UNIVERSIDAD NACIONAL AUTÓNOMA DE MÉXICO CENTRO DE CIENCIAS GENÓMICAS DOCTORADO EN CIENCIAS BIOMÉDICAS BIOSÍNTESIS Y FUNCIONES DE LOS LÍPIDOS DE ORNITINA T E S I S QUE PARA OBTENER EL GRADO DE DOCTOR EN CIENCIAS P R E S E N T A: NAPOLEÓN GONZÁLEZ SILVA DIRECTOR DE TESIS: DR. OTTO GEIGER CUERNAVACA, MOR. NOVIEMBRE DEL 2011 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. UNIVERSIDAD NACIONAL AUTÓNOMA DE MÉXICO CENTRO DE CIENCIAS GENÓMICAS Tesis Doctoral: “Biosíntesis y funciones de los lípidos de ornitina” que para obtener el grado de Doctor en Ciencias presenta: Napoleón González Silva Este trabajo se realizó en el laboratorio del Programa de Ecología Genómica del Centro de Ciencias Genómicas/UNAM, bajo la tutoría del Dr. Otto Geiger, El comité tutor que evaluó el avance del presente proyecto de investigación estuvo integrado por el Dr. Otto Geiger, la Dra. Gloria Soberón Chávez y la Dra. Isabel Ma. López Lara. El jurado de la réplica oral de este trabajo de investigación se integró por el Dr. Michael F. Dunn, el Dr. Jorge Luis Folch Mallol, la Dra. Ma. Del Carmen Gómez Eichelmann, el Dr. Dimitris Georgellis y el Dr. Otto Geiger. El trabajo experimental de esta tesis lo financió la Dirección General de Asuntos del Personal Académico/UNAM (200806 y 218009) y el Consejo Nacional de Ciencia y Tecnología de México (CONACyT 82614). Durante la realización de esta tesis Napoleón González Silva recibió una beca de doctorado otorgada por CONACyT del periodo de Febrero de 2005 a Enero de 2010. Agradecimientos Académicos Un especial y sincero agradecimiento al Dr. Otto Geiger por compartir su conocimiento y experiencia, por guiarme durante el desarrollo de este proyecto y por todo el tiempo dedicado. De igual manara agradezco a la Dra. Ma. Isabel Ma. López Lara por sus sugerencias, aportaciones y enseñanzas. Hago extenso mi agradecimiento al Dr. Christian Sohlenkamp por sus sugerencias, por su disponibilidad para resolver dudas amablemente y por ser amigo y maestro. También agradezco a la Dra. Gloria Soberón Chávez por sus cuestionamientos, por el esfuerzo que realizó de venir a Cuernavaca semestralmente a discutir los avances de este proyecto y por asumir con gran responsabilidad y seriedad su papel de tutora. Al doctor Dr. Michael F. Dunn, a la Dra. Ma. Del Carmen Gómez Eichelmann, el Dr. Dimitris Georgellis y al Dr. Jorge Luis Folch Mallol por la cuidadosa revisión de esta tesis y por sus sugerencias para mejorarla. A la Dra. Esperanza Martínez Romero por proveer lo que se necesita en el laboratorio y por propiciar la interacción académica en el Programa de Ecología Genómica. A José Espíritu, a Martin, a Lucila, a Gladys y a Ma. Dolores por todo el apoyo brindado. Al CONACyT, a la UNAM y al CCG por todos los recursos invertidos en mi formación. Personales A mis padres, hermanos, tíos, primos, sobrinos y especialmente a mi tío Víctor y a mis tías María Trinidad y Eduviges por todos sus consejos y el apoyo brindado. A la Familia Romero-Gallardo por su amistad por el apoyo que me brindo antes y durante el desarrollo de este proyecto. Especialmente a Mario M. Romero Tinoco por motivarme con su optimismo a seguirme superando. A mis amigos Abel Vázquez, David Martínez y a sus familias, por sus consejos y amistad. A mis amigos del CCG J. Luis acosta, Miguel, Ángel, Rosa, Mario, Diana Sahonero, Uriel, Daniela, Roberto, Diana Vera, Emilio, Ramón, Ernesto, Marco, Aline, Martha, Tania, Tabita, Luis E. Servín, Luis M. Bolaños y Wendy, así como a todos mis excompañeros y amigos del CCG por generar un ambiente de trabajo agradable. Dedicatoria A toda mi familia y a la memoria de mis abuelos Josefina y Severo. Contenido Índice de figuras ....................................................................................................................... 4 Índice de tablas ........................................................................................................................ 5 Abreviaturas............................................................................................................................. 6 1. Resumen .............................................................................................................................. 7 2. Abstract (resumen en inglés) ............................................................................................... 10 3. Introducción ....................................................................................................................... 13 3.1. Envolturas celulares de las bacterias ............................................................................. 13 3.1.1. Membranas celulares de las bacterias..................................................................... 14 3.1.1.1. Membrana interna y externa de bacterias Gram negativas................................ 16 3.1.2. Lípidos de membrana de bacterias ......................................................................... 18 3.1.2.1. Lípidos de membrana con la estructura aciloxi-acil de bacterias........................ 19 3.1.2.1.1. Lipopolisacárido........................................................................................ 19 3.1.2.1.2. Lípidos de lisina ........................................................................................ 21 3.1.2.1.3. Lípidos de glicina....................................................................................... 22 3.1.2.1.4. Lípidos de glicinaserina ............................................................................. 24 3.1.2.1.5. Lípidos de ornitina .................................................................................... 25 3.1.2.1.5.1. Distribución y estructura de los lípidos de ornitina .............................. 25 3.1.2.1.5.2. Biosíntesis de los lípidos de ornitina.................................................... 26 3.1.2.1.5.2.1. La N-aciltransferasa OlsB.............................................................. 27 3.1.2.1.5.2.2. La O-aciltransferasa OlsA.............................................................. 30 3.1.2.1.5.2.3. Las hidroxilasas de lípidos de ornitina OlsC y OlsE ......................... 32 3.1.2.1.5.3. Lípidos de ornitinataurina hidoxilados ................................................ 35 3.1.2.2. La respuesta inmune innata hacia los lípidos de membrana con la estructura aciloxi-acil de bacterias ................................................................................................ 36 4. Antecedentes...................................................................................................................... 38 5. Hipótesis ............................................................................................................................ 41 6. Objetivo general ................................................................................................................. 41 6.1. Objetivos particulares................................................................................................ ... 41 7. Procedimientos experimentales de los resultados adicionales .............................................. 42 7.1. Deacilación química de lípidos de ornitina .................................................................... 42 1 7.2. Ensayo in vitro con OlsD usando como sustrato a los 2-OH-OLs ...................................... 42 7.3. Inactivación de bcam1214 ............................................................................................ 43 7.4. Clonación y expresión de los genes bcal0511 y bcam2775 de B. cenocepacia J2315 homólogos a PHYH.............................................................................................................. 45 7.5. Complementación con genes olsBs rizobiales de la mutante deficiente en olsB derivada de B. cenocepacia J2315 (NG1)................................................................................................ . 47 7.6. Ensayo de sensibilidad a antibióticos con cepas derivadas de B. cenocepacia J2315........ 48 7.7. Clonación de las dos versiones de olsC en pNG23 para cooexpresarlas con olsB de B. cenocepacia J2315 .............................................................................................................. 49 7.8. Clonación de olsC en pET16b, su expresión, y los ensayos enzimáticos para OlsC............ 50 7.9. Análisis in silico y antecedentes bibliográficos del gen smc02490 de S. meliloti 1021 ...... 50 7.10. Inactivación del gen smc02490 de S. meliloti 1021 ....................................................... 51 8. Resultados.......................................................................................................................... 54 8.1. Artículos publicados ..................................................................................................... 54 8.1.1. I Artículo................................................................................................................ 54 8.1.2. II Artículo............................................................................................................... 70 8.1.3. III Artículo.............................................................................................................. 94 8.2. Resultados adicionales ............................................................................................... 119 8.2.1. Los lisolípidos (LOLs) derivados de los OLs modificados por OlsD tienen menor movilidad en TLC con respecto a los LOLs derivados de OLs no modificados .................... 121 8.2.2. OlsD también modifica in vitro a los 2-OH-OL........................................................ 122 8.2.3. La mutante en bcam1214 derivada de B. cenocepacia J2315 aún forma los derivados 2-hidroxilados de los OLs y PE........................................................................................ 123 8.2.4. La expresión de los genes bcal0511 y bcam2775 de B. cenocepacia J2315 que codifican para productos homólogos a PHYH sugiere que estos genes no están involucrados en modificar a los OLs ni a PE......................................................................................... 124 8.2.5. Los genes olsBs de rizobiales no complementan a la mutante NG1 derivada de B. cenocepacia J2315......................................................................................................... 126 8.2.6. Sensibilidad a antibióticos de cepas derivadas de B. cenocepacia J2315................. 127 8.2.7. La coexpresión en la cepa CS111 de la versión corta de olsC y olsB de B. cenocepacia J2315 formaron un nuevo lípido..................................................................................... 128 8.2.8. OlsC con cola de histidinas es funcional y aparentemente es una proteína soluble . 129 8.2.9. Los homólogos a SMc02490 se encuentran fusionados con homólogos a OlsB en muchas bacterias. ......................................................................................................... 131 2 8.2.10. Las mutantes en el gen smc02490 de S. meliloti 1021 .......................................... 132 9. Discusión .......................................................................................................................... 134 10. Conclusiones................................................................................................................... 141 11. Perspectivas.................................................................................................................... 141 12. Referencias bibliográficas ................................................................................................ 142 3 Índice de figuras  Figura 1. Estructura esquemática de las envolturas celulares de la bacteria Gram negativa E. coli K‐12 ....................... 14  Figura 2. Estructura del lípido A‐Kdo2 de E. coli y la reacción catalizada por LpxO propuesta por Gibbons et al. (2008) 21  Figura 3. Lípido de lisina (LL) de Agrobacterium tumefaciens....................................................................................... 22  Figura 4. Lípido de Glicina (GL) de Cyclobacterium marinus WH .................................................................................. 23  Figura 5. Lípido de glicinaserina (SGL) de Flavobacterium meningosepticum ............................................................... 25  Figura 6. Biosíntesis y estructura del lípido de ornitina de S. meliloti 1021 .................................................................. 27  Figura 7. Síntesis de 3‐ceto‐dodecanoil‐homoserina lactona, molécula señal del quórum sensing de Pseudomonas  aeruginosa .................................................................................................................................................................. 28  Figura 8. Árbol filogenético sin raíz de OlsB de Sinorhizobium meliloti, OlsB de Burkholderia cenocepacia y ORFs  similares a OlsB ........................................................................................................................................................... 29  Figura 9. Biosíntesis de los lípidos de ornitina en Rhizobium tropici CIAT899 ............................................................... 34  Figura 10. Lípido de ornitinataurina 2‐hidroxilado (2‐OH‐TOL) de Gluconobacter cerinus ............................................. 36  Figura 11. Separación por TLC de lípidos de ornitina no modificados y modificados por OlsD, así como sus respectivos  lisolípidos (LOLs) ....................................................................................................................................................... 122  Figura 12. Modificación in vitro de [14C]2‐OH‐OL por OlsD ......................................................................................... 122  Figura 13. Perfil de hibridación tipo Southern de candidatas a mutantes en el gen bcam1214 ................................... 123  Figura 14. Análisis de extractos crudos libres de células por SDS‐PAGE ...................................................................... 125  Figura 15. Separación de lípidos totales de membrana por 2D‐TLC marcados con acetato de sodio [1‐14C] de cepas  derivadas de la complementación con olsBSm y olsBBc de la mutante en olsB de S. meliloti 1021 (AAK1) .................... 127  Figura 16. Separación de lípidos totales de membrana por 2D‐TLC marcados con acetato de sodio [1‐14C] de cepas  derivadas de la cepa CS111 ....................................................................................................................................... 129  Figura 17. Separación de lípidos totales de membrana por 2D‐TLC marcados con acetato de sodio [1‐14C] de la cepa  899‐olsCΔ1 x pNG43, el plásmido contiene el gen olsC .............................................................................................. 130  Figura 18. OlsC es una proteína soluble que introduce un grupo hidroxilo en el C2 del acilo esterificado de los OLs .. 130  Figura 19. Perfil de hibridación tipo Southern de candidatas a mutantes en el gen smc02490 ................................... 133  Figura 20. Predicción de hélices transmembranales de OlsD con el algoritmo TMHMM [Krogh et al., 2001] .............. 139          4 Índice de tablas  Tabla 1. Antibióticos usados en los ensayos de sensibilidad ........................................................................................ 48  Tabla 2. Cepas y plásmidos, y sus características relevantes ...................................................................................... 119         5 Abreviaturas  ACP: acyl carrier protein (proteína acarreadora de acilos)  AcpS: acyl carrier protein synthase (ACP sintasa)  CL: cardiolipin (cardiolipina)  CM: cytoplasmic membrane (membrana citoplasmática)  CoA: coenzyme A (coenzima A)  cpm: counts per minute (cuentas por minuto)  DGTS: diacylglyceryl‐N,N,N‐trimethylhomoserine (diacilgliceril N,N,N‐trimetilhomoserina)  DMPE: dimethylphosphatidylethanolamine (dimetilfosfatidiletanolamina)  GL: glycine‐containing lipid (lípido de glicina)  IM: Inner membrane (membrana interna)  LL: lysine‐containing lipid (lípido de lisina)  LOL: lyso‐ornithine lipid (lisolípido de ornitina)  LPS: lipopolysaccharide (lipopolisacárido)  MDO: membrane‐derived oligosaccharides (oligosacáridos derivados de membrana)  MMPE: monomethylphosphatidylethanolamine (monometilfosfatidiletanolamina)  OL: ornithine‐containing lipid (lípido de ornitina)  OM: outer membrane (membrana externa)  PC: phosphatidylcholine (fosfatidilcolina)  PCR: the polymerase chain reaction (la reacción en cadena de la polimerasa)  PE: phosphatidylethanolamine (fosfatidiletanolamina)  PG: phosphatidylglycerol (fosfatidilglicerol)  SDS‐PAGE: polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate (electroforesis en  gel de poliacrilamida con dodecilsulfato sódico)  SGL: serineglycine‐containing lipid (lípido de glicinaserina)  TLC: thin layer chromatography (cromatografía en capa fina)  TLR: toll‐like receptor (receptor tipo toll)  TOLs: tauro‐ornithine‐containing lipid (lípido de ornitinataurina)  2D: two‐dimensional (en dos dimenciones)  2‐OH: a hydroxyl group at C2 of an esterified fatty acyl residue (un grupo hidroxilo en el C2 de un acilo  esterificado)  2‐OH‐OL: OL hidroxilado en el C2 de un acilo esterificado  2‐OH‐PE: PE hidroxilada en el C2 de un acilo esterificado  6 1.!Resumen  Burkholderia cenocepacia J2315 es una bacteria Gram negativa conocida como un  patógeno oportunista de humanos. Una característica distintiva del género Burkholderia es  su repertorio de lípidos polares de membrana, entre ellos fosfatidiletanolamina (PE) y los  lípidos  de  ornitina  (OLs),  así  como  sus  derivados  2‐hidroxilados  (2‐OH‐PE  y  2‐OH‐OLs,  respectivamente), los cuales difieren de las versiones típicas por poseer un grupo hidroxilo  en  el  C2  (2‐OH),  que  sustituye  a  un  átomo  de  hidrógeno,  en  un  residuo  de  acilo  esterificado. Similarmente, residuos de miristato esterificados del  lípido A de Salmonella  typhimurium pueden tener la modificación 2‐OH, la cual es realizada por la enzima LpxO.  Sin embargo,  la enzima  responsable de hacer  la sustitución 2‐OH en  los 2‐OH‐OLs de B.  cenocepacia  J2315 todavía no se conoce. Nosotros postulamos que  la 2‐hidroxilación de  los  2‐OH‐OLs  podría  ser  realizada  por  una  nueva  dioxigenasa  homóloga  a  LpxO.  En  el  genoma de B. cenocepacia J2315 identificamos dos marcos abiertos de lectura (BCAM1214  y BCAM2401) homólogos a LpxO de S. typhimurium.  Al  introducir bcam2401 (nombrado olsD) en Sinorhizobium meliloti 1021 se formó  un nuevo  lípido  y en B.  cenocepacia  J2315 dos nuevos  lípidos.  Sorprendentemente,  los  datos  obtenidos  por  espectrometría  de  masas  sugieren  que  OlsD  modifica  el  acilo  amidificado a la ornitina de los OLs con un grupo hidroxilo. Lo anterior es consistente con  la menor movilidad en cromatografía en capa fina (TLC) de los lisolípidos (LOLs) derivados  de OLs modificados por OlsD con respecto a los LOLs derivados de los OLs no modificados.  Esto sugiere que los LOLs derivados de OLs modificados por OlsD son más polares porque  probablemente están hidroxilados en el ácido graso amidificado, el cual forma parte de los  LOLs. Este es el primer reporte de una enzima que hidroxila el acilo amidificado de los OLs.  La  hidroxilación  por  OlsD  de  los  OLs  en  el  ácido  graso  amidificado  ocurre  cuando  B.  cenocepacia  J2315  se  crece  en  estrés  ácido.  Hemos  desarrollado  un  ensayo  para  la  dioxigenasa OlsD y presentamos una caracterización  inicial de  la enzima. Con  los ensayos  enzimáticos demostramos que OlsD puede modificar  in  vitro  a  las dos  clases de  lípidos  ornitina (OLs y 2‐OH‐OLs) que produce B. cenocepacia J2315.  7 En contraste,  la expresión del bcam1214 en S. meliloti 1021 y en B. cenocepacia  J2315 no formó ningún lípido adicional y la mutante en este gen tiene un perfil de lípidos  de membrana similar a la cepa silvestre. Este resultado indica que la proteína BCAM1214  no está  involucrada en hidroxilar a  los OLs ni a PE. Al no encontrar el gen que codificara  para  la  enzima  responsable  de  la  actividad  de  hidroxilar  a  los  OLs  ni  a  PE  en  el  C2,  estudiamos  otros  genes  candidatos  que  codifican  para  productos  homólogos  a  la  2‐ hidroxilasa PHYH de humano.  En el  genoma de B.  cenocepacia  J2315  encontramos dos  homólogos  a  PHYH  de  humano:  bcal0511  y  bcam2775,  los  cuales  se  clonaron  y  se  expresaron en cepas de Escherichia coli, S. meliloti y B. cenocepacia. Ninguno de  los dos  genes alteró el perfil de  lípidos de membrana en  las cepas en  las que se expresaron. Lo  que sugería que estos genes no están involucrados en hidroxilar a los OLs ni a PE.  El gen olsB codifica para la enzima que participa en el primer paso de la biosíntesis  de  los OLs. Una mutante en ese gen  (NG1) derivada de B.  cenocepacia  J2315 no  forma  ninguna clase de OLs, aún cuando se introdujo el gen olsD en esa mutante. Lo que sugiere  que la formación de los OLs hidroxilados ocurre solamente cuando la ruta para biosíntesis  de  los OLs no modificados está  integra. Sorprendentemente,  la mutante NG1 no restauró  la formación de OLs al introducir independientemente los genes olsB de S. meliloti 1021 y  de R.  tropici CIAT899, ni al coexpresar a olsB y acpP de S. meliloti 1021. Sin embargo,  la  mutante en olsB  (AAK1) derivada de S. meliloti 1021 si  restauró  la producción de OLs al  complementarla con el olsB de B. cenocepacia J2315. Actualmente desconocemos porqué  los genes olsB rizobiales no complementan a la mutante NG1.  También  realizamos una evaluación de  la  sensibilidad a antibióticos de  las  cepas  mutantes AQ3 (mutante en olsD) y NG1, así como de la cepa silvestre. Pero no observamos  diferencias en la sensibilidad entre las mutantes y la cepa silvestre.   Recientemente se demostró que el gen olsC de R. tropici CIAT899 codifica para  la  enzima que está involucrada en hidroxilar el C2 del ácido graso esterificado de los OLs de  esta bacteria. El gen olsC  tiene dos probables codones de  inicio de  la  traducción  lo que  teóricamente  podría  producir  dos  versiones  (la  corta  y  la  larga)  de  la  enzima  OlsC.  La  coexpresión de olsC  (versión  corta)  con olsB de B.  cenocepacia  J2315 en  la  cepa CS111  8 formó un nuevo lípido, lo que no ocurrió al expresar la versión larga de olsC con olsB de B.  cenocepacia  J2315. Eso  sugiere que  la versión corta de OlsC es  la  funcional. Mostramos  indicios  experimentales  que  indican  que  OlsC,  al  igual  que  LpxO,  podría  ser  una  2‐ hidroxilasa dependiente de Fe2+/O2/α‐cetoglutarato.  Existe el antecedente de que la coexpresión de olsB y olsA de S. meliloti 1021 en E.  coli no produce OLs. Debido al  resultado anterior  se especuló que otra proteína podría  estar  involucrada en  la biosíntesis de  los OLs. En S. meliloti 1021 el gen smc02490 podría  ser  el  que  codifica  para  dicha  enzima,  ya  que  homólogos  a  este  gen  coexiste  o  se  encuentran  fusionados  con homólogos del  gen olsB en muchas bacterias. Actualmente,  hemos  construido  una  mutante  en  el  gen  smc02490  y  se  harán  experimentos  para  determinar si este gen está involucrado en la biosíntesis de los OLs.  9 2. Abstract (resumen en inglés)  Burkholderia cenocepacia  is an  important opportunistic pathogen, and one of the  most striking features of the Burkholderia genus is the repertoire of polar lipids present in  its  membrane,  including  phosphatidylethanolamine  (PE)  and  ornithine‐containing  lipids  (OLs),  as well  as  the  2‐hydroxylated  derivatives  of  PE  and OLs  (2‐OH‐PE  and  2‐OH‐OLs,  respectively),  which  differ  from  the  standard  versions  by  virtue  of  the  presence  of  a  hydroxyl group at C2 (2‐OH) of an esterified fatty acyl residue. Similarly, a lipid A‐esterified  myristoyl group  from Salmonella  typhimurium can have a 2‐hydroxy modification  that  is  due to the LpxO enzyme. We thus postulated that 2‐hydroxylation of 2‐OH‐OLs might be  catalyzed by a novel dioxygenase homologue of  LpxO.  In B.  cenocepacia, we have now  identified  two  open  reading  frames  (BCAM1214  and  BCAM2401)  homologous  to  LpxO  from S. typhimurium.  The  introduction of bcam2401  (named olsD)  into  Sinorhizobium meliloti  leads  to  the formation of one new  lipid and  in B. cenocepacia of two new  lipids. Surprisingly, the  lipid modifications on OLs due to OlsD occur  in the amide‐linked  fatty acyl chain. This  is  consistent with  the  lower mobility  in  thin  layer  chromatography of  lyso‐ornithine  lipids  (LOLs)  derived  from  OLs  modified  by  OlsD,  when  compared  to  the  LOLs  derived  from  unmodified standard OLs. OlsD‐modified LOLs are more polar than standard OLs due to an  extra hydroxyl group on the amide‐linked fatty acyl moiety. The hydroxyl modification of  OLs on the amide‐linked fatty acyl moiety due to OlsD occurs under acid stress conditions.  An assay has been developed for the OlsD dioxygenase, and an  initial characterization of  the enzyme is presented. With enzyme assays we show that OlsD can modify in vitro the  two classes of ornithine lipids (OLs and 2‐OH‐OLs), which produces B. cenocepacia J2315.  In contrast, expression of bcam1214 in S. meliloti 1021 or B. cenocepacia J2315 did  not  cause  the  formation of any additional  lipid. The B.  cenocepacia bcam1214‐deficient  mutant  has  a  membrane  lipid  profile  similar  to  wild  type.  These  results  suggest  that  BCAM1214 does not hydroxylate OLs or PE. Also,  two B. cenocepacia  J2315 homologues  (bcal0511  and  bcam2775)  of  the  human  phytanoyl‐CoA  hydroxylase  PHYH,  which  hydroxylates at the 2‐position of fatty acids, were studied for their capacity to hydroxylate  10 OL‐ or PE‐bound fatty acyl residues at the 2‐position. After cloning and expressing either  bcal0511  or  bcam2775,  in  Escherichia  coli,  S. meliloti,  or B.  cenocepacia  no  changes  in  membrane  lipid profiles could be observed, suggesting that these genes are not  involved  in hydroxylating OLs or PE.  The N‐acyltransferase‐encoding OlsB catalyzes the first step of OL biosynthesis and  the OlsB‐deficient mutant NG1 of B.  cenocepacia  J2315 does not  form any kind of OLs.  Therefore, formation of hydroxylated OLs occurs only when the biosynthesis pathway for  nonmodified  standard  OLs  is  intact.  Surprisingly,  the  NG1  mutant  cannot  be  complemented for the formation of OLs by introducing the olsB gene from S. meliloti 1021  or  from R.  tropici CIAT899 or by  combined expression of olsB and acpP  from S. meliloti  1021 in NG1. In contrast, the olsB mutant AAK1 of S. meliloti 1021 can be complemented  for OL production upon expression of the olsB gene from B. cenocepacia J2315. Currently,  we  don’t  know why  rhizobial  olsB  genes  cannot  complement  the  olsB‐deficient mutant  NG1 of B. cenocepacia.  B. cenocepacia J2315 mutants deficient in olsD (AQ3) or olsB (NG1) were not more  sensitive than the wild type to a wide range of antibiotics tested.  The  gene  olsC  from  R.  tropici  CIAT899  encodes  the  enzyme  that  hydroxylates  esterified fatty acyl residues of OLs at the 2‐position. The olsC gene has two potential start  codons for translation which theoretically could produce two versions (a short and a long  version) of the enzyme OlsC. The combined expression of the short version rhizobial olsC  and olsB from B. cenocepacia J2315  in S. meliloti strain CS111 caused the formation of a  new  lipid, which did not occur when the  long version of olsC was coexpressed with olsB  from  B.  cenocepacia  J2315.  This  result  suggests  that  the  short  version  of  OlsC  is  the  functional one. Also, we have developed an assay for the OlsC dioxygenase and, like LpxO,  it seems to be a 2‐hydroxylase dependent on Fe2+, O2, and α‐ketoglutarate.  Surprisingly, the combined expression of olsB and olsA  from S. meliloti 1021  in E.  coli does not  cause  the  formation of OLs  leading  to  the  suggestion  that more  rhizobial  genes/proteins  might  be  required  for  a  complete  OL  biosynthesis.  Homologues  of  the  smc02490  gene  are  fused  to  or  coexist  with  olsB.  Therefore,  we  speculated  that  the  11 smc02490 product might be required for OL biosynthesis. We have constructed a mutant  in the smc02490 gene and currently perform experiments to determine whether this gene  is involved in OL biosynthesis.  12 3. Introducción  3.1. Envolturas celulares de las bacterias  Las envolturas celulares de  las bacterias consisten de  las estructuras celulares que  envuelven al citoplasma. La membrana citoplasmática  (CM) o membrana  interna  (IM), el  espacio  periplasmático  con  la  pared  celular  y  la  membrana  externa  (OM)  son  las  envolturas celulares de  las bacterias Gram negativas  (fig. 1). En contraste,  las envolturas  celulares de las bacterias Gram positivas consisten solamente de una CM y la pared celular.  Las envolturas celulares de las bacterias están constituidas por lípidos, proteínas, péptidos,  carbohidratos o combinaciones de estas moléculas.  