reaction with 2,4-dinitrophenyl- hydrazine (DNPH) to form hydrazones. Briely, 0.5mg of mitochondria was incubated with 20mM DNPH solution for 1 h; then proteins were precipitated with 20% (w/v) of trichloroacetic acid and redissolved in DNPH. In brief, the proteins were precipitated by the addition of 20% (w/v) of trichloroacetate; protein pellet was washed three times with ethanol : ethyl acetate (1 : 1) and resuspended in 1mL of 6M guanidine. he absorbance was recorded at 370 nm (εHydrazone� 22×10 3M−1·cm−1). 2.17.Measurement ofHO-1Activity. Fresh livers were placed in prechilled Dounce homogenizer, and cold homogeniza- tion bufer containing 100mM potassium phosphate bufer (pH 7.4), 2mM MgCl2, 250mM sucrose, and a protease inhibitor cocktail (10 μg/mL leupeptin, 10 μg/mL trypsin inhibitor, 2 μg/mL aprotinin, and 1mM PMSF) was added. he homogenate was centrifuged at 10,000 g for 30min at 4°C, followed by the supernatant centrifugation at 100,000 g for 60min at 4°C, to obtain the microsomal fraction as a pellet. HO-1 activity was spectrophotometrically measured as described previously [46]. he microsomal fraction (50 μL) was added to the reaction mixture (500 μL) con- taining 0.8mM NADPH, 2mM glucose-6-phosphate, 0.2 unit of glucose-6-phosphate dehydrogenase, 20 μM hemin, 100mM potassium phosphate bufer (pH 7.4), and 2mg of rat liver cytosol as a source of biliverdin reductase. he mixture was incubated at 37°C for 60min in dark, and samples were left in an ice bath for at least 2min to stop the reaction. Bilirubin product was determined by calculation from diference in optical density (OD) at 464 nm and 530 nm (OD464–OD530 nm) of the sample. HO activity is expressed as pmol/min/mg protein. 2.18. Data Analysis. he obtained data are represented as mean± standard deviation of three independent determi- nations, using the Sigma Plot software version 11.0. Dif- ferences between means were obtained by analysis of variance (ANOVA) and multiple comparison tests. P values< 0.05 were considered as signiicant. 3. Results and Discussion he efectiveness ofM. oleifera extract in alleviating diabetes was assessed in the STZ-induced diabetic model in Wistar rats. In response to STZ, rats showed increased water uptake, increased urine production, increased blood glucose levels, and reduced weight gain (D group� 229± 9.05mg/dL and 156± 12 g), which were unaltered in the control group (C group� 78± 5.5mg/dL and 187± 18 g), while M group sig- niicantly alleviated all parameters of diabetes (86± 4.2mg/dl and 194± 8 g). hese results suggest that M. oleifera leaf may be a potential agent in the treatment of type 1 diabetes and are agreed with the observations that suggest the beneicial efects ofMoringa oleifera supplementation on diabetes [47, 48]. Hence, these results led us to investigate the valuable efects of the leaf extract on STZ-induced mitochondrial changes, in liver, evaluating STZ injury on both, respiratory and enzyme activities from respiratory chain and some of the antioxidant system comparing them with those from M. oleifera treatment. 3.1. M. oleifera Attenuates Oxidant Stress and the Decrease in the Glutathione System in Liver Mitochondria. Diabetic cells and tissues have the capacity to invoke adaptive mechanisms that evolved to defend against oxidative stress [49]. One putative mechanism is a defense system that would protect against ROS into mitochondria. hese include the super- oxide conversion to hydrogen peroxide (H2O2) by man- ganese superoxide dismutase (SOD) and scavenging H2O2 by catalase, glutathione peroxidase (GPx), or peroxiredoxin III [50]. Reduced glutathione (GSH) scavenges H2O2 via GPx, ubiquitously expressed both in the mitochondria matrix and intermembrane space [51]. In turn, the reduction of oxidized glutathione (GSSG) to GSH is catalyzed by glutathione reductase (GR), which requires NADPH. hus, increased ROS removal results in increased NADPH turn- over. Also, GSH can also be used in conjugation reactions to protect mitochondria enzymes from various toxins, for 4 Biochemistry Research International example, by-products in lipid peroxidation such as 4-hydroxynonenal (HNE) [52]. To assess the inluence of STZ injection on redox state, mitochondrial GSH levels and GR of liver were examined. A single dose of STZ caused a signiicant decrease in GSH and total GSH contents of diabetic rats (Table 1). Basal levels of total GSH were 178.8± 5.7 μmol/mg of protein in control mitochondria, whereas total GSH levels in isolated mito- chondria from STZ-treated rats (D group) were signiicantly decreased by 70% compared with control (100.2± 1.3 μmol/mg of protein). In contrast, M. oleifera treatment prevented a STZ-mediated decrease in GSH levels (M group� 223.7± 2.9 μmol/mg of protein), which correspond to an increase of 25% compared with C group. It is worth to mention that M group rats showed a signiicant increase in values of GSH, total GSH (2 times), and GSH/GSSG ratio (P< 0.05) compared with D group. However, the M group ratio was 12 times reduced with respect to C group (Table 1). One possible explanation for this phenomenon may be the inactivation of mitochondrial GR activity. However, as observed in Table 1, STZ administration did not alter the GR activity in liver mitochondria when compared with control rats. However, M group samples signiicantly increased GR activity when compared with values of D and C groups (Table 1). Hence, these results show that GR inactivation is not the main mechanism of GSSG accumulation into the mitochondria. he oxidative stress implications in diabetes pathogenesis are suggested to be produced not only by ROS generation but also by a nonenzymatic protein glycation, autoxidation of glucose, impairment of antioxidant enzymes, and peroxides formation. herefore, GSH level decline is associated with oxidative damage to macromolecules, such as lipids and proteins. ROS-mediated lipid peroxidation is a crucial factor in the development of diabetic liver complications. In addi- tion, GSH depletion induces heme oxygenase-1 (HO-1), a key microsomal enzyme in heme degradation to carbon mon- oxide (CO), iron (Fe2+), and biliverdin; this latter being converted into bilirubin by the cytosolic biliverdin re- ductase [53, 54]. Moreover, some observations suggest the cytoprotective mechanism of HO-1 against oxidative stress involving an increase in mitochondrial carrier levels and antiapoptotic proteins as well as in cytochrome c oxidase activity [55]. In order to evaluate this possibility, we measure car- bonyls concentration, lipoperoxidation, and HO-1 activity. As observed in Table 2, C group showed the lowest levels of carbonylation and MDA. In contrast, STZ treatment in- creased lipid peroxidation and protein carbonyl content (Table 2). Besides, M group showed a signiicant decrease in lipoperoxidation in liver mitochondria (P< 0.05) when compared with D group. he carbonylation level in mito- chondria of M group was signiicantly lower and showed signiicant diference when compared with D group (Table 2). In contrast, M. oleifera extract administration did not prevent the HO-1 induction provoked by STZ, where its enzymatic activity remained signiicantly higher. Our results clearly demonstrated that M. oleifera signiicantly sup- pressed both lipoperoxidation and protein carbonylation. However, M. oleifera did not lower the HO-1 activity, and little is known about the molecular mechanisms responsible for its activation, which requires further investigation. It has been reported thatM. oleifera exhibits bifunctional antioxidant properties related to its ability to react directly with ROS and to induce antioxidant enzymes expression such as superoxide dismutase, catalase, glutathione re- ductase, and glutathione peroxidase [56–58]. We conirmed previous data and showed that Moringa not only decreased lipoperoxidation and protein carbonylation levels in rat livers but also increased HO-1 activity, parameters associ- ated with a cytoprotective mechanism against oxidative stress [59]. 3.2. Efects of STZ and M. oleifera on Oxygen Consumption. Although many previous studies have re- ported pharmacological properties of M. oleifera, particu- larly as antioxidant and antidiabetic properties that may provide beneits for diabetic patients [25, 60, 61], there are no reports that showM. oleifera extract efect on mitochondria functionality. To determine the changes of mitochondrial respiration in STZ-induced diabetic rats and M. oleifera Table 1: GSH and GSSG levels by HPLC-DAD and GR enzymatic activity in liver mitochondria from diferent groups. Group GSH (µmol/mg protein) GSSG (µmol/mg protein) GSH/GSSG ratio Total GSH (µmol/mg protein) GR (U/min) C 174.1± 35.1 4.8± 3.4 36.2± 0.19 178.8± 5.7 267.2± 11.7 D 50.4± 1∗ 49.8± 1.1∗ 1± 0.08∗ 100.2± 1.3∗ 236± 14.5 M 169± 1.2∗∗ 54.7± 2.2∗∗ 3± 0.05∗∗ 223.7± 2.9∗∗ 366.7± 23.8∗∗ C� control; D� diabetic; M� diabetic plus Moringa. ∗Signiicant diference versus control (P< 0.05). ∗∗Signiicant diference versus control and diabetic (P< 0.05). Table 2: Levels of MDA and carbonyl groups in liver mitochondria from diferent treatments. Group MDA (nmol/mg prot) Carbonyl groups (nmol/mg) HO-1 (pmol/min/mg) C 0.4317± 0.009 3.7232± 0.57 40.9± 4.9 D 0.5028± 0.06 12.738± 0.28# 85.7± 2.1∗ M 0.3851± 0.02∗ 4.2645± 0.98 105.2± 3.4∗# C� control; D� diabetic; M� diabetic plus Moringa. ∗Signiicant diference versus diabetic (P< 0.05). #Signiicant diference versus control (P< 0.05). Biochemistry Research International 5 protective efect, we measured the mitochondrial respiratory chain using Clark-type oxygen electrode and determined enzymatic activity of each complex by spectrophotometric methods. Figure 1(a) shows the respiratory activity of all groups in presence of complex I, II, and IV substrates. he combi- nation of pyruvate +malate indirectly investigates the monocarboxylate and dicarboxylate transporters and the pyruvate dehydrogenase activities. he substrate combina- tion produces NADH which donates electrons to complex I. In D group, the state 4 respiration with pyruvate +malate was not afected. However, Moringa treatment resulted in 15% decrease in the state 4 respiration. By contrast, succinate donates electrons to FAD+ in complex II and yields sig- niicantly high respiration state 4 rates in both D and M groups compared with C group. Otherwise, in diabetic rats, state 4 respiration with succinate increased by 80% com- pared with control rats. he observed change in the diabetic animals was rectiied by Moringa treatment (Figure 1(a)). Additionally, functional analysis of complex IV (cytochrome c oxidase) maximal activity was assayed with ascorbate (Asc) and N,N,N′,N′-tetramethyl-p-phenylenediamine (TMPD), which is an artiicial redox mediator that assists the electron transfer from ascorbate to cytochrome c. Complex IV res- piration was calculated as the portion sensitive to cyanide potassium (KCN), a speciic inhibitor of cytochrome c ox- idase (Figure 1(a)), revealing no diferences among all ex- perimental groups. Additionally, we measured the speciic activities of re- spiratory chain complexes in liver mitochondria. Spectro- photometric analysis showed a signiicant increase in complex I and ATPase activities in D group (Figure 1(c)), while Moringa treatment was efectively reversing this al- teration nearly to control values. Complex II showed no signiicant change in mitochondrial fraction of D group (Figure 1(b)). hese data show that individual activities of ⁎ # # ⁎ C 200 CI+II+III+IV CII+III+IV CIV 150 100 50 n ga t O 2 (m g· m in ) 0 D M C D M C D M (a) ⁎ CI C A ct iv it y (μ m o l/ m in /m g p ro te in ) D M CII CI CII CI CII 0.18 0.16 0.14 0.12 0.10 0.08 0.06 0.04 0.02 0.00 (b) ⁎⁎ ⁎ ⁎ C D + Oligo + Oligo + Oligo M 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 A T P h yd ro ly si s (μ m o l/ m in /m g p ro te in ) (c) Figure 1: Moringa oleifera efect on mitochondrial respiratory chain. (a) Oxygen consumption (∗P< 0.05 versus D; #P< 0.05 versus C); (b) enzymatic activity of complex I and II (∗P< 0.05 versus C andM); (c) F1/F0ATPase activity ( ∗ P< 0.05 versus C andM; ∗∗P< 0.05 versus M+oligo) from liver mitochondria. C: control; D: diabetic; M: Moringa-treated diabetic groups. 6 Biochemistry Research International mitochondrial electron transport chain (ETC) enzymes were not negatively modiied in diabetic treatment. In addition, the respiratory properties of D and M groups have ap- proximately 1.5–2 times succinate-respiratory rates com- pared with that of C group. herefore, our results suggest that respiratory complex activities were not decreased in liver mitochondria in STZ-induced diabetic rats. hese indings, which may appear, at irst glance, contradictory, may be interpreted in terms of higher ETC eiciency, thus, avoiding energy losses by electron leakage in response to change in physiological functions and body energy re- quirements; ETC undergoes some modiications either during pathology development or disease [62]. In addition, several studies concerning STZ-treated rats have been performed with animals of diferent strains and diferent amounts of STZ [63, 64]. In addition, in isolated hepatocytes, increasing glucose concentration does not increase ΔµH+, mitochondrial respiratory rate, or cytosolic NADH/NAD+ ratio; instead, most of glucose excess is converted to gly- cogen [65]. In fact, some authors have recently suggested that mitochondria overstimulation is a probable risk factor for insulin resistance, while moderate mitochondrial dys- function may actually be protective under certain conditions, suggesting the mitochondrial modulation as a prospective therapy for metabolic diseases [66]. For this reason, it is important that future research clariies the true energy functional state of isolated mitochondria from diabetic ani- mals [67]. 3.3. Modulation of Mitochondrial Complexes by STZ. As mitochondrial content can substantially impact on respiratory capacity, protein components of individual respiratory com- plexes were quantiied. To test whether hyperglycemia andM. oleifera extract altered the composition of the ETC, we ana- lyzed the expression level of nuclear-encoded mitochondrial complex I subunit NDUFB8, complex II subunit SDHB, complex III UQCRC2, complex IV MTCO1, and complex V ATP5A of each group. Interestingly, in diabetic rats, the NDUFB8 subunit resulted in an increase of complex I ex- pression, while complex II and III were unaltered (Figure 2). hese data which resulted in increased expression of NDUFB8 and MTCO1 support the suggestion that increased activity of mitochondrial respiratory chain could result from a proteome alteration leading to modulation of expression/activity of a range of mitochondrial components. More importantly, an upregulation of hepatic CI-NDUFB8 and CIV-MTCO1 was found in diabetic rats, consistentwith changes in-gel activity of these complexes. hus, it is plausible that increased NDUFB8 and MTCO1 contents observed in the STZ group, resulting from diabetes mellitus type 1, may account, in part, for the mitochondrial morphological changes observed, which could have downstream efects on mitochondrial functionality leading to hepatic dysfunction. 3.4. Loss of Redox State Does Not Destabilize Mitochondrial Supercomplexes. It is now widely accepted that mitochon- drial respiratory chain is organized with stable and func- tional entities called supercomplexes (SC) [68]. SC consist of various ratios of copies of individual complexes (I, III, IV, and V) to form stable, supramolecular structures; for in- stance, CI forms a supercomplex with CIII2 and CIV (known as the respirasome), as well as with CIII2 alone (SC I + III2). CIII2 forms a supercomplex with CIV (SC III2+ IV), and CV forms dimers (CV2). In addition, another recent advance is that the discovery of respiratory megacomplex (MC I2III2IV2) represents the highest-order assembly of re- spiratory complexes [69], and it allows mitochondria to respond to energy requiring conditions and to minimize ROS generation during electron transfer reactions [70], as well as the sequestering of vulnerable sites of mitochondrial complexes from oxidative damage as a protective mecha- nism that prevents tight interactions between the individual complexes [71]. It is fairly well established in rectus abdominis muscle of diabetic obese patients. BN-PAGE revealed a striking de- crease in complex I, III, and IV containing mitochondrial SC [72]. According to these results, Lenaz and Genova [71] suggest that oxidative stress acts primarily by disassembling supercomplex associations thereby establishing a vicious circle of oxidative stress and energy failure, ultimately leading to cell damage and disease. It is interesting to mention that there are diverse speciic regulatory proteins for the supramolecular organization of individual complexes that include CIV [73], respiratory SC factors 1 and 2 (Rcf1 and 2) [74], protein Cox interacting (Coi) [75], and COX7a2L [76]. hese proteins down- regulation can impair the formation of SC; for instance, some studies show that diverse pathologies decrease CIV subunit levels afecting stoichiometry and assembly of SC [77, 78]. In addition, diabetes induces mitochondrial ge- nome damage by an increased free radical production de- pleting antioxidant status [79]. Moreover, other structural components as cardiolipin have been shown to be crucial for functionality and SC formation and might be involved in the pathophysiology of diabetes [80]. hus, the impact of complex IV failure and other enzymes may cause an energy crisis due to a lower ATP synthesis and an increased ROS production. Figure 3(a) shows the Coomassie blue staining of the gels for all treatments and the colorimetric enzymatic staining of NADH, Succinate, COX, and ATPase complexes after de- tergent extraction and BN-PAGE or hrCN-PAGE (only for CV). Figure 3(b) clearly indicates that the major form of supercomplexes is present in all samples. In contrast, the amount of free complex I and IV were decreased in D group, and these values did not change in mitochondria isolated from M group. Otherwise, the in-gel activity of complex II was signiicantly lower in C group compared with those in the D and M groups. In addition, the brown bands indicate the presence of complex IV in all groups and its increase in D group (Figure 3(d)). On the other hand, in-gel ATP hydrolysis/lead phosphate precipitation assay revealed bands representing the F1F0monomer and F1F0 dimer bands in Figures 3(e) and 3(f) showing the same functional patterns as the in-solution assays, indicating the level of intrinsic activity driven by complex V. In support, we have shown by comparing the in-gel enzyme activities that the ATPase Biochemistry Research International 7 activity of the F1F0-ATP synthase is speciically and sig- niicantly increased in D group when compared with C group. Our results are concerned with the changes in the amount of CIV subunits; for example, Cox6b1 is involved in the regulation of mitochondrial function by promoting SC formation, suggesting its antiaging efects of calorie re- striction [81]. In addition, heart failure in dogs induced by coronary microembolism resulted in loss of complex IV containing SC of the electron transport chain [77, 78]. Similarly, in RAW 264.7 macrophages, knockdown of either subunit cytochrome c oxidase (CcO) Vb or CcO IV resulted in a signiicant decrease in CcO containing supercomplexes [78]. Liver mitochondria from ethanol-treated rat also showed a lower level of supercomplexes with a concomitant loss of CcO protein [82]. herefore, complex IV has been shown to be necessary for maintaining the stability of complex I in SC, as shown in mouse ibroblast cell lines, where a reduced expression of subunit IVi1 or nonsense mutation in subunit I not only resulted in lower CcO content but also caused signiicant reduction in complex I [83]. Structural defects in complex III also afected the amount of complex I, whereas chemical inhibition did not. Patients with defects in cytochrome b not only lose complex III but also show decreased amounts of complex I, while main- taining a normal enzymatic activity [84]. Conversely, the disruption of complex I function caused by nonsense mu- tations in NDUFS4, a subunit of this large multimeric complex, led to the partial loss of complex III activity in skin ibroblast cultures obtained from Leigh-like patients [85, 86]. However, defects in the complex I subunit ND5 did not cause a loss of complex III in the I-III supercomplex [87]. KDa C D M CIV CI 75 50 37 25 20 15 (a) 3.0 ⁎ ⁎ CIV-MTCO1 2.5 2.0 1.5 1.0 0.5 0.0 Control Diabetic Moringa R el at iv e p ro te in e xp re ss io n (f o ld v er su s co n tr o l) (b) CII-SDHB Control Diabetic Moringa R el at iv e p ro te in e xp re ss io n (f o ld v er su s co n tr o l) 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 (c) CIII-UQCRC2 Control Diabetic Moringa R el at iv e p ro te in e xp re ss io n (f o ld v er su s co n tr o l) 1.2 1.0 0.8 0.6 0.4 0.2 0.0 (d) ⁎⁎ ⁎# CI-NDUFB8 Control Diabetic Moringa R el at iv e p ro te in e xp re ss io n (f o ld v er su s co n tr o l) 60 50 40 30 20 10 0 (e) CV-ATP5A Control Diabetic Moringa R el at iv e p ro te in e xp re ss io n (f o ld v er su s co n tr o l) 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 (f) Figure 2: Characterization of OXPHOS proteins expressed in liver mitochondria during diabetes. (a) OXPHOS cocktail speciicity demonstrated by a Western blot from liver-isolated mitochondria of diabetic rats and treated withMoringa extract. Relative expression of (b)MTC01 subunit of CIV, (c) SDHB subunit of CII, (d) UQCRC2 subunit of CIII, (e) NDUFB8 subunit of CI, and (f) ATP5A subunit of CV was performed by densitometric analysis from gel. Molecular mass standards are shown on the left panel of the gel. C: control; D: diabetic; M:Moringa-treated group. Data are shown as mean band density normalized relative to UQCR2. Signiicant diferences are represented by ∗ P< 0.05 versus C; #P< 0.05 versus D. 8 Biochemistry Research International 3.5. H2O2 Production by Liver Mitochondria Oxidizing Complex I and Complex II Substrates. Several studies have reported that ROS overproduction by mitochondrial ETC is responsible for hyperglycemia-induced oxidative stress and the pathogenesis of diabetic complications [88, 89]; however, it is not clear whether mitochondria of diabetic origin really generate ROS independently of the surrounding diabetic milieu. Herlein et al. [90] showed that the gastrocnemius, heart, and liver mitochondria of STZ-diabetic rats were not irrevocably altered to produce superoxide excess either by complex I or complex III. Moreover, gastrocnemius and heart mitochondria demonstrated an increased respiratory coupling, instead of a decrement. In addition, mitochondria of insulin-deicient diabetic rats did show signs of ROS overproduction. hus, the detailed molecular mecha- nism and sites of ROS generation during diabetes are controversial. In isolated mitochondria, the rate of mitochondrial ROS generation is directly governed by membrane potential (ΔΨm) and pH gradient across the inner membrane, favored only by a state 4 condition [91]. Hence, H2O2 production rate was measured in liver mitochondria using luorescent dye Amplex Red and pyruvate plus malate or succinate, as complex I-III and II-III linked substrates, respectively, and the results are shown in Figure 4. Data presented in Figure 4 (a) show that the major source of ROS is complex I for liver mitochondria of C group, using pyruvate plus malate as substrate, relecting the generation of superoxide anion. In contrast, with M group-isolated mitochondria oxidizing succinate in state 4 produced 6 times more ROS compared with other treatments (C and D groups) (Figure 4(b)). herefore, the mitochondrial ROS production rates varied dramatically among the three experimental groups in re- sponse to addition of respiratory inhibitors. At the level of ROS formation, all groups have the same basal formation using malate plus pyruvate or succinate, but the addition of respiratory inhibitors had a varied efect on ROS production in the diferent groups. In case of the C group, the addition of rotenone and antimycin stimulated ROS production by using pyruvate and malate (Figure 4(a)). However, in both D group and M group, the addition of rotenone has no efect, while antimycin caused only a slight increase in ROS formation (Figure 4(a)). Nevertheless, with succinate, inhibitors have a diferent pattern. In the C group, the addition of rotenone causes no efect on ROS pro- duction, and antimycin favors its increase; but in this case, malonate has no efect (Figure 4(b)). However, in the di- abetic group, antimycin has a greater efect on ROS pro- duction than control, and malonate adversely afects ROS formation (Figure 4(b)). Finally, M group sensitivity to individual training in- hibitors was unaltered in case of rotenone and malonate, but adding antimycin in this treatment favored ROS production (Figure 4(b)). hus, measurements of ROS with Amplex Red cannot be used for sites of ROS generation from liver mi- tochondria treated with STZ and/or M. oleifera. his situ- ation could be attributable to experimental conditions because complex I (rotenone), complex II (malonate), and complex III (antimycin A) inhibitors have been commonly used. However, the inal concentration being used is not stationary, causing experimental errors that are diferent from one method to other. In addition, it is necessary to use other respiratory inhibitors, as stigmatellin andmyxothiazol. No obstant, this does not deny other possible explanations that can afect ROS production as diferences in the stoichiometry-activity ratios of the respiratory complexes [92], the susceptibility to proton pump slip at complex IV [93], or other mechanisms. Damage to complex I, themost vulnerable ETC complex, increases ROS production, leading to a vicious circle of further mitochondrial dysfunction. It is important to note that complex I injury has a stronger impact on mitochon- drial function compared with the damage to other com- plexes because mitochondria possess smaller amounts of complex I than other ETC complexes [94]. Superoxide production by complex I is much higher during reverse electron transport from succinate to NAD+ [95]. In addition, it was found that defective complex I produces more ROS CB SC I V IV II III2 SC I V IV II III2 V2 D M 10% 4% (a) C D M (b) C D M (c) C D M (d) C D M (e) C D M (f) Figure 3: Electrophoretic representative pattern of liver mitochondrial solubilized of the diferent groups (5 g de digitonin/g protein). (a) Blue native polyacrylamide gel electrophoresis (BN-PAGE) stained with Coomassie blue; (b) complex I in-gel activity; (c) complex II in- gel activity; (d) complex IV in-gel activity; (e) complex IV in-gel activity; (f) high resolution clear native polyacrylamide gel electrophoresis (hrCN-PAGE) of complex V. B: bovine heart solubilized mitochondria (positive control); C: control; D: diabetic; M: Moringa-treated diabetic groups. Biochemistry Research International 9 [96], suggesting that structural modiications of the enzyme may play a crucial role in ROS production process. Recently, it was reported that pancreatic mitochondrial complex I showed aberrant hyperactivity in type 1 and 2 STZ-diabetic mice and rat and in cultured β cells [97]. Further experi- ments focusing on STZ-induced diabetes in rats revealed that complex I′s hyperactivity could be attenuated by metformin. Interestingly, in this study, no changes were reported in complex I activity in brain, liver, and heart by BN-PAGE [97]. However, the reason why complex I activity did not exhibit detectable increases in these tissues is unknown. Our results in Figure 1(b) show that complex I activity was signiicantly higher in diabetes than in healthy in- dividuals. his increased activity was apparently contrib- uted by an increased NDUFB8 subunit protein content as shown in Figure 2(e). Moreover, the complex I hyperac- tivity also imposed pressure in complex IV (Figures 2(b) and 3(d)). hese indings suggest that these elevated ac- tivities could be attributed for ROS production, given that higher ETC activity can also increase mitochondrial ROS generation [98, 99]. However, our results on speciic sites of ROS generation along with ETC are controversial (Figure 4). his may explain why in some tissues seem that inhibition of electron transfer at complex I (by rotenone) may generate an increase in radical formation, whereas, in others, rote- none will reduce radical generation by preventing passage of electron further into the distal part of the chain. However, the basis for such a diference is obscure and presumed to be related to ΔΨm changes and radicals leakage across the membranes [100]. Polyphenols have been traditionally viewed as antioxi- dants; however, increasing evidence has emerged supporting the ability of certain polyphenols to exert numerous ROS- scavenging independent actions. hen all these natural compounds modulate mitochondrial functions by inhibiting organelle enzymes or metabolic pathways, by altering the production of ROS and modulating the activity of tran- scription factors which regulate the expression of mito- chondrial proteins [101]. hus, some particular polyphenols are now recognized as molecules capable of modulate pathways that deine mitochondrial biogenesis (i.e., in- ducing sirtuins), mitochondrial membrane potential (i.e., mitochondrial permeability transition pore opening and uncoupling efects), mitochondrial electron trans- port chain and ATP synthesis (i.e., modulating complex I to V activity), intramitochondrial oxidative status (i.e., inhibiting/inducing ROS formation/removal enzymes), and ultimately mitochondrial-triggered cell death (modu- lating intrinsic apoptosis) (review in [102]). hus, some studies have indicated that mitochondria may be the target organelle of phenolic compounds [103, 104]. Recently, it was reported that galangin (natural lavonoid) could maintain liver mitochondrial function in diabetic rats through oxi- dative stress reduction and both antioxidant enzymes and respiratory complexes activities enhancement [79]. here- fore, the likely role of mitochondrial ROS in diabetes has led to eforts for developing efective antioxidant compounds targeted to mitochondria. his study was designed to investigate the protective efect ofM. oleifera on liver bioenergetics and to elucidate its potential mechanism.M. oleifera resulted in a well-preserved mitochondrial redox potential, signiicantly by elevating heme oxygenase-1 and decreasing ROS formation and lip- operoxidation. hese observations indicated that STZ- induced mitochondrial oxidative damage was remarkably attenuated. hus, to our knowledge, we have shown for the irst time that M. oleifera extract modulates mitochondrial respiratory activity, an efect that may account for some of the protective properties of phytochemicals. hese efects may be of physiological signiicance since it seems that some phytochemicals are concentrated into mitochondria. he 160 140 120 100 80 60 40 20 0 C D M A m p le x u lt ra re d /r es o ru i n l u o re sc en ce ( U A ) R R R R + A R + A R + A (a) R R + A R + A + M I R R + A R R + A R + A + M I R + A + M I C D M 200 180 160 140 120 100 80 60 40 20 0 A m p le x u lt ra re d /r es o ru i n fl u o re sc en ce ( U A ) (b) Figure 4: H2O2 production by mitochondrial respiratory chain measured by Amplex Red (UA) in liver mitochondria from diferent treatments (C: control; D: diabetic; M:Moringa-treated diabetic groups) oxidizing (a) pyruvate plus malate or (b) succinate as substrates and the efects of respiratory chain inhibitors (R: rotenone; A: antimycin A; Ml: malonate). Mitochondria were studied during state 4 respiration. To correct for the increase in background luorescence of the Amplex Red/HRP detection system overtime, luorescence was monitored for a period of ten minutes. his background was subtracted from resoruin trace. Data are means± SEM (n� 5). 10 Biochemistry Research International results also support a pharmacological use of M. oleifera extract in drug to reduce mitochondrial damage in vivo. However, the details about mechanism of action require further investigation. 4. Conclusions We provide experimental evidence indicating that M. olei- fera extract targeting mitochondria can be used therapeu- tically to alleviate diabetes. herefore, it will be important to identify regulatory proteins involved in the adjustment of respiratory chain complex organization/activity in response to altered redox state. In liver, the alteration of mitochon- drial enzymatic activities and oxidative stress induced by STZ suggested of a compensatory response. In addition, M. oleifera extract upregulated mitochondrial genes linked with respiratory chain. Our data show an increased mitochon- drial function and activity/expression of respiratory com- plexes in liver of STZ-diabetic rats, which can be normalized by M. oleifera at levels that do not markedly alter the consequences of hyperglycemia. Conflicts of Interest he authors declare no conlicts of interest. Authors’ Contributions Erick Sierra-Campos, Mónica Andrea Valdez-Solana, and Óscar Flores-Herrera contributed to conceptualization, lit- erature review, and writing the original draft. Maŕıa Ale- jandra Sánchez-Muñoz, Mara Ibeth Campos-Almazán, and Guadalupe Garćıa-Arenas conducted the animal studies and performed the experiments. Erick Sierra-Campos, Maŕıa Alejandra Sánchez-Muñoz, Mercedes Esparza-Perusquı́a, Soia Olvera-Sánchez, and Óscar Flores-Herrera analyzed the data. Óscar Flores-Herrera, Alfredo Téllez-Valencia, Claudia Avitia-Domı́nguez, and Mónica Andrea Valdez- Solana contributed reagents/material/analysis tools. Maŕıa Alejandra Sánchez-Muñoz and Erick Sierra-Campos wrote the paper. All authors read and approved the inal manuscript. Acknowledgments he authors would like to express their sincere gratitude to the Consejo Nacional de Ciencia y Tecnologı́a (Conacyt, México) for inancial support to Erick Sierra-Campos (Grant 268184). he authors thank Héctor Vázquez-Meza for providing technical assistance. Maŕıa Alejandra Sánchez Muñoz is also grateful to Conacyt, México, for the inancial support for her Masters studies (Grant 601630). he authors are also grateful to the Mexican Moringa oleifera producers (Akuanandi) for providing all samples for this study. Supplementary Materials Figure S1: Possible mitochondrial processes that are mod- ulated byMoringa oleifera extract. It is now well established that the individual mitochondrial respiratory complexes can be organized into supercomplexes, but the composition and abundance of these may vary among organisms and tissues depending on the metabolic and physiological conditions. Alteration of mitochondrial electron transport chain is a recognized hallmark of the diabetic-associated decline in liver bioenergetics; however, the molecular events involved are only poorly understood.Moringa oleifera is used for the treatment of diabetes. However, its role on mitochondrial functionality is not yet established. his study was aimed to evaluate the efect of M. oleifera extract on supercomplex formation, ATPase activity, ROS production, GSH levels, lipid peroxidation, and protein carbonylation. he levels of lipid peroxidation and protein carbonylation were increased in the diabetic group. However, the levels were decreased in M. oleifera-treated diabetic rats. Analysis of in-gel activity showed an increase in all complexes activities in the diabetic group, but spectrophotometric determinations of complex II and IV activities were unafected in this treatment. However, we found an oxygen consumption abolition through com- plex I-III-IV pathway in the diabetic group treated with Moringa. Respiration with succinate feeding into complex II-III-IV was increased in the diabetic group. We have shown for the irst time that M. oleifera extract modulates mitochondrial respiratory activity, an efect that may ac- count for some of the protective properties of phyto- chemicals. hese efects may be of physiological signiicance since it seems that some phytochemicals are concentrated into mitochondria. he results also support a pharmaco- logical use of M. oleifera extract in drug to reduce mito- chondrial damage in vivo. (Supplementary Materials) References [1] L. D. Osellame, T. S. Blacker, and M. R. Duchen, “Cellular and molecular mechanisms of mitochondrial function,” Best Practice & Research Clinical Endocrinology & Metabolism, vol. 26, no. 6, pp. 711–723, 2012. [2] D. C. Chan, “Mitochondria: dynamic organelles in disease, aging, and development,” Cell, vol. 125, no. 7, pp. 1241–1252, 2006. [3] C. Greggio, P. Jha, S. S. Kulkarni et al., “Enhanced respiratory chain supercomplex formation in response to exercise in human skeletal muscle,” Cell Metabolism, vol. 25, no. 2, pp. 301–311, 2017. [4] S. J. Ramı́rez-Aguilar, M. Keuthe, M. Rocha et al., “he composition of plant mitochondrial supercomplexes changes with oxygen availability,” Journal of Biological Chemistry, vol. 286, no. 50, pp. 43045–43053, 2011. [5] A. O. Helbig, M. J. L. de Groot, R. A. van Gestel et al., “A three-way proteomics strategy allows diferential analysis of yeast mitochondrial membrane protein complexes under anaerobic and aerobic conditions,” Proteomics, vol. 9, no. 20, pp. 4787–4798, 2009. [6] L. Böttinger, S. E. Horvath, T. Kleinschroth et al., “Phos- phatidylethanolamine and cardiolipin diferentially afect the stability of mitochondrial respiratory chain super- complexes,” Journal of Molecular Biology, vol. 423, no. 5, pp. 677–686, 2012. [7] J. E. Vance and G. Tasseva, “Formation and function of phosphatidylserine and phosphatidylethanolamine in mammalian cells,” Biochimica et Biophysica Acta (BBA)- Biochemistry Research International 11 Molecular and Cell Biology of Lipids, vol. 1831, no. 3, pp. 543–554, 2013. [8] A. Görlach, E. Y. Dimova, A. Petry et al., “Reactive oxygen species, nutrition, hypoxia and diseases: problems solved?,” Redox Biology, vol. 6, pp. 372–385, 2015. [9] M. P. Murphy and R. A. J. Smith, “Targeting antioxidants to mitochondria by conjugation to lipophilic cations,” Annual Review of Pharmacology and Toxicology, vol. 47, pp. 629–656, 2007. [10] B. B. Lowell and G. I. Shulman, “Mitochondrial dysfunction and type 2 diabetes,” Science, vol. 307, no. 5708, pp. 384–387, 2005. [11] K. Morino, K. F. Petersen, and G. I. Shulman, “Molecular mechanisms of insulin resistance in humans and their po- tential links with mitochondrial dysfunction,” Diabetes, vol. 55, no. 2, pp. S9–S15, 2006. [12] J. A. Brignone, C. M. C. de Brignone, R. R. Rodriguez, B. N. Badano, and A. O. M. Stoppani, “Modiied oscillation behavior and decreased D-3-hydroxybutyrate dehydroge- nase activity in diabetic rat liver mitochondria,” Archives of Biochemistry and Biophysics, vol. 214, no. 2, pp. 581–588, 1982. [13] J. A. Brignone, C. M. C. de Brignone, C. R. Ricci, I. R. de Mignone, M. C. Susemihl, and R. R. Rodŕıguez, “Favourable, signiicant efect of the dose-dependent treatment with RU 38486 (RU) on the alterations of the hepatic mitochondrial function of diabetic rats,” Diabetes Research and Clinical Practice, vol. 32, no. 3, pp. 141–148, 1996. [14] H. Raza and A. John, “Implications of altered glutathione metabolism in aspirin-induced oxidative stress and mito- chondrial dysfunction in HepG2 cells,” PLoS One, vol. 7, no. 4, article e36325, 2012. [15] G. K. Bandyopadhyay, G. Y. Joseph, J. Ofrecio, and J. M. Olefsky, “Increased malonyl-CoA levels in muscle from obese and type 2 diabetic subjects lead to decreased fatty acid oxidation and increased lipogenesis; thiazolidinedione treat- ment reverses these defects,” Diabetes, vol. 55, no. 8, pp. 2277–2285, 2006. [16] J. G. Duncan, “Mitochondrial dysfunction in diabetic car- diomyopathy,” Biochimica et Biophysica Acta (BBA)- Molecular Cell Research, vol. 1813, no. 7, pp. 1351–1359, 2011. [17] L. Rato, A. I. Duarte, G. D. Tomás et al., “Pre-diabetes alters testicular PGC1-α/SIRT3 axis modulating mitochondrial bioenergetics and oxidative stress,” Biochimica et Biophysica Acta (BBA)-Bioenergetics, vol. 1837, no. 3, pp. 335–344, 2014. [18] W. C. Parks and R. L. Drake, “Insulin mediates the stimu- lation of pyruvate kinase by a dual mechanism,” Biochemical Journal, vol. 208, no. 2, pp. 333–337, 1982. [19] G. Antoun, F. McMurray, A. B. hrush et al., “Erratum to: Impaired mitochondrial oxidative phosphorylation and supercomplex assembly in rectus abdominis muscle of di- abetic obese individuals,” Diabetologia, vol. 59, no. 2, pp. 396-397, 2016. [20] C. D. Berdanier, H. B. Everts, C. Hermoyian, and C. E. Mathews, “Role of vitamin A in mitochondrial gene expression,” Diabetes Research and Clinical Practice, vol. 54, no. 2, pp. S11–S27, 2001. [21] D. Loven, H. Schedl, H. Wilson et al., “Efect of insulin and oral glutathione on glutathione levels and superoxide dis- mutase activities in organs of rats with streptozocin-induced diabetes,” Diabetes, vol. 35, no. 5, pp. 503–507, 1986. [22] P. S. Samiec, C. Drews-Botsch, E. W. Flagg et al., “Gluta- thione in human plasma: decline in association with aging, age-related macular degeneration, and diabetes,” Free Rad- ical Biology and Medicine, vol. 24, no. 5, pp. 699–704, 1998. [23] A. Abd El Latif, B. E. S. El Bialy, H. D. Mahboub, and M. A. Abd Eldaim, “Moringa oleifera leaf extract ameliorates alloxan-induced diabetes in rats by regeneration of β cells and reduction of pyruvate carboxylase expression,” Bio- chemistry and Cell Biology, vol. 92, no. 5, pp. 413–419, 2014. [24] S. Sreelatha and P. R. Padma, “Antioxidant activity and total phenolic content of Moringa oleifera leaves in two stages of maturity,” Plant Foods for Human Nutrition, vol. 64, no. 4, pp. 303–311, 2009. [25] P. Chumark, P. Khunawat, Y. Sanvarinda et al., “he in vitro and ex vivo antioxidant properties, hypolipidaemic and antiatherosclerotic activities of water extract of Moringa oleifera Lam. leaves,” Journal of Ethnopharmacology, vol. 116, no. 3, pp. 439–446, 2008. [26] A. L. Al-Malki and H. A. El Rabey, “he antidiabetic efect of low doses of Moringa oleifera Lam. seeds on streptozotocin induced diabetes and diabetic nephropathy in male rats,” BioMed Research International, vol. 2015, Article ID 381040, 13 pages, 2015. [27] D. J. Dorta, A. A. Pigoso, F. E. Mingatto et al., “he in- teraction of lavonoids with mitochondria: efects on ener- getic processes,” Chemico-Biological Interactions, vol. 152, no. 2-3, pp. 67–78, 2005. [28] I. Elingold, M. P. Isollabella, M. B. Casanova et al., “Mito- chondrial toxicity and antioxidant activity of a prenylated lavonoid isolated from Dalea elegans,” Chemico-Biological Interactions, vol. 171, no. 3, pp. 294–305, 2008. [29] R. Lagoa, I. Graziani, C. Lopez-Sanchez, V. Garcia-Martinez, and C. Gutierrez-Merino, “Complex I and cytochrome c are molecular targets of lavonoids that inhibit hydrogen per- oxide production by mitochondria,” Biochimica et Bio- physica Acta (BBA)-Bioenergetics, vol. 1807, no. 12, pp. 1562–1572, 2011. [30] N. Madi, M. Dany, S. Abdoun, and J. Usta, “Moringa oleifera’s nutritious aqueous leaf extract has anticancerous efects by compromising mitochondrial viability in an ROS-dependent manner,” Journal of the American College of Nutrition, vol. 35, no. 7, pp. 604–613, 2016. [31] W. Khan, R. Parveen, K. Chester, S. Parveen, and S. Ahmad, “Hypoglycemic potential of aqueous extract of Moringa oleifera leaf and in vivo GC-MS metabolomics,” Frontiers in Pharmacology, vol. 8, p. 577, 2017. [32] R. Hovius, H. Lambrechts, K. Nicolay, and B. de Kruijf, “Improved methods to isolate and subfractionate rat liver mitochondria. Lipid composition of the inner and outer mem- brane,” Biochimica et Biophysica Acta (BBA)-Biomembranes, vol. 1021, no. 2, pp. 217–226, 1990. [33] E. Sierra-Campos, I. Velazquez, D. Matuz-Mares, A. Villavicencio-Queijeiro, and J. P. Pardo, “Functional properties of the Ustilago maydis alternative oxidase under oxidative stress conditions,” Mitochondrion, vol. 9, no. 2, pp. 96–102, 2009. [34] D. D. L. R. Castillo, M. Zarco-Zavala, S. Olvera-Sanchez et al., “Atypical cristae morphology of human syncytiotrophoblast mitochondria role for complex V,” Journal of Biological Chemistry, vol. 286, no. 27, pp. 23911–23919, 2011. [35] M. Esparza-Perusquı́a, S. Olvera-Sánchez, J. P. Pardo, G. Mendoza-Hernández, F. Mart́ınez, and O. Flores-Herrera, “Structural and kinetics characterization of the F1F0-ATP synthase dimer. New repercussion of monomer-monomer contact,” Biochimica et Biophysica Acta (BBA)-Bioenergetics, vol. 1858, no. 12, pp. 975–981, 2017. 