La CM es muy similar entre bacterias Gram negativas y Gram positivas, y consiste  de una bicapa  lipídica  formada por dos monocapas  lipídicas constituidas principalmente  de glicerofosfolípidos, así como de proteínas de membrana integrales y periféricas.  El  espacio  periplasmático  es  el  espacio  que  existe  entre  la membrana  interna  y  externa de las bacterias Gram negativas. Este es un ambiente oxidante donde la estructura  de  las  proteínas  puede  ser  estabilizada  por  enlaces  disulfuro.  Dependiendo  del  microorganismo,  el  espacio  contiene  enzimas  hidrolíticas  que  degradan  alimentos  y  algunas sustancias tóxicas, proteínas de unión, las cuales inician el proceso de transporte  de sustratos, quimiorreceptores,  las cuales son proteínas  involucradas en  la quimiotaxis,  así  como  enzimas  involucradas  en  diversas  vías  bioquímicas  incluyendo  la  síntesis  de  peptidoglicano  [Madigan  y  Martinko,  2006].  En  el  espacio  periplasmático  de  las  Gram  negativas  también se encuentra  la pared celular. La pared celular en  las bacterias Gram  negativas es mucho más delgada respecto a la de bacterias Gram positivas, esta previene  de  la  lisis  osmótica  [Madigan  y  Martinko,  2006].  La  pared  celular  está  compuesta  principalmente de peptidoglicano que está constituido por polímeros  lineales de β‐(1,4)‐ N‐acetilglucosamina y ácido N‐acetilmurámico unidos por pequeños péptidos  [Vollmer y  Bertsche,  2008].  En  el  espacio  periplasmático  también  hay  oligosacáridos  derivados  de  membrana  (MDO),  y  los  MDO  de  Escherichia  coli  están  compuestos  de  seis  a  ocho  residuos de glucosa unidos por enlaces β‐(1‐2) y β‐(1‐6) [Rumley et al., 1992].  13 La OM es exclusiva de  las bacterias Gram negativas y es  la envoltura más externa  que forma  la superficie celular de este grupo de bacterias. La OM también está formada  por dos monocapas. La monocapa  interna de  la OM al  igual que  las monocapas de  la  IM  está constituida principalmente de glicerofosfolípidos. En contraste,  la monocapa externa  de la OM está constituida principalmente por monómeros de lipopolisacárido (LPS) [Raetz  et al., 2007].      Figura 1. Estructura esquemática de  las envolturas celulares de  la bacteria Gram negativa E. coli K‐12. La membrana  interna, el espacio periplasmático y la membrana externa se indican, así como sus subcomponentes. Kdo (ácido 2‐ceto‐3‐ desoxi‐D‐mano‐octulosónico). Tomada y traducida de Raetz et al. (2007).    3.1.1. Membranas celulares de las bacterias  Las membranas  celulares de  las bacterias  son estructuras dinámicas que permiten a  estos  microorganismos  interactuar  y  comunicarse  con  el  medio  extracelular.  Las  membranas  celulares  de  las  bacterias  son  imprescindibles  porque  delimitan  el  cuerpo  celular, lo cual, permite establecer gradientes químicos o potenciales de membrana entre  14 el citoplasma y el ambiente exterior, que a través de  la respiración proveen de  la energía  que las células destinan a procesos esenciales. Los gradientes se generan debido a que las  membranas celulares  funcionan como barreras de permeabilidad selectiva muy efectivas  para  iones,  compuestos  hidrofílicos  y  hidrofóbicos,  lo  que  consecuentemente  regula  el  flujo  de  muchas  sustancias  hacia  fuera  y  hacia  dentro  de  las  células.  Aunque  algunos  compuestos  hidrofóbicos  pueden  difundirse  a  través  de  las  membranas  con  relativa  facilidad.  Las membranas se forman por la incorporación de lípidos y proteínas sintetizados por  las células,  los cuales se  insertan en  las membranas preexistentes. Las membranas de  las  bacterias Gram negativas están constituidas por ciertas proporciones de clases de lípidos,  y proteínas, los que determinan su identidad estructural y funcional [Madigan y Martinko,  2006]. Las membranas celulares consisten de una bicapa lipídica y de proteínas integrales  y periféricas, algunas de estas son glicoproteínas o lipoprotínas. La estructura de bicapa de  la membrana  representa  la  conformación más  estable  de  las moléculas  lipídicas  en  un  ambiente acuoso [Madigan y Martinko, 2006].  Los lípidos y las proteínas de membrana están unidos principalmente por interacciones  no covalentes como enlaces de hidrógeno e interacciones hidrofóbicas. En adición cationes  tales  como Mg+  y  Ca+  ayudan  a  estabilizar  las membranas  por  combinación  iónica  con  cargas  negativas  de  los  lípidos  [Madigan  y  Martinko,  2006].  Las  bicapas  lipídicas  están  constituidas por dos monocapas de  lípidos que están orientados en direcciones opuestas  formando  un  dominio  hidrofóbico  con  sus  residuos  de  acilo  o  alquilo  y  dos  dominios  hidrofílicos  determinados  por  los  grupos  cabeza  polares  de  los  lípidos,  unos  se  dirigen  hacia el citosol y otros hacia el exterior de la célula (fig. 1) [Madigan y Martinko, 2006]. Los  dominios  hidrofílicos  y  hidrofóbicos  formados  por  el  ensamble  macromolecular  de  las  bicapas  son  los  que  impiden  la  difusión  espontánea  de  iones  o  de  compuestos  de  naturaleza opuesta a esos dominios.  Muchos  de  los  materiales  que  se  mueven  del  citoplasma  celular  al  ambiente  extracelular,  y  viceversa,  lo  hacen  a  través  del  transporte  activo,  el  cual  usa  proteínas  transportadoras  que  consumen  energía  metabólica  al  realizar  su  actividad.  Otros  15 materiales  se  trasladan  a  través  de  transporte  pasivo  mediante  canales  de  difusión  generales  o  específicos  formados  por  proteínas  que  forman  poros  (porinas)  en  las  membranas. La difusión a través de  las membranas está determinada por el tamaño y  la  naturaleza de los materiales en cuestión.  Algunos  componentes  de  las membranas  suplen  de  sustratos  para  la  biosíntesis  de  otros  componentes  celulares  y  de  moléculas  señales.  Mientras  tanto,  las  proteínas  de  membrana  están  involucradas  en  funciones  esenciales  para  las  células  tales  como  el  trasporte selectivo de iones o moléculas, en la conversión y generación de energía a través  de la cadena respiratoria, división celular, motilidad, en la señalización celular funcionando  como  receptores  censando  señales  intracelulares  y  extracelulares,  y  en  el metabolismo  participando como enzimas en diversos procesos del catabolismo y anabolismo  [Facey y  Kuhn, 2010; Madigan y Martinko, 2006].  3.1.1.1. Membrana interna y externa de bacterias Gram negativas  Las bacterias Gram negativas están rodeadas por dos membranas diferentes: la IM y la  OM.  Las dos membranas están  separadas por el espacio periplasmático y  se  componen  principalmente de una mezcla de glicerofosfolípidos pero difieren considerablemente en  su estructura y composición. Se asume que en E. coli la IM está compuesta por una bicapa  asimétrica  con  la  siguiente  composición  de  glicerofosfolípidos:  alrededor  de  80  %  de  fosfatidiletanolamina (PE), 15 % de fosfatidilglicerol (PG) y 5 % de cardiolipina (CL) [Kadner,  1996].  La  OM,  también  es  una  bicapa  asimétrica  y  su  monocapa  interna  tiene  una  composición  lipídica similar a  la membrana citoplasmática;  la monocapa externa de esta  membrana en la mayoría de las especies de las bacterias Gram negativas está compuesta  de LPS y en unas pocas especies también por sulfonolípidos o glicoesfingolípidos, además  del  LPS  o  en  su  lugar  [Nikaido,  2003].  Aunque  glicerofosfolípidos  también  han  sido  detectados en esa monocapa en casos extraordinarios.  Ambas  membranas  contienen  muchas  proteínas  que  son  esenciales  para  muchas  funciones  celulares.  Estas  proteínas  desempeñan  funciones  protagónicas  en  la  transducción de energía, también están involucradas en el censado y en la transducción de  16 señales de estímulos ambientales, así  como en el  influjo y eflujo de  sustancias  [Facey y  Kuhn, 2010].  La  biogénesis  de  estas  proteínas  requiere  de  el  direccionamiento  coordinado  a  la  membrana,  la  inserción  y  el  desplazamiento  a  través  de  la  membrana,  y  subsecuentemente  el  ensamble  en  complejos  multiprotéicos.  Las  proteínas  de  la  membrana tienen que incorporarse en la IM o en la OM y ser plegadas correctamente para  funcionar  en  la  célula.  Para  facilitar  el  direccionamiento  se  requiere  del  plegamiento  y  ensamble,  muchas  chaperonas  moleculares,  de  la  translocación  y  de  maquinaria  de  inserción [Facey y Kuhn, 2010]. La información de porqué algunas proteínas son dirigidas a  las membranas como destino final está contenida dentro de la secuencia de aminoácidos  de esas proteínas. En general, las proteínas con hélices α con segmentos que atraviesan la  membrana se localizan en la IM, mientras que proteínas anfipáticas con la forma de barril  β (llamadas proteínas de membrana externa u Omps) están localizadas en la OM. En E. coli  aproximadamente de 20‐30 % de todas las proteínas codificadas en su genoma (alrededor  de 4000) son proteínas de la IM y aproximadamente 2 % son proteínas de la OM [Facey y  Kuhn, 2010].  Las membranas de las células eucariontes albergan microdominios conocidos como  balsas lipídicas que contienen proteínas involucradas en el transporte y en la señalización.  Las  balsas  lipídicas  de  eucariontes  son  enriquecidas  en  lípidos  tales  como  esteroles  y  esfingolípidos.  Recientemente  López  y  Kolter  (2011)  mostraron  evidencia  de  que  las  bacterias  también  contienen  microdominios  que  son  similares  funcionalmente  a  los  reportados en eucariontes. En eucariontes la proteína flotilina‐1 se ha localizado en balsas  lipídicas y reportes previos indican que las proteínas de bacterias homólogas a la flotilina‐1  de eucariontes están distribuidas heterogéneamente en  la membrana  citoplasmática en  un patrón punteado [Zhang et al. 2005; Donovan y Bramkamp 2009]. El ácido zaragocico  en  Bacillus  subtilis  inhibe  la  formación  de  biofilm  y  la  secreción  de  proteínas  pero  no  viabilidad de las células [López y Kolter, 2011]. El ácido zaragocico es un conocido inhibidor  de  la escualeno sintasa y debido al efecto que causa en B. subtilis se ha sugerido que  los  lípidos asociados con  las balsas  lipídicas bacterianas probablemente son poliisopreniodes  17 [López  y Kolter, 2011]. Otra evidencia que  López  y Kolter  (2011) mostraron es que una  mutante de B.  subtilis  en  las proteínas homólogas  a  la  flotilina‐1 de  eucariontes  se  vio  afectada en  la ruta de  traducción de señal que conduce a  la  formación de biofilm, en  la  que  está  involucrada  la  censora  cinasa  (KinC).  La  proteína  KinC  presumiblemente  está  localizada en balsas lipídicas al igual que las proteínas bacterianas homólogas a la flotilina‐ 1.  3.1.2. Lípidos de membrana de bacterias  La función estructural de los lípidos de membrana es esencial para los organismos  vivos. Los lípidos de membrana normalmente son anfifílicos y cilíndricos que consisten de  dos cadenas hidrofóbicas  largas de acilo o alquilo y un grupo cabeza hidrofílico [Dowhan,  2008].  El  área  cubierta  por  los  grupos  cabeza  de  estos  lípidos  suele  ser  similar  al  área  cubierta  por  la  de  sus  cadenas  hidrocarbonadas,  lo  que  permite,  que  los  monómeros  lipídicos interactúen, favoreciendo la formación de la bicapa en lugar de micelas [Dowhan,  2008].  En  las bacterias  los glicerofosfolípidos comprenden alrededor del 10 % del peso seco  de la célula, y cada mol de lípido requiere para su biosíntesis de alrededor de 32 moles de  adenosín  trifosfato  [Nelson  y  Cox,  2010].  En  la  bacteria  modelo  E.  colí  los  principales  lípidos  de  membrana  son  los  glicerofosfolípidos:  PE,  PG  y  CL  [Rock,  2008].  Otros  glicerofosfolípidos  tales  como:  fosfatidilserina,  fosfatidilcolina  (PC)  y  fosfatidilinositol  también están presentes en grupos específicos de bacterias. Algunas bacterias  tienen  la  capacidad  de  modificar  ciertos  glicerofosfolípidos  con  residuos  de  aminoácidos  o  con  grupos hidroxilo  [Geiger et al., 2010]. Formas deaciladas de CL  se han encontrado en  la  membrana celular de especies del género Streptomyces [Hoischen et. al., 1997; Sandoval‐ Calderón et al., 2009].  Los glicolípidos también están presentes en las membranas de las bacterias, así como  lípidos esteroides y hopanoides [López‐Lara et al., 2003; Geiger et al., 2010]. Otros lípidos  sin fósforo en su estructura como el sulfolípido sulfoquinovosildiacilglicerol y diacilgliceril  N,N,N‐trimetilhomoserina  (DGTS) son sintetizados principalmente en grandes cantidades  18 cuando algunas bacterias se crecen en condiciones limitadas en fósforo [López‐Lara et al.,  2003; Geiger et al., 2010].  E. coli y casi todas las bacterias Gram negativas contienen LPS (fig. 2). El lípido A es una  parte  del  LPS  y  forma  la  monocapa  lipídica  externa  de  la  OM  de  estas  bacterias,  la  biosíntesis y el sistema de modificación del lípido A ha sido descrito recientemente [Raetz  et al., 2007]. En algunas bacterias Gram negativas el LPS es remplazado por esfingolípidos  [Kawahara et al., 1991]. Los sulfonolípidos también se han reportado que están presentes  en la OM de bacterias Gram negativas [Pitta et al., 1989].  Además de los lípidos mencionados en algunas bacterias, se ha reportado la presencia  de lípidos con la estructura aciloxi‐acil que contienen aminoácidos, entre ellos, los lípidos  que contienen lisina, glicina, glicinaserina, ornitina u ornitinataurina. La estructura aciloxi‐ acil también está presente en  lípido A. El lípido A de  las Enterobacteriacea es un potente  inmunoactivador de  la  inmunidad  innata de mamíferos y se ha demostrado que algunos  de los lípidos con la estructura aciloxi‐acil que contienen aminoácidos también lo son.  3.1.2.1. Lípidos de membrana con la estructura aciloxi‐acil de bacterias  3.1.2.1.1. Lipopolisacárido  Los  LPSs  son  glucolípidos  anfipáticos  complejos  que  forman  la  mayor  parte  de  la  superficie  (75  %)  de  la  mayoría  de  las  bacterias  Gram  negativas.  Los  LPSs  son  inmunogénicos e interaccionan con organismos hospederos eucariotas y con el ambiente  exterior [Raetz y Whitfield, 2002; Raetz et al., 2007]. Una molécula típica de LPS consiste  de  tres  partes  (fig.  1):  1)  el  lípido  A,  2)  el  oligosacárido  central,  el  cual  se  divide  conceptualmente en núcleo interno y externo, y 3) un polisacárido distal (antígeno O).  El lípido A es la parte del LPS que ancla a estas moléculas a la OM. El lípido A de E. coli  K‐12 está compuesto por un disacárido de glucosamina unido por enlaces β−(1’,6), el cual  es  típicamente  fosforilado  en  las  posiciones  1  y  4’,  y  acilado  con  ácido  (R)‐3‐ hidroximirístico en  las posiciones 2, 2′, 3 y 3′, mediante enlaces tipo amida (2, 2’) y éster  (3,3’) (fig. 2) [Raetz y Whitfield, 2002]. Los ácidos grasos de las posiciones 2´ y 3´ también  son acilados en su C3 formando enlaces éster y produciendo la estructura aciloxi‐acil (fig.  19 2). El  lípido A de E. coli y de muchas otras bacterias difiere de un glicerofosfolípido por  tener  seis  ácidos  grasos  saturados  en  lugar  de  dos  saturados  o  insaturados.  Esta  característica hace a la OM mucho más hidrofóbica [Nikaido, 1996] que una bicapa típica  de glicerofosfolípidos. La hidrofóbicidad de esta membrana también es promovida por las  interacciones  laterales  fuertes  entre  las  moléculas  de  LPS  produciendo  baja  fluidez  [Nikaido, 2003].  El  núcleo  interno  del  oligosacárido  central  contiene  típicamente  residuos  de  Kdo  (ácido 2‐ceto‐3‐desoxi‐D‐mano‐octulosónico) y uno normalmente está unido al disacárido  de  glucosamina  y  por  otro  lado  a  un  residuo  de  L‐glicero‐D‐manoheptosa  (heptosa).  Además, la estructura del núcleo interno frecuentemente es decorada con otros azúcares,  grupos fosfato o residuos de etanolamina difosfato (fig. 1). Los múltiples grupos aniónicos  (entre  ellos,  grupos  fosfato)  del  esqueleto  del  disacárido  de  glucosamina  y  del  núcleo  interno  son  conocidos  por  interactuar  fuertemente  con  cationes  divalentes,  los  cuales  compensan  la  repulsión electrostática entre  las moléculas adyacentes de LPSs, debida a  esos grupos aniónicos. El núcleo interno de E. coli K‐12 se conecta con el núcleo externo a  través de un residuo de heptosa conservado. El núcleo externo del oligosacárido central  muestra una mayor diversidad estructural respecto al núcleo  interno y contiene residuos  de  azúcar  entre  ellos  heptosa  y  hexosas  como  glucosa,  galactosa, N‐acetilglucosamina,  [Raetz y Whitfield, 2002].  El antígeno O está unido al núcleo externo y es  la parte más variable del LPS, es un  polisacárido  constituido por unidades  repetitivas de monosacáridos u oligosacáridos.  La  longitud del antígeno O difiere aún en un mismo organismo, pudiendo estar ausentes en  unos y en otros consistir de más de 60 monosacáridos. Los monosacáridos que componen  las unidades  repetitivas  son  generalmente,  azúcares neutros  y  acídicos,  aminoazúcares,  además  de  los  azúcares  inusuales,  tales  como  6‐desoxihexosas  ó  3,6‐didesoxihexosas  [Raetz y Whitfield, 2002].  Algunos ácidos grasos esterificados del  lípido A de ciertas bacterias  son modificados  con un grupo hidroxilo. En  S.  typhimurium existen esos ácidos grasos  S‐2‐hidroxilados  y  están  esterificados  al  C3  del  acilo  unido  a  la  posición  3´  del  segundo  residuo  de  20 glucosamina del lípido A y se ha pensado que son importantes para la patogénesis de este  organismo  (fig.  2)  [Gibbons  et  al.,  2000;  Gibbons  et  al.,  2008].  La  S‐2‐hidroxilación  es  realizada  después  de  que  el  residuo  de  acilo  es  unido  a  la  molécula  del  lípido  A  y  es  catalizada por  la dioxigenasa LpxO dependiente de Fe2+/O2/α‐cetoglutarato para producir  el 2‐OH‐lípido A‐Kdo2 (fig. 2) [Gibbons et al., 2000; Gibbons et al., 2008]. La expresión del  gen  lpxO  se  incrementó  cuando  S.  typhimurium  se  creció  en  bajas  concentraciones  de  Mg2+  [Gibbons  et  al.,  2005].  Se ha  especulado que  los  grupos hidroxilos  extras podrían  incrementar  los enlaces de hidrógeno entre  las moléculas adyacentes de  lípidos que  los  poseen aumentando  la estabilidad e  impermeabilidad de  la membrana externa a ciertos  compuestos en algunas condiciones de crecimiento [Nikaido, 2003; Gibbons et al., 2008].     2‐OH‐lípido A‐Kdo2 lípido A‐Kdo2  Figura 2. Estructura del lípido A‐Kdo2 de E. coli y la reacción catalizada por LpxO propuesta por Gibbons et al. (2008).  LpxO cataliza la 2‐hidroxilación del residuo de miristato unido al C3 del acilo que está unido a la posición 3´ del lípido A‐ Kdo2 hexaacilado para producir el 2‐OH‐lípido A‐Kdo2. Tomada de Gibbons et al. (2008).    3.1.2.1.2. Lípidos de lisina  Los  lípidos de  lisina  (LLs) se describieron en una cepa de Agrobacterium tumefaciens  [Tahara  et  al.,  1976a].  La  estructura  de  los  LLs  se  reportó  como  α‐N‐(aciloxi‐acil)‐lisina  [Tahara et al., 1976a]. En esta estructura el grupo α‐amino de  la  lisina está N‐acilado con  un residuo de palmitato y un segundo acilo está esterificado al C3 del palmitato formando  21 la  estructura  aciloxi‐acil  (fig.  3)  [Tahara  et  al.,  1976a].  La  configuración  del  carbono  asimétrico  del  palmitato  no  se  ha  determinado.  No  se  han  hecho  ensayos  sobre  la  bioactividad  de  estos  lípidos  y  tampoco  se  conocen  los  genes  involucrados  en  su  biosíntesis.    LL  Figura  3.  Lípido  de  lisina  (LL)  de  Agrobacterium  tumefaciens. Tomada de Geiger et al. (2010).    3.1.2.1.3. Lípidos de glicina  Los  lípidos de glicina  (GLs) se  identificaron en  la bacteria que se mueve por “gliding”  Cytophaga  johnsonae C21 y en  la bacteria marina Cyclobacterium marinus WH [Kawazoe  et al., 1991; Batrakov et al., 1999]. Los GLs son análogos a los LLs descritos anteriormente  pero en lugar de la lisina contienen una glicina. La estructura de la especie mayoritaria de  GL  de  C.  marinus  WH  se  determinó  como  N‐[3‐D‐(13‐metiltetradecanoiloxi)‐15‐ metilhexadecanoil]glicina  [Batrakov et al., 1999]. En esta estructura  (fig. 4) el grupo  iso‐ acilo  amidificado  está  esterificado  en  el  C3  por  otro  grupo  iso‐acilo.  La  configuración  absoluta  del  C3  asimétrico  del  acilo  amidificado  se  determinó  como D  [Batrakov  et  al.,  1999].  22 Los GLs  también  son  llamados  citolipinas  porque  se  identificaron  inicialmente  en  el  género Cytophaga [Batrakov et al., 1999]. Estos constituyen alrededor del 6 y 5 % del total  de lípidos en C. johnsonae C21 [Kawazoe et al., 1991] y en C. marinus WH [Batrakov et al.,  1999], respectivamente. Basados en análisis por métodos cromatográficos de varias cepas  se asumió que  lipoaminoácidos estructuralmente  similares a  los GLs están presentes en  varias bacterias del género Cytophaga que se mueven por “gliding” [Kawazoe et al., 1992],  y  en  algunas  bacterias  Gram  negativas  de  agua  dulce  pertenecientes  a  los  géneros  Arcocella [Nikitin et al., 1994] y Flectobacillus [Raj y Maloy, 1990] relacionados al género  Cyclobacterium.  Los  últimos  tres  géneros  son  distintos  sistemáticamente  de  la  familia  Cytophagaceae.  Debido  a  eso  se  sugirió  que  los  GLs  pudieran  estar  ampliamente  distribuidos entre bacterias acuáticas Gram negativas [Batrakov et al., 1999]. Sin embargo,  una explicación alternativa podría ser que  los genes estructurales que codifican para  las  proteínas  que  participan  en  la  biosíntesis  de  los  GLs  hayan  sido  transferidos  horizontalmente.  Hasta la fecha los genes involucrados en la biosíntesis de los GLs no se conocen aunque  nosotros  esperamos  que  esos  genes  sean  similares  a  los  genes  involucrados  en  la  biosíntesis  de  los  lípidos  de  ornitina,  debido  a  la  similitud  estructural  entre  ambas  moléculas  [Geiger et al., 2010].  Interesantemente,  fusiones homólogas del gen olsB, que  se  describirán  más  adelante,  que  se  han  detectado  en  especies  de  Alteromonadales,  también están en los genomas de varios miembros del género Cytophaga. Actualmente de  los  GLs  solamente  conocemos  que  funcionan  como  unidades  estructurales  de  las  membranas celulares de algunas bacterias.    GL  Figura 4. Lípido de Glicina (GL) de Cyclobacterium  marinus WH. Tomada de Geiger et al. (2010).  23 3.1.2.1.4. Lípidos de glicinaserina  El  lípido de  glicinaserina  (SGL)  se  aisló del patógeno oportunista  Flavobacterium  meningosepticum y se le llamo “flavolipina”, debido al nombre del género de la bacteria de  la  cual  se  aisló  por  primera  vez  [Kawai  et  al.,  1988].  La  flavolipina  se  caracterizó  por  métodos químicos  y  fisicoquímicos, e  inicialmente  se propusó  la estructura N‐(3‐aciloxi‐ acil)serina,  la cual es  incorrecta [Kawai et al., 1988; Shiozaki et al., 1998a] porque  le falta  un residuo de glicina. Intentos por sintetizar químicamente  la flavolipina para estudiar su  actividad biológica  llevó a  la reinvestigación estructural y se determinó que  la  flavolipina  (fig.  5)  contenía  los  aminoácidos  serina  y  glicina  condensados  y  dos  grupos  acilos,  uno  amidificado a  la glicina y el otro esterificado al C3 del acilo amidificado  [Shiozaki et al.,  1998b].  La  configuración  absoluta  del C3  asimétrico  del  acilo  amidificado  se  determinó  como  (R)  [Shiozaki  et  al.,  1998b; Gomi  et  al.,  2002].  Flavolipina  comparte  la  estructura  básica con  los GLs pero tiene un residuo adicional de serina  (figs. 4 y 5),  lo cual, sugiere  que  los  GLs  pudieran  ser  precursores  biosintéticos  directos  para  la  formación  de  flavolipinas en C. marinus WH [Batrakov et al., 1999]. A diferencia de C. marinus WH, siete  especies  analizadas  del  género  Cytophaga  no  sintetizan  flavolipinas.  Hasta  la  fecha  los  genes involucrados en la biosíntesis de flavolipina no han sido identificados.  Por  cierto  tiempo  se  pensó  que  flavolipina  era  exclusiva  de  especies  del  género  Flavobacterium, posteriormente se encontró en C. marinus WH  [Batrakov et al., 1998] y  por  lo  tanto, es probable que  también esté presente en otras bacterias de agua marina.  Flavolipina constituye alrededor de 21 y 11 % del total de  lípidos de F. meningosepticum  [Kawai et al., 1988] y de C. marinus WH [Batrakov et al., 1998], respectivamente.    24 SGL  Figura  5.  Lípido  de  glicinaserina  (SGL)  de  Flavobacterium meningosepticum. Tomada de Geiger  et al. (2010).    3.1.2.1.5. Lípidos de ornitina  3.1.2.1.5.1. Distribución y estructura de los lípidos de ornitina  Los  lípidos de ornitina (OLs) están ampliamente distribuidos entre  las bacterias Gram  negativas y también han sido reportados en algunas Gram positivas, entre ellas especies  de  Mycobacterium  y  Streptomyces  [López‐Lara  et  al.,  2003]  pero  están  ausentes  en  arqueas y eucariontes. La estructura de  los OLs de diferentes bacterias ha sido reportada  como α‐N‐(aciloxi‐acil)‐ornitina [Knoche y Shively, 1972; Geiger et al., 1999]. Los OLs son  análogos a los LLs y a los GLs, descritos anteriormente, pero en lugar de la lisina o glicina  contienen un residuo de ornitina. Su estructura consiste de un grupo acilo que está unido  al grupo α‐amino de la ornitina por enlace amida. Un segundo grupo acilo está ligado por  enlace  éster  al  C3  del  acilo  amidificado  (fig.  6).  La  estructura  de  los  OLs  que  se  ha  elucidado  por  resonancia  magnética  nuclear  es  consistente  con  la  que  se  ha  descrito  anteriormente  [Okuyama  y Monde,  1996; Maneerat  et  al.,  2006;  Keck  et  al.,  2011].  En  algunas  bacterias  los  grupos  acilos  unidos  por  enlace  éster  están  hidroxilados  en  los  carbonos de las posiciones 2 o 3 [Asselineau, 1991].  Recientemente  también se han reportado acilos amidificados hidroxilados en  los OLs  de B. cenocepacia  J2315  [González‐Silva et al., 2011] y OLs hidroxilados en el residuo de  25 ornitina en R. tropici CIAT899 [Vences‐Guzmán et al., 2011]. Los acilos amidificados de los  OLs de diferentes bacterias normalmente son de C16 o de C18 y saturados, mientras que  los esterificados  son muy variables en  su  longitud y  suelen  tener una  insaturación o un  grupo ciclopropano, y en algunas bacterias ambos acilos son iso‐acilos [Palacios‐Chaves et  al.,  2011;  González‐Silva  et  al.,  2011].  La  configuración  del  C3  asimétrico  del  acilo  amidificado de los OLs es D o (R) [Kawai et al., 1999]. Aunque los OLs están presentes en  ambas  membranas  de  las  bacterias  Gram  negativas,  estos  se  han  encontrado  mayoritariamente en  la membrana externa  [Dees y Shively, 1982; Palacios‐Chaves et al.,  2011; Vences‐Guzman et al., 2011].  3.1.2.1.5.2. Biosíntesis de los lípidos de ornitina  La biosíntesis de un OL estándar  se da en dos pasos. En el primer paso, participa  la  enzima OlsB que es una N‐aciltransferasa y cataliza  la  transferencia de un grupo acilo 3‐ hidroxilado de una proteína acarreadora de acilos  (ACP) al grupo  α‐amino de  la ornitina  formando el  lisolípido de ornitina (LOL) [Gao et al., 2004]. En el segundo paso,  la enzima  OlsA que es una O‐aciltransferasa cataliza la transferencia de otro grupo acilo de una ACP  al  grupo  3‐hidroxilo  del  LOL  formando  el  OL  (fig.  6)  [Weissenmayer  et  al.,  2002].  La  transcripción  del  gen  olsB  en  S.  meliloti  1021  se  incrementa  cuando  se  crece  en  condiciones  limitadas  en  fósforo  y  está  regulada  por  el  regulador  transcripcional  PhoB  [Krol y Becker, 2004]. La expresión constitutiva de olsB en S. meliloti 1021  incrementó  la  formación de OLs independientemente de si se crece en concentraciones altas o bajas de  fósforo, sugiriendo que en S. meliloti 1021 la formación de OlsB suele ser limitante para la  cantidad de OL que  se  forma en esta bacteria  [Gao et  al., 2004].  Sin embargo, muchas  otras bacterias sintetizan OLs en cantidades relativamente altas aún cuando se crecen en  medios  de  cultivo  ricos  en  fósforo  [Palacios‐Chaves  et  al.,  2011; Vences‐Guzman  et  al.,  2011; González‐Silva et al., 2011].    26   Figura 6. Biosíntesis y estructura del lípido de ornitina de S. meliloti 1021. Tomada y traducida de Geiger et al. (2010).    3.1.2.1.5.2.1. La N‐aciltransferasa OlsB  Mutantes  deficientes  en  el  gen  olsB  derivadas  de  S.  meliloti  [Gao  et  al.,  2004],  Rhodobacter  capsulatus  [Aygun‐Sunar et al., 2006], Brucella abortus  [Palacios‐Chaves et  al., 2011] y B. cenocepacia [González‐Silva et al., 2011] son incapaces de formar LOLs, OLs  y cualquier derivado de estos. La expresión de olsB de S. meliloti 1021 en E. coli causó  la  formación  de  LOL  [Gao  et  al.,  2004].  El  resultado  anterior  sugiere  que OlsB  es  una N‐ aciltransferasa que condensa ornitina con un grupo acilo 3‐hidroxilado para formar el LOL,  y por lo tanto, cataliza el primer paso de la biosíntesis de los OLs. Las predicciones indican  que  la enzima OlsB es de 296 aminoácidos y que es  soluble en agua. La caracterización  funcional  de  OlsB  definió  una  función  concreta  para  un  grupo  completo  de  proteínas  ortólogas  (COG3176)  previamente  asignadas  como  hipotéticas  o  como  hemolisinas  putativas [Gao et al., 2004]. OlsB pertenece a la superfamilia de acil‐CoA N‐aciltransferasas  [Gough et al., 2001].  