12 Biochemistry Research International [36] H. Schägger and G. von Jagow, “Blue native electrophoresis for isolation of membrane protein complexes in enzymati- cally active form,” Analytical Biochemistry, vol. 199, no. 2, pp. 223–231, 1991. [37] I. Wittig, M. Karas, and H. Schägger, “High resolution clear native electrophoresis for in-gel functional assays and luorescence studies of membrane protein complexes,” Molecular & Cellular Proteomics, vol. 6, no. 7, pp. 1215–1225, 2007. [38] I. Wittig and H. Schägger, “Electrophoretic methods to isolate protein complexes from mitochondria,” Methods in Cell Biology, vol. 80, pp. 723–741, 2007. [39] S. J. Couoh-Cardel, S. Uribe-Carvajal, S. Wilkens, and J. J. Garcı́a-Trejo, “Structure of dimeric F1F0-ATP synthase,” Journal of Biological Chemistry, vol. 285, no. 47, pp. 36447– 36455, 2010. [40] U. K. Laemmli, “Cleavage of structural proteins during the assembly of the head of bacteriophage T4,” Nature, vol. 227, no. 5259, pp. 680–685, 1970. [41] O. H. Lowry, N. J. Rosebrough, A. L. Farr, and R. J. Randall, “Protein measurement with the Folin phenol reagent,” Journal of Biological Chemistry, vol. 193, no. 1, pp. 265–275, 1951. [42] O. Yilmaz, S. Keser, M. Tuzcu et al., “A practical HPLC method to measure reduced (GSH) and oxidized (GSSG) glutathione concentrations in animal tissues,” Journal of Animal and Veterinary Advances, vol. 8, no. 2, pp. 343–347, 2009. [43] J. A. Buege and S. D. Aust, “[30] Microsomal lipid perox- idation,”Methods in Enzymology, vol. 52, pp. 302–310, 1978. [44] A. A. Starkov, “Measurement of mitochondrial ROS pro- duction,”Methods in Molecular Biology, vol. 648, pp. 245–2553 , 2010. [45] R. L. Levine, D. Garland, C. N. Oliver et al., “[49] De- termination of carbonyl content in oxidatively modiied proteins,” Methods in Enzymology, vol. 186, pp. 464–478, 1990. [46] P. Naughton, R. Foresti, S. K. Bains, M. Hoque, C. J. Green, and R. Motterlini, “Induction of heme oxygenase 1 by nitrosative stress. A role for nitroxyl anion,” Journal of Bi- ological Chemistry, vol. 277, no. 43, pp. 40666–40674, 2002. [47] P. Kiranmayi and B. Babitha, “Efect ofMoringa oleifera Leaf powder on blood glucose levels in type II diabetes mellitus patients,”Vegetos-An International Journal of Plant Research, vol. 24, no. 2, pp. 114–116, 2011. [48] P. Anthanont, N. Lumlerdkij, P. Akarasereenont, S. Vannasaeng, and A. Sriwijitkamol, “Moringa oleifera leaf increases insulin secretion after single dose administration: a preliminary study in healthy subjects,” Journal of the Medical Association of hailand, vol. 99, no. 3, pp. 308–313, 2016. [49] A. C. Maritim, R. A. Sanders, and J. B. Watkins III, “Efects of α-lipoic acid on biomarkers of oxidative stress in streptozotocin-induced diabetic rats,” Journal of Nutritional Biochemistry, vol. 14, no. 5, pp. 288–294, 2003. [50] A. C. Maritim, R. A. Sanders, and J. B.Watkins III, “Diabetes, oxidative stress, and antioxidants: a review,” Journal of Biochemical andMolecular Toxicology, vol. 17, no. 1, pp. 24–38, 2003. [51] A. B. Egorova, Y. A. Uspenskaya, and V. P. Nefedov, “NAD and glutathione modulate sensitivity of bone marrow cells to oxidative stress,” Bulletin of Experimental Biology and Medicine, vol. 132, no. 1, pp. 637–640, 2001. [52] W. Völkel, R. Alvarez-Sánchez, I. Weick, A. Mallya, W. Dekanta, and A. Pähler, “Glutathione conjugates of 4-hydroxy-2 (E)-nonenal as biomarkers of hepatic oxidative stress-induced lipid peroxidation in rats,” Free Radical Biology and Medicine, vol. 38, no. 11, pp. 1526–1536, 2005. [53] R. Tenhunen, H. S. Marver, and R. Schmid, “he enzymatic conversion of heme to bilirubin by microsomal heme oxy- genase,” Proceedings of the National Academy of Sciences, vol. 61, no. 2, pp. 748–755, 1968. [54] R. K. Kutty and M. D. Maines, “Puriication and charac- terization of biliverdin reductase from rat liver,” Journal of Biological Chemistry, vol. 256, no. 8, pp. 3956–3962, 1981. [55] M. A. Di Noia, S. Van Driesche, F. Palmieri et al., “Heme oxygenase-1 enhances renal mitochondrial transport carriers and cytochrome c oxidase activity in experimental diabetes,” Journal of Biological Chemistry, vol. 281, no. 23, pp. 15687– 15693, 2006. [56] D. Jaiswal, P. K. Rai, S. Mehta et al., “Role ofMoringa oleifera in regulation of diabetes-induced oxidative stress,” Asian Paciic Journal of Tropical Medicine, vol. 6, no. 6, pp. 426– 432, 2013. [57] R. Gupta, M. Mathur, V. K. Bajaj et al., “Evaluation of an- tidiabetic and antioxidant activity of Moringa oleifera in experimental diabetes,” Journal of Diabetes, vol. 4, no. 2, pp. 164–171, 2012. [58] A. F. Santos, A. C. Argolo, P. M. Paiva, and L. C. Coelho, “Antioxidant activity of Moringa oleifera tissue extracts,” Phytotherapy Research, vol. 26, no. 9, pp. 1366–1370, 2012. [59] C. S. T. Origassa and N. O. S. Câmara, “Cytoprotective role of heme oxygenase-1 and heme degradation derived end products in liver injury,”World Journal of Hepatology, vol. 5, no. 10, p. 541, 2013. [60] D. Jaiswal, P. K. Rai, A. Kumar, S. Mehta, and G. Watal, “Efect of Moringa oleifera Lam. leaves aqueous extract therapy on hyperglycemic rats,” Journal of Ethno- pharmacology, vol. 123, no. 3, pp. 392–396, 2009. [61] S. Adisakwattana and B. Chanathong, “Alpha-glucosidase inhibitory activity and lipid-lowering mechanisms of Mor- inga oleifera leaf extract,” European Review for Medical and Pharmacological Sciences, vol. 15, no. 7, pp. 803–808, 2011. [62] F. M. Ferreira, C. M. Palmeira, R. Seiça, A. J. Moreno, and M. S. Santos, “Diabetes and mitochondrial bioenergetics: alterations with age,” Journal of Biochemical and Molecular Toxicology, vol. 17, no. 4, pp. 214–222, 2003. [63] S. Ardestani, D. L. Deskins, and P. P. Young, “Membrane TNF-alpha-activated programmed necrosis is mediated by Ceramide-induced reactive oxygen species,” Journal of Molecular Signaling, vol. 8, p. 12, 2013. [64] D.-D. Zhao, N. Yu, X.-K. Li et al., “Antidiabetic and anti- oxidative efect of Jiang Tang Xiao Ke granule in high-fat diet and low-dose streptozotocin induced diabetic rats,” Evidence-Based Complementary and Alternative Medicine, vol. 2014, Article ID 475192, 8 pages, 2014. [65] E. K. Ainscow and M. D. Brand, “Top-down control analysis of ATP turnover, glycolysis and oxidative phosphorylation in rat hepatocytes,” FEBS Journal, vol. 263, no. 3, pp. 671–685, 1999. [66] C. Desler, T. L. Hansen, J. B. Frederiksen, M. L. Marcker, K. K. Singh, and L. J. Rasmussen, “Is there a link between mitochondrial reserve respiratory capacity and aging?,” Journal of Aging Research, vol. 2012, Article ID 192503, 9 pages, 2012. [67] D. A. Rendon, K. Kotedia, S. F. Afshar et al., “Mapping radiation injury and recovery in bone marrow using 18F-FLT Biochemistry Research International 13 PET/CTand USPIO MRI in a rat model,” Journal of Nuclear Medicine, vol. 57, no. 2, pp. 266–271, 2016. [68] T. Lobo-Jarne and C. Ugalde, “Respiratory chain super- complexes: structures, function and biogenesis,” Seminars in Cell & Developmental Biology, 2017, in press. [69] R. Guo, S. Zong, M.Wu, J. Gu, andM. Yang, “Architecture of human mitochondrial respiratory megacomplex I2III2IV2,” Cell, vol. 170, no. 6, pp. 1247–1257, 2017. [70] D. R. Winge, “Sealing the mitochondrial respirasome,” Molecular and Cellular Biology, vol. 32, no. 14, pp. 2647– 2652, 2012. [71] G. Lenaz and M. L. Genova, “Supramolecular organisation of the mitochondrial respiratory chain: a new challenge for the mechanism and control of oxidative phosphorylation,” Advances in Experimental Medicine and Biology, vol. 748, pp. 107–144, 2012. [72] G. Antoun, F. McMurray, A. B. hrush et al., “Impaired mitochondrial oxidative phosphorylation and supercomplex assembly in rectus abdominis muscle of diabetic obese in- dividuals,” Diabetologia, vol. 58, no. 12, pp. 2861–2866, 2015. [73] C. Oswald, U. Krause-Buchholz, and G. Rödel, “Knockdown of human COX17 afects assembly and supramolecular or- ganization of cytochrome c oxidase,” Journal of Molecular Biology, vol. 389, no. 3, pp. 470–479, 2009. [74] C. R. Lundin, C. von Ballmoos, M. Ott, P. Ädelrotha, and P. Brzezinski, “Regulatory role of the respiratory super- complex factors in Saccharomyces cerevisiae,” Proceedings of the National Academy of Sciences of the United States of America, vol. 113, no. 31, pp. E4476–E4485, 2016. [75] R. K. Singhal, C. Kruse, J. Heidler et al., “Coi1 is a novel assembly factor of the yeast complex III–complex IV supercomplex,”Molecular Biology of the Cell, vol. 28, no. 20, pp. 2609–2622, 2017. [76] R. Pérez-Pérez, T. Lobo-Jarne, D. Milenkovic et al., “COX7A2L is a mitochondrial complex III binding protein that stabilizes the III2 + IV supercomplex without afecting respirasome formation,” Cell Reports, vol. 16, no. 9, pp. 2387–2398, 2016. [77] M. G. Rosca, E. J. Vazquez, J. Kerner et al., “Cardiac mi- tochondria in heart failure: decrease in respirasomes and oxidative phosphorylation,”Cardiovascular Research, vol. 80, no. 1, pp. 30–39, 2008. [78] D. Galati, S. Srinivasan, H. Raza et al., “Role of nuclear- encoded subunit Vb in the assembly and stability of cyto- chrome c oxidase complex: implications in mitochondrial dysfunction and ROS production,” Biochemical Journal, vol. 420, no. 3, pp. 439–449, 2009. [79] A. A. Aloud, C. Veeramani, C. Govindasamy, M. A. Alsaif, A. S. El Newehy, and K. S. Al-Numair, “Galangin, a dietary lavonoid, improves antioxidant status and reduces hyperglycemia-mediated oxidative stress in streptozotocin- induced diabetic rats,” Redox Report, vol. 22, no. 6, pp. 290–300, 2017. [80] V. B. Ritov, E. V. Menshikova, K. Azuma et al., “Deiciency of electron transport chain in human skeletal muscle mito- chondria in type 2 diabetes mellitus and obesity,” American Journal of Physiology-Endocrinology and Metabolism, vol. 298, no. 1, pp. E49–E58, 2010. [81] S.-E. Kim, R. Mori, T. Komatsu et al., “Upregulation of cytochrome c oxidase subunit 6b1 (Cox6b1) and formation of mitochondrial supercomplexes: implication of Cox6b1 in the efect of calorie restriction,”Age, vol. 37, no. 3, p. 45, 2015. [82] S. Bansal, M. Siddarth, D. Chawla, B. D. Banerjee, S. V. Madhu, and A. K. Tripathi, “Advanced glycation end products enhance reactive oxygen and nitrogen species generation in neutrophils in vitro,” Molecular and Cellular Biochemistry, vol. 361, no. 1-2, pp. 289–296, 2012. [83] Y. Li, J.-S. Park, J.-H. Deng, and Y. Bai, “Cytochrome c oxidase subunit IV is essential for assembly and respiratory function of the enzyme complex,” Journal of Bioenergetics and Biomembranes, vol. 38, no. 5-6, pp. 283–291, 2006. [84] H. Schägger, R. de Coo, M. F. Bauer, S. Hofmann, C. Godinot, and U. Brandt, “Signiicance of respirasomes for the assembly/stability of human respiratory chain complex I,” Journal of Biological Chemistry, vol. 279, no. 35, pp. 36349–36353, 2004. [85] S. Budde, L. van den Heuvel, R. Smeets et al., “Clinical heterogeneity in patients with mutations in the NDUFS4 gene of mitochondrial complex I,” Journal of Inherited Metabolic Disease, vol. 26, no. 8, pp. 813–815, 2003. [86] S. Scacco, V. Petruzzella, S. Budde et al., “Pathological mutations of the human NDUFS4 gene of the 18-kDa (AQDQ) subunit of complex I afect the expression of the protein and the assembly and function of the complex,” Journal of Biological Chemistry, vol. 278, no. 45, pp. 44161– 44167, 2003. [87] P. Cardol, L. Boutafala, S. Memmi, B. Devreese, R. F. Matagnea, and C. Remaclea, “In Chlamydomonas, the loss of ND5 subunit prevents the assembly of whole mito- chondrial complex I and leads to the formation of a low abundant 700 kDa subcomplex,” Biochimica et Biophysica Acta (BBA)-Bioenergetics, vol. 1777, no. 4, pp. 388–396, 2008. [88] M. Brownlee, “he pathobiology of diabetic complications,” Diabetes, vol. 54, no. 6, pp. 1615–1625, 2005. [89] M. R. Duchen, “Roles of mitochondria in health and disease,” Diabetes, vol. 53, no. 1, pp. S96–S102, 2004. [90] J. A. Herlein, B. D. Fink, Y. O’Malley, and W. I. Sivitz, “Superoxide and respiratory coupling in mitochondria of insulin-deicient diabetic rats,” Endocrinology, vol. 150, no. 1, pp. 46–55, 2008. [91] M. D. Brand, C. Afourtit, T. C. Esteves et al., “Mitochondrial superoxide: production, biological efects, and activation of uncoupling proteins,” Free Radical Biology and Medicine, vol. 37, no. 6, pp. 755–767, 2004. [92] L. K. Kwong and R. S. Sohal, “Substrate and site speciicity of hydrogen peroxide generation in mouse mitochondria,” Archives of Biochemistry and Biophysics, vol. 350, no. 1, pp. 118–126, 1998. [93] B. Kadenbach, “Intrinsic and extrinsic uncoupling of oxi- dative phosphorylation,” Biochimica et Biophysica Acta (BBA)-Bioenergetics, vol. 1604, no. 2, pp. 77–94, 2003. [94] A. Musatov and N. C. Robinson, “Susceptibility of mito- chondrial electron-transport complexes to oxidative damage. Focus on cytochrome c oxidase,” Free Radical Research, vol. 46, no. 11, pp. 1313–1326, 2012. [95] A. J. Lambert and M. D. Brand, “Inhibitors of the quinone- binding site allow rapid superoxide production from mito- chondrial NADH: ubiquinone oxidoreductase (complex I),” Journal of Biological Chemistry, vol. 279, no. 38, pp. 39414– 39420, 2004. [96] S. Raha and B. H. Robinson, “Mitochondria, oxygen free radicals, disease and ageing,” Trends in Biochemical Sciences, vol. 25, no. 10, pp. 502–508, 2000. [97] J. Wu, X. Luo, N. hangthaeng et al., “Pancreatic mito- chondrial complex I exhibits aberrant hyperactivity in di- abetes,” Biochemistry and Biophysics Reports, vol. 11, pp. 119–129, 2017. 14 Biochemistry Research International [98] H. Raza, S. K. Prabu, A. John, and N. G. Avadhani, “Impaired mitochondrial respiratory functions and oxidative stress in streptozotocin-induced diabetic rats,” International Journal of Molecular Sciences, vol. 12, no. 5, pp. 3133–3147, 2011. [99] J. Škrha Jr., M. Kalousová, J. Švarcová et al., “Relationship of soluble RAGE and RAGE ligands HMGB1 and EN-RAGE to endothelial dysfunction in type 1 and type 2 diabetes mel- litus,” Experimental and Clinical Endocrinology & Diabetes, vol. 120, no. 5, pp. 277–281, 2012. [100] O. Demin, H. Westerhof, and B. Kholodenko, “Mathe- matical modelling of superoxide generation with the bc1 complex of mitochondria,” Biochemistry. Biokhimiia, vol. 63, no. 6, pp. 634–649, 1998. [101] L. Gibellini, E. Bianchini, S. De Biasi, M. Nasi, A. Cossarizza, and M. Pinti, “Natural compounds modulating mitochon- drial functions,” Evidence-Based Complementary and Al- ternative Medicine, vol. 2015, Article ID 527209, 13 pages, 2015. [102] C. Sandoval-Acuña, J. Ferreira, and H. Speisky, “Polyphenols and mitochondria: an update on their increasingly emerging ROS-scavenging independent actions,” Archives of Bio- chemistry and Biophysics, vol. 559, pp. 75–90, 2014. [103] L. Raudone, D. Burdulis, R. Raudonis et al., “Efect of Perilla Frutescens extracts and rosmarinic acid on rat heart mito- chondrial functions,” Acta Poloniae Pharmaceutica, vol. 73, no. 1, pp. 135–145, 2016. [104] G. B. Melo, R. L. Silva, V. A.Melo et al., “Efect of the aqueous extract of Hyptis pectinata on liver mitochondrial respira- tion,” Phytomedicine, vol. 12, no. 5, pp. 359–362, 2005. Biochemistry Research International 15 Composition Comments 1. We have rephrased the sentence “In addition, impairment ... has been reported” for grammatical correctness and clarity. Please conirm that this is your intended meaning. 2. Note that citation for Supplemetary materials section has been added here. Please conirm. 3. Please conirm the journal title for References [44] and [71]. 16 Biochemistry Research International It is very important to conirm the author(s) last and irst names in order to be displayed correctly on our website as well as in the indexing databases: Author 1 Given Names: Maŕıa Last Name: Alejandra Sánchez-Muñoz Author 2 Given Names: Mónica Andrea Last Name: Valdez-Solana Author 3 Given Names: Mara Ibeth Last Name: Campos-Almazán Author 4 Given Names: Óscar Last Name: Flores-Herrera Author 5 Given Names: Mercedes Last Name: Esparza-Perusquı́a Author 6 Given Names: Soia Last Name: Olvera-Sánchez Author 7 Given Names: Guadalupe Last Name: Garćıa-Arenas Author 8 Given Names: Claudia Last Name: Avitia-Domı́nguez Author 9 Given Names: Alfredo Last Name: Téllez-Valencia Author 10 Given Names: Erick Last Name: Sierra-Campos It is also very important for each author to provide an ORCID (Open Researcher and Contributor ID). ORCID aims to solve the name ambiguity problem in scholarly communications by creating a registry of persistent unique identiiers for individual researchers. To register an ORCID, please go to the Account Update page (http://mts.hindawi.com/update/) in our Manuscript Tracking System and after you have logged in click on the ORCID link at the top of the page.his link will take you to the ORCID website where you will be able to create an account for yourself. Once you have done so, your new ORCID will be saved in our Manuscript Tracking System automatically. Author(s) Name(s) Contents lists available at ScienceDirect Free Radical Biology and Medicine journal homepage: www.elsevier.com/locate/freeradbiomed Original article Cardioprotective strategies preserve the stability of respiratory chain supercomplexes and reduce oxidative stress in reperfused ischemic hearts I. Ramírez-Camachoa, F. Correaa, M. El Hafidia, A. Silva-Palaciosa, M. Ostolga-Chavarríaa, M. Esparza-Perusquíab, S. Olvera-Sánchezb, O. Flores-Herrerab, C. Zazuetaa, ⁎ a Departamento de Biomedicina Cardiovascular, Instituto Nacional de Cardiología. I. Ch., 14080 Mexico, D.F., Mexico bDepartamento de Bioquímica, Facultad de Medicina, Universidad Nacional Autónoma de México, 04510 Mexico, D.F., Mexico A R T I C L E I N F O Keywords: Mitochondrial respiratory chain Supercomplexes Reactive oxygen species N-acetylcysteine Postconditioning A B S T R A C T Electron leakage from dysfunctional respiratory chain and consequent superoxide formation leads to mi- tochondrial and cell injury during ischemia and reperfusion (IR). In this work we evaluate if the supramolecular assembly of the respiratory complexes into supercomplexes (SCs) is associated with preserved energy efficiency and diminished oxidative stress in post-ischemic hearts treated with the antioxidant N-acetylcysteine (NAC) and the cardioprotective maneuver of Postconditioning (PostC). Hemodynamic variables, infarct size, oxidative stress markers, oxygen consumption and the activity/stability of SCs were compared between groups. We found that mitochondrial oxygen consumption and the activity of respiratory complexes are preserved in mitochondria from reperfused hearts treated with both NAC and PostC. Both treatments contribute to recover the activity of individual complexes. NAC reduced oxidative stress and maintained SCs assemblies containing Complex I, Complex III, Complex IV and the adapter protein SCAFI more effectively than PostC. On the other hand, the activities of CI, CIII and CIV associated to SCs assemblies were preserved by this maneuver, suggesting that the activation of other cardioprotective mechanisms besides oxidative stress contention might participate in maintaining the activity of the mitochondrial respiratory complexes in such superstructures. We conclude that both the monomeric and the SCs assembly of the respiratory chain contribute to the in vivo functionality of the mitochondria. However, although the ROS-induced damage and the consequent increased production of ROS affect the assembly of SCs, other levels of regulation as those induced by PostC, might participate in maintaining the activity of the respiratory complexes in such superstructures. 1. Introduction The organization of the respiratory chain complexes has been re- presented into different models: the fluid state one, in which individual entities and mobile electron carriers diffuse freely in the mitochondrial inner membrane interacting through random collisions [1]; the solid state model, in which the respiratory complexes are arranged into compact conglomerates known as supercomplexes (SCs) or respira- somes [2] and the currently accepted pattern that integrates both types of organization, known as the plasticity model [3]. The controversy on the possible artifacts that mild detergents cause on respiratory com- plexes' association, has been tempered by experimental evidences that include the identification of SCs in mitochondria from many organisms [4,5]; the demonstration that almost all Complex I is bound to Complex III in the absence of detergents [5,6] and that the purified SCs are stable and catalytically active [7]. Mitochondrial respiratory complexes are the main source of reactive oxygen species (ROS) production and in consequence, particularly sensitive to their effects. Superoxide anion (O2• ―) is produced after one- electron reduction of O2 in two sites of Complex I: at flavin in the NADH-oxidizing site (site IF) and in the ubiquinone-reducing site (site IQ) [8], as well as in the outer quinol-binding site of mitochondrial complex III (site IIIQo) [9]. The proposal that the structural organization of the respiratory complexes into supramolecular arrangements https://doi.org/10.1016/j.freeradbiomed.2018.09.047 Received 6 April 2018; Received in revised form 20 September 2018; Accepted 30 September 2018 Abbreviations: ROS, reactive oxygen species; O2, oxygen; O2 •, Superoxide anion; IR, ischemia and reperfusion; SC's, supercomplexes; SCAFI, supercomplex assembly factor I; NAC, N-acetylcysteine; PostC, postconditioning; H2O2, Hydrogen peroxide; NADH, Nicotinamide adenine dinucleotide; HR, Heart rate; LVDP, Left ven- tricular developed pressure; TTC, Triphenyltetrazolium Chloride ⁎ Correspondence to: Departamento de Biomedicina Cardiovascular Instituto Nacional de Cardiología, Ignacio Chávez, Juan Badiano No. 1. Colonia Sección XVI, Mexico, D.F. 14080, Mexico. E-mail address: ana.zazueta@cardiologia.org.mx (C. Zazueta). Free Radical Biology and Medicine 129 (2018) 407–417 Available online 11 October 2018 0891-5849/ © 2018 Elsevier Inc. All rights reserved. T decreases electron leakage and controls ROS production [10] is sup- ported by reports showing that disruption of the association between Complex I and Complex III with dodecyl maltoside augments ROS generation in vitro [11] and that low levels of the III2IV2 supercomplex is associated with increased ROS production in yeast mitochondria [12]. It has been established that ROS generated in Complexes I and III contribute to myocardial damage during ischemia and reperfusion (IR) [13–15], but few studies have addressed that their arrangement into SCs might be related with diminution in oxidative stress and with cardioprotection in reperfused hearts. What is known is that the in- crease of the cytochrome-c-oxidase subunit VIb in SCs, concurs with left ventricular pressure recovery in IR hearts subjected to preconditioning [16] and, that isoflurane-conferred protection in ischemic injury is partially related with the stabilization of oligomers from complexes III/ IV [17]. In this work we evaluated the effect of the antioxidant N- acetylcysteine (NAC) and of the mechanical manoeuver of Post- conditioning (PostC) on the functional properties of the mitochondrial respiratory chain and stability of SCs in mitochondria from IR hearts, to provide further evidences on the causative link between ROS produc- tion, energy deficiency and decrease in mitochondrial respirasome formation. 2. Material and methods This investigation was approved by the Ethics Committee of the National Institute of Cardiology, “Ignacio Chávez” (INC-13806). The experimental protocols followed the guidelines of Norma Oficial Mexicana for the use and care of laboratory animals (NOM-062-ZOO- 1999) and for disposal of biological residues (NOM-087-SEMAR- NAT-SSA1-2002). 2.1. Experimental design Male Wistar rats (300–350 g) were anaesthetized by injecting in- traperitoneally a single dose of sodium pentobarbital (60mg/kg i.p) plus sodium heparin and complete lack of pain response was assessed by determining pedal withdrawal reflex. Hearts were perfused retro- gradely in a Langendorff heart perfusion system via the aorta at a constant flow rate of 13mL/min with Krebs-Henseleit solution (118mM NaCl, 4.75mM KCl, 1.18mM KH2PO4, 1.18mM MgSO4·7H2O, 2.5 mM CaCl2, 25mM NaHCO3, 5 mM glucose and 0.1 mM sodium oc- tanoate, pH 7.4), which was continuously bubbled with 95% O2 and 5% CO2 at 37 °C. Cardiac performance was measured at left ventricular end-diastolic pressure (LVEDP) of 10mmHg using a latex balloon in- serted into the left ventricle and connected to a pressure transducer. Throughout the experiment, left ventricular developed pressure (LVDP) was recorded using the software LabChart8-Pro v8.1.5 from ADInstruments. Heart rate (HR) expressed as beat number x min−1 was obtained from the left ventricular pressure waveform. Hearts were perfused for 20min to reach a steady state and then subjected to the different protocols. The experimental groups were: 1) Control, hearts continuously perfused for additional 90min 2) IR, hearts subjected to global ischemia for 30min by turning off the pumping system and then to 60min of reperfusion. 3) IR+NAC hearts, that were subjected to 30min of ischemia and that during the first 10 min of reperfusion received 0.25mM of NAC in the Krebs-Henseleit Buffer and 4) IR+PostC, hearts subjected to 30min of ischemia, to postconditioning (5 cycles of 30 s reperfusion and 30 s ischemia) and to 60min of reperfusion (Fig. 1A). 2.2. Infarct size Infarct size was evaluated by staining with triphenyltetrazolium chloride (TTC). At the end of the experiments the hearts to be used for infarct size calculations were frozen at −20 °C. Heart slices of ~3mm were obtained and immersed in 1% TTC solution in phosphate buffer (8.8 mM Na2HPO4, 1.8 mM NaH2PO4, pH 7.4) for 10min at 37 °C. Digital images of heart slices were analyzed using the ImageJ® 1.48 software (NIH, MD, USA). Infarct size was expressed as percentage of total heart area [18]. 2.3. Mitochondrial isolation Fresh cardiac tissue from the different groups was placed in cold buffer solution containing 250mM sucrose, 10mM HEPES, and 1mM ethylenediaminetetraacetic acid, pH 7.4. The hearts were minced and incubated for 10min with the same buffer plus subtilisin A (2mg/g of tissue) in an ice bath. Then the tissue was washed, suspended in the same buffer without the enzyme and homogenized. Mitochondria were obtained by differential centrifugation as previously described [19]. 2.4. ROS production and oxidative stress markers in mitochondria Hydrogen peroxide production was evaluated as an indicator of ROS generation. Mitochondria (0.250mg/mL) were incubated in reaction buffer containing 100mM sucrose, 75mM KCl, 5 mM Tris, 3 mMMgCl2, 10 μM EGTA, and 1mM KH2PO4, pH 7.4, plus 0.1 μM dihydrodi- chlorofluorescein (DCF) and 1 U/mL horseradish peroxidase. Mitochondria were energized with malate/glutamate (5:3 mM) or with succinate (10mM). Antimycin A (AA; 5 μM) or rotenone (1 μM) were used to enhance ROS production. DCF oxidation was monitored at λex=475 nm (4-nm slit) and λem=525 nm (4-nm slit) in a Perkin- Elmer LS50B spectrofluorometer at 30 °C. The slopes of the traces ob- tained after substrate and/or inhibitor addition were compared against a standard curve of H2O2 generated by adding known amounts of H2O2 to the buffer containing DCF, horseradish peroxidase and mitochondria as previously described [20]. Malondialdehyde (MDA) was measured as lipoperoxidation marker according with Guerrero-Beltrán et al. [21]. Briefly, 1mg of mi- tochondrial protein was added to a medium containing 1-methyl-2- phenylindole. The reaction was started by adding 37% HCl and in- cubated for 40min at 45 °C. Then the samples were centrifuged at 3000 g for 5min and the optical density of the supernatant was mea- sured. The method is based on the formation of a stable chromophore with a maximal intensity of absorbance at 586 nm. Results are ex- pressed in nmol of MDA per milligram of protein. Reduced glutathione (GSH) was evaluated fluorometrically as de- scribed by Galván-Arzate et al. [22]. Briefly, fresh mitochondria (15 µg) were derivatized with 100 μL of o-phthalaldehyde (OPA, 1mg/mL) and incubated for 20min at room temperature in 2-mL final volume. Fluorescence was measured at λex=350 nm and λem=420 nm in a LS50B Luminescence Spectrophotometer (Perkin Elmer, Waltham, MA). Twin samples were used in parallel to evaluate oxidized glutathione (GSSG). Mitochondria were incubated for 30min with 100 μL of 40mM N-ethylmaleimide (NEM) to prevent that glutathione free SH group reacts with OPA. Then, the samples were mixed with 4.3mL of 0.1 N NaOH (pH 12) to allow GSSG reduction. Finally an aliquot was with- drawn, incubated with 50 μg of OPA for 20min and fluorescence was measured [23]. Results were normalized per mg protein and expressed as GSH/GSSG ratio. We also measured protein oxidation with the OxyBlot™ protein oxidation detection kit (Merck-Millipore, Darmstadt, Germany). Briefly, two aliquots of each mitochondrial sample (20–40 μg) were transferred to Eppendorf tubes and denatured by adding a final concentration of 10% SDS. One aliquot was derivatized to 2,4-dinitrophenylhydrazone (DNP) with 2,4-dinitrophenylhydrazine (DNPH), while the aliquot used as the negative control was incubated with the same volume of control solution. Then, samples were neutralized and separated by SDS-PAGE, transferred to PDVF membranes and incubated with rabbit anti-DNP antibodies (1:150 dilution) in PBS–Tween containing 1% bovine serum albumin. Horseradish peroxidase-conjugated secondary antibodies I. Ramírez-Camacho et al. Free Radical Biology and Medicine 129 (2018) 407–417 408 (1:300 dilution) and a chemiluminescent reagent were used for signal detection. 2.5. Mitochondrial oxygen consumption Mitochondrial oxygen consumption was determined using a Clark- type oxygen electrode (Yellow Springs Instruments, Yellow Springs, OH, USA). The experiments were carried out in 1.5mL of basic medium, containing 125mM KCl, 10mM Hepes and 3mM Pi, pH 7.3. State 4 respiration was evaluated in the presence of 5mM sodium glutamate and 5mM sodium malate. State 3 respiration was stimulated by adding 200 μM adenosine diphosphate (ADP). Respiratory rates are expressed as ng-atoms oxygen/min/mg protein (ngAO/min/mg). Respiratory control ratio (RCR) was calculated as the ratio between the State 3 and State 4 rates. Phosphorylation efficiency (ADP/O ratio) was calculated from the added amount of ADP and total amount of oxygen consumed during state 3. Uncoupled respiration was observed after incubating with 1 µM of CCCP (Carbonyl cyanide m-chlorophenyl hy- drazone). 2.6. Mitochondria solubilization Separation of electron transport chain complexes was performed using BN-PAGE according to the protocol described by Schäfer et al. [24]. Mitochondrial proteins were solubilized with digitonin using a detergent to protein ratio of 2:1 (mg:mg) in 200 μL of 3×-buffer con- taining 150mM BIS-TRIS, 1.5mM aminocaproic acid, pH 7. After in- cubating during 30min, the samples were centrifuged at 18,000 g for 2 h and supernatants were recovered. All procedures were performed at 4 °C. 2.7. Blue native polyacrylamide gel electrophoresis (BN-PAGE) and in-gel activity assays Blue-Native (BN) 3.5–10% gradient gels were loaded with 100 µg of mitochondrial protein in the presence of Coomassie blue G-250 dye and were run overnight at 30 V in cold. The anode buffer contained 50mM BIS-TRIS, pH 7.0 and the cathode buffer 50mM tricine, 15mM BIS- TRIS/HCl plus 0.002% Coomassie blue G-250. At the end of the elec- trophoresis, some gels were further stained with Coomasie blue. Other gels were used to measure the enzymatic activities of each mitochon- drial respiratory complex as follows: NADH dehydrogenase activity of Complex I (CI) was evaluated by formazan production, after incubating the gels in 10mM TRIS, 1mg/mL (w/v) of 4-nitro blue tetrazolium chloride and 0.14mM NADH, pH 7.0. Succinate dehydrogenase activity of Complex II (CII) was visualized as purple bands in the presence of 80mM succinate, 2 mg/mL (w/v) of 4-nitro blue tetrazolium chloride and 4.5mM EDTA, pH 7.0. Complex III (CIII) activity was determined with 0.2mM TMPD (N,N,N′,N′-tetramethyl-p-phenylenediamine) in 70% methanol, 100mM sodium acetate and 100 μM H2O2, pH 7.0. Cytochrome c oxidase activity of Complex IV (CIV) was visualized by precipitation of 3,3-diaminobenzidine oxides in a buffer containing 50mM sodium phosphate, 0.5 mg/mL of 3,3-diaminobenzidine-tetra- hydrochloride, 0.5mg/mL cytochrome c, 20 units/mL catalase, and 75mg/mL sucrose, pH 7.0. Finally, ATPase activity (CV) was detected in a medium containing 30mg of lead II perchlorate trihydrate, 30mM Tris-Glycerol, 5 mM ATP and 15mM CaCl2 in 20mL until maximal signal of the lead phosphate band was obtained [25]. The developed bands were analyzed by densitometry using the Image J program from the National Institutes of Health. Fig. 1. A) Schematic representation of the experimental protocols in the isolated heart model. All hearts were stabilized during 20min. Control hearts were continuously perfused during 110min; IR hearts were subjected to 30min of global ischemia and 60min of reperfusion; IR+NAC hearts received 0.25mM NAC during the first 10min of reperfusion and IR+PostC hearts were subjected to 5 cycles of 30 s reperfusion and 30 s ischemia and then, to 60min of reperfusion; B) infarct size; C) heart rate (beats*min−1) and D) left ventricular developed pressure (LVDP). Data are means ± SD of at least six different experiments. *P < 0.05 vs. Control and **P < 0.05 vs. IR. I. Ramírez-Camacho et al. Free Radical Biology and Medicine 129 (2018) 407–417 409 2.8. Activity of the respiratory chain complexes The enzymatic function of the mitochondrial respiratory complexes was also measured spectrophotometrically as described by Spinazzi et al. [26]. Complex I activity was evaluated in 1mL of 10mM TRIS buffer pH 7.4, supplemented with 100 µM NADH and mitochondrial protein (50 µg). The reaction was started by adding 0.95 µM 2,6-di- chloroindophenol (DCPIP) as terminal electron acceptor. Reduction of DPCIP was followed at 600 nm during 3min in the presence or absence of 1 µM rotenone (ε=21mM−1 cm−1). Specific activity is expressed as nmol/min/mg. Complex II activity was determined incubating 40 μg of mitochondrial protein in 0.5M potassium phosphate buffer (pH 7.5) supplemented with 20mM succinate and 150 μL of DCPIP 0.015% (w/ v). Final volume was adjusted to 1mL with distilled water. Reduction of DPCIP was followed at 600 nm during 3min (ε=21mM−1 cm−1). Specific activity is expressed as nmol succinateox/min/mg, due that 1mol of DCPIP is reduced as 1mol of succinate is oxidized. Complex III was measured as follows: 40 μg of mitochondrial protein was incubated in 1mL of potassium phosphate buffer 0.5M (pH 7.5) plus 1mM EDTA, 0.5 mM KCN and 50 μM of oxidized cytochrome c at 30 °C for 15min. Cytochrome c reduction was initiated by adding 50 µM of decylubi- quinol (DBH2) in the presence or absence of 10 µg of antimycin A. The increase in absorbance was followed for 1–2min at 550 nm (ε=18.5mM−1 cm−1). Specific complex III activity is the antimycin A–sensitive-activity. Complex IV was evaluated after adding 40 μg of mitochondrial protein to 1mL of 50mM potassium phosphate buffer, (pH 7.0) containing 50 μM of reduced cytochrome c, with or without 0.5 mM KCN. Decrease in absorbance was immediately registered at 550 nm (ε=18.5mM−1 cm−1). Complex IV activity is expressed as first order rate constant k (min−1 ×mg protein−1). Oligomycin-sen- sitive ATPase activity (CV) was evaluated by measuring ATP hydrolysis by a coupled enzymatic assay using lactate dehydrogenase and pyr- uvate kinase [27]. Forty μg of mitochondrial protein were incubated in 200 μL of a medium containing 50mM Tris·HCl, pH 8.0, 5 mg/mL BSA, 20mM MgCl2, 50 mM KCl, 15 μM CCCP, 5 μM AA, 10mM phosphoe- nolpyruvate (PEP), 2.5mM ATP, 4U lactate dehydrogenase, 4U pyr- uvate kinase and 1mM NADH during 5min at 37 °C. The reaction was continuously recorded at 340 nm during 3min, then 3 μM oligomycin was added and measured for additional 3min.hydrolysis NADH con- centration was calculated by using an extinction coefficient of 6220M−1 cm-l. As the hydrolysis of 1mol ATP produces the oxidation of 1mol NADH through this coupling system, the activity is expressed as nmol ATP/mg protein/min. 2.9. Immunodetection in first dimension (1-D) and second dimension (2-D) native gels BN gels were incubated in 20mL of a solution containing 0.192M glycine, 0.025M TRIS plus 1% SDS during 60min under constant agi- tation and transferred onto polyvinylidine difluoride membranes (PVDF) for 50min at 25 V. Then, the membranes were blocked with 5% defat milk in TBS-T and incubated overnight with anti-NADH dehy- drogenase subunit Ndufs4 (ab55540, Abcam, Cambridge, MA, USA;1:1000); anti-ubiquinol cytochrome c reductase core protein UQCRFS1 (ab14746, Abcam, Cambridge, MA, USA; 1:1000); anti-cy- tochrome c oxidase subunit II (sc-514489, Santa Cruz Biotechnology, USA; 1:1000) and anti-COX7A2L subunit (Proteintech, Rosemont, IL, USA, 1:1000). Secondary antibodies were used at 1:25,000 dilution. The signals were visualized with C-Digit blot scanner and analyzed by ImageJ (NIH). 2.10. Statistical analysis Data were analyzed using one or two-way ANOVA followed by Tukey's post-hoc test with the graph Pad PRISM 5.03 software; P < 0.05 was considered statistically significant. 3. Results 3.1. N-acetylcysteine and postconditioning maintain cardiac function and reduce infarct size in reperfused hearts NAC and PostC reduced infarct size to 26% and 18% respectively, in comparison with almost 59% of cell death observed in IR hearts (P < 0.05) (Fig. 1B). Heart rate and left-ventricular developed pressure (LVDP) were maintained in Control hearts during 110min of constant perfusion, while those parameters decreased from the first minutes and until the end of reperfusion in IR hearts. LVDP in the IR group was more affected than heart rate. Both NAC and PostC prevented almost com- pletely from heart dysfunction at the end of reperfusion, being more evident cardiac pressure recovery (Fig. 1C and Fig. 1D). 3.2. Cardioprotection conferred by N-acetylcysteine and postconditioning is related with decreased H2O2 content and reduction of oxidative stress Succinate oxidation in the presence of both Antimycin A (AA) and rotenone (Rot) augmented hydrogen peroxide levels in IR mitochondria in comparison with Control mitochondria (10 ± 3.5 and 10.1 ± 1.3 pmol H2O2/min/mg vs. 4.7 ± 1.7 and 1.8 ± 2.3 pmol H2O2/min/mg respectively). Both NAC and PostC, diminished hydrogen peroxide at different extent, although PostC showed higher efficiency even without inhibitors. On the other hand, mitochondria from IR hearts energized with glutamate/malate showed higher rate of hydrogen peroxide gen- eration than Control mitochondria, that further increase in the presence of AA, confirming the participation of Complex III in ROS production. Again, both treatments diminished ROS production in comparison with IR mitochondria. Also, the combination of glutamate/malate plus ro- tenone enhanced H2O2 production in IR mitochondria as compared with the Control group (Fig. 2). Lipid peroxidation and protein carbo- nylation increased in mitochondria from the IR group in correlation with decreased redox state (GSH/GSSG) as compared to the Control group (P < 0.05) (Fig. 3). NAC reduced MDA levels by 51% (1.3 ± 0.5 vs. 2.7 ± 0.8 nmol MDA/mg protein in IR mitochondria) (P < 0.05) and PostC to 1.65 ± 0.2 nmol MDA/mg protein (48%) (Fig. 3A). GSH/GSSG increased significantly in IR+PostC in compar- ison with IR mitochondria (P < 0.05; Fig. 3B); whereas protein oxi- dation diminished in both IR+NAC and IR+PostC groups (Fig. 3C). Fig. 2. Effect of NAC and PostC on the rate of hydrogen peroxide production in mitochondria from reperfused hearts. H2O2 content was determined by fol- lowing HRP-catalyzed dichlorofluorescein (DCF) oxidation in mitochondria plus succinate or glutamate/malate in the presence of Antimycin A and/or rotenone. Data are means ± SD of at least three independent experiments per group. *P < 0.05 vs. Control and * * P < 0.05 vs. IR. # P < 0.05 vs. mi- tochondria without inhibitors. I. Ramírez-Camacho et al. Free Radical Biology and Medicine 129 (2018) 407–417 410 3.3. Mitochondrial function is preserved by diminishing oxidative stress in reperfused hearts The association between oxidative stress and oxidative phosphor- ylation (OXPHOS) was evaluated in isolated mitochondria by per- forming oxygen consumption experiments. No changes were observed in basal oxygen consumption (State 4) with NADH-linked substrates between Control, IR and IR+NAC mitochondria (Fig. 4A), whereas ADP-stimulated respiration (State 3) diminished in the IR group and was recovered in both IR+NAC and IR+PostC mitochondria. RCR, the single most useful general measure of mitochondrial function, dimin- ished from 2.89± 0.43 to 1.6±0.19 (Control vs. IR mitochondria; P < 0.05) and increased significantly in mitochondria from hearts subjected to both treatments (Fig. 4B). ADP/O, which represents the maximum number of ATP molecules made as an electron pair passes down the respiratory chain from substrate to oxygen, followed a similar pattern, but was only preserved with NAC treatment (P < 0.05). ADP- stimulated (State 3) and uncoupled respiration diminished in the IR group suggesting inhibition of substrate oxidation (Fig. 4C). Fig. 3. Oxidative stress markers in isolated mitochondria from reperfused heart treated with NAC and subjected to PostC. A) Malondialdehyde (MDA) levels; B) GSH/ GSSG and C) Representative immunoblot of carbonylated proteins and densitometric ratio between total protein carbonylation (DPN) and adenine nucleotide translocator (ANT) content. It is also shown a Coomassie blue staining of the gel. Data are representative of at least five different experiments per group. Data are means ± SD. *P < 0.05 vs. Control and **P < 0.05 vs. IR. I. Ramírez-Camacho et al. Free Radical Biology and Medicine 129 (2018) 407–417 411 3.4. Enzymatic activities of the respiratory chain complexes from reperfused hearts treated with N-acetylcysteine and postconditioning The activity of CI diminished 59% in IR mitochondria when com- pared with that of Control mitochondria. IR+NAC and IR+PostC groups recovered CI activity, but the difference was significant only in the last one (Fig. 5A). The activities of CII (Fig. 5B) and CIII (Fig. 5C) were also affected by IR, although the counteracting effect of NAC was more evident in CII, than in CIII activity. Conversely, PostC increased CIII activity to higher values than NAC and even than in Control (Fig. 5C). Despite the lack of significant differences, IR mitochondria showed a slight decrease in CIV activity in comparison with mi- tochondria from the other groups (Fig. 5D). CV activity was fully re- covered in both IR+NAC and in IR+PostC mitochondria as compared with IR mitochondria (Fig. 5E). 3.5. Detection of respiratory complex activities in blue native gels The relative contribution of the monomeric and SCs forms on the total activity of the respiratory chain complexes and the effect of ROS- reducing treatments in mitochondria from IR hearts were evaluated in BN gels. SCs assemblies in which the activities of the three complexes were detected and the additional bands that contained higher mass complexes than its corresponding monomers were included in the SCs analysis (Fig. 6). To identify putative productive SCs, we performed a careful alignment of the activity gels and label the group of bands in which Complex I, Complex III and Complex IV comigrate. As Complex III assay might render unspecific staining, we correlate the activity bands with the spots obtained after immunodetection with Anti-Com- plex III subunit UQCRFS1 in a two-dimension SDS gel. To further cor- roborate the identity of the respiratory complexes and its associations in our preparations, a comparison was made with the electrophoretic pattern of bovine heart mitochondria (Supplementary material). Complex I in situ activity associated to the monomeric form di- minished in IR mitochondria and was maintained in the IR+PostC group (P < 0.05); this treatment also maintained the activity of CI in SCs at the levels observed in Control mitochondria (P < 0.05; Fig. 6A and D). CIII2 activity diminished in IR mitochondria and was fully re- covered in both IR+NAC and IR+PostC groups (P < 0.05); whereas the activity associated to SCs increased only in mitochondria from PostC hearts as compared with IR mitochondria (P < 0.05; Fig. 6B and D). CIV activity decreased in the monomer and in SCs from IR mi- tochondria as compared with the other groups. The activity associated to the monomer and to SCs increased significantly only in mitochondria from PostC hearts (P < 0.05; Fig. 6C and D). Although Complex II and Complex V did not contribute to SCs Fig. 4. Oxygen consumption in isolated mitochondria with glutamate plus malate. A) State 4 respiration and State 3 respiration; B) Respiratory Control Rate (RCR) and ADP/O; C) uncoupled respiration. Data are means of at least three different experiments ± SD. *P < 0.05 vs. Control and **P < 0.05 vs. IR. I. Ramírez-Camacho et al. Free Radical Biology and Medicine 129 (2018) 407–417 412 assemblies, their activities were also evaluated. No changes in Complex II activity were detected in any of the experimental groups (Fig. 7A), neither in the activity of the dimeric and monomeric forms of Complex V (Fig. 7B). 3.6. Relative content of respiratory complexes in supercomplex assemblies in reperfused hearts treated with N-acetylcysteine and subjected to postconditioning We also examined whether IR and ROS-reducing treatments affect respiratory complexes content in SCs by performing (1D-BN) PAGE and subsequent immunoblotting. A I+III2+IVn complex was identified in Fig. 5. Spectrophotometric analysis of the respiratory complexes activities in isolated mitochondria. A) Activity of Complex I; B) Complex II; C) Complex III; D) Complex IV and E) Complex V. Values are the means ± SD of at least four different experiments. *P < 0.05 vs. Control and **P < 0.05 vs. IR. I. Ramírez-Camacho et al. Free Radical Biology and Medicine 129 (2018) 407–417 413 mitochondria from the Control group, which also contained the adapter protein SCAFI (Supercomplex assembly factor I). The relative content of the subunits NDUFS4 (Complex I), UQCRFS1 (Complex III), COX IV (Complex IV) and COX7A2L (SCAFI) clearly diminished in mitochon- dria from the IR group and were preserved differentially in the IR +NAC and in the IR+PostC groups (Fig. 8). 4. Discussion In this work we show that oxidative stress contention promotes heart performance, decreases infarct size, preserves mitochondrial function and maintains the assembly/activities of both SCs and the individual complexes of the respiratory chain. These results reinforce the well known paradigm that mitochondria are both producers and ROS targets and incorporate SCs stability as a main factor in mitochondrial function. The association between SCs' assembly and lower ROS generation has been observed in different conditions. Fibroblasts exposed to re- spiratory chain complex inhibitors increased ROS in correlation with SCs levels diminution [28], the knockdown of Complex I subunit NDUFS1 decreases the integration of complex I into SCs and augments ROS levels in neurons [29]; whereas age-related changes in SCs archi- tecture modify ROS production [30]. Such association might be the consequence of a more efficient electron transfer in SCs than in Fig. 6. In gel activities of the respiratory chain complexes (I, III and IV) in BN-gels. Representative images of at least three different experiments of the in-gel activities of re- spiratory complexes in blue native gels. Gels were pre- pared and loaded with solubilized mitochondria. Parallel running was performed to optimize the comparison be- tween the activities and location of the different re- spiratory complexes and SCs. The symbol (&) groups the bands in which activities for Complex I, III and IV were observed. The bands labeled as CIVn represent possible oligomeric forms of cytochrome c oxidase. A) Complex I activity; B) Complex III activity; C) Complex IV activity and D) densitometric analysis of the activity associated to monomeric (Complex) and SCs forms. Values are the means ± SD of at least three different experiments. *P < 0.05 vs. Control and **P < 0.05 vs. IR. Fig. 7. In gel activities of the respiratory chain complexes (II and V) in BN-gels and Coomassie blue staining. Representative images of at least three different experi- ments of the in-gel activities of respiratory complexes in blue native gels. Gels were prepared, loaded with solubi- lized mitochondria and run in parallel to optimize the comparison between the activities and location of the different respiratory complexes and SCs. A) Complex II; B) Complex V (CVd = CV dimer; CVm = CV monomer); C) Coomassie blue staining. I. Ramírez-Camacho et al. Free Radical Biology and Medicine 129 (2018) 407–417 414 individual complexes [31], as the distances between binding sites of coenzyme Q at complexes I-III and of cytochrome c at complexes III-IV in these supramolecular arrangements are shorter [32] than the diffu- sion lengths calculated by random collision mechanisms for individual complexes [33]. The site of ROS production is still a non-resolved issue. It is gen- erally accepted that during succinate oxidation, ROS are produced to some extent by complex III [34] but mainly through reverse electron transport (RET) to complex I [35]. Indeed, Chouchani et al. [36] re- ported that succinate is accumulated during ischemia and that is me- tabolized in parallel with ROS production from Complex I by RET during early reperfusion in vivo. Although it was tempting to speculate that the observed diminution in succinate dehydrogenase activity (Fig. 5B) could drive succinate accumulation, our results show that H2O2 production was not sensitive to rotenone, therefore RET is not the main mechanism of ROS production in our experimental setting. In this sense, it was reported that the damage to complex I during ischemia/ reperfusion might induce the loss of protein-protein interaction be- tween this complex and complex III, enhancing ROS production by complex III [37]. This scenario might explain our observation of com- plex III diminution in SCs in association with increased ROS production. Our finding that individual Complex I and Complex III activities and those associated to SCs assemblies are higher in mitochondria from the PostC group than in the NAC group, might reflect the activation of protective mechanisms independent of those related with oxidative stress contention. PostC modulates the activity of several protein ki- nases associated with cardioprotection [38]. Results from our group and others show that activation of extracellular signal-regulated kinases 1/2 (ERK1/2) correlates with increased levels of phosphorylated pro- teins in mitochondria from PostC hearts [39], and that this maneuver increases cGMP levels preventing the opening of the mitochondrial permeability transition pore (mPTP) during the early stage of reperfu- sion [40]. Phosphate groups attached to residues of subunits IVa and IVb from Complex IV has been detected in mitochondria isolated from pre-conditioned hearts, suggesting a possible role of phosphorylation in the stability of these proteins [16], that might be extended to SCs. In this work not only the activity, but the immunoreactivity of some of complexes in both its individual and SCs forms was compromised during IR. Loss of integrity of particular complex subunits after oxida- tive injury might account for these results, as it has been reported that ischemia diminishes the amount of detectable NDUFA9 subunit from Complex I in both native and denaturing conditions in association with cardiolipin oxidation [41]. We found that COX7A2L was mainly present in I+III2+IVn and in much lower extent in free complex IV. This protein was detected in superstructures composed of III2+IV and I+III2+IVn, but not in free complexes III and IV [42]; whereas more recently, Perez-Perez et al. [43] reported that COX7A2L is not exclusively associated with SCs, but it also co-migrates with CIII2, free complex IV and with different SCs containing complex III. This authors propose that although COX7A2L stabilizes SCs III2+IV, has not role on respirasome formation. Further controversy on the possible relevance of COX7A2L was added by Boe- kema et al. [44], which sustain that point mutations within genes en- coding subunits of one oxidative phosphorylation complex affect the stability of other complexes. The idea that dissociation of SCs has pathophysiological implica- tions was first suggested by Lenaz and Genova [45]. Their speculation that ROS reduce electron channeling, lowers electron transfer and/or proton translocation, eliciting further ROS production, fits with the vicious circle of oxidative stress and energetic decline observed in re- perfusion damage. However, changes in SCs' assembly has rendered polemic results in the framework of cardiovascular diseases. While Wong et al. [16] reported that SCs levels were similar in mitochondria isolated from IR and from preconditioned hearts; the cardioprotective effect of isoflurane was related, at least in part with maintenance of Complex III and Complex IV activities in SCs [17]. On the other hand, SCs disassembly was observed after long-time (60min) reperfusion and prevented by inhibiting the permeability transition pore [46], whereas lower levels of supercomplex I+III2+IV1 were detected without changes in the individual activities of the respiratory chain complexes Fig. 8. Immunodetection of SCs assemblies in mitochondria from reperfused hearts treated with NAC and subjected to PostC. Representative western blot images of at least three different experiments in which mi- tochondrial protein were separated in 1D blue native gels. The same membrane was stripped and confronted to all the antibodies. A) Complex I was detected with anti-NADH de- hydrogenase subunit NDUFS4; B) Complex III with anti-Ubiquinol cytochrome C reductase core protein UQCRFS1; C) Complex IV with anti-cytochrome c Oxidase subunit IV; D) SCAFI with anti-COX7A2L supercomplex III-IV subunit; E) Coomassie blue staining of the PVDF membrane used in this set of experi- ments used as control load. * SCs in which SCAFI was not detected. I. Ramírez-Camacho et al. Free Radical Biology and Medicine 129 (2018) 407–417 415 in a canine heart failure model [47]. Here, we find that oxidative stress diminution preserves mitochondrial function and that cardioprotection is related with maintenance of the activity and content of Complex I, Complex III and Complex IV in supramolecular assemblies. 5. Conclusion We conclude that both the monomeric forms and the SCs of the respiratory chain contribute to the in vivo functionality of the mi- tochondria. However, although the ROS-induced damage and the con- sequent production of ROS affect the assembly of SCs, other levels of regulation as those induced by PostC, might participate in maintaining the activity of the mitochondrial respiratory complexes in such super- structures. Acknowledgments Jazmín Ixchel Ramírez Camacho is a doctoral student from Programa de Doctorado en Ciencias Biomédicas, Universidad Nacional Autónoma de México (UNAM) and received fellowship 581606 from Consejo Nacional de Ciencia y Tecnologia (CONACYT). This work was partially supported by Grant 283363 to CZ from CONACYT, Mexico and Instituto Nacional de Cardiología, I. Ch. and PAPIIT IN222617 to OFH from the Universidad Nacional Autónoma de México. Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at doi:10.1016/j.freeradbiomed.2018.09.047. References [1] C.R. Hackenbrock, B. Chazotte, S.S. Gupte, The random collision model and a cri- tical assessment of diffusion and collision in mitochondrial electron transport, J. Bioenerg. Biomembr. 18 (1986) 331–368, https://doi.org/10.1007/BF00743010. [2] H. Schägger, K. Pfeiffer, Supercomplexes in the respiratory chains of yeast and mammalian mitochondria, EMBO J. 19 (2000) 1777–1783, https://doi.org/10. 1093/emboj/19.8.1777. [3] R. Acin-Perez, J.A. Enríquez, The function of the respiratory supercomplexes: the plasticity model, Biochim. Biophys. Acta 1837 (2014) 444–450, https://doi.org/10. 1016/j.bbabio.2013.12.009. [4] L. Robles-Martínez, M.G. Guerra- Sánchez, O. Flores-Herrera, A.N. Hernández- Lauzardo, M.G. Velázquez-Del Valle, J.P. Pardo, The mitochondrial respiratory chain of Rizopus Stolonifer (Ehrenb.:fr.) Vuill, Arch. Microbiol. 195 (2013) 51–61, https://doi.org/10.1007/s00203-012-0845-7. [5] H. Schägger, K. Pfeiffer, The ratio of oxidative phosphorylation complexes I-V in bovine heart mitochondria and the composition of respiratory chain super- complexes, J. Biol. Chem. 276 (2001) 37861–37867, https://doi.org/10.1074/jbc. M106474200. [6] G. Lenaz, G. Tioli, A.I. Falasca, M.L. Genova, Complex I function in mitochondrial supercomplexes, Biochim. Biophys. Acta 2016 (1857) 991–1000, https://doi.org/ 10.1016/j.bbabio.2016.01.013. [7] R. Acín-Pérez, P. Fernández, M.L. Peleato, A. Pérez-Martos, J.A. Enríquez, Respiratory active mitochondrial supercomplexes, Mol. Cell. Biol. 32 (2008) 529–539, https://doi.org/10.1016/j.molcel.2008.10.021. [8] J.R. Treberg, C.L. Quinlan, M.D. Brand, Evidence for two sites of superoxide pro- duction by mitochondrial NADH-ubiquinone oxidoreductase (complex I), J. Biol. Chem. 286 (2011) 27103–27110, https://doi.org/10.1074/jbc.M111.252502. [9] M.P. Murphy, How mitochondria produce reactive oxygen species, Biochem. J. 417 (2009) 1–13, https://doi.org/10.1042/BJ20081386. [10] G. Lenaz, A. Baracca A, G. Barbero, C. Bergamini, M.E. Dalmonte, M. Del Sole, M. Faccioli, A. Falasca, R. Fato, M.L. Genova, G. Sgarbi, G. Solaini, Mitochondrial respiratory chain super-complex I-III in physiology and pathology, Biochim. Biophys. Acta 1797 (2010) 633–640, https://doi.org/10.1016/j.bbabio.2010.01. 025. [11] E. Maranzana, G. Barbero, A.I. Falasca, G. Lenaz, M.L. Genova, Mitochondrial re- spiratory supercomplex association limits production of reactive oxygen species from complex I, Antioxid. Redox Signal. 19 (2013) 1469–1480, https://doi.org/10. 1089/ars.2012.4845. [12] M. Vukotic, S. Oeljeklaus, S. Wiese, F.N. Vögtle, C. Meisinger, H.E. Meyer, A.D.M. Katschinski, D.C. Jans, S. Jakobs, B. Warscheid, P. Rehling, M. Deckers, Rcf1 mediates cytochrome oxidase assembly and respirasome formation, revealing het- erogeneity of the enzyme complex, Cell Metab. 15 (2012) 336–347, https://doi. org/10.1016/j.cmet.2012.01.016. [13] R. Ferrari, C. Ceconi, S. Curello, A. Cargnoni, E. Pasini, F. De Giuli, A. Albertini, Role of oxygen free radicals in ischemic and reperfused myocardium, Am. J. Clin. Nutr. 53 (1991) 215S–222S. [14] F. Arreguín, N. García, S. Hernández-Reséndiz, M. Buelna-Chontal, F. Correa, V. Olín-Sandoval, O.N. Medina-Campos, J. Pedraza-Chaverri, C. Zazueta, Attenuation of oxidant damage in the postconditioned heart involves non-enzy- matic response and partial catalytic protection, Exp. Physiol. 97 (2012) 1119–1130, https://doi.org/10.1113/expphysiol.2012.065763. [15] G. Paradies, G. Petrosillo, M. Pistolese, N. Di Venosa, A. Federici, F.M. Ruggiero, Decrease in mitochondrial complex I activity in ischemic/reperfused rat heart: in- volvement of reactive oxygen species and cardiolipin, Circ. Res. 94 (2004) 53–59, https://doi.org/10.1161/01.RES.0000109416.56608.64. [16] R. Wong, A.M. Aponte, C. Steenbergen, E. Murphy, Cardioprotection leads to novel changes in the mitochondrial proteome, Am. J. Physiol. Heart Circ. Physiol. 298 (2010) H75–H91, https://doi.org/10.1152/ajpheart.00515.2009. [17] C. Lotz, J. Zhang, C. Fang, D. Liem, P. Ping, Isoflurane protects the myocardium against ischemic injury via the preservation of mitochondrial respiration and its supramolecular organization, Anesth. Analg. 120 (2015) 265–274, https://doi.org/ 10.1213/ANE.0000000000000494. [18] F. Correa, M. Buelna-Chontal, V. Chagoya, G. García-Rivas, R.M. Vigueras, J. Pedraza-Chaverri, W.R. García-Niño, R. Hernández-Pando, J.C. León-Contreras, C. Zazueta, Inhibition of the nitric oxide/cyclic guanosine monophosphate pathway limited the cardioprotective effect of post-conditioning in hearts with apical myo- cardial infarction, Eur. J. Pharmacol. 765 (2015) 472–481, https://doi.org/10. 1016/j.ejphar.2015.09.018. [19] E. Parra, D. Cruz, G. García, C. Zazueta, F. Correa, N. García, E. Chávez, Myocardial protective effect of octylguanidine against the damage induced by ischemia re- perfusion in rat heart, Mol. Cell. Biochem. 269 (2005) 19–26, https://doi.org/10. 1007/s11010-005-2989-0. [20] A. Ruíz-Ramírez, M. Chávez-Salgado, J.A. Peñeda-Flores, E. Zapata, F. Masso, M. El- Hafidi, High-sucrose diet increases ROS generation, FFA accumulation, UCP2 level, and proton leak in liver mitochondria, Am. J. Physiol. Endocrinol. Metab. 301 (2011) E1198–E1207, https://doi.org/10.1152/ajpendo.00631.2010. [21] C.E. Guerrero-Beltrán, M. Calderón-Oliver, E. Tapia, O.N. Medina-Campos, D.J. Sánchez-González, C.M. Martínez-Martínez, K.M. Ortiz-Vega, M. Franco, J. Pedraza-Chaverri, Sulforaphane protects against cisplatin-induced ne- phrotoxicity, Toxicol. Lett. 192 (2010) 278–285, https://doi.org/10.1016/j.toxlet. 2009.11.007. [22] S. Galván-Arzate, J. Pedraza-Chaverrí, O.N. Medina-Campos, P.D. Maldonado, B. Vázquez-Román, C. Ríos, A. Santamaría, Delayed effects of thallium in the rat brain: regional changes in lipid peroxidation and behavioral markers, but moderate alterations in antioxidants, after a single administration, Food Chem. Toxicol. 43 (2005) 1037–1045, https://doi.org/10.1016/j.fct.2005.02.006. [23] P.J. Hissin, R. Hilf, A fluorometric method for determination of oxidized and re- duced glutathione in tissues, Anal. Biochem. 74 (1976) 214–226, https://doi.org/ 10.1016/0160-5402(84)90059-7. [24] E. Schäfer, H. Seelert, N.H. Reifschneider, F. Krause, N.A. Dencher, J. Vonck, Architecture of active mammalian respiratory chain supercomplexes, J. Biol. Chem. 281 (2006) 15370–15375, https://doi.org/10.1074/jbc.M513525200. [25] C. Jung, C.M. Higgins, Z. Xu, Measuring the quantity and activity of mitochondrial electron transport chain complexes in tissues of central nervous system using blue native polyacrylamide gel electrophoresis, Anal. Biochem. 286 (2000) 214–223, https://doi.org/10.1006/abio.2000.4813. [26] M. Spinazzi, A. Casarin, V. Pertegato, L. Salviati, C. Angelini, Assessment of mi- tochondrial respiratory chain enzymatic activities on tissues and cultured cells, Nat. Protoc. 7 (2012) 1235–1246, https://doi.org/10.1038/nprot.2012.058. [27] A. Barrientos, In vivo and in organelle assessment of oxphos activities, Methods 26 (2002) 307–316, https://doi.org/10.1016/S1046-2023(02)00036-1. [28] F. Diaz, J.A. Enriquez, C.T. Morales, Cells lacking Rieske iron-sulfur protein have a reactive oxygen species-associated decrease in respiratory complexes I and IV, Mol. Cell. Biol. 32 (2012) 415–429, https://doi.org/10.1128/MCB.06051-11. [29] I. López-Fabuel, J. Le Douce, A. Logan, A.M. James, G. Bonvento, M.P. Murphy, A. Almeida, J.P. Bolaños, Complex I assembly into supercomplexes determines differential mitochondrial ROS production in neurons and astrocytes, Proc. Natl. Acad. Sci. USA 113 (2016) 13063–13068, https://doi.org/10.1073/pnas. 1613701113. [30] N.A. Dencher, M. Frenzel, N.H. Reifschneider, M. Sugawa, F. Krause, Proteome alterations in rat mitochondria caused by aging, Ann. N. Y. Acad. Sci. 1100 (2007) 291–298, https://doi.org/10.1196/annals.1395.030. [31] A. Panov, S. Dikalov, N. Shalbuyeva, R. Hemendinger, J.T. Greenamyre, J. Rosenfeld, Species- and tissue-specific relationships between mitochondrial per- meability transition and generation of ROS in brain and liver mitochondria of rats and mice, Am. J. Physiol. Cell Physiol. 292 (2007) C708–C718, https://doi.org/10. 1152/ajpcell.00202.2006. [32] T. Althoff, D.J. Mills, J.L. Popot, W. Kühlbrandt, Arrangement of electron transport chain components in bovine mitochondrial supercomplex I1III2IV1, EMBO J. 30 (2011) 4652–4664, https://doi.org/10.1038/emboj.2011.324. [33] B. Chazotte, C.R. Hackenbrock, The multicollisional, obstructed, long-range diffu- sional nature of mitochondrial electron transport, J. Biol. Chem. 263 (1988) 14359–14367, https://doi.org/10.1038/nature19794. [34] Q. Chen Q, E.J. Vazquez, S. Moghaddas, C.L. Hoppel, E.J. Lesnefsky, Production of reactive oxygen species by mitochondria: central role of complex III, J. Biol. Chem. 278 (2003) 36027–36031. [35] L.K. Kwong, R.S. Sohal, Substrate and site specificity of hydrogen peroxide gen- eration in mouse mitochondria, Arch. Biochem. Biophys. 350 (1998) 118–126. [36] E.T. Chouchani, V.R. Pell, E. Gaude, D. Aksentijević, S.Y. Sundier, E.L. Robb, A. Logan, S.M. Nadtochiy, E.N.J. Ord, A.C. Smith, F. Eyassu, R. Shirley, C.H. Hu, A.J. Dare, A.M. James, S. Rogatti, R.C. Hartley, S. Eaton, A.S.H. Costa, P.D. Brookes, I. Ramírez-Camacho et al. Free Radical Biology and Medicine 129 (2018) 407–417 416 S.M. Davidson, M.R. Duchen, K. Saeb-Parsy, M.J. Shattock, A.J. Robinson, L.M. Work, C. Frezza, T. Krieg, M.P. Murphy, Ischaemic accumulation of succinate controls reperfusion injury through mitochondrial ROS, Nature 515 (2014) 431–435, https://doi.org/10.1038/nature13909. [37] H.L. Lee, C.L. Chen, S.T. Yeh, J.L. Zweier, Y.R. Chen, Biphasic modulation of the mitochondrial electron transport chain in myocardial ischemia and reperfusion, Am. J. Physiol. Heart Circ. Physiol. 302 (2012) H1410–H1422, https://doi.org/10. 1152/ajpheart.00731.2011. [38] M. Ovize, G.F. Baxter, F. Di Lisa, P. Ferdinandy, D. Garcia-Dorado, D.J. Hausenloy, G. Heusch, J. Vinten-Johansen, D.M. Yellon, R. Schulz, working Group of cellular biology of heart of European Society of Cardiology. postconditioning and protection from reperfusion injury: where do we stand? Position paper from the working group of cellular biology of the heart of the European Society of Cardiology, Cardiovasc. Res. 87 (2010) 406–423, https://doi.org/10.1093/cvr/cvq129. [39] S. Hernández-Reséndiz, C. Zazueta, PHO-ERK1/2 interaction with mitochondria regulates the permeability transition pore in cardioprotective signaling, Life Sci. 108 (2014) 13–21, https://doi.org/10.1016/j.lfs.2014.04.037. [40] M. Fujita, H. Asanuma, A. Hirata, M. Wakeno, H. Takahama, H. Sasaki, J. Kim, S. Takashima, O. Tsukamoto, T. Minamino, Y. Shinozaki, H. Tomoike, M. Hori, M. Kitakaze, Prolonged transient acidosis during early reperfusion contributes to the cardioprotective effects of postconditioning, Am. J. Physiol. 29 (2007) H2004–H2008, https://doi.org/10.1152/ajpheart.01051.2006. [41] A.K. Gardichela, D.F. Stowe, W.E. Antholine, M. Yang, A.K. Camara, Damage to mitochondrial complex I during cardiac ischemia reperfusion injury is reduced in- directly by anti-anginal drug ranolazine, Biochim. Biophys. Acta 2012 (1817) 419–429, https://doi.org/10.1016/j.bbabio.2011.11.021. [42] E. Lapuente-Brun, R. Moreno-Loshuetos, R. Acín-Pérez, A. Latorre-Pellicer, C. Colás, E. Balsa, E. Perales-Clemente, P.M. Quirós, E. Calvo, M.A. Rodríguez-Hernández, P. Navas, R. Cruz, A. Carracedo, C. López-Otín, A. Pérez-Martos, P. Fernández-Silva, E. Fernández-Vizarra, J.A. Enríquez, Supercomplex assembly determines electron flux in the mitochondrial electron transport chain, Science 340 (2013) 1567–1570, https://doi.org/10.1126/science.1230381. [43] R. Perez-Perez, T. Lobo-Jarne, D. Milenkovic, A. Mourier, A. Bratic, A. Garcia- Bartolome, E. Fernández-Vizarra, S. Cadenas, A. Delmiro, I. García-Consuegra, J. Arenas, M.A. Martín, N.G. Larsson, C. Ugalde, COX7A2L is a mitochondrial complex III binding protein that stabilizes the III2 + IV supercomplex without affecting respirasome formation, Cell Rep. 16 (2016) 2387–2398, https://doi.org/ 10.1016/j.celrep.2016.07.081. [44] E.J. Boekema, H.P. Braun, Supramolecular structure of the mitochondrial oxidative phosphorylation system, J. Biol. Chem. 282 (2007) 1–4, https://doi.org/10.1074/ jbc.R600031200. [45] G. Lenaz, M.L. Genova, Kinetics of integrated electron transfer in the mitochondrial respiratory chain: random collisions vs. solid state electron channeling, Am. J. Physiol. Cell Physiol. 292 (2007) C1221–C1239, https://doi.org/10.1152/ajpcell. 00263.2006. [46] S. Jang, T.S. Lewis, C. Powers, Z. Khuchua, C.P. Baines, P. Wipf, S. Javadov, Elucidating mitochondrial electron transport chain supercomplexes in the heart During ischemia-reperfusion, Antioxid. Redox Signal. 27 (2017) 57–69, https://doi. org/10.1089/ars.2016.6635. [47] M.G. Rosca, E.J. Vazquez, J. Kerner, W. Parland, M.P. Chandler, W. Stanley, H.N. Sabbah, C.L. Hoppel, Cardiac mitochondria in heart failure: decrease in re- spirasomes and oxidative phosphorylation, Cardiovasc. Res. 80 (2008) 30–39, https://doi.org/10.1093/cvr/cvn184. I. Ramírez-Camacho et al. Free Radical Biology and Medicine 129 (2018) 407–417 417 Contents lists available at ScienceDirect BBA - Bioenergetics journal homepage: www.elsevier.com/locate/bbabio Mitochondrial respirasome works as a single unit and the cross-talk between complexes I, III2 and IV stimulates NADH dehydrogenase activity Reyes-Galindo Meztli1, Suarez Roselia1, Esparza-Perusquía Mercedes, de Lira-Sánchez Jaime, Pardo J. Pablo, Martínez Federico, Flores-Herrera Oscar⁎ Departamento de Bioquímica, Facultad de Medicina, Universidad Nacional Autónoma de México, 04510 México City, México A R T I C L E I N F O Keywords: Respirasome Mitochondrial supercomplexes Complex I activity Ustilago maydis mitochondria ROS production A B S T R A C T Ustilago maydis is an aerobic basidiomycete that depends on oxidative phosphorylation for its ATP supply, pointing to the mitochondrion as a key player in its energy metabolism. Mitochondrial respiratory complexes I, III2, and IV occur in supramolecular structures named respirasome. In this work, we characterized the subunit composition and the kinetics of NADH:Q oxidoreductase activity of the digitonine-solubilized respirasome (1600 kDa) and the free-complex I (990 kDa). In the presence of 2,6-dimethoxy-1,4-benzoquinone (DBQ) and cytochrome c, both the respirasome NADH:O2 and the NADH:DBQ oxidoreductase activities were inhibited by rotenone, antimycin A or cyanide. A value of 2.4 for the NADH oxidized/oxygen reduced ratio was determined for the respirasome activity, while ROS production was less than 0.001% of the oxygen consumption rate. Analysis of the NADH:DBQ oxidoreductase activity showed that respirasome was 3-times more active and showed higher ainity than free-complex I. The results suggest that the contacts between complexes I, III2 and IV in the respirasome increase the catalytic eiciency of complex I and regulate its activity to prevent ROS pro- duction. 1. Introduction The proton electrochemical potential, ΔμH+, across energy trans- ducing membranes is the basis of the chemiosmotic hypothesis for en- ergy coupling [1–3]. In mitochondria this electrochemical potential is used for heat production, ion and substrate transport, ATP/ADP ex- change, and especially ATP synthesis [4]. The proton translocation across inner mitochondrial membrane occurs through three protein complexes termed NADH:coenzyme Q oxidoreductase (complex I), coenzyme Q:cytochrome c oxidoreductase (complex III2, which is a functional dimer), and cytochrome c oxidase (complex IV); ad- ditionally, the succinate:coenzyme Q oxidoreductase (complex II), that belongs to the electron transport chain, doesn't translocate protons but produce ubiquinol which is a mobile lipid electron carrier [4]. Three models have been proposed to explain the organization of the electron transport chain complexes: 1) “Random collision model”, proposed by Hackenbrock et al. [5] in which individual respiratory complexes in the inner membrane difuse freely, and electron transfer is based on random collisions between complexes and two small electron carriers, coenzyme Q and cytochrome c; 2) “Solid state model”, in which complexes are attached in supra-structures called super- complexes [6], which have been founded in mitochondria from mam- mals, plants, fungi, and bacteria; and 3) “Plasticity model”, which in- volves both previous models [7]. Supercomplexes have a wide distribution in the natural kingdoms, from bacteria to plants and animals. In Paracoccus denitrificans the su- percomplexes III2:IV1 [8] and I1:III2:IV1 have been reported [9]; while in Saccharomyces cerevisiae complex III2 could be attached to one (III2:IV1) or two (III2:IV2) monomers of complex IV [10,11]. In Neuro- spora crassa supercomplexes are composed of complexes I, III2, and IV in diferent proportions [12]. Supercomplex I1:III2 is the most abundant in potato (Solanum tuberosum), bean (Phaseolus vulgaris), barley (Hor- deum vulgare), and Arabidopsis thaliana [13–15]. Bovine heart mi- tochondrial supercomplexes described are I1:III2:IV1, I1III2, and III2IV1 [16]. If complexes I, III2 and IV are present in the supercomplex and NADH oxidation and oxygen reduction occur, they are called respira- some [10]. Actually, the architecture of respirasomes from porcine (Sus scrofa) heart mitochondria [17] and ovine (Ovis aries) heart mitochondria [18] has been determined by cryo-electron microscopy with a resolution of https://doi.org/10.1016/j.bbabio.2019.06.017 Received 1 November 2018; Received in revised form 20 June 2019; Accepted 22 June 2019 ⁎ Corresponding author at: Departamento de Bioquímica, Facultad de Medicina, Universidad Nacional Autónoma de México, Apdo. Postal 70-159, Coyoacán, 04510 México, Cd. Mx., México. E-mail address: olores@bq.unam.mx (O. Flores-Herrera). 1M. Reyes-Galindo and R. Suarez contributed equally to this paper. BBA - Bioenergetics 1860 (2019) 618–627 Available online 25 June 2019 0005-2728/ © 2019 Elsevier B.V. All rights reserved. T 5.4 Å and 5.8 Å, respectively. Initially, respirasomes were associated with electron channeling from NADH to oxygen and the enhancement of electron low between complexes [5,19]; however this hypothesis has been challenged by new structural and functional evidence [17,18,20]. Currently, the accepted role is related to structural stabilization of complex I [6,21], and pre- vention of oxygen radicals production [7,21–28]. However, the precise role of supercomplexes remains to be deined. In this work, digitonin-solubilized respirasomes from Ustilago maydis mitochondria were isolated and their subunit composition and activity were characterized. U. maydis is an aerobic basidiomycete which infects the corn and teocinte plants in a biotrophic way [29]. In the laboratory the non-pathogenic yeast form of U. maydis is easily maintained in standard growth conditions [30]. U. maydis contains the four classic mitochondrial respiratory complexes and depends on the oxidative phosphorylation for the supply of ATP [31]. Supercomplexes isolated from U. maydis contained complexes I, III2 and IV, as well as coenzyme Q and cytochrome c. NADH oxidation supported oxygen uptake and was sensitive to KCN, antimycin A or rotenone. Ad- ditionally, complex I activity from respirasome was inhibited by anti- mycin A or KCN, even upon the addition of coenzyme Q and cyto- chrome c, suggesting a tight functional interaction between complexes. Kinetic characterization of NADH:Q oxidoreductase activity showed that respirasome was 3-times more active than free-complex I, sug- gesting a stimulatory efect of the contacts between complexes in the respirasome. 2. Materials and methods 2.1. Cell culture and mitochondria isolation U. maydis cells (strain FB2) were prepared as previously described [31]. U. maydis mitochondria were isolated using the method described by Waterield and Sisler [32]. For details see Supplemental material. 2.2. Solubilization of respiratory supercomplexes The respiratory supercomplexes and complexes were solubilized from U. maydis mitochondria using digitonin (a very-mild detergent) as described by [33–35], with minor modiications [36]. Briely, U. maydis mitochondria (10mg/ml) were suspended in 3.5ml of 50mM Bis-Tris and 500mM 6-aminocaproic acid, pH 7.0 and 140 μl digitonin (50% stock) were added to reach a detergent/protein ratio of 2:1. Digitonin was added drop by drop while the mixture was gentle stirred in an ice bath and then incubated in this condition during 30min. The mixture was centrifuged at 100,000g for 30min at 4 °C and supernatant con- taining the supercomplexes and individual complexes was recovered and immediately loaded into a sucrose gradient (16–42%) for super- complexes isolation (vide infra). 2.3. Respirasome isolation Mitochondrial digitonin extract (16mg protein) was loaded on 24ml of a continuous sucrose gradient (16–42% sucrose, 15mM Tris, pH 7.4, 20mM KCl and 0.2% digitonin) and centrifuged at 131,000g for 16 h at 4 °C [36]. Afterward, 500 μl fractions were collected from the bottom of the gradient. Fractions containing respirasomes were iden- tiied by BN-PAGE (vide infra). These respirasomes samples were pooled and diluted 7-fold with 30mM HEPES, pH 8.0 and 5% glycerol; then were concentrated using a Centrifugal Filters Units (100K, Millipore Amicon Utra) to a inal volume of 100 μl, and stored at –70 °C until used. 2.4. Blue Native-PAGE and in-gel catalytic activity assays Samples from the sucrose gradient and the later supercomplexes fraction were loaded on a linear polyacrylamide gradient gel (4–10%) for Blue Native PAGE (BN-PAGE) [35]. The BN-PAGE bufers were 50mM Bis-Tris/HCl, pH 7.0 for the anode electrode, and 50mM tricine, 15mM Bis-Tris, pH 7.0 and the anionic Coomassie© Brilliant Blue R- 125 dye (0.02%) for the cathode electrode [35]. The voltage was set to 35 V for 10 h at 4 °C and the run was stopped when the sharp line of the dye approached the gel front. Molecular weight of the respiratory complexes and supercomplexes was determined by their electrophoretic mobility and in-gel catalytic activity, using the complexes of digitonine- solubilized bovine heart mitochondria as standards. The in-gel assays were performed as described by Jung [34] using gel loaded with isolated digitonine-solubilized supercomplexes from of U. maydis mitochondria. NADH dehydrogenase activity (NADH:- methylthiazolyldiphenyl tetrazolium bromide (MTT) oxidoreductase) was assayed at 20–25 °C in a bufer containing 1.2mMMTT and 1.0 mM NADH in 10mM Tris/HCl, pH 7.4. For succinate dehydrogenase activity (Succinate:MTT oxidoreductase) NADH was replaced by 10mM succi- nate, 0.2mM phenazine methosulfate (PMS), 5mM EDTA in 10mM K2HPO4, pH 7.4. NADH or succinate dehydrogenase activity was cor- related with the development of purple precipitates on the gel. When activity-staining appear (10–20min) the reaction was stopped with ixing solution (50% methanol, 10% acetic acid). To assay the activity of complex IV the gel was incubated in 50mM K2HPO4, pH 7.2, 4.7 mM 3,3′diaminobenzidine tetrahydrochloride (DAB) and 16 μM horse heart cytochrome c. After 30–40min of incubation at 20–25 °C, the activity was observed as a brown precipitate and the reaction was stopped with the ixing solution. Activity of complex V was assayed in 50mM glycine (adjusted to pH 8.0 with triethanolamine), 10mM MgCl2, 0.15% Pb (ClO4)2 and 5mM ATP. ATP hydrolysis correlated with the develop- ment of white lead phosphate precipitates. The reaction was stopped using 50% methanol, and subsequently the gel was transferred to water and scanned against a dark background as described previously [36,37]. 2.5. Kinetic characterization of NADH dehydrogenase activity from respirasomes and free-complex Activity of complex I (NADH:2,6-dimethoxy-1,4-benzoquinone (DBQ) oxidoreductase activity) from respirasomes or free-complex I was determined spectrophotometrically at 340 nm by following the oxidation of NADH (εNADH=6.22mM−1 cm−1) in an Agilent 8453 UV–visible spectrophotometer (Agilent Technologies, USA). Activity of isolated respirasomes or free-complex I was performed in a reaction mixture containing 120mM KCl, 5mM MgCl2, 1 mM EGTA, 30mM KH2PO4, pH 7.4, at 25 °C. Where indicated, isolated respirasomes or free-complex I, were added to the bufer described above plus 10 μM of horse heart cytochrome c [38], 10–1000 μM of DBQ and 10–150 μM NADH to start the reaction. Where indicated, rotenone (1–10 μM) [39–41], antimycin A (0.1–1 μM) [42–44], or cyanide (1–3mM) [45,46] were added. Addition of cyanide increased the pH to 7.8, but the activity of complex I and the respirasome was the same at pH 7.4 and pH 7.8 (data no shown). To explore the efect of phospholipids on the activity of free-complex I and respirasomes, the protocol reported by [47] was followed using asolectin or lecithin. Protein concentration of respirasome or free-complex I was 50 μg/ ml and the reaction was started by the addition of NADH. NADH ab- sorbance was continuously monitored and the time response was less than 1 s. Kinetic analysis of changes in NADH dehydrogenase activity (initial velocity) was carried out using the direct spectrophotometric recording. Initial velocities were further obtained from the slope of the linear region in each spectrophotometric recording, and the linear re- gion of the traces was corroborated with the plot of the irst derivative against time. Data were analyzed by robust, weighted, non-linear re- gression analysis using the SigmaPlot software (Systat Software, Inc., version 10.0). The data represent the average of eight independent experiments. Rotenone was added to inhibit NADH:DBQ M. Reyes-Galindo, et al. BBA - Bioenergetics 1860 (2019) 618–627 619 oxidoreductase activity of complex I; samples used in this work were 100% inhibited by rotenone, conirming that complex I was the only NADH dehydrogenase present in both samples. The concentration of complex I in respirasome and free-complex I samples was determined by a densitometry analysis of Coomassie© Brilliant Blue R-125 stained NUAM 77 kDa-subunits (ID Scafold-NCBI Q4P4Z1; ID KEGG/ PENDANT UMAG_10695, Table 1) from an SDS-Tricine-PAGE, using Coomassie stained BSA as a standard (see Supplemental material sec- tion). The gel was scanned and the stain-intensity of NUAM subunit and BSA was determined by the Image Analysis software version 1.0 (Thermo Fisher Scientiic Inc.). The intensities of NUAM subunit and BSA were measured by peak integration after densitometry analyses. The mol of NUAM subunit was determined using the molecular weight of the mature protein (Table 1). The amount of complex I in respira- somes and free-complex I samples was 3.3 ± 0.7 μg/10 μg total protein and 2.8 ± 0.5 μg/10 μg total protein, respectively; these amounts of complex I in respirasomes and free-complex I were used to kinetics parameters estimation. 2.6. Oxygen consumption by mitochondrial respirasomes Oxygen consumption by isolated respirasomes was determined using a type Clark electrode in the bufer described above at 30 °C. Mixture reaction was supplemented with 10 μM horse heart cytochrome c, 70 μM DBQ and 20–100 μM NADH. Maximum activity of complex IV from respirasomes was assayed with 4mM ascorbate and 6mM 2,3,5,6- tetramethyl-p-phenylendiamine (TMPD) to reduce the horse heart cy- tochrome c. Where indicated, rotenone (10 μM), antimycin A (1 μM), or cyanide (3mM) were added. 2.7. Quantification of hydrogen peroxide produced by respirasomes Quantiication of hydrogen peroxide was performed with Amplex® Red hydrogen peroxide assay kit (Invitrogen, Molecular Probes, USA), following the manufacturer instructions. Experimental conditions used were similar to those described in Section 2.5 (vide supra). Superoxide dismutase (50 U/ml) was added to the reaction mixture to accelerate the production of hydrogen peroxide from the superoxide anion. 2.8. Tandem mass spectrometry (LC/ESI–MS/MS) Protein identiication of isolated supercomplexes was determined by mass spectrometry performed by the Arizona Proteomics Consortium (Cancer Center and by the BIO5 Institute of the University of Arizona). Samples were prepared following the speciications of the Proteomics Core Laboratory. Scafold program (version Scafold_4.8.9, Proteome Software Inc., Portland, OR) was used to validate MS/MS based peptide and protein identiications. Peptide identiications were accepted if they could be established at greater than 95.0% probability by the Scafold Local FDR algorithm. Protein identiications were accepted if they could be established at greater than 99.0% probability and con- tained at least 2 identiied peptides. Protein probabilities were assigned by the Protein Prophet algorithm [48]. Proteins that contained similar peptides and could not be diferentiated based on MS/MS analysis alone were grouped to satisfy the principles of parsimony. Proteins sharing signiicant peptide evidence were grouped into clusters. Proteins were annotated with GO terms from NCBI (downloaded Apr 24, 2018) [49]. 2.9. Determination of protein concentration Samples were treated with 0.4% deoxycholate and the protein content was determined as described by Lowry et al. [50]. Bovine Table 1 Subunit identity and molecular mass of Ustilago maydis mitochondrial respirasomes. The identity of each subunit was determined by LC/ESI-MS/MS. Subunit identity MW mature protein (kDa) Exclusive unique peptides Unique exclusive spectra/total spectra Coverage (%) ID (Scafold-NCBI) ID (KEGG/PENDANT) Complex I NUAM 77.2 33 51/103 53 Q4P4Z1 UMAG_10695 NUBM 51.6 18 20/32 31 Q4PGP5 UMAG_11170 NUCM 48.2 17 27/47 45 Q4P4N9 UMAG_11162 NUGM 28.3 8 13/18 21 Q4PDY2 UMAG_11896 NUHM 24.3 7 8/15 32 Q4PGX9 UMAG_00634 NUKM 20.5 5 9/18 26 Q4P1W1 UMAG_11038 NUFM 13.4 5 7/15 51 Q4P7I2 UMAG_11517 NB4M 15.0 3 4/4 31 Q4PBS6 UMAG_02437 NUPM 17.2 4 6/8 44 UMAG_05598 NUEM 35.1 17 23/40 57 Q4PHN2 UMAG_00381 N7BM 14.1 5 6/11 41 Q4P6C6 UMAG_10847 NI2M 10.1 4 5/8 38 Q4P2N8 UMAG_05625 NUYM 17.2 6 10/16 34 Q4PHA1 UMAG_00512 NUJM 14.1 4 4/7 55 Q4PAH0 UMAG_11495 NUXM 20.9 5 8/11 46 Q4P5K7 UMAG_10989 NUZM 22.2 5 8/21 40 Q4P0U1 UMAG_12039 NB6M 10.7 5 6/14 52 Q4PF02 UMAG_01311 Complex III2 QCR2 43.2 17 27/49 56 Q4PEI5 UMAG_01478 Cyt1 30.4 10 15/29 36 Q4P5I2 UMAG_11534 QCR7 10.5 8 11/14 70 Q4P6M6 UMAG_04237 RIP1 26.5 6 9/22 21 Q4P7T8 UMAG_10507 Complex IV Cox2 28.6 5 8/14 30 Q0H8Y7 Q0H8Y7 Cox4 12.8 3 4/9 41 Q4P511 UMAG_04802 Cox5A 16.8 7 10/20 46 Q4P348 UMAG_05465 Cytochrome c 11.9 3 4/7 39 XP_011389077 UMAG_02708 The identity of each protein was determined by mass spectrometry. The subunit molecular weight of the mature protein was determined by 2D-Tricine-SDS-PAGE and corroborated with the molecular weight obtained from U. maydis genome analysis (Biomax informatics ag; http://pedant.helmholtz-muenchen.de/ pedant3htmlview/pedant3view?Method=analysis&Db=p3_t237631_Ust_maydi_v2GB). M. Reyes-Galindo, et al. BBA - Bioenergetics 1860 (2019) 618–627 620 serum albumin (BSA) was used as standard. 2.10. Materials Analytical grade reagents were purchased from Sigma Chemical Co. (St. Louis, MO, USA), E. Merck (Darmstadt, Germany), and BioRad (Hercules, CA, USA). Strain FB2 of U. maydis was obtained from the American Type Cell Collection (Manassas, VA, USA). 3. Results 3.1. Composition of isolated respirasomes Respiratory complexes and supercomplexes from U. maydis mi- tochondria were eiciently solubilized with digitonin preserving their activity (Fig. 1A). In gel-activity of individual complex I, II, IV and V was associated with a protein band of a molecular mass of 960, 150, 240 and 640 kDa, respectively; additionally, ATPase activity of complex V was associated with a single protein band of 1260 kDa, which has been reported as the dimer of F1F0-ATP synthase [36]. Activities of complexes I and IV were associated with several bands with molecular masses from 1400 to 1900 kDa (Fig. 1A). MS/MS analysis conirmed the presence of complex III2 in these supercomplexes. Digitonin-solubilized respirasomes from U. maydis were isolated by sucrose density gradient centrifugation (Fig. 1B). NADH:MTT oxidor- eductase activity from complex I was distributed from fraction 3 to 30; however, fractions 3–11 contained exclusively the respirasome, and this pattern was highly reproducible (Fig. 1B). Free-complex I was re- covered from fractions 25–30. The fractions containing respirasomes and free-complex I were pooled separately and concentrated as de- scribed in the Materials and methods section and their purity, in terms of NADH dehydrogenase activity, was analyzed by BN-PAGE (Fig. 1C). For the respirasomes, activities of complex I and IV were associated with a main protein band of 1600 kDa, using bovine mitochondrial respiratory complexes solubilized with digitonin as standard (Fig. 1C). No activity of monomeric complex I and IV, or complex III2 stained with Coomassie was observed, demonstrating that supercomplexes were isolated without contamination by individual complexes. Free-complex I activity was located around 970 kDa as a single band (Fig. 1C, right lane). Although activities of complexes I and IV were observed in the upper zone of the gel (Fig. 1B), suggesting a broad spectrum of supercomplexes stoichiometries, a main protein band of 1630 kDa was observed in the gel stained with Coomassie (Fig. 1C). Using the molecular weight obtained from U. maydis genome database (Biomax informatics ag; http://pedant.helmholtz-muenchen.de/ pedant3htmlview/pedant3view?Method=analysis&Db=p3_t237631_ Ust_maydi_v2GB) for complex I (900 kDa), dimer of complex III2 (473 kDa), and complex IV (203 kDa), we hypothesize that the minimum and most probable stoichiometry of this 1630 kDa super- complex is I1:(III2)1:IV1. Seventeen subunits for complex I, 4 subunits for complex III2, and 3 for complex IV were identiied by MS/MS analysis of isolated respirasomes (Table 1), conirming their composi- tion. Additionally, cytochrome c was detected in the respirasome Fig. 1. Isolation and in-gel activity of the respirasome. Respiratory complex and supercomplexes from U. maydis mitochondria were solubilized with digitonin (A). Left panel shows the Coomassie-stained native gel strips; CI, CII, CIV, and CV corresponding to in-gel activities assay of complexes I, II, IV, and V, respectively. Respirasome was isolated by sucrose-gradient ultracentrifugation and its in-gel NADH dehydrogenase activity analyzed by BN-PAGE (B). Fractions from the bottom [3–11] and the top [25–30] of the sucrose gradient were used to obtain isolated the respirasome and free-complex I, respectively (C). Where is showed Bos taurus respiratory complex and supercomplexes were solubilized with digitonin and used as standard. U. maydis respirasome sample was used to subunits identiication by MS/MS (Table 1). M. Reyes-Galindo, et al. BBA - Bioenergetics 1860 (2019) 618–627 621 sample (Table 1). Complex I amount was determined as described in Materials and methods section and Supplementary material. 3.2. Oxygen consumption by respirasomes Surprisingly, isolated respirasomes reduced oxygen in the presence of NADH (133 ± 17 nmol O2 reduced·(mg of CI·min)−1), suggesting that they contain, in addition to functional respiratory complexes I, III2 and IV, the mobile elements, coenzyme Q and cytochrome c, allowing the electron lux from NADH to oxygen (Fig. 2A). If mobile electron carriers (i.e. 60 μg cytochrome c/ml and 65 μM DBQ) were added to the reaction mixture, oxygen consumption increased (397 ± 41 nmol O2 reduced·(mg of CI·min)−1; Fig. 2B–E), suggesting that these could be used as substrates by isolated respirasomes. Once NADH has been oxidized, oxygen uptake decreases (Fig. 2A–C), and a new NADH ad- dition promoted respiration again. Although oxygen uptake was sup- ported by NADH oxidation, the maximum complex IV activity was reached with ascorbate and TMPD addition (992 ± 203 nmol O2 re- duced·(mg of CI·min)−1; Fig. 2C–E), indicating that lux control could belong to complexes I or III2. Oxygen reduction by respirasome in the presence of NADH, cytochrome c and DBQ was inhibited by rotenone (Fig. 2D), antimycin A (Fig. 2E), or cyanide (Fig. 2A–E). The total in- hibition of electron lux by these speciic inhibitors discards the pre- sence of rotenone-insensitive alternative NADH dehydrogenases (i.e. Nde1, Nde2, or Ndi1) or cyanide-resistant alternative oxidase [31] in isolated respirasomes. Additionally, MS/MS analysis conirms the ab- sence of these alternative respiratory elements in the U. maydis re- spirasome. The absence of hydrogen peroxide (H2O2) as a result of electron leak during NADH oxidation was demonstrated by the addition of catalase (Fig. 2A and B). Alternatively, H2O2 production by re- spirasomes, assayed with the Amplex Red probe, was of 240 ± 4 pmol of H2O2·(mg of CI·min)−1 in the presence of 150 nmol of NADH to as- sess the maximum rate of oxygen consumption; Cyanide addition increased ROS production to 550 ± 10 pmol of H2O2·(mg of CI·min)−1, and 484 ± 24 pmol of H2O2·(mg of CI·min)−1 in the absence or pre- sence of superoxide dismutase, respectively. These values represent 0.001% of the maximum rate of oxygen consumption supported by NADH. Since H2O2 production was negligible (i.e. less than 0.001%) during maximum oxygen consumption by respirasome, one question is arising, the NADH:DBQ oxido-reductase activity of complex I occur even if electron low in complex III2 and IV in supercomplexes is interrupted by antimycin A or cyanide? To answer this question, NADH:DBQ oxido- reductase activity by respirasomes was monitored in the presence of inhibitors of complex I, III2 or IV. 3.3. NADH oxidation by respirasome Oxidation of NADH by respirasomes was recorded spectro- photometrically at 340 nm in the presence of cytochrome c and DBQ (Fig. 3). NADH dehydrogenase activity of respirasomes was inhibited by rotenone (Fig. 3A) demonstrating that this activity belongs ex- clusively to complex I and alternative NADH dehydrogenases were absent. Interestingly, reduction of DBQ by complex I was stopped when complex III2 and complex IV were inhibited by antimycin A or cyanide, respectively, even in the presence of an excess of coenzyme Q and cy- tochrome c (Fig. 3B and C). Inhibition of the NADH:DBQ oxidor- eductase activity occurs even if inhibitor (i.e. Antimycin A or cyanide) was added before NADH (Fig. 3B and C). This observation indicates that activity of complex I is tightly coordinated with the activities of com- plexes III2 and IV in the respirasome (Fig. 2). Antimycin A and cyanide have no efect on free-complex I activity (Fig. 3E and F). Remarkably, the ratio between NADH oxidation (i.e. 959 ± 197 nmol NADH oxi- dize/mg of CI·min; see Fig. 3) and oxygen reduction (i.e. 397 ± 81 nmol O2 reduced/mg of CI·min, see Fig. 2) by respirasomes was 2.42 ± 0.3, very close to the theoretical value of 2 for the electron Fig. 2. Oxygen consumption by isolated Ustilago maydis respirasome. (A) Electron lux into respira- some was started by 150 nmol of NADH addition in 120mM KCl, 5mM MgCl2, 1 mM EGTA, 30mM KH2PO4, pH 7.4, at 30 °C. (B) Mobile electron car- riers stimulate respirasome oxygen uptake (i.e. 65 μM DBQ and 60 μg cytochrome c/ml). Maximum activity of complex IV was reached by ascorbate (4mM) and TMPD (6mM) addition (C). Where is showed 15 μM rotenone (D), 3 μM antimycin A (E) or 3 mM cyanide (A–E) were added to inhibit respira- tion through complexes I, III2 or IV, respectively. Catalase=2000–4000 Units. Numbers below each recording represent the velocity of oxygen con- sumption=nmol O2/mg of CI·min. Diferent NADH concentrations were added to assay the respirasomes response and the oxygen reduction mean ± S.D. was: 133 ± 17 nmol O2 reduced/mg of CI·min with 30 nmol of NADH added (n=6); 266 ± 30 nmol O2 reduced/mg of CI·min with 75 nmol of NADH added (n=6); and 397 ± 81 nmol O2 reduced/mg of CI·min with 150 nmol of NADH added (n=10). The x-axis scale showed at bottom of C and E is the same for the upper recording (A, B and D). Results were obtained from 7 diferent preparations. M. Reyes-Galindo, et al. BBA - Bioenergetics 1860 (2019) 618–627 622 lux from NADH to oxygen through respiratory complexes I-III2-IV. Oxygen uptake stimulation by the addition of DBQ and cytochrome c (Fig. 2A and B) suggests that mobile electron carriers were taken from the medium, and their microdifusion between complexes could play an important role in electron lux. Interestingly, inhibition of complexes III2 or IV stops the NADH:DBQ oxidoreductase activity of the respira- some (Fig. 3), suggesting that protein-protein contacts between com- plexes I, III2 and IV could play an important role in the regulation of respirasome activity. In this sense, we have previously characterized the role of protein-protein interactions in the activity of the complex V dimer [36]. Then, the characterization of NADH dehydrogenase activity in respirasomes and free-complex I could be helpful to elucidate the role of the interactions between complexes in the respirasomes. 