27 Heath  y  Rock  [1998]  describieron  un  motivo  consenso  (H(X)4D)  común  de  aciltransferasas de glicerolípidos y demostraron que un cambio en la H conservada de este  motivo por una A eliminó  la actividad de  la glicerol‐3‐fosfato O‐aciltransferasa PlsB de E.  coli. En OlsB, H87 y D92 podría ser ese motivo, el cual, está conservado en  las proteínas  homólogas  a  OlsB  [Gao  et  al.,  2004].  Una  búsqueda  en  PhyloFacts  con  OlsB  de  B.  cenocepacia J2315 identificó a las sintasas de acil homoserina lactona EsaI [Watson et al.,  2002] y LasI [Gould et al., 2004], y las predicciones sugieren que la estructura terciaria de  OlsB es  similar a  la de estas  sintasas de autoinductores. Como OlsB,  las  sintasas de acil  homoserina  lactona  son  N‐aciltransferasas  que  usan  acil‐ACPs  y  un  derivado  de  un  aminoácido  (SAM)  como  sustrato  (fig.  7).  La  sintasa  FeeM  de N‐acil  aminoácido  de  un  microbio  de  suelo  no  cultivado  se  une  a  la  proteína  portadora  de  grupos  acilos  Feel  catalizando  la formación de N‐acil tirosina y su estructura se asemeja a  la de sintasas de  acil homoserina lactona EsaI y LasI [Wagoner y Clardy, 2006].      Figura  7.  Síntesis  de  3‐ceto‐dodecanoil‐homoserina  lactona,  molécula  señal  del  quórum  sensing  de  Pseudomonas  aeruginosa. Tomada y traducida de Geiger et al. (2010).    Análisis  genómico  indica  que  como  muchos  otros  grupos  de  bacterias,  el  orden  Rhodobacterales tienen un operón olsBA (contiene el olsB1) y además tienen otro gen que  codifica para otro producto homólogo a OlsB (olsB2) (fig. 8). El gen olsB2 no está localizado  28 físicamente  cercano  al operón olsBA pero  está presente  en  su  genoma.  Los homólogos  OlsB2  de  diferentes  Rhodobacterales  están más  cercanamente  relacionados  entre  ellos  que  con  los  verdaderos  homólogos  OlsB1  (fig.8).  Interesantemente,  Rhodobacter  sphaeroides forma  lípidos con  la estructura aciloxi‐acil que contienen glutamina, además,  de los OLs conocidos [Zhang et al., 2009]. El primer paso de la biosíntesis de los lípidos que  contiene  glutamina  podría  ser  catalizado  por  el  homólogo  OlsB2  encontrado  exclusivamente en Rhodobacterales [Geiger et al., 2010].      Figura  8.  Árbol  filogenético  sin  raíz  de  OlsB  de  Sinorhizobium  meliloti,  OlsB  de  Burkholderia  cenocepacia  y  ORFs  similares a OlsB. El árbol fue construido usando el programa CLUSTALW (http://www.expasy.ch/). Las distancias entre las  secuencias están expresadas  como 0.08  cambios por aminoácido.  Los números de acceso  son  los  siguientes: OlsB de  Sinorhizobium meliloti 1021 (NP_384499), OlsB de Burkholderia cenocepacia J2315 (YP_002230419), HlyC de Escherichia  coli CFT073 (NP_755444), EsaI de Pantoea stewartii subsp. stewartii (AAA82096), LasI de Pseudomonas aeruginosa PAO1  (NP_250123),  la N‐aciltransferasa FeeM de una bacteria no cultivada (AAM97306), Agrobacterium tumefaciens str. C58  (NP_353376), Brucella melitensis 16M  (NP_540717), Dinoroseobacter shibae DFL 12  (YP_001532815 y YP_001533050),  Loktanella vestfoldensis SKA53 (ZP_01003503 y ZP_01003690), Magnetospirillum magnetotacticum MS‐1 (ZP_00050271,  ZP_00055184,  y  ZP_00053729),  Mesorhizobium  loti  MAFF303099  (NP_104372),  Mycobacterium  tuberculosis  H37Rv  (NP_217543),  Oceanicola  batsensis  HTCC2597  (ZP_00999281  y  ZP_00999194),  Oceanobulbus  indolifex  HEL‐45  (ZP_02153706 y ZP_02153839), Octadecabacter antarticus 307 (EDY79302 y EDY80135), Paracoccus denitrificans PD1222  (ZP_00631241  y  ZP_00629627),  Phaeobacter  gallaeciensis  2.10  (ZP_02148428  y  ZP_02148788),  Pseudomonas  aeruginosa PAO1 (NP_253040), Rhodobacter sphaeroides 2.4.1 (YP_354511 y YP_352676), Roseobacter denitrificans OCh  114  (YP_682889 y YP_683344), and Silicibacter pomeroyi DSS‐3  (YP_167215 y YP_167705). OlsB1 y OlsB2  son  los dos  subgrupos diferentes de homólogos OlsB presentes en las bacterias del orden Rhodobacterales. Tomada y traducida de  Geiger et al. (2010).  29 Los únicos otros ejemplos donde múltiples están presentes homólogos de olsB en un  mismo  genoma  son  las  α‐proteobacterias  Magnetospirillum  magneticum  y  Magnetospirillum  magnetotacticum  (fig.  8).  Además  de  los  OLs,  otros  aminolípidos  desconocidos han sido descritos en  la membrana del magnetosoma de Magnetospirillum  los  cuales  podrían  ser  sintetizados  por  el  segundo  homólogo  [Schüler,  2004].  Notablemente,  la  expresión  heteróloga  de  DNA  microbiano  extraído  directamente  de  muestras ambientales condujo a la identificación de N‐acilos de cadena larga derivados de  tirosina, fenilalanina, triptófano y arginina, todos ellos tienen actividad antibiótica [Brady  et al., 2002; Brady y Clardy, 2005; Clardy y Brady, 2007]. Es posible que algunas de  las N‐ aciltransferasas que transfieren acilos cadena  larga  involucradas en  la formación de estos  compuestos catalicen el paso inicial en la biosíntesis de otros lípidos de membrana con la  estructura aciloxi‐acil [Geiger et al., 2010].  Geiger  et  al.  (2010)  han  especulado  que  la  biosíntesis  de  OLs  pudiera  ser  más  complicada  de  lo  que  se  ha  descrito,  debido  a  que  en  un  análisis  de  los  genomas  de  bacterias secuenciadas encontraron que varias bacterias del orden Alteromonadales y en  la ε‐proteobacteria Arcobacter butzleri la proteína OlsB está fusionada en su N‐terminal a  un dominio de una proteína hipotética de función desconocida.   3.1.2.1.5.2.2. La O‐aciltransferasa OlsA  Mutantes  deficientes  en  olsA  derivadas  de  S.  meliloti  1021  [Weissenmayer  et  al.,  2002], R. capsulatus  [Aygun‐Sunar et al., 2006], Brucella abortus  [Palacios‐Chaves et al.,  2011]  y  Pseudomonas  aeruginosa  [Lewenza  et  al.,  2011]  también  fueron  incapaces  de  formar  OLs.  La  sobreexpresión  olsB  en  una  mutante  deficiente  de  olsA  derivada  de  S.  meliloti 1021 provocó la acumulación del LOL [Gao et al., 2004]. Las predicciones sugieren  que  OlsA  de  S.  meliloti  es  una  proteína  de  292  aminoácidos  con  una  probable  hélice  transmembranal cercana al N‐terminal.  Un alineamiento de OlsA con algunas enzimas procariotas que exhiben las actividades  de  aciltransferasas  de  ácido  lisofosfatídico  [Aygun‐Sunar  et  al.,  2006]  muestra  que  hay  principalmente dos regiones bien conservadas  (los aminoácidos 67‐83 y 139‐154) donde  OlsA  tiene  la mayor  similitud  con  otros miembros  de  ese  grupo.  En  aciltransferasas  de  30 ácido lisofosfatídico, dos motivos, NHQS y PEGTR, están conservados [West et al., 1997], y  se  encuentran  en  formas  modificadas  (NHVS,  aminoácidos  72‐75;  PEGTT,  aminoácidos  143‐147)  en  la  secuencia  de  OlsA  [Weissenmayer  et  al.,  2002].  Basados  en  el  motivo  (H(X)4D) común a aciltransferasas de glicerolípidos en OlsA, H73 y D78 podrían formar ese  motivo. Es sorprendente, sin embargo, que las secuencias que codifican para las enzimas  con actividades de aciltransferasas de ácido lisofosfatídico en general son muy diferentes.  Por ejemplo, en la bacteria Neisseria meningitidis hay tres enzimas (NlaA, NlaB, y la tercera  actividad que se detectó en  la mutante doble deficiente en nlaA y nlaB) con actividad de  aciltransferasas de ácido lisofosfatídico in vitro [Shih et al., 1999]. Aunque NlaA y NlaB son  del mismo organismo, ellas  son muy diferentes y es  interesante observar que mutantes  deficientes  en  nlaA  y  mutantes  deficientes  en  nlaB  muestran  diferentes  fenotipos  sugiriendo  que  NlaA  y  NlaB  realizan  diferentes  reacciones  bioquímicas  in  vivo  en  N.  meningitidis.  Por  su  secuencia, OlsA  claramente  se agrupa dentro de  las aciltransferasas de ácido  lisofosfatídico caracterizadas. Sin embargo, mutantes deficientes en olsA de S. meliloti son  incapaces  de  formar  OLs,  y  no  muestran  acumulación  de  ácido  lisofosfatídico  ni  debilitamiento  en  la  biosíntesis  de  glicerofosfolípidos.  Por  lo  tanto,  debe  de  haber  una  actividad  idéntica  a  la  de  PlsC  en  S.  meliloti  responsable  de  estas  últimas  funciones,  presumiblemente SMc00714 [Basconcillo et al., 2009]. OlsA es requerida para la actividad  de  O‐aciltransferasa  dependiente  de  acil‐AcpP  y  LOL,  que  convierte  el  LOL  en  OL  [Weissenmayer  et  al.,  2002].  Además  de  OlsA,  Pseudomonas  fluorescens  posee  dos  homólogas a aciltransferasas de ácido  lisofosfatídico, HdtS y PatB [Cullinane et al., 2005].  HdtS y PatB son capaces de complementar el crecimiento de una mutante deficiente en  PlsC de E. coli, mientras que OlsA de P. fluorescens no complementa. Aunque HdtS o PatB  pueden realizar la función de PlsC in vivo ellas no son idénticas funcionalmente. Mutantes  carentes  de  PatB  muestran  crecimiento  reducido  en  altas  temperaturas  mientras  que  mutantes deficientes en HdtS están afectadas en crecimiento, motilidad y tienen reducida  cantidad de ácido cis‐vaccénico [Cullinane et al., 2005].  31 También  Rhodobacter  capsulatus  posee  tres  homólogos  a  aciltransferasas  de  ácido  lisofosfatídico, OlsA, PlsC316  y PlsC3498  [Aygun‐Sunar et  al., 2007]. OlsA  y PlsC316  son  capaces de  complementar  el  crecimiento de una mutante deficiente de PlsC de  E.  coli,  mientras  que  PlsC3498  no  complementa.  Una  mutante  deficiente  de  PlsC316  tiene  cantidades  reducidas de  ácidos  grasos C16  [Aygun‐Sunar et  al., 2007].  Se ha observado  que OlsA de R. capsulatus también es capaz de acilar a 1‐acilo‐sn‐glicerol‐3‐fosfato además  de  al  LOL,  exhibiendo  especificidad  relajada  hacia  el  sustrato  aceptor de  acilos  [Aygun‐ Sunar et al., 2007]. Se espera que estudios futuros más detallados con el grupo actual de  “aciltransferasas de ácido lisofosfatídico” revelaran numerosos subgrupos con actividades  bioquímicas ligeramente diferentes [Geiger et al., 2010].  Actualmente,  se  conoce  poco  sobre  las  funciones  que  desempeñan  los  OLs  en  las  bacterias que  los poseen. Las mutantes de S. meliloti [Weissenmayer et al., 2002; López‐ Lara et al., 2005] y de P. aeruginosa [Lewenza et al., 2011] deficientes en la biosíntesis de  OLs no muestran ninguna alteración sobresaliente en su fenotipo macroscópico cuando se  compara con  la cepa silvestre. Solamente en una doble mutante de S. meliloti deficiente  en  la  biosíntesis  de  OLs  y  de  DGTS  se  observó  reducida  producción  celular  cuando  se  creció en condiciones  limitadas en fósforo  [López‐Lara et al., 2005]. En R. capsulatus,  los  OLs se requieren para el óptimo estado de equilibrio de las cantidades de citocromos tipo‐ c  en  la  membrana  [Aygun‐Sunar  et  al.,  2006].  Estudios  en  otras  bacterias  han  correlacionado  un  incremento  en  la  resistencia  a  péptidos  antimicrobiales  con  la  producción  OLs  [Minnikin  y  Abdolrahimzadeh,  1974;  Dorrer  y  Teuber,  1977].  Pero  en  estudios recientes han demostrado que en Brucella abortus y P. aeruginosa los OLs no se  requieren para  incrementar esa  resistencia  [Palacios‐Chaves et al., 2011; Lewenza et al.,  2011].  3.1.2.1.5.2.3. Las hidroxilasas de lípidos de ornitina OlsC y OlsE  En  algunas  especies de  los  géneros Burkholderia,  Flavobacterium  [Asselineau,  1991;  Kawai  et  al.,  1988],  Thiobacillus  [Knoche  y  Shively,  1972], Gluconobacter  [Tahara  et  al.,  1976b],  Streptomyces  [Asselineau,  1991],  Ralstonia  [Galbraith  et  al.,  1999]  y  Rhizobium  [Vences‐Guzmán et al., 2011], existen OLs en  los cuales el grupo acilo unido por enlace  32 éster posee un grupo hidroxilo en el C2 (2‐OH‐OLs). Estos acilos esterificados son similares  a  los que existen en el  lípido A de algunas bacterias, mencionados anteriormente (fig. 2).  Los residuos de acilo 2‐OH no se forman mediante la biosíntesis normal de ácidos grasos y  una actividad enzimática específica es necesaria para formar el grupo hidroxilo en el C2 de  un residuo de ácido graso.  Rhizobium tropici CIAT899 es un simbionte de plantas de frijol, tolerante a pHs ácidos y  a  altas  temperatura  [Martínez‐Romero  et  al.,  1991; Vences‐Guzmán  et  al.,  2011]  forma  cuatro clases moleculares de OLs (nombrados S1, S2, P1 y P2) [Rojas‐Jiménez et al., 2005].  Una mutante deficiente en olsC (899‐olsCΔ1) no forma P1 ni P2. El producto codificado por  olsC es una putativa dioxigenasa similar a LpxO que puede convertir las dos formas menos  polares de  los OLs (sustratos, S1 y S2) a  las dos formas más polares (productos, P1 y P2)  [Rojas‐Jiménez et al., 2005]. Recientemente se demostró que la conversión de S1 y S2 a P1  y P2,  respectivamente, es por una hidroxilación en el C2 del  acilo esterificado  [Vences‐ Guzmán et al., 2011].  Una predicción  indica que OlsC de R. tropici CIAT899 es una proteína soluble en agua  de 281 aminoácidos  [Rojas‐Jiménez et al., 2005]. Recientemente también se  identificó el  gen olsE que codifica para la enzima que modifica a los OLs S1 para convertirlo en los OLs  S2  [Vences‐Guzmán  et  al.,  2011].  OlsE  presumiblemente  hidroxila  a  los  OLs  S1  en  el  residuo de ornitina y  se ha especulado que probablemente OlsE  también modifica a  los  OLs  P1  para  convertirlos  en  los OLs  P2  [Vences‐Guzmán  et  al.,  2011].  Las  predicciones  indican que la proteína OlsE (331 aminoácidos) es muy hidrofóbica y puede formar entre 4  y 6 hélices transmembranales [Vences‐Guzmán et al., 2011]. La mutante en olsE (MAV04)  no formó los OLs S2 y P2, y la doble mutante olsC/olsE (MAV05) no formó a los OLs S2, P1  y  P2  [Vences‐Guzmán  et  al.,  2011].  En  R.  tropici  CIAT899  las  cuatro  clases  de  OLs  se  localizan principalmente en la membrana externa [Vences‐Guzmán et al., 2011].  33   Fig. 9. Biosíntesis de los lípidos de ornitina en Rhizobium tropici CIAT899. Los genes que codifican para las enzimas OlsB  y OlsA se identificaron por primera vez en S. meliloti 1021 [Weissenmayer et al., 2002; Gao et al., 2004], mientras que el  gen que codifica para la hidroxilasa OlsC de OLs se describió por primera vez en R. tropici CIAT899 [Rojas‐Jiménez et al.,  2005].  Vences‐Guzmán  et  al.  (2011)  determinaron  que  OlsC  hidroxila  al  C2  del  ácido  graso  secundario  y  también  identificaron el gen que codifica para la hidroxilasa OlsE, la cual introduce un grupo hidroxilo en el residuo de ornitina de  los OLs. Lisolípido de ornitina (LOL), lípido de ornitina (OL). Tomada y traducida de Vences‐Guzman et al. (2011).    Notablemente,  grupos  2‐OH  en  acilos  esterificados  han  sido  reportados  en  los  esfingolípidos y PE de bacterias, otros componentes mayoritarios de la membrana externa  de bacterias Gram negativas.  Especies moleculares de PE hidroxilada  en  el C2 del  acilo  unido a la posición sn‐2 (2‐OH‐PE) se forman en miembros del género Burkholderia [Phung  et  al.,  1995;  Taylor  et  al.,  1998].  En  Burkholderia  cepacia  NCTC  10661  los  lípidos  2‐ hidroxilados  (2‐OH‐OLs  y  2‐OH‐PE)  se  incrementaron  cuando  esa  cepa  creció  en  altas  temperaturas [Taylor et al., 1998]. La bacteria Thiobacillus thiooxidans crece en ambientes  34 muy ácidos y debido a la localización de OLs en la membrana externa se especuló que esos  lípidos podrían estar involucrados en la resistencia a acidez [Dees y Shively, 1982].  Rojas‐Jiménez et al. (2005) reportaron que  la cepa 899‐olsCΔ1 creció a pH 4.5 de una  forma  similar  a  la  cepa  silvestre,  ambas  cepas  alcanzaron  una  OD  a  600  nm  de  aproximadamente 1.6. Sin embargo,  la complementación de  la mutante en olsC a pH 4.5  con  un  fragmento  que  contenía  a  olsC  en  un  vector  de  alto  número  de  copias,  afectó  seriamente el crecimiento de dicha cepa alcanzando una OD a 600 nm menor a 0.4 [Rojas‐ Jiménez et al., 2005]. Al crecer en medio TY y marcar los lípidos radiactivamente de la cepa  899‐olsCΔ1 complementada, Rojas‐Jiménez et al.  (2005) reportaron que esta cepa formó  mayoritariamente los OLs P1 y P2, y una ausencia casi total de S1 y S2. La cepa 899‐olsCΔ1  derivada de R. tropici CIAT899 formó nódulos en las plantas de frijol poco desarrollados 21  días después de la inoculación con la bacteria, los nódulos carecían de lenticelas y fijaron  dos veces menos nitrógeno [Rojas‐Jiménez et al., 2005].  Recientemente Vences‐Guzmán et al.  (2011)  reportaron estudios más detallados del  perfil lipídico de R. tropici CIAT899 y de las cepas mutantes (899‐olsCΔ1, MAV04 y MAV05)  crecidas en medio TY a diferentes temperaturas (30 ⁰C, 37 ⁰C y 42 ⁰C) y pHs (7.0, 4.5 y 4.0).  Mutantes en olsC que no hidroxilan OLs son más sensibles a estrés térmico y ácido, y  las  tres  mutantes  carentes  de  hidroxilasas  de  OL  establecieron  una  simbiosis  deficiente  [Vences‐Guzmán  et  al.,  2011].  Las  diferencias  obtenidas  entre  los  estudios  de  Rojas‐ Jiménez  et  al.  (2005)  y  el  de  Vences‐Guzmán  et  al.  (2011)  posiblemente  se  deben  a  diferencias en los medios de cultivo [Vences‐Guzmán et al., 2011].  3.1.2.1.5.3. Lípidos de ornitinataurina hidoxilados  En Gluconobacter cerinus,  los 2‐OH‐OLs son parcialmente modificados con un residuo  de taurina el cual está  ligado por enlace amida a el grupo α‐carboxilo de  la ornitina  (fig.  10)  [Tahara  et  al.,  1978].  Se  ha  observado  que  una  fracción  particular  de  G.  cerinus  requiere ATP y Mn2+ para condensar taurina con los 2‐OH‐OLs para formar los lípidos que  contienen ornitinataurina 2‐hidroxilados (2‐OH‐TOLs). Los 2‐OH‐TOLs han sido nombrados  cerilipina  debido  al  nombre  de  la  bacteria  de  la  cual  se  aisló  [Tahara  et  al.,  1978].  35 Actualmente desconocemos el gen que codifica para  la enzima que hace  la actividad de  condensar taurina con los 2‐OH‐OLs.    2‐OH‐TOL  Figura 10. Lípido de ornitinataurina 2‐hidroxilado (2‐ OH‐TOL) de Gluconobacter cerinus. Tomada de Geiger  et al. (2010).    3.1.2.2.  La  respuesta  inmune  innata  hacia  los  lípidos  de membrana  con  la  estructura  aciloxi‐acil de bacterias  Los  lípidos  de  membrana  con  la  estructura  aciloxi‐acil  de  bacterias  inducen  drásticamente  la  respuesta  inmune de mamíferos  al  ser  reconocidos por  los  receptores  tipo  toll  (TLRs)  como  patrones  moleculares  asociados  a  patógenos.  El  ejemplo  mejor  estudiado es la inducción que ocasiona el lípido A, la parte reactiva del LPS, que estimula a  las células de la inmunidad innata usando como mediadores de la señal al TLR4 y al factor  nuclear кB. En la ruta de señalización del lípido A también está involucrada la proteína MD‐ 2,  la  cual,  se  asocia  físicamente  con  TLR4  en  la  superficie  celular,  confiriéndoles  sensibilidad a  los receptores TLR4s hacia  las moléculas de LPSs [Shimazu et al., 1999]. Al  igual que el lípido A de E. coli, la estructura 3‐aciloxi‐acilamida con la configuración (R) en  su  carbono  asimétrico  también  está  presente  en  los  GLs  [Batrakov  et  al.,  1999],  SGLs  [Shiozaki et al., 1998ab; Gomi et al., 2002] y en OLs [Kawai et al., 1999] (figs. 2, 4, 5 y 6).  36 Sin  embargo,  solamente  se  ha  evaluado  la  capacidad  de  inducir  la  respuesta  inmune  innata  de  los  SGLs  y  de  los  OLs.  La  respuesta  inmune  innata  se  midió  por  la  producción  de  prostaglandina  E2,  interleucina  1β  y  del  factor  α  de  la  necrosis  tumoral  cuando  los  lípidos  se expusieron a  los macrófagos  [Kawai  y Akagawa 1989]. Un estudio  reciente  sugiere  que  el  estado  físico  en  el  que  se  aplican  el  lípido  A  y  el  OL  a  los  macrófagos, afecta la actividad biológica de estos lípidos [Okemoto et al., 2008]. Los OLs y  los SGLs pueden ser usados como coadyuvantes  [Kato y Goto, 1997; Kawai et al., 1999;  Kawai et al., 2002] y cuando se  inyectan en ratones antes que el  lípido A puede prevenir  los  efectos  letales  de  este  último,  aunque  el  efecto  protector  de  SGLs  fue más  débil  y  debido  a  eso  se  especuló  que  los  OLs  y  SGLs  podrían  funcionar  como  antagonistas  bloqueadores de los eventos provocados por el agonista lípido A [Kawai et al., 1992].  Como  el  lípido  A,  la  señal  que  causa  la  respuesta  inmune  innata  por  SGLs  es  traducida vía el complejo TLR4/MD‐2 [Gomi et al., 2002]. La señalización mediada por el  complejo TLR4/MD‐2 se incremento por los SGLs cuando se expresó CD14 [Kawasaki et al.,  2003]. La proteína CD14 actúa como un correceptor  (junto con el TLR4 y MD‐2) para  la  detección del LPS. Kawasaki et al.  (2003) también demostró que  la configuración  (R) del  residuo  lipídico  de  los  SGLs  es  esencial  para  la  inducción  de  la  señal  mediada  por  el  complejo  TLR4/MD‐2,  y  que  la  configuración  del  residuo  de  serina  y  la  secuencia  de  aminoácidos en el dominio de aminoácidos no  son  importantes para  la  inducción de  la  señal. Los componentes que participan en traducción de la señal inducida por los OLs aún  no han sido identificados y sugerimos que debido a la similitud estructural de los OLs con  el  lípido A y con  los SGLs, probablemente  los OLs usan  los mismos componentes que el  lípido A y los SGLs ya que la respuesta inmune que provocan también es similar.      37 4. Antecedentes  La  bacteria  Gram  negativa  Burkholderia  cenocepacia  J2315  es  un  miembro  del  complejo Burkholderia cepacia (BCC), un subgrupo de importantes patógenos oportunistas  muy virulentos del género Burkholderia que infecta a humanos con fibrosis quística, con la  enfermedad granulomatosa  crónica y a  individuos  inmunosuprimidos  causando una alta  tasa de mortalidad  [Mahenthiralingan et al., 2005]. Aunque se conocen muchos  factores  que  contribuyen a  la  virulencia del BCC,  frecuentemente  infecciones  con el BCC no  son  eficientemente eliminadas por los tratamientos con antibióticos comunes.  Los lípidos de membrana de B. cenocepacia J2315 son: CL, PG, PE y OLs, así como  los derivados atípicos 2‐hidroxilados de PE y OLs (2‐OH‐PE y 2‐OH‐OLs, respectivamente),  los cuales difieren de las versiones estándar por poseer un grupo hidroxilo en el C2 (2‐OH),  que sustituye a un átomo de hidrógeno, en un residuo de acilo esterificado [Phung et al.,  1995; Taylor et al., 1998]. Un OL estándar  se  compone de un  residuo de ornitina y dos  residuos  de  acilos  sin  grupos  hidroxilos  (fig.  6),  uno  de  estos  acilos  está  amidificado  al  grupo  α‐amino de  la ornitina y el otro acilo está esterificado al C3 del acilo amidificado  [Knoche y Shively, 1972; Geiger et al., 1999].  La ruta biosintética para los OLs sin hidroxilaciones se elucidó en S. meliloti 1021 y  consiste  de  dos  pasos.  En  el  primer  paso,  la  enzima OlsB  que  es  una N‐aciltransferasa  cataliza la transferencia de un grupo acilo 3‐hidroxilado de una ACP al grupo α‐amino de la  ornitina formando el lisolípido de ornitina (LOL) [Gao et al., 2004]. En el segundo paso, la  enzima OlsA que es una O‐aciltransferasa  cataliza  la  transferencia de un  segundo grupo  acilo de una ACP al grupo 3‐hidroxilo del LOL formando el OL (Fig. 6) [Weissenmayer et al.,  2002].  B.  cenocepacia  J2315  tiene  los  genes  ortólogos  que  codifican  para  las  proteínas  OlsA  y OlsB.  En este  trabajo mostramos que bcal1281  codifica para  la N‐aciltransferasa  OlsB y probablemente bcal3137 codifica para la O‐aciltransferasa OlsA.  En  2‐OH‐PE  de  B.  cenocepacia  J2315  el  residuo  de  acilo  2‐OH  está  ligado  exclusivamente a la posición sn‐2 en ese glicerofosfolípido, mientras que en los 2‐OH‐OLs  los residuos de acilos 2‐hidroxilados son los acilos esterificados [Phung et al., 1995; Taylor  et  al.,  1998].  Los  residuos  de  acilo  2‐hidroxilados  no  se  forman  durante  la  biosíntesis  38 estándar de ácidos grasos y se requieren actividades enzimáticas específicas para formar  un grupo hidroxilo en el C2.  En Salmonella enterica serovar Typhimurium (S. typhimurium) una proteína (LpxO)  se  identificó  como  la  responsable  de  formar  el  grupo  2‐OH  en  el  residuo  de  miristato  esterificado al ácido graso de la posición 3´del lípido A [Gibbons et al., 2000]. La reacción  de  hidroxilación  es  catalizada  por  la  dioxigenasa  LpxO  dependiente  de  Fe2+/O2/α‐ cetoglutarato (fig. 2) [Gibbons et al., 2008]. Se ha especulado que tales grupos hidroxilos  extras podrían  incrementar  los enlaces de hidrógeno entre  las moléculas adyacentes de  lípido  A  aumentando  la  estabilidad  de  la  membrana  externa  y  consecuentemente  disminuyendo  la  permeabilidad  a  ciertos  compuestos  en  algunas  condiciones  de  crecimiento [Nikaido, 2003; Gibbons et al., 2008]. Una enzima homóloga a LpxO, OlsC de  R. tropici CIAT899 está  involucrada en  la formación de  las dos clases de OLs más polares  (P1 y P2), a partir de las dos menos polares (S1 y S2), existentes en ese organismo [Rojas‐ Jiménez et al., 2005]. Mientras se desarrollaba este trabajo, Vences‐Guzmán et al. (2011)  reportaron que OlsC hidroxila  in vivo a  los  lípidos S1 y S2 en el C2 del acilo esterificado  para convertirlos en P1 y P2, respectivamente.  Como  el  lípido A,  los OLs  están  en  la membrana  externa  [Dees  y  Shively,  1982;  Palacios‐Chaves et al., 2011; Vences‐Guzmán et al., 2011] y por  lo  tanto en  la superficie  celular. Cuando las bacterias entran a un mamífero, sus patrones moleculares asociados a  patógenos  de  la  superficie  celular  activan  el  sistema  inmune  de  los  mamíferos  y  normalmente  las  bacterias  son  fagocitadas  por  los  macrófagos  y  dirigidas  al  compartimento  lisosomal  para  digerirlas.  En  ese  compartimento  los  nutrimentos  son  restrictivos  para  las  bacterias  y  en  esas  condiciones  teóricamente  se  favorece  la  producción de OLs. Por lo tanto, los OLs podrían desempeñar papeles protagónicos para la  sobrevivencia en el  interior de macrófagos de algunas bacterias que poseen  la capacidad  de sintetizarlos.  La  función que desempeña  la modificación 2‐OH en PE  y OLs en B.  cenocepacia  J2315  no  se  conoce.  Para  estudiarla  optamos  por  tratar  de  identificar  los  genes  que  potencialmente  podrían  hacer  la  modificación  2‐OH  en  los  residuos  de  acilo  2‐ 39 hidroxilados,  generar mutantes  en  esos  genes  y  crecerlas  en  diferentes  condiciones.  El  genoma de B. cenocepacia  J2315 contiene dos marcos abiertos de  lectura  (BCAM1214 y  BCAM2401  [OlsD])  homólogos  a  LpxO  de  S.  typhimurium,  anotados  como  putativas  β‐ hidroxilasas.      40 5. Hipótesis:    Los  2‐OH‐OLs  en  Burkholderia  cenocepacia  J2315  se  forman  a  partir  de  OLs  no  hidroxilados  mediante  una  nueva  2‐hidroxilasa  homóloga  a  LpxO  de  Salmonella  typhimurium.    6. Objetivo general:    Identificar genes que codifiquen para enzimas  involucradas en hidroxilar a  los OLs  de  B.  cenocepacia  J2315  y  determinar  en  qué  parte  de  la  molécula  de  los  OLs  los  hidroxilan, así como caracterizar las enzimas que tengan esa capacidad.  6.1. Objetivos particulares:  1. Determinar si  las putativas hidroxilasas de B. cenocepacia  J2315 homólogas a  LpxO, hidroxilan a los OLs.  2. Determinar  la  parte  de  las  moléculas  de  los  OLs  hidroxilada  por  la(s)  hidroxilasa(s).  3. Desarrollar un ensayo enzimático para la(s) hidroxilasa(s).  4. Construir mutantes de B. cenocepacia J2315 en los genes homólogos a lpxO de  S. typhimurium así como en el gen olsB y analizar los fenotipos de las mutantes  generadas.      41 7. Procedimientos experimentales de los resultados adicionales  7.1. Deacilación química de lípidos de ornitina  Se  modificó  OL  radiactivo  in  vitro  con  OlsD,  como  se  ha  descrito  previamente  [González‐Silva at al., 2011] y los lípidos modificados y no modificados se extrajeron por el  método de Bligh y Dyer (1959). Después se separaron por TLC con el sistema de solventes  cloroformo: metanol: ácido acético glacial (130:50:20, v/v/v). El gel de silica que contenía  los lípidos se raspó y los lípidos se extrajeron del gel con el método de Bligh y Dyer (1959).  Posteriormente  los  lípidos  se  secaron  y  se  sometieron  a  un  tratamiento  alcalino  suave  (KOH 0.3 M en metanol a 37 ⁰C por 3 h); las condiciones anteriores hidrolizan los enlaces  éster. Después el pH del hidrolizado se neutralizó con ácido acético glacial y el metanol de  las  muestras  se  volatilizó  con  un  flujo  de  nitrógeno.  Posteriormente,  los  lípidos  se  extrajeron  con  acetato de etilo,  se disolvieron en metanol:  cloroformo  (1:1,  v/v),  y una  alícuota  se  cuantificó en el  contador de  centelleo. Subsiguientemente  se  cargaron en  la  capa  fina  alrededor  de  11,000  cuentas  por minuto  (cpm)  de  los OLs  no modificados  y  modificados, y 23,000 cpm de  los LOLs no modificados y modificados. Posteriormente se  corrió  la capa fina con el sistema de solventes cloroformo: metanol: acido acético glacial  (130:50:20, v/v/v). Por último, la capa fina se expuso a una pantalla del Phosphoimager y  se escaneó 3 días después.  