3.4. NADH dehydrogenase activity from complex I versus NADH dehydrogenase activity from respirasome As a irst step, activity of NADH:DBQ oxidoreductase from respira- some and free-complex I was determined following the change in ab- sorbance at 340 nm. Activity of NADH:DBQ oxidoreductase by respirasome (Fig. 4A and B) or free-complex I (Fig. 4C and D) increased as the NADH or DBQ concentrations were raised. The data were itted to the Michaelis- Menten equation. Respirasome showed a Vmax=3340 ± 150 nmol NADH oxidized·(mg of CI·min)−1, a KM-NADH=19 ± 4 μM and a KM- DBQ=76 ± 15 μM. In contrast, free-complex I showed a Vmax value of 1000 ± 80 nmol NADH oxidized·(mg of CI·min)−1, a KM- NADH=50 ± 11 μM and a KM-DBQ=103 ± 30 μM (Table 2). It has been reported that phospholipid reconstitution could increase the ac- tivity of DDM-isolated complex I from Yarrowia lipolytica [47]; how- ever, this efect was not observed in the free-complex I or respirasome from U. maydis (see Suppl material). The Lineweaver-Burk plot for free- complex I as well as complex I from respirasome was consistent with a random Bi Bi mechanism, which predicts a ternary complex (i.e. NADH- CI-DBQ). The kcat values for free-complex I and respirasome were 15 ± 1 s−1 and 49 ± 2 s−1, respectively (Table 2), conirming that respirasome is 3-times more active than free-complex I. Additionally, the kcat/KM-NADH values of free-complex I and respirasome were 2.9×105 ± 0.2× 105 and 2.6×106 ± 0.6×106M−1 s−1, respectively; while the kcat/KM- DBQ values were 1.4× 105 ± 0.3×105 and 6.5×105 ± 1.3×105M−1 s−1 for free-complex I and respirasome, respectively (Table 2), indicating a higher speciicity of respirasome for NADH and DBQ. These results indicate that incorporation of complex I into respirasomes and the interaction between complexes I, III2 and IV, has a stimulatory efect on the activity and ainity of complex I. 3.5. Proteins associated to respirasomes The MS/MS analysis of isolated respirasomes showed ive proteins related with the organization of mitochondrial architecture: Prohibitins 1 (UMAG_11092) and 2 (UMAG_05030); Fcj1 (UMAG_00635), Rcf2 (UMAG_03929) and the mitochondrial inner membrane organizing system protein 1 (UMAG_10488). All these proteins showed Fig. 3. Efect of rotenone, antimycin A and cyanide on NADH:DBQ oxidoreductase activity of respirasome. Respirasomes were incubated in the bufer described in Fig. 2. NADH oxidation was monitored by NADH absorbance at 340 nm. Where is indicated respirasome, DBQ (65 μM), cytochrome c (60 μg/ml), NADH (150 nmol), and (A) rotenone (15 μM), (B) antimycin A (5 μg/ml), or (C) cyanide (3mM), were added. Addition of inhibitor before NADH is showed as +Rotenone (A), +Antimycin A (B), or +CN- (C). Activity of free-complex I was assayed in similar conditions in the presence of rotenone (D), antimycin A (E) or cyanide (F). The number below each recording represents the velocity of NADH oxidation. The mean ± S.D. value of NADH:DBQ oxidoreductase activity for respirasome in this experimental condition was 959 ± 197 nmol NADH oxidize/mg of CI·min (n= 9). Results were obtained from 7 diferent preparations. M. Reyes-Galindo, et al. BBA - Bioenergetics 1860 (2019) 618–627 623 transmembrane domains but only prohibitin 2 and Rcf2 presented a glycine zipper motif in the transmembrane domains. It has been sug- gested that proteins with transmembrane glycine zippers play a struc- tural role as glue between the membrane proteins [51]. Finally, three proteins with the pentatricopeptide repeat domain (PPR domain: UMAG_06347 and UMAG_11282; PET127: UMAG_02275) were asso- ciated with the respirasome sample. 4. Discussion Mitochondrial respiratory complex I, III2 and IV are membrane proton-pumps that transform the energy of NADH into the proton electrochemical gradient (ΔμH+) across the inner membrane. The free energy stored in the ΔμH+ is utilized for ATP synthesis. Actually, stable interactions between the respiratory complexes have been determined, and these new structures are named supercomplexes. Although com- position and stoichiometry of supercomplexes are diverse, if complexes I, III2 and IV are presents, the electron low could occur from NADH to oxygen (i.e. a respirasome). The respirasome has been described in mitochondria from diferent eukaryotes such as bovine [16], Neurospora crassa [12], and prokaryotes as the α-proteobacteria Paracoccus deni- trificans [9]. In this work the U. maydis respirasome was eiciently solubilized with digitonine and isolated as a highly stable unit (Fig. 1). Composi- tion of U. maydis respirasome was assessed by MS/MS analysis (Table 1), in-gel activity (i.e. complex I and IV), and NADH:O2 oxi- doreductase activity (i.e. electron lux from NADH to oxygen). The minimal stoichiometry of the respirasome was I1III2IV1 with an ap- parent molecular mass of 1630 kDa (Fig. 1). Oxygen consumption supported by NADH oxidation strongly in- dicates that cytochrome c and coenzyme Q were present in isolated U. maydis respirasome (Fig. 2A), similar to that described for the respira- some obtained from bovine heart mitochondria [52]. This could be Fig. 4. Kinetic characterization of NADH:DBQ activity from respirasome and free-complex I. The dependence of respirasome (A) and free-complex I (C) activity on NADH at diferent ixed concentrations of DBQ, and dependence of respirasome (B) and free-complex I (D) activity on DBQ at diferent ixed NADH concentrations, were itted to the Michaelis-Menten equation. In (A) and (C) the ixed concentration of DBQ was (●)= 10; (○)= 50; (▲)= 100; (Δ)= 200; (■)= 400; (□)= 600; (▼)= 800; and (∇)= 1000 μM. In (B) and (D) the ixed concentration of NADH was (●)= 10; (○)= 50; (▲)= 75; (Δ)= 100; and (■)= 150 μM. The data are the average of four replicates from ive independent preparations. The activity was corrected by complex I amount in each preparation of respirasomes and free-complex I as described in Materials and methods section. Error bars represent S.D. Table 2 Kinetics parameters of NADH:DBQ oxide-reductase activity of the respirasome and the free-complex I from Ustilago maydis mitochondria. Free-complex Ia Respirasomea Vmax (nmol NADH oxidized/ mg complex I·min−1) 1000 ± 80 3340 ± 150 kcat (s−1) 15 ± 1 49 ± 2 NADH KM (μM) 50 ± 11 19 ± 4 kcat/Km (M−1 s−1) 2.9× 105 ± 0.2×105 2.6× 106 ± 0.6×106 DBQ KM (μM) 103 ± 30 76 ± 15 kcat/Km (M−1 s−1) 1.4× 105 ± 0.3×105 6.5× 105 ± 1.3×105 a Complex I mol in free-Complex I and respirasome samples was determined as described in Materials and methods section, and kinetics parameters were showed as mg of Complex I. M. Reyes-Galindo, et al. BBA - Bioenergetics 1860 (2019) 618–627 624 possible because the binding of cytochrome c to complex III2 is strong [53,54], and coenzyme Q was embedded in the lipid annulus of the respirasome. However, incorporation of fresh cytochrome c and DBQ into isolated respirasomes increased the electron low from NADH to oxygen, suggesting that the respirasome can take the quinone and cy- tochrome c from the medium. Recent experimental evidences support that substrate channeling inside the respirasome doesn't occur. Electron microscopy showed that each monomer in complex III dimer (III2) contains two Q binding sites (QP and QN), which are located on opposite sides of the dimer. The activity of complex III2 involves oxidation of QH2 and reduction of cytochrome c, in a process called the Q-cycle. QH2 binds at the QP site and its electrons are transferred, one to cytochrome c in the inter- membrane space via the Rieske protein, and the other to a Q bound at the QN site, generating a semiquinone (Q%). A new QH2 binds at the QP site and a second cytochrome c is reduced, and the Q% bound at the QN site is fully reduced to QH2, which is released into the ubiquinone pool. Kinetic analysis has demonstrated the alternating activity of each monomer [55]. The cryo-electron microscopy analysis of the respirasome shows that the Q binding site of one monomer of complex III2 is facing the Q site of complex I but separated by ~10 nm (~100 Ǻ), while the other monomer is facing complex IV [18]. Also, these studies showed that there were no proteins involved in substrate channeling between the Q sites of complex I and complex III2 [18]. Indeed, a time-resolved kinetic study of the respiratory chain in submitochondrial particles or mitochondrial membranes showed that Q exists as a single, common pool which is exchanged freely between respiratory complexes, including supercomplexes [56]. An elegant ap- proach to test the electron channeling in the respirasome was provided by Fedor and Hirst [20], who hypothesized that in the presence of electron channeling between complexes I and III2, the respiratory ac- tivity supported by NADH should be insensitive to enzymes that take the QH2 from the quinone pool. Using a cyanide-insensitive, non-elec- trogenic quinol oxidase (AOX), they probed that adding AOX to bovine submitochondrial particles the NADH oxidation rate increases and be- comes cyanide insensitive. They concluded that channeling doesn't occur because quinol produced by complex I is released into the qui- none pool and oxidized by AOX [20]. However, a recent report shows that in the inner mitochondrial membrane complex I is heterogeneously distributed [57]. In bovine heart about 44% of complex I occurs as a single copy (I1); 16% as I1III2 supercomplexes, and 40% as I1III2IV1–2 respirasome [57]. In Yarrowia lipolytica the arrangements of complex I founded were: complex I by itself (40%); supercomplexes I1:III2 (13%); and I1III2IV1–2 (47%) [57]. In this sense, the kinetic behavior of complex I in the diferent su- percomplexes found in the membrane is not necessarily the same. An approach to determine the properties of respirasome is to isolate it. Interestingly, although the NADH:O2 oxidoreductase activity of U. maydis respirasome was sensitive to classical respiratory inhibitors (Fig. 2), inhibition of complexes III2 or IV prevented NADH oxidation (Fig. 3) in the presence of an excess of DBQ and cytochrome c. Parti- cularly, inhibition of NADH:DBQ oxidoreductase activity by antimycin A or cyanide strongly supports the idea that interactions between complex I and complexes III2 and IV might regulate the activity of complex I, and therefore the respirasome activity. In line with this hypothesis, inhibition of NADH:O2 oxidoreductase activity in the re- spirasome with antimycin A or cyanide does not induce H2O2 or su- peroxide production, as conirmed by catalase addition (Fig. 2) or Amplex Red assay in the presence or absence of superoxide dismutase. Since complex IV doesn't use the product of complex I, and its active site is ~20 nm away from complex I Q site [18], the cross-talk (i.e. complex-complex contacts) between complexes IV and I is a working hypothesis to explain that inhibition of complex IV induces the in- hibition of complex I. It has been described that intersubunit contacts play a signiicant role in the catalytic properties of many enzymes. Particularly, in the 3- deoxy-d-manno-octulosonate-8-P synthase the subunit interphase is important for substrate selectivity and binding [58]; and in the gluco- samine-6-P deaminase from Escherichia coli the contacts between the subunits modify the allosteric equilibrium between the R and T-state [59]. In U. maydis the contacts between monomers in the dimer of F1F0- ATP synthase increase the ATPase activity and decrease the IC50 for oligomycin [36]. In the respirasome, interactions between complex IV and complex I involve subunit 7A of complex IV and either ND5 or the 39-kDa subunit of complex I. Additionally, complexes IV and III2 interact through subunits cox 7A and cox 8B and regions of subunits QCR9, QCR8 of complexes IV and III2, respectively [57]. To explore the efect of complex-complex contacts in the respira- some, we decided to determine NADH:DBQ oxidoreductase activity of complex I in two states, free as well as incorporated in the respirasome. Free-complex I from U. maydis showed a Vmax of 1000 ± 80 nmol NADH oxidized·(mg of complex I·min)−1; similar values (420–840 nmol NADH·(mg·min)−1) have been reported for Yarrowia lipolitica [44,60]. However, kinetic analysis showed that the activity of U. maydis re- spirasomal complex I was 3-time higher than that of the individual complex I (Vmax=3340 ± 150 nmol NADH oxidized·(mg of complex I·min)−1), and the ainity for NADH and DBQ was also higher. These observations indicate that incorporation of complex I in super- complexes (i.e. respirasome) increased their catalytic (kcat) and the speciicity (kcat/KM) constants, illustrating the possible role of the tight interactions between complexes I, III2 and IV. In contrast with our re- sults, the group of Shinzawa-Itoh described that the supercomplexes isolated from bovine heart mitochondria showed an NADH:Q1 oxidor- eductase activity of 700–1120 nmol NADH·(mg of SC·min)−1 in the presence of 150 μM NADH, 37 μM Q1, 200 μM cytochrome c and 2mM KCN [61]. The activity of U. maydis respirasome was 3-time higher than the activity of bovine supercomplexes, and under similar concentra- tions of NADH, coenzyme Q, and cytochrome c the U. maydis respira- some showed only 30% of its Vmax (Fig. 4). It's important to note that KCN was present in the mixture assay of the bovine supercomplexes activity [61], while KCN had a diferent efect on U. maydis respira- some. An important diference between bovine and U. maydis respirasomes is the dependence of complex I activity on active complexes III2 and IV. Here we showed that NADH:DBQ oxidoreductase activity of complex I was depended on active complex III2 and IV, even in the presence of an excess of DBQ and cytochrome c; moreover, inhibition of complex IV abolished complex I activity (Fig. 3). Bovine respirasome showed si- milar behavior only if coenzyme Q concentration was limiting, sug- gesting that activities of complex III2 and IV were important to remove the products [61]. Under this condition electron leak at mammalian complex I or contribution to QH2 pool would be probable [61]. Ad- ditionally, complex-complex contacts in U. maydis respirasome pro- motes NADH:DBQ activity, while in bovine respirasome this stimulating efect was not observed. To explain the efect of cyanide or antimycin A on the activity of complex I in respirasome, we hypothesized that the respirasome in U. maydis is “tighter”, and the interactions between the complexes pro- moted a conformational change in complex I (i.e. formation of deactive status of complex I). Studies on the deactive state of complex I, which is formed during ischemia, showed that an unstructured region in the loop of ND3 prevents coenzyme Q binding [62]. Actually, this deactive state is considered a regulatory mechanism of complex I to minimize ischemia-reperfusion injury [63]. Thus, although a deactive state of complex I in U. maydis must be determined experimentally, in the re- spirasome the cyanide inhibition of complex IV could recreate the ischemia condition, promoting the “deactive state” of complex I and preventing NADH oxidation and coenzyme Q reduction. Regarding the identity of proteins associated whit the U. maydis M. Reyes-Galindo, et al. BBA - Bioenergetics 1860 (2019) 618–627 625 respirasomes sample, prohibitins 1 and 2 [64,65] and Rcf2 [66,67] have an important role in the formation and stabilization of super- complexes. Although Fcj1 has a role in the formation of the mi- tochondrial crista [68], it has not been reported a role of this protein in respirasome formation. The common features of the pentatricopeptide repeat domain pro- teins and Pet127 detected in the respirasome sample are the mi- tochondrial localization and mRNA binding during the synthesis of proteins in mitochondria. Although deletion of some members of these proteins results in a complete loss of mitochondrial respiratory capacity [69,70], their precise role in the U. maydis respirasome should be ex- perimentally determined. Under physiological conditions, the interaction between complexes I-III2-IV in the respirasome may have two possible roles: 1) stimulate the NADH:DBQ oxidoreductase activity of complex I, and 2) provide a mechanism of regulation of complex I, stabilizing its deactive/active status to prevent ROS production. If cellular ATP decreases, individual mitochondrial respiratory complexes could associate into respirasomes to improve the electron lux, the proton pumping and generate the ΔμH+ needed for ATP synthesis. Contrary, if there is not a cellular de- mand for ATP (or an ischemic state) complex I can be in a deactive state to decrease NADH consumption and cytochrome c reduction, and then drop the electron lux reducing the ROS production. Author contribution Reyes-Galindo M, Suarez R and Flores-Herrera O designed and performed principal experiments; Esparza-Perusquía M, respirasomes isolation and technical assistance; de Lira-Sánchez J, ROS production analysis and technical assistance; Pardo JP and Martínez F analyzed data; Flores-Herrera O supervised project and wrote the paper with contributions from all authors. Transparency document The Transparency document associated with this article can be found, in online version. Declaration of Competing Interest The authors declare no conlict of interest. Acknowledgments We dedicate this work as a memorial to Professor PhD. Edmundo Chávez-Cossío, an exceptional friend and colleague, who passed away suddenly on December 24, 2018. This work was supported by Dirección General de Asuntos del Personal Académico (DGAPA) (IN222617, IN222117 and IN211715) from Universidad Nacional Autónoma de México (UNAM) and Consejo Nacional de Ciencia y Tecnología (CONACyT) (168025 and 254904). MR-G is a student of Facultad de Medicina (309210120) from UNAM and fellowship from Programa de Apoyo y Fomento a la Investigación Estudiantil (AFINES); RS is a stu- dent of Ingeniería Bioquímica (11120199) from Instituto Tecnológico de Morelia, México; ME-P is a PhD student of the Programa de Doctorado en Ciencias Biológicas (511021118) from UNAM and sup- ported by the CONACyT through a doctoral scholarship (254400). JdL- S is a PhD student of the Programa de Doctorado en Ciencias Biomédicas (518024299) from UNAM and supported by the CONACyT through a doctoral scholarship (666472). The authors thank to the Posgrado en Ciencias Biológicas and Ciencias Biomédicas from UNAM for the academic support. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.bbabio.2019.06.017. References [1] P. Mitchell, Coupling of photophosphorylation to electron and hydrogen transfer by a chemiosmotic type of mechanism, Nature. 191 (1961) 144–148. [2] P. Mitchell, Chemiosmotic Coupling and Energy Transduction, Glynn Research, LTD., Bodmin, Cornwall, U.K, 1968, pp. 1–111. [3] P. Mitchell, Chemiosmotic coupling in oxidative and phosphosynthetic phosphor- ylation, Biochim. Biophys. Acta 1807 (2011) 1507–1538. [4] D.G. Nicholls, S.J. Ferguson, Bioenergetics 4. London, Academic Press, 2013. [5] C.R. Hackenbrock, B. Chazotte, S.S. Gupte, The random collision model and a cri- tical assessment of difusion and collision in mitochondrial electron transport, J. Bioenerg. Biomembr. 18 (1986) 331–368. [6] J. Vonck, E. Schäfer, Supramolecular organization of protein complexes in the mitochondrial inner membrane, Biochim. Biophys. Acta 1793 (2009) 117–124. [7] R. Acín-Perez, J.A. Enriquez, The function of the respiratory supercomplexes: the plasticity model, Biochim. Biophys. Acta 1837 (2014) 444–450. [8] E.A. Berry, B.L. Trumpower, Isolation of ubiquinol oxidase from Paracoccus deni- tricans and resolution into cytochrome bc1 and cytochrome c-aa3 complexes, J. Biol. Chem. 260 (1985) 2458–2467. [9] A. Stroh, O. Anderka, K. Pfeifer, T. Yagi, M. Finel, B. Ludwig, H. Schagger, Assembly of respiratory complexes I, III, and IV into NADH oxidase supercomplex stabilizes complex I in Paracoccus denitrificans, J. Biol. Chem. 279 (2004) 5000–5007. [10] H. Schägger, K. Pfeifer, Supercomplexes in the respiratory chains of yeast and mammalian mitochondria, EMBO J. 19 (2000) 1777–1783. [11] R.A. Stuart, Supercomplex organization of the oxidative phosphorylation enzymes in yeast mitochondria, J. Bioenerg. Biomembr. 40 (2008) 411–417. [12] I. Marques, N.A. Dencher, A. Videira, F. Krause, Supramolecular organization of the respiratory chain in Neurospora crassa mitochondria, Eukaryot. Cell 6 (2007) 2391–2405. [13] H. Eubel, L. Jänsch, H.P. Braun, New insights into the respiratory chain of plant mitochondria. Supercomplexes and a unique composition of complex II, Plant Physiol. 133 (2003) 274–286. [14] H. Eubel, J. Heinemeyer, H.P. Braun, Identiication and characterization of re- spirasomes in potato mitochondria, Plant Physiol. 134 (2004) 1450–1459. [15] N.V. Dudkina, J. Heinemeyer, S. Sunderhaus, E.J. Boekema, H.P. Braun, Respiratory chain supercomplexes in the plant mitochondrial membrane, Trends Plant Sci. 11 (2006) 232–240. [16] H. Schägger, K. Pfeifer, The ratio of oxidative phosphorylation complexes I–V in bovine heart mitochondria and the composition of respiratory chain super- complexes, J. Biol. Chem. 276 (2001) 37861–37867. [17] G. Jinke, W. Meng, G. Runyu, Y. Kaige, L. Jianlin, G. Ning, Y. Maojun, The archi- tecture of the mammalian respirasome, Nature. 537 (2016) 639–643. [18] J.A. Letts, K. Fiedorczuk, L.A. Sazanov, The architecture of respiratory super- complexes, Nature. 537 (2016) 644–648. [19] M.L. Genova, A. Baracca, A. Biondi, G. Casalena, M. Faccioli, A.I. Falasca, G. Formiggini, G. Sgarbi, G. Solaini, G. Lenaz, Is supercomplex organization of the respiratory chain required for optimal electron transfer activity? Biochim. Biophys. Acta 1777 (2008) 740–746. [20] J.G. Fedor, J. Hirst, Mitochondrial supercomplexes do not enhance catalysis by quinone channeling, Cell Metab. 28 (2018) 525–531. [21] R. Acin-Perez, M.P. Bayona-Bafaluy, P. Fernandez-Silva, R. Moreno-Loshuertos, A. Perez-Martos, C. Bruno, C.T. Morales, J.A. Enriquez, Respiratory complex III is required to maintain complex I in mammalian mitochondria, Mol. Cell 13 (2004) 805–815. [22] F. Diaz, H. Fukui, S. Garcia, C.T. Moraes, Cytochrome c oxidase is required for the assembly/stability of respiratory complex I in mouse ibroblasts, Mol. Cell. Biol. 26 (2006) 4872–4881. [23] Y.C. Chen, E.B. Taylor, N. Dephoure, J.M. Heo, A. Tonhato, I. Papandreou, N. Nath, N.C. Denko, S.P. Gygi, J. Rutter, Identiication of a protein mediating respiratory supercomplex stability, Cell Metab. 15 (2012) 348–360. [24] E. Maranzana, G. Barbero, A.I. Falasca, G. Lenaz, M.L. Genova, Mitochondrial re- spiratory supercomplex association limits production of reactive oxygen species from complex I, Antioxid. Redox Signal. 19 (2013) 1469–1480. [25] E. Lapuente-Brun, R. Moreno-Loshuertos, R. Acín-Pérez, A. Latorre-Pellicer, C. Colás, E. Balsa, E. Perales-Clemente, P.M. Quirós, E. Calvo, M.A. Rodríguez- Hernández, P. Navas, R. Cruz, Á. Carracedo, C. López-Otín, A. Pérez-Martos, P. Fernández-Silva, E. Fernández-Vizarra, J.A. Enríquez, Supercomplex assembly determines electron lux in the mitochondrial electron transport chain, Science. 340 (2013) 1567–1570. [26] K. Ikeda, S. Shiba, K. Horie-Inoue, K. Shimokata, S. Inoue, A stabilizing factor for mitochondrial respiratory supercomplex assembly regulates energy metabolism in muscle, Nat. Commun. 4 (2013) 2147–2156. [27] M.L. Genova, G. Lenaz, Functional role of mitochondrial respiratory super- complexes, Biochim. Biophys. Acta 1837 (2014) 427–443. [28] Y. Chaban, E.J. Boekema, N.V. Dudkina, Structures of mitochondrial oxidative phosphorylation supercomplexes and mechanisms for their stabilization, Biochim. Biophys. Acta 1837 (2014) 418–426. [29] M.P. McCann, K.M. Snetselaar, A genome-based analysis of amino acid metabolism in the biotrophic plant pathogen Ustilago maydis, Fungal Genet. Biol. 45 (Suppl. 1) (2008) S77–S87. [30] J. Ruiz-Herrera, C.G. León, L. Guevara-Olvera, A. Cárabez-Trejo, A yeast-mycelial dimorphism of haploid and diploid strains of Ustilago maydis in liquid culture, M. Reyes-Galindo, et al. BBA - Bioenergetics 1860 (2019) 618–627 626 Microbiology. 1421 (1995) 695–703. [31] O. Juárez, G. Guerra, F. Martínez, J.P. Pardo, The mitochondrial respiratory chain of Ustilago maydis, Biochim. Biophys. Acta 1658 (2004) 244–251. [32] W.F. Waterield, H.D. Sisler, A convenient procedure for rapid release of protoplasts from Ustilago maydis, Biotechniques. 6 (1988) 832–834. [33] H. Schägger, W.A. Cramer, G. von Jagow, Analysis of molecular masses and oli- gomeric states of protein complexes by blue native electrophoresis and isolation of membrane protein complexes by two-dimensional native electrophoresis, Anal. Biochem. 217 (1994) 220–230. [34] C. Jung, C.M. Higgins, Z. Xu, Measuring the quantity and activity of mitochondrial electron transport chain complexes in tissues of central nervous system using blue native polyacrylamide gel electrophoresis, Anal. Biochem. 286 (2000) 214–223. [35] I. Wittig, M. Karas, H. Schägger, High resolution clear native electrophoresis for in- gel functional assays and luorescence studies of membrane protein complexes, Mol. Cell. Proteomics 6 (2007) 1215–1225. [36] M. Esparza-Perusquia, S. Olvera-Sanchez, J.P. Pardo, G. Mendoza-Hernandez, F. Martinez, O. Flores-Herrera, Structural and kinetics characterization of the F1F0- ATP synthase dimer. New repercussion of monomer-monomer contact, Biochim. Biophys. Acta Bioenerg. 1858 (2017) 975–981. [37] D. De los Ríos-Castillo, M. Zarco-Zavala, S. Olvera-Sanchez, J.P. Pardo, O. Juarez, F. Martinez, G. Mendoza-Hernandez, J.J. García-Trejo, O. Flores-Herrera, Atypical cristae morphology of human syncytiotrophoblast mitochondria: role for complex V, J. Biol. Chem. 286 (2011) 23911–23919. [38] K.M.C. Sinjorgo, T.B.M. Hakvoort, I. Durak, J.W. Draijer, J.K.P. Post, A.O. Muijsers, Human cytochrome c oxidase isoenzymes from heart and skeletal muscle: pur- iication and properties, Biochim. Biophys. Acta 850 (1987) 144–150. [39] Y. Nakashima, K. Shinzawa-Itoh, K. Watanabe, K. Naoki, N. Hano, S. Yoshikawa, Steady-state kinetics of NADH:coenzyme Q oxidoreductase isolated from bovine heart mitochondria, J. Bioenerg. Biomembr. 34 (2002) 11–19. [40] A.D. Vinogradov, V.G. Grivennikova, Generation of superoxide-radical by the NADH:ubiquinone oxidoreductase of heart mitochondria, Biochemistry (Mosc) 70 (2005) 120–127. [41] S. Dröse, A. Galkin, U. Brandt, Proton pumping by complex I (NADH:ubiquinone oxidoreductase) from Yarrowia lipolytica reconstituted into proteoliposomes, Biochim. Biophys. Acta 1710 (2005) 87–95. [42] M. Candela, E. Zaccherini, D. Zannoni, Respiratory electron transport and light- induced energy transduction in membranes from the aerobic photosynthetic bac- terium Roseobacter denitrificans, Arch. Microbiol. 175 (2001) 168–177. [43] A. Galkin, U. Brandt, Superoxide radical formation by pure complex I (NADH:ubiquinone oxidoreductase) from Yarrowia lipolytica, J. Biol. Chem. 280 (2005) 30129–30135. [44] F.A. Rotsaert, M.G. Ding, B.L. Trumpower, Diferential eicacy of inhibition of mitochondrial and bacterial cytochrome bc1 complexes by center N inhibitors an- timycin, ilicicolin H and funiculosin, Biochim. Biophys. Acta 1777 (2008) 211–219. [45] D. Guo, T. Nguyen, M. Ogbi, H. Tawik, G. Ma, Q. Yu, R.W. Caldwell, J.A. Johnson, Protein kinase C-epsilon coimmunoprecipitates with cytochrome oxidase subunit IV and is associated with improved cytochrome-c oxidase activity and cardioprotec- tion, Am. J. Physiol. Heart Circ. Physiol. 293 (2007) H2219–H2230. [46] T. Sugio, M. Fujii, Y. Ninomiya, T. Kanao, A. Negishi, F. Takeuchi, Reduction of Hg2+ with reduced mammalian cytochrome c by cytochrome c oxidase puriied from a mercury-resistant Acidithiobacillus ferrooxidans strain, MON-1, Biosci. Biotechnol. Biochem. 72 (2008) 1756–1763. [47] S. Dröse, K. Zwicker, U. Brandt, Full recovery of the NADH:ubiquinone activity of complex I (NADH:ubiquinone oxidoreductase) from Yarrowia lipolytica by the ad- dition of phospholipids, Biochim. Biophys. Acta 1156 (2002) 65–72. [48] A.I. Nesvizhskii, A. Keller, E. Kolker, R. Aebersold, A statistical model for identi- fying proteins by tandem mass spectrometry, Anal. Chem. 75 (2003) 4646–4658. [49] M. Ashburner, C.A. Ball, J.A. Blake, D. Botstein, H. Butler, J.M. Cherry, A.P. Davis, K. Dolinski, S.S. Dwight, J.T. Eppig, M.A. Harris, D.P. Hill, L. Issel-Tarver, A. Kasarskis, S. Lewis, J.C. Matese, J.E. Richardson, M. Ringwald, G.M. Rubin, G. Sherlock, Gene ontology: tool for the uniication of biology. The Gene Ontology Consortium, Nat. Genet. 25 (2000) 25–29. [50] O.H. Lowry, N.J. Rosebrough, A.L. Rarr, R.J. Randall, Protein measurement with the Folin-phenol reagent, J. Biol. Chem. 193 (1951) 265–275. [51] S. Kim, T.-J. Jeon, A. Oberai, D. Yang, J.J. Schmidt, J.U. Bowie, Transmembrane glycine zippers: physiological and pathological roles in membrane proteins, Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 14278–14283. [52] T. Althof, D.J. Mills, J.-L. Popot, W. Kühlbrandt, Arrangement of electron transport chain components in bovine mitochondrial supercomplex I1III2IV1, EMBO J. 30 (2011) 4652–4664. [53] J. Heinemeyer, H.-P. Braun, E.J. Boekema, R. Kouril, A structural model of the cytochrome c reductase/oxidase supercomplex from yeast mitochondria, J. Biol. Chem. 282 (2007) 12240–12248. [54] A. Nyola, C. Hunte, A structural analysis of the transient interaction between the cytochrome bc1 complex and its substrate cytochrome c, Biochem. Soc. Trans. 36 (2008) 981–985. [55] M. Castellani, R. Covian, T. Kleinschroth, O. Anderka, B. Ludwig, B.L. Trumpower, Direct demonstration of half-of-the-sites reactivity in the dimeric cytochrome bc1 complex. Enzyme with one inactive monomer is fully active but unable to activate the second ubiquinol oxidation site in response to ligand binding at the ubiquinone reduction site, J. Biol. Chem. 285 (2010) 502–510. [56] J.N. Blaza, R. Serreli, A.J.Y. Jones, K. Mohammed, J. Hirst, Kinetic evidence against partitioning of the ubiquinone pool and the catalytic relevance of respiratory-chain supercomplexes, Proc. Natl. Acad. Sci. U. S. A. 111 (2014) 15735–15740. [57] K.M. Daviesa, T.B. Bluma, W. Kühlbrandt, Conserved in situ arrangement of com- plex I and III2 in mitochondrial respiratory chain supercomplexes of mammals, yeast, and plants, Proc. Natl. Acad. Sci. U. S. A. 115 (2018) 3024–3029. [58] T.M. Allison, F.C. Cochrane, G.B. Jameson, E.J. Parker, Examining the role of in- tersubunit contacts in catalysis by 3-deoxy-d-manno-octulosonate 8-phosphate synthase, Biochemistry. 52 (2013) 4676–4686. [59] D.A. Cisnerosa, G.M. Montero-Moran, S. Lara-Gonzalez, M.L. Calcagno, Inversion of the allosteric response of Escherichia coli glucosamine-6-P deaminase to N-acet- ylglucosamine 6-P, by single amino acid replacements, Arch. Biochem. Biophys. 421 (2004) 77–84. [60] A. Waletko, K. Zwicker, A. Abdrakhmanova, V. Zickermann, U. Brandt, S. Kerscher, Histidine 129 in the 75-kDa subunit of mitochondrial complex I from Yarrowia li- polytica is not a ligand for [Fe4S4] cluster N5 but is required for catalytic activity, J. Biol. Chem. 280 (2005) 5622–5625. [61] K. Shinzawa-Itoh, H. Shimomura, S. Yanagisawa, S. Shimada, R. Takahashi, M. Oosaki, T. Ogura, T. Tsukihara, Puriication of active respiratory supercomplex from bovine heart mitochondria enables functional studies, J. Biol. Chem. 291 (2016) 4178–4184. [62] J.N. Blaza, K.R. Vinothkumar, J. Hirst, Structure of the deactive state of mammalian respiratory complex I, Structure. 26 (2018) 312–319. [63] E.T. Chouchani, V.R. Pell, E. Gaude, D. Aksentijevic, S.Y. Sundier, E.L. Robb, A. Logan, S.M. Nadtochiy, E.N.J. Ord, A.C. Smith, et al., Ischaemic accumulation of succinate controls reperfusion injury through mitochondrial ROS, Nature. 515 (2014) 431–435. [64] L.G. Nijtmans, S.M. Artal, L.A. Grivell, P.J. Coates, The mitochondrial PHB complex: roles in mitochondrial respiratory complex assembly, ageing and degenerative disease, Cell. Mol. Life Sci. 59 (2002) 143–155. [65] C. Jian, F. Xu, T. Hou, T. Sun, J. Li, H. Cheng, X. Wang, Deiciency of PHB complex impairs respiratory supercomplex formation and activates mitochondrial lashes, J. Cell Sci. 130 (2017) 2620–2630. [66] V. Strogolova, A. Furness, M. Robb-McGrath, J. Garlich, R.A. Stuart, Rcf1 and Rcf2, members of the hypoxia-induced gene 1 protein family, are critical components of the mitochondrial cytochrome bc1-cytochrome c oxidase supercomplex, Mol. Cell. Biol. 32 (2012) 1363–1373. [67] M. Vukotic, S. Oeljeklaus, S. Wiese, F.N. Vogtle, C. Meisinger, H.E. Meyer, A. Zieseniss, D.M. Katschinski, D.C. Jans, S. Jakobs, B. Warscheid, P. Rehling, M. Deckers, Rcf1 mediates cytochrome oxidase assembly and respirasome forma- tion, revealing heterogeneity of the enzyme complex, Cell Metab. 15 (2012) 336–347. [68] R. Rab, V. Soubannier, R. Scholz, F. Vogel, N. Mendl, A. Vasiljev-Neumeyer, C. Körner, R. Jagasia, T. Keil, W. Baumeister, M. Cyrklaf, W. Neupert, A.S. Reichert, Formation of cristae and crista junctions in mitochondria depends on antagonism between Fcj1 and Su e/g, J. Cell Biol. 185 (2009) 1047–1063. [69] O. Puchta, M. Lubas, K.A. Lipinski, J. Piatkowski, M.M.P. Golik, DMR1 (CCM1/ YGR150C) of Saccharomyces cerevisiae encodes an RNA-binding protein from the pentatricopeptide repeat family required for the maintenance of the mitochondrial 15S ribosomal RNA, Genetics. 184 (2010) 959–973. [70] J.P. Mayorga, Y. Camacho-Villasana, M. Shingú-Vázquez, R. García-Villegas, A. Zamudio-Ochoa, A.E. García-Guerrero, G. Hernández, X. Pérez-Martínez, A novel function of Pet54 in regulation of cox1 synthesis in Saccharomyces cerevisiae mi- tochondria, J. Biol. Chem. 291 (2016) 9343–9355. M. Reyes-Galindo, et al. BBA - Bioenergetics 1860 (2019) 618–627 627 Aspectos generales del transporte de colesterol en la esteroidogénesis de la placenta humana Nota: Artículo recibido el 08 de diciembre del 2018 y aceptado el xx de xxxx del 2019. ARTÍCULO DE REVISIÓN Resumen La placenta humana requiere de colesterol para sintetizar la progesterona que mantiene la relación entre el feto y la madre, lo que le permite concluir de manera exitosa el embarazo. La placenta incorpora el colesterol principalmente a través de las lipoproteínas de baja densidad (LDL) que se obtienen del torrente circulatorio materno por un mecanismo de endocitosis. A los endosomas que se generan en este proceso se les unen varias proteínas conformando los endosomas tardíos, que degradan las LDL y liberan el colesterol a las mitocondrias del sinciciotrofoblasto que lo transforman en pregnenolona y posteriormente en progesterona. Las proteínas de fusión de membranas denominados complejos SNARE participan en la liberación del colesterol en sitios de contacto especíicos en donde se localizan las proteínas mitocondriales responsables de la esteroidogénesis. Palabras Clave: placenta humana, receptores de lipoproteínas, mitocondrias, transporte de colesterol, esteroidogénesis, sitios de contacto. General aspects of cholesterol transport in the steroidogenesis of the human placenta Abstract The human placenta requires cholesterol to synthesize the progesterone that maintains the relationship between the fetus and the mother, which allows it to successfully conclude pregnancy. The placenta incorporates cholesterol through the LDL obtained from the maternal blood stream by a mechanism of endocytosis. Endosomes formed by this process degrade the LDL, bind several proteins forming the late endosomes and releasing the cholesterol to syncytiotrophoblast mitochondria to transform it into pregnenolone and then, into progesterone. The soluble attachment proteins denominated SNARE participates in the transport of cholesterol in speciic contact sites where the mitochondrial proteins responsible for steroidogenesis are located. Key Words: human placenta, lipoprotein receptors, mitochondria, cholesterol transport, steroidogenesis, contact sites. © 2019 Universidad Nacional Autónoma de México, Facultad de Estudios Superiores Zaragoza. Este es un artículo Open Access bajo la licencia CC BY-NC-ND (http://creativecommons.org/licenses/by-nc-nd/4.0/). TIP Revista Especializada en Ciencias Químico-Biológicas, 22: 1-9, 2019. DOI: 10.22201/fesz.23958723e. Sofía Olvera-Sánchez, Mercedes Esparza-Perusquía, Oscar Flores-Herrera, Viviana A. Urban-Sosa y Federico Martínez* Departamento de Bioquímica, Facultad de Medicina, Universidad Nacional Autónoma de México, Alcaldía de Coyoacán, 04510, Ciudad de México, México. E-mail: *fedem@bq.unam.mx PUBLICACIÓN CONTINUA TIP Rev.Esp.Cienc.Quím.Biol.2 Vol. 22 DOI: 10.22201/fesz.23958723e. a placenta humana tiene un papel primordial al establecer mecanismos de comunicación entre el feto y la madre durante el embarazo. En este proceso, la progesterona adquiere un papel central Introducción L al favorecer la implantación del blastocisto, la remodelación de la matriz extracelular y la promoción de la migración del trofoblasto (Halasz & Szekeres-Bartho, 2013). Además, la síntesis de progesterona se incrementa conforme avanza el embarazo (Morel et al., 2016). Estas funciones garantizan que el producto alcance la madurez que le permita vivir fuera del vientre materno (Burton & Fowden, 2015; Vaughan & Fowden, 2016; Ashary, Tiwari & Modi, 2018). La placenta es un órgano excepcional capaz de realizar de manera simultánea las funciones de diferentes tejidos, por ejemplo: sintetiza hormonas como las glándulas endocrinas, realiza el intercambio gaseoso como en los pulmones y iltra metabolitos como en los riñones, entre otras funciones (Chatuphonprasert, Jarukamjorn & Ellinger, 2018). A pesar del papel que desempeña la placenta, hasta el momento no se conocen con certeza los mecanismos maternos o fetales que la controlan, aunque se han sugerido posibles señales autocrinas, paracrinas o intracrinas. El linaje celular funcional de la placenta es el trofoblasto, que se divide en cito y sinciciotrofoblasto, este último es responsable del transporte de nutrientes y desechos al estar en contacto con las lagunas maternas y al mismo tiempo, el responsable del metabolismo esteroidogénico (en este artículo se considera la esteroidogénesis como la transformación del colesterol en progesterona). Si bien no se conocen varios aspectos del control del metabolismo placentario, lo que sí es claro es que la placenta humana tiene un papel central en las funciones del embarazo (Byrns, 2014), para mantener la relación materno- fetal al sintetizar cantidades elevadas de progesterona, una hormona esteroide que se produce en las mitocondrias del sinciciotrofoblasto y que regula, entre otras funciones, los procesos inmunológicos que evitan que el producto sea rechazado, ya que la carencia o disminución de progesterona podría provocar un aborto (Holland et al., 2017a). Se ha demostrado que los explantes de la placenta humana (pequeños fragmentos que contienen a las células del trofoblasto), son resistentes a condiciones físicas no isiológicas que un tejido “normal” no podría soportar. Por ejemplo, luego de la incubación en un medio hipotónico o a temperaturas superiores a los 45°C, los explantes realizan todas sus funciones metabólicas, incluyendo la síntesis de progesterona, de manera similar a los explantes que no fueron expuestos a estos tratamientos. Los resultados muestran que la placenta humana contiene mecanismos de autopreservación y como estrategia isiológica, un control metabólico y funcional independiente a los estímulos externos que desarrolló como una forma para proteger al producto (Paul, Jailkhani &Talwar, 1978, Paul, Gupta, Jailkhani &Talwar, 1980). En este sentido, se ha reportado que las mitocondrias de la placenta se adaptan de acuerdo con el avance del embarazo (Holland et al., 2017b). Por otro lado, aunque existen abundantes evidencias del metabolismo energético, aún falta mucho por conocer del funcionamiento de la placenta en el ser humano (Kaiser, 2014; Bianco-Miotto et al., 2016). Receptores a lipoproteínas Para la síntesis de progesterona es necesario el colesterol; sin embargo, debido a que la placenta no tiene la maquinaria para sintetizar las cantidades de colesterol que se requieren para la producción diaria de progesterona (van Leusden & Villee, 1965; Zelewski & Villee 1966), éste es proporcionado por las lipoproteínas maternas a través de los receptores especíicos del sinciciotrofoblasto. En las células del trofoblasto se ha descrito la presencia de varios tipos de receptores para las lipoproteínas, como el de las lipoproteínas de baja densidad (LDL) que son los que se expresan de manera principal (Winkel,Snyder, MacDonald & Simpson,1980; Furuhashi et al., 1989; Chatuphonprasert et al., 2018). Las lipoproteínas de alta densidad (HDL) se han identiicado en las células del trofoblasto (Wadsack et al., 2003), en las células BeWo (Pagler et al., 2006) y en varios tejidos esteroidogénicos. Sin embargo, a diferencia de las LDL, las HDL se procesan por una vía diferente a la lisosomal (Sanderson, 2009). Debido a que la síntesis de progesterona está sujeta a la capacidad de incorporar el colesterol de las LDL (Figura 1), se determinó que la concentración de los receptores para las lipoproteínas está regulada por la presencia de las mismas LDL. En células aisladas del trofoblasto y mantenidas en cultivo, la incubación con LDL-125 por tiempos largos, disminuyó cerca del 90% la captación y degradación de las LDL, lo que sugiere que las células del trofoblasto pueden regular el número de receptores de las LDL en la membrana plasmática. Este efecto no se observó cuando las células se incubaron en presencia de HDL (Winkel, MacDonald & Simpson, 1981). Sin embargo, datos recientes sugieren que la placenta usa tanto las LDL como las HDL para obtener el colesterol necesario para la síntesis de progesterona (Sanderson, 2009). La capacidad de la placenta de emplear ambas lipoproteínas asegura que el aporte de colesterol sea constante para mantener la esteroidogénesis, aun en condiciones que puedan afectar la concentración de cualquiera de éstas, asegurando así que el embarazo no esté en riesgo. El aporte del colesterol materno no se usa tan solo para la producción de progesterona, ya que también se requiere para satisfacer las necesidades del desarrollo del feto en crecimiento (Woollett, 2011; Bartels & O’Donoghue, 2011; Bhattacharjee et al., 2012; Baardman et al., 2013). Olvera Sánchez, S. et al.: Transporte de colesterol en la esteroidogénesis de la placenta 32019 DOI: 10.22201/fesz.23958723e. celular. Las diferencias en la concentración del colesterol en las membranas sugieren la existencia de un mecanismo de regulación celular mediado por al menos dos vías de transporte: vesículas y proteínas especíicas (Flis & Daum, 2013). En los tejidos esteroidogénicos, el transporte de colesterol al citocromo P450scc mitocondrial es relevante, ya que cuando se estimula la síntesis de pregnenolona, la primera hormona esteroide que se genera por la actividad de la cadena acoplada al P450scc, el transporte de colesterol hacia la mitocondria deber ser vectorial, eiciente y en la concentración necesaria para dar respuesta a las necesidades celulares y sin alterar sus funciones (Chien, Rosal & Chung, 2017). Tejidos esteroidogénicos y transporte mitocondrial del colesterol Aunque el colesterol es una molécula hidrofóbica que las células requieren para varias funciones, su concentración es altamente regulada, ya que los cambios en las concentraciones pueden ser fatales para la viabilidad celular; como en el caso de la estabilidad de las membranas. Se ha descrito que, en las células de los mamíferos, la presencia de colesterol en el retículo endoplásmico es baja, pues su concentración corresponde al 5% de los fosfolípidos; sin embargo, en el aparato de Golgi su concentración es mayor al 30% (Mesmin, Antonny & Drin, 2013). La mayoría del colesterol se localiza en la membrana plasmática, con un 60% del total del esterol Figura 1. Esquema de transporte del colesterol en el torrente circulatorio materno a las mitocondrias de las células de la placenta humana. Las LDL maternas se asocian a los receptores de las LDL (LDL-R) que son endocitadas conformando el endosoma, en donde el colesterol se libera de las LDL. Posteriormente, se asocian proteínas para constituir el endosoma tardío (Endosoma-T) que permite su asociación a la membrana externa mitocondrial (MEM), en donde se encuentran los sitios de contacto esteroidogénicos. En estos sitios se realiza el transporte de colesterol que ingresa a la membrana interna en donde se localiza la cadena del citocromo P450scc para su transformación en pregnenolona (P5) y posteriormente a progesterona (P4). Una vez sintetizada la P4, ésta se libera al torrente circulatorio para mantener la relación materno-fetal. Figura elaborada por los autores. TIP Rev.Esp.Cienc.Quím.Biol.4 Vol. 22 DOI: 10.22201/fesz.23958723e. sin embargo, en las mitocondrias de la placenta humana los niveles de colesterol son altos, semejantes incluso a los de la membrana plasmática (Navarrete, Flores-Herrera, Uribe & Martínez, 1999). Esto adquiere relevancia, ya que hasta el momento no se han descrito moduladores de la esteroidogénesis placentaria; sin embargo, esta característica habla de un transporte permanente de colesterol que permite mantener una producción de progesterona de manera sostenida durante el embarazo. Una vez que el colesterol ha sido transportado a la membrana externa mitocondrial, éste debe continuar su tránsito hasta la membrana interna donde se encuentra el citocromo P450scc para ser transformado en pregnenolona (Martínez & Strauss,1997).Esto sugiere que un paso limitante en la esteroidogénesis podría ser la transferencia de colesterol de la membrana externa a la interna mitocondrial. Se ha reportado que para el transporte de colesterol entre las membranas mitocondriales se requiere de los sitios de contacto (Strauss, Kishida, Christenson, Fujimoto & Hiroi, 2003; Miller & Bose, 2011). No obstante que la STARD1 es la principal proteína asociada al lujo de colesterol hacia la mitocondria y descrita para diferentes tejidos esteroidogénicos de regulación aguda, la placenta humana no la expresa (Miller, 2013). Sin embargo, la placenta sintetiza entre 250 y 500 mg de progesterona al día durante el último trimestre del embarazo (Simpson & MacDonald, 1981; LaVoie & King, 2009); esto implica un lujo continuo de colesterol hacia la mitocondria. En la última década se identiicó en las células del sinciciotrofoblasto a la proteína STARD3 (Watari et al., 1997), descrita originalmente en un tumor de mama, la cual presenta el dominio START de unión al colesterol característico de esta familia de proteínas. La STARD3 es una proteína endosomal con 4 segmentos transmembranales en su extremo amino terminal (dominio MENTAL). Alpy &Tomasetto (2014), reportaron que, aunque la STARD1 y STARD3 presentan un dominio START y comparten ciertas propiedades bioquímicas, también son diferentes en varios aspectos, que van desde su localización hasta su estructura (van der Kant, Zondervan, Janssen & Neefjes, 2013) demostraron que los endosomas tardíos y los lisosomas contienen la mitad del colesterol celular y que de ellos depende su distribución. También observaron que hay dos tipos de endosomas tardíos, uno presenta a la proteína ORP1L que transporta oxisteroles y otro contiene a la STARD3 y a la ABCA3 (una proteína transportadora que usa ATP para la translocación de sustratos, incluyendo lípidos), (van der Kant, Zondervan, Janssen & Neefjes, 2013). Esto les permitió sugerir que la STARD3 facilita el lujo de colesterol de los endosomas tardíos a la mitocondria, constituyendo un tipo de transporte vesicular. De igual manera, se ha descrito que la STARD3 requiere de los aminoácidos M307 y N311 Dentro de las vías de transporte del colesterol celular, se ha descrito una familia de proteínas denominadas STARD, con capacidad de unir al colesterol y participar en su distribución intracelular (Alpy et al., 2013). Estas proteínas se asocian a complejos multiproteicos especíicos que aseguran la precisa distribución del esterol. Aunado a esto, se han descrito asociaciones entre las membranas de los diferentes organelos celulares (ver más adelante), conformados por transportadores, enzimas y proteínas de andamiaje (Mesmin, Antonny & Drin, 2013), formando microdominios entre las membranas denominados sitios de contacto (Poderoso et al., 2013; Soientini & Graham, 2016). Se han descrito varios tipos de sitios de contacto, por ejemplo, las membranas asociadas a las mitocondrias (mitochondria- associated ER membrane, MAMs), que son un microdominio de asociación entre el retículo endoplásmico (RE) y las mitocondrias. Es importante destacar que la mitocondria tiene un papel relevante en el proceso esteroidogénico, ya que es ahí en donde reside la transformación del colesterol en pregnenolona (Papadopoulos & Miller, 2012), además de participar en otros procesos (Tait & Green, 2012), como la síntesis de ATP, señalización intracrina, la apoptosis, la inmunidad innata, la autofagia o la respuesta al estrés (Galluzzi, Kepp & Kroemer, 2012). Este tipo de microdominios se ha sugerido que contribuye al transporte de diferentes lípidos entre las membranas celulares. En la esteroidogénesis, parecen conjuntarse un sistema mediado por vesículas y otro por proteínas. Dentro del sistema proteico, se ha descrito a la STARD1 como la principal proteína que participa en la esteroidogénesis de gónadas y glándulas suprarrenales. Se ha demostrado que la STARD1 transiere colesterol eicientemente a la mitocondria (Tuckey, Headlam, Bose & Miller, 2002) y que su interacción con el colesterol induce un cambio conformacional de su estructura nativa a glóbulo fundido (‘moltenglobule’), que se ha sugerido como un estado dinámico que permite la transferencia del esterol (Rajapaksha, Kaur, Bose, Whittal & Bose, 2013). En la interacción de la STARD1 con la mitocondria durante el transporte del colesterol se han reportado complejos multiproteicos formados por un grupo de diversas proteínas entre las que se encuentran la VDAC (Bose, Whittal, Miller & Bose, 2008; Prasad et al., 2015), el receptor σ-1 (Marriott, Prasad, Thapliyal & Bose, 2012) y la ATPasa ATAD3 (Issop et al., 2015), por mencionar algunas. En este sentido, los complejos multiproteicos a su vez pueden interaccionar con elementos proteicos del retículo endoplásmico, con gotas lipídicas o endosomas tardíos (Beller, Thiel, Thul & Jäckle, 2010; Yang, Galea, Sytnyk & Crossley, 2012). Aunque hay un gran lujo de colesterol en los tejidos esteroidogénicos de regulación aguda, en las mitocondrias la cantidad de colesterol es baja (Horvath & Daum, 2013); Olvera Sánchez, S. et al.: Transporte de colesterol en la esteroidogénesis de la placenta 52019 DOI: 10.22201/fesz.23958723e. para unir al colesterol (van der Kant, Zondervan, Janssen & Neefjes, 2013). Asimismo, construyeron el knock-out del gen que codiica para la STARD3 y observaron que no se eliminó de manera total la síntesis de progesterona, sugiriendo que el transporte de colesterol mediado por la STARD3 no es el único camino que emplea la placenta humana para mantener la esteroidogénesis (van der Kant, Zondervan, Janssen & Neefjes, 2013). En el laboratorio, hemos determinado que la proteína HSP60 comparte epítopes con la STARD3 y que su presencia induce el incremento en la síntesis de progesterona de las mitocondrias del sinciciotrofoblasto (Olvera-Sánchez,Espinosa-García, Monreal, Flores-Herrera & Martínez, 2011). Recientemente se determinó que la HSP60 participa junto con la MLN64 en el transporte de colesterol para llevar a cabo la esteroidogénesis en la placenta humana (Monreal-Flores, Espinosa-García, García-Regalado, Arechavaleta-Velasco & Martínez, 2017). Se han descrito complejos multiproteicos responsables de conducir el colesterol hasta el P450scc, entre los que se encuentra el transduceosoma, compuesto de diversas proteínas membranales y citoplasmáticas y el metabolón esteroidogénico descrito como un conjunto de proteínas de las membranas mitocondriales (Fan & Papadopoulos, 2013). Al respecto, se reportó que la STARD3 es capaz de generar puntos de contacto entre el RE y los endosomas, como un paso previo al transporte de colesterol a las mitocondrias (Alpy et al., 2013) y un mecanismo particular para la esteroidogénesis de la placenta humana (Tuckey, Bose, Czerwionka & Miller, 2004; Martin, Kennedy & Karten, 2014). En su conjunto, la información muestra que de manera general el transporte de colesterol para la esteroidogénesis requiere de múltiples sistemas que involucran proteínas como la STARD1 y la STARD3, los endosomas tardíos, las gotas lipídicas, el citoesqueleto y los elementos MAMs, MINOS (mitocondrial inner rmembrane organizing system) y MCSs (mitocondrial contact sites), así como los sitios de contacto entre las membranas mitocondriales para realizar con éxito el lujo de colesterol y su transformación en pregnenolona en los tejidos esteroidogénicos de regulación aguda y en progesterona en el caso de la placenta humana, un tejido de regulación crónica. Sitios de contacto entre las membranas mitocondriales Hackenbrock & Miller (1975), describieron por primera vez que las mitocondrias poseen diferentes tipos de sitios de contacto y que son complejos multienzimáticos que realizan diferentes funciones mitocondriales. Su formación se lleva a cabo mediante un proceso dinámico que requiere de la asociación de proteínas y enzimas especíicas. Para la fosforilación oxidativa se ha descrito la asociación de varias proteínas, entre las cuales están el acarreador de los adenín-nucleótidos (ANT), la porina o el canal aniónico dependiente del voltaje (VDAC), la creatina cinasa y la hexocinasa, entre otros (Adams, Bosch, Schelege, Wallimann & Brdiczka, 1989; Papadopoulos, 1993). En la apoptosis, se asocian proteínas como el ANT, la VDAC, la cicloilina D y la hexocinasa II. Además, la membrana externa posee proteínas que pueden establecer, junto con la actina, uniones dependientes de ATP y proteínas antiapoptóticas de la familia Bcl-2, cuya sobreexpresión tiene como consecuencia el bloqueo de la apoptosis (Adams & Cory, 2001; Suen, Norris & Youle, 2008). Otro ejemplo es el poro de la transición de la permeabilidad (Rasola & Bernardi, 2014), que es un complejo multiproteico que forma poros no selectivos en la membrana interna y que tiene los siguientes componentes estructurales: el ANT, la cicloilina D, y el VDAC. Adicionalmente, se ha identiicado al translocador de proteínas (TPSO, antes conocido como el receptor periférico a benzodiacepinas) unido al poro en la membrana externa, la creatina cinasa en el espacio intermembranal y la hexocinasa II ligadas a la VDAC en la cara citosólica de la membrana externa, así como las proteínas Bax/Bcl-2 (Zamzami & Kroemer, 2001). Finalmente se ha sugerido que para la esteroidogénesis en las glándulas suprarrenales y gónadas existe un punto de contacto especíico, compuesto de un octámero de la creatina cinasa (CK), el VDAC, la TPSO y el ANT, además del IP3R (ER- resident inositol triphosphate receptor), la Mfn2 y la Mfn1 (mitofusina 1 y 2), entre otras (Martin, Kennedy & Karten, 2014).La asociación de las proteínas de la membrana externa como VDAC1 y TSPO con las proteínas de la familia de las ATPasas (ATAD3a) en la membrana interna, forman el núcleo de un complejo que regula la importación del colesterol mitocondrial (Issop et al., 2015). Datos recientes sugieren que la TPSO, aunque forma parte del sitio de contacto, no es indispensable para el transporte del colesterol (Tu et al., 2014; Morohaku et al., 2014). Adicionalmente, se ha descrito que la asociación entre las proteínas Tom22, Tim23, Tim 50, y la 3βHSD2 puede repercutir en la esteroidogénesis (Pawlak, Prasad, Thomas, Whittal & Bose, 2011; Rajapaksha et al., 2016). Nuestro grupo de trabajo sugirió un modelo para la participación de la STARD3 en las mitocondrias de la placenta humana (Esparza-Perusquía et al., 2015) donde se propone que las gotas lipídicas contienen proteínas constituyentes de los complejos de fusión de membranas denominados SNARE (soluble N-ethylmaleimide-sensitive factor attachmentprotein receptor), entre las que incluyen las α-SNAP, syntaxina 3, 7 y 12, syntaxina-Binding Protein-2, syntaxina-Binding Protein-3, y VAMP-8, que podrían estar participando en el transporte del colesterol en las mitocondrias de la placenta humana. Se ha demostrado que la proteína SNAP promueve la interacción entre las gotas de lípidos y las mitocondrias, y que en las células esteroidogénicas las TIP Rev.Esp.Cienc.Quím.Biol.6 Vol. 22 DOI: 10.22201/fesz.23958723e. SNARE se expresan como Syntaxina-17, SNAP-23 y SNAP- 25 (Grant et al., 1999; Steegmaier,Oorschot, Klumperman & Scheller, 2000; Jägerström et al., 2009). Estas observaciones sugieren que las proteínas SNARE podrían contribuir en el mecanismo de transporte del colesterol a las mitocondrias esteroidogénicas, muy probablemente mediante la promoción de la interacción funcional entre las gotas de lípidos, el retículo endoplásmico, los endosomas y las mitocondrias. En las células del sinciciotrofoblasto, la proteína STARD3 puede ser incorporada en las mitocondrias a través de estos complejos y ser transformadaa su forma activa por proteasas mitocondriales (Esparza-Perusquía et al., 2015), favoreciendo su incorporación o participación en los sitios de contacto (Uribe, Strauss & Martínez, 2003; Miller & Bose 2011) y promover el transporte del colesterol para la síntesis de la progesterona (Strauss, Kishida, Christenson, Fujimoto & Hiroi ,2003). Agradecimientos Parte de este trabajo fue apoyado por el programa PAPIIT de la UNAM con los proyectos IN211912, IN211715, IN215518 y IN222617 y del proyecto 168025 de CONACYT. Referencias Adams, J.M. & Cory, S. (2001). Life-or-death decisions by the Bcl-2 protein family. Trends Biochem. Sci., 26, 61-66. DOI: 10.1016/S0968-0004(00)01740-0. Adams, V., Bosch, W., Schelege. J., Wallimann, T. & Brdiczka, D. (1989). Further characterization of contact sites from mitochondria of diferent tissues: topology of peripheral kinases. Biochim. Biophys. Acta, 981(2), 213-225. DOI: 10.1016/0005-2736(89)90031-X. Alpy, F. & Tomasetto, C. (2014). START ships lipids across interorganelle space. Biochimie, 96, 85-95. DOI: 10.1016/j.biochi.2013.09.015. Alpy, F., Rousseau, A., Schwab, Y., Legueux, F., Stoll, I., Wendling, C., Spiegelhalter, C., Kessler, P., Mathelin, C., Rio, M.C., Levine, T.P. & Tomasetto, C. (2013). STARD3 or STARD3NL and VAP form a novel molecular tether between late endosomes and the ER. J. Cell Sci., 126, 5500–5512. DOI: 10.1242/jcs.139295. Ashary, N., Tiwari, A. & Modi, D. (2018). Embryo Implantation: Warin times of love. Endocrinology,159, 1188–1198. DOI: 10.1210/en.2017-03082. Baardman, M.E., Kerstjens-Frederikse, W.S., Berger, R.M.F., Bakker, M.K., Hofstra, R.M.W. & Plösch, T. (2013).The role of maternal-fetal cholesterol transport in early fetal life: current insights. Biol. Reprod., 88(24), 1-9. DOI: 10.1095/biolreprod.112.102442. Bartels, A. & O’Donoghue, K. (2011). Cholesterol in pregnancy: a review of knowns and unknowns. Obstet. Med., 4, 147–151. DOI: 10.1258/om.2011.110003. Beller, M., Thiel, K., Thul, P.J. & Jäckle, H. (2010). Lipid droplets: A dynamic organelle moves into focus. FEBS Lett, 584, 2176-82. DOI: 10.1016/j. febslet.2010.03.022. Bhattacharjee, J., Ietta, F., Romagnoli, R., Bechi, N., Caniggia, I. & Paulesu, L. (2012). ABC transporters in human placenta and their role in maternal-fetal cholesterol transfer: ABCA1 candidate target. En Jing Zheng (Ed.). Recent Advances in Research on the Human Placenta (pp. 336-354).London: IntechOpen. ISBN 978-953-51- 0194-9 DOI: 10.5772/1211. Bianco-Miotto, T., Blundell, C., Buckberry, S., Chamley, L., Chong, S., Cottrell, E., Dawson, P., Hanna, C., Holland, O., Lewis, R.M., Moritz, K., Myatt, L., Perkins, A.V., Powell, T., Safery, R., Sferruzzi-Perri, A., Sibley, C., Simmons, D. & O’Tierney-Ginn, P.F. (2016). IFPA meeting 2015 workshop report I: Placental mitochondrial function, transport systems and epigenetics. Placenta, 48, Suppl. 1, Trophoblast Res., 30, S3-S6. DOI: 10.1016/j. placenta.2015.11.014. Bose, M., Whittal, R.M., Miller, W.L. & Bose, H.S. (2008). Steroidogenic activity of StAR requires contact with mitochondrial VDAC1 and phosphate carrier protein. J. Biol. Chem.,283, 8837–8845. DOI: 10.1074/jbc. M709221200. Burton, G.J. & Fowden, A.L. (2015). The placenta: a multifaceted, transient organ. Phil. Trans. R. Soc. B, 370: 20140066. DOI: 10.1098/rstb.2014.0066. Byrns, M.C. (2014). Regulation of progesterone signaling during pregnancy: Implications for the use of progestins for the prevention of preterm birth. J. Steroid Biochem.& Mol. Biol.,139, 173– 181. DOI: 10.1016/j. jsbmb.2013.01.015. Chatuphonprasert, W., Jarukamjorn, K. & Ellinger, I. (2018). Physiology and pathophysiology of steroid biosynthesis, transport and metabolism in the human placenta. Front. Pharmacol., 9, 1027. DOI: 10.3389/fphar.2018.01027. Chien, Y., Rosal, K. & Chung, B. (2017). Function of CYP11A1 in the mitochondria. Mol. Cell Endocrinol., 441, 55-61. DOI: 10.1016/j.mce.2016.10.030. Esparza-Perusquía, M., Olvera-Sánchez, S., Flores-Herrera, O., Flores-Herrera, H., Guevara-Flores, A., Pardo, J.P., Espinosa-García, M.T. & Martinez, F. (2015). Mitochondrial proteases act on STARD3 to activate progesterone synthesis in human syncytiotrophoblast. Biochim. Biophys. Acta, 1850(1), 107-117. DOI: 10.1016/j.bbagen.2014.10.009. Fan, J. & Papadopoulos, V. (2013). Evolutionary origin of the mitochondrial cholesterol transport machinery reveals a universal mechanism of steroid hormone biosynthesis in animals. PLoS ONE, 8, e76701. DOI:10.1371/journal. pone.0076701. Flis, V.V. & Daum, G. (2013). Lipid transport between the endoplasmic reticulum and mitochondria. Cold Spring Harb Perspect Biol., 5(6), pii: a013235. DOI: 10.1101/ cshperspect.a013235. Olvera Sánchez, S. et al.: Transporte de colesterol en la esteroidogénesis de la placenta 72019 DOI: 10.22201/fesz.23958723e. Furuhashi, M., Seo, H., Mizutani, S., Narita, O., Tomoda, Y. &, Matsui, N. (1989). Expression of low density lipoprotein receptor gene in human placenta during pregnancy. Mol. Endocrinol., 3, 1252-6. DOI: 10.1210/mend-3-8-1252. Galluzzi, L., Kepp, O. & Kroemer, G. (2012). Mitochondria: master regulators of danger signalling. Nat. Rev. Moll. Cell Biol., 13, 780-788. DOI: 10.1038/nrm3479. * Grant, N.J., Hepp, R., Krause, W., Aunis, D., Oehme, P. & Langley, K. (1999). Diferential expression of SNAP- 25 isoforms and SNAP-23 in the adrenal gland. J. Neurochem., 72(1), 363-372. DOI: 10.1046/j.1471- 4159.1999.0720363.x. Hackenbrock, C.R. & Miller, K.J. (1975). The distribution of anionic sites on the surfaces of mitochondrial membranes. Visual probing with polycationic ferritin. J. Cell. Biol., 65(3), 615-630. DOI: 10.1083/jcb.65.3.615. Halasz, M. & Szekeres-Bartho, J. (2013). The role of progesterone in implantation and trophoblast invasion. J.Reprod. Immunol., 97, 43– 50. DOI: 10.1016/j. jri.2012.10.011. Holland, O., Nitert, M.D., Gallo, L.A., Vejzovic, M., Fisher, J.J. & Perkins A.V. (2017a). Placental mitochondrial function and structure in gestational Disorders.Placenta, 54, 2-9.DOI: 10.1016/j.placenta.2016.12.012. Holland, O.J., Hickey, A.J.J., Alvsaker, A., Moran, S., Hedges, C., Chamley, L.W. & Perkins, A.V. (2017b). Changes in mitochondrial respiration in the human placenta over gestation. Placenta, 57, 102-112. DOI: 10.1016/j. placenta.2017.06.011. Horvath, S.E. & Daum, G. (2013). Lipids of mitochondria. Prog. Lipid Res., 52, 590-614. DOI: 10.1016/j. plipres.2013.07.002. Issop, L., Fan, J., Lee, S., Rone, M.B., Basu, K., Mui, J. & Papadopoulos, V. (2015).Mitochondria-associated membrane formation in hormone-stimulated Leydig cell steroidogenesis: Role of ATAD3. Endocrinology, 1: 334- 345. DOI: 10.1210/en.2014-1503. Jägerström, M.S., Polesie, S., Wickström, Y., Johansson, B.R., Schroder, H.D., Højlund, K. & Boström, P. (2009). Lipid droplets interact with mitochondria using SNAP23. Cell Biol Int., 33(9), 934-940. DOI: 10.1016/j. cellbi.2009.06.011. Kaiser, J. (2014). Gearing up for a closer look at the human placenta. Science, 344(6188), 1073. DOI: 10.1126/ science.344.6188.1073. LaVoie, H.A. & King, S.R. (2009). Transcriptional regulation of steroidogenic genes: STARD1, CYP11A1 and HSD3B. Exp. Biol. Med. (Maywood), 234, 880-907. DOI: 10.3181/0903-MR-97. Marriott, K.S., Prasad, M., Thapliyal, V. & Bose, H.S. (2012). σ-1 Receptor at the mitochondrial-associated endoplasmic reticulum membrane is responsible for mitochondrial metabolic regulation. J. Pharmacol. Exp.Ther., 343, 578- 586. DOI: 10.1124/jpet.112.198168. Martin, L.A., Kennedy, B.E. & Karten, B. (2014). Mitochondrial cholesterol: mechanisms of import and efects on mitochondrial function. J. Bioenerg. Biomembr., 48(2), 137-151. DOI: 10.1007/s10863-014-9592-6. Martínez, F. & Strauss, J.F. 3rd (1997). Regulation of mitochondrial cholesterol metabolism.Subcell. Biochem., 28, 205-234. PMID: 9090296. Mesmin, B., Antonny, B. & Drin, G. (2013). Insights into the mechanisms of sterol transport between organelles. Cell. Mol. Life Sci., 70, 3405–3421.DOI: 10.1007/s00018-012- 1247-3. Miller, W.L. & Bose, H.S. (2011). Early steps in steroidogenesis: intracellular cholesterol traicking. J. Lipid Res., 52, 2111-2135. DOI: 10.1194/jlr.R016675. Miller, W.L. (2013).Steroid hormone synthesis in mitochondria. Mol. Cell Endocrinol.,379, 62-73. DOI. 10.1016/j.mce.2013.04.014. Monreal-Flores, J., Espinosa-García, M.T., García-Regalado, A., Arechavaleta-Velasco, F. & Martínez, F. (2017). The heat shock protein 60 promotes progesterone synthesis in mitochondria of JEG-3 cells. Reprod. Biol., 17, 154-161. DOI: 10.1016/j.repbio.2017.04.001. Morel, Y., Roucher, F., Plotton, I., Goursaud, C., Tardy, V. & Mallet, D. (2016). Evolution of steroids during pregnancy: Maternal, placental and fetal synthesis. Ann. Endocrinol. (Paris), 77(2), 82-89. DOI: 10.1016/j.ando.2016.04.023. 2016. Morohaku, K., Pelton, S.H., Daugherty, D.J., Butler, W.R., Deng, W. & Selvaraj, V. (2014). Translocator protein/ peripheral benzodiazepine receptor is not required for steroid hormone biosynthesis. Endocrinology, 155(1), 89–97. DOI: 10.1210/en.2013-1556. Navarrete, J., Flores-Herrera, O., Uribe, A. & Martínez, F. (1999). Diferences in cholesterol incorporation into mitochondria from hepatoma AS-30D and human term placenta. Placenta, 20(4), 285-291. DOI: 10.1053/ plac.1998.0374. Olvera-Sánchez, S., Espinosa-García, M.T., Monreal, J., Flores-Herrera, O. & Martinez, F. (2011). Mitochondrial heat shock protein participates in placental steroidogenesis. Placenta, 32, 222-229. DOI: 10.1016/j. placenta.2010.12.018. Pagler, T.A., Golsabahi, S., Doringer, M., Rhode, S., Schutz, G.J., Pavelka, M., Wadsack, C., Gauster, M., Lohninger, A., Laggner, H., Strobl, W. & Stangl, H. (2006). A Chinese hamster ovarian cell line imports cholesterol by high density lipoprotein degradation. J. Biol. Chem., 281, 38159-38171. DOI: 10.1074/jbc.M603334200 Papadopoulos, V. & Miller, W.L. (2012). Role of mitochondria in steroidogenesis. Best Pract. Res. Clin. Endocrinol.Metab., 26(6), 771-790. DOI: 10.1016/j. beem.2012.05.002. Papadopoulos, V. (1993). Peripheral-type benzodiazepine/ diazepam binding inhibitor receptor: biological role in TIP Rev.Esp.Cienc.Quím.Biol.8 Vol. 22 DOI: 10.22201/fesz.23958723e. steroidogenic cell function. Endocr.Rev., 14, 222-240. DOI: 10.1210/edrv-14-2-222. Paul, S., Gupta, P.D., Jailkhani, B.L. & Talwar, G.P. (1980). Resistance of human syncytiotrophoblast to hypotonic and thermal stress. J. Reprod.Fertil., 58(1), 183-187. PMID: 7359476. Paul, S., Jailkhani, B.L. & Talwar, G.P. (1978). Isolation and functional maintenance in culture of syncytiotrophoblasts from human placenta. Indian J. Exp. Biol., 16, 1226- 1235. PMID: 750412. Pawlak, K.J., Prasad, M., Thomas, J.L., Whittal, R.M. & Bose, H.S. (2011). Inner mitochondrial translocase Tim50 interacts with 3β-hydroxysteroid dehydrogenase type 2 to regulate adrenal and gonadal steroidogenesis. J. Biol.Chem., 286(45), 39130-39140. DOI: 10.1074/jbc. M111.290031. Poderoso, C., Duarte, A., Cooke, M., Orlando, U., Gottifredi, V., Solano, A.R. Lemos, J.R. & Podestá, E.J. (2013). The spatial and temporal regulation of the hormonal signal. Role of mitochondria in the formation of a protein complex required for the activation of cholesterol transport and steroids synthesis. Mol. Cell. Endocrinol.,371, 26–33. DOI: 10.1016/j.mce.2012.12.024. Prasad, M., Kaur, J., Pawlak, K.J., Bose, M., Whittal, R.M. & Bose, H.S. (2015).Mitochondria associated ER- Membrane (MAM) regulates steroidogenic activity via StAR-VDAC2 interaction. J. Biol. Chem.,290, 2604- 2616. DOI: 10.1074/jbc.M114.605808. Rajapaksha, M., Kaur, J., Bose, M., Whittal, R.M. & Bose, H.S. (2013). Cholesterol-mediated conformational changes in the steroidogenic acute regulatory protein are essential for steroidogenesis. Biochemistry,52, 7242−7253. DOI: 10.1021/bi401125v. Rajapaksha, M., Kaur, J., Prasad, M., Pawlak, K.J., Marshall, B., Perry, E.W., Whittal, R.M. & Bose, H.S. (2016). An outer mitochondrial translocase, Tom22, is crucial for inner mitochondrial steroidogenic regulation in adrenal and gonadal tissues. Mol. Cell Biol., 36(6), 1032-1047. DOI: 10.1128/MCB.01107-15; Rasola, A. & Bernardi, P. (2014). The mitochondrial permeability transition pore and its adaptive responses in tumor cells. Cell Calcium, 56(6), 437-445. DOI: 10.1016/j.ceca.2014.10.003. Sanderson, J.T. (2009). Placental and Fetal Steroidogenesis Human. En: Lafond, J. & Vaillancourt, C. (Ed.). Embryogenesis: Methods and Protocols. (pp. 127-136) Nueva York:HumanaPress.DOI 10.1007/978-1-60327- 009-0 7. Simpson, E.R. & MacDonald, P.C. (1981). Endocrine physiology of the placenta. Annu. Rev. Physiol., 43, 163- 188. DOI: 10.1146/annurev.ph.43.030181.001115 Soientini, U. & Graham, A. (2016). Intracellular cholesterol transport proteins: roles in health and disease. Clin. Sci., 130, 1843–1859 DOI: 10.1042/CS20160339 Steegmaier, M., Oorschot, V., Klumperman, J. & Scheller, R.H. (2000). Syntaxin 17 is abundant in steroidogenic cells and implicated in smooth endoplasmic reticulum membrane dynamics. Mol. Biol. Cell, 11(8), 2719-2731. DOI: 10.1091/mbc.11.8.2719. Strauss, J.F.3rd, Kishida, T., Christenson, L.K., Fujimoto, T. & Hiroi, H. (2003). StAR domain proteins and the intracellular traicking of cholesterol in steroidogenic cells. Mol. Cell, Endocrinol., 202, 59-65. DOI: 10.1016/ S0303-7207(03)00063-7. Suen, D.F., Norris, K.L. & Youle, R.J. (2008). Mitochondrial dynamics and apoptosis. Genes Dev., 22, 1577-1590. DOI: 10.1101/gad.1658508. Tait, S.W. & Green, D.R. (2012). Mitochondria and cell signalling. J. Cell Sci., 125, 807-815. DOI: 10.1242/ jcs.099234. Tu, L.N., Morohaku, K., Manna, P.R., Pelton, S.H., Butler, W.R., Stocco, D.M. & Selvaraj, V. (2014). Peripheral benzodiazepine receptor/translocator protein global knock-out mice are viable with no efects on steroid hormone biosynthesis. J. Biol. Chem., 289, 40, 27444– 27454. DOI: 10.1074/jbc.M114.578286. Tuckey, R.C., Headlam, M.J, Bose, H.S. & Miller, W.L. (2002). Transfer of cholesterol between phospholipid vesicles mediated by the steroidogenic acute regulatory protein (StAR). J. Biol. Chem., 277, 47123-28. DOI: 10.1074/jbc.M206965200. Tuckey, R.C., Bose, H.S., Czerwionka, I. & Miller, W.L. (2004). Molten globule structure and steroidogenic activity of N-218 MLN64 in human placental mitochondria. Endocrinology, 145, 1700-1707. DOI: 10.1210/en.2003- 1034. Uribe, A. Strauss, J.F. 3rd & Martínez, F. (2003). Contact sites from human placental mitochondria: characterization and role in progesterone synthesis. Arch. Biochem. Biophys., 413(2), 172-181. DOI: 10.1016/S0003-9861(03)00097-3. van der Kant, R., Zondervan, I., Janssen, L. & Neefjes, J. (2013). Cholesterol-binding molecules MLN64 and ORP1L mark distinct late endosomes with transporters ABCA3 and NPC1. J. Lipid Res., 54, 2153–2165. DOI: 10.1194/jlr.M037325 van Leusden, H. & Villee, C.A. (1965).The de novo synthesis of sterols and steroids -from acetate by preparations of human term placenta. Steroids, 6, 31-45. DOI: 10.1016/0039-128X(65)90031-0. Vaughan, O.R. & Fowden, A.L. (2016). Placental metabolism: substrate requirements and the response to stress. Reprod. Dom. Anim., 51 (Suppl. 2), 25-35.DOI: 10.1111/ rda.12797. Wadsack, C., Hammer, A., Levak-Frank, S., Desoye, G., Kozarsky, K.F., Hirschmugl, B., Sattler, W. & Malle, E. (2003). Selective cholesteryl ester uptake from high density lipoprotein by human irst trimester and term Olvera Sánchez, S. et al.: Transporte de colesterol en la esteroidogénesis de la placenta 92019 DOI: 10.22201/fesz.23958723e. villous trophoblast cells. Placenta, 24, 31-43. DOI: 10.1053/plac.2002.0912. Watari, H., Arakane, F., Moog-Lutz, C., Kallen, C.B., Tomasetto, C., Gerton, G.L., Rio, M.C., Baker, M.E. & Strauss, J.F. 3rd. (1997). MLN64 contains a domain with homology to the steroidogenic acute regulatory protein (StAR) that stimulates steroidogenesis. Proc. Natl. Acad. Sci. USA, 94(16), 8462-8467doi.org/10.1073/ pnas.94.16.8462 Winkel, C.A., MacDonald, P.C. & Simpson, E.R. (1981). The role of receptor-mediated low-density lipoprotein uptake and degradation in the regulation of progesterone biosynthesis and cholesterol metabolism by human trophoblasts. Placenta, (Suppl. 3), 133-143. PMID: 6306642. Winkel, C.A., Snyder, J.M., MacDonald, P.C. & Simpson, E.R. (1980). Regulation of cholesterol and progesterone synthesis in human placental cells in culture by serum lipoproteins. Endocrinology, 106, 1054-1060. DOI: 10.1210/endo-106-4-1054. Woollett, L.A. (2011). Transport of maternal cholesterol to the fetal circulation. Placenta, 32 (Suppl. 2), S218-21. DOI: 10.1016/j.placenta.2011.01.011. Yang, H., Galea, A., Sytnyk, V. & Crossley, M. (2012). Controlling the size of lipid droplets: lipid and protein factors. Curr. Opin. Cell Biol., 24, 509-516. DOI: 10.1016/j.ceb.2012.05.012. Zelewski, L. & Villee, C.A. (1966). The biosynthesis of squalene, lanosterol, and cholesterol by minced human placenta.Biochemistry,5,1805-1814. PMID: 5963423. Zamzami, N. & Kroemer, G. (2001). The mitochondrion in apoptosis: how Pandora’s Box opens. Nat. Rev. Mol. Cell Biol., 2(1), 67-71.DOI: 10.1038/35048073 Viruses 2020, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/viruses Article 1 Steady-state persistence of respiratory syncytial virus 2 in a macrophage-like cell line and identification of 3 non-synonymous mutations through sequencing of 4 the persistent viral genome 5 Ximena Ruiz-Gómez1, Joel Armando Vázquez-Pérez2, Oscar Flores-Herrera3, Mercedes 6 Esparza-Perusquía3, Carlos Santiago-Olivares1, Jorge Gaona5, Beatriz Gómez1, Carmen Méndez4, 7 Evelyn Rivera-Toledo1,* 8 1 Departamento de Microbiología y Parasitología, Facultad de Medicina, Universidad Nacional Autónoma 9 de México, Ciudad Universitaria, Coyoacán 04510, Mexico City, Mexico; menaxrg@hotmail.com (XRG); 10 carlosantiagolivares@yahoo.com.mx (CASO); bgomez2017@gmail.com (BG); evelyn.rivera@unam.mx 11 (ERT) 12 2 Instituto Nacional de Enfermedades Respiratorias Ismael Cosío Villegas, Mexico City, Mexico; 13 joevazpe@gmail.com 14 3 Departamento de Bioquímica, Facultad de Medicina, Universidad Nacional Autónoma de México, Ciudad 15 Universitaria, Coyoacán 04510, Mexico City, Mexico; oflores@bq.unam.mx (OFH); mesparza@bq.unam.mx 16 (MEP) 17 4 Departamento de Embriología y Genética, Facultad de Medicina, Universidad Nacional Autónoma de 18 México, Ciudad Universitaria, Coyoacán 04510, Mexico City, Mexico; mendezmc@unam.mx 19 5 Departamento de Microbiología y Patología, Centro Universitario de Ciencias de la Salud, Universidad de 20 Guadalajara, Guadalajara, Jalisco, Mexico; jgaber2007@gmail.com 21 22 23 * Correspondence: evelyn.rivera@unam.mx (ERT); Tel.: +52 5556232467 24 Received: date; Accepted: date; Published: date 25 Abstract: Persistent viral infection in cell cultures have been categorized at least as in 26 “carrier-state”, where there exists a low proportion of cells infected by a lytic virus, and as in 27 “steady-state”, where most of cells are infected, but in absence of cytophatic effect. Here, we 28 showed that human respiratory syncytial virus (hRSV) maintained a steady-state persistence in 29 the macrophage-like cell line P388D1 after 120 passages, since the viral genome was detected in all 30 of the cells analyzed by fluorescence in situ hybridization, whereas only defective viruses were 31 identified by sucrose gradients and titration assay. Interestingly, up to 8.0±7.0% of the cells 32 harboring the hRSV genome revealed undetectable expression of the viral nucleoprotein N; 33 however, when this cell population was sorted by flow cytometry and independently cultured, 34 hRSV protein expression was induced at detectable levels since the first post-sorting passage, 35 supporting that sorted cells harbored the viral genome. Finally, sequencing of the persistent hRSV 36 genome obtained from virus particles collected from cell-culture supernatants, allowed 37 assembling of a complete genome sequence that displayed 38 nonsynonymous mutations 38 distributed in ten of the eleven viral proteins. Eight of the 38 nonsynonymous mutations have 39 been previously characterized and they were related to alterations in viral assembly or budding, 40 membrane fusion and evasion of the antiviral response. 41 Keywords: Respiratory syncytial virus; viral persistence; steady-state persistence; defective 42 viruses; viral genome sequencing. 43 44 Viruses 2020, 12, x FOR PEER REVIEW 2 of 21 1. Introduction 45 Persistent infection is a relevant strategy for many viruses to be long-term maintained within a host 46 population, since silent viral transmission may be allowed from healthy carriers, although viral 47 persistence could also lead to life-threatening chronic diseases. Human respiratory syncytial virus 48 (hRSV) is an enveloped Orthopneumovirus from the Pneumoviridae family, with genome encoded in a 49 non-segmented negative strand RNA with approximately 15,200 nucleotides of size and tropism for 50 epithelial cells and macrophages of the respiratory tract [1–3]. The hRSV genome consists of ten 51 genes, NS1-NS2-N-P-M-SH-G-F-M2-L, encoding eleven proteins, since the M2 gene has two 52 overlapping open reading frames expressing M2-1 and M2-2 [4]. Eight RSV proteins are structural, 53 whereas the NS1, NS2 and M2-2 are non-structural proteins with relevant roles in immune evasion 54 and regulation of genome transcription/replication, respectively [5–7]. 55 Globally, hRSV is the main etiological agent of severe acute lower respiratory infection (ALRI) in 56 children under five years and it was estimated in 2015 a total of 33.1 million of RSV-ALRI, with 3.2 57 million of hospital admissions and 59,600 in-hospital deaths; of relevance, forty-five percent of 58 hospitalizations occurred in infants younger than six months [8]. RSV-ALRI early in life has been 59 associated to respiratory diseases such as wheezing and asthma-like symptoms in later childhood 60 [9–13], which is highly prevented by Palivizumab prophylaxis in children without a family history 61 of atopy, suggesting that RSV predisposes to long-term airway hyperreactivity in an 62 atopy-independent mechanism [13,14]. It has been proposed that hRSV persistence in the 63 respiratory tract may be at least one mechanism associated to recurrent asthma-like symptoms [15]. 64 Experimental models in rats, mice and guinea pigs have shown that hRSV persists in vivo, 65 according to long-standing detection of viral antigens, recovery of viral RNA and isolation of 66 infectious virus in pulmonary tissue [16–19]. Evidence of hRSV persistence in humans is limited, 67 but some studies have reported viral RNA in respiratory and non-respiratory tissues of 68 asymptomatic children and adults [20–23], as well as in respiratory secretions of 69 immunocompromised patients [24]. Cell lines as in vitro models are useful tools to evaluate 70 mechanisms and potential target cells for virus persistence; in such a way, epithelial cells, dendritic 71 cells and macrophages have been persistently infected by hRSV [25–27]. 72 Persistent viral infection in cell cultures has been classified in at least two types: carrier-state and 73 steady-state. In the former, only a small proportion of cells are productively infected, functioning as 74 a permanent source of virus that is delivered to the extracellular medium to be transmitted 75 horizontally to the constantly growing uninfected cell population. In steady-state persistence, most 76 cells are infected and both, virus and cell multiplication, are maintained without cell destruction 77 [28,29]. 78 We have studied persistence of hRSV in the macrophage-like cell line P388D1 for over 120 passages 79 and our observations indicate hRSV alters cell survival, IFN-I response and nitric oxide production 80 to persist [30–32]. Most of our persistently hRSV-infected macrophage-like cultures are constituted 81 by >90% of cells expressing viral proteins [31,32]; however, we have also identified some cultures 82 with ≤70% of hRSV positive cells after storage in liquid nitrogen, suggesting it could be associated 83 to a carrier-state. Herein, we report that hRSV established no a carrier-state, but a steady-state 84 persistence in the P388D1 cell line, since viral genomic RNA was detected in all of the cells analyzed 85 by fluorescence in situ hybridization, whereas only defective viruses were identified through viral 86 titration and sucrose gradients. Additionally, genomic RNA was isolated from virus particles 87 Viruses 2020, 12, x FOR PEER REVIEW 3 of 21 collected from cell-culture supernatants and sequenced to determine changes experienced in the 88 hRSV genome after several passages. 89 2. Materials and Methods 90 2.1. Cell lines and virus 91 Persistently RSV infected macrophage cultures (MΦP) were obtained and characterized in our 92 laboratory from surviving cells after acute infection of the mouse macrophage-like cell line P388D1 93 at multiplicity of infection of 1 (m.o.i. of 1), with the wild type hRSV Long strain (VR-26, ATCC), as 94 previously described [25]. Viral persistence was constantly monitored (every two or three weeks) 95 through expression of three structural proteins by direct immunofluorescence (section 2.2). MΦP 96 cultures showed non-cytophatic effects and have been subcultured or passaged for over 120 times 97 in RPMI-1640 culture medium (Gibco), supplemented with 5% fetal bovine serum (Biowest), 1% 98 penicillin-streptomycin (Invitro) and 1 µM 2-mercaptoethanol (Sigma). This work was performed 99 with passages 120–150. The original P388D1 cell line was used as a non-infected control (MΦN) and 100 was maintained under similar culture conditions as MΦP. The human epithelial cell line HEp-2 and 101 the monkey kidney epithelial cell line Vero E6 were cultured in Dulbecco’s Modified Eagle Medium 102 (DMEM; Gibco), supplemented with 5% fetal bovine serum, 10 nM HEPES (Sigma) and 1% 103 penicillin-streptomycin. HEp-2 cells were acutely infected at m.o.i. of 0.5 with the RSV Long strain 104 for 72 h to compare buoyant density of viruses produced under this condition and buoyant density 105 of viruses in MΦP cultures (see section 2.5). 106 Infectious hRSV from MΦP cultures were quantified in HEp-2 and Vero E6 cells through a limiting 107 dilution method, with viral titers expressed as the 50% tissues culture infectious dose per ml 108 (TCID50/ml). 109 110 2.2. Direct immunofluorescence 111 Percentage of persistently RSV infected cells in MΦP cultures was evaluated by direct 112 immunofluorescence. MΦP were fixed with 4% p-formaldehyde on ice and permeabilized with 113 0.3% saponine at room temperature (15 min each step). After blocking non-specific binding sites (30 114 min) with blocking solution (5% neonate bovine serum (Bioexport) in PBS), a mix of FITC-labeled 115 monoclonal antibodies against three hRSV structural proteins was added by using the IMAGEN 116 Respiratory syncytial virus kit (anti-hRSV-FITC; Oxoid), diluted 1:10 in blocking solution. MΦN 117 were stained under the same protocol to confirm binding specificity of the FITC-labeled anti-hRSV 118 antibody (internal negative control). Percentage of FITC+ cells (hRSV-protein+ cells) was determined 119 by flow cytometry in a FACS Calibur (BD Biosciences). Unstained MΦP or MΦN were used as 120 negative controls, to determine background fluorescence. 121 122 2.3. Cell sorting 123 FITC– and FITC+ cells were isolated from MΦP cultures with ≤70% of hRSV-protein+ cells (passages 124 120, 123 and 126) in a FACS Aria II cell sorter (BD Biosciences), after staining with the 125 anti-hRSV-FITC, although cells were not permeabilized. Each subset was defined in a fluorescence 126 histogram and extreme opposed regions, corresponding to FITC– and FITC+ populations were 127 selected for sorting. Recovered cells were independently cultured in supplemented RPMI during 128 four to five days before the first, post-sorting passage. Immunofluorescence to determine 129 Viruses 2020, 12, x FOR PEER REVIEW 4 of 21 percentage of hRSV-protein+ cells in consecutive post-sorting passages was performed in fixed and 130 permeabilized cells (as described in section 2.2). 131 132 2.4. Fluorescence in situ hybridization and indirect immunofluorescence 133 Fluorescence in situ hybridization (FISH) for hRSV genomic RNA (hRSV-gRNA) in MΦP cultures 134 was performed with a pool of 29 single stranded DNA probes (20 nucleotides), detecting targeted 135 positions from 1140 to 2285, encompassing the nucleoprotein N gene (GenBank: AY911262.1). Each 136 probe was labeled with the red fluorophore Quasar 570 (Biosearch Technologies). MΦP and MΦN 137 were plated onto coverslips (2.5×105) in 48-well plates and incubated overnight to allow adherence. 138 Cells were washed with PBS and fixed with 4% p-formaldehyde (10 min) and then permeabilized in 139 0.3% saponine (15 min). Permeabilized cells were treated with wash buffer containing 10% 140 formamide and 2× saline sodium citrate (2× SSC; 0.3 M NaCl and 0.03 M sodium citrate, pH 7.4). 141 Heat denaturalized DNA from salmon testes (Sigma), was added at 0.5 mg/ml in hybridization 142 buffer (wash buffer plus 5% Ficoll 400) for 2 h at 37°C. After washing twice, specific DNA probes 143 were added at 4 nM in hybridization buffer overnight at 37°C under constant agitation. Two 144 stringency washes were performed to eliminate unbound probes (30 min each). Indirect 145 immunofluorescence was performed immediately with the anti-N primary antibody (sc-58001; 146 Santa Cruz Biotechnology), dilution 1:100 in 3% neonate bovine serum in PBS (30 min), followed by 147 a FITC-labeled secondary antibody 1:100 (sc-2010; Santa Cruz Biotechnology). Nuclear staining was 148 performed with 0.8 µg/ml Hoechst 33342 (Invitrogen). Cover slides were mounted with 6 µl of 149 mounting fluid (Oxoid). Fluorescence imaging acquisition was performed in a microscope Leica Las 150 X (Leica). Auto-fluorescence of unstained cells was a reference to set the threshold for positive 151 staining of probes and antibodies. 152 Percent of hRSV-protein+ cells by confocal microscopy was calculated by counting cells expressing 153 the N protein in at least three different microscopic fields, from three independent experiments and 154 compared to percentages obtained from immunofluorescence analyzed through flow cytometry. 155 156 2.5. Sucrose gradients and buoyant density 157 Virus buoyant density was determined by analysis of viral pellets through sucrose gradients. 158 Supernatants from MΦP cultures or acutely infected HEp-2 cells (72 h post infection) were collected 159 and centrifuged at 59,000×g, 4°C (2.5 h) in a SW45 Ti rotor. Pellets were resuspended in 10 mM 160 Tris-HCl pH 7.5, loaded in a lineal sucrose gradient (20–60%) and centrifuged at 59,000×g, 4°C (2.5 161 h) in a SW28 rotor. Gradients were fractionated from bottom to top in 0.5 ml and optical density at 162 280 nm (O.D. 280 nm) was determined by spectrophotometry, whereas refraction units were 163 determined with a hand-held refractometer (Atago). Buoyant density (ρ) was calculated as 164 ρ20°C=2.7329η - 2.6425, where η=refraction units [25]. 165 166 2.6. Virus genome sequencing 167 Viral genome sequence was obtained from RNA extracted from viral pellets enriched through 168 centrifugation of supernatants collected along five passages from MΦP cultures sorted as FITC– and 169 FITC+ (10–15 post-sorting passages), at 59,000×g, 4°C (2.5 h). Viral pellets were resuspended in 10 170 mM Tris-HCl pH 7.5 (90 µl) and treated with 5 U DNase I (Invitrogen) at 37°C (30 min) and 100 171 µg/ml RNase A (Invitrogen) at 37°C (15 min). After nucleases treatment, RNA was immediately 172 extracted with the QIAmp Viral RNA kit (Qiagen), according to the manufacturer’s instructions. 173 Viruses 2020, 12, x FOR PEER REVIEW 5 of 21 Depleted RNA was directly fragmented by using the Illumina TruSeq Stranded mRNA Sample 174 Preparation Kit (Illumina) and two libraries, (Persistent_hRSV_F– and Persistent_hRSV_F+), were 175 constructed according to the manufaturer’s instructions. 176 Five millions reads per sample were obtained and RSV genomes were assembled de novo using 177 Spades (v.3.12.0.); contigs were mapped against the VR-26 strain of the hRSV (GenBank accession 178 no. AY911262.1). Best alignment was for hRSV with 98.8% of coverage for each sample, with depth 179 of 769× and 552× for Persistent_hRSV_F– and Persistent_hRSV_F+, respectively; genome sequences 180 are available at the National Center for Biotechnology Information (GenBank accession no. 181 MT492012 and MT492011). 182 2.7. Statistical analysis 183 Mean±standard deviation from three independent experiments was determined by using GraphPad 184 Prism software. 185 186 3. Results 187 3.1. Detection of the hRSV genome in MΦP cultures through fluorescence in situ hybridization 188 189 The MΦP cultures have been previously studied to determine alterations in the biology of the host 190 cell, associated to maintenance of hRSV persistent infection [30–32]. Our previous reports were 191 performed in cultures of MΦP with >90% of hRSV-protein+ cells, according to direct 192 immunofluorescence with anti-hRSV-FITC antibodies and analysis by flow cytometry (Figure 1). 193 However, we also observed few MΦP cultures with ≤70% cells expressing viral proteins after 194 storage in liquid nitrogen (Figure 1); this type of cultures were of interest in this work to study 195 whether MΦP cultures were carrier- or steady-state. 196 197 198 199 200 201 Figure 1. Persistently hRSV infected macrophages (MΦP) were analyzed by direct immunofluorescence 202 and flow cytometry to determine percentage of cells expressing viral proteins. Three different MΦP cultures 203 were identified with high a) or intermediate b), c) percentage of cells expressing hRSV proteins after one week 204 of their independent thawing from liquid nitrogen. 