7.2. Ensayo in vitro con OlsD usando como sustrato a los 2‐OH‐OLs  Al  introducir el plásmido pSphx04 que contenía al gen olsD en  la cepa silvestre B.  cenocepacia J2315 se formaron dos nuevos lípidos de membrana (NL1 y NL2), y el análisis  por espectrometría de masas de NL1 y NL2 sugería que estos  lípidos tenían sustituido un  átomo de hidrógeno por un grupo hidroxilo en un carbono del acilo amidificado [González‐ Silva  et  al.,  2011].  Este  resultado  indicaba  que  probablemente  OlsD  puede  usar  como  sustratos a  las dos clases de  lípidos de ornitina  (OLs y 2‐OH‐OLs) que produce esta cepa  [González‐Silva et al., 2011]. Para  reafirmar  lo anterior, procedimos a  realizar un ensayo  enzimático  in  vitro  con  OlsD  usando  como  sustrato  a  los  2‐OH‐OLs  de  B.  cenocepacia  J2315.  El  sustrato  (2‐OH‐OLs)  se  preparó  a  partir  10 ml  de  cultivo  en  LB  de  la  cepa  B.  cenocepacia pNG24, marcado con 10 μCi de acetato de sodio [1‐14C] (60 mCi/mmol) y se  42 cosechó  a  una  absorbancia  de  1.0  a  620  nm.  Los  lípidos  totales  de  membrana  se  extrajeron por  el método  de Bligh  y Dyer  (1959).  La  fase  de  cloroformo  se  separó  por  cromatografía en capa fina de dos dimensiones (2D‐TLC) [González‐Silva et al., 2011]. Los  2‐OH‐OLs radiactivos  fueron  localizados por autoradiografía y extraídos de  la  fracción de  gel de  la silica, y  los 2‐OH‐OLs se diluyeron en una mezcla de metanol: cloroformo  (1:1,  v/v). Después una  alícuota  se  cuantificó en el  contador de  centelleo. Posteriormente el  ensayo  se  realizó  similarmente a como  lo  reportaron previamente  [González‐Silva et al.,  2011]. La diferencia consistió en usar como sustrato a  los 2‐OH‐OLs en  lugar de  los OLs y  en hacerlo a un pH de 7.5. Después del ensayo,  los  lípidos radiactivos se extrajeron y se  separaron  por  TLC  usando  la  siguiente mezcla de  solventes  cloroformo: metanol:  ácido  acético  (130:50:20,  v/v/v),  y  los  lípidos  se  detectaron  y  cuantificaron  con  el  phosphorimager.  7.3. Inactivación de bcam1214  La expresión heteróloga del gen bcam1214 en S. meliloti 1021 y en B. cenocepacia  J2315  sugirió  que  este  gen  no  estaba  involucrado  en modificar  a  los OLs  ni  a  PE.  Para  demostrar convincentemente lo anterior se procedió a generar una mutante por remplazo  génico  en  dicho  gen.  Los  oligonucleótidos  oLOP91  (ATGTTGATATCTGATGGTGTATCCGATGGG)  y  oLOP92  (AAAGGATCCTATGAAAAGCAGTTCAGCG) que introducen los sitios de corte EcoRV y BamHI  (subrayados), respectivamente, se usaron en la reacción en cadena de la polimerasa (PCR)  para amplificar un  fragmento de 1076 pares de bases  (pb)  localizado  río arriba del gen  bcam1214.  Similarmente,  los  oligonucleótidos  oLOP147  (AAGGGATCCATCGTCGTCGACGGCGAATC)  y  oLOP148  (ACTCTCTAGATCCTC  GATCGCACGGTGGTC)  que  introducen  los  sitios  de  corte  BamHI  y  XbaI  (subrayados),  respectivamente,  se  usaron  en  la  PCR  para  amplificar  un  fragmento  de  641  pb  que  comprende parte del bcam1214 y de su región río abajo. El producto de la PCR amplificado  con  oLOP91  y  oLOP92  se  digirió  con  EcoRV/BamHI  y  se  clonó  en  pBluescriptSK+  previamente digerido con las mismas enzimas, resultando el plásmido pNG12. El producto  43 de  la PCR amplificado con oLOP147 y oLOP148 se digirió con BamHI/XbaI, y se clonó en  pUC18 previamente digerido con las mismas enzimas, resultando el plásmido pNG49.  Por secuenciación de los insertos de los plásmidos pNG12 y pNG49 se confirmó que  se  habían  clonado  los  fragmentos  deseados.  El  plásmido  pNG49  se  movilizó  a  la  cepa  JM110  (no metila el DNA) debido a que en DH5α  se metila el último nucleótido  (A) del  sitio  de  corte  XbaI,  lo  cual  no  permitía  liberar  el  inserto.  Posteriormente  el  plásmido  pNG12 se digirió con BamHI y XbaI y se ligó con el fragmento de DNA de 641 pb obtenido  después  de  la  digestión  con  BamHI/XbaI  de  pNG49  derivado  de  JM110,  resultando  el  plásmido  pNG51.  Después  el  plásmido  pNG51  se  linearizó  con  BamHI  y  se  ligó  con  el  fragmento BamHI del pCAT que contiene el gen para la cloranfenicol acetiltransferasa (cat)  que  confiere  resistencia  a  cloranfenicol.  El  plásmido  resultante  contenía  las  regiones  flanqueantes  de  bcam1214  interrumpidas  por  un  gen  que  confiere  resistencia  a  cloranfenicol y se llamó pNG53. El plásmido pNG53 se movilizo a la cepa JM110. Después  el  pNG53  se  extrajo  de  esa  cepa,  se  digirió  con  EcoRV/XbaI  para  reclonar  las  regiones  flanqueantes  de  bcam1214  y  el  gen  que  confiere  resistencia  a  cloranfenicol  localizado  entre  las  dos  regiones.  El  fragmento  EcoRV/XbaI  se  clonó  en  el  vector  suicida  pK18mobsacB  [Schäfer  et.  al.,  1994]  que  previamente  se  digirió  con  SmaI/XbaI,  produciendo el plásmido pNG55.  El plásmido pNG55 se introdujo en B. cenocepacia J2315 (cepa silvestre) mediante  una  cruza  biparental,  usando  como  donadora  a  la  cepa  de  E.  coli  S17‐1.  Las  transconjugantes  se  seleccionaron  en  el  medio  de  cultivo  LB  con  cloranfenicol  y  ácido  nalidíxico,  lo que también permitió contraseleccionar a E. coli S17‐1. Posteriormente una  recombinante  sencilla  se  creció  en  condiciones  no  selectivas  en  LB  hasta  alcanzar  una  absorbancia de 1.0 a 620 nm, se hicieron diluciones y se sembraron en LB con sacarosa al  10 % (peso/volumen) y cloranfenicol para seleccionar recombinantes dobles. Después de 8  días  de  crecimiento  en  el  medio  selectivo,  las  5  colonias  que  crecieron  más  rápido  se  seleccionaron para analizar  su genotipo por hibridación  tipo Southern. En  la hibridación  tipo  Southern  se  usó  como  sonda marcada  con  digoxigenina  el  fragmento  de  1076  pb,  localizado río arriba del gen bcam1214. El DNA de  la cepa silvestre y de  las candidatas a  44 mutantes en el gen bcam1214 se digirió con NcoI y PstI. Se esperaba que  la sonda en  la  cepa silvestre hibridara con un fragmento de 1654 pb y en las candidatas a mutantes con  uno  de  2130  pb.  Después  de  obtener  las  mutantes  se  analizó  su  perfil  de  lípidos  de  membrana como se ha descrito anteriormente [González‐Silva et al., 2011].  7.4. Clonación y expresión de  los genes bcal0511 y bcam2775 de B. cenocepacia  J2315  homólogos a PHYH  Ninguno de los dos genes de B. cenocepacia J2315 homólogos a lpxO (bcam1214 y  olsD)  fueron  responsables  de  codificar  para  una  enzima  que  modificara  con  un  grupo  hidroxilo al C2 del acilo esterificado de los OLs o de PE de B. cenocepacia J2315. Nosotros  también  estudiamos  otros  genes  candidatos  que  podrían  codificar  para  enzimas  responsables de hacer 2‐hidroxilaciones en el ácido graso esterificado de los OLs. Ellos son  los que codifican productos homólogos a  la enzima peroxisómica  fitanoil‐CoA hidroxilasa  (PHYH)  de  humano,  la  cual  cataliza  la  α‐oxidación  (2‐hidroxilación)  del  ácido  fitánico  proveniente del consumo de ciertos alimentos [Jansen et al., 1997; Jansen et al., 2000]. En  el  genoma  de  B.  cenocepacia  J2315  se  identificaron  dos  genes  que  codifican  para  productos homólogos a PHYH de humano: bcal0511 y bcam2775, y se procedió a clonarlos  y a expresarlos.  Los  oligonucleótidos  oLOP107  (AGGAATACATATGTCGTCTCCATTGCAGTCGG)  y  oLOP108  (AAAGGATCCCTAGAACTGCACTTCCGGC)  que  introducen  sitios  de  corte  NdeI  y  BamHI (subrayados), respectivamente, se usaron en la PCR para amplificar el gen bcal0511  (774  pb).  Los  oligonucleótidos  oLOP97  (AGGAATACATATGACTCATTCGATCGACAGG)  y  oLOP98  (AAATAAGATCTCAGATCGCGCCGACCTGCC)  que  introducen  sitios  de  corte NdeI  y  BglII (subrayados), respectivamente, se usaron en la PCR para amplificar el gen bcam2775  (875 pb). El gen bcam2775 se encuentra en la hebra reversa y su primer nucleótido era G,  en el oligonucleótido oLOP97 se sustituyó por A, para poderlo clonar con el sitio de corte  NdeI y quedara direccionalmente en fase con el marco de lectura del vector de expresión  pET9a. En ambas reacciones de PCR se usó como molde DNA genómico de B. cenocepacia  J2315. Después de  la amplificación  los productos  se digirieron  con  las  correspondientes  45 enzimas  y  los  fragmentos  obtenidos  se  clonaron  en  pET9a,  resultando  los  plásmidos  pNG26 y pNG18, respectivamente.  Por secuenciación de los insertos de los plásmidos se confirmó que pNG26 porta el  gen bcal0511 y pNG18 a bcam2775. Con los plásmidos pNG26 y pNG18 se transformaron  células competentes de  la cepa de expresión de E. coli BL21 (DE3) pLysS. Para expresar a  bcal0511 y a bcam2775, cultivos en fase estacionaria de cepas derivadas BL21 (DE3) pLysS  que  contenían  los plásmidos pET9a vacío  (control negativo), pNG26 y pNG18 en  LB  con  kanamicina  y  cloranfenicol,  se  diluyeron  1:100  en  20  ml  de  LB  con  kanamicina  y  cloranfenicol,  y  se  crecieron  a  30  ⁰C  hasta  alcanzar  una  absorbancia  de  0.3  a  620  nm.  Posteriormente se adicionó  IPTG (0.2 mM), se trasfirió 1 ml de cada cepa a un tubo tipo  Falcon  de  10 ml  y  se  adicionaron  2  µCi  de  acetato  de  sodio  [1‐14C]  (60 mCi/mmol).  El  crecimiento de todos los cultivos se continúo por 3 h. Las células de los cultivos radiactivos  se cosecharon por centrifugación a 13,400 g por 5 min, se desechó el sobrenadante y se  resuspendieron en 100 µl de agua.  Posteriormente los lípidos se extrajeron por el método de Bligh y Dyer (1959) y se  analizaron 2D‐TLC [González‐Silva et al., 2011]. Las células de los cultivos no radiactivos se  cosecharon  por  centrifugación  a  8,228  g  por  15  min,  se  desechó  el  sobrenadante,  las  células  se  resuspendieron  en  1.9  ml  de  agua  y  se  almacenaron  a  ‐20  ⁰C.  Después  la  suspensión  celular  se  congeló  y  descongeló  3  veces  para  que  las  células  se  lisaran,  y  después  el  lisado  celular  se  centrifugó  a  4,629  g  por  15  min  y  se  desechó  el  pelet.  Posteriormente, 50 µl del sobrenadante (extracto crudo) de cada cepa se mezcló con 5 µl  DNAsa  (2.5  unidades/µl)  y  se  incubó  por  37  ⁰C  por  15  min.  Subsiguientemente,  a  las  muestras anteriores se les adicionó 55 µl del bufer 2X treatment (0.125 M de Tris/HCl pH  6.8, 4 % de SDS, 20 % de glicerol y 10 % de mercaptoetanol) y se incubaron a 95 ⁰C por 5  min. Después  se  les dió un pulso en  la  centrifuga. Por último,  se  cargaron 5 µl de  cada  muestra en geles desnaturalizantes de poliacrilamida al 10 %.  Los plásmidos pNG26 y pNG18 se linearizaron con BamHI y BglII, respectivamente,  y se ligaron al vector de amplio rango de hospedero pRK404, previamente linearizado con  BamHI. Resultando los plásmidos pNG27 y pNG20 y se verificó por digestión que portaran  46 a  los genes bcal0511 y bcam2775,  respectivamente. El plásmido pNG27  se movilizó por  conjugación  a  la  cepa CS111  x pNG25  y el pNG20  a  la  cepa  S. meliloti 1021  x pILAS03,  ambas cepas  receptoras  sobreproducen OLs. Los plásmidos pNG27 y pNG20  también  se  movilizaron por conjugación a las cepas B. cenocepacia J2315 y NG1.  7.5.  Complementación  con  genes  olsBs  rizobiales  de  la  mutante  deficiente  en  olsB  derivada de B. cenocepacia J2315 (NG1)  Para conocer si la mutante NG1 se complementaba con los genes olsB de S. meliloti  1021  o  de  R.  tropici CIAT899,  se  introdujeron  los  plásmidos  pJG21  [Gao  et  al.,  2004]  y  pERMAV32 (construido por Miguel A. Vences‐Guzmán, datos no publicados) el cual resultó  de clonar el pEMAV31 (que contiene a olsB de R. tropici CIAT899 clonado en pET9a como  NdeI‐BamHI) en el pRK404  con BamHI,  a B.  cenocepacia NG1,  respectivamente. Ambos  plásmidos  se movilizaron  a  la mutante NG1  y  a  la  cepa  silvestre B.  cenocepacia  J2315,  mediante una  cruza biparental, usando  como donadora a  la  cepa de E.  coli  S17‐1. Para  comprobar fehacientemente que el plásmido pJG21 contenía el gen funcional después de  haberlo movilizado a la mutante NG1, el plásmido se extrajo por minipreparaciones de esa  cepa y  se movilizó a  la mutante deficiente en olsB derivada de  S. meliloti 1021  (AAK1),  mediante una cruza biparental, usando como donadora a la cepa de E. coli S17‐1. También  se  construyó  el  plásmido  pNG33  que  se  produjo  al  digerir  el  pJG20  con  BamHI/BglII  y  clonar el  fragmento  liberado  (olsBSm + promotor T7) en el plásmido pIML21  (contiene a  acpP  de  S. meliloti  1021  ligado  al  pRK404  como  BamHI/BglII)  construido  por  Isabel M.  López‐Lara (datos no publicados) previamente digerido con BamHI. Los plásmidos pIML21  y pNG33 se movilizaron a  las cepas NG1 y AAK1, y  los plásmidos pNG28 y pNG24 sólo se  movilizaron a  la cepa AAK1, mediante una cruza biparental, usando como donadora a  la  cepa de E. coli S17‐1. Todas  las transconjugantes se seleccionaron en el medio de cultivo  LB o PY CaCl2  [Beringer, 1974]  con  tetraciclina  y  ácido nalidíxico.  La presencia de  ácido  nalidíxico, permitió contraseleccionar a E. coli S17‐1. Posteriormente se analizó el perfil de  lípidos de todas  las transconjugantes mencionadas en este párrafo como ha sido descrito  anteriormente [González‐Silva et al., 2011].  47 7.6. Ensayo de sensibilidad a antibióticos con cepas derivadas de B. cenocepacia J2315  Mutaciones que causan cambios en la composición lipídica de la membrana de las  bacterias  frecuentemente  alteran  la  sensibilidad  a  antibióticos.  Para  comparar  la  sensibilidad a antibióticos de  las cepas  J2315, AQ3, AQ3 x pJG16, AQ3 x pSphx04, NG1,  NG1  x  pNG28  y  NG1  x  pNG24  se  crecieron  por  3  días  en  cajas  de  LB,  se  adicionó  tetraciclina  si  contenían  plásmido.  Posteriormente  una  colonia  aislada  de  cada  cepa  se  inoculó en 10 ml de LB líquido, con tetraciclina para las cepas que contenían plásmido y se  dejaron  crecer  hasta  una  absorbancia  de  0.3  a  620  nm.  Después  se  distribuyeron  uniformemente  50  µl  de  cada  cepa  sobre  la  superficie  de  LB  de  cajas  petri  (todas  sin  antibióticos). Posteriormente se colocaron los discos que contenían los antibióticos (tabla  1) del kit (Oxoid), se incubaron a 30 ⁰C por 4 días y finalmente se comparó la sensibilidad  mediante  la  medición  del  diámetro  de  las  zonas  transparentes  (halos  de  inhibición)  alrededor de los discos.  Tabla 1. Antibióticos usados en los ensayos de sensibilidad.  Código de los antibióticos  Nombre Unidades/µg Blanco TE30  tetraciclina 30 subunidad del ribosoma 30S  C30  cloranfenicol 30 subunidad del ribosoma 50S  VA30  vancomicina 30 pared celular PB300  polimixina B 300 membranas celulares  B10  bacitracina 10 membranas celulares  E15  eritromicina 15 subunidad del ribosoma 50S  F300  nitrofurantoína 300 DNA y metabolismo energético  TOB10  tobramicina 10 subunidad del ribosoma 30S  W5  trimetoprima 5 metabolismo del ácido fólico  N30  neomicina 30 subunidad del ribosoma 30S  CN10  gentamicina 10 subunidad del ribosoma 30S  S10  estreptomicina 10 subunidad del ribosoma 30S  AMP10  ampicilina 10 pared celular FOX30  cefoxitina 30 pared celular DA2  clindamicina 2 subunidad del ribosoma 50S  K30  kanamicina 30 subunidad del ribosoma 30S  PRL1000  piperacilina 1000 pared celular NA30  ácido nalidíxico 30 DNA girasa 48 7.7. Clonación de las dos versiones de olsC en pNG23 para cooexpresarlas con olsB de B.  cenocepacia J2315  Rojas‐Jiménez et al. (2005) construyeron una mutante en el gen olsC por deleción  parcial  de  211  pb  que  comprendía  la  región  predicha  que  codificaba  para  el  dominio  catalítico de  la enzima OlsC. Rojas‐Jiménez et al.  (2005)  también demostraron que OlsC  está  involucrada en  convertir  las dos  clases de OLs menos polares  (S1  y  S2) en  las dos  clases  de  OLs  más  polares  (P1  y  P2)  y  complementó  la  mutante  con  el  plásmido  pBBR1,6BE,  el  cual  contenía  un  fragmento  de  1660  bp  BamHI‐EcoRI  clonado  en  pBBR‐ MCS5. El  fragmento clonado contenía al gen olsC y su  región  rio arriba que  incluye una  región  promotora  predicha.  Basados  en  análisis  con  diferentes  programas  computacionales para predecir ORFs, Rojas‐Jiménez et al. (2005) determinaron que el gen  estructural es de 846 pb en el marco +1 y que comprendía de la posición 2,611 a la 3456  de la secuencia de 3,761 b depositada en GeneBank con el número de acceso AY954450.  Sin  embargo,  Rojas‐Jiménez  et  al.  (2005)  no  mostraron  evidencia  experimental  con  el  marco  individual  predicho  para  olsC.  Extraordinariamente,  algunos  programas  computacionales para la predicción de ORFs también indican que el gen olsC es de 900 pb  y que inicia en posición 2557 en el marco +1. Por  lo tanto, se predicen teóricamente dos  versiones una corta (846 pb) y una larga (900 pb) de olsC.  Para determinar qué versión es funcional o si ambas lo son, Christian Sohlenkamp  amplificó  y  clonó  ambas  versiones  de  olsC  en  pET9a  como  NdeI‐BamHI  (datos  no  publicados). Los plásmidos pCCS98 y pCCS82 que contienen las dos versiones corta y larga,  respectivamente, se digirieron con BamHI‐BglII para liberar los insertos y se reclonaron en  el pNG23, previamente digerido con BamHI, produciendo  los plásmidos pNG29 y pNG30,  respectivamente. Los plásmidos pNG29 y pNG30 se linearizaron con BamHI y se ligaron al  pRK404,  previamente  digerido  BamHI,  produciendo  los  plásmidos  pNG31  y  pNG32,  respectivamente. La  ligación de  los plásmidos pNG29 y pNG30 al pRK404 se verificó por  digestión de pNG31 y pNG32, respectivamente. Los plásmidos pNG24, pNG31 y pNG32 se  introdujeron por transformación en células competentes de E. coli S17‐1 y de esa cepa se  49 movilizaron por conjugación a la cepa CS111. Posteriormente se analizó el perfil de lípidos  de membrana de las cepas CS111 x pNG24, CS111 x pNG31 y CS111 x pNG32.  7.8. Clonación de olsC en pET16b, su expresión, y los ensayos enzimáticos para OlsC  El fragmento liberado por la digestión del plásmidos pCCS98 con NdeI/BamHI (olsC)  se ligó al pET16b, previamente digerido con NdeI/BamHI, produciendo el plásmido pNG42.  Posteriormente  el  plásmido  pNG42  se  linearizó  con  BamHI  y  se  ligó  al  pRK404,  previamente digeridó con la misma enzima, produciendo el plásmido pNG43. El pNG43 se  movilizó  por  conjugación  a  la  cepa  mutante  en  olsC  899‐olsCΔ1  derivada  de  R.  tropici  CIAT899, usando como  cepa donadora a  la E. coli S17‐1. Después  se analizó el perfil de  lípidos de membrana de  la cepa 899‐olsCΔ1 con el plásmido pNG43. El plásmido pNG42  también  se  introdujo  por  transformación  a  la  cepa  BL21  (DE3)  x  pLysS,  se  indujo  su  expresión y se prepararon  los extractos crudos, una parte del extracto crudo se usó para  separar por ultracentrifugación la fracción soluble y la insoluble. Se cuantificó una alícuota  del  extracto  crudo  y  de  las  fracciones  soluble  e  insoluble  para  estimar  la  cantidad  de  proteína  total  que  contenían.  Después  se  realizaron  los  ensayos  enzimáticos  con  los  extractos crudos y con  las fracciones soluble e  insoluble. A partir de  la expresión de olsC  todos  los  procedimientos  posteriores  se  realizaron  similarmente  a  como  los  describió  González‐Silva  et  al.  (2011)  para  OlsD;  como  control  negativo  se  usaron  los  extractos  crudos de la cepa BL21 (DE3) x pLysS x pET16b. Las únicas diferencias entre los ensayos de  OlsD y OlsC es que en los ensayos de OlsC pusimos una menor cantidad de proteína total  (0.4 mg/ml) del extracto crudo, de la fracción soluble y de membranas, así como 0.2% de  Triton X‐100 en lugar de 0.1%. Después del ensayo los lípidos radiactivos se extrajeron y se  separaron  por  TLC  usando  la  mezcla  de  solventes  cloroformo:  metanol:  ácido  acético  (130:50:20, v/v/v), y los lípidos se detectaron y cuantificaron con el phosphorimager.  7.9. Análisis in silico y antecedentes bibliográficos del gen smc02490 de S. meliloti 1021  Los genes olsA y olsB se  identificaron por primera vez en S. meliloti 1021  [Gao et  al.,  2004; Weissenmayer  et  al.,  2002].  Sin  embargo,  la  expresión  simultanea  de  ambos  genes en la cepa de expresión BL21 (DE3) x pLysS de E. coli, no produjo OLs (Jun‐Lian Gao,  datos no publicados). Una posible explicación, es que otro gen aún no  identificado esté  50 involucrado en  la biosíntesis de  los OLs. Geiger et al.  (2010) analizaron  las  interacciones  físicas  y  funcionales  del  gen  olsB  con  el  software  “string  database”  disponible  en  la  dirección  electrónica  del  genoma  de  S.  meliloti  1021  (http://iant.toulouse.inra.fr/bacteria/annotation/cgi/rhime.cgi?docid=SMc01127).  Geiger  et al. (2010) encontraron que genes homólogos a los genes olsB y smc02490 de S. meliloti  1021 coexiste en varias bacterias o se encuentran fusionados en Arcobacter butzleri y en  especies del orden Alteromonadales. Sin embargo, en S. meliloti 1021 esos genes no  se  encuentran fusionados, ni tampoco lo están en otros miembros de las Alfaproteobacterias  y  de  las  Betaproteobacterias.  Debido  a  lo  anterior  procedimos  a  repetir  el  análisis  realizado por Geiger et al. (2010) para actualizar los datos y también hicimos un blastp con  la  proteína  SMc02490  (307  aminoácidos)  de  S. meliloti  1021  con  el  número  de  acceso  NP_387151.1. También procedimos a buscar los antecedentes bibliográficos de smc02490  y a inactivarlo.  7.10. Inactivación del gen smc02490 de S. meliloti 1021  Los  oligonucleótidos  oLOP163  (ATGTTGATATCGATGCGTCGCGAAAGAGTGG)  y  oLOP164 (AAAGGATCCCAGCGTATCGACGATATGGC) que introducen los sitios de corte EcoRV  y BamHI (subrayados), respectivamente, se usaron en la PCR para amplificar un fragmento  de  1072  pb  localizado  río  arriba  del  gen  smc02490.  Similarmente,  los  oligonucleótidos  oLOP165  (AAAGGATCCCCGATGGAAGTGACACGTGC)  y  oLOP166  (ACTCTCTAGAGAAGACTGGTCTCCGGTTCG) que introducen los sitios de corte BamHI y XbaI  (subrayados), respectivamente, se usaron en la PCR para amplificar un fragmento de 1045  pb localizado río abajo del gen smc02490. En todas reacciones de PCR se usó como molde  DNA  genómico  de  S.  meliloti  1021.  El  producto  de  la  PCR  amplificado  con  los  oligonucleótidos oLOP163 y oLOP164 se clonó en el vector pCR®2.1‐TOPO®, como lo indicó  el  fabricante  de  éste  vector,  resultando  el  plásmido  pNG50.  El  producto  de  la  PCR  amplificado  con  oLOP165  y  oLOP166  se  digirió  con  BamHI/XbaI  y  se  clonó  en  pBluescriptSK+  previamente  digerido  con  las  mismas  enzimas,  resultando  el  plásmido  pNG45.  51 Por secuenciación de los insertos de los plásmidos se confirmó que pNG50 y pNG45  portan la región río arriba y río abajo del gen smc02490, respectivamente. Posteriormente  el plásmido pNG45 se digirió con EcoRV/BamHI y se ligó con el fragmento de DNA de 1072  pb obtenido después de la digestión con EcoRV/BamHI de pNG50, resultando el plásmido  pNG52. Después, el plásmido pNG52  se  linearizó  con BamHI y  se  ligó  con el  fragmento  BamHI  del  pCAT  que  contiene  el  gen  cat  que  confiere  resistencia  a  cloranfenicol.  El  plásmido  resultante contenía  las  regiones  flanqueantes del gen smc02490  interrumpidas  por  el  gen  cat  y  se  llamó  pNG54.  El  plásmido  pNG54  se  digirió  con  EcoRV/XbaI  para  reclonar las regiones flanqueantes del gen smc02490 y el gen cat localizado entre las dos  regiones. El fragmento EcoRV/XbaI se clonó en el vector suicida pK18mobsacB [Schäfer et.  al., 1994] previamente digerido con SmaI/XbaI, produciendo el plásmido pNG56.   El plásmido pNG56 se  introdujo en S. meliloti 1021  (cepa silvestre) mediante una  cruza diparental, usando como donadora a la cepa de E. coli S17‐1. Las transconjugantes se  seleccionaron  en  el  medio  de  cultivo  PY  CaCl2  con  cloranfenicol,  neomicina  y  ácido  nalidíxico,  lo que también permitió contraseleccionar a E. coli S17‐1. Posteriormente una  recombinante  sencilla  se  creció en  condiciones no  selectivas en PY CaCl2 hasta alcanzar  una absorbancia de 1.0 a 620 nm, se hicieron diluciones y se sembraron en PY CaCl2 con  sacarosa al 10 %  (peso/volumen) y cloranfenicol para seleccionar  recombinantes dobles.  Después de 3 días de crecimiento en el medio selectivo, se seleccionaron 100 colonias que  se resembraron simultáneamente en cajas de PY CaCl2 con cloranfenicol o con neomicina.  Después se seleccionaron 10 colonias que crecieron solamente en el medio de cultivo que  contenía cloranfenicol y que no crecieron en el medio de cultivo que contenía neomicina.  Las  10  colonias  se  analizaron  genotípicamente  por  hibridación  tipo  Southern.  En  la  hibridación tipo Southern se usó como sonda marcada con digoxigenina el fragmento de  1072  pb,  localizado  río  arriba  del  gen  smc02490.  El  DNA  de  la  cepa  silvestre  y  de  las  candidatas a mutantes en el gen smc02490 se digirió con EcoRI. Se esperaba que la sonda  en  la  cepa  silvestre  hibridara  con  un  fragmento  de  12477  pb  y  en  las  candidatas  a  mutantes con uno de 2320 pb.  52 Tambien  hemos  diseñado  los  oligonucleótidos  oLOP161  (ACCTTATCCATGGCGAGACGTGATTCGGC)  y  oLOP162  (AAAGGATCCTTAGCCCGCTACAACCGCCC)  para  amplificar  al  gen  smc02490,  los  cuales  introducen los sitios de corte NcoI y BamHI (subrayados), el gen se clonará en el vector de  expresión pET3d. Además se construyó el plásmido que se usara como control el pNG39  que  resultó  de  la  ligación  del  pET3d  con  el  pRK404,  ambos  digeridos  previamente  con  BamHI.      53 8. Resultados  Los resultados obtenidos en esta tesis doctoral se detallan en las subsecciones 8.1  en la que se encuentran los artículos publicados y en la 8.2 en la que están los resultados  adicionales. El  III artículo contiene una discusión de  los resultados principales,  la cual es  ampliada en  la discusión de esta  tesis. En  la  subsección 8.2  se muestran  los  resultados  adicionales  relacionados a  la  temática de  los artículos  II y  III, pero que aún no han  sido  publicados.    8.1. Artículos publicados  8.1.1. I Artículo  Geiger, O., González‐Silva N., López‐Lara, I. M., and Sohlenkamp, C. (2010) Amino acid‐ containing membrane lipids in bacteria. Prog. Lipid Res. 49, 46‐60.  La  bacteria  modelo  Escherichia  coli  solamente  forma  tres  lípidos  principales  de  membrana  fosfatidiletanolamina,  fosfatidilglicerol  y  cardiolipina,  y  todos  son  glicerofosfolípidos.  Fosfatidilserina  contiene  un  residuo  de  aminoácido  y  es  uno  de  los  lípidos  principales  en  las  membranas  eucariontes  pero  en  muchas  bacterias  éste  solamente  es  un  intermediario  biosintético  minoritario.  En  algunas  bacterias,  los  glicerofosfolípidos  aniónicos  fosfatidilglicerol  y  cardiolipina  pueden  ser  decorados  con  residuos aminoacil. Por ejemplo, fosfatidilglicerol puede ser decorado con lisina, alanina o  arginina mientras que en caso de cardiolipina se conocen modificaciones con  lisina o D‐ alanina. En pocas bacterias,  lípidos derivados de diacilglicerol pueden ser sustituidos con  lisina u homoserina. Lípidos aciloxi‐acil en los cuales la parte lipídica está unida por enlace  amida  al  grupo  α‐amino  de  un  aminoácido  están  ampliamente  distribuidos  entre  las  bacterias  y  los  lípidos  de  ornitina  son  la  versión  más  común  de  este  tipo  de  lípidos.  Solamente  pocos  grupos  bacteriales  forman  lípidos  de  glicina,  lípidos  de  glicinaserina,  esfingolípidos, o sulfonolípidos. Aunque muchos de estos  lípidos de membrana bacterial  que  contienen aminoácidos  son producidos en  ciertas  condiciones de estrés,  se  conoce  poco sobre las funciones moleculares específicas de estos lípidos.  54 Review Amino acid-containing membrane lipids in bacteria Otto Geiger *, Napoleón González-Silva, Isabel M. López-Lara, Christian Sohlenkamp Centro de Ciencias Genómicas, Universidad Nacional Autónoma de México, Apdo. Postal 565-A, Cuernavaca, Morelos CP62210, Mexico a r t i c l e i n f o Keywords: Bacterial membrane Amino acid-containing lipid Phosphatidylserine Lysyl-phosphatidylglycerol Betaine lipid Ornithine-containing lipid Acyl-oxyacyl lipid Sulfonolipid a b s t r a c t In the bacterial model organism Escherichia coli only the three major membrane lipids phosphatidyleth- anolamine, phosphatidylglycerol, and cardiolipin occur, all of which belong to the glycerophospholipids. The amino acid-containing phosphatidylserine is a major lipid in eukaryotic membranes but in most bac- teria it occurs only as a minor biosynthetic intermediate. In some bacteria, the anionic glycerophospho- lipids phosphatidylglycerol and cardiolipin can be decorated with aminoacyl residues. For example, phosphatidylglycerol can be decorated with lysine, alanine, or arginine whereas in the case of cardiolipin, lysine or D-alanine modifications are known. In few bacteria, diacylglycerol-derived lipids can be substi- tuted with lysine or homoserine. Acyl-oxyacyl lipids in which the lipidic part is amide-linked to the a- amino group of an amino acid are widely distributed among bacteria and ornithine-containing lipids are the most common version of this lipid type. Only few bacterial groups form glycine-containing lipids, serineglycine-containing lipids, sphingolipids, or sulfonolipids. Although many of these amino acid-con- taining bacterial membrane lipids are produced in response to certain stress conditions, little is known about the specific molecular functions of these lipids.  