205 206 RSV-FITC C o u n t RSV-FITC C o u n t RSV-FITC C o u n t (a) (b) (c) Viruses 2020, 12, x FOR PEER REVIEW 6 of 21 First, we evaluated the amount of cells harboring the hRSV-gRNA through FISH and confocal 207 microscopy by using DNA probes directed against the whole hRSV N gene, along with analysis of 208 N protein expression by indirect immunofluorescence (Figure 2a). This assay was performed with 209 MΦP cultures displaying less than 82% cells expressing viral proteins, according to flow cytometry 210 (Table 1). Interestingly, confocal microscopy revealed a regular cytoplasmic distribution of viral 211 RNA in every single cell from MΦP cultures, whereas the nucleoprotein N was observed with 212 variable expression and forming small or large aggregates (Figure 2a and 2b), in agreement with 213 morphology of inclusion bodies, previously described during hRSV replication [33]. We also 214 observed some cells with cytoplasmic hRSV-gRNA but undetectable expression of N protein 215 (Figure 2a). In merged images, red fluorescence from the DNA probes was deeply intense, whereas 216 green fluorescence from the anti-N antibody was in some cases almost imperceptible, mainly in 217 cells with low levels of N protein. As expected, mock-infected macrophages (MΦN) did not bind 218 either DNA probe against the hRSV-gRNA or the anti-N antibody. 219 Besides, we calculated the percentage of cells expressing the N protein in at least three different 220 microscopic fields from three independent MΦP cultures and such value was compared to that 221 obtained by flow cytometry; bright field was useful to improve visualization of cells with low N 222 protein expression (Figure 2b). Results showed higher percentage of hRSV-protein+ cells analyzed 223 by confocal microscopy than that by flow cytometry (Table 1), although this difference was not 224 statistically significant (p=0.058). It is important to consider that data from table 1 were calculated 225 with 358.1±79.6 cells from each of the three independent MΦP cultures, whereas 10,000 events were 226 considered for analysis by flow cytometry. Thus, we observed hRSV-gRNA in all of the cells in 227 MΦP cultures, as expected in steady-state persistence, despite the fact that viral protein expression 228 was not always detected either by confocal microscopy or flow cytometry. 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 Viruses 2020, 12, x FOR PEER REVIEW 7 of 21 245 246 Figure 2. Intracellular distribution of hRSV genomic RNA (hRSV-gRNA) in MΦP cultures was analyzed 247 by FISH and expression of the N protein determined by indirect immunofluorescence. Non-infected 248 macrophages (MΦN) were used as negative control. (a) Nucleus staining was performed with Hoechst 33342 249 (blue), N-protein was detected with an anti-N antibody followed by a FITC-labeled secondary antibody 250 (green) and the hRSV-gRNA detection was achieved with Quasar 570-labeled DNA probes directed against the 251 N gene (red). White arrows point to a cell with low N protein expression, high level of hRSV-gRNA and 252 imperceptible green signal in merge image. Yellow arrows point to a cell with undetectable N protein 253 expression, even though hRSV-gRNA was distinguished. Images were captured with a 63× objective. Scale 254 bars: 15 µm. (b) Brigh field was used for contrast green fluorescence, in order to distinguish cells expressing 255 low levels of N protein. Black arrows point to cells with few, tiny N protein aggregates; every cell with low or 256 high expression of N protein was counted to calculate percentage of hRSV-protein+ cells by confocal 257 microscopy and this value was compared to percentages obtained through flow cytometry (Table 1). 258 Percentages were calculated from three independent MΦP cultures. 259 260 261 262 3.2. Flow cytometric cell sorting of hRSV-protein positive and hRSV-protein negative cells from MΦP 263 cultures 264 265 Next, we used anti-hRSV-FITC antibodies to sort FITC– and FITC+ cells from MΦP cultures by using 266 flow cytometry; afterwards, cells were independently cultured to evaluate their behavior regarding 267 viral protein expression throughout subsequent subcultures. High purity of FITC– cells was 268 achieved by sorting (96.4±1.2%), whereas purity of FITC+ cells was much lower (61.0±6.8%), 269 M Φ N M Φ P Hoechst Anti-N (FITC) gRNA (Quasar 570) Merge hRSV-protein + cells (%) MΦP culture Flow cytometry Microscopy 1 67.50 95.10 2 79.30 84.00 3 82.10 97.00 Mean±SD 76.3±7.7 92.0±7.0 (a) (b) Table 1. Percentages of hRSV-protein + cells by two methods Viruses 2020, 12, x FOR PEER REVIEW 8 of 21 indicating that they were contaminated with negative cells (Figure 3). After their independent 270 culture during four to five days, the first post-sorting passage was evaluated regarding hRSV 271 protein expression. Under such condition, 53.9±16.4% of hRSV-protein+ cells were observed in the 272 cultures sorted as FITC– (hereafter called as csFITC–), with a progressive increase up to 94.3±2.9% at 273 passage six and then, a slight but constant decrease up to 79.2±8.2% at passage 15 (Figure 3). In 274 cultures sorted as FITC+ (hereafter called as csFITC+) 68.5±10.8% hRSV-protein+ cells were observed 275 in the first post-sorting passage, with gradual augment to 97.2±2.0% at passage six and decline to 276 82.5±12.6% in the last passage. These results suggested that FITC– cells harbored the hRSV-gRNA, 277 whereas protein expression was low-to-undetectable but with potential to be reactivated to higher 278 levels after sorting and independent culture. 279 280 281 282 Figure 3. Sorting and independent culture of cells with positive and negative expression of hRSV 283 proteins. Cells expressing or not hRSV proteins in MΦP cultures were identified by direct immunofluorescence 284 and flow cytometry with anti-hRSV-FITC labeled antibodies. FITC– and FITC+ cells were sorted, cultured 285 independently and percentage of hRSV-protein+ cells was evaluated during 14 consecutive post-sorting 286 passages. Percentage of FITC– and FITC+ cells at time cero was determined according to purity of each cell 287 population recovered. Data represent average±SD of the three independent MΦP cultures. 288 289 290 3.3. Buoyant density and tittering of persistent hRSV from MΦP cultures 291 292 In persistent carrier-cultures, infectious viruses are normally detected in supernatants [34,35]; 293 therefore, we examined the presence of infectious hRSV in supernatants from MΦP cultures. 294 Through titration in HEp-2 cells by limiting dilution (TCID50), we did not detect infectious viruses 295 in at least five different passages (120, 122, 135, 149 and 150), since cytopathic effect was 296 undetectable. Nevertheless, titration in Vero E6 cells showed cytopathic effect sometimes in the first 297 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 10 20 30 40 50 60 70 80 90 100 110 FITC+ at sorting FITC– at sorting Sorting and seeding Passage number after sorting (% ) h R S V -p ro te in + c el ls Viruses 2020, 12, x FOR PEER REVIEW 9 of 21 log10 dilution resulting in viral titers of 255.7±202.6 TCID50/ml, indicating infectious viruses were 298 produced at low levels. Accordingly, we determined virus buoyant density by sucrose gradients. 299 Reported buoyant density for hRSV is 1.17–1.20 g/ml [36] and we calculated average density of 300 1.072±0.009 g/ml for viruses in MΦP cultures (Figure 4), indicating primarily production of 301 low-density, defective virus. As a control, we also determined buoyant density of hRSV produced 302 in acutely infected HEp-2 cells (72 hours post infection), obtaining an average density of 1.195±0.18 303 g/ml, with low frequency of defective viruses (Figure 4); besides, viral titer during acute infection 304 was of 2.40±0.55×106 TCID50/ml. Accordingly, absence of infectious viruses support our previous 305 observations that RSV established a steady-state persistence in macrophage-like cells. 306 307 308 309 310 Figure 4. Buoyant density of viral particles in MΦP cultures and in HEp-2 cells infected at m.o.i. of 0.5 for 311 72 h was determined through sucrose gradients. Supernatants from hRSV-infected cells were fractioned in 312 sucrose gradients and buoyant density of each fraction was calculated; optical density at 280 nm was also 313 evaluated. Buoyant densities are indicated in fractions that show absorbance at 280 nm. Data represent the 314 average±SD of three independent MΦP or acutely infected HEp-2 cells. 315 316 317 3.4. Sequencing of the persistent hRSV genome 318 319 Finally, we sequenced the persistent hRSV genome obtained from viral particles in supernatants of 320 two independent MΦP cultures, csFITC– and csFITC+. Supernatants were collected from both types 321 of MΦP cultures along 10–15 post-sorting passages that showed approximately 80% hRSV-protein+ 322 cells. After RNA was extracted from virus pellets, two sequencing libraries, Persistent_hRSV_F– and 323 Persistent_hRSV_F+, were constructed for samples from csFITC– and csFITC+, respectively. One 324 contig of 15,227 nucleotides per sample was obtained with 98% identity with respect to the 325 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.05 1.06 1.25 1.07 1.23 1.20 MΦP HEp-2 + hRSV (72 h post-infection) 1.10 Fraction number O .D . 2 8 0 n m Viruses 2020, 12, x FOR PEER REVIEW 10 of 21 reference sequence, strain Long VR-26 (GeneBank accession no. AY911262.1), used to establish our 326 persistently hRSV infected culture. 327 A total of 38 nonsynonymous mutations were detected in the persistent hRSV genome (Figure 5), of 328 which, 34.2% occurred in protein F, 18.4% in protein G, 10.5% in protein NS1 and 7.9% or 10.5% in 329 the polymerase L. Results were identical for the genome of virus collected from csFITC– and 330 csFITC+ cultures, except in protein L of virus from the last culture that showed an additional 331 substitution (Q1112R, Table 2). Frequency of nonsynonymous mutations in each viral protein was 332 calculated in regards to total amino acids (protein length), indicating similar mutation frequency for 333 F and G proteins (Figure 5), while protein NS1 showed the higher mutation frequency with respect 334 to all other viral proteins. 335 336 337 338 339 Figure 5. Schematic representation of hRSV genome and number of nonsynonymous mutations identified 340 after sequencing the persistent viral genome of MΦP cultures. Viral RNA was obtained from hRSV recovered 341 by ultracentrifugation of supernatants from two independent MΦP cultures (csFITC– and csFITC+). 342 Independent sequencing was performed and nonsynonymous mutations were identified. Numbers on the top 343 of the diagram correspond to quantity of nonsynonymous mutations identified in the corresponding protein, 344 while numbers into parenthesis represent the **frequency of nonsynonymous mutations, calculated as a 345 percentage per total amino acid number of each viral protein. *3 and 4 mutations were observed in the 346 polymerase L, in csFITC– and csFITC+ respectively. 347 348 349 We found previous reports describing biological effects of substitutions in eight of the 38 positions 350 identified by us with nonsynonymous mutations (Table 2). Thereby, substitutions in proteins N 351 (I270V), P (S116P), G (D6V) and F (K68Q, I79M, N126D, M526L, C550S), might alter viral assembly 352 or budding, antibody recognition, triggering of virus-cell fusion and cell-cell fusion, as well as 353 replication of the viral genome [24,37–42], in agreement with our observation of defective viruses. 354 The other 30 substitutions, most of them located in proteins G and F, have not been described and 355 their biological effect and contribution to hRSV persistence remains unknown; however, we 356 included in table 2 commentary for some of them with regard to their localization or proximity to 357 other characterized mutations. 358 359 360 361 NS1 NS2 N P M SH G F M2-1 L 4 1 3 2 2 1 7 13 1 0 *3–4 M2-1 (2.87) (0.80) (0.76) (0.82) (0.78) (1.60) (2.34) (2.26) (0.51) (0.00) (0.14–0.18) Number of nonsynonymous mutations in hRSV proteins Nonsynonymous mutation frequency** (%) Viruses 2020, 12, x FOR PEER REVIEW 11 of 21 Table 2. Substitutions observed in the persistent hRSV genome and potential biological effects. 362 Gen Substitutions Biological effect Commentaries References NS1 E21K Uncharacterized – – V82A Uncharacterized – – E91G Uncharacterized – – T123S Uncharacterized –NS1 α-helix-3 is constituted by residues 119-134. –Substitutions Y125, L132/L133 and α-helix-3 truncation, reduce NS1-mediated suppression of the antiviral response. [55] NS2 T90A Uncharacterized –P36, L52 and P92 residues are required for NS2 ubiquitination activity, targeting STAT2 for proteasomal degradation. [57] N K60E – – – H89Q Uncharacterized – – I270V Directly involved in interaction with viral P protein –Core RSV polymerase complex consists of L and P, but N and M2-1 are also required. –I270A abrogates interaction with P protein and significantly reduces polymerase activity. [37] P N75D Uncharacterized – – S116P Phosphorylation site –S116, S117, S119, S232 and S237 are phosphorylation sites; mutation or dephosphorylation reduce N–P interaction and hRSV budding from infected HEp-2 cells. –M2-2-dependent transcriptional inhibition requires simultaneous phosphorylation of S116, S117, S119. [47,50,51] M M43I Uncharacterized – – T133I Uncharacterized – – SH T38A Uncharacterized – – G D6V Involved in interaction with viral M protein –D6A substitution reduced 3-fold the interaction of G with matrix M protein (cytoplasmic domain). [38] G29C Uncharacterized –Position inside cytoplasmic domain [50] I50T Uncharacterized –Position inside transmembrane domain [50] T113I Uncharacterized – – T133I Uncharacterized – – I189V Uncharacterized – – N294K Uncharacterized – – F K68Q Triggering of –A positively charged residue in position 68 is [39,46] Viruses 2020, 12, x FOR PEER REVIEW 12 of 21 *Only observed in hRSV from csFITC+ 363 **SCID, severe combined immune deficiency; RdRp, RNA-dependent RNA polymerase domain; PRNTase, 364 polyribonucleotidyl transferase or capping domain 365 366 membrane fusion and recognition by a neutralizing antibody essential to mediate membrane fusion and viral entry. –K68N substitution is associated to low binding affinity of the neutralizing mAb MEDI8897. I79M Stability of the pre-fusion conformation –I79A resulted in absence of membrane fusion, possibly by premature transition to the post-fusion conformation. [40] N80K Uncharacterized – – T101P Uncharacterized – – N126D N-glycosylation site –N126Q substitution does not alter either F protein expression or membrane fusion (position within p27 peptide). [41] I152T Uncharacterized -Position inside fusion domain [54] I292V Uncharacterized – – T323S Uncharacterized – – Y391H Uncharacterized – – E497A Uncharacterized – – H515N Uncharacterized – – M526L Uncharacterizedb ut M526I has been observed –M526I substitution was observed in a SCID** patient with persistent RSV infection, after BM transplant and rising in CD8 T cell count. [24] C550S Palmitoylation site –C550 is a palmitoylated residue of the cytoplasmic domain. –C550S does not alter F expression, although membrane fusion increased partially. [42] M2-1 T110A Uncharacterized – – M2-2 None – – – L G412S Uncharacterized –Position inside **RdRp domain [47] Q1112R* Uncharacterized –Position inside **PRNTase domain V1124I Uncharacterized –Position inside **PRNTase domain K1602R Uncharacterized –Position inside connector domain Viruses 2020, 12, x FOR PEER REVIEW 13 of 21 4. Discussion 367 Considering that epithelial cells and macrophages are susceptible to hRSV infection in the 368 respiratory tract [1–3], we previously studied alterations in the biology of macrophage-like cells 369 during hRSV persistence and described some mechanisms related to maintenance of such long-term 370 viral infection [30–32]. Herein, we focused on features of the virus present in MΦP cultures after 371 several passages and present evidence that hRSV established a steady-state persistence, since 372 hRSV-gRNA was detected in all of the cells in culture, despite the fact that some of them displayed 373 low-to-undetectable N protein expression, according to flow cytometry and confocal microscopy. 374 We also observed a predominant production of defective viral particles and presented for the first 375 time the sequence of a persistent hRSV genome that has been maintained in vitro throughout 120 376 passages. The persistent hRSV genome showed 38 non-synonymous mutations distributed in ten 377 viral proteins; only eight substitutions have been previously studied and are mainly related to 378 alterations in viral assembly or budding, membrane fusion and possibly in evasion of the antiviral 379 response. 380 MΦP has been subcultured for more than 120 passages by means of repeated freezing and thawing; 381 normally we observe cultures with more than 90% of cells expressing hRSV proteins after thawing 382 from liquid nitrogen. However, we also detected some cultures from different passages with 383 relatively low percentage of infected cells (≤70%) and such observation initially suggested that the 384 virus might have sometimes low rates of replication, allowing high proliferation index of the 385 uninfected cell population, as in a carrier-state persistence. However, detection of hRSV-gRNA by 386 FISH in the whole culture and the evidence of a small cell population without N protein expression, 387 suggested hRSV established a steady-state persistence in the macrophage-like cell line P388D1 and 388 we propose the hypothesis that some cells may preserve the viral genome in silent form. Actually, 389 we calculated by confocal microscopy up to 8.0±7.0% hRSV-gRNA+ cells, with undetectable N 390 protein expression. Sorting and independent culture of FITC– cells showed since the first 391 post-sorting passage, about 50% hRSV-protein+ cells and a progressive increase at subsequent 392 subcultures, supporting the evidence that this cell population preserved intracellularly the viral 393 genome. We do not discard that the sorted hRSV– population may have included cells with very 394 low levels of viral proteins, essentially because we did not permeabilized them during 395 immunofluorescence to preserve viability, then, selection of FITC– and FITC+ cells was set up 396 according to their surface protein expression. When we compared percentages of hRSV-protein+ 397 cells analyzed by flow cytometry and confocal microscopy after cell permeabilization (Table 1), the 398 last method provided higher values though without statistical significance. As percentage of 399 hRSV-protein+ cells was determined by flow cytometry from 10,000 events, whereas percentage by 400 confocal microscopy was calculated from <500 cells, the former method could be considered more 401 reliable. However, overlapping between the autofluorescent cell population (control) and cells with 402 low expression of viral proteins by flow cytometry, could avoid accurate definition of positives and 403 negatives during analysis, resulting in underestimation of hRSV-protein+ cells. 404 It has been advised that numerous cycles of freezing and thawing could lead to phenotypic and 405 functional changes in cell lines, although mechanisms have not been described [43]; we consider 406 such effect as a possibility in MΦP, but if this was the case, alterations in hRSV-gRNA transcription 407 and translation were apparently reversible after independent culture of FITC– cells. 408 Viruses 2020, 12, x FOR PEER REVIEW 14 of 21 Previous reports have shown that positive-sense single stranded RNA viruses maintain their 409 genome during persistence as double stranded RNA, in absence or with low level of viral protein 410 expression (e.g. porcine reproductive and respiratory syndrome virus, and Seneca virus A) [34,44]. 411 In negative-sense RNA viruses, both dsRNA and structural viral proteins have been observed, 412 although studies have been performed only during acute infection [45]. We did not discard that 413 FITC– cells from MΦP cultures might be silent reservoirs of hRSV-gRNA; therefore our observations 414 warrant further research to determine mechanisms associated to viral genome maintenance in the 415 host cell during persistence. 416 On the other hand, we previously reported that relative expression of hRSV genome is 11-fold 417 higher in MΦP at 24 h of culture with respect to acutely infected macrophages at 24 h post-infection, 418 and 2000-fold greater in MΦP compared with MΦN (non-infected control) [32]. Hence, despite 419 hRSV genome is actively replicated in MΦP, infectious viral particles are absent as we identified 420 only low-density defective virus (Figure 4). Such defective viruses were detected since early 421 passages of the MΦP culture (after passage 11; [25]); however, they were mixed with infectious 422 viruses displaying densities of 1.18 g/ml. Thus, after passage 120 we only identified viruses with 423 undetectable infectivity in HEp-2 cells and very low infectivity in Vero cells (which do not produce 424 type-I interferon), suggesting that progressive subculture of MΦP produced cumulative changes in 425 the viral genome. 426 Sequencing of the genomic RNA from defective viruses in MΦP supernatants, showed a full-length 427 genome with 38 nonsynonymous substitutions in ten of the eleven viral proteins, being F, G and 428 NS1 the proteins with higher mutation frequency (Figure 5); previous reports have described eight 429 of the 38 identified mutations [37–42,46,47]. 430 Substitutions in the polymerase complex. The polymerase complex required for replication and 431 transcription of the hRSV genome consists of the large protein or L polymerase, the phosphoprotein 432 P, the nucleoprotein N and the processivity factor M2-1. N protein associates with nascent genomic 433 RNA possibly to modulate elongation during replication (increasing polymerase processivity) and 434 avoiding viral RNA destruction by cellular nucleases and/or recognition by the innate immune 435 system [48]. P protein is a fundamental polymerase-cofactor that tethers L polymerase to the N–436 RNA complex. Other roles of P are as chaperone, preventing association of the recently synthesized 437 N with host cell RNAs, and recruiting of M2-1 to the polymerase complex for efficient transcription 438 [49]. Also, P protein has at least five phosphorylation sites of which, S116, S117 and S119 are 439 required for M2-2-mediated transcriptional inhibition [50], whereas this phosphorylated form of P 440 is also able to interact with N protein, contributing to viral assembly and budding from infected 441 cells [50,51]. 442 We observed a total of ten nonsynonymous mutations in P, N, M2-1 and L proteins of which, only 443 two have been previously studied: 1) substitution I270A in N is related to diminished interaction 444 with P and reduced polymerase activity [37]. Our results showed the comparable substitution I270V 445 in protein N from the persistent hRSV; however, we have not evidence of altered polymerase 446 activity since, as mentioned before, hRSV genome is actively replicated in MΦP [32]. 2) 447 Substitutions S116L/A produce only a slight effect in P protein function, although S116D abolishes 448 completely its role during transcription and replication [47]; additionally, the former substitutions 449 partially affect N–P protein interaction [47,50]. We observed the substitution S116P in P protein and 450 we do not discard that such proline might be altering N–P interaction and, in consequence, viral 451 Viruses 2020, 12, x FOR PEER REVIEW 15 of 21 budding during hRSV persistence, although further experiments are necessary to determine 452 relevance of this particular mutation. 453 Unexpectedly, hRSV collected from csFITC+ showed a fourth mutation in the polymerase L with 454 respect to virus from csFITC– that displayed only three mutations. Thus, virus from csFITC+ 455 exhibited two mutations (Q1112R and V1124I) localized in the conserved region V (CRV, residues 456 956-1437) of the polymerase, involved in addition of a methylated cap to viral mRNAs [52]. Even 457 tough we need to study biological relevance of having one or two mutations in the CRV, we 458 observed phenotypic differences between cells from csFITC+ and csFITC–, as the former culture 459 presented abundance of cells with a more rounded morphology and less firmly attached to surface 460 of culture dishes. 461 Substitutions in G protein. The attachment G protein is one of three hRSV envelope 462 glycoproteins (F and SH are the other two) and displays the highest variability with regards to 463 other proteins; for this reason it is a reference for virus classification into antigenic subtypes (A and 464 B) and genotypes [53]. Structurally, G is a type II glycoprotein with cytoplasmic and 465 transmembrane domains localized in the N-terminal region, and an extracellular domain displaying 466 only 44% of amino acid identity between subgroups [54,55]. A secreted form lacking the 467 cytoplasmic and part of the transmembrane (TM) domain is also produced in hRSV-infected cells 468 and escape mutations in these two domains have been observed in infected cultures under selective 469 pressure by a neutralizing anti-G antibody, resulting in reduced amounts of full-length G protein 470 on the cell-surface and into the viral envelope [38]. 471 We found two nonsynonymous substitutions in the cytoplasmic domain of G (D6V and G29C) and 472 one localized in the TM domain (I50T); position D6 has been described as involved in interaction 473 with protein M during matrix assembly (located under the viral envelope), although it has been 474 reported that protein F has a redundant role through its direct interaction with M [38]. Considering 475 the three substitutions observed in the cytoplasmic and TM domains of G in viruses from MΦP 476 cultures, we do not discard main production of this glycoprotein in its soluble form and assembly 477 of virions with low levels of this envelope protein. Interestingly, virions released from cells infected 478 with a G-deleted hRSV, show up to 60% less quantity of SH protein in their envelope [56], 479 suggesting that a lack of G protein may alter hRSV assembly and possibly its buoyant density. 480 Substitutions in F protein. Fusion protein F is essential for viral entry through fusion of the 481 viral envelope with the host-cell plasma membrane, allowing access of the capsid to the cytoplasm 482 to initiate viral replication; also mediates cell-cell fusion and development of syncytia [57]. Protein F 483 is expressed in a pre-fusion state on the viral envelope and undergoes conformational changes to a 484 post-fusion form, upon binding to hRSV receptors that potentially are ICAM-I and nucleolin [58]. 485 Five of the 13 sites with nonsynonymous mutations identified in the persistent hRSV-F protein have 486 been previously studied. Of relevance, K68 provides a positive charge that is essential for lipid 487 bilayer fusion [39]; our finding of the K68Q substitution might be related to absence of syncytia in 488 MΦP cultures and the low-to-undetectable viral infectivity. Also, I79 is related to stability of the 489 pre-fusion conformation and an earlier report indicates I79A substitution avoids membrane fusion, 490 probably by premature transition to the post-fusion conformation [40]; we observed the I79M 491 substitution, which may have similar consequences. It is also notable the uncharacterized 492 substitution I152T localized inside the fusion peptide [59], which might affect the hydrophobic 493 property of this domain, since threonine is a polar amino acid. In summary, substitutions K68Q, 494 Viruses 2020, 12, x FOR PEER REVIEW 16 of 21 I79M and I152T may be associated to altered viral-cell fusion and cell-cell fusion processes in MΦP 495 cultures. 496 Furthermore, a substitution M526I in F was reported in an infant with severe combined 497 immunodeficiency syndrome and persistent hRSV infection (>70 days virus detection in respiratory 498 samples), but only after bone marrow transplant [24]. Biological relevance of this substitution was 499 not studied; however, authors proposed it might be associated to specific or nonspecific immune 500 responses [24]. Considering it was observed a M526L substitution in the persistent hRSV from MΦP 501 cultures, we suggest it might be associated to innate immune pressure. 502 Substitutions in NS1. The two nonstructural proteins, NS1 and NS2, are hRSV molecules 503 responsible of suppressing the type I interferon (IFN-I)-mediated antiviral response [60]. Even 504 though the substitutions we observed in NS1 protein have not been previously characterized, it is 505 notable that the T123S (Table 2) is inside the NS1 helix α-3, which is a determinant domain for 506 innate immune modulation [61]. We previously reported that nuclear localization of IRF-3 and 507 synthesis of IFN-β are constitutive in MΦP, however, autocrine response to this cytokine is blocked, 508 and recombinant-IFN-β does not induce either STAT-1 phosphorylation or antiviral gene 509 transcription [31]. Further research is necessary to determine whether the T123S substitution in NS1 510 altered its role, as suppressor of IFN-I synthesis in MΦP, enabling NS2 as main antiviral effector 511 [62] contributing to maintenance of hRSV persistence. Notably, replication of defective viruses 512 (from supernatants of MΦP cultures) in HEp-2 cells was not significantly improved in Vero cells, 513 suggesting that mutations in NS1 and NS2 proteins were not principal determinants for lowering 514 infectious capacity, as may be mutations in proteins F and G. 515 5. Conclusions 516 Our results provided evidence that hRSV established a steady-state persistent infection in 517 macrophage-like cells, since all of the cells studied from MΦP cultures harbored the viral genome, 518 whereas non-infectious viral particles were mainly produced. However, it is still important to study 519 mechanisms associated to viral genome transmission and regulation of gene expression during 520 hRSV persistence. After sequencing the persistent viral genome isolated from virus particles from 521 MΦP culture supernatants we did not detect either gene deletions or insertions. However, the 522 nonsynonymous mutations identified and previously characterized, suggested that persistent hRSV 523 is defective in our in vitro model mainly by alterations at the level of viral assembly, viral budding 524 and membrane fusion; such alterations may be associated to the long-term infection. Further 525 characterization of the 30 novel nonsynonymous mutations would be useful to understand the 526 complex virus-host cell interaction related to hRSV persistence in macrophage-like cells, and as a 527 possibility, to identify molecular markers of persistent hRSV-infections. 528 529 Author Contributions: Conceptualization, ERT, JAVP; experiments XRG, ERT, JAVP, CSO, MEP, OFH; 530 confocal microscopy analysis, CM; data analysis, ERT, XRG, JAVP, JGB, BG; manuscript preparation ERT; 531 funding acquisition, ERT. 532 Funding: This research was supported by a grant from the School of Medicine of the National Autonomous 533 University of Mexico. 534 535 Acknowledgments: We thank Ana Flisser for providing the infrastructure and for review of this 536 manuscript. We thank María Jose Gómora Herrera for microscopy technical assistance. We thank Tanya Plett 537 Torres, Adriana Gaspar Rodríguez and Elizabeth Linares Alcántara for scientific discussions. Finally, we thank 538 Viruses 2020, 12, x FOR PEER REVIEW 17 of 21 Ricardo Alfredo Grande Cano for technical assistance in preparing samples for RNA-seq. 539 540 Conflicts of Interest: The authors declare no conflict of interest. 541 542 References 543 1. Panuska, J.R.; Hertz, M.I.; Taraf, H.; Villani, A.; Cirino, N.M. Respiratory Syncytial Virus 544 Infection of Alveolar Macrophages in Adult Transplant Patients. Am Rev Respir Dis 1992, 145, 934–545 939, doi:10.1164/ajrccm/145.4_Pt_1.934. 546 2. Papin, J.F.; Wolf, R.F.; Kosanke, S.D.; Jenkins, J.D.; Moore, S.N.; Anderson, M.P.; Welliver, R.C. 547 Infant baboons infected with respiratory syncytial virus develop clinical and pathological changes 548 that parallel those of human infants. American Journal of Physiology-Lung Cellular and Molecular 549 Physiology 2013, 304, L530–L539, doi:10.1152/ajplung.00173.2012. 550 3. Johnson, J.E.; Gonzales, R.A.; Olson, S.J.; Wright, P.F.; Graham, B.S. The histopathology of fatal 551 untreated human respiratory syncytial virus infection. Mod Pathol 2007, 20, 108–119, 552 doi:10.1038/modpathol.3800725. 553 4. Gould, P.S.; Easton, A.J. Coupled Translation of the Respiratory Syncytial Virus M2 Open 554 Reading Frames Requires Upstream Sequences. J. Biol. Chem. 2005, 280, 21972–21980, 555 doi:10.1074/jbc.M502276200. 556 5. Cartee, T.L.; Wertz, G.W. Respiratory Syncytial Virus M2-1 Protein Requires Phosphorylation 557 for Efficient Function and Binds Viral RNA during Infection. J. Virol. 2001, 75, 12188–12197, 558 doi:10.1128/JVI.75.24.12188-12197.2001. 559 6. Cheng, X.; Park, H.; Zhou, H.; Jin, H. Overexpression of the M2-2 Protein of Respiratory 560 Syncytial Virus Inhibits Viral Replication. JVI 2005, 79, 13943–13952, 561 doi:10.1128/JVI.79.22.13943-13952.2005. 562 7. Sedeyn, K.; Schepens, B.; Saelens, X. Respiratory syncytial virus nonstructural proteins 1 and 2: 563 Exceptional disrupters of innate immune responses. PLoS Pathog 2019, 15, e1007984, 564 doi:10.1371/journal.ppat.1007984. 565 8. Shi, T.; McAllister, D.A.; O’Brien, K.L.; Simoes, E.A.F.; Madhi, S.A.; Gessner, B.D.; Polack, F.P.; 566 Balsells, E.; Acacio, S.; Aguayo, C.; et al. Global, regional, and national disease burden estimates of 567 acute lower respiratory infections due to respiratory syncytial virus in young children in 2015: a 568 systematic review and modelling study. The Lancet 2017, 390, 946–958, 569 doi:10.1016/S0140-6736(17)30938-8. 570 9. Sigurs, N.; Bjarnason, R.; Sigurbergsson, F.; Kjellman, B. Respiratory Syncytial Virus 571 Bronchiolitis in Infancy Is an Important Risk Factor for Asthma and Allergy at Age 7. Am J Respir 572 Crit Care Med 2000, 161, 1501–1507, doi:10.1164/ajrccm.161.5.9906076. 573 10. Sigurs, N.; Aljassim, F.; Kjellman, B.; Robinson, P.D.; Sigurbergsson, F.; Bjarnason, R.; 574 Gustafsson, P.M. Asthma and allergy patterns over 18 years after severe RSV bronchiolitis in the 575 first year of life. Thorax 2010, 65, 1045–1052, doi:10.1136/thx.2009.121582. 576 11. Schauer, U.; Hoffjan, S.; Bittscheidt, J.; Kochling, A.; Hemmis, S.; Bongartz, S.; Stephan, V. RSV 577 bronchiolitis and risk of wheeze and allergic sensitisation in the first year of life. European 578 Respiratory Journal 2002, 20, 1277–1283, doi:10.1183/09031936.02.00019902. 579 Viruses 2020, 12, x FOR PEER REVIEW 18 of 21 12. Stein, R.T.; Sherrill, D.; Morgan, W.J.; Holberg, C.J.; Halonen, M.; Taussig, L.M.; Wright, A.L.; 580 Martinez, F.D. Respiratory syncytial virus in early life and risk of wheeze and allergy by age 13 581 years. The Lancet 1999, 354, 541–545, doi:10.1016/S0140-6736(98)10321-5. 582 13. Zomer-Kooijker, K.; Uiterwaal, C.S.P.M.; van der Gugten, A.C.; Wilbrink, B.; Bont, L.J.; van der 583 Ent, C.K. Decreased lung function precedes severe respiratory syncytial virus infection and 584 post-respiratory syncytial virus wheeze in term infants. European Respiratory Journal 2014, 44, 666–585 674, doi:10.1183/09031936.00009314. 586 14. Simões, E.A.F.; Carbonell-Estrany, X.; Rieger, C.H.L.; Mitchell, I.; Fredrick, L.; Groothuis, J.R. 587 The effect of respiratory syncytial virus on subsequent recurrent wheezing in atopic and nonatopic 588 children. Journal of Allergy and Clinical Immunology 2010, 126, 256–262, doi:10.1016/j.jaci.2010.05.026. 589 15. Sikkel, M.B.; Quint, J.K.; Mallia, P.; Wedzicha, J.A.; Johnston, S.L. Respiratory Syncytial Virus 590 Persistence in Chronic Obstructive Pulmonary Disease: The Pediatric Infectious Disease Journal 2008, 591 27, S63–S70, doi:10.1097/INF.0b013e3181684d67. 592 16. Bramley, A.M.; Vitalis, T.Z.; Wiggs, B.R.; Hegele, R.G. Effects of respiratory syncytial virus 593 persistence on airway responsiveness and inflammation in guinea-pigs. European Respiratory Journal 594 1999, 14, 1061–1067, doi:10.1183/09031936.99.14510619. 595 17. Estripeaut, D.; Torres, J.P.; Somers, C.S.; Tagliabue, C.; Khokhar, S.; Bhoj, V.G.; Grube, S.M.; 596 Wozniakowski, A.; Gomez, A.M.; Ramilo, O.; et al. Respiratory Syncytial Virus Persistence in the 597 Lungs Correlates with Airway Hyperreactivity in the Mouse Model. J INFECT DIS 2008, 198, 1435–598 1443, doi:10.1086/592714. 599 18. Piedimonte, G.; Walton, C.; Samsell, L. Vertical Transmission of Respiratory Syncytial Virus 600 Modulates Pre- and Postnatal Innervation and Reactivity of Rat Airways. PLoS ONE 2013, 8, e61309, 601 doi:10.1371/journal.pone.0061309. 602 19. Schwarze, J.; O’Donnell, D.R.; Rohwedder, A.; Openshaw, P.J.M. Latency and Persistence of 603 Respiratory Syncytial Virus Despite T Cell Immunity. Am J Respir Crit Care Med 2004, 169, 801–805, 604 doi:10.1164/rccm.200308-1203OC. 605 20. Jartti, T.; Jartti, L.; Peltola, V.; Waris, M.; Ruuskanen, O. Identification of Respiratory Viruses in 606 Asymptomatic Subjects: Asymptomatic Respiratory Viral Infections. The Pediatric Infectious Disease 607 Journal 2008, 27, 1103–1107, doi:10.1097/INF.0b013e31817e695d. 608 21. Borg, I.; Rohde, G.; Löseke, S.; Bittscheidt, J.; Schultze-Werninghaus, G.; Stephan, V.; Bufe, A. 609 Evaluation of a quantitative real-time PCR for the detection of respiratory syncytial virus in 610 pulmonary diseases. Eur Respir J 2003, 21, 944–951, doi:10.1183/09031936.03.00088102. 611 22. Rezaee, F.; Gibson, L.F.; Piktel, D.; Othumpangat, S.; Piedimonte, G. Respiratory Syncytial 612 Virus Infection in Human Bone Marrow Stromal Cells. Am J Respir Cell Mol Biol 2011, 45, 277–286, 613 doi:10.1165/rcmb.2010-0121OC. 614 23. Cubie, H.A.; Duncan, L.A.; Marshall, L.A.; Smith, N.M. Detection of respiratory syncytial virus 615 nucleic acid in archival postmortem tissue from infants. Pediatr Pathol Lab Med 1997, 17, 927–938. 616 24. Grad, Y.H.; Newman, R.; Zody, M.; Yang, X.; Murphy, R.; Qu, J.; Malboeuf, C.M.; Levin, J.Z.; 617 Lipsitch, M.; DeVincenzo, J. Within-Host Whole-Genome Deep Sequencing and Diversity Analysis 618 of Human Respiratory Syncytial Virus Infection Reveals Dynamics of Genomic Diversity in the 619 Absence and Presence of Immune Pressure. Journal of Virology 2014, 88, 7286–7293, 620 doi:10.1128/JVI.00038-14. 621 Viruses 2020, 12, x FOR PEER REVIEW 19 of 21 25. Sarmiento, R.E.; Tirado, R.; Gómez, B. Characteristics of a respiratory syncytial virus 622 persistently infected macrophage-like culture. Virus Research 2002, 84, 45–58, 623 doi:10.1016/S0168-1702(01)00420-8. 624 26. Martínez, I.; Lombardía, L.; Herranz, C.; García-Barreno, B.; Domínguez, O.; Melero, J.A. 625 Cultures of HEp-2 cells persistently infected by human respiratory syncytial virus differ in 626 chemokine expression and resistance to apoptosis as compared to lytic infections of the same cell 627 type. Virology 2009, 388, 31–41, doi:10.1016/j.virol.2009.03.008. 628 27. Hobson, L.; Everard, M.L. Persistent of respiratory syncytial virus in human dendritic cells and 629 influence of nitric oxide. Clin. Exp. Immunol. 2008, 151, 359–366, doi:10.1111/j.1365-2249.2007.03560.x. 630 28. Ito, M.; Takeuchi, T.; Nishio, M.; Kawano, M.; Komada, H.; Tsurudome, M.; Ito, Y. Early Stage 631 of Establishment of Persistent Sendai Virus Infection: Unstable Dynamic Phase and Then Selection 632 of Viruses Which Are Tightly Cell Associated, Temperature Sensitive, and Capable of Establishing 633 Persistent Infection. JVI 2004, 78, 11939–11951, doi:10.1128/JVI.78.21.11939-11951.2004. 634 29. Woodbury, E.L. A review of the possible mechanisms for the persistence of foot-and-mouth 635 disease virus. Epidemiol. Infect. 1995, 114, 1–13, doi:10.1017/S0950268800051864. 636 30. Nakamura-López, Y.; Villegas-Sepúlveda, N.; Sarmiento-Silva, R.E.; Gómez, B. Intrinsic 637 apoptotic pathway is subverted in mouse macrophages persistently infected by RSV. Virus Research 638 2011, 158, 98–107, doi:10.1016/j.virusres.2011.03.016. 639 31. Rivera-Toledo, E.; Torres-González, L.; Gómez, B. Respiratory Syncytial Virus Persistence in 640 Murine Macrophages Impairs IFN-β Response but Not Synthesis. Viruses 2015, 7, 5361–5374, 641 doi:10.3390/v7102879. 642 32. Santiago-Olivares, C.; Rivera-Toledo, E.; Gómez, B. Nitric oxide production is downregulated 643 during respiratory syncytial virus persistence by constitutive expression of arginase 1. Arch Virol 644 2019, 164, 2231–2241, doi:10.1007/s00705-019-04259-0. 645 33. Lifland, A.W.; Jung, J.; Alonas, E.; Zurla, C.; Crowe, J.E.; Santangelo, P.J. Human Respiratory 646 Syncytial Virus Nucleoprotein and Inclusion Bodies Antagonize the Innate Immune Response 647 Mediated by MDA5 and MAVS. Journal of Virology 2012, 86, 8245–8258, doi:10.1128/JVI.00215-12. 648 34. Guo, R.; Shang, P.; Carrillo, C.A.; Sun, Z.; Lakshmanappa, Y.S.; Yan, X.; Renukaradhya, G.J.; 649 McGill, J.; Jaing, C.J.; Niederwerder, M.C.; et al. Double-stranded viral RNA persists in vitro and in 650 vivo during prolonged infection of porcine reproductive and respiratory syndrome virus. Virology 651 2018, 524, 78–89, doi:10.1016/j.virol.2018.08.006. 652 35. Pinkert, S.; Klingel, K.; Lindig, V.; Dorner, A.; Zeichhardt, H.; Spiller, O.B.; Fechner, H. 653 Virus-Host Coevolution in a Persistently Coxsackievirus B3-Infected Cardiomyocyte Cell Line. 654 Journal of Virology 2011, 85, 13409–13419, doi:10.1128/JVI.00621-11. 655 36. Mbiguino, A.; Menezes, J. Purification of human respiratory syncytial virus: superiority of 656 sucrose gradient over percoll, renografin, and metrizamide gradients. Journal of Virological Methods 657 1991, 31, 161–170, doi:10.1016/0166-0934(91)90154-R. 658 37. Esneau, C.; Raynal, B.; Roblin, P.; Brûlé, S.; Richard, C.-A.; Fix, J.; Eléouët, J.-F.; Galloux, M. 659 Biochemical characterization of the respiratory syncytial virus N 0 -P complex in solution. J. Biol. 660 Chem. 2019, 294, 3647–3660, doi:10.1074/jbc.RA118.006453. 661 38. Ghildyal, R.; Li, D.; Peroulis, I.; Shields, B.; Bardin, P.G.; Teng, M.N.; Collins, P.L.; Meanger, J.; 662 Mills, J. Interaction between the respiratory syncytial virus G glycoprotein cytoplasmic domain and 663 Viruses 2020, 12, x FOR PEER REVIEW 20 of 21 the matrix protein. Journal of General Virology 2005, 86, 1879–1884, doi:10.1099/vir.0.80829-0. 664 39. Hicks, S.N.; Chaiwatpongsakorn, S.; Costello, H.M.; McLellan, J.S.; Ray, W.; Peeples, M.E. Five 665 Residues in the Apical Loop of the Respiratory Syncytial Virus Fusion Protein F2 Subunit Are 666 Critical for Its Fusion Activity. Journal of Virology 2018, 92, 16. 667 40. Bermingham, I.M.; Chappell, K.J.; Watterson, D.; Young, P.R. The Heptad Repeat C Domain of 668 the Respiratory Syncytial Virus Fusion Protein Plays a Key Role in Membrane Fusion. J. Virol. 2017, 669 JVI.01323-17, doi:10.1128/JVI.01323-17. 670 41. Leemans, A.; Boeren, M.; Van der Gucht, W.; Pintelon, I.; Roose, K.; Schepens, B.; Saelens, X.; 671 Bailey, D.; Martinet, W.; Caljon, G.; et al. Removal of the N-Glycosylation Sequon at Position N116 672 Located in p27 of the Respiratory Syncytial Virus Fusion Protein Elicits Enhanced Antibody 673 Responses after DNA Immunization. Viruses 2018, 10, 426, doi:10.3390/v10080426. 674 42. Branigan, P.J. The cytoplasmic domain of the F protein of Human respiratory syncytial virus is 675 not required for cell fusion. Journal of General Virology 2006, 87, 395–398, doi:10.1099/vir.0.81481-0. 676 43. Fazekas, J.; Grunt, T.W.; Jensen-Jarolim, E.; Singer, J. Long term storage in liquid nitrogen leads 677 to only minor phenotypic and gene expression changes in the mammary carcinoma model cell line 678 BT474. Oncotarget 2017, 8, 35076–35087, doi:10.18632/oncotarget.16623. 679 44. Maggioli, M.F.; Fernandes, M.H.V.; Joshi, L.R.; Sharma, B.; Tweet, M.M.; Noll, J.C.G.; 680 Bauermann, F.V.; Diel, D.G. Persistent Infection and Transmission of Senecavirus A from Carrier 681 Sows to Contact Piglets. J Virol 2019, 93, e00819-19, /jvi/93/21/JVI.00819-19.atom, 682 doi:10.1128/JVI.00819-19. 683 45. Mateer, E.; Paessler, S.; Huang, C. Confocal Imaging of Double-Stranded RNA and Pattern 684 Recognition Receptors in Negative-Sense RNA Virus Infection. JoVE 2019, 59095, doi:10.3791/59095. 685 46. Zhu, Q.; Lu, B.; McTamney, P.; Palaszynski, S.; Diallo, S.; Ren, K.; Ulbrandt, N.D.; Kallewaard, 686 N.; Wang, W.; Fernandes, F.; et al. Prevalence and Significance of Substitutions in the Fusion 687 Protein of Respiratory Syncytial Virus Resulting in Neutralization Escape From Antibody 688 MEDI8897. The Journal of Infectious Diseases 2018, 218, 572–580, doi:10.1093/infdis/jiy189. 689 47. Lu, B.; Ma, C.-H.; Brazas, R.; Jin, H. The Major Phosphorylation Sites of the Respiratory 690 Syncytial Virus Phosphoprotein Are Dispensable for Virus Replication In Vitro. JVI 2002, 76, 10776–691 10784, doi:10.1128/JVI.76.21.10776-10784.2002. 692 48. Fearns, R.; Deval, J. New antiviral approaches for respiratory syncytial virus and other 693 mononegaviruses: Inhibiting the RNA polymerase. Antiviral Research 2016, 134, 63–76, 694 doi:10.1016/j.antiviral.2016.08.006. 695 49. Gilman, M.S.A.; Liu, C.; Fung, A.; Behera, I.; Jordan, P.; Rigaux, P.; Ysebaert, N.; Tcherniuk, S.; 696 Sourimant, J.; Eléouët, J.-F.; et al. Structure of the Respiratory Syncytial Virus Polymerase Complex. 697 Cell 2019, 179, 193-204.e14, doi:10.1016/j.cell.2019.08.014. 698 50. Asenjo, A.; Villanueva, N. Phosphorylation of the human respiratory syncytial virus P protein 699 mediates M2-2 regulation of viral RNA synthesis, a process that involves two P proteins. Virus 700 Research 2016, 211, 117–125, doi:10.1016/j.virusres.2015.10.011. 701 51. Villanueva, N.; Navarro, J.; Cubero, E. Antiviral effects of xanthate D609 on the human 702 respiratory syncytial virus growth cycle. Virology 1991, 181, 101–108, 703 doi:10.1016/0042-6822(91)90474-P. 704 52. Braun, M.R.; Deflubé, L.R.; Noton, S.L.; Mawhorter, M.E.; Tremaglio, C.Z.; Fearns, R. RNA 705 Viruses 2020, 12, x FOR PEER REVIEW 21 of 21 elongation by respiratory syncytial virus polymerase is calibrated by conserved region V. PLoS 706 Pathog 2017, 13, e1006803, doi:10.1371/journal.ppat.1006803. 