2009 Elsevier Ltd. All rights reserved. Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 2. Amino acid-containing glycerophospholipids. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 2.1. Phosphatidylserine and its derivates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 2.2. Aminoacyl modifications of phosphatidylglycerol. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 2.3. Aminoacyl modifications of cardiolipin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3. Amino acid-containing diacylglycerols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 3.1. Homoserine-containing betaine lipids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 3.2. Mycobacterial lysine-containing lipid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 4. Amino acid-containing acyl-oxyacyl lipids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 4.1. Ornithine-containing lipids. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 4.1.1. Distribution and structure of ornithine-containing lipids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 4.1.2. Biosynthesis of ornithine-containing lipids. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 4.1.3. Hydroxylated ornithine-containing lipids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 4.1.4. Tauro-ornithine- and lysine-containing lipids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 4.2. Glycine-containing lipids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 4.3. Serineglycine-containing lipids. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 4.4. The innate immune response to amino acid-containing acyl-oxyacyl lipids. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 5. Bacterial sphingolipids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 6. Sulfonolipids in the Cytophaga group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 7. Stress causes changes in bacterial membranes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 8. Conclusions and perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 0163-7827/$ - see front matter  2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.plipres.2009.08.002 * Corresponding author. Tel.: +52 7773 290815; fax: +52 7773 175581. E-mail address: otto@ccg.unam.mx (O. Geiger). Progress in Lipid Research 49 (2010) 46–60 Contents lists available at ScienceDirect Progress in Lipid Research journal homepage: www.elsevier .com/locate /pl ipres 55 1. Introduction A primary role of lipids in cellular function is to form the lipid bilayer permeability barrier of cells. Glycerophospholipids are the primary building blocks of membranes but other lipids are impor- tant components. In the bacterial model organism Escherichia coli only the three major membrane lipids phosphatidylethanolamine, phosphatidylglycerol, and cardiolipin occur [1]. In addition, E. coli and almost all other Gram-negative bacteria usually have the lipid A-containing lipopolysaccharide in the outer monolayer of their outer membrane and lipid A modification systems have been re- viewed recently [2]. However, in other bacteria, additional and alternative membrane lipids are found and in many cases neither their biosyntheses nor their functionalities are understood. Some Gram-negative bacteria have phosphatidylcholine [3] or sphingoli- pids [4] in their standard repertoire, whereas many Gram-positives have glycosylated diacylglycerols [5] and lysyl-phosphatidylglyc- erol [6] in their membranes. Notably, phosphatidylinositol is an essential lipid forMycobacterium tuberculosis [7]. Steroid and hopa- noid lipids only occur in some bacteria [8]. Under certain stress conditions specific membrane lipids can be formed and in some cases existing membrane lipids can suffer modifications in order to minimize the stress exerted. For example, under phosphorus- limiting conditions of growth, some bacteria formmembrane lipids lacking phosphorus such as glycolipids, sulfolipids, betaine lipids, or ornithine-containing lipids [9]. Challenge of proteobacteria with acid causes modifications of membrane lipids, such as formation of lysyl-phosphatidylglycerol [10] or hydroxylation of ornithine-con- taining lipids [11]. Numerous examples of bacterial lipids containing amino acids or peptides are known [12] and many of them display interesting properties as antibiotics [13] and biosurfactants [14]. However, in order to form lipidic bilayer membranes, amphiphilic lipids usu- ally need to have at least two long-chain acyl or alkyl residues and the molecules should be roughly of cylindric shape, i.e. in cross- section, the area covered by their hydrophilic head group should be similar to the area covered by the hydrophobic acyl or alkyl chains and therefore upon assembly of monomers, bilayer instead of micelle formation is favoured [15]. In this review we will focus on membrane-forming lipids containing aminoacyl residues. 2. Amino acid-containing glycerophospholipids 2.1. Phosphatidylserine and its derivates In bacteria, CDP-diacylglycerol is the central activated interme- diate for the biosynthesis of glycerophospholipids [3] (Fig. 1). Dis- tinct and specific enzymes belonging to the CDP-alcohol phosphotransferase family condense alcohols such as glycerol-3- phosphate, inositol, choline, or serine to CDP-diacylglycerol form- ing phosphatidylglycerol phosphate, phosphatidylinositol, phos- phatidylcholine, or phosphatidylserine (PS), respectively, (Fig. 1). The only amino acid converted by a CDP-alcohol phosphotransfer- ase into a phospholipid head group is serine, and the condensation of CDP-diacylglycerol with serine to form phosphatidylserine (PS) constitutes the first step for the synthesis for phosphatidylethanol- amine (PE) [1,16]. In most bacteria, this condensation is catalyzed by the membrane-bound type II PS synthase (Pss) but in some Gram-negative bacteria, like the Enterobacteriaceae, PS is synthe- sized by a type I Pss, which are soluble enzymes that form part of a distinct superfamily that furthermore includes cardiolipin (CL) synthases, poxvirus envelope proteins, phospholipases D, and nucleases [17]. In a second step, the decarboxylation of PS is catalyzed by PS decarboxylase (Psd) to yield PE. Although PS ac- counts for 5–15% of the phospholipids in eukaryotic cells [18], in most bacteria, PS is a biosynthetic intermediate and is a very minor membrane lipid. In some bacteria, however, the pool of PS seems to be larger and in Bacillus megaterium PS comprises some 5–10% [19] of the total membrane lipids. Also, in Bdellovibrio bacteriovorus, which parasitizes larger Gram-negative bacteria, PS is a major membrane phospholipid [20]. Psd-deficient bacterial mutants are unable to form PE, and as PS is not consumed they accumulate sig- nificant amounts of PS. In such mutants, PS can comprise up to 34% in E. coli [21], or up to 29% in Bacillus subtilis [22] of the total mem- brane lipids. A Psd-deficient mutant of Sinorhizobium meliloti lacks PE but forms up to 18% PS [23]. This sinorhizobial Psd-deficient mutant resembles in many vegetative aspects a Pss-deficient mu- tant of S. meliloti [16] which also lacks PE but does not form PS either. Surprisingly, Pss-deficient mutants lacking PE form nitro- gen-fixing root nodules on alfalfa host plants nearly as efficiently as the wild type. In contrast, the Psd-deficient sinorhizobial mutant accumulates significant amounts of PS and only few empty nodules that are unable to fix nitrogen are formed by this mutant with much delay on the host plant [23]. In animal systems, PS plays a key role in physiological and pathological events. For example, PS exposed on activated platelets promotes the blood coagulation cas- cade and the aggregation of platelets, and the externalization of PS to the cell surface is a hallmark of apoptotic cells [24]. Although PS is a major membrane lipid in plants [25] its specific roles or func- tions are unknown. However, the presence of PS in the Psd-defi- cient sinorhizobial mutant interferes with the accommodation of this mutant bacterium inside the nodule, possibly due to a plant- mediated mechanism. In E. coli, the PS formed is distributed equally between the inner and outer membrane [26], but it is not clear whether bacteria expose their PS on the outer surface of the outermost membrane. 2.2. Aminoacyl modifications of phosphatidylglycerol Phosphatidylglycerol (PG) and cardiolipin (CL) are the major an- ionic membrane lipids in most bacteria and their synthesis is well understood [1]. Modified forms of PG and CL have been described in different bacteria. Lysyl-phosphatidylglycerol (lysyl-PG) and other aminoacyl esters of PG, such as alanyl-PG, or ornithyl-PG, are major membrane lipids in several Gram-positive bacteria (fir- micutes) [28]. Lysyl-PG constitutes a major membrane lipid in Staphylococcus aureus [29], B. subtilis [30], Bacillus anthracis [31], Listeria monocytogenes [32], and Lactococcus plantarum. In addition to lysyl-PG, some bacteria form ornithyl-PG (M. tuberculosis) or ala- nyl-PG (Clostridium perfringens) [6]. Moreover, aminoacylation of PG with arginine [33] or glycine [34] has been described. A large variety of PG-derived lipids are present in Enterococcus faecalis (formerly known as Streptococcus faecalis) which probably has ala- nyl-PG, 20-lysyl-PG, 30-lysyl-PG, 20,30-dilysyl-PG, arginyl-PG, and a diglucosyl derivative of PG [35]. As with other acylated glycerol derivatives, a 20-lysyl-PG can undergo acyl migration to yield 30-ly- syl-PG [36]. The protein MprF (‘‘multiple peptide resistance fac- tor”), responsible for lysyl-PG formation, was first described in S. aureus during a screen for transposon mutants more susceptible to cationic peptides of the innate immune response than the wild type [37]. MprF from S. aureus is able to transfer lysine from charged lysyl-tRNA to PG forming lysyl-PG [38,39] (Fig. 1). Also in B. subtilis [40] or in B. anthracis [31], MprF is required for the syn- thesis of lysyl-PG and for resistance to cationic antimicrobial pep- tides. It has been thought that the presence of lysyl-PG is restricted mainly to Gram-positive bacteria although lysyl-PG is present in a strain of Pseudomonas aeruginosa [41] and in Caulobacter crescentus [42]. In screens for mutants more susceptible to acidic growth con- ditions, Reeve et al. [43] and Vinuesa et al. [44] identified genes coding for MprF homologues in the a-proteobacteria Sinorhizobium medicae and Rhizobium tropici that are called lpiA (‘‘low pH-induc- O. Geiger et al. / Progress in Lipid Research 49 (2010) 46–60 47 56 ible A”). In both organisms, lpiA is transcriptionally induced under acidic growth conditions [43–45]. Later, genes encoding for homo- logues of LpiA were identified in several other Gram-negative bac- teria, like Agrobacterium tumefaciens, Aeromonas hydrophila, Xanthomonas campestris, Xylella fastidiosa and several species of the genera Brucella, Burkholderia, Pseudomonas, among others. Interestingly, most of these species interact with eukaryotic hosts as symbionts, pathogens, or commensals. Lysyl-PG formation in R. tropici increases the resistance to cationic peptides [10]. Notably, in most Gram-negative bacteria the lpiA gene probably forms an operon with the gene atvA (‘‘acid tolerance and virulence A”). The biochemical function of AtvA is not known. More recently, genes encoding proteins responsible for the for- mation of alanyl-PG have been identified. The pathogen C. perfrin- gens has two phylogenetically distinct MprF paralogues, one responsible for the formation of lysyl-PG and the other for the syn- thesis of alanyl-PG [6] (Fig. 1). Alanyl-PG also occurs in P. aerugin- osa and the responsible gene is identified [46]. Formation of Fig. 1. Biosynthesis of glycerophospholipids in bacteria. (A) Formation of the activated intermediate CDP-diacylglycerol. Glycerol-3-phosphate forms the backbone of all glycerophospholipid molecules and it can be synthesized by two different pathways, either from glycerol directly by glycerol kinase (GlpK) or by reduction of the glycolytic intermediate dihydroxyacetone phosphate with NADH catalyzed by biosynthetic glycerol-3-phosphate dehydrogenase (GpsA) [1]. In E. coli, the glycerol-3-phosphate acyltransferase PlsB can use acyl-CoA or acyl-ACP as acyl donors and is the major activity for catalyzing the first acylation at position 1 of glycerol-3-phosphate thereby forming 1-acyl-glycerol-3-phosphate. However, the more widespread pathway to achieve the initial acylation of glycerol-3-phosphate among bacteria seems to involve PlsX and PlsY [27]. In this pathway, PlsX catalyzes the conversion of acyl-ACP and inorganic phosphate to acyl-phosphate and ACP. In a second step, PlsY transfers the acyl group from acyl-phosphate to glycerol-3-phosphate forming inorganic phosphate (Pi) and 1-acyl-glycerol-3-phosphate [27]. The second fatty acyl residue is added by another enzyme, the 1-acyl-glycerol-3-phosphate acyltransferase PlsC, to form phosphatidic acid. The conversion of phosphatidic acid to CDP-diacylglycerol (CDP-diglyceride) is catalyzed by CDP-diglyceride synthase CdsA [1]. (B) Diversification of phospholipid head groups. The first step in the synthesis of phosphatidylethanolamine (PE) is the condensation of CDP-diacylglycerol with serine to form phosphatidylserine (PS) catalyzed by PS synthase (Pss). In a second step, the decarboxylation of PS is catalyzed by PS decarboxylase (Psd) to yield PE. A well-known pathway for PC formation occurs by threefold methylation of PE using S-adenosylmethionine (SAM) as methyl donor and catalyzed by phospholipid N-methyltransferase (PmtA). Many PC-containing bacteria have a second pathway for PC formation, catalyzed by PC synthase (Pcs), in which choline is condensed directly to CDP-diacylglycerol forming PC and CMP [3]. In the initial step of phosphatidylglycerol (PG) and cardiolipin (CL) biosynthesis, phosphatidylglycerol phosphate synthase (PgsA) transfers glycerol-3-phosphate to CDP-diacylglycerol under the release of CMP thereby producing phosphatidylglycerol phosphate (PGP). There are at least two enzymes with PGP phosphatase activity (PgpA and PgpB) in E. coli, releasing inorganic phosphate from PGP to form phosphatidylglycerol (PG) [1]. Lysyl-phosphatidylglycerol (lysyl-PG) is a well-known membrane lipid in many Gram-positive bacteria and MprF can transfer lysine from charged lysyl-tRNA to PG forming lysyl-PG. The pathogen Clostridium perfringens has two phylogenetically distinct MprF paralogues, one responsible for the formation of lysyl-PG (MprF1) and the other causing the synthesis of alanyl-phosphatidylglycerol (MprF2) [6]. In E. coli and probably most other bacteria, a cardiolipin synthase (Cls) condenses two PG molecules to yield cardiolipin (CL) and free glycerol in a transesterification reaction. Although a MprF homologue is required for the lysinylation of CL, it is not known whether lysyl-CL is formed by lysinylation of CL or by Cls-catalyzed condensation of lysyl-PG with PG. The causative agent of tuberculosis Mycobacterium tuberculosis has phosphatidylinositol (PI) and derivatives thereof as major components in its membrane. In M. tuberculosis, PI is formed by condensing myo-inositol to CDP- diacylglyceride in a reaction catalyzed by PI synthase [7]. 48 O. Geiger et al. / Progress in Lipid Research 49 (2010) 46–60 57 20-alanyl-PG in P. aeruginosa is increased under acidic growth con- ditions. Alanyl-PG is found in the inner and the outer membrane of P. aeruginosa and it increases the resistance of the bacterium to- wards the b-lactam antibiotic cefsulodin, the heavy metal ion Cr3+, the osmolyte sodium lactate, and the cationic antimicrobial peptide protamine sulphate [46]. Two genes coding for LpiA/MprF homologues are also present in Sphingomonas wittichii and Enterococcus faecium and even three LpiA/MprF homologues are encoded by the Kineococcus radiodurans genome, but nothing is known about their functions. Although the presence of aminoacylated PG has been described only in bacteria, a gene coding for an LpiA/MprF homologue is present in the gen- ome of the moss Physcomitrella patens. 2.3. Aminoacyl modifications of cardiolipin Cardiolipin (CL) can also be substituted on the sn-2 hydroxyl of the middle glycerol moiety with a-D-glucopyranosyl, D-alanyl, or L- lysyl residues [47,48] in the group N streptococcus Vagococcus flu- vialis or with L-lysyl in Listeria species [32] (Fig. 1). The MprF homo- logue Lmo1695 of L. monocytogenes is required for the lysinylation of both, PG and CL [49]. However, it is not known whether lysyl-CL is formed by lysinylation of CL or by CL synthase-catalyzed con- densation of lysyl-PG with PG. Therefore, the precise biosynthetic steps that form these modified versions of CL are not clear. Usually, CL is only partially ionized at physiological conditions due to the hydrogen bonding of the sn-2 hydroxyl group of the middle glyc- Fig. 1 (continued) O. Geiger et al. / Progress in Lipid Research 49 (2010) 46–60 49 58 erol moiety with the neighboring phosphate groups. Therefore, CL can function as a proton sink or a conduit for protons in transfer processes [15]. Upon derivatization of the sn-2 hydroxyl of the middle glycerol with the above-mentioned residues the hydrogen bonding should be impeded and the special property of CL as a pro- ton sink will be lost. 3. Amino acid-containing diacylglycerols 3.1. Homoserine-containing betaine lipids Although PC is known to be the major membrane lipid in eukaryotes, some lower eukaryotic organisms possess the betaine lipid diacylglyceryl-N,N,N-trimethylhomoserine (DGTS) instead. DGTS occurs in a wide variety of lower green plants (green algae, bryophytes and pteridophytes), chromophytes, fungi, and amoebae (reviewed in [3]). In some a-proteobacteria, DGTS acts as a phos- phorus-free membrane lipid [50] that substitutes for PC under con- ditions of phosphate limitation [51]. There is an apparent reciprocity between the content of PC and the content of DGTS, i.e. when PC is a major membrane lipid often no DGTS is detected in the same organism, whereas in organisms where DGTS is a ma- jor lipid, PC is only found in trace levels. This correlation suggests, that DGTS and PC, both zwitterionic at physiological pH, are inter- changeable at least with regard to essential functions for the respective organism. The mutual replacement of PC or DGTS occurs in culture conditions but also in natural environments [51a]. Phy- toplankton communities from the phosphorus-poor Sargasso Sea have much higher betaine lipid/PC ratios than communities from the phosphorus-replete South Pacific due to the respective adjust- ments in eukaryotic phytoplankton [51a]. Two structural genes from Rhodobacter sphaeroides, btaAB, cod- ing for two enzymes BtaA and BtaB involved in DGTS biosynthesis have been characterized [52,53]. The BtaA S-adenosylmethionine/ diacylglycerol 3-amino-3-carboxypropyl transferase converts diac- ylglycerol (DAG) into diacylglyceryl-homoserine (DGHS) and dur- ing the formation of the ether bond, S-adenosylmethionine functions as donor of the homoseryl group. Subsequently, the S- adenosylmethionine: diacylglyceryl-homoserine-N-methyltrans- ferase BtaB catalyzes threefold methylation of DGHS in order to yield DGTS (Fig. 2). Orthologues of BtaA and BtaB exist in S. meliloti, and a BtaA-deficient mutant of S. meliloti is unable to produce DGTS [54]. In the eukaryotic green alga Chlamydomonas reinhardtii, the betaine lipid synthase BTA1Cr is a bifunctional protein that can perform the homoseryl modification of diacylglycerol as well as the subsequent methylations of the homoseryl amino group [55]. In other lower eukaryotes where DGTS is present, genes coding for homologues of the bifunctional C. reinhardtii enzyme are pres- ent. Heterologous expression of BTA1Cr in E. coli leads to DGTS accumulation and the two domains of BTA1Cr are functionally equivalent to BtaA and BtaB [55]. The wealth of genome sequencing data indicates that the occur- rence of DGTS-like betaine lipids is limited in bacteria. Homo- logues of rhodobacterial BtaA (above 42% identity and 55% similarity) and BtaB are mainly found in some orders of the a-pro- teobacteria, such as the Rhodobacterales (Rhodobacter, Roseobacter, Sagittula, Stappia), the Sphingomonadales (Sphingomonas, Erythrob- acter), the Rhizobiales (Rhizobium, Agrobacterium, Sinorhizobium, Ochrobactrum, Mesorhizobium, Beijerinckia, Rhodopseudomonas), and in members of the Planctomycetes (Planctomyces, Blastopirell- ula, Rhodopirellula) (Fig. 3). More distantly related genes are pres- ent in the d-proteobacterium Plesiocystis pacifica SIR-I and in Chthoniobacter flavus (Chlamydia/Verrucomicrobia). Genes coding for homologues of the bifunctional enzymes from Chlamydomonas reinhardtii are present in several lower eukaryotes such as Candida albicans, Cryptococcus neoformans, Neurospora crassa and Physc- omitrella patens (Fig. 3). Planctomycetes represent a distinct bacte- rial phylum as they show absence of a peptidoglycan cell wall and extensive cell compartmentalization, in some cases even a Fig. 2. Biosynthesis of diacylglyceryl-N,N,N-trimethylhomoserine (DGTS) [52,53]. DAG: diacylglycerol; SAM: S-adenosylmethionine; SAH: S-adenosylhomocysteine; DGHS: diacylglyceryl-homoserine. 50 O. Geiger et al. / Progress in Lipid Research 49 (2010) 46–60 59 membrane-enclosed nuclear structure [56]. Interestingly, the BtaAB homologues from the order Planctomycetales seem to be closer re- lated to the sequences from eukaryotic origin than to the other bacterial sequences (Fig. 3). It has been speculated that the biosyn- thesis of certain phosphorus-free membrane lipids such as DGTS might improve the survival of bacteria under phosphorus-depleted conditions [54]. The fact that intracellular pathogens of the genus Brucella apparently lack the genes needed for DGTS biosynthesis but that the closely related, and normally free-living (opportunis- tic) pathogen Ochrobacter anthropi has the respective genes sup- ports this idea. 3.2. Mycobacterial lysine-containing lipid A diacylglycerol-based lysine-containing lipid was isolated from Mycobacterium phlei strain IST [57]. Lysine is esterified to 1,2- diglyceride via an ester linkage (Fig. 4) and the major fatty acyl substitutions are palmitic and tuberculostearic acid [57]. The mycobacterial lysine-containing lipid in strain IST is not detected it in a reference strain (ATCC 19249), and it has been suggested that this lipid is involved in lysine uptake to the cell [57]. 4. Amino acid-containing acyl-oxyacyl lipids 4.1. Ornithine-containing lipids 4.1.1. Distribution and structure of ornithine-containing lipids Ornithine-containing lipids (OL) are widespread among Gram- negative bacteria and have been reported in some Gram-positives, like Mycobacterium and Streptomyces species (reviewed in [9]) but are absent in Archaea and Eukarya. The a-N-(acyloxyacyl)-ornith- ines have a 3-hydroxyfatty acyl group that is attached in amide linkage to the a-amino group of ornithine [51,58]. A second fatty acyl group is ester-linked to the 3-hydroxy position of the first fatty acid (Fig. 5A). In some bacteria the fatty acyls joined by ester linkage are hydroxylated at the 2 or 3 positions [12]. The configu- ration of the asymmetric carbon of 3-hydoxyfatty acyls of the OLs is D or (R) [59]. Although OLs are found in both membranes of Gram-negative bacteria, they seem to be enriched in the outer membrane [60]. 4.1.2. Biosynthesis of ornithine-containing lipids The biosynthesis of OLs occurs in two-steps. The N-acyltransfer- ase OlsB catalyzes the transfer of a 3-hydroxy fatty acyl group from 3-hydroxy fatty acyl–acyl carrier protein to the a-amino group of ornithine forming lyso-ornithine lipid [61]. Next, the O-acyltrans- ferase OlsA catalyzes the transfer of an acyl group from fatty acyl–acyl carrier protein to the hydroxy group of lyso-ornithine li- pid forming OL [62] (Fig. 5A). OlsB-deficient mutants of S. meliloti [61] or Rhodobacter capsula- tus [63] are unable to form OL. Expression of olsB from S. meliloti in Escherichia coli causes the formation of lyso-ornithine lipid [61]. The OlsB of S. meliloti is predicted to be a water-soluble protein of 296 amino acids that encodes anN-acyltransferase converting ornithine Fig. 4. Structure of mycobacterial lysyl-diacylglycerol. Fig. 3. Unrooted phylogenetic tree of Rhodobacter sphaeroides BtaA, Sinorhizobium meliloti BtaA, the N-terminal domain of Chlamydomonas reinhardtii Bta1CR and BtaA- like ORFs from other genomes. The tree was constructed using the program CLUSTALW (http://www.expasy.ch/). Distances between sequences are expressed as 0.06 changes per amino acid residue. The asterisks indicate that only the N- terminal domains of the respective sequences corresponding to BtaA were used for the construction of the tree. Accession numbers are as follows: Rhodobacter sphaeroides BtaA (ABA80038), Sinorhizobium meliloti 1021 BtaA (NP_386300), Chlamydomonas reinhardtii Bta1CR (XP_001700879), Agrobacterium tumefaciens str. C58 (NP_355081), Beijerinckia indica subsp. indica ATCC 9039 (YP_001834100), Blastopirellula marina DSM 3645 (ZP_01088661), Candida albicans SC5314 (XP_713069), Chthoniobacter flavus Ellin428 (ZP_03133405), Erythrobacter sp. NAP1 (ZP_01039410), Hoeflea phototrophica DFL-43 (ZP_02168757), Laccaria bicolor S238N-H82 (XP_001877704), Loktanella vestfoldensis SKA53 (ZP_01002837), Meso- rhizobium loti MAFF303099 (NP_103130), Mesorhizobium sp. BNC1 (YP_674579), Nocardia farcinica IFM 10152 (YP_117653), Ochrobactrum anthropi ATCC 49188 (YP_001370273), Parvibaculum lavamentivorans DS-1 (ABS62442), Physcomitrella patens subsp. patens (XP_001757434), Planctomyces maris DSM 8797 (ZP_01856778), Plesiocystis pacifica SIR-1 (ZP_01909206), Pseudovibrio sp. JE062 (EEA94326), Rhizobium etli CFN 42 (YP_470361), Rhodopirellula baltica SH 1 (NP_863860), Rhodopseudomonas palustris BisB18 (YP_531155), Roseobacter sp. MED193 (EAQ46984), Sagittula stellata E-37 (EBA07014), Sphingomonas sp. SKA58 (ZP_01304163), Stappia aggregata IAM 12614 (ZP_01546622), and Thiomicrospira crunogena XCL-2 (YP_390494). A – Actinomycetales, Euk – Eukaryotes, V – Verrucomicrobia, Pla – Planctomycetales, Rhi – Rhizobiales, Rho – Rhodobacterales, Sph – Sphingomonadales, c – gamma proteobacteria, d – delta proteobacteria. O. Geiger et al. / Progress in Lipid Research 49 (2010) 46–60 51 60 to lyso-ornithine lipid thereby catalyzing first step of OL biosynthe- sis. OlsB defines a concrete function for a whole cluster of ortholo- gous group of proteins (COG3176) previously assigned as hypothetical or as putative hemolysins [61]. OlsB belongs to the acyl-CoA N-acyltransferase superfamily [64]. Heath and Rock [65] described a consensus peptide motif (H(X)4D) common to glycer- olipid acyltransferases and demonstrated that an exchange of this conserved H for A eliminated the activity of E. coli PlsB. In OlsB, H87 and D92might form such a motif that is conserved within OlsB homologues [61]. A search of PhyloFacts with OlsB identifies acyl- homoserine lactone synthases EsaI [66] and LasI [67] and the ter- tiary structure of OlsB is expected to be quite similar to the autoin- ducer synthase. Like OlsB, acyl-homoserine lactone synthases are N-acyl transferases that use acyl-ACPs and an amino acid derivative (SAM) as substrates (Fig. 5B). The N-acyl amino acid synthase FeeM from an uncultured soil microbe binds the acyl carrier protein FeeL, catalyzes the formation of N-acyl tyrosine, and its structure resem- bles that of acyl-homoserine lactone synthases EsaI and LasI [68]. Genome analyses indicate that like many bacterial groups, the order Rhodobacterales have an olsBA operon (containing olsB1) plus a gene coding for a second homologue of OlsB (olsB2) (Fig. 6). The olsB2 gene is not located physically close to olsBA, but at another site in the genome. The OlsB2 homologues of differ- ent Rhodobacterales are more closely related to each other than to the OlsB1 homologues (Fig. 6). Interestingly, R. sphaeroides forms lipids that contain glutamine in addition to the well-known OL [69]. The initial step in the biosynthesis of glutamine-containing lipids might be catalyzed by the OlsB2 homologue found exclu- sively in the Rhodobacterales. The only other examples where multiple olsB homologues are present within one genome are the a-proteobacteria Magnetospir- illum magneticum and M. magnetotacticum (Fig. 6). In addition to OL, another unknown amino lipid has been described in the mag- netosome membrane of Magnetospirillum which might be synthe- sized by the second homologue [70]. Remarkably, heterologous expression of microbial DNA extracted from environmental sam- ples led to the identification of long-chain N-acyl derivatives of tyrosine, phenylalanine, tryptophan, and arginine, all of which have antibiotic activity [13,71,72]. It is possible that some of the long-chain N-acyl transferases involved in the formation of these compounds catalyze initial steps in the biosyntheses of other still unknown acyl–oxyacyl membrane lipids. Genome sequencing data indicate, that in several organisms from the order Alteromonadales and in the e-proteobacterium Fig. 5. Biosyntheses of ornithine-containing lipids (A) and of the Pseudomonas aeruginosa quorum sensing signal 3-oxo-dodecanoyl-homoserine lactone (B). 52 O. Geiger et al. / Progress in Lipid Research 49 (2010) 46–60 61 Arcobacter butzleri, the OlsB protein is N-terminally fused to a hypothetical protein domain of unknown function, suggesting that OL biosynthesis might turn out to be more complicated. The pres- ence of OL has not been described in any of these species. Also mutants of S. meliloti [62] or R. capsulatus [63] deficient in OlsA are unable to form OL. Overexpression of olsB in an olsA-defi- cient mutant of S. meliloti leads to the accumulation of lyso-orni- thine lipid [61]. The OlsA of S. meliloti is a protein of 292 amino acids with a probable transmembrane helix close to the N-termi- nus. An alignment of OlsA with some prokaryotic enzymes display- ing lysophosphatidic acid acyltransferase activities [62] demonstrates that there are two conserved regions (amino acids 67–83 and 139–154) where OlsA has the highest similarity to other members of this group. In lysophosphatidic acid acyltransferases, two motifs, NHQS and PEGTR, are conserved [73], and they are found in modified forms (NHVS, amino acids 72–75; PEGTT, amino acids 143–147) in the OlsA sequence [62]. Based on the consensus peptide motif (H(X)4D) [65] common to glycerolipid acyltransfer- ases, in OlsA, H73 and D78 might form such a motif. It is striking, however, that sequences coding for enzymes with lysophospha- tidic acid acyltransferase activities are overall quite dissimilar. For example, in the bacterium Neisseria meningitidis there are three enzymes (NlaA, NlaB, and a third activity detectable in nlaA-, nlaB- deficient double mutants) with lysophosphatidic acid acyltransfer- ase activity in vitro [74]. Though NlaA and NlaB are from the same organism they are quite dissimilar and it is interesting to note that nlaA-deficient mutants and nlaB-deficient mutants show different phenotypes suggesting that NlaA and NlaB perform different bio- chemical functions in N. meningitidis in vivo. From its sequence, OlsA clearly groups within the present lysophosphatidic acid acyl- transferases. Because olsA-deficient mutants are unable to form OL, but show no accumulation of lysophosphatidic acid and no impair- ment of glycerophospholipid biosynthesis, there must be a PlsC activity in S. meliloti responsible for these latter functions, presum- ably SMc00714 (http://sequence.toulouse.inra.fr/meliloti.html). OlsA is required for the enzymatic activity of a lyso-ornithine lipid- and acyl-AcpP-dependent O-acyltransferase that converts lyso- ornithine lipid into OL [62]. In addition to OlsA, Pseudomonas fluo- rescens possesses two more lysophosphatidic acid acyltransferase homologues, HdtS and PatB [75]. Either HdtS or PatB complement an E. coli PlsC-deficient mutant for growth, while the OlsA from P. fluorescens does not. Although HdtS or PatB can provide the PlsC function in vivo, they are not functionally identical. Mutants lack- ing PatB show reduced growth at elevated temperatures while HdtS-deficient mutants are affected in growth, motility and have reduced amounts of cis-vaccenic acid [75]. Also, Rhodobacter cap- sulatus possesses three lysophosphatidic acid acyltransferase homologues, OlsA, PlsC316, and PlsC3498 [76]. Either OlsA or PlsC316 fromR. capsulatus complement anE. coli PlsC-deficient mu- tant for growth, while PlsC3498 does not. A PlsC316-deficient mu- tant has reduced amounts of C16 fatty acids [76]. Therefore, OlsA from R. capsulatus is able to acylate 1-acyl-sn-glycerol-3-phosphate in addition to lyso-ornithine lipid and therefore exhibits relaxed substrate specificity towards the acyl acceptor substrate [76]. It is expected that future studies of the present group of ‘‘lysophos- phatidic acid acyltransferases” will reveal numerous subgroups with slightly different biochemical activities. Little is known about the functions of OLs. S. meliloti mutants deficient in OL biosynthesis do not show any alteration of their macroscopic phenotype. The unability to form OL must be com- bined with deficient DGTS biosynthesis to obtain reduced cell yields when S. meliloti is grown under phosphorus-limiting condi- tions [54]. In R. capsulatus, OLs are required for optimal steady- state amounts of c-type cytochromes [63]. 4.1.3. Hydroxylated ornithine-containing lipids In bacteria like Burkholderia cepacia, Flavobacterium [12,77], Thiobacillus [58], Gluconobacter [78], Streptomyces [12], some Rals- tonia spec. [79], and R. tropici [11] OL also have ester-linked fatty acyl groups with a hydroxyl group at the 2-position. The 2-hydrox- yfatty acyl residues are not formed during standard fatty acid bio- synthesis and specific enzymatic activities are required to introduce a hydroxyl group onto the 2-position of a fatty acyl res- idue. Similar S-2-hydoxyfatty acyl moieties are integral parts of Salmonella typhimurium lipid A and are thought to be of importance for pathogenesis of this organism. The S-2-hydroxylation is intro- duced after the fatty acyl group had been attached to the lipid A molecule and is catalyzed by the Fe2+/O2/a-ketoglutarate-depen- dent LpxO-encoded dioxygenase [80,81]. It has been speculated that the hydroxyl groups might increase hydrogen bonding be- tween adjacent lipid A molecules decreasing the outer membrane’s permeability to lipophilic compounds under some growth condi- tions [81]. A similar dioxygenase (OlsC) might be responsible for the introduction of 2-hydroxy substitutions on the ester-linked fatty acyl group of OL [11] (Fig. 5). 2-Hydroxy substitutions on es- ter-linked fatty acyl groups occur also in bacterial sphingolipids and PE, other major components of the outer membrane in Gram-negative bacteria. PE is 2-hydroxylated on its sn-2-fatty acyl residue in Burkholderia [82,83]. Homologues of S. typhimurium Fig. 6. Unrooted phylogenetic tree of Sinorhizobium meliloti OlsB, Burkholderia cenocepacia OlsB and OlsB-like ORFs. The tree was constructed using the program CLUSTALW (http://www.expasy.ch/). Distances between sequences are expressed as 0.08 changes per amino acid residue. Accession numbers are as follows: Sinorhizobium meliloti 1021 OlsB (NP_384499), Burkholderia cenocepacia J2315 OlsB (YP_002230419), Escherichia coli CFT073 HlyC (NP_755444), Pantoea stewartii subsp. stewartii EsaI (AAA82096), Pseudomonas aeruginosa PAO1 LasI (NP_250123), N-acyl transferase FeeM from an uncultured bacterium (AAM97306), Agrobacterium tumefaciens str. C58 (NP_353376), Brucella melitensis 16M (NP_540717), Dinoroseobacter shibae DFL 12 (YP_001532815 and YP_001533050), Loktanella vestfoldensis SKA53 (ZP_01003503 and ZP_01003690), Magnetospirillum magnetotacticum MS-1 (ZP_00050271, ZP_00055184, and ZP_00053729), Mesorhizobium loti MAFF303099 (NP_104372), Mycobacterium tuberculosis H37Rv (NP_217543), Oceanicola batsensis HTCC2597 (ZP_00999281 and ZP_00999194), Oceanobulbus indolifex HEL-45 (ZP_02153706 and ZP_02153839), Octadecabacter antarticus 307 (EDY79302 and EDY80135), Paracoc- cus denitrificans PD1222 (ZP_00631241 and ZP_00629627), Phaeobacter gallaeciensis 2.10 (ZP_02148428 and ZP_02148788), Pseudomonas aeruginosa PAO1 (NP_253040), Rhodobacter sphaeroides 2.4.1 (YP_354511 and YP_352676), Roseob- acter denitrificans OCh 114 (YP_682889 and YP_683344), and Silicibacter pomeroyi DSS-3 (YP_167215 and YP_167705). OlsB1 and OlsB2 assign the two different subgroups of OlsB homologues present in bacteria from the order Rhodobacterales. O. Geiger et al. / Progress in Lipid Research 49 (2010) 46–60 53 62 LpxO are found in the genome of Burkholderia cenocepacia J2315 (BCAM1214 and BCAM2401) and they might be candidates for introducing hydroxyl groups into the esterified fatty acyl residue of OL. Rhizobium tropici CIAT899, an efficient symbiont of bean plants, is highly tolerant to acid, and produces four different classes of OL (termed S1, S2, P1, and P2) [11]. A mutant deficient in olsC is sym- biotically defective and does not form P1 or P2. Overexpression of the olsC gene in the olsC-deficient mutant yielded P1 and P2 as ma- jor OLs, coupled with a near-complete lack of S1 and S2 and an acid-sensitive phenotype [11]. These results suggest that some classes of OL are important for acid tolerance (S1 and S2) and oth- ers for symbiotic effectiveness (P1 and P2), but in order to optimize both traits, an adequate balance of the four distinct classes of OLs is required [11]. The product encoded by olsC is a putative LpxO-like dioxygenase that might convert the two less polar forms of OLs (S1 and S2) to the two more polar forms (P1 and P2) [11] by hydroxyl- ation at an unknown position. The OlsC of R. tropici is predicted to be a water-soluble protein of 281 amino acids [11]. 4.1.4. Tauro-ornithine- and lysine-containing lipids In Gluconobacter cerinus, ornithine-containing lipids hydroxyl- ated in the 2-position of the ester-linked fatty acyl residue (2- OH-OL) are partially modified with a taurine residue that is amide-linked to the a-carboxy group of ornithine [78] (Fig. 7). The particulate fraction from G. cerinus requires ATP and Mn2+ to condense taurine to 2-OH-OL leading to the formation of tauro- ornithine lipid [84]. This tauro-ornithine lipid is also called cerili- pin after the species of the bacterium from which it was isolated [78]. The gene encoding the taurine-condensing activity is unknown. A lysine-containing lipid (LL) from an Agrobacterium tumefaciens strain [85] has the a-amino group of lysine N-acylated with a 3- hydroxypalmitoyl residue that is esterified with a fatty acid (Fig. 7). This LL is analogous to the OL with lysine instead of orni- thine as a building block. 4.2. Glycine-containing lipids The glycine-containing lipids (GLs) were identified in the glid- ing bacterium Cytophaga johnsonae C21 and the Gram-negative sea-water bacterium Cyclobacterium marinus WH [86,87]. GLs con- sist of the amino acid glycine and two fatty acyl residues, using the acyl-oxyacyl or piggyback structure. The structure of GL from C. marinus WH is principally a N-[3-D-(13-methyltetradecanoyloxy)- 15-methylhexadecanoyl]glycine [87]. In this structure (Fig. 8), an iso-3-hydroxyfatty acyl group is amide-linked to glycine and its 3-hydroxy group is esterified to another iso-fatty acid. The abso- lute configuration of the hydroxy ester is 3-D [87]. This type of GL is called cytolipin because it was initially identified in the genus Cytophaga [86]. It constitutes about 6% and 5% of the total lipids in C. johnsonae C21 [86] and C. marinus WH [87], respectively. Based on chromatographic methods and specific stains it was assumed that lipoamino acids structurally similar to GL are presents in sev- eral gliding bacteria of the genus Cytophaga [88], and in some strains of Gram-negative fresh-water bacteria belonging to the genera Arcocella [89] and Flectobacillus [90] related to the genus Cyclobacterium. The three latter genera are systematically distant from the family Cytophagaceae and therefore it was suggested that GL might be widely distributed among Gram-negative aquatic bac- teria [87]. However, an alternative explanation might be that the structural genes for GL formation were transferred horizontally. Fig. 7. Structures of tauro-ornithine-containing lipid (cerilipin) and of agrobacterial lysine-containing lipid (LL). Fig. 8. Structures of glycine-containing lipid (GL) and of serineglycine-containing lipid (SGL; flavolipin). 54 O. Geiger et al. / Progress in Lipid Research 49 (2010) 46–60 63 The genes involved in the GL biosynthesis are not known although one might expect that they are similar to the genes involved in OL biosynthesis, due to the structural similarity between both mole- cules. Homologues to the above-described olsB fusions detected in species of the Alteromonadales are also found in genomes of sev- eral members of the genus Cytophaga. So far nothing is known about functions associated with GL. 4.3. Serineglycine-containing lipids A serineglycine-containing lipid (SGL) was isolated from the opportunistic pathogen Flavobacterium meningosepticum [77]. This SGL was called ‘‘flavolipin” based on the genus name of the bacte- rium from which it was first isolated [77]. The initially proposed flavolipin structure of an N-(3-acyloxyacyl)serine, was incorrect [77,91] because it lacked a glycine residue. The synthesis of flavol- ipin to study its biological activities led to the correct structural assignment as a serineglycine-containing lipid (SGL) (Fig. 8) [92]. Flavolipin is not unique to Flavobacterium species, is also found in C. marinus WH [93], and therefore might be present in other sea-water bacteria as well. Flavolipin constitutes about 21% and 11% of the total lipids in F. meningosepticum [77] and C. marinus WH [93], respectively. Flavolipin shares the GL basic structure but has an additional serine residue (Fig. 8) which suggests that GL is a direct biosynthetic precursor for flavolipin formation in C. marinus WH [87]. In contrast to C. marinus, the seven species ana- lyzed from the Cytophaga genus produce no flavolipin [88]. The genes involved in flavolipin biosynthesis are not known. 4.4. The innate immune response to amino acid-containing acyl- oxyacyl lipids Bacterial lipids with an acyl-oxyacyl structure are recognized by toll-like receptors (TLRs) as pathogen-associated molecular pat- terns and trigger the innate immune response of mammals. The 3-acyl-oxyacylamide structure with (R)-configuration is present in OL, SGL, and lipid A. The best studied example is the bacterial endotoxin lipid A. Lipid A is the reactive part of LPS that stimulates Toll-like receptor 4 (TLR4) and the nuclear factor jB (NF-jB) to produce inflammatory cytokines. MD-2, a molecule that physically associates with TLR4 on the cell surface, confers the LPS respon- siveness on the TLR4 receptor [94]. OL and SGL also induce inflam- matory immune responses, measured by the formation of PGE2, IL- 1b, and tumor necrosis factor a by macrophages [95]. A recent study suggests that even the physical state of lipid A or OL affect their biological activities [96]. OL and SGL can be used as adjuvants [59,97–101], and when injected into mice before exposure to the endotoxin lipid A they prevent the lethal effects of the latter [102]. Because of the structural similarities between the two mol- ecules, OL might function as an antagonistic blocker of lipid A-pro- voked events [102]. Like LPS, the inflammatory immune response- causing SGL signal is transduced via the TLR4–MD-2 complex [103]. 5. Bacterial sphingolipids Although sphingolipids are not amino acid-containing lipids in a strict sense, they are formed by condensing an amino acid (serine) to the fatty acyl-CoA forming the sphingolipid precursor 3-oxo-sphinganine, CoA and CO2 (Fig. 9A). In eukaryotes, sphingo- lipids are ubiquitous and essential components of the plasma membrane and are crucial for signaling and organization of lipid rafts. In contrast, sphingolipids occur only in few bacteria, particu- larly some anaerobes, where they functionally replace other bacterial membrane lipids. Sphingolipids are found in the gen- era Pedobacter [104], Bacteroides, Prevotella, Porphyromonas, Fuso- bacterium, Sphingomonas, Sphingobacterium, Bdellovibrio, Cystobacter, Mycoplasma, Flectobacillus, and possibly Acetobacter [4]. Their Fig. 9. Bacterial sphingolipids. The serine palmitoyl transferase (Spt)-catalyzed initial step of sphingolipid biosynthesis in bacteria (A) and the structure of glycosphingolipid GSL-1 from Sphingomonas (B). O. Geiger et al. / Progress in Lipid Research 49 (2010) 46–60 55 64 occurrence in bacteria is thought so unusual that the genus name of the respective bacterium harbours the prefix ‘‘Sphingo”, i.e. in Sphingomonas and Sphingobacterium. Some Gram-negative bacte- ria, such as Sphingomonas capsulata, lack lipopolysaccharide in their outer membrane and instead have glycosphingolipids as functional replacements. In Sphingomonas paucomobilis, two glyco- sphingolipids differing in their ceramide structures are substituted with the tetrasaccharide Man-Gal-GlcNAc-GlcA [105]. The chirality at carbon atoms C-2 and C-3 of the sphingoid base is D-erythro [4,106]. The variability of glycosphingolipids in the outer bacterial membrane of the Sphingomonadaceae is considerable [107], and only some of these glycosphingolipids, such as GSL-1 (Fig. 9B), are recognized by natural killer T cells which provide an innate- type immune response towards glycosphingolipid-containing bacteria [107,108]. Major molecular species of ceramides in sphingobacteria have been identified as 2-N-20-hydroxy- 130-methyltetradecanoyl-15-methylhexadecasphinganine, 2-N- 130-methyltetradecanoyl-15-methylhexadecasphinganine, and 2-N-130-methyltetradecanoyl-hexadecasphinganine [109–111]. Many Bacteroides species have two types of phosphosphingolipids, ceramide phosphorylethanolamine and ceramide phosphorylglyc- erol [112]. In eukaryotes, the biosynthesis of sphingolipids takes place in five stages. It begins with the condensation of serine and a fatty acyl-CoA to form 3-oxo-sphinganine (Stage 1), followed by its reduction to sphinganine (Stage 2), acylation to N-acylsphinganine (dihydroceramide) (Stage 3), and desaturation to ceramide (Stage 4) [106,113]. In Stage 5, ceramide is modified with different polar groups to form the great diversity of sphingolipids. Although the eukaryotic genes involved in the sphingolipid biosynthetis are known [114,115], little is known in bacteria. An exception is sphin- golipid biosynthesis Step 1 catalyzed by serine palmitoyltransfer- ase (EC 2.3.1.50) (Fig. 9A). Like other oxoamine synthases, the bacterial soluble serine palmitoyltransferase is pyridoxal 50-phos- phate-dependent and performs a Claisen condensation between serine and the acyl-CoA thioester with concomitant decarboxyl- ation [116]. Although the serine palmitoyltransferase from Sphingomonas seems to be cytosolic, the serine palmitoyltransfe- rases from Sphingobacterium multivorum and from Bdellovibrio stol- pii are peripherically associated with the cytoplasmic side of the inner membrane [117,118]. The S. paucimobilis serine palmitoyl- transferase crystal structure [119] at 1.3 Å resolution shows that the enzyme is a symmetrical homodimer with two active sites composed of monomers consisting of three domains. The pyridoxal 50-phosphate cofactor is bound covalently to lysine 265 as an inter- nal aldimine/Schiff base, and the active site is composed of resi- dues from both subunits, located at the bottom of a deep cleft. Other bacterial a-oxoamine synthases are 8-amino-7-oxononano- ate synthase (BioF; EC 2.3.1.47) which catalyzes the formation of 8-amino-7-oxononanoate from 6-carboxyhexanoyl-CoA and L-ala- nine during biotin biosynthesis, 5-aminolevulinate synthase (HemA; EC 2.3.1.37), which catalyzes the formation of 5-aminolev- ulinate from succinyl-CoA and glycine during tetrapyrrole and heme biosynthesis in a-proteobacteria [120], and 2-amino-3- oxobutyrate coenzyme A ligase (Kbl; EC 2.3.1.29), which cleaves 2-amino-3-oxobutyrate into acetyl-CoA and glycine during threo- nine degradation [116]. Phylogenetic analysis of bacterial a-oxo- amine synthases (Fig. 10) suggests that distinct subgroups of serine palmitoyltransferases exist and that the encoding genes fre- quently form an operon with a putative acyl carrier protein gene. This finding suggests specialized acyl carrier proteins, instead of CoA, are used in some cases during the initial step of sphingolipid biosynthesis in bacteria. Based on our analysis (Fig. 10), the ability to form sphingolipids is more widespread in a-proteobacteria (Gluconobacter, Granulibacter, Caulobacter) than previously thought and might occur even in the b-proteobacterium Nitrosomonas and in several pathogenic Escherichia coli strains, i.e. in the enterotoxi- genic E. coli (ETEC) B7A. 6. Sulfonolipids in the Cytophaga group Gram-negative bacteria of the Cytophaga groupmove by gliding. Major lipids in the membranes of Cytophaga johnsonae are sulfon- olipids, OL, and PE. Sulfonolipids and OL are predominantly local- ized to the outer membrane whereas PE is the predominant lipid of the inner membrane [121]. Sulfonolipids contain capnine that is formed by the condensation of cysteate with fatty acyl-CoA un- der the release of CO2 [122,123] (Fig. 11A), in a reaction analogous to the one catalyzed by serine palmitoyltransferase [119]. Capnine is then converted to N-acyl-capnine, the membrane-forming sul- fonolipid. The N-acylated residues are C14, C15, and C16 3-hydrox- ylated iso-fatty acids [124]. Mutants of C. johnsonae, deficient in gliding and sulfonolipid biosynthesis, were isolated and restoration of the sulfonolipid content by providing cysteate resulted in recov- ery of the ability to glide [123]. Therefore, sulfonolipids might be required for gliding motility. A structural variant of capnine exists in another member of the Cytophaga group, Salinibacter ruber [125]. The Salinibacter sulfonolipid contains an extra carboxylate at car- bon 2 and an O-acyl group at carbon 3 (Fig. 11B) that is diagnostic for this extremely halophilic bacterial genus [125]. Fig. 10. Unrooted phylogenetic tree of selected bacterial serine palmitoyltransfe- rases and other a-oxoamine synthases from bacteria. The tree was constructed using the program CLUSTALW (http://www.expasy.ch/). Distances between sequences are expressed as 0.05 changes per amino acid residue. The asterisks label species in which the serine palmitoyltransferase gene forms an operon with a putative acyl carrier protein. Accession numbers are as follows: Bacteroides thetaiotaomicron VPI-5482 (NP_809783, NP_810284, NP_810356), Bacteriovorax stolpii Spt (BAF73753), Caulobacter crescentus CB15 (NP_419978, NP_420168, NP_420387), Escherichia coli B7A (EDV60350, ZP_03029941, ZP_03030227), Gluco- nobacter oxydans 621H (AAW61792, YP_192033, YP_191153), Granulibacter bethesdensis CGDNIH1 (YP_744060, YP_744129, YP_744319), Nitrosomonas eutropha C91 (YP_746703), Porphyromonas gingivalis ATCC 33277 (YP_001929837, YP_001929605, YP_001929054), Sphingobacterium multivorum Spt (BAF73751), Sphingomonas paucimobilis Spt (BAB56013), Sphingomonas wittichii RW1 (YP_001264383, YP_001264306, YP_001261757), and Zymomonas mobilis subsp. mobilis ZM4 (YP_163005, YP_162933, YP_163652). Annotations of genes are: 8- amino-7-oxononanoate synthase (BioF), 2-amino-3-ketobutyrate coenzyme A ligase (Kbl), 5-aminolevulinate synthase (HemA), and serine palmitoyltransferase (Spt). 56 O. Geiger et al. / Progress in Lipid Research 49 (2010) 46–60 65 7. Stress causes changes in bacterial membranes Membrane lipid compositions of bacteria have usually been determined after the organisms had been grown on complex or de- fined culture media. Such determinations are reproducible and have led to the conviction that membrane lipid compositions are characteristic invariable traits of organisms. Unlike animal cells, however, plant and bacterial cells are not embedded in a controlled environment but are subject to many environmental changes and stresses. The bacterial membrane adapts to changing environ- ments by altering the membrane lipid components by which it is formed. It has long been known that reduced temperatures [1] or in- creased hydrostatic pressure [126] cause a reduction of membrane fluidity. In an attempt to maintain the fluidity of their membranes, bacteria include more unsaturated or branched fatty acyl chains into their membrane lipids thereby increasing packing disorder and fluidity of their membranes [127]. Acid stress in proteobacteria causes modifications of membrane lipids, such as formation of ly- syl-PG [10], alanyl-PG [46] or the hydroxylation of OL [11]. Under phosphorus-limiting conditions, membrane phospholipids of some bacteria are partially replaced by lipids without phosphorus as demonstrated in Bacillus subtilis [128], Pseudomonas diminuta [129], P. fluorescens [130], and Rhodobacter sphaeroides [50]. In S. meliloti, these phosphorus-free lipids are sulfoquinovosyl diacyl- glycerol, OL, and DGTS [51]. In other bacteria, these phosphorus- free lipids include glycolipids as well. The ability to form OL or DGTS contributes to increased cell yields when S. meliloti is grown under the phosphorus-limiting condition [54]. In S. meliloti, the amounts of OL formed are strongly dependent on growth condi- tions [61]. However, in clinical isolates of Flavobacterium [77], Burkholderia isolates [82] and in pathogenic Brucella and Bordetella species [131], OL are normally major membrane lipids. Lipid A-containing lipopolysaccharides (LPS) usually cover the outer surface of the outer membrane in Gram-negative bacteria and pose a major permeability barrier for hydrophilic and hydro- phobic compounds. It is assumed that the hydrocarbon regions of the outer membrane are in a gel-like state of very low fluidity un- der physiological conditions [132]. Strong interactions between the lipid molecules forming the outer membrane are probably key to this gel-like behaviour and to its functions as a permeability bar- rier. Different environments/stresses require adjustments in the outer membrane that are accomplished by certain chemical modi- fications of LPS. The absence of divalent cations (Mg2+ and Ca2+) will destabilize the outer membrane, and low Mg2+ concentrations activate the PhoPQ system to trigger a number of modifications of the LPS of S. typhimurium to stabilize the outer membrane [133]. Among these membrane-stabilizing modifications is the LpxO-cat- alyzed 2-hydroxylation of an esterified acyl residue of the lipid A of LPS [80,81]. Introduction of an additional hydroxyl group into the fatty acyl chain of a membrane lipid increases hydrogen bonding with neighboring molecules leading to membrane stabilization. In some bacterial groups, other lipids occur in the outer membrane either in addition to or in place of LPS. These include sphingolipids, sulfonolipids [132], and OLs [60]. Each of these outer membrane lipids have fatty acids with hydroxyl groups at the 2- and/or 3-po- sition to stabilize the membrane. In addition, PE, which is enriched in the outer membrane, is 2-hydroxylated on its sn-2 fatty acid in Burkholderia [83]. 