707 53. Do, L.A.H.; Wilm, A.; van Doorn, H.R.; Lam, H.M.; Sim, S.; Sukumaran, R.; Tran, A.T.; 708 Nguyen, B.H.; Tran, T.T.L.; Tran, Q.H.; et al. Direct whole-genome deep-sequencing of human 709 respiratory syncytial virus A and B from Vietnamese children identifies distinct patterns of inter- 710 and intra-host evolution. Journal of General Virology 2015, 96, 3470–3483, doi:10.1099/jgv.0.000298. 711 54. Kwilas, S.; Liesman, R.M.; Zhang, L.; Walsh, E.; Pickles, R.J.; Peeples, M.E. Respiratory 712 Syncytial Virus Grown in Vero Cells Contains a Truncated Attachment Protein That Alters Its 713 Infectivity and Dependence on Glycosaminoglycans. JVI 2009, 83, 10710–10718, 714 doi:10.1128/JVI.00986-09. 715 55. Collins, P.L. O glycosylation of glycoprotein G of human respiratory syncytial virus is 716 specified within the divergent ectodomain. Journal of Virology 1990, 64, 4007–4012, 717 doi:10.1128/JVI.64.8.4007-4012.1990. 718 56. Batonick, M.; Wertz, G.W. Requirements for Human Respiratory Syncytial Virus Glycoproteins 719 in Assembly and Egress from Infected Cells. Advances in Virology 2011, 2011, 1–11, 720 doi:10.1155/2011/343408. 721 57. Rossey, I.; McLellan, J.S.; Saelens, X.; Schepens, B. Clinical Potential of Prefusion RSV F-specific 722 Antibodies. Trends in Microbiology 2018, 26, 209–219, doi:10.1016/j.tim.2017.09.009. 723 58. Mastrangelo, P.; Hegele, R.G. Host and Viral Factors in Respiratory Syncytial Virus Infection. 724 Curr Pediatr Rep 2013, 1, 149–157, doi:10.1007/s40124-013-0019-3. 725 59. Sun, Z.; Pan, Y.; Jiang, S.; Lu, L. Respiratory Syncytial Virus Entry Inhibitors Targeting the F 726 Protein. Viruses 2013, 5, 211–225, doi:10.3390/v5010211. 727 60. Goswami, R.; Majumdar, T.; Dhar, J.; Chattopadhyay, S.; Bandyopadhyay, S.K.; Verbovetskaya, 728 V.; Sen, G.C.; Barik, S. Viral degradasome hijacks mitochondria to suppress innate immunity. Cell 729 Res 2013, 23, 1025–1042, doi:10.1038/cr.2013.98. 730 61. Chatterjee, S.; Luthra, P.; Esaulova, E.; Agapov, E.; Yen, B.C.; Borek, D.M.; Edwards, M.R.; 731 Mittal, A.; Jordan, D.S.; Ramanan, P.; et al. Structural basis for human respiratory syncytial virus 732 NS1-mediated modulation of host responses. Nat Microbiol 2017, 2, 17101, 733 doi:10.1038/nmicrobiol.2017.101. 734 62. Whelan, J.N.; Tran, K.C.; van Rossum, D.B.; Teng, M.N. Identification of Respiratory Syncytial 735 Virus Nonstructural Protein 2 Residues Essential for Exploitation of the Host Ubiquitin System and 736 Inhibition of Innate Immune Responses. J. Virol. 2016, 90, 6453–6463, doi:10.1128/JVI.00423-16. 737 738 © 2020 by the authors. Submitted for possible open access publication under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 739 305 “Lehninger Preparations” Odra Noel Capítulos de Libro 306 12. CAPÍTULOS DE LIBRO • “Proteomics of lignocellulosic substrates bioconversion in anaerobic digesters to increase the carbon recovery as methane” Alicia Guadalupe Talavera- Caro, María Alejandra Sánchez-Muñoz, Inty Omar Hernández-De Lira, Lilia Ernestina Montañez-Hernández, Jesús Antonio Morlett-Chávez, María de las Mercedes Esparza-Perusquía, Nagamani Balagurusamy. Biological Approaches for the Management of Agro-industrial Residues Applied Environmental Science and Engineering for a Sustainable Future. (2020) Editor: Zainul Akmar Zakaria, Ramaraj Boopathy, Julian Rafael Dib, Reeta Rani Singhania. Springer, Cham. ISBN 978-3-030-39137-9. Chapter 4 Proteomics of Lignocellulosic Substrates Bioconversion in Anaerobic Digesters to Increase Carbon Recovery as Methane Alicia Guadalupe Talavera-Caro, María Alejandra Sánchez-Muñoz, Inty Omar Hernández-De Lira, Lilia Ernestina Montañez-Hernández, Ayerim Yedid Hernández-Almanza, Jésus Antonio Morlett-Chávez, María de las Mercedes Esparza-Perusquia, and Nagamani Balagurusamy Abstract Anaerobic digestion (AD) is a cost-effective treatment for management of lignocellulosic substrates, viz., agricultural wastes and animal manures, which also aids in generation of methane as biofuel. Although the application of AD technology is increasing, one of the major limitations of the process is that the rate of fermen- tation is higher than the rate of methanogenesis, which significantly affects process stability and methane yield. Normally, the souring of digesters can be observed after 2–4 weeks after the initiation of the volatile fatty acids accumulation, which makes it difficult for early detection and consequently resulting in acidification of digesters. Of late, metagenomic approaches are gaining importance due to their ability to reveal the microbial diversity and their dynamics in a relatively short time. However, their functional nature could not be clearly explained due to the lack of data on their activity. Recent advances in proteomic studies show its potential as a complemen- tary technology to metagenomic studies for efficient management of digesters. Metaproteomic analyses aid in identifying a shift in metabolic paths and in metabolic networks under stress conditions. This provides insights on functionality, microbial interactions, and provides data on spatiotemporal variations and their dynamics of proteins crucial for efficient performance of the digester. Besides, this technique has A. G. Talavera-Caro · M. A. Sánchez-Muñoz · I. O. H.-D. Lira · L. E. Montañez-Hernández · A. Y. Hernández-Almanza · N. Balagurusamy (*) Facultad de Ciencias Biológicas, Laboratorio de Bioremediación, Universidad Autónoma de Coahuila, Torreón, Mexico e-mail: bnagamani@uadec.edu.mx J. A. Morlett-Chávez Facultad de Ciencias Químicas, Laboratorio de Biología Molecular, Universidad Autónoma de Coahuila, Saltillo, Mexico M. d. l. M. Esparza-Perusquia Facultad de Medicina, Departamento de Bioquímica, Universidad Nacional Autónoma de México, Ciudad de México, Mexico © Springer Nature Switzerland AG 2020 Z. A. Zakaria et al. (eds.), Valorisation of Agro-industrial Residues – Volume I: Biological Approaches, Applied Environmental Science and Engineering for a Sustainable Future, https://doi.org/10.1007/978-3-030-39137-9_4 81 led to identify novel phylotypes with novel functions among the microbial commu- nities of the anaerobic digesters, which suggest the potential of proteomics in bioprospection of novel enzymes for industrial purposes. How proteomics along with metagenomics and transcriptomics data could aid in early detection of distur- bances in the digesters helps in formulating recovery strategies as well as to increase the methane content of biogas will be discussed in this chapter. Keywords Anaerobic digestion · Metabolic networks · Methane · Proteomics 4.1 Introduction The recent use of lignocellulosic biomass as a renewable energy source has been of increasing interest due to the environmental crisis and alarming decline in fossil fuel reserves. In the next few decades, bioenergy will be considered as one of the potential renewable energy sources, in addition to other renewable energy sources, e.g., wind and solar. Wide-scale practice of these technologies depends on the economics involved in their infrastructure andmaintenance. Among all, bioenergy from biomass is constantly investigated due to their easiness in installation and operation. Anaer- obic digestion (AD) is one of the processes widely used to recover energy from biomass sources, like animal manures, solid municipal wastes, paper industry wastes, energy crops, or agricultural wastes, in the form of methane (Nallathambi Gunaseelan 1997). Anaerobic digestion process degrade/oxidize organic matter under anaerobic con- ditions by several consortia of different metabolic groups of microorganisms, where methane (60–70%) and carbon dioxide (30–40%), and other trace gases (<1% hydro- gen, nitrogen, ammonia, and hydrogen sulfide) are major end products. However, methane is the most valuable product because it can be used to generate electricity and heat (Angelidaki et al. 2003). Methane production in AD process involves four different steps: hydrolysis, acidogenesis, acetogenesis, and methanogenesis. Nevertheless, microorganisms carrying out this degradation/transformation reac- tions differs in their physiology, nutritional requirement, growth kinetics, and are sensitive to the environment. This characteristics lead to a delicate balance between all group of microorganisms involved in AD, and any modification can cause instability and consequently, low methane yield (Adekunle and Okolie 2015). For this reason, over the last decade, there has been a rapid development in state-of-the- art techniques to understand the microbial community dynamics, interactions, and functionality to achieve a proper digester efficiency and stability. Besides, it is necessary to study the non-culturable microorganisms involved in the process and its proteome to fill the knowledge gaps in our understanding of the complex microbial interplay and functionality in AD process. The study of microbial diversity and gene expression in AD, through metagenomic and metatranscriptomic analysis, has helped us to reveal “black box” contents and their role to a certain level. Still, there is limited understanding of all the possible metabolic pathways that are active throughout the biomethanation process 82 A. G. Talavera-Caro et al. and without which, successful operation and maintenance of biodigesters for higher methane yield. Hence, the interest on metaproteomics of digesters is gaining atten- tion. This tool can evaluate growth and activity of different microorganisms in relation to their environment (protein expression and localization), to identify posttranslational modification, to infer certain protein–protein interactions, amino acid sequences, and genotypes, besides protein identification (Vanwonterghem et al. 2014). As well, metaproteomic databases can permit to examine targeted biomarkers from microbial communities for evaluation of the biodigester functioning. In this chapter, we focus on the contribution of metaproteomic approaches to gain an insight on the composition of microorganisms sharing similar metabolic structure, and the shift in their dynamics and functions under certain environmental or induced conditions in biodigesters employing lignocellulosic substrates as main feedstock. 4.2 Anaerobic Digestion of Lignocellulosic Substrates Global annual production of available lignocellulosic biomass is 181.5 billion tonnes. In USA alone, about 1.25 billion tonnes of lignocellulosic biomass is produced annually, while in Canada, about 69.25 million tonnes are generated (Paul and Dutta 2018). The use of biomass residues as sources of renewable energy has increased. Recently, the lignocellulosic-rich biomass feedstocks such as fibrous food wastes, animal manures, paper industry wastes, agro residues, and energy crops are mostly used as feedstocks for bioenergy production (Sawatdeenarunat et al. 2015). Apart from biogas production, ethanol and butanol production from ligno- cellulosic biomass, such as wheat straw, corn cob, and sugarcane bagasse is also being studied (Jiang et al. 2017). The major components of the lignocellulosic biomasses are cellulose, hemicellu- lose, and lignin, which are hydrolyzed through a series of reactions by microorgan- isms. Cellulose and hemicellulose are the predominant polysaccharides in these biomass materials. Whereas, lignin is conformed of phenolic polymers, which add recalcitrance to the complex structure of lignocellulose substrates and limits the accessibility of polysaccharides by microbial enzymes (Isikgor and Becer 2015; Liu and Chen 2015). Li et al. (2018) reported the interaction of cellulose, hemicel- lulose, and lignin components on biodegradability of different lignocellulosic bio- masses and observed that methane production was favored and correlated with decomposition of substrates rich in cellulose and hemicellulose, whereas, lignin was not completely digested (Li et al. 2018). Nevertheless, the lignocellulosic biomass shows low rate of polysaccharide hydrolysis due to the presence of lignin (Cesarino et al. 2012). Therefore, to increase hydrolysis rate, methods of pretreatment have been developed (Ariunbaatar et al. 2014) and becoming crucial to anaerobic digestion process. Pretreatments aid to overcome limitations and eliminate the barriers to access polysaccharides for deg- radation, augment digestibility, and consequently increase biogas production from lignocellulosic biomass residues (Chen et al. 2014). Pretreatment techniques such as physical (steam explosion, hydro-thermolysis, thermochemical), chemical (alkalis, 4 Proteomics of Lignocellulosic Substrates Bioconversion in Anaerobic Digesters. . . 83 acids, oxidants as organic solvents), nonconventional (ionic liquids), and biological processes are mostly applied to polysaccharide decomposition (Singh et al. 2015; Carrere et al. 2016). Enhancement in methane yield has been reported for several lignocellulosic residues, which reveal the significant improvements on lignin depolymerization. However, type of pretreatment depends on the composition of lignocellulosic feed- stocks (Table 4.1), since the type and accumulation of products after pretreatment process can either be beneficial or harmful to the microbial consortium of AD process (Poudel et al. 2012; Ahring et al. 2015). Additionally, C:N ratio of lignocellulosic-rich substrates is an important param- eter as high and low ratios were found to have negative impact on the process by altering pH, and consequently inhibiting growth and activity of microbial commu- nities in the biodigester (Rahman et al. 2017). In general, to overcome the limitations due to C:N ratios, the addition of a co-substrate rich in carbon or nitrogen provide optimum conditions for biomethanation and this process is known as co-digestion. Selection of suitable co-substrate is important, in order to enhance synergisms, dilute detrimental compounds, and optimize the methane yield without affecting digestate quality (Mata-Alvarez et al. 2014; Siddique and Wahid 2018). Hence, research on the AD process evolves continuously to identify optimum operational conditions and their relation to the microbial diversity, their function to increase methane yield. 4.3 Recognizing Important Pathways of AD In terms of energy, anaerobic digestion is a green technology, where biogas produc- tion is a more efficient method for energy generation from biomass than other biological and thermochemical conversion processes (Deublein 2009). AD is an alternative to landfills as a means of organic waste management as AD process generates energy apart from reducing methane emissions. Similarly, traditional management of burning conventional forage residues results in atmospheric pollu- tion and the application of AD process can recover energy from these lignocellulosic biomass (Braun et al. 2008). The conversion of agricultural waste is commonly performed in large parallel or serial biodigester systems of different sizes and designs, known as biogas plants (BGPs). The biodigesters are classified depending on some conditions; such as temperature of the process (psychrophilic, mesophilic, or thermophilic), the type of substrate (e.g., silage, animal manure, or dung), and consistency (e.g., Wet process with low solid content or Dry digestion process with high solids content) (Mcinerney et al. 1979; Weiland 2010; Angelidaki et al. 2005). AD is a complex multistep process that is performed by a large consortium of microorganisms composed of four major groups as mentioned previously, hydro- lytic, fermentative, syntrophic acetogenic bacteria, and methanogenic archaea (Fig. 4.1) (Ferry 1993; Zheng et al. 2014). First, hydrolytic bacteria hydrolyze biopolymers (lipids, proteins, and carbohy- drates) to soluble oligomers and monomers (long-chain fatty acids, glycerol, amino 84 A. G. Talavera-Caro et al. T a b le 4 .1 B io ch em ic al co m p o si ti o n o f d if fe re n t li g n o ce ll u lo si c su b st ra te s an d th ei r p o te n ti al m et h an e p ro d u ct io n T y p e o f su b st ra te R ea ct o r v o lu m e (L ) C el lu lo se (% ) H C (% ) L ig n in (% ) C (% ) N (% ) C :N T S (% ) V S (% ) M et h an e p ro d u ct io n (m L C H 4 /g V S ) R ef er en ce s R ic e st ra w 1 3 5 – 4 4 2 7 – 3 4 1 2 – 1 3 3 9 .7 0 .9 4 7 – 6 7 9 2 .9 8 1 .6 2 8 1 P au l an d D u tt a (2 0 1 8 ), S aw at d ee n ar u n at et al . ( 2 0 1 5 ), L i et al . (2 0 1 3 a) W h ea t st ra w 1 3 8 – 4 2 2 0 – 2 7 2 0 – 2 2 3 9 .9 0 .4 5 0 – 6 0 9 0 .5 7 7 .9 2 4 5 C o rn st o v er 1 4 0 2 5 – 3 1 1 4 – 1 7 4 3 .2 0 .8 5 0 – 6 3 8 4 .9 7 6 .9 2 4 1 P au l an d D u tt a ( 2 0 1 8 ), S aw at d ee n ar u n at et al .( 2 0 1 5 ), S in g h et al . (2 0 1 5 ), L i et al . (2 0 1 3 b ) C o rn co b 0 .5 7 5 4 5 2 5 1 5 4 1 .2 6 0 .4 5 1 2 3 a 8 1 .2 2 2 5 4 .2 P au l an d D u tt a (2 0 1 8 ), P ér ez - R o d rí g u ez et al . (2 0 1 6 ), K an w al et al . ( 2 0 1 9 ) S u g ar ca n e b ag as se 2 4 0 – 4 5 2 0 – 2 4 2 5 – 3 0 4 6 .0 8 0 .7 4 1 1 8 – 1 5 0 7 5 – 1 6 7 3 .5 5 8 4 .7 5 P au l an d D u tt a ( 2 0 1 8 ), In y an g et al . (2 0 1 0 ), M u st af a et al . (2 0 1 8 ) S w it ch g ra ss 1 3 6 – 4 5 2 8 – 3 0 1 2 – 2 6 4 3 .6 0 .4 9 0 9 1 .3 8 7 .4 2 4 6 P au l an d D u tt a ( 2 0 1 8 ), L i et al . (2 0 1 3 b ) C h ic k en m an u re 1 2 0 2 3 .2 1 .6 3 5 .9 3 .4 1 0 .9 2 5 .9 1 9 .5 2 9 5 P au l an d D u tt a (2 0 1 8 ), L i et al . ( 2 0 1 3 b ) D ai ry m an u re 1 1 9 .5 1 5 .2 1 7 .4 3 7 .6 2 .8 1 3 .4 3 8 .5 2 8 .8 5 1 L i et al . (2 0 1 3 b ) S w in e m an u re 1 1 1 .3 2 7 .7 4 .3 3 4 .8 2 .2 1 5 .8 3 0 .4 2 2 3 2 2 L i et al . ( 2 0 1 3 b ) F o o d w as te 1 1 2 5 .9 7 .9 4 3 .3 3 .3 3 – 1 7 3 .7 3 .3 3 4 2 L i et al . ( 2 0 1 3 b ), D iv y a et al . (2 0 1 5 ) H C H em ic el lu lo se a N o t d et er m in ed 4 Proteomics of Lignocellulosic Substrates Bioconversion in Anaerobic Digesters. . . 85 acids, and monosaccharides) by extracellular enzymes. These compounds are further converted to volatile fatty acids (VFAs; butyrate, propionate, acetate, among others), alcohol (ethanol and methanol), H2, and CO2 by fermentative bacteria. Eventually, VFAs greater than C2 and alcohols are oxidized to acetate, hydrogen (H2), formate, and CO2 by syntrophic acetogens. Finally, the last group of methanogenic archaea converts acetate and CO2 to methane (Gujer and Zehnder 1983). In the last AD step (methanogenesis), a complex interplay among different func- tional microorganisms occurs. Hydrogenotrophic methanogens oxidize H2 into meth- ane (CH4) by using CO2.While themethyl group of acetate ormethylamines is reduced to CH4 by acetoclastic methanogenesis (Schink 1997). Syntrophic acetate oxidation (SAO) (Schnürer et al. 1999) also occurs under anaerobic conditions, yielding CO2 and H2, which feed the hydrogenotrophic methanogens (Fig. 4.1). Methanogenic stage is one of the rate-limiting steps as the growth rate of methanogens is low as well as they are sensible to environmental fluctuations such as pH, temperature, and VFAs concen- tration (Chen et al. 2008). In addition, several other factors influence biogas yield, mainly the recalcitrant nature of the substrate, the binding of bacteria on the substrate during the hydrolysis stage (Angelidaki et al. 2011) and high ammonia concentrations (Appels et al. 2011). It is worth to mention, methanogenesis involve an optimal organization and interaction among different bacterial and archaeal communities, specific syntrophic interactions, and an imbalance can affect growth and activity of the microbial communities and could cause a deterioration in reactor performance, and thereby decreasing the methane yield (Krause et al. 2008; Rastogi et al. 2008; Akuzawa et al. 2011). Understanding the structure of microbial communities, the possible interac- tions among different microbial groups, and the active metabolic pathways could help in improving the methane yield. As mentioned earlier, the metaproteomics is a useful Fig. 4.1 General map of the four main stages of anaerobic digestion with the most abundant intermediates influencing methane yield 86 A. G. Talavera-Caro et al. tool that can provide information on transcription and translation, giving discernment between regulation of gene expression, protein synthesis, stability, and turnover of mRNA and proteins synthesized in situ. Of late, this approach has been successfully applied to laboratory- and full-scale anaerobic systems with potential in biogas production from lignocellulose substrates, as seen in Table 4.2. Laboratory-scale digesters systems are scaled-down models to investigate new different substrate composition, microbial diversity, and its efficiency on organic matter removal and consequently methane yield potential and for testing new reactor configurations (Herrmann et al. 2011). These systems allow us to add control tools to handle several operational parameters and to control a malfunction, if there is any. In contrast, full-scale BGP represents a bigger challenge in operation and maintenance and some disturbances are hard to manipulate (Gerardi 2003). Table 4.2 Overview of metaproteomic studies on AD of lignocellulosic biomass Reactor type Substrate Major findings References Laboratory-scale reactors Thermophilic 8 L- stirred tank reactor (55 C) Mixture of beet silage (95%) and chopped rye (5%) Proteins involved in acetoclastic and hydrogenotrophic methanogenesis, energy con- servation, and a heat shock protein were identified. Most of them belong to Methanosarcinales, and others to Methanomicrobiales and Synergistales Hanreich et al. (2012) Mesophilic 2 L digester under acid stress conditions (35 C) Blended Taihu blue algae Proteins involved in methane production and energy metab- olism were identified (MCR: methyl-coenzyme M reduc- tase, alcohol dehydrogenase, coenzyme-B sulfoethyl thiotransferase) from Methanosarcinales, Methanomicrobiales, and Clostridiales Yan et al. (2012) Mesophilic 500 mL batch digesters (38 C) Cut straw and hay Members of Bacteroidetes were responsible for carbohy- drate metabolism, while fla- gellins from Firmicutes showed its prevalence among the community. Otherwise, abundant enzymes from methanogenesis were detected from Methanobacteriales, Methanosarcinales, and Methanomicrobiales Hanreich et al. (2013) (continued) 4 Proteomics of Lignocellulosic Substrates Bioconversion in Anaerobic Digesters. . . 87 Table 4.2 (continued) Reactor type Substrate Major findings References Thermophilic 1 L digester (55 C) Unprinted office paper and anaerobic sludge from thermophilic industrial digester fed with municipal solid wastes Cellulose and hemicellulose hydrolysis and fermentation enzymes were strongly related to Caldicellulosiruptor spp. and Clostridium thermocellum. Hydrogenotrophic pathway enzymes assigned to Methanobacteriales. Coprothermobacter proteolyticus recognized to perform proteolysis and fermentation Lü et al. (2014) Mesophilic 5 L con- tinuous stirred tank reactors (CSTRs, 37 C) designed as a “bovid-like” digestive system Aerobic sludge from a wastewater treatment plant inoculated with cow digestive tract contents (RU) and cow manure (CO) Glycogen-accumulating microorganisms (Competibacteraceae with two-phase metabolism from aerobic to anaerobic) domi- nated the CSTRs. Hydrogenotrophic methanogenic proteins domi- nated the RU. While CO reac- tor was affiliated to subunits from acetyl-CoA decarbonylase/synthase (ACDS) complex and acetate kinase from Methanosarcinales and Methanomicrobiales Bize et al. (2015) Thermophilic digester (2 L; 55 C) Fresh and digested swine manure The most abundant proteins belong to energy production and conversion, carbohydrates, lipids, and amino acid metab- olism, followed by information storage and cellular processing proteins. The proteins related to energy production were ATPases subunits, MCR, and acetyl CoA decarbonylase Lin et al. (2016) Mesophilic and ther- mophilic anaerobic digesters (5 L); 37 and 55 C) Acidified grass Both mesophilic and thermo- philic reactors contained high abundance of glycolytic pro- teins, sugar transport systems, and phosphotransferase sys- tems affiliated to Firmicutes. While in thermophilic condi- tions chaperons and heat shock proteins were overexpressed Abendroth et al. (2017) (continued) 88 A. G. Talavera-Caro et al. Table 4.2 (continued) Reactor type Substrate Major findings References Mesophilic 2 L batch reactors (37 C) Reed straw (pretreated with cellulase) and swine manure sludge Bacterial proteins (Firmicutes, Proteobacteria, Actinobacteria, and Bacteroidetes) were mainly affiliated to polymer metabo- lism. Ferredoxin-NADP reductase for H2 production was assigned to Azotobacter. The most abundant metaproteins were acetyl-CoA decarbonylase, MCR, and ace- tate pathway from Methanosarcinales Jia et al. (2017a) Mesophilic 2 L reac- tors (35 C) Food waste with short- term hydrothermal pretreatment Carbohydrate and energy metabolism were the most active functions during the H2 production stage of Firmicutes and Bacteroidetes. Proteins from acetate metabolism, methylotrophic and acetoclastic pathways increased during the methanogenic stage were assigned to Methanosarcinales, Methanobacteriales, and Methanotococcales Jia et al. (2017b) Mesophilic (R1-2), thermophilic (R3-4), and high ammonia levels (R5-6) of multibioreactor sys- tem (500 mL) Sludge from BGP fed with corn, silage, pressed and pulp turnip, chicken dung, liquid manure, and iron sludge Enzymes of glycolysis and amino acid biosynthesis were found in R1. In reactors 3 and 4 decreased MCR from Methanosarcinales and Methanobacteriales. While in R5-6 increased the expression of 5,10-ethylene H4MTP reductase and subunits of ACDS complex from Methanobacteriales and Methanosarinales Kohrs et al. (2017) Mesophilic 500 mL reactor Cut filter paper The most upregulated proteins included carbohydrates hydro- lases, ABC transporter pro- teins, outer binding proteins, copper amine oxidases, trans- lation elongation factors, car- boxyl transferase, glyceraldehyde 3-phosphate dehydrogenase, and flagellins, mostly belonging to Firmicutes, Synergistetes, and Bacteroidetes Speda et al. (2017) (continued) 4 Proteomics of Lignocellulosic Substrates Bioconversion in Anaerobic Digesters. . . 89 Table 4.2 (continued) Reactor type Substrate Major findings References Mesophilic 4 L leach bed reactors (37 C) Acidified ensiled rye- grass and granular sludge Proteins related to carbohy- drate hydrolysis, glycolysis, and transport proteins were assigned to Firmicutes, Bacteroidetes, Spirochaetes, and Proteobacteria. ATPases and oxidoreductases belong mostly to Firmicutes, Proteobacteria, and Bacteroidetes. Meanwhile, proteins from lipid and amino acid metabolisms and environ- mental stress were affiliated to Firmicutes and Bacteroidetes Joyce et al. (2018) Mesophilic 10 L stirred tank reactor (37 C) Dried distiller grains feedstock under trace element deprivation Trace element (TE) deprivation causes a decrease of hydrogenotrophic metaproteins from Methanomicrobiales. Only coenzyme F420-reducing hydrogenase and methyl- H4MTP increased its abun- dance upon the addition of TE. Methylotrophic and acetoclastic metaproteins decreased while formylmethanofuran dehydro- genase from Methanosarcinales increased Wintsche et al. (2018) Full-scale biogas plant Mesophilic biogas plants (BGPs; 270–2280 m3) Corn-/grass-/rye whole crop silages Piglet manure/cattle manure/cattle slurry Peptidases, glycolytic enzymes, glucose transporters, ribosomal proteins, chaperons, amino acid metabolism, and energy conservation proteins were identified from bacteria. No cellulolytic enzymes were detected. For Archaea, hydrogenotrophic and acetoclastic metaproteins from Methanobacteriales and Methanosarcinales were iden- tified. Changes in protein pro- files correlated to MCR decrease upon acidification Heyer et al. (2013) (continued) 90 A. G. Talavera-Caro et al. Table 4.2 (continued) Reactor type Substrate Major findings References Mesophilic (43 C) and thermophilic (52 C) BGPs (1500–1600 m3) M: Whole crop silages of maize, forage rye, cattle manure, and slurry T: Mix of maize whole crop silage and poultry manure Carbohydrate hydrolases, sugar transporters, glycolytic enzymes, and primary fermen- tation enzymes were identified. Most of the mesophilic pro- teins were affiliated to Methanosarcinales. Whereas Firmicutes and Thermotogales were assigned to thermophilic BGP, as well as Methanobacteriales Kohrs et al. (2014) Mesophilic BGP (43 C; 1500 m3) Whole crop silages of maize and rye, cattle manure and cattle slurry Proteins as H4MPT S-methyltransferase, V-type H+-transporting ATPase and MCR from both acetoclastic and hydrogenotrophic path- ways were dominant and belonged to Methanomicrobiales and Methanosarcinales. Subunits of ACDS complex were affili- ated to Methanosarcinales Theuerl et al. (2015) 35 mesophilic and thermophilic BGPs (min. 33 C, max. 55 C; 20–4000 m3) Agricultural substrates, industrial wastes, slaughterhouse wastes, sewage sludge, munici- pal waste, mixed and unknown substrates The 40 BGPs were dominated by methanogenic enzymes related to nutrient transport and one-carbon metabolism. The most abundant metaproteins (MCR and 5,10-methylene H4MTP reductase) belonged to Methanobacteriales and Methanosarcinales. At 33 C, proteins from short fatty acid metabolism, lipid and one-carbon metabolism were abundant. At 55 C, proteins from DNA recombination and repair, and amino acid biosyn- thesis were abundant Heyer et al. (2016) Mesophilic and ther- mophilic BGPs (1–3: 37 and 4: 54 C; 105 m3) BGP1: Maize silage, sugar beet, and poultry manure. BGP2: Maize silage, grass, and pig/cattle manure. BGP3: Maize silage and pig manure. BGP4: Maize silage, grass, and pig manure ABC transporters, carbon and methane enzymes were assigned to BGP3. ABC trans- porters were highly expressed and affiliated to Firmicutes and Bacteroidetes, as well as, hypothetical substrate-binding proteins. Glycolytic enzymes were identified from Firmicutes and Bacteroidetes. While members of Methanosarcinales and Ortseifen et al. (2016) (continued) 4 Proteomics of Lignocellulosic Substrates Bioconversion in Anaerobic Digesters. . . 91 4.4 Metaproteomics in AD of Lignocellulosic Substrates Metaproteomics was first defined by Wilmes and Bond (2004) as “the large-scale characterization of the entire protein complement of environmental microbiota at a given point of time.” Through the years several denominations have been used depending on the different experimental procedure, the complexity of the environmen- tal sample or the outcomes. Terms include environmental proteomics,metaproteomics, community proteomics, proteogenomics, and proteotyping. However, not all are syn- onyms. Schneider and Riedel (2010) mentioned that environmental proteomics refers to the proteome analysis of environmental samples, whilemetaproteomics is the study of highly complex biological systems containing a large number of proteins, which is hard to assign to species within a phylotype. In contrast, community proteomics infers that most of the proteins identified are assigned specifically to members of the com- munity. Proteogenomics links the gene function to the identified protein, giving the accurate information about a biological system functionality. On the other hand, proteotyping refers to a gel-free approach, supported by the rapid protein resolution bymass spectrometers for the characterization ofmixedmicrobial communities (Kohrs et al. 2017). Despite several definitions, proteomics englobes a large-scale study of proteins, which allows the understanding the metabolic networks, syntrophic interac- tions, carbon and nitrogen fluxes, and novel pathways. Several analytical methods have been applied to provide an insight into microbial communities in AD, commonly genomic approaches. Cloning and sequencing of Table 4.2 (continued) Reactor type Substrate Major findings References Methanomicrobiales were responsible for hydrogenotrophic and methylotrophic methanogenesis Thermophilic indus- trial biogas reactor (60 C; 2200 m3) Food waste with high levels of free ammonia Dictyoglomales and Planctomycetes were highly active in polysaccharide hydrolysis. Proteins from obli- gate hydrogenotrophic methanogens dominated over acetoclastic methanogens. Novel phylotypes of SAO-bacteria (unFi_c1 and unFi_c2) were identified and perform β-oxidation of buty- rate and other longer chain fatty acids, as well as in acetate oxidation Hagen et al. (2017) 92 A. G. Talavera-Caro et al. DNA or fingerprint target of 16S rRNA gene have been usually applied to explore communities of Archaea and Bacteria (Clement et al. 1998; Schlüter et al. 2008). However, metaproteomics emerged as a complementary approach to give a full vision of the physiological and biochemical functions of microbial population. General metaproteomic workflow comprises biogas community sampling, protein extraction, protein gel separation, tryptic digestion of proteins, mass spectrometry of resulting peptides, and database searching of mass spectra (Hassa et al. 2018). However, as mentioned before, new gel-free approaches have led to rapid resolving mass spectrometers (MS) for rapid identification and characterization of microbial communities employing tandem MS and MS/MS-based shotgun proteomics (Karlsson et al. 2015). 4.4.1 Hydrolysis As mentioned previously, microbial communities degrade polymeric biomass into monomers by hydrolytic enzymes during the first step of AD process in order that simpler compounds are available for the next steps of biomethanation process. The three primary substrates for hydrolysis are polysaccharides, lipids, and proteins, which are generally present in majority of the wastes or feedstocks of anaerobic digesters (Tong et al. 1990). In the case of polysaccharides, there exist two basic types of enzyme system for its hydrolysis: complex systems as cellulosomes, produced by anaerobic bacteria and nonassociated, free enzyme systems produced by aerobic microorganisms (Fig. 4.2) (Felix and Ljungdahl 1993). The first metaproteomic study conducted by Hanreich et al. (2013), demonstrated the presence of α-amylase and glycoside hydrolase only. α-amylase is an endoamylase, which acts on α-1,4 glycosidic bonds in amylose or amylopectin of starch, releasing oligosaccharides of different length. While, glycoside hydrolases are capable of hydrolyzing cellulose, hemicellulose, and starch. This study employed maize-digestate from a biogas plant fermenting maize and a mix of cut straw and hay as feedstock. The metaproteome was dominated only for few proteins from genus Thermoanaerobacterum and Microscilla and in less abundance, the proteins of Cytophaga, which synthetizes pectate lyase. This enzyme catalyzes the eliminative cleavage of pectate, a main component of cell walls in plants. In contrast, a study of a biogas plant treating silage or whole maize crop showed a set of hydrolytic enzymes performing degradation of high-molecular carbohydrates like cellulose, hemicellu- lose, xylan, and arabinan (Heyer et al. 2013). Other nonagricultural substrates, as waste papers, with a composition of about 70% of cellulose and 30% of hemicellulose has led to the identification of structural and catalytic components of C. themocellum cellulosome; CelS and CelJ, as well as hydrolytic enzymes degrading high-molecular carbohydrates. Other enzymes as β-mannanase, acetyl xylan esterase, and endoxylanase, specialized in hemicellulose degradation were related to Caldicellulosiruptor genus (Lü et al. 2014). On the other hand, food wastes as feedstock, with high levels of free ammonia, indicated that an 4 Proteomics of Lignocellulosic Substrates Bioconversion in Anaerobic Digesters. . . 93 uncultured Atribacteria was mainly responsible for the hydrolysis of polysaccha- rides synthesizing enzymes as β-glucoside, galactose mutarotase, L-fucose isomer- ase, and xylose isomerase potentially related with hemicellulose degradation. For protein hydrolysis, proteases, and endopeptidases belonging to C. proteolyticus are reported frequently, when protein-rich biomass is treated in Fig. 4.2 Schematic diagram of the nonassociated and associated cellulase systems. Most aerobic microorganisms degrade cellulose by secreting a set of complex enzymes viz., endoglucanase, exoglucanase, and β-glucosidase. On the contrary, most of the anaerobic microorganisms produce cellulosomes. The cellulosome is an extracellular multienzymatic complex present on the cell wall and binds to the substrate for its hydrolysis. It can incorporate several hydrolases through cohesin– dockerin interaction, while the carbohydrate-binding module keeps the cellulosome attached to the lignocellulosic substrate. This figure was modified from Zhu and McBride (2017) 94 A. G. Talavera-Caro et al. thermophilic biodigester (Heyer et al. 2013). Other enzymes also related to protein degradation are trypsin-like serine protease and it is reported that Planctomycetes and Atribacteria groups are metabolically active in carbohydrate and protein degra- dation (Hagen et al. 2017). Recently, metaproteomics is employed as a bioprospecting tool for identifying novel enzymes. Speda et al. (2017) showed this tool is useful to identify and select novel enzymes from consortia, that are specifically upregulated upon its induction. Cellulolytic activity was targeted in a defined medium containing filter paper, instead of glucose, and compared with a non-induced sample. Cellulose induction led to the identification of 1,4 β-cellobiosidase, 1,4 β-xylanase, cellobiose phosphorylase, β-glucosidase, and hypothetical Ig domain proteins from several species. 4.4.2 Nutrient Transport Substrate transporter systems are of great relevance for the following steps of methanogenesis. The main mechanism by which a microorganism can obtain nutri- ents from the environment is by means of these proteins. Diverse studies have evidenced transport proteins in anaerobic reactors. Two classes of proteins involved in nutrient transport are TonB-dependent receptors and ATP-binding cassette (ABC) transporters. These proteins have been previously reported in AD from several members of Bacteroidetes and Spirochaetes phylum (Hanreich et al. 2013). The TonB-dependent transporter (TBDT) is a bacterial outer protein that can actively transport siderophores, as well as, vitamin B12, nickel complexes, and carbohy- drates. This receptor is part of a starch utilization system that has been recognized by its efficiency to transport oligosaccharides to its further degradation. As mentioned before, TBDT is deployed for more complex substrates and uses a proton motive force for the uptake of oligomers that are too large to diffuse via porins (Lü et al. 2014). Interestingly, TBDT also can degrade polymers as polysaccharides, proteins, proteoglycans and via substrate-binding hydrolytic proteins. In contrast, ABC proteins are a family of primary transporters that hydrolyzes ATP to transport organic and inorganic compounds. ABC systems are bioenergetically expensive as ATP hydrolysis is needed to translocate the substrate across the membrane. Consequently, investment of ATP in this transport mechanism limits binding and transport, especially when higher concentration metabolites are present. ABC transporters have been reported to be related to peptide transport, maltose, and other metabolites, such as glycerol 3-phosphate (Speda et al. 2017; Kohrs et al. 2014; Hagen et al. 2017). Ortseifen and colleagues (2016) showed in an integrated metagenome-proteome research digesting maize silage and pig manure, that mostly ABC-transporters of peptides, oligopeptides, monosaccharides, and iron of the phylum Firmicutes were upregulated, and as well as other translocating proteins from Spirochaetes, Thermotogae, and Thermococcus phylum. While Jia and coworkers (2017a) made an extensive work of the metaproteome evaluating the four different stages of methanogenesis (peak stage of hydrogen production, late 4 Proteomics of Lignocellulosic Substrates Bioconversion in Anaerobic Digesters. . . 95 stage of hydrogen production, peak methanogenic stage, and late methanogenic stage) in a bioreactor fed with cellulase-pretreated reed straw. They found ABC protein expression increased during the peak methanogenic stage where methane production potentials and methane production rate reached 2709.94 mL and 9.71 mL/h. Otherwise, components of sugar transport systems (like the phosphotransferase system) were identified mainly from Firmicutes species, in a biodigester fed with grass, and with a separate acidification step at thermophilic and mesophilic condi- tions (Abendroth et al. 2017). This active transport is used by bacteria for uptake of carbohydrates, particularly hexoses, hexitols, and disaccharides, where the source of energy is from phosphoenolpyruvate (Roseman 1969). 4.4.3 Acidogenesis After hydrolysis and nutrient transport, monosaccharides and amino acids are the most abundant substrates for fermentation and a wide range of microorganisms can metabolize both, mostly Clostridia and other Gram-positive bacteria (Madigan et al. 2008; Ramsay and Pullammanappallil 2001). Monosaccharides are channeled to catabolic pathways for the production of pyru- vate via the Embden–Meyerhof–Parnas (EMP; glycolysis) or Entner Doudoroff (ED) pathway. During glycolysis, reducing equivalents like NADH and H2 are produced, and pyruvate is further metabolized to acetate, CO2, and H2 (at low partial pressure) or subsequently to C3 products (lactate or propionate), or C2/C4/C6 products (acetate/butyrate/caproate) via acetyl-CoA (at high partial pressure). At low partial pressure of H2, the flow of electrons (NADH) lead to H2 production which leads to pyruvate degradation. As partial pressure increases, the flow of electrons shift to the generation of reduced electron fermentation products (volatile fatty acids, VFAs) such as propionate and long-chain fatty acids, lactate, or ethanol. Thus, in a system where methanogens are effectively consuming H2, low concentrations of ethanol, lactate, and butyrate are maintained (Bräsen et al. 2014). At first evaluation of a full-scale agricultural BGP metaproteome, Heyer et al. (2013) demonstrated the identification of metabolic enzymes involved in glycolysis as glyceraldehyde-3-phosphate dehydrogenase (G3PD), enolase, phosphoglycerate kinase, glycerol kinase, and lactate dehydrogenase (LDH), mostly associated to Clostridia. Similar results were found in agricultural BGPs in mesophilic and thermo- philic conditions assigning proteins from glycolysis as 6-phosphofructokinase, aldol- ase, G3PD, 3-phosphoglycerate kinase, phosphoglycerate mutase, and enolase, as well as enzymes from the primary fermentation: LDH (assigned to Lactobacillus), NADP- dependent isopropanol dehydrogenase, and aldehyde dehydrogenase also related to Clostridia (Kohrs et al. 2017). Jia et al. (2017a) also demonstrated the presence of LDH in Streptococcus during the hydrogen production stage. Other BGP digesting maize silage and pig manure revealed highly abundant proteins in fermentation. In spite of the low abundance reported, the use of an 96 A. G. Talavera-Caro et al. integrated approach for combining metaproteomics and metagenomics tools aid in the identification of glycolysis enzymes (enolase, aldolase, and G3PD) (Ortseifen et al. 2016). Similar results were found in a case study evaluating a proteome from a thermophilic reactor degrading swine manure, where enolase from different species of phyla Proteobacteria were identified (Lin et al. 2016). On the other hand, an industrial biogas reactor predominantly fed with food waste and high levels of free ammonia, few enzymes from glycolysis were identified, and attributed to uncultured phylotypes Atribacteria, Planctomycetes, and Dictyglomus (Hagen et al. 2017). In contrast, a mesophilic reactor containing pretreated food waste (under short-term hydrothermal) recorded a higher proportion of proteins of carbo- hydrate metabolism to Proteobacteria, Firmicutes, Actinobacteria, Bacteroidetes, and Cyanobacteria. The proteins involved belong to glycolysis, pyruvate, propio- nate, glyoxylate, and dicarboxylate metabolism (Jia et al. 2017b). In lab-scale models, anaerobic digestion of office paper, led to the identification and assignation of all glycolytic enzymes to C. thermocellum, C. proteolyticus, and Caldicellulosiruptor genus. Also, proteins involved in the synthesis of fermentation products such as lactate, ethanol, butanol, acetate, formate, and butanoate were also assigned to the same genus (Lü et al. 2014). Corresponding to the metabolism of grass, as lignocellulosic biomass, a set of glycolytic enzymes were identified specif- ically in the phase of sugar assimilation (Abendroth et al. 2017). Speda et al. (2017) were able to identify carboxyl transferase (gluconeogenesis) and G3PD (glycolysis) in a cellulose-rich (paper filter) biodigester. These findings showed that the EMP is one of the key glycolytic pathways functional during anaerobic digestion, as well as in the formation of intermediary products of fermentation, leading to the next step of methane production (Wilmes and Bond 2009; Abram et al. 2009). 4.4.4 Acetogenesis During this stage, syntrophic bacteria oxidize VFAs greater than C2 to produce key intermediates (25% acetate and 11% H2) of the process. In this acetogenesis step, obligatory hydrogen forming syntrophic bacteria can cause toxic effects by accu- mulating high hydrogen pressure on the system. Consequently, the microbial con- sortium is not capable to survive under those conditions. Hence, symbiosis, as a syntrophic relationship, is necessary between acetogenic bacteria and autotrophic methanogens or sulfate-reducing bacteria for hydrogen consumption. The hydrogenotrophic methanogens keep the hydrogen pressure low, which contributes to a thermodynamically controlled condition for the fermentative bacteria to con- tinue oxidizing the organic compounds (e.g., ethanol, propionate, and butyrate into acetate) (Barua and Dhar 2017). This oxidizing activity is related to the genera of Syntrophomonas and Syntrophobacter, as well to Chloroflexi, Actinobacteria, and Spirochaetes. Nonetheless, more or less abundance is related as well to Gelria, Lachnospiraceae (uncultured), Ruminococcaeae, Incertae sedis, Sporanaerobacter, and Petrobacter (Ziemiński and Frąc 2012; Wang et al. 2017; Jain et al. 2015). 4 Proteomics of Lignocellulosic Substrates Bioconversion in Anaerobic Digesters. . . 