8. Conclusions and perspectives Membrane lipids act to form the lipid bilayer surrounding every cell and interact with other biomolecules based on their distinct chemical nature. Although phosphatidylserine has been Fig. 11. Proposed pathway for sulfonolipid biosynthesis and an unusual sulfonolipid from Salinibacter. O. Geiger et al. / Progress in Lipid Research 49 (2010) 46–60 57 66 extensively studied in eukaryotes, other, less universal amino acid- containing membrane lipids are less well-known. Ornithine-con- taining lipids are formed in a direct two-step pathway whereas more steps are needed to form any of the glycerophospholipids. Therefore, making ornithine-containing lipids and other amino acid-containing acyl-oxyacyl lipids might be an easy way to build membranes in primitive biological systems. However, OL are cer- tainly less resistant to extreme environmental conditions than archaeal ether lipids and probably would not have been of use when life originated on earth. Also, no OL-containing bacterium is known that is totally devoid of glycerophospholipids, leaving the question open whether a functional membrane can be formed by phosphorus-free membrane lipids only. Resolving the genetics and biochemistry of lysyl-phosphatidylglycerol, diacylglyceryl trimethylhomoserine, and ornithine-containing lipids in recent years has revealed the importance of these lipids in adapting to stress conditions and for the survival of bacteria. Nevertheless, we are only beginning to understand the functions of some of the amino acid-containing bacterial membrane lipids. The addition of amino acids into the structure of membrane lipids increases structural and chemical diversity, modifies net charge and polarity, and permits interaction with elements in the environment. For many minor amino acid-containing membrane lipids not much more than their structure and the producing bacterium are known. However, the immense information on bacterial genomes and im- proved bioinformatic tools will accelerate the detection of struc- tural genes for many amino acid-containing bacterial membrane lipids. In addition to the more traditional approaches, another ave- nue for discovering new amino acid-containing bacterial mem- brane lipids will be the expression of metagenomic libraries and a subsequent screening for lipids. Finally, more biochemical stud- ies are needed on the biosynthesis pathways as well as structural studies on the enzymes involved to provide feedback to impove bioinformatic predictions for ORFs involved in the biosynthesis of amino acid-containing bacterial membrane lipids. Acknowledgments This review is dedicated to Eugene P. Kennedy on the occasion of his 90th birthday. Work in our laboratory was supported by grants from DGAPA/UNAM (IN200806, IN217907, and IN218009) and the Consejo Nacional de Ciencia y Tecnología de México (CONACyT 46020-N, 49738-Q, and 82614). References [1] Rock CO. Fatty acid and phospholipid metabolism in prokaryotes. In: Vance DE, Vance JE, editors. Biochemistry of lipids lipoproteins and membranes. Amsterdam: Elsevier; 2008. p. 59–96. [2] Raetz CR, Reynolds CM, Trent MS, Bishop RE. Lipid A modification systems in Gram-negative bacteria. Annu Rev Biochem 2007;76:295–329. [3] Sohlenkamp C, López-Lara IM, Geiger O. Biosynthesis of phosphatidylcholine in bacteria. Prog Lipid Res 2003;42:115–62. [4] Olsen I, Jantzen E. Sphingolipids in bacteria and fungi. Anaerobe 2001;7:103–12. [5] Jorasch P, Wolter FP, Zähringer U, Heinz E. A UDP glucosyltransferase from Bacillus subtilis successively transfers up to four glucose residues to 1,2- diacylglycerol: expression of ypfP in Escherichia coli and structural analysis of its reaction products. Mol Microbiol 1998;29:419–30. [6] Roy H, Ibba M. RNA-dependent lipid remodeling by bacterial multiple peptide resistance factors. Proc Natl Acad Sci USA 2008;105:4667–72. [7] Jackson M, Crick DC, Brennan PJ. Phosphatidylinositol is an essential phospholipid in mycobacteria. J Biol Chem 2000;275:30092–9. [8] Summons RE, Bradley AS, Jahnke LL, Waldbauer JR. Steroids, triterpenoids and molecular oxygen. Philos Trans Roy Soc B 2006;361:951–68. [9] López-Lara IM, Sohlenkamp C, Geiger O. Membrane lipids in plant-associated bacteria: their biosyntheses and possible functions. Mol Plant Microbe Interact 2003;16:567–79. [10] Sohlenkamp C, Galindo-Lagunas KA, Guan Z, Vinuesa P, Robinson S, Thomas Oates J, et al. The lipid lysyl-phosphatidylglycerol is present in membranes of Rhizobium tropici CIAT899 and confers increased resistance to polymyxin B under acidic growth conditions. Mol Plant Microbe Interact 2007;20:1421–30. [11] Rojas-Jiménez K, Sohlenkamp C, Geiger O, Martínez-Romero E, Werner D, Vinuesa P. A CIC chloride channel homolog and ornithine-containing membrane lipids of Rhizobium tropici CIAT899 are involved in symbiotic efficiency and acid tolerance. Mol Plant Microbe Interact 2005;18: 1175–85. [12] Asselineau J. Bacterial lipids containing amino acids or peptides linked by amide bonds. Fortschr Chem Org Naturst 1991;56:1–85. [13] Brady SF, Chao CJ, Clardy J. New natural product families from an environmental DNA (eDNA) gene cluster. J Am Chem Soc 2002;124:9968–9. [14] Desai JD, Banat IM. Microbial production of surfactants and their commercial potential. Microbiol Mol Biol Rev 1997;61:47–64. [15] Dowhan W, Bogdanov M, Mileykovskaya E. Functional roles of lipids in membranes. In: Vance DE, Vance JE, editors. Biochemistry of lipids, lipoproteins and membranes. Amsterdam, The Netherlands: Elsevier; 2008. p. 1–37. [16] Sohlenkamp C, de Rudder KEE, Geiger O. Phosphatidylethanolamine is not essential for growth of Sinorhizobium meliloti on complex culture media. J Bacteriol 2004;186:1667–77. [17] Koonin EV. A duplicated catalytic motif in a new superfamily of phosphohydrolases and phospholipid synthases that includes poxvirus envelope proteins. Trends Biochem Sci 1996;21:242–3. [18] Vance DE, Vance JE. Phospholipid biosynthesis in eukaryotes. In: Vance DE, Vance JE, editors. Biochemistry of lipids, lipoproteins and membranes. Amsterdam, The Netherlands: Elsevier; 2008. p. 213–44. [19] Langley KE, Yaffe MP, Kennedy EP. Biosynthesis of phospholipids in Bacillus megaterium. J Bacteriol 1979;140:996–1007. [20] Nguyen NA, Sallans L, Kaneshiro ES. The major glycerophospholipids of the predatory and parasitic bacterium Bdellovibrio bacteriovorus HID5. Lipids 2008;43:1053–63. [21] Hawrot E, Kennedy EP. Phospholipid composition and membrane function in phosphatidylserine decarboxylase mutants of Escherichia coli. J Biol Chem 1978;253:8213–20. [22] Matsumoto K, Okada M, Horikoshi Y, Matsuzaki H, Kishi T, Itaya M, et al. Cloning, sequencing, and disruption of the Bacillus subtilis psd gene coding for phosphatidylserine synthase. J Bacteriol 1998;180:100–6. [23] Vences-Guzmán MA, Geiger O, Sohlenkamp C. Sinorhizobium meliloti mutants deficient in phosphatidylserine decarboxylase accumulate phosphatidylserine and are strongly affected during symbiosis with alfalfa. J Bacteriol 2008;190:6846–56. [24] Vance JE, Steenbergen R. Metabolism and functions of phosphatidylserine. Prog Lipid Res 2005;44:207–34. [25] Rontein D, WuWI, Voelker DR, Hanson AD. Mitochondrial phosphatidylserine decarboxylase from higher plants. Functional complementation in yeast, localization in plants, and overexpression in Arabidopsis. Plant Physiol 2003;132:1678–87. [26] Langley KE, Hawrot E, Kennedy EP. Membrane assembly: movement of phosphatidylserine between the cytoplasmic and outer membranes of Escherichia coli. J Bacteriol 1982;152:1033–41. [27] Lu Y-J, Zhang Y-M, Grimes KD, Qi J, Lee RE, Rock CO. Acyl-phosphates initiate membrane phospholipid synthesis in Gram-positive pathogens. Mol Cell 2006;23:765–72. [28] O’Leary WM, Wilkinson SG. Gram-positive bacteria. In: Ratledge S, Wilkinson SG, editors. Microbial lipids. London: Academic Press; 1988. p. 117–201. [29] Nahaie MR, Goodfellow M, Minnikin DE, Hajek V. Polar lipid and isoprenoid quinine composition in the classification of Staphylococcus. J Gen Microbiol 1984;130:2427–37. [30] Op den Kamp JA, Redai I, van Deenen LL. Phospholipid composition of Bacillus subtilis. J Bacteriol 1969;99:298–303. [31] Samant S, Hsu FF, Neyfakh AA, Lee H. The Bacillus anthracis protein MprF is required for synthesis of lysylphosphatidylglycerols and for resistance to cationic antimicrobial peptides. J Bacteriol 2009;191:1311–9. [32] Fischer W, Leopold K. Polar lipids of four Listeria species containing L- lysylcardiolipin, a novel lipid structure, and other unique phospholipids. Int J Syst Bacteriol 1999;49:653–62. [33] Houtsmuller UM, van Deenen L. Identification of a bacterial phospholipid as an O-ornithine ester of phosphatidyl glycerol. Biochim Biophys Acta 1963;70:211–3. [34] Gould RM, Lennarz WJ. Biosynthesis of aminoacyl derivatives of phosphatidylglycerol. Biochem Biophys Res Commun 1967;26:512–4. [35] Dos Santos Mota JM, Op den Kamp JAF, Verheij HM, van Deenen LLM. Phospholipids of Streptococcus faecalis. J Bacteriol 1970;104:611–9. [36] Tocanne JF, Verheij HM, den Kamp JA, van Deenen LL. Chemical and physicochemical studies of lysylphosphatidylglycerol derivatives. Occurrence of a 20 yields 30 lysyl migration. Chem Phys Lipids 1974;13:389–403. [37] Peschel A, Jack RW, Otto M, Collins VL, Staubitz P, Nicholson G, et al. Staphylococcus aureus resistance to human defensins and evasion of neutrophil killing via the novel virulence factor MprF is based on modification of membrane lipids with L-lysine. J Exp Med 2001;193:1067–76. [38] Oku Y, Kurokawa K, Ichihashi N, Sekimuzu K. Characterization of the Staphylococcus aureus mprF gene, involved in lysinylation of phosphatidylglycerol. Microbiology 2004;150:45–51. [39] Staubitz P, Neumann H, Schneider T, Wiedemann L, Peschel A. MprF- mediated biosynthesis of lysylphosphatidylglycerol, an important 58 O. Geiger et al. / Progress in Lipid Research 49 (2010) 46–60 67 determinant in staphylococcal defensing resistance. FEMS Microbiol Lett 2004;231:67–71. [40] Salzberg LI, Helmann JD. Phenotypic and transcriptomic characterization of Bacillus subtilis mutants with grossly altered membrane composition. J Bacteriol 2008;190:7797–807. [41] Kenward MA, Brown MRW, Fryer JJ. The influence of calcium or manganese on the resistance to EDTA, polymyxin B or cold shock, and the composition of Pseudomonas aeruginosa grown in glucose- or magnesium-depleted batch cultures. J Appl Bacteriol 1979;47:489–503. [42] Jones DE, Smith JD. Phospholipids of the differentiating bacterium Caulobacter crescentus. Can J Biochem 1979;57:424–8. [43] Reeve WG, Tiwari RP, Worsley PS, Dilworth MJ, Glenn AR, Howieson JG. Constructs for insertional mutagenesis, transcriptional signal localisation and gene regulation studies in root nodule and other bacteria. Microbiology 1999;145:1307–16. [44] Vinuesa P, Neumann-Silkow F, Pacios-Bras C, Spaink HP, Martinez-Romero E, Werner D. Genetic analysis of a pH-regulated operon from Rhizobium tropici CIAT899 involved in acid tolerant and nodulation competitiveness. Mol Plant Microbe Interact 2003;16:159–68. [45] Reeve WG, Tiwari RP, Bräu l, Castelli J, Garau G, Dilworth MJ, et al. The Sinorhizobium medicae WSM419 lpiA gene is transcriptionally activated by low pH and regulated by FsrR. Microbiology 2006;152:3049–59. [46] Klein S, Lorenzo C, Hoffmann S, Walther JM, Storbeck S, Piekarski T, et al. Adaptation of Pseudomonas aeruginosa to various conditions includes tRNA- dependent formation of alanyl-phosphatidylglycerol. Mol Microbiol 2009;71:551–65. [47] Fischer W, Arneth-Seifert D. D-alanylcardiolipin, a major component of the unique lipid pattern of Vagococcus fluvialis. J Bacteriol 1998;180:2950–7. [48] Peter-Katalinic J, Fischer W. F-D-Glucopyranosyl-, D-alanyl- and L- lysylcardiolipin from Gram-positive bacteria: analysis by fast atom bombardment mass spectrometry. J Lipid Res 1998;39:2286–92. [49] Thedieck K, Hain T, Mohamed W, Tindall BJ, Nimtz M, Chakraborty T, et al. The MprF protein is required for lysinylation of phospholipids in listerial membranes and confers resistance to cationic antimicrobial peptides (CAMPs) on Listeria monocytogenes. Mol Microbiol 2006;62: 1325–39. [50] Benning C, Huang ZH, Gage DA. Accumulation of a novel glycolipid and a betaine lipid in cells of Rhodobacter sphaeroides grown under phosphate limitation. Arch Biochem Biophys 1995;317:103–11. [51] Geiger O, Röhrs V, Weissenmayer B, Finan TM, Thomas-Oates JE. The regulator gene phoB mediates phosphate stress-controlled synthesis of the membrane lipid diacylglyceryl-N,N,N-trimethylhomoserine in Rhizobium (Sinorhizobium) meliloti. Mol Microbiol 1999;32:63–73. [51a] Van Mooy BA, Fredricks HF, Pedler BE, Dyhrman ST, Karl DM, Koblízek M, et al. Phytoplankton in the ocean use non-phosphorus lipids in response to phosphorus scarcity. Nature 2009;458:69–72. [52] Klug RM, Benning C. Two enzymes of diacylglyceryl-O-40-(N,N,N,-trimethyl) homoserine biosynthesis are encoded by btaA and btaB in the purple bacterium Rhodobacter sphaeroides. Proc Natl Acad Sci USA 2001;98: 5910–5915. [53] Riekhof WR, Andre C, Benning C. Two enzymes, BtaA and BtaB, are sufficient for betaine lipid biosynthesis in bacteria. Arch Biochem Biophys 2005;441:96–105. [54] López-Lara IM, Gao JL, Soto MJ, Solares-Pérez A, Weissenmayer B, Sohlenkamp C, et al. Phosphorus-free membrane lipids of Sinorhizobium meliloti are not required for the symbiosis with alfalfa but contribute to increased cell yields under phosphorus-limiting conditions of growth. Mol Plant Microbe Interact 2005;18:973–82. [55] Riekhof WR, Sears BB, Benning C. Annotation of genes involved in glycerolipid biosynthesis in Chlamydomonas reinhardtii: discovery of the betaine lipid synthase BTA1Cr. Eukaryot Cell 2005;4:242–52. [56] Madigan MT, Martinko JM, Dunlap PV, Clark DP. Brock biology of microorganisms. 12th ed. San Francisco, CA: Pearson Benjamin Cummings; 2008. [57] Lerouge P, Lebas MH, Agapakis-Caussé C, Promé JC. Isolation and structural characterization of a new non-phosphorylated lipoamino acid from Mycobacterium phlei. Chem Phys Lipids 1988;49:161–6. [58] Knoche HW, Shively JM. The structure of an ornithine-containing lipid from Thiobacillus thiooxidans. J Biol Chem 1972;247:170–8. [59] Kawai Y, Nakagawa Y, Matuyama T, Akagawa K, Itagawa K, Fukase K, et al. A typical bacterial ornithine-containing lipid Na-(D)-[3- (hexadecanoyloxy)hexadecanoyl]-ornithine is a strong stimulant for macrophages and a useful adjuvant. FEMS Immunol Med Microbiol 1999;23:67–73. [60] Dees C, Shively JM. Localization and quantitation of the ornithine lipid of Thiobacillus thiooxidans. J Bacteriol 1982;149:798–9. [61] Gao JL, Weissenmayer B, Taylor AM, Thomas-Oates J, López-Lara IM, Geiger O. Identification of a gene required for the formation of lyso-ornithine lipid, an intermediate in the biosynthesis of ornithine-containing lipids. Mol Microbiol 2004;53:1757–70. [62] Weissenmayer B, Gao JL, López-Lara IM, Geiger O. Identification of a gene required for the biosynthesis of ornithine-derived lipids. Mol Microbiol 2002;45:721–33. [63] Aygun-Sunar S, Mandaci S, Koch HG, Murria IVJ, Goldfine H, Daldal F. Ornithine lipid is required for optimal steady-state amounts of c-type cytochromes in Rhodobacter capsulatus. Mol Microbiol 2006;61:418–35. [64] Gough J, Karplus K, Hughey R, Chothia C. Assignment of homology to genome sequences using a library of hidden Markov models that represent all proteins of known structure. J Mol Biol 2001;313:903–19. [65] Heath RJ, Rock CO. A conserved histidine is essential for glycerolipid acyltransferase catalysis. J Bacteriol 1998;180:1425–30. [66] Watson WT, Minogue TD, Val DL, Beck von Bodman S, Churchill MEA. Structural basis and specificity of acyl-homoserine lactone signal production in bacterial quorum sensing. Mol Cell 2002;9:685–94. [67] Gould TA, Schweizer HP, Churchill MEA. Structure of the Pseudomonas aeruginosa acylhomoserine lactone synthase LasI. Mol Microbiol 2004;53:1135–46. [68] Van Wagoner RM, Clardy J. FeeM, an N-acyl amino acid synthase from an uncultured soil microbe: Structure, mechanism, and acyl carrier protein binding. Structure 2006;14:1425–35. [69] Zhang X, Ferguson-Miller SM, Reid GE. Characterization of ornithine and glutamine lipids extracted from cell membranes of Rhodobacter sphaeroides. J Am Soc Mass Spectrom 2009;20:198–212. [70] Schüler D. Molecular analysis of a subcellular compartment: the magnetosome membrane in Magnetospirillum gryphiswaldense. Arch Microbiol 2004;181:1–7. [71] Brady SF, Clardy J. N-acyl derivatives of arginine and tryptophan isolated from environmental DNA expressed in Escherichia coli. Org Lett 2005;7:3613–6. [72] Clardy J, Brady SF. Cyclic AMP directly activates NasP, an N-acyl amino acid antibiotic biosynthetic enzyme cloned from an uncultured b- proteobacterium. J Bacteriol 2007;189:6487–9. [73] West J, Tompkins CK, Balantac N, Nudelman E, Meengs B, White T, et al. Cloning and expression of two human lysophosphatidic acid acyltransferase cDNAs that enhance cytokine-induced signaling responses in cells. DNA Cell Biol 1997;16:691–701. [74] Shih GC, Kahler CM, Swartley JS, Rahman MM, Coleman J, Carlson RW, et al. Multiple lysophosphatidic acid acyltransferases in Neisseria meningitidis. Mol Microbiol 1999;32:942–52. [75] Cullinane M, Baysse C, Morrissey JP, O’Gara F. Identification of two lysophosphatidic acid acyl transferase genes with overlapping function in Pseudomonas fluorescens. Microbiology 2005;151:3071–80. [76] Aygun-Sunar S, Bilaloglu R, Goldfine H, Daldal F. Rhodobacter capsulatus OlsA is a bifunctional enyzme active in both ornithine lipid and phosphatidic acid biosynthesis. J Bacteriol 2007;189:8564–74. [77] Kawai Y, Yano I, Kaneda K. Various kinds of lipoamino acids including a novel serine-containing lipid in an opportunistic pathogen Flavobacterium. Their structures and biological activities on erythrocytes. Eur J Biochem 1988;171:73–80. [78] Tahara Y, Kameda M, Yamada Y, Kondo K. A new lipid; the ornithine and taurine-containing ‘‘cerilipin”. Agric Biol Chem 1976;40:243–4. [79] Galbraith L, Jonsson MH, Rudhe LC, Wilkinson SG. Lipids and fatty acids of Burkholderia and Ralstonia species. FEMS Microbiol Lett 1999;173:359–64. [80] Gibbons HS, Lin S, Cotter RJ, Raetz CR. Oxygen requirement for the biosynthesis of the S-2-hydroxymyristate moiety in Salmonella typhimurium lipid A. Function of LpxO, a new Fe2+/alpha-ketoglutarate-dependent dioxygenase homologue. J Biol Chem 2000;275:32940–9. [81] Gibbons HS, Reynolds CM, Guan Z, Raetz CRH. An inner membrane dioxygenase that generates the 2-hydroxymyristate moiety of Salmonella lipid A. Biochemistry 2008;47:2814–25. [82] Phung LV, Tran TB, Hotta H, Yabuuchi E, Yano I. Cellular lipid and fatty acid compositions of Burkholderia pseudomallei strains isolated from human and environment in Viet Nam. Microbiol Immunol 1995;39:105–16. [83] Taylor CJ, Anderson AJ, Wilkinson SG. Phenotypic variation of lipid composition in Burkholderia cepacia: a response to increased growth temperature is a greater content of 2-hydroxy acids in phosphatidylethanolamine and ornithine amide lipid. Microbiology 1998;144:1737–45. [84] Tahara Y, Shinmoto K, Yamada Y, Kondo K. Enzymatic synthesis of tauro- ornithine lipid in Gluconobacter cerinus. Agric Biol Chem 1978;42:205–6. [85] Tahara Y, Yamada Y, Kondo K. A new lysine-containing lipid isolated from Agrobacterium tumefaciens. Agric Biol Chem 1976;40:1449–50. [86] Kawazoe R, Okuyama H, Reichardt W, Sasaki S. Phospholipids and a novel glycine-containing lipoamino acid in Cytophaga johnsonae Stanier strain C21. J Bacteriol 1991;173:5470–5. [87] Batrakov SG, Nikitin DI, Mosezhnyi AE, Ruzhitsky AO. A glycine-containing phosphorus-free lipoaminoacid from the Gram-negative marine bacterium Cyclobacterium marinus WH. Chem Phys Lipids 1999;99:139–43. [88] Kawazoe R, Monde K, Reichardt W, Okuyama H. Lipoamino acids and sulfonolipids in Cytophaga johnsonae Stanier strain C21 and six related species of Cytophaga. Arch Microbiol 1992;158:171–5. [89] Nikitin DI, Oranskaya MS, Pitryuk IA, Chernykh NA, Lysenko AM. A new ring- forming bacterium Arcocella aquatica gen. et sp. nov. Microbiology (Moscow) 1994;63:87–90 [Engl. Transl.]. [90] Raj HD, Maloy SR. Proposal of Cyclobacterium marinus gen. nov., comb. nov. for a marine bacterium previously assigned to the genus Flectobacillus. Int J Syst Bacteriol 1990;40:337–47. [91] Shiozaki M, Deguchi N, Ishikawa T, Haruyama H, Kawai Y, Nishijima M. Revised structure of flavolipin and synthesis of its isomers. Tetrahedron Lett 1998;39:4497–500. [92] Shiozaki M, Deguchi N, Mochizuki T, Wakabayashi T, Ishikawa T, Haruyama H, et al. Revised structure and synthesis of flavolipin. Tetrahedron 1998;54:11861–76. O. Geiger et al. / Progress in Lipid Research 49 (2010) 46–60 59 68 [93] Batrakov SG, Nikitin DI, Sheichenko VI, Ruzhitsky AO. A novel sulfonic-acid analogue of ceramide is the major extractable lipid of the Gram-negative marine bacterium Cyclobacterium marinus WH. Biochim Biophys Acta 1998;1391:79–91. [94] Shimazu R, Akashi S, Ogata H, Nagai Y, Fukudome K, Miyake K, et al. MD-2, a molecule that confers lipopolysaccharide responsiveness on Toll-like receptor 4. J Exp Med 1999;189:1777–82. [95] Kawai Y, Akagawa K. Macrophage activation by an ornithine-containing lipid or a serine-containing lipid. Infect Immun 1989;57:2086–91. [96] Okemoto K, Hanada K, Nishijima M, Kawasaki K. The preparation of a lipidic endotoxin affects its biological activities. Biol Pharm Bull 2008;31:1952–4. [97] Kato H, Goto N. Adjuvanticity of an ornithine-containing lipid of Flavobacterium meningosepticum as a candidate vaccine adjuvant. Microbiol Immunol 1997;41:101–6. [98] Kawai Y, Kamoshita K, Akagawa K. B-lymphocyte mitogenicity and adjuvanticity of an ornithine-containing lipid or a serine-containing lipid. FEMS Microbiol Lett 1991;67:127–9. [99] Kawai Y, Takasuka N, Inoue K, Akagawa K, Nishijima M. Ornithine-containing lipids stimulate CD14-dependent TNF-alpha production from murine macrophage-like J774.1 and RAW 264.7 cells. FEMS Immunol Med Microbiol 2000;28:197–203. [100] Kawai Y, Okawarab AI, Okuyama H, Kura F, Suzuki K. Modulation of chemotaxis, O(2)() production and myeloperoxidase release from human polymorphonuclear leukocytes by the ornithine-containing lipid and the serineglycine-containing lipid of Flavobacterium. FEMS Immunol Med Microbiol 2000;28:205–9. [101] Kawai Y, Watanabe M, Matsuura M, Nishijima M, Kawahara K. The partially degraded lipopolysaccharide of Burkholderia cepacia and ornithine- containing lipids derived from some Gram-negative bacteria are useful complex lipid adjuvants. FEMS Immunol Med Microbiol 2002;34:173–9. [102] Kawai Y, Kaneda K, Morisawa Y, Akagawa K. Protection of mice from lethal endotoxemia by use of an ornithine-containing lipid or a serine-containing lipid. Infect Immun 1991;59:2560–6 [Erratum: Infect Immun 1992;60:320]. [103] Gomi K, Kawasaki K, Kawai Y, Shiozaki M, Nishijima M. Toll-like receptor 4- MD-2 complex mediates the signal transduction induced by flavolipin, an amino acid-containing lipid unique to Flavobacterium meningosepticum. J Immunol 2002;168:2939–43. [104] Steyn PL, Segers P, Vancanneyt M, Sandra P, Kersters K, Joubert JJ. Classification of heparinolytic bacteria into a new genus, Pedobacter, comprising four species: Pedobacter heparinus comb. nov., Pedobacter piscium comb. nov., Pedobacter africanus sp. nov. and Pedobacter saltans sp. nov. proposal of the family Sphingobacteriaceae fam. nov.. Int J Syst Bacteriol 1998;48:165–77. [105] Kawahara K, Seydel U, Matsuura M, Danbara H, Rietschel ET, Zähringer U. Chemical structure of glycosphingolipids isolated from Sphingomonas paucimobilis. FEBS Lett 1991;292:107–10. [106] Hanada K. Serine palmitoyltransferase, a key enzyme of sphingolipid metabolism. Biochim Biophys Acta 2003;1632:16–30. [107] Kinjo Y, Pei B, Bufali S, Raju R, Richardson SK, Imamura M, et al. Natural Sphingomonas glycolipids vary greatly in their ability to activate natural killer T cells. Chem Biol 2008;15:654–64. [108] Wu D, Zajonc DM, Fujio M, Sullivan BA, Kinjo Y, Kronenberg M, et al. Design of natural killer T cell activators: Structure and function of a microbial glycosphingolipid bound to mouse CD1d. Proc Natl Acad Sci USA 2006;103:3972–7. [109] Yano I, Tomiyasu I, Yabuuchi E. Long chain base composition of strains of three species of Sphingobacterium gen. nov.. FEMS Microbiol Lett 1982;15:303–7. [110] Yano I, Imaizumi S, Tomiyasu I, Yabuuchi E. Separation and analysis of free ceramides containing 2-hydroxy fatty acids in Sphingobacterium species. FEMS Microbiol Lett 1983;20:449–53. [111] Naka T, Fujiwara N, Yano I, Maeda S, Doe M, Minamino M, et al. Structural analysis of sphingophospholipids derived from Sphingobacterium spiritivorum, the type species of genus Sphingobacterium. Biochim Biophys Acta 2003;1635:83–92. [112] Kato M, Muto Y, Tanaka-Bandoh K, Watanabe K, Ueno K. Sphingolipid composition in Bacteroides species. Anaerobe 1995;1:135–9. [113] Nieto FL, Pescio LG, Favale NO, Adamo AM, Sterin-Speziale NB. Sphingolipid metabolism is a crucial determinant of cellular fate in non-stimulated proliferating Madin–Darby canine kidney (MDCK) cells. J Biol Chem 2008;283:25682–91. [114] Kihara A, Mitsutake S, Mizutani Y, Igarashi Y. Metabolism and biological functions of two phosphorylated sphingolipids, sphingosine 1-phosphate and ceramide 1-phosphate. Prog Lipid Res 2007;46:126–44. [115] Hirabayashi Y, Furuya S. Roles of L-serine and sphingolipid synthesis in brain development and neuronal survival. Prog Lipid Res 2008;47:188–203. [116] Kerbarh O, Campopiano DJ, Baxter RL. Mechanism of a-oxoamine synthases: identification of the intermediate Claisen product in the 8-amino-7- oxononanoate synthase reaction. Chem Commun 2006:60–2. [117] Ikushiro H, Hayashi H, Kagamiyama H. A water-soluble homodimeric serine palmitoyltransferase from Sphingomonas paucimobilis EY2395T strain. Purification, characterization, cloning, and overproduction. J Biol Chem 2001;276:18249–56. [118] Ikushiro H, Islam MM, Tojo H, Hayashi H. Molecular characterization of membrane-associated soluble serine palmitoyltransferases from Sphingobacterium multivorum and Bdellovibrio stolpii. J Bacteriol 2007;198:5749–61. [119] Yard BA, Carter LG, Johnson KA, Overton IM, Dorward M, Liu H, et al. The structure of serine palmitoyltransferase; gateway to sphingolipid biosynthesis. J Mol Biol 2007;370:870–86. [120] Astner I, Schulze JO, van den Heuvel J, Jahn D, Schubert WD, Heinz DW. Crystal structure of 5-aminolevulinate synthase, the first enzyme of heme biosynthesis, and its link to XLSA in humans. EMBO J 2005;24:3166–77. [121] Pitta TP, Leadbetter ER, Godchaux III W. Increase of ornithine amino lipid content in a sulfonolipid-deficient mutant of Cytophaga johnsonae. J Bacteriol 1989;171:952–7. [122] White RH. Biosynthesis of the sulfonolipid 2-amino-3-hydroxy-15- methylhexadecane-1-sulfonic acid in the gliding bacterium Cytophaga johnsonae. J Bacteriol 1984;159:42–6. [123] Abbanat DR, Leadbetter ER, Godchaux III W, Escher A. Sulphonolipids are molecular determinants of gliding motility. Nature 1986;324:367–9. [124] Godchaux III W, Leadbetter ER. Sulfonolipids of gliding bacteria: structure of the N-acylaminosulfonates. J Biol Chem 1984;259:2982–90. [125] Corcelli A, Lattanzio VMT, Mascolo G, Babudri F, Oren A, Kates M. Novel sulfonolipid in the extremely halophilic bacterium Salinibacter ruber. Appl Environ Microbiol 2004;70:6678–85. [126] DeLong EF, Yayanos AA. Adaptation of the membrane lipids of a deep-sea bacterium to changes in hydrostatic pressure. Science 1985;228:1101–3. [127] Zhang YM, Rock CO. Membrane lipid homeostasis in bacteria. Nat Rev Microbiol 2008;6:222–33. [128] Minnikin DE, Abdolrahimzadeh H, Baddiley J. Variation of polar lipid composition of Bacillus subtilis Marburg with different growth conditions. FEBS Lett 1972;27:16–8. [129] Minnikin DE, Abdolrahimzadeh H, Baddiley J. Replacement of acidic phospholipids by acidic glycolipids in Pseudomonas diminuta. Nature (London) 1974;249:268–9. [130] Minnikin DE, Abdolrahimzadeh H. The replacement of phosphatidylethanolamine and acidic phospholipids by an ornithine-amide lipid and a minor phosphorus-free lipid in Pseudomonas fluorescens NCMB 129. FEBS Lett 1974;43:257–60. [131] Thiehle OW, Schwinn G. The free lipids of Brucella melitensis and Bordetella pertussis. Eur J Biochem 1973;34:333–44. [132] Nikaido H. Molecular basis of bacterial outer membrane permeability revisited. Microbiol Mol Biol Rev 2003;67:593–656. [133] Murata T, Tseng W, Guina T, Millar SI, Nikaido H. PhoPQ-mediated regulation produces a more robust permeability barrier in the outer membrana of Salmonella enterica serovar typhimurium. J Bacteriol 2007;189:7213–22. 60 O. Geiger et al. / Progress in Lipid Research 49 (2010) 46–60 69 8.1.2. II Artículo  Vences‐Guzmán, M. A., Guan, Z., Ormeño‐Orrillo, E., González‐Silva, N., López‐Lara,  I.  M., Martínez‐Romero, E., Geiger, O., and Sohlenkamp, C. (2011) Hydroxylated ornithine  lipids increase stress tolerance in Rhizobium tropici CIAT899. Mol. Microbiol. 79, 1496‐  1514.  Los  lípidos  de  ornitina  (OLs)  están  ampliamente  distribuidos  en  las  bacterias  Gram  negativas. Los OLs estándar consisten de un residuo de ornitina y dos ácidos grasos. Un  ácido graso está unido al grupo α‐amino de  la ornitina por enlace amida y el otro ácido  graso está esterificado al C3 del ácido graso amidificado. Los OLs pueden ser hidroxilados  en el ácido graso esterificado y esa modificación ha sido relacionada con un  incremento  en la tolerancia a estrés. Rhizobium tropici CIAT899 es resistente a estreses como pH bajo  y altas temperaturas y forma nódulos en  las raíces de plantas de  frijol, en  los cuales  fija  nitrógeno. R. tropici CIAT899 forma cuatro clases de OLs. Estudios de  la función de estos  OLs han  sido obstaculizados debido a carencia de conocimiento  sobre su biosíntesis. En  este trabajo se describe que  la biosíntesis de OLs se  incrementa bajo estrés ácido y que  OLs  estan  enriquecidos  en  la  membrana  externa.  Usando  un  escrutinio  de  expresión  funcional, se identificó la hidroxilasa OlsE de OLs, la cual en combinación con la hidroxilasa  OlsC  son  responsables de  la biosíntesis de  los OLs modificados en R.  tropici CIAT899. A  diferencia de otras hidroxilaciones descritas, la hidroxilación catalizada por OlsE ocurre en  el  residuo de ornitina. Se caracterizaron Mutantes deficientes en OlsE u OlsC y  la doble  mutante  OlsC/OlsE.  Mutantes  derivadas  de  R.  tropici  CIAT899  deficientes  en  la  hidroxilación  de  OLs  que  se  debe  a  OlsC  son  más  susceptibles  al  estrés  ácido  y  a  la  temperatura. Las tres mutantes que carecen de  las hidroxilasas de OLs se ven afectadas  durante la simbiosis.      70 Hydroxylated ornithine lipids increase stress tolerance in Rhizobium tropici CIAT899mmi_7535 1496..1514 Miguel Á. Vences-Guzmán,1 Ziqiang Guan,2 Ernesto Ormeño-Orrillo,1 Napoleón González-Silva,1 Isabel M. López-Lara,1 Esperanza Martínez-Romero,1 Otto Geiger1 and Christian Sohlenkamp1* 1Centro de Ciencias Genómicas, Universidad Nacional Autónoma de México, Av. Universidad s/n, Apdo. Postal 565-A, Cuernavaca, Morelos, CP62210, Mexico. 