97 In the case of butyrate, the oxidizing pathways are through the β-oxidation, for propionate oxidation will proceed through the methyl-malonyl-CoA (MMC) path- way and for syntrophic oxidation will be associated to the Wood-Ljungdahl (WL) pathway. According to Hagen et al. (2017), Syntrophomonas genus is the major phylotype in AD, where S. wolfei required all classes of the β-oxidation enzymes (acyl-CoA dehydrogenase, enoyl-CoA hydratase, 3-hydroxyacyl-CoA dehydrogenase, and 3-ketoacyl-CoA thiolase). While for the propionate degradation, Pelotomaculum thermopropionicum identification suggests the requirement of the methyl-malonyl-CoA (MMC) cluster and propionate CoA transferase (PCT) cluster in high abundance. Further, this study reported that Thermoacetogenium phaeum for acetate oxidation by WL pathway (formyltetrahydrofolate synthase, 5,10- methylenetetrahydrofolate dehydrogenase, methylenetetrahydrofolate reductase, trimethylamine:corrinoid methyltransferase, carbon monoxide dehydrogenase/ace- tyl-CoA, phosphotransacetylase, and acetate kinase) (Hagen et al. 2017). Role of all enzymes participating in AD has been poorly reported, due to the large amounts of proteins and other interfering substances present in the sample.Nevertheless, a relatively high abundance of proteins involved in acetogenesis has been revealed.Mostly, WL-like formyltetrahydrofolate synthase, 5,10-methylenetetrahydrofolate dehydroge- nase, methylenetetrahydrofolate reductase, trimethylamine-corrinoid methyltransferase, carbon monoxide dehydrogenase-acetyl CoA, phosphotransacetylase, and acetate kinase have been identified as crucial for energy transport and microbial interactions performed mainly in syntrophic acetate oxidizers. A protein cluster encoding Fe–S oxidoreductase and an electron transfer flavoprotein were also identified, both related as well as electron transfer mechanisms (Hagen et al. 2017). 4.4.5 Interspecies Hydrogen Transfer in Syntrophs Interspecies hydrogen transfer mechanism is vital in syntrophic relationships, where the latter groups such as hydrogen consumers are strongly influencing the syntrophic bacteria (Gomez Camacho and Ruggeri 2018). Acetate produced in the process could be converted to methane either directly by methyl reduction or by following a two-step reaction, where acetate is first oxidized to CO2 and H2, by syntrophic acetate oxidation (SAO), and then this hydrogen is used to reduce CO2 into CH4 (Mulat et al. 2014). Heyer et al. (2019) studied the interactions between microorganisms and the metabolic interchangeability of the different microorganisms. This work suggested that under specific anaerobic digestion conditions the thermodynamic equilibrium of CO2, H2, and acetate will decide on the metabolic pathway shift, either between SAO or homoacetogenesis. Understanding this will explain how some archaeal species have major enzyme affinity on acetate and could suppress other acetate-consuming phylotypes. Thus, competition on the substrate between certain microorganisms such as Methanosaetaceae may kill or suppress other species due to the expression of bacteriocins, which inhibit the competitor (Heyer et al. 2019). The H2 produced by 98 A. G. Talavera-Caro et al. non-methanogenic syntrophic microorganisms from key fermentation products (eth- anol and C2 and greater than C2 volatile fatty acids) are reduced to methane by hydrogenotrophic methanogens or to H2S by sulfate-reducing bacteria. This inter- species microbial exchange of hydrogen suggests that syntrophs are incapable of independently oxidizing alcohols and C2 and greater than C2 volatile fatty acids under anaerobic conditions and need a partner that consumes hydrogen to keep the partial pressure of hydrogen under control and facilitate their metabolic activity. Syntrophic interactions consist generally on the intercellular transport of reducing equivalents, like H2 and/or formate, coupled with H2/formate consumers, also referred as interspecies hydrogen transfer (IHT) (Shrestha and Rotaru 2014; Summers et al. 2010). In addition, formate often serves as a substitute for H2 in interspecies electron transfer. The electron reduced carriers on this type of mechanism are additionally regenerated to an oxidized state (Shrestha and Rotaru 2014; Kouzuma et al. 2015). Westerholm et al. (2016) reported that SAO bacteria are principally classified in the group of homoacetogens, which perform the Wood Ljungdahl (WL) pathway during growth, in presence of autotrophic and/or heterotrophic substrates, and produce acetate as main by-product. In this pathway, they suggest that the gene fhs encodes the enzyme formyl tetrahydrofolate synthetase (FTHFS) to catalyze the ATP-dependent activation of formate, postulating as well, the reverse WL perfor- mance for acetate oxidation. Interestingly, the Pseudothermotoga lettingae acetate oxidizer can combine the methyl branch of the latter pathway with a glycine cleavage system (Westerholm et al. 2016). Recent discoveries reported that some bacteria could directly transfer electrons to methanogens, as a unique cell-to-cell electron transfer mechanism, in a thermody- namically efficient manner (Cheng and Call 2016). The electron transfer, between microorganisms mediating electron carriers, is referred as direct interspecies electron transport (DIET), where three mechanisms have recently been suggested: (1) the via conductive pili, by association from two bacteria with a conductive pili (conductive nanowires), (2) the membrane-bound mechanism with electron transport proteins, by electron transmission which represents close cell connections by using a multiheme outer surface cytochrome (OmcZ), and finally (3), the more recently studied the magnetite particle which form chains for electrically connecting cells involved in DIET (Park et al. 2018). 4.4.6 Methanogenesis Methanogenesis is one of the critical steps in the process of AD, characterized by slow reaction rate on the energy workflow. Syntrophic interactions between acetogenic bacteria and methanogens as well as methyl group reduction are essential to CH4 production. Thus, understanding the microbial community involved in electron trans- ferring dynamics is key to biogas production improvement. Archaeal methanogens are dominant groups in this phase, generally performing the aceticlastic, hydrogenotrophic 4 Proteomics of Lignocellulosic Substrates Bioconversion in Anaerobic Digesters. . . 99 and methylotrophic pathways. The microorganisms usually found are the strict hydrogenotrophic Methanomicrobiales, Methanobacteriales, Methanococcales, and the acetate-utilizing microorganisms Methanosarcinales with specific predominance ofMethanosarcina andMethanosaeta. Methanogens are quite sensitive for changes in environmental and operational conditions of AD process and many factors (i.e., high concentrations of volatile fatty acids, ammonium, sulfide, sodium an heavy metals) could inhibit the process (Al Seadi et al. 2008; Ziganshin et al. 2016). As mentioned earlier, methane can be produced either by hydrogenotrophic pathway by reducing CO2 using hydrogen, the methylotrophic pathway where methylated compounds like methanol or methylamine are reduced, and in acetoclastic pathway, methyl group of acetate is directly reduced to methane. Recently, the class of Thermoplasmata has been described to be capable to reduce methanol with H2 and may use methylamines as well, suggesting that methanogenic diversity could be higher (Wintsche et al. 2018). From the above three pathways, studies were performed using isotope assays to know the metabolic contribution from each one. It was reported that syntrophic acetogenic process and hydrogenotrophic methanogenesis accounted for 41 and 50% of methane formation at 37 C and 55 C, respectively (Yin et al. 2018). Another study using isotopes indicated that the non-aceticlastic oxidizers performed, approximately 80% of the pathway of the acetate decomposition in the reactor, which indicated the role of syntrophic acetate-oxidizing bacteria. Pseudothermotoga lettingae (previously as Termotoga lettingae) strain was reported to show syntrophic acetate oxidizing activity without sulfate ions and under co-culture conditions in relation with hydrogenotrophic methanogens (Sasaki et al. 2011). Identification of the pathways and enzymes involved in the methanogenesis networks via metaproteomic approach has been carried out (Table 4.2). During the hydrogenotrophic pathway, CO2 is reduced to methane through the intermediates formyl, methylene, and methyl. These residues are transferred to the coenzyme M, forming a methyl-CoM molecule further reduced to CH4 by the key methyl coen- zyme M reductase (MCR). Meanwhile, the energetically coenzyme F420 acts as an electron acceptor for hydrogenase, formate dehydrogenase, and carbon monoxide dehydrogenase, as well as donor electron for reductase NADP+. Moreover, this coenzyme utilize H2 and formic acid, as electron donor to produce the methane by CO2 reduction (Jia et al. 2017b). In the aceticlastic pathway, methyl group of acetate is reduced to methane by methyl reductase enzyme. In the case of methylotrophic pathway, the methyl groups are transferred to a methanol-specific corrinoid protein, then reduced by the MCR (Jia et al. 2017b; Guo et al. 2015). It is important to highlight that methyl-coenzyme M (methyl-CoM) reductase is active for all the three pathways (Hanreich et al. 2012). In a full-scale BGP, CODH/ACS and energy-converting hydrogenase (Ech), pro- teins, and enzymes involved inmetabolismofmethanol andmethylated amines, aswell as, V-type ATP synthase for energy conversion were assigned to Methanosarcinales. While F420-dependentN5,N10-methyleneH4MPT reductase (Mer),Mtr, and F420 reduc- ing hydrogenase (Frh) were assigned toMethanobacteriales (Heyer et al. 2016). In the hydrogenotrophic pathway, Lü et al. (2014) found proteins from strains of 100 A. G. Talavera-Caro et al. Methanothermobacter as H2-forming methylene H4MPT dehydrogenase, F420-depen- dent methyleneH4MPT dehydrogenase, Mtr, Mer, MCR, Frh, and heterodisulfide reductase. As well as from the methylotrophic pathway, monomethylamine methyltransferase, and a large subunit of the corrinoid/iron–sulfur protein and methylcobamide:CoMmetyltransferase were detected. In contrast, none of acetoclastic enzymes were identified in this study. Furthermore, trace elements on the anaerobic digestion have an impact on the performance process to carry out cell metabolisms and is critical to the final stage of the methane yield. For methanogens, the presence of Fe, Zn, Ni, Cu, Co, Mo, and Mn are essential. First, Fe, is important in stimulant as growth factor and formation of cytochromes and ferroxins vital for energy metabolism. Additionally, trace elements form the active site in metalloproteins, act as a cofactor and give the structure. Enzymes such as Mtr and MCR require Co and a nickel-containing cofactor F430 in their active sites, respectively (Choong et al. 2016). Trace element deprivation has shown to decrease hydrogenotrophic metaproteins abundance from Methanomicrobiales and methylotrophic and aceticlastic metaproteins from Methanosarcinales. However, trace elements may have negative impact when present in high concentrations. It has been demonstrated to cause enzyme disruptions and changes on functional structure, and microbial composition as well. Therefore, effectiveness of the anaerobic digestion performance by using trace elements will further depend on its optimum bioavailability fraction (Bourven et al. 2017). 4.5 Stress Responses and Biomarkers Although AD is an economic way of waste management combining with renewable energy production in the form of methane, the process has certain thresholds and one of them is high sensitivity to the presence of certain substances at high concentrations during the process. Most frequently, a reactor turns “sour” due to accumulation of volatile fatty acids (VFAs). Further, ammonia, high partial pressure of H2 have recorded negative effects on AD process, among several other factors (Chen et al. 2014). Therefore, several studies have focused on efforts to overcome stress condi- tions by detecting different key enzymes to those conditions in the reactor (Table 4.3). Formerly, when the process contains simple sugars, which are easy to degrade, VFAs are accumulated and decrease in pH results in the imbalance of AD process. Interestingly, under this condition, Theuerl et al. (2015) could detect proteins involved on aceticlastic and hydrogenotrophic methanogenesis from Methanosarcinales, Methanobacteriales, and Methanomicrobiales, with an abundance ranging from 55 to 77%. On the other hand, high concentrations of total ammonia nitrogen due to the high rate of protein hydrolysis could negatively affect reactor operation. Nevertheless, lignocellulosic-rich matter is a convenient substrate to slow the rates of process start- up, diminishing the possible ammonia accumulation. Also, trace elements have demonstrated to overcome such problems. It has demonstrated that nickel which is contained in coenzyme F430 enhanced methane potential and overcome such 4 Proteomics of Lignocellulosic Substrates Bioconversion in Anaerobic Digesters. . . 101 problems (Capson-Tojo et al. 2017). Additionally, under concentrations of 6–7 g-N/ L ammonium, inhibition of Methanomicrobiales and Methanosaetacea activities was observed along with a shift in microbial community (Lü et al. 2014). The lignocellulosic-rich feedstocks with high lignin content could also cause inhibitory stress conditions. The negative impacts on these types of substrate, especially due to the by-products of lignin decomposition, inhibitory aromatic compounds. These are mainly composed of furanic acid (5-HMF, fufural) and phenolic compounds leading to a metabolic shift of the H2-producing pathways to a nonproducing pathway (Monlau et al. 2014). However, several numbers of enzymes are involved in the catabolic pathway of lignocellulosic compounds. The Table 4.3 General overview on protein abundance under varying environmental and operational conditions Reactor Key enzyme/ biomarkers Functional role RA Species related to expression References Lab-scale anaerobic moving bed reactor at 37 C Cofactor F430 Nickel hydrocorrinoid pros- thetic group of the methyl-CoM reduc- tase. It decreases the toxic effect of higher VFAs content and increases methane production ++ Methanococcus jannaschii Methanococcus maripaludis Methanococcus vaneilii Passaris et al. (2018) Lab-scale mesophilic digester treating starch Methyl coenzyme-M reductase It reduces the methyl CoM with hydrogen for methane production + Methanobacterium Methanosaeta Zhang et al. (2018) Low-temper- ature granu- lar sludge reactor oper- ated at 15 C aOxygen-sen- sitive alcohol dehydrogenase Interconversion from alcohol to acetaldehyde ++ P. propionicus Abram et al. (2011) aTransketolase Constitutes the reversible link between glycolysis and pentose phos- phate pathways + + + Sequencing batch reactor with alter- ations in phosphorus level aPeroxiredoxin Protects the cell against reactive oxy- gen species ++ + Accumulibacter phosphatis Wilmes and Bond (2009) aThioredoxins/ chaperon proteins Responsible to maintain disulfide bonds within the cytoplasmatic pro- teins in a reduced state + + + RA Relative abundance aAbundance of proteins depending on the stress condition or inhibitory factor + low abundance, ++ medium abundance, +++ high abundance 102 A. G. Talavera-Caro et al. benzoyl-CoA reductase class II (BamBCDEFGHI) could function as a functional marker, which is expressed and detected when mono-aromatic compounds are degraded (von Netzer et al. 2016). A well-studied biomarker to understand methane production is the expression of mcrA gene, which encodes the key enzyme methyl coenzyme M reductase (MCR). MCR catalyses the last step of methanogenesis and related to methane yield. This biomarker can provide useful information and by monitoring its activity functional performance of a biodigester can be studied (Morris et al. 2014). Expression ofmcrA has been found related to the presence of concentration of volatile fatty acid such as acetate and propionate (Aguinaga Casañas et al. 2015). However, new protein biomarkers are drawing attention, as proteins can be synthesized and then folded immediately after a stimulus, which could vary under different conditions. Hence, protein identification with potential in biomarker fingerprint could be revealing tool for instant physiological responses (Lacerda et al. 2007). 4.6 Challenges and Future Perspectives Metaproteomics analysis is the most recently developed tool, which indicates the protein assignment to specific microorganisms and contributes to understand the relationship between phylogenetic analyses and the proteome (Abram et al. 2011). However, its application has not been completely exploited since there is still more “black holes” and methodological challenges to solve the whole metabolic pathways (Wilmes et al. 2015). Principal problems are related to methodological issues involved in the correct isolation of the proteome, since humic acids and their impurities make the quanti- fication and separation of the proteins very challenging. Humic acids are commonly present in environmental samples and bind to the proteins which hamper protein separation (Wilmes and Bond 2004). Fractioning proteins by gel separation, also involves several difficulties, as poor separation of acidic, alkaline, and hydrophobic proteins and a low load capacity, which affects the resolution and analysis of the gel (Li et al. 2016). Secondly, peptides’ identification is one of the most challenging steps in metaproteomics. Some proteins’ sequences are among the most highly conserved across different microorganisms, which entails the same peptide sequences from several different species. These leads to have many proteins identified in the same information storage and processing group. Besides, MS/MS data uses the top of the 10–20 most abundant peptide ions acquired and does not cover all available peptide ions, which misses a plenty of valuable information (Heyer et al. 2017). Thus, metaproteomic approaches have to overcome the cleanup and characterization of peptides and proteins by intensive purification and pre-fractionation methods (Wenzel et al. 2018). The key to the metaproteomic studies is the resolution level of the peptide identification. In future, considerations on the use of the best algorithm which 4 Proteomics of Lignocellulosic Substrates Bioconversion in Anaerobic Digesters. . . 103 could identify the proteins covered on the database to evaluate for de novo results (Heyer et al. 2015). The molecular techniques could provide the microbial informa- tion and linking the species on the proteomic data by complementary studies of the microbial consortium, could aid to overcome the challenges encountered in the coverage of the samples. Further, improvement in methodologies to identify proteins along with improvement in methods of data analyses are essential to apply metaproteomics as an effective tool to understand the microbial diversity and their function of biodigesters (Herbst et al. 2016). References Abendroth C, Simeonov C, Peretó J, Antúnez O, Gavidia R, Luschnig O, Porcar M (2017) From grass to gas: microbiome dynamics of grass biomass acidification under mesophilic and thermophilic temperatures. Biotechnol Biofuels 10:171. https://doi.org/10.1186/s13068-017- 0859-0 Abram F, Gunnigle E, O’Flaherty V (2009) Optimisation of protein extraction and 2-DE for metaproteomics of microbial communities from anaerobic wastewater treatment biofilms. Electrophoresis 30:4149–4151. https://doi.org/10.1002/elps.200900474 Abram F, Enright A-M, O’Reilly J, Botting CH, Collins G, O’Flaherty V (2011) A metaproteomic approach gives functional insights into anaerobic digestion. J Appl Microbiol 110:1550–1560. https://doi.org/10.1111/j.1365-2672.2011.05011.x Adekunle KF, Okolie JA (2015) A review of biochemical process of anaerobic digestion. Adv Biosci Biotechnol 06:205. https://doi.org/10.4236/abb.2015.63020 Aguinaga Casañas MA, Rangkasenee N, Krattenmacher N, Thaller G, Metges CC, Kuhla B (2015) Methyl-coenzyme M reductase A as an indicator to estimate methane production from dairy cows. J Dairy Sci 98:4074–4083. https://doi.org/10.3168/jds.2015-9310 Ahring BK, Biswas R, Ahamed A, Teller PJ, Uellendahl H (2015) Making lignin accessible for anaerobic digestion by wet-explosion pretreatment. Bioresour Technol 175:182–188. https:// doi.org/10.1016/j.biortech.2014.10.082 Akuzawa M, Hori T, Haruta S, Ueno Y, Ishii M, Igarashi Y (2011) Distinctive responses of metabolically active microbiota to acidification in a thermophilic anaerobic digester. Microb Ecol 61:595–605. https://doi.org/10.1007/s00248-010-9788-1 Al Seadi T, Rutz D, Prassl H, Köttner M, Finsterwalder T, Volk S, Janssen R (2008) Biogas handbook. University of Southern Denmark Esbjerg, Esbjerg Angelidaki I, Ellegaard L, Ahring BK (2003) Applications of the anaerobic digestion process. Adv Biochem Eng Biotechnol 82:1–33. https://doi.org/10.1007/3-540-45838-7_1 Angelidaki I, Boe K, Ellegaard L (2005) Effect of operating conditions and reactor configuration on efficiency of full-scale biogas plants. Water Sci Technol 52:189–194. https://doi.org/10.2166/ wst.2005.0516 Angelidaki I, Karakashev D, Batstone DJ, Plugge CM, Stams AJM (2011) Biomethanation and its potential. Methods Enzymol 494:327–351. https://doi.org/10.1016/B978-0-12-385112-3. 00016-0 Appels L, Lauwers J, Degrève J, Helsen L, Lievens B, Willems K, Van Impe J, Dewil R (2011) Anaerobic digestion in global bio-energy production: potential and research challenges. Renew Sust Energ Rev 15:4295–4301. https://doi.org/10.1016/j.rser.2011.07.121 Ariunbaatar J, Panico A, Esposito G, Pirozzi F, Lens PNL (2014) Pretreatment methods to enhance anaerobic digestion of organic solid waste. Appl Energy 123:143–156. https://doi.org/10.1016/ j.apenergy.2014.02.035 104 A. G. Talavera-Caro et al. Barua S, Dhar BR (2017) Advances towards understanding and engineering direct interspecies electron transfer in anaerobic digestion. Bioresour Technol 244:698–707. https://doi.org/10. 1016/j.biortech.2017.08.023 Bize A, Cardona L, Desmond-Le Quéméner E, Battimelli A, Badalato N, Bureau C, Madigou C, Chevret D, Guillot A, Monnet V, Godon J-J, Bouchez T (2015) Shotgun metaproteomic profiling of biomimetic anaerobic digestion processes treating sewage sludge. Proteomics 15:3532–3543. https://doi.org/10.1002/pmic.201500041 Bourven I, Casellas M, Buzier R, Lesieur J, Lenain J-F, Faix A, Bressolier P, Maftah C, Guibaud G (2017) Potential of DGT in a new fractionation approach for studying trace metal element impact on anaerobic digestion: the example of cadmium. Int Biodeterior Biodegrad 119:188–195. https://doi.org/10.1016/j.ibiod.2016.11.007 Bräsen C, Esser D, Rauch B, Siebers B (2014) Carbohydrate metabolism in archaea: current insights into unusual enzymes and pathways and their regulation. Microbiol Mol Biol Rev 78:89–175. https://doi.org/10.1128/MMBR.00041-13 Braun R, Weiland P, Wellinger A (2008) Biogas from energy crop digestion. IEA Bioenergy Task 37:1–20 Capson-Tojo G, Ruiz D, Rouez M, Crest M, Steyer J-P, Bernet N, Delgenès J-P, Escudié R (2017) Accumulation of propionic acid during consecutive batch anaerobic digestion of commercial food waste. Bioresour Technol 245:724–733. https://doi.org/10.1016/j.biortech.2017.08.149 Carrere H, Antonopoulou G, Affes R, Passos F, Battimelli A, Lyberatos G, Ferrer I (2016) Review of feedstock pretreatment strategies for improved anaerobic digestion: from lab-scale research to full-scale application. Bioresour Technol 199:386–397. https://doi.org/10.1016/j.biortech.2015. 09.007 Cesarino I, Araújo P, Domingues Júnior AP, Mazzafera P (2012) An overview of lignin metabolism and its effect on biomass recalcitrance. Braz J Bot 35:303–311. https://doi.org/10.1590/S0100- 84042012000400003 Chen Y, Cheng JJ, Creamer KS (2008) Inhibition of anaerobic digestion process: a review. Bioresour Technol 99:4044–4064. https://doi.org/10.1016/j.biortech.2007.01.057 Chen JL, Ortiz R, Steele TWJ, Stuckey DC (2014) Toxicants inhibiting anaerobic digestion: a review. Biotechnol Adv 32:1523–1534. https://doi.org/10.1016/j.biotechadv.2014.10.005 Cheng Q, Call DF (2016) Hardwiring microbes via direct interspecies electron transfer: mecha- nisms and applications. Environ Sci Process Impacts 18:968–980. https://doi.org/10.1039/ C6EM00219F Choong YY, Norli I, Abdullah AZ, Yhaya MF (2016) Impacts of trace element supplementation on the performance of anaerobic digestion process: a critical review. Bioresour Technol 209:369–379. https://doi.org/10.1016/j.biortech.2016.03.028 Clement BG, Kehl LE, DeBord KL, Kitts CL (1998) Terminal restriction fragment patterns (TRFPs), a rapid, PCR-based method for the comparison of complex bacterial communities. J Microbiol Methods 31:135–142. https://doi.org/10.1016/S0167-7012(97)00105-X Deublein S (2009) Biogas from waste and renewable resources: an introduction. Choice Rev Online 46:2682–2682. https://doi.org/10.5860/CHOICE.46-2682 Divya D, Gopinath LR, Merlin Christy P (2015) A review on current aspects and diverse prospects for enhancing biogas production in sustainable means. Renew Sust Energ Rev 42:690–699. https://doi.org/10.1016/j.rser.2014.10.055 Felix CR, Ljungdahl LG (1993) The cellulosome: the exocellular organelle of Clostridium. Annu Rev Microbiol 47:791–819. https://doi.org/10.1146/annurev.mi.47.100193.004043 Ferry JG (1993) Fermentation of acetate. In: Ferry JG (ed) Methanogenesis: ecology, physiology, biochemistry and genetics. Springer, Boston, pp 304–334 Gerardi MH (2003) The microbiology of anaerobic digesters. Wiley-Intersciences, Canada Gomez Camacho CE, Ruggeri B (2018) Syntrophic microorganisms interactions in anaerobic digestion (AD): a critical review in the light of increase the energy production. Chem Eng Trans 64:391–396. https://doi.org/10.3303/CET1864066 4 Proteomics of Lignocellulosic Substrates Bioconversion in Anaerobic Digesters. . . 105 Gujer W, Zehnder AJB (1983) Conversion processes in anaerobic digestion. Water Sci Technol 15:127–167. https://doi.org/10.2166/wst.1983.0164 Guo J, Peng Y, Ni B-J, Han X, Fan L, Yuan Z (2015) Dissecting microbial community structure and methane-producing pathways of a full-scale anaerobic reactor digesting activated sludge from wastewater treatment by metagenomic sequencing. Microb Cell Factories 14:33. https://doi.org/ 10.1186/s12934-015-0218-4 Hagen LH, Frank JA, Zamanzadeh M, Eijsink VG, Pope PB, Horn SJ, Arntzen MØ (2017) Quantitative metaproteomics highlight the metabolic contributions of uncultured phylotypes in a thermophilic anaerobic digester. Appl Env Microbiol 83:e01955–e01916. https://doi.org/ 10.1128/AEM.01955-16 Hanreich A, Heyer R, Benndorf D, Rapp E, Pioch M, Reichl U, Klocke M (2012) Metaproteome analysis to determine the metabolically active part of a thermophilic microbial community producing biogas from agricultural biomass. Can J Microbiol 58:917–922. https://doi.org/10. 1139/w2012-058 Hanreich A, Schimpf U, Zakrzewski M, Schlüter A, Benndorf D, Heyer R, Rapp E, Pühler A, Reichl U, Klocke M (2013) Metagenome and metaproteome analyses of microbial communities in mesophilic biogas-producing anaerobic batch fermentations indicate concerted plant carbo- hydrate degradation. Syst Appl Microbiol 36:330–338. https://doi.org/10.1016/j.syapm.2013. 03.006 Hassa J, Maus I, Off S, Pühler A, Scherer P, Klocke M, Schlüter A (2018) Metagenome, metatranscriptome, and metaproteome approaches unraveled compositions and functional rela- tionships of microbial communities residing in biogas plants. Appl Microbiol Biotechnol 102:5045–5063. https://doi.org/10.1007/s00253-018-8976-7 Herbst F-A, Lünsmann V, Kjeldal H, Jehmlich N, Tholey A, von Bergen M, Nielsen JL, Hettich RL, Seifert J, Nielsen PH (2016) Enhancing metaproteomics—the value of models and defined environmental microbial systems. Proteomics 16:783–798. https://doi.org/10.1002/pmic. 201500305 Herrmann C, Heiermann M, Idler C (2011) Effects of ensiling, silage additives and storage period on methane formation of biogas crops. Bioresour Technol 102:5153–5161. https://doi.org/10. 1016/j.biortech.2011.01.012 Heyer R, Kohrs F, Benndorf D, Rapp E, Kausmann R, Heiermann M, Klocke M, Reichl U (2013) Metaproteome analysis of the microbial communities in agricultural biogas plants. New Biotechnol 30:614–622. https://doi.org/10.1016/j.nbt.2013.01.002 Heyer R, Kohrs F, Reichl U, Benndorf D (2015) Metaproteomics of complex microbial commu- nities in biogas plants. Microb Biotechnol 8:749–763. https://doi.org/10.1111/1751-7915. 12276 Heyer R, Benndorf D, Kohrs F, De Vrieze J, Boon N, Hoffmann M, Rapp E, Schlüter A, Sczyrba A, Reichl U (2016) Proteotyping of biogas plant microbiomes separates biogas plants according to process temperature and reactor type. Biotechnol Biofuels 9:155. https://doi.org/10.1186/ s13068-016-0572-4 Heyer R, Schallert K, Zoun R, Becher B, Saake G, Benndorf D (2017) Challenges and perspectives of metaproteomic data analysis. J Biotechnol 261:24–36. https://doi.org/10.1016/j.jbiotec.2017. 06.1201 Heyer R, Schallert K, Siewert C, Kohrs F, Greve J, Maus I, Klang J, Klocke M, Heiermann M, Hoffmann M, Püttker S, Calusinska M, Zoun R, Saake G, Benndorf D, Reichl U (2019) Metaproteome analysis reveals that syntrophy, competition, and phage-host interaction shape microbial communities in biogas plants. Microbiome 7:69. https://doi.org/10.1186/s40168-019- 0673-y Inyang M, Gao B, Pullammanappallil P, Ding W, Zimmerman AR (2010) Biochar from anaerobi- cally digested sugarcane bagasse. Bioresour Technol 101:8868–8872. https://doi.org/10.1016/j. biortech.2010.06.088 Isikgor FH, Becer CR (2015) Lignocellulosic biomass: a sustainable platform for the production of bio-based chemicals and polymers. Polym Chem 6:4497–4559. https://doi.org/10.1039/ c5py00263j 106 A. G. Talavera-Caro et al. Jain S, Jain S, Tim Wolf I, Lee J, Wah Tong Y (2015) A comprehensive review on operating parameters and different pretreatment methodologies for anaerobic digestion of municipal solid waste. Renew Sust Energ Rev 52:142–154. https://doi.org/10.1016/j.rser.2015.07.091 Jia X, Xi B-D, Li M-X, Yang Y, Wang Y (2017a) Metaproteomics analysis of the functional insights into microbial communities of combined hydrogen and methane production by anaer- obic fermentation from reed straw. PLoS One 12:e0183158. https://doi.org/10.1371/journal. pone.0183158 Jia X, Xi B, Li M, Liu D, Hou J, Hao Y, Meng F (2017b) Metaproteomic analysis of the relationship between microbial community phylogeny, function and metabolic activity during biohydrogen- methane coproduction under short-term hydrothermal pretreatment from food waste. Bioresour Technol 245:1030–1039. https://doi.org/10.1016/j.biortech.2017.08.180 Jiang Y, Xin F, Lu J, Dong W, Zhang W, Zhang M, Wu H, Ma J, Jiang M (2017) State of the art review of biofuels production from lignocellulose by thermophilic bacteria. Bioresour Technol 245:1498–1506. https://doi.org/10.1016/j.biortech.2017.05.142 Joyce A, Ijaz UZ, Nzeteu C, Vaughan A, Shirran SL, Botting CH, Quince C, O’Flaherty V, Abram F (2018) Linking microbial community structure and function during the acidified anaerobic digestion of grass. Front Microbiol 9:540. https://doi.org/10.3389/fmicb.2018.00540 Kanwal S, Chaudhry N, Munir S, Sana H (2019) Effect of torrefaction conditions on the physico- chemical characterization of agricultural waste (sugarcane bagasse). Waste Manag 88:280–290. https://doi.org/10.1016/j.wasman.2019.03.053 Karlsson R, Gonzales-Siles L, Boulund F, Svensson-Stadler L, Skovbjerg S, Karlsson A, Davidson M, Hulth S, Kristiansson E, Moore ERB (2015) Proteotyping: proteomic character- ization, classification and identification of microorganisms – a prospectus. Syst Appl Microbiol 38:246–257. https://doi.org/10.1016/j.syapm.2015.03.006 Kohrs F, Heyer R, Magnussen A, Benndorf D, Muth T, Behne A, Rapp E, Kausmann R, Heiermann M, Klocke M, Reichl U (2014) Sample prefractionation with liquid isoelectric focusing enables in depth microbial metaproteome analysis of mesophilic and thermophilic biogas plants. Anaerobe 29:59–67. https://doi.org/10.1016/j.anaerobe.2013.11.009 Kohrs F, Heyer R, Bissinger T, Kottler R, Schallert K, Püttker S, Behne A, Rapp E, Benndorf D, Reichl U (2017) Proteotyping of laboratory-scale biogas plants reveals multiple steady-states in community composition. Anaerobe 46:56–68. https://doi.org/10.1016/j.anaerobe.2017.02.005 Kouzuma A, Kato S, Watanabe K (2015) Microbial interspecies interactions: recent findings in syntrophic consortia. Front Microbiol 6:477. https://doi.org/10.3389/fmicb.2015.00477 Krause L, Diaz NN, Edwards RA, Gartemann K-H, Krömeke H, Neuweger H, Pühler A, Runte KJ, Schlüter A, Stoye J, Szczepanowski R, Tauch A, Goesmann A (2008) Taxonomic composition and gene content of a methane-producing microbial community isolated from a biogas reactor. J Biotechnol 136:91–101. https://doi.org/10.1016/j.jbiotec.2008.06.003 Lacerda CMR, Choe LH, Reardon KF (2007) Metaproteomic analysis of a bacterial community response to cadmium exposure. J Proteome Res 6:1145–1152. https://doi.org/10.1021/ pr060477v Li Y, Zhang R, Liu X, Chen C, Xiao X, Feng L, He Y, Liu G (2013a) Evaluating methane production from anaerobic mono- and co-digestion of kitchen waste, corn stover, and chicken manure. Energy Fuel 27:2085–2091. https://doi.org/10.1021/ef400117f Li Y, Zhang R, Liu G, Chen C, He Y, Liu X (2013b) Comparison of methane production potential, biodegradability, and kinetics of different organic substrates. Bioresour Technol 149:565–569. https://doi.org/10.1016/j.biortech.2013.09.063 Li R, Wu Z, Wangb Y, Ding L, Wang Y (2016) Role of pH-induced structural change in protein aggregation in foam fractionation of bovine serum albumin. Biotechnol Rep 9:46–52. https:// doi.org/10.1016/j.btre.2016.01.002 Li W, Khalid H, Zhu Z, Zhang R, Liu G, Chen C, Thorin E (2018) Methane production through anaerobic digestion: participation and digestion characteristics of cellulose, hemicellulose and lignin. Appl Energy 226:1219–1228. https://doi.org/10.1016/j.apenergy.2018.05.055 4 Proteomics of Lignocellulosic Substrates Bioconversion in Anaerobic Digesters. . . 107 Lin Y-W, Tuan N, Huang S-L (2016) Metaproteomic analysis of the microbial community present in a thermophilic swine manure digester to allow functional characterization: a case study. Int Biodeterior Biodegrad 115:64–73. https://doi.org/10.1016/j.ibiod.2016.06.013 Liu Z-H, Chen H-Z (2015) Xylose production from corn stover biomass by steam explosion combined with enzymatic digestibility. Bioresour Technol 193:345–356. https://doi.org/10. 1016/j.biortech.2015.06.114 Lü F, Bize A, Guillot A, Monnet V, Madigou C, Chapleur O, Mazéas L, He P, Bouchez T (2014) Metaproteomics of cellulose methanisation under thermophilic conditions reveals a surprisingly high proteolytic activity. ISME J 8:88–102. https://doi.org/10.1038/ismej.2013.120 Madigan M, Martinko J, Dunlap PV, Clark DP (2008) Brock biology of microorganisms. Int Microbiol 11:65–73 Mata-Alvarez J, Dosta J, Romero-Güiza MS, Fonoll X, Peces M, Astals S (2014) A critical review on anaerobic co-digestion achievements between 2010 and 2013. Renew Sust Energ Rev 36:412–427. https://doi.org/10.1016/j.rser.2014.04.039 Mcinerney M, Bryant MP, Pfennig N (1979) Anaerobic bacterium that degrades fatty acids in syntrophic association with methanogens. Arch Microbiol 122:129–135. https://doi.org/10. 1007/BF00411351 Monlau F, Sambusiti C, Barakat A, Quéméneur M, Trably E, Steyer J-P, Carrère H (2014) Do furanic and phenolic compounds of lignocellulosic and algae biomass hydrolyzate inhibit anaerobic mixed cultures? A comprehensive review. Biotechnol Adv 32:934–951. https://doi. org/10.1016/j.biotechadv.2014.04.007 Morris R, Schauer-Gimenez A, Bhattad U, Kearney C, Struble CA, Zitomer D, Maki JS (2014) Methyl coenzymeM reductase (mcrA) gene abundance correlates with activity measurements of methanogenic H2/CO2-enriched anaerobic biomass. Microb Biotechnol 7:77–84. https://doi. org/10.1111/1751-7915.12094 Mulat DG, Ward AJ, Adamsen APS, Voigt NV, Nielsen JL, Feilberg A (2014) Quantifying contribution of synthrophic acetate oxidation to methane production in thermophilic anaerobic reactors by membrane inlet mass spectrometry. Environ Sci Technol 140130145609003. https:// doi.org/10.1021/es403144e Mustafa AM, Li H, Radwan AA, Sheng K, Chen X (2018) Effect of hydrothermal and Ca(OH)2 pretreatments on anaerobic digestion of sugarcane bagasse for biogas production. Bioresour Technol 259:54–60. https://doi.org/10.1016/j.biortech.2018.03.028 Nallathambi Gunaseelan V (1997) Anaerobic digestion of biomass for methane production: a review. Biomass Bioenergy 13:83–114. https://doi.org/10.1016/S0961-9534(97)00020-2 Ortseifen V, Stolze Y, Maus I, Sczyrba A, Bremges A, Albaum SP, Jaenicke S, Fracowiak J, Pühler A, Schlüter A (2016) An integrated metagenome and-proteome analysis of the microbial community residing in a biogas production plant. J Biotechnol 231:268–279. https://doi.org/10. 1016/j.jbiotec.2016.06.014 Park J, Lee B, Shi P, Kwon H, Jeong SM, Jun H (2018) Methanol metabolism and archaeal community changes in a bioelectrochemical anaerobic digestion sequencing batch reactor with copper-coated graphite cathode. Bioresour Technol 259:398–406. https://doi.org/10. 1016/j.biortech.2018.03.009 Passaris I, Van Gaelen P, Cornelissen R, Simoens K, Grauwels D, Vanhaecke L, Springael D, Smets I (2018) Cofactor F430 as a biomarker for methanogenic activity: application to an anaerobic bioreactor system. Appl Microbiol Biotechnol 102:1191–1201. https://doi.org/10. 1007/s00253-017-8681-y Paul S, Dutta A (2018) Challenges and opportunities of lignocellulosic biomass for anaerobic digestion. Resour Conserv Recycl 130:164–174. https://doi.org/10.1016/j.resconrec.2017.12.005 Pérez-Rodríguez N, García-Bernet D, Domínguez JM (2016) Effects of enzymatic hydrolysis and ultrasounds pretreatments on corn cob and vine trimming shoots for biogas production. Bioresour Technol 221:130–138. https://doi.org/10.1016/j.biortech.2016.09.013 Poudel BN, Paudel KP, Timilsina G, Zilberman D (2012) Providing numbers for a food versus fuel debate: an analysis of a future biofuel production scenario. Appl Econ Perspect Policy 34:637–668. https://doi.org/10.1093/aepp/pps039 108 A. G. Talavera-Caro et al. Rahman MA, Møller HB, Saha CK, Alam MM, Wahid R, Feng L (2017) Optimal ratio for anaerobic co-digestion of poultry droppings and lignocellulosic-rich substrates for enhanced biogas production. Energy Sustain Dev 39:59–66. https://doi.org/10.1016/j.esd.2017.04.004 Ramsay IR, Pullammanappallil PC (2001) Protein degradation during anaerobic wastewater treat- ment: derivation of stoichiometry. Biodegradation 12:247–256. https://doi.org/10.1023/ A:1013116728817 Rastogi G, Ranade DR, Yeole TY, Patole MS, Shouche YS (2008) Investigation of methanogen population structure in biogas reactor by molecular characterization of methyl-coenzyme M reductase A (mcrA) genes. Bioresour Technol 99:5317–5326. https://doi.org/10.1016/j. biortech.2007.11.024 Roseman S (1969) The transport of carbohydrates by a bacterial phosphotransferase system. J Gen Physiol 54:138–184. https://doi.org/10.1085/jgp.54.1.138 Sasaki D, Hori T, Haruta S, Ueno Y, Ishii M, Igarashi Y (2011) Methanogenic pathway and community structure in a thermophilic anaerobic digestion process of organic solid waste. J Biosci Bioeng 111:41–46. https://doi.org/10.1016/j.jbiosc.2010.08.011 Sawatdeenarunat C, Surendra KC, Takara D, Oechsner H, Khanal SK (2015) Anaerobic digestion of lignocellulosic biomass: challenges and opportunities. Bioresour Technol 178:178–186. https://doi.org/10.1016/j.biortech.2014.09.103 Schink B (1997) Energetics of syntrophic cooperation in methanogenic degradation. Microbiol Mol Biol Rev 61:262–280 Schlüter A, Bekel T, Diaz NN, Dondrup M, Eichenlaub R, Gartemann K-H, Krahn I, Krause L, Krömeke H, Kruse O, Mussgnug JH, Neuweger H, Niehaus K, Pühler A, Runte KJ, Szczepanowski R, Tauch A, Tilker A, Viehöver P, Goesmann A (2008) The metagenome of a biogas-producing microbial community of a production-scale biogas plant fermenter analysed by the 454-pyrosequencing technology. J Biotechnol 136:77–90. https://doi.org/10.1016/j. jbiotec.2008.05.008 Schneider T, Riedel K (2010) Environmental proteomics: analysis of structure and function of microbial communities. Proteomics 10:785–798. https://doi.org/10.1002/pmic.200900450 Schnürer A, Zellner G, Svensson BH (1999) Mesophilic syntrophic acetate oxidation during methane formation in biogas reactors. FEMS Microbiol Ecol 29:249–261. https://doi.org/10. 1111/j.1574-6941.1999.tb00616.x Shrestha PM, Rotaru A-E (2014) Plugging in or going wireless: strategies for interspecies electron transfer. Front Microbiol 5:237. https://doi.org/10.3389/fmicb.2014.00237 Siddique MNI, Wahid ZA (2018) Achievements and perspectives of anaerobic co-digestion: a review. J Clean Prod 194:359–371. https://doi.org/10.1016/j.jclepro.2018.05.155 Singh S, Cheng G, Sathitsuksanoh N, Wu D, Varanasi P, George A, Balan V, Gao X, Kumar R, Dale BE, Wyman CE, Simmons BA (2015) Comparison of different biomass pretreatment techniques and their impact on chemistry and structure. Front Energy Res 2:62. https://doi.org/ 10.3389/fenrg.2014.00062 Speda J, Jonsson B-H, Carlsson U, Karlsson M (2017) Metaproteomics-guided selection of targeted enzymes for bioprospecting of mixed microbial communities. Biotechnol Biofuels 10:128. https://doi.org/10.1186/s13068-017-0815-z Summers ZM, Fogarty HE, Leang C, Franks AE, Malvankar NS, Lovley DR (2010) Direct exchange of electrons within aggregates of an evolved syntrophic coculture of anaerobic bacteria. Science 330:1413–1415. https://doi.org/10.1126/science.1196526 Theuerl S, Kohrs F, Benndorf D, Maus I, Wibberg D, Schlüter A, Kausmann R, Heiermann M, Rapp E, Reichl U, Pühler A, Klocke M (2015) Community shifts in a well-operating agricultural biogas plant: how process variations are handled by the microbiome. Appl Microbiol Biotechnol 99:7791–7803. https://doi.org/10.1007/s00253-015-6627-9 Tong X, Smith LH, McCarty PL (1990) Methane fermentation of selected lignocellulosic materials. Biomass 21:239–255. https://doi.org/10.1016/0144-4565(90)90075-U 4 Proteomics of Lignocellulosic Substrates Bioconversion in Anaerobic Digesters. . . 109 Vanwonterghem I, Jensen PD, Ho DP, Batstone DJ, Tyson GW (2014) Linking microbial com- munity structure, interactions and function in anaerobic digesters using new molecular tech- niques. Curr Opin Biotechnol 27:55–64. https://doi.org/10.1016/j.copbio.2013.11.004 von Netzer F, Kuntze K, Vogt C, Richnow HH, Boll M, Lueders T (2016) Functional gene markers for fumarate-adding and dearomatizing key enzymes in anaerobic aromatic hydrocarbon deg- radation in terrestrial environments. J Mol Microbiol Biotechnol 26:180–194. https://doi.org/10. 1159/000441946 Wang D, Liu Y, Ngo HH, Zhang C, Yang Q, Peng L, He D, Zeng G, Li X, Ni B-J (2017) Approach of describing dynamic production of volatile fatty acids from sludge alkaline fermentation. Bioresour Technol 238:343–351. https://doi.org/10.1016/j.biortech.2017.04.054 Weiland P (2010) Biogas production: current state and perspectives. Appl Microbiol Biotechnol 85:849–860. https://doi.org/10.1007/s00253-009-2246-7 Wenzel L, Heyer R, Schallert K, Löser L, Wünschiers R, Reichl U, Benndorf D (2018) SDS-PAGE fractionation to increase metaproteomic insight into the taxonomic and functional composition of microbial communities for biogas plant samples. Eng Life Sci 18:498–509. https://doi.org/10. 1002/elsc.201800062 Westerholm M, Moestedt J, Schnürer A (2016) Biogas production through syntrophic acetate oxidation and deliberate operating strategies for improved digester performance. Appl Energy 179:124–135. https://doi.org/10.1016/j.apenergy.2016.06.061 Wilmes P, Bond PL (2004) The application of two-dimensional polyacrylamide gel electrophoresis and downstream analyses to a mixed community of prokaryotic microorganisms. Environ Microbiol 6:911–920. https://doi.org/10.1111/j.1462-2920.2004.00687.x Wilmes P, Bond PL (2009) Microbial community proteomics: elucidating the catalysts and metabolic mechanisms that drive the Earth’s biogeochemical cycles. Curr Opin Microbiol 12:310–317. https://doi.org/10.1016/j.mib.2009.03.004 Wilmes P, Heintz-Buschart A, Bond PL (2015) A decade of metaproteomics: where we stand and what the future holds. Proteomics 15:3409–3417. https://doi.org/10.1002/pmic.201500183 Wintsche B, Jehmlich N, Popp D, Harms H, Kleinsteuber S (2018) Metabolic adaptation of methanogens in anaerobic digesters upon trace element limitation. Front Microbiol 9:405. https://doi.org/10.3389/fmicb.2018.00405 Yan Q, Li Y, Huang B, Wang A, Zou H, Miao H, Li R (2012) Proteomic profiling of the acid tolerance response (ATR) during the enhanced biomethanation process from Taihu Blue Algae with butyrate stress on anaerobic sludge. J Hazard Mater 235–236:286–290. https://doi.org/10. 1016/j.jhazmat.2012.07.062 Yin D-M, Westerholm M, Qiao W, Bi S-J, Wandera SM, Fan R, Jiang M-M, Dong R-J (2018) An explanation of the methanogenic pathway for methane production in anaerobic digestion of nitrogen-rich materials under mesophilic and thermophilic conditions. Bioresour Technol 264:42–50. https://doi.org/10.1016/j.biortech.2018.05.062 Zhang W, Dai K, Xia X-Y, Wang H-J, Chen Y, Lu Y-Z, Zhang F, Zeng RJ (2018) Free acetic acid as the key factor for the inhibition of hydrogenotrophic methanogenesis in mesophilic mixed culture fermentation. Bioresour Technol 264:17–23. https://doi.org/10.1016/j.biortech.2018.05.049 Zheng Y, Zhao J, Xu F, Li Y (2014) Pretreatment of lignocellulosic biomass for enhanced biogas production. Prog Energy Combust Sci 42:35. https://doi.org/10.1016/j.pecs.2014.01.001 Zhu Y, McBride MJ (2017) The unusual cellulose utilization system of the aerobic soil bacterium Cytophaga hutchinsonii. Appl Microbiol Biotechnol 101:7113–7127. https://doi.org/10.1007/ s00253-017-8467-2 Ziemiński K, Frąc M (2012) Methane fermentation process as anaerobic digestion of biomass: transformations, stages and microorganisms. Afr J Biotechnol 11:4127–4139. https://doi.org/ 10.5897/AJBX11.054 Ziganshin AM, Ziganshina EE, Kleinsteuber S, Nikolausz M (2016) Comparative analysis of methanogenic communities in different laboratory-scale anaerobic digesters. Archaea 2016:1–12. https://doi.org/10.1155/2016/3401272 110 A. G. Talavera-Caro et al.