2Department of Biochemistry, Duke University Medical Center, Durham, NC 27710, USA. Summary Ornithine lipids (OLs) are widespread among Gram- negative bacteria. Their basic structure consists of a 3-hydroxy fatty acyl group attached in amide linkage to the a-amino group of ornithine and a second fatty acyl group ester-linked to the 3-hydroxy position of the first fatty acid. OLs can be hydroxylated within the secondary fatty acyl moiety and this modification has been related to increased stress tolerance. Rhizobium tropici, a nodule-forming a-proteobacterium known for its stress tolerance, forms four different OLs. Studies of the function of these OLs have been hampered due to lack of knowledge about their biosynthesis. Here we describe that OL biosynthesis increases under acid stress and that OLs are enriched in the outer membrane. Using a functional expression screen, the OL hydroxylase OlsE was identified, which in combination with the OL hydroxy- lase OlsC is responsible for the synthesis of modified OLs in R. tropici. Unlike described OL hydroxylations, the OlsE-catalysed hydroxylation occurs within the ornithine moiety. Mutants deficient in OlsE or OlsC and double mutants deficient in OlsC/OlsE were characterized. R. tropici mutants deficient in OlsC- mediated OL hydroxylation are more susceptible to acid and temperature stress. All three mutants lacking OL hydroxylases are affected during symbiosis. Introduction Membranes of the Gram-negative model organism Escherichia coli only contain three major phospholipids, that is phosphatidylethanolamine, phosphatidylglycerol and cardiolipin (Heath et al., 2002). Some other bacteria also form the membrane lipids phosphatidylinositol or phosphatidylcholine (Jackson et al., 2000; Sohlenkamp et al., 2003). In addition to phospholipids, many bacteria also present phosphorus-free membrane lipids such as ornithine lipids (OLs), diacylglyceryl-N, N, N-trimethylhomoserine (DGTS) or sulpholipids (SLs) in their membranes (López-Lara et al., 2003; Geiger et al., 2010). In some cases, like for example Rhodobacter sphaeroides or Sinorhizobium meliloti, the formation of these phosphorus-free membrane lipids is induced by phosphate-limiting growth conditions (Benning et al., 1995; Geiger et al., 1999). Some bacteria such as Bru- cella abortus (Comerci et al., 2006; Bukata et al., 2008) or Rhizobium tropici (Rojas-Jiménez et al., 2005; Sohlen- kamp et al., 2007) also form significant amounts of OLs during growth in standard laboratory media such as LB which contain phosphate in concentrations that are not growth-limiting. Ornithine lipids are widespread among Gram-negative bacteria and have also been reported in some Gram- positive bacteria, like Mycobacterium and Streptomyces species, but seem to be absent from Archaea and Eukarya (López-Lara et al., 2003; Geiger et al., 2010). OLs contain a 3-hydroxy fatty acyl group that is attached in amide linkage to the a-amino group of ornithine. A second fatty acyl group is ester-linked to the 3-hydroxy position of the first fatty acid. It has been reported that in some bacteria the ester-linked fatty acid is hydroxylated at the 2 or 3 position (Asselineau, 1991). The genes olsB and olsA encoding the two enzymes essential for OL biosynthesis from ornithine and acyl-ACPs have been first described in S. meliloti (Weissenmayer et al., 2002; Gao et al., 2004). Although OLs are probably found in both membranes of Gram-negative bacteria, they seem to be enriched in the outer membrane (OM) as was shown in the acid-resistant species Thiobacillus thiooxidans (Dees and Shively, 1982). Therefore, Dees and Shively specu- lated about a role of OLs in acid resistance (Dees and Shively, 1982). Accepted 29 December, 2010. *For correspondence. E-mail chsohlen@ccg.unam.mx; Tel. (+52) 777 3131697; Fax (+52) 777 3175581. Molecular Microbiology (2011) 79(6), 1496–1514  doi:10.1111/j.1365-2958.2011.07535.x First published online 23 January 2011 © 2011 Blackwell Publishing Ltd 71 Rhizobium tropici CIAT899 is highly tolerant to many environmental stresses such as acidity or high temperatures. It can grow on acidified media down to pH 4.0, and it is a good competitor for nodule occupancy in Phaseolus vulgaris (common bean) and other hosts under acidic conditions (Martínez-Romero et al., 1991). A gene responsible for the hydroxylation of OL has been isolated in R. tropici using a transposon mutagenesis approach looking for mutants affected in their capacity to grow at pH 4.5 (Vinuesa et al., 2003; Rojas-Jiménez et al., 2005). Rojas-Jiménez et al. described the presence of four dif- ferent species of OL in R. tropici membranes which were called S1, S2, P1 and P2. They showed that the putative hydroxylase OlsC is responsible for the formation of P1 and P2, presumably from OLs S1 and S2 functioning as substrates (Fig. 1), but they did not investigate on the position of the OlsC-dependent hydroxylation. No acid growth phenotype was observed for the olsC-deficient mutant, but constitutive expression of olsC was associ- ated with the inability of the strain to grow at pH 4.5. Upon inoculation of the olsC mutant onto bean plants only poorly developed nodules were observed (Ndv-) 21 days after inoculation of the plants (Rojas-Jiménez et al., 2005). In an earlier study, Taylor et al. (1998) had observed an increased formation of hydroxylated OLs at an elevated temperature in Burkholderia cepacia. These two previous results indicated a role of modified OLs in stress tolerance, and prompted us to investigate the syn- thesis of modified OLs and their role in stress tolerance in R. tropici in more detail. In this study we describe the isolation of the OL hydroxylase OlsE and the construction of R. tropici mutants deficient in the hydroxylation of OLs. We show that OlsC is introducing a hydroxyl group in the 2 position of the secondary fatty acid of OLs and that OlsE introduces a hydroxylation in the ornithine moiety of OLs. The characterization of these mutants shows that hydroxylated OLs are important for adaptation to stress conditions in R. tropici. Fig. 1. Biosynthesis of ornithine lipids in Rhizobium tropici CIAT899. The genes coding for OlsB and OlsA have been first identified in Sinorhizobium meliloti, whereas the gene encoding the OL hydroxylase OlsC has been described first in R. tropici. Here we describe that the hydroxylation introduced by OlsC is in the 2 position of the secondary fatty acid. We also describe the identification of the gene encoding the OL hydroxylase OlsE introducing a hydroxyl group in the ornithine moiety of OL. Lyso-ornithine lipid (LOL), ornithine lipid (OL). Hydroxylated ornithine lipids in Rhizobium tropici 1497 © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 72 Results Stress conditions alter the amount of modified OLs in R. tropici indicating a role of OLs in stress adaptation Rhizobium tropici CIAT899 is a nodule-forming rhizobium well known for its ability to resist stress conditions such as acidic pH or high temperatures (Martínez-Romero et al., 1991). In an earlier study Rojas-Jiménez et al. (2005) had observed that R. tropici forms four different OLs. In addi- tion to the unmodified OL which was named S1 (for sub- strate 1) three additional modified OLs probably derived from S1 are present. Taylor et al. (1998) had observed an increase in the relative amounts of hydroxylated OL when B. cepacia was grown at increased temperatures. To find out if the modification of OL also occurs as a stress response in R. tropici and if these modifications might have a role in stress adaptation, R. tropici CIAT899 was grown at 30°C, 37°C and 42°C and its lipid composition was analysed (Fig. 2A–C, Table 1). At the standard growth temperature of 30°C, all four OLs can be detected, with P1 being the most abundant OL. An increase in growth temperature to 37°C causes a decrease in the OLs S2 and P2 and a simultaneous increase in S1. When grown at 42°C the amounts of S1 and P1 decrease slightly. The OLs S2 and P2 cannot be detected in cells grown at 42°C. An unknown lipid which migrates similarly as the sulpholipid sulphoquinovosyl diacylglycerol is apparently formed at 42°C but not at lower growth temperatures. The decrease in OLs is accompanied by changes in the phospholipid composition: phosphatidyle- thanolamine (PE) decreases whereas phosphatidylcho- line (PC), phosphatidylglycerol (PG) and cardiolipin (CL) increase. R. tropici CIAT899 was also grown in complex TY medium adjusted to different pH values (compare Fig. 2A, D and E, Table 2). In R. tropici cells grown at pH 4.5 the OLs S1 and S2 are not detectable, whereas P2 is Fig. 2. Separation of [14C]acetate-labelled lipids from Rhizobium tropici CIAT899 grown in complex TY medium at 30°C (A), at 37°C (B), at 42°C (C), at 30°C at pH 4.5 (D) or at 30°C at pH 4.0 (E) by two-dimensional thin-layer chromatography. The phospholipids phosphatidylethanolamine (PE), phosphatidylcholine (PC), phosphatidylglycerol (PG), cardiolipin (CL), monomethyl PE (MMPE), dimethyl PE (DMPE) and the ornithine lipids (OLs) S1, S2, P1 and P2 are indicated. U, unknown lipid. 1498 M. Á. Vences-Guzmán et al.  © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 73 Ta b le 1. M em br an e lip id co m po si tio n of R h iz o b iu m tr o p ic iw ild -t yp e C IA T 89 9, o ls E -d efi ci en tm ut an tM A V 04 ,o ls C -d efi ci en tm ut an t8 99 -o ls C D 1 an d o ls C /o ls E -d efi ci en td ou bl e m ut an tM A V 05 af te r gr ow th on co m pl ex T Y m ed iu m at 30 °C , 37 °C or 42 °C . Li pi d C om po si tio n (% of to ta l14 C ) 30 °C 37 °C 42 °C C IA T 89 9 M A V 04 89 9- o ls C D 1 M A V 05 C IA T 89 9 M A V 04 89 9- o ls C D 1 M A V 05 C IA T 89 9 M A V 04 89 9- o ls C D 1 M A V 05 P C 22 .6  0. 6 24 .8  2. 9 20 .7  0. 7 24 .8  0. 1 28 .9  4. 0 27 .2  4. 8 23 .5  4. 3 24 .6  1. 2 26 .8  1. 2 28 .5  1. 6 27 .0  2. 0 27 .5  2. 1 P E 25 .0  0. 1 24 .6  1. 4 24 .9  2. 1 25 .6  0. 6 23 .5  5. 0 23 .2  3. 7 19 .1  1. 0 20 .4  4. 8 16 .2  1. 3 17 .9  1. 9 11 .9  1. 3 11 .8  1. 1 D M P E 1. 1  0. 1 1. 7  0. 2 1. 6  0. 6 1. 6  0. 4 1. 4  0. 6 1. 1  0. 2 1. 4  0. 4 1. 3  0. 2 0. 4  0. 0 0. 4  0. 1 0. 6  0. 2 0. 7  0. 1 P G 15 .8  0. 8 13 .9  2. 8 17 .4  1. 5 16 .0  0. 6 12 .5  0. 9 11 .8  1. 2 18 .8  1. 7 16 .9  1. 6 23 .8  0. 3 22 .1  0. 8 33 .9  1. 0 29 .2  1. 1 C L 4. 0  0. 1 4. 8  0. 1 3. 0  0. 1 5. 6  0. 3 5. 3  1. 3 4. 8  1. 0 8. 0  2. 0 6. 7  2. 4 7. 5  0. 5 8. 2  0. 3 11 .1  2. 5 10 .7  1. 3 S 1 2. 9  0. 2 6. 8  1. 0 26 .5  1. 7 26 .4  1. 8 7. 3  1. 0 9. 0  1. 4 27 .7  0. 9 30 .1  5. 6 5. 6  0. 2 4. 7  0. 9 10 .8  0. 8 16 .0  1. 5 S 2 3. 5  0. 5 n. d. 5. 9  0. 5 n. d. 1. 6  0. 3 n. d. 1. 5  0. 2 n. d. n. d. n. d. n. d. n. d. P 1 20 .6  0. 3 23 .4  1. 8 n. d. n. d. 18 .2  2. 8 22 .9  0. 1 n. d. n. d. 15 .2  0. 5 15 .2  0. 7 n. d. n. d. P 2 4. 5  0. 3 n. d. n. d. n. d. 1. 3  0. 4 n. d. n. d. n. d. n. d. n. d. n. d. n. d. U n. d. n. d. n. d. n. d. n. d. n. d. n. d. n. d. 4. 5  0. 5 3. 0  0. 7 4. 7  0. 7 4. 1  1. 0 T he va lu es sh ow n ar e m ea n va lu es  st an da rd de vi at io n de riv ed fr om at le as t th re e in de pe nd en t ex pe rim en ts . P C : ph os ph at id yl ch ol in e; P E : ph os ph at id yl et ha no la m in e; M M P E : m on om et hy l ph os ph at id yl et ha no la m in e; D M P E : di m et hy l ph os ph at id yl et ha no la m in e; P G : ph os ph at id yl gl yc er ol ; C L: ca rd io lip in ; S 1: su bs tr at e 1, un m od ifi ed or ni th in e lip id ; S 2, P 1, P 2: hy dr ox yl at ed or ni th in e lip id s; U : un id en tifi ed lip id ; n. d. : no t de te ct ed . Ta b le 2. M em br an e lip id co m po si tio n of R h iz o b iu m tr o p ic iw ild -t yp e C IA T 89 9, o ls E -d efi ci en tm ut an tM A V 04 ,o ls C -d efi ci en tm ut an t8 99 -o ls C D 1 an d o ls C /o ls E -d efi ci en td ou bl e m ut an tM A V 05 af te r gr ow th on co m pl ex T Y m ed iu m ad ju st ed to pH 7. 0, pH 4. 5 or pH 4. 0. Li pi d C om po si tio n (% of to ta l14 C ) pH 7. 0 pH 4. 5 pH 4. 0 C IA T 89 9 M A V 04 89 9- o ls C D 1 M A V 05 C IA T 89 9 M A V 04 89 9- o ls C D 1 M A V 05 C IA T 89 9 M A V 04 89 9- o ls C D 1 M A V 05 P C 27 .6  0. 7 25 .0  0. 2 21 .3  0. 3 24 .6  0. 4 27 .7  0. 2 28 .1  1. 7 23 .3  2. 2 23 .8  3. 8 26 .2  0. 3 26 .6  0. 4 19 .5  0. 5 18 .1  2. 0 P E 26 .7  0. 3 26 .5  0. 1 21 .7  0. 1 24 .7  0. 5 27 .4  0. 3 27 .3  2. 1 12 .8  1. 5 10 .9  0. 4 22 .3  0. 4 25 .6  0. 3 14 .9  0. 4 12 .3  2. 0 D M P E 0. 4  0. 1 0. 6  0. 1 0. 6  0. 1 0. 9  0. 0 0. 8  0. 1 0. 5  0. 1 1. 0  0. 2 1. 0  0. 1 0. 8  0. 2 0. 2  0. 0 0. 2  0. 1 0. 6  0. 2 P G 12 .2  0. 4 10 .0  0. 1 13 .1  0. 2 11 .5  0. 1 8. 2  0. 2 4. 7  0. 4 10 .1  0. 9 9. 2  0. 9 6. 8  0. 1 7. 4  1. 3 12 .4  1. 2 12 .9  1. 1 C L 4. 4  0. 2 5. 7  0. 4 5. 5  0. 3 3. 7  0. 3 6. 9  0. 3 2. 4  0. 2 4. 2  0. 7 4. 3  0. 1 3. 7  0. 5 4. 2  1. 0 5. 7  0. 4 4. 7  0. 7 S 1 2. 9  1. 7 8. 9  0. 2 32 .5  0. 2 34 .6  0. 2 n. d. 2. 0  0. 5 40 .6  2. 4 46 .0  2. 9 n. d. 1. 6  0. 3 40 .7  1. 1 44 .6  3. 6 S 2 3. 3  0. 3 n. d. 5. 3  0. 5 n. d. n. d. n. d. 2. 5  0. 1 n. d. n. d. n. d. n. d. n. d. P 1 18 .0  0. 4 23 .3  0. 1 n. d. n. d. 19 .4  0. 3 32 .0  1. 0 n. d. n. d. 30 .2  0. 1 31 .9  2. 5 n. d. n. d. P 2 4. 5  0. 4 n. d. n. d. n. d. 8. 3  0. 2 n. d. n. d. n. d. 8. 1  0. 0 n. d. n. d. n. d. U n. d. n. d. n. d. n. d. 1. 3  0. 1 3. 0  0. 3 5. 5  1. 0 4. 8  0. 5 1. 9  0. 1 2. 5  0. 4 6. 6  0. 6 6. 8  0. 5 T he va lu es sh ow n ar e m ea n va lu es  st an da rd de vi at io n de riv ed fr om at le as t th re e in de pe nd en t ex pe rim en ts . F or ab br ev ia tio ns se e Ta bl e 1. Hydroxylated ornithine lipids in Rhizobium tropici 1499 © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 74 increased and no changes are detected for P1. When grown at pH 4.0 again OLs S1 and S2 cannot be detected, but P1 increases drastically and becomes the major mem- brane lipid (Fig. 2E). OLs are enriched in the OM of R. tropici CIAT899 Dees and Shively (1982) had shown that in the acid- resistant species T. thiooxidans OL is present mainly in the OM and they had therefore speculated that it might play a role in conferring acid resistance to these bacteria. If such a hypothesis were true one would expect an accumulation of OLs also in the OM of the acid-tolerant bacterium R. tropici. Inner membrane (IM) and OM from R. tropici were separated and the lipids of both membranes were extracted and separated using two two-dimensional TLC (Fig. 3). The protein content of the fractions was estimated using absorption measurements at 280 nm. The protein- enriched fractions formed two peaks corresponding to the IM and OM (Fig. 3A). KDO (2-keto-3-deoxyoctanoate) Fig. 3. Localization of OLs in membranes of wild-type Rhizobium tropici CIAT899. A and B. Results of a sucrose density gradient centrifugation of cell membranes of R. tropici CIAT899. (A) A280 readings of the gradient fractions. (B) 2-Keto-3-oxyoctonate content (closed triangles) and NADH oxidase (closed squares) activity of the fractions. C and D. Separation of membrane lipids extracted from the inner (C) and outer membrane (D). Fractions corresponding to the inner and outer membranes were pooled, lipids were extracted with 1-butanol and subsequently analysed using two-dimensional TLC. Lipids were visualized by spraying with ceric sulphate in sulphuric acid. The phospholipids phosphatidylethanolamine (PE), phosphatidylcholine (PC), phosphatidylglycerol (PG), cardiolipin (CL), monomethyl PE (MMPE), dimethyl PE (DMPE) and the ornithine lipids (OLs) S1, S2, P1 and P2 are indicated. A quantification of the lipids is shown in Table 3. 1500 M. Á. Vences-Guzmán et al.  © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 75 content and NADH oxidase activity that were used as markers for the OM and IM, respectively, indicated that the IM was contaminated to some extent by the OM, but that the OM was almost free of contamination by the IM (Fig. 3B). The TLC analysis of the Bligh-Dyer extracts showed that phospholipids are the major membrane lipids of the IM but are present in much smaller relative amounts in the OM (Fig. 3C and D, Table 3). A quantifica- tion of the lipids showed that phospholipids form more than 70% of the membrane lipids of the IM but only about 40% of the membrane lipids of the OM, excluding lipopolysac- charide (LPS). OLs form less than 30% of the membrane lipids of the IM but about 60% of the membrane lipids of the OM (again excluding LPS). Taking the contamination of the IM fractions with OM material into account the result overestimates the real concentration of OLs in the IM. Assuming that the outer leaflet of the OM is composed mainly of the lipid A moiety of LPS, this result indicates that the major proportion of the inner leaflet of the OM is composed of OLs. Expression cloning of the OL-modifying enzyme OlsE from R. tropici The experiments described earlier indicated a possible role for the different OLs in the R. tropici stress response. In S. meliloti only one type of OL is present. In contrast, four different types of OLs called S1, S2, P1 and P2 are present in R. tropici CIAT899 (Fig. 1). The gene olsC encoding the enzyme OlsC responsible for the synthesis of OLs P1 and P2 from the substrates S1 and S2 has been described earlier (Rojas-Jiménez et al., 2005). It was not known, however, which gene encodes the hypothetical enzyme OlsE responsible for the synthesis of S2 and possibly also for the synthesis of P2 (Fig. 1). We suspected that S1, corresponding to the OL present in S. meliloti, was a substrate for the OlsE-catalysed reaction. The S. meliloti strain CS111.pNG25 lacking the ninhydrin-positive lipid PE and producing increased amounts of the OL S1 was con- structed and transconjugants of CS111.pNG25 harbour- ing cosmids containing R. tropici CIAT899 genomic DNA were assayed for the presence of a second ninhydrin- positive lipid in addition to S1. In the transconjugant referred to as CS111.pNG25.pCos94, two ninhydrin- positive lipids with the expected Rf values for S1 and S2 were detected. A restriction analysis of pCos94 showed that it contains about 18–20 kb of inserted DNA. Restric- tion fragments of the pCos94 insert were subcloned into a broad-host-range vector and again conjugated into CS111.pNG25. The resulting transconjugants were analysed as described above for the cosmid bank (data not shown). A plasmid conferring the formation of the OL S2 was identified and its insert was sequenced. In addi- tion to three predicted complete ORFs it contained two incomplete ORFs (GenBank Accession No. HM010770). BLAST searches using the NCBI database with the amino acid sequences of the three complete ORFs as query were made (Altschul et al., 1997). The first ORF was annotated as a putative acetyltransferase, the second ORF as a putative aminoglycoside N(6′) acetyltrans- ferase and the third ORF as a putative hydroxylase. The three candidate ORFs were cloned into a broad-host- range plasmid and the resulting plasmids were conju- gated into CS111.pNG25. Labelling of the lipids of the three transconjugants with [14C]acetate showed that ORF3 codes for the putative hydroxylase OlsE which is responsible for the formation of S2 (Fig. 4). Table 3. Membrane lipid composition of the inner and outer mem- brane of R. tropici CIAT899. Lipid Inner membrane Outer membrane PC 23.9 9.1 PE 8.2 6.6 MMPE 5.4 5.9 DMPE 4.8 5.0 PG 17.2 7.7 CL 12.2 6.7 S1 6.2 11.4 S2 5.0 11.0 P1 11.1 25.3 P2 6.0 11.3 The data were obtained from the TLC plates shown in Fig. 3C and D using the program ImageQuant. Numbers present per cent of total lipids present in the TLC. For abbreviations see Table 1. Fig. 4. Expression cloning of olsE from R. tropici. Lipids of Sinorhizobium meliloti CS111.pNG25 containing different plasmids or cosmids were radiolabelled with [14C]acetate and separated by one-dimensional TLC. The following strains were analysed: CS111.pNG25.pCos94 (cosmid, lane 1), CS111.pNG25.pERMAV04 (ORF1 to 3/3.5 kb insert, lane 2), CS111.pNG25.pERMAV13 (ORF3, lane 3), CS111.pNG25.pERMAV12 (ORF2, lane 4), CS111.pNG25.pERMAV11 (ORF1, lane 5) and CS111.pNG25.pERMAV06 (negative control, lane 6). The phospholipids phosphatidylcholine (PC), phosphatidylglycerol (PG), cardiolipin (CL) and the ornithine lipids (OLs) S1 and S2 are indicated. Hydroxylated ornithine lipids in Rhizobium tropici 1501 © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 76 OlsE belongs to the fatty acyl hydroxylase superfamily and is responsible for the hydroxylation of OL within the ornithine moiety The gene olsE encodes a very hydrophobic protein of 330 amino acids predicted to form between four and six trans- membrane helices. An analysis of the amino acid sequence shows that OlsE belongs to the fatty acyl hydroxylase superfamily (cl01132) which is characterized by the presence of two copies of the HXHH motif. This superfamily includes fatty acid and carotene hydroxy- lases, sterol desaturases (Mitchell and Martin, 1997), C-5 sterol desaturase (Arthington et al., 1991) and C-4 sterol methyl oxidase (Bard et al., 1996; Kennedy et al., 2000). A similar motif (HX3–4H, HX2–3HH, HX2–3H) can be found in membrane-bound fatty acid desaturases such as OLE1 from Saccharomyces cerevisiae and is also present in bacterial alkane hydroxylase (Kok et al., 1989) and xylene monooxygenase (Suzuki et al., 1991). In these proteins the conserved histidine residues act to co-ordinate an oxo-bridged diiron cluster (Fe–O–Fe) that functions as part of the reaction centre (Fox et al., 1993; Shanklin et al., 1994). The annotation of OlsE as a fatty acyl hydroxylase indicates that OlsE introduces a hydroxyl group into OL at an unknown position. To localize the hydroxyl group on the OL S2, lipids were extracted according to Bligh and Dyer (1959) from a 1 l culture of the olsC-deficient R. tropici mutant 899-olsCD1. OLs S1 and S2 were purified from the total lipid extract and analysed by normal-phase LC-coupled electrospray ionization (ESI) mass spectrom- etry (MS) in the negative ion mode. Prior to fragmentation ions with m/z 691 and 707 corresponding to OLs S1 and S2 were detected. The molecular ion was shifted in case of S2 to an m/z 16 amu higher in comparison with S1 indicating the presence of an additional oxygen suggest- ing the presence of an additional hydroxyl group. Com- paring the fragmentation patterns of S1 and S2 it was observed that the modification present in S2 is located within the ornithine moiety and not in the fatty acyl chains (Fig. 5). When assaying the two-dimensional TLC plates with R. tropici lipids with ninhydrin it was noticed that S2 and P2 react with delay in comparison with S1 and P1, and that the developed colour is different. While S1 and P1 upon reaction with ninhydrin develop a red to purple colour, the reaction of S2 and P2 causes the formation of an orange colour. OlsC introduces a hydroxyl group at the 2 position of the secondary fatty acid of OL OlsC is a homologue of the hydroxylase LpxO from Sal- monella typhimurium that is responsible for the addition of a 2-hydroxy group to the myristate residue present at the 3′ position of lipid A. Rojas-Jiménez et al. (2005) had discovered the gene olsC and had shown that OlsC is a putative hydroxylase responsible for the formation of the OLs P1 and P2 from the OLs S1 and S2 in R. tropici (Fig. 1). However, it was not known in what part of the OL structure the OlsC-dependent hydroxylation occurs. To localize the hydroxyl group on the OL P1, OLs S1 and P1 were purified from the total lipid extracts and analysed by normal-phase LC-coupled ESI-MS in the negative ion mode. Prior to fragmentation ions with m/z 691 and 707 corresponding to OLs S1 and P1 were detected. The molecular ion of P1 was shifted to an m/z 16 amu higher in comparison with S1 indicating the presence of an addi- tional oxygen suggesting the presence of an additional hydroxyl group. Comparing the fragmentation patterns of S1 and P1 it was observed that the modification present in P1 is located within the secondary fatty acyl chain (data not shown) which in case of S1 is mainly lactobacillic acid and in case of P1 hydroxy lactobacillic acid. In order to determine the position of the OlsC-dependent hydroxyla- tion in P1 its fatty acids were transmethylated before the hydroxyl groups were derivatized to trimethylsilyl (TMS) ethers similar to the procedures described by Gibbons et al. (2008). Alpha- and beta-hydroxy fatty acid standards of 16 and 18 carbons were processed in parallel with the samples (Fig. S1A–D). GC/MS analysis of the derivatized fatty acids shows the presence of three peaks present in the samples derived from P1 that are not present in the samples derived from S1 (Fig. S1E and F). Their fragmentation pattern indicates that the OlsC-dependent hydroxylation occurs in the 2 position (Fig. S1G). Lipid composition analysis of olsE and olsE/olsC mutants To study the role of OLs in R. tropici in more detail, mutants deficient in olsE and double mutants deficient in olsC and olsE were constructed. Their lipid compositions were compared with the wild-type strain CIAT899 and the OlsC-deficient mutant 899-olsCD1 (Fig. 6, Table 1). As expected the olsE-deficient mutant MAV04 lacked the OLs S2 and P2, the olsC-deficient mutant 899-olsCD1 lacked P1 and P2 and in the double mutant MAV05 (DolsCDolsE) no S2, P1 or P2 were detectable. Appar- ently, the amount of OLs, being the sum of S1, S2, P1 and P2, is more or less stable between 20% and 35% when R. tropici is grown in complex TY medium at 30°C. No sig- nificant differences in the relative amounts of the phos- pholipids PE, PC, PG and CL were observed between the different strains. To show that the observed phenotypes were caused by the absence of the deleted genes, mutants MAV04 (DolsE) and MAV05 (DolsCDolsE) were also complemented. When olsE was present in trans in MAV04 again formation of S2 and P2 was detected and 1502 M. Á. Vences-Guzmán et al.  © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 77 Fig. 5. Collision-induced dissociation mass spectra of ornithine lipids S1 and S2 detected in lipid extract of R. tropici mutant 899-olsCD1. Negative ion collision induced dissociation mass spectra of [M-H]- ions at m/z 671 (A) obtained from OL S1 and m/z 707 (B) obtained from OL S2. The structures of major fragment ions are indicated. The position of the hydroxyl group introduced in the ornithine moiety is assigned tentatively. Complete structures of the OLs are shown in Fig. 1. Hydroxylated ornithine lipids in Rhizobium tropici 1503 © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 78 when mutant MAV05 was complemented with olsE the OLs S2 and P2 could be detected, whereas S1 and P1 did not accumulate (data not shown). Constitutive expression of olsC and olsE together in MAV05 caused the accumu- lation of P2 while only trace amounts of the other OLs were observed (data not shown). Such an over- complementation leading to the accumulation of the reac- tion product(s) while almost completely consuming the substrate(s) had also been observed earlier for the complementation of the olsC-deficient mutant 899-olsCD1 (Rojas-Jiménez et al., 2005). Growth and characterization of the lipid composition of mutants 899-olsCD1, MAV04 and MAV05: OlsC is important in conferring stress resistance An earlier study in B. cepacia (Taylor et al., 1998) had shown an increase in formation of hydroxylated OL when the bacteria were grown at higher temperature. In this study we have shown that the relative amounts of the different OLs shift in response to a change in growth temperature or pH of the medium. Rojas-Jiménez et al. (2005) had observed that a R. tropici strain constitutively expressing olsC was not able to grow at pH 4.5 any more. Therefore, it was expected that the R. tropici mutants deficient in OL hydroxylation would show a phenotype under conditions of acid or temperature stress. The wild- type R. tropici CIAT899 and the three mutants 899- olsCD1, MAV04 and MAV05 were cultivated in complex TY medium at pH 4.0, 4.5 and 7.0. At pH 7.0 all four strains divide at a similar rate (Fig. 7A). At pH 4.5 the wild-type CIAT899 and the mutant MAV04 (DolsE) grow at a similar rate compared with pH 7.0 whereas the other two mutants seem to present a longer generation time (Fig. 7B). At pH 4.0 the wild-type CIAT899 and the mutant MAV04 (DolsE) grow significantly slower than at pH 4.5 but still both cultures reach a final optical density larger than 1.0, whereas the mutants 899-olsCD1 and MAV05 (DolsCDolsE) at most undergo one single division (Fig. 7C). To determine if the observed differences are related to changes in lipid composition, wild-type and mutant cells were grown and labelled in the correspond- ing media and analysed by TLC in two dimensions (Table 2). At pH 7.0 all four strains show similar concen- trations of phospholipids and the distinct patterns of the different OLs typical for each mutant described above. At pH 4.5 both OlsC-deficient mutants (899-olsCD1 and MAV05) show a drastic reduction in PE content and a strong increase in S1 to up to more than 40%. At pH 4.0 again, both OlsC-deficient mutants show a very similar lipid composition with S1 being the major membrane lipid and PE being drastically reduced. The wild-type CIAT899 apparently forms more P1 under these conditions. It seems that low-pH conditions cause the accumulation of OLs in all strains: in the wild-type and the mutant MAV04 (DolsE) the major lipid accumulated is P1, whereas in the mutants 899-olsCD1 and MAV05 (DolsCDolsE) the major lipid is S1. When the wild-type CIAT899 and the three mutants deficient in OL hydroxylation were cultivated in TY medium at 30°C, no differences in generation time can be observed between them (Fig. 7D). At 37°C, both strains lacking olsC (899-olsCD1 and MAV05) seem to grow slightly slower than the other two strains (Fig. 7E). At 42°C Fig. 6. Analysis of membrane lipid composition of R. tropici wild-type CIAT899 (A), olsC-deficient mutant 899-olsCD1 (B), olsE-deficient mutant MAV04 (C) and olsC/olsE-deficient double mutant MAV05 (D). Lipids were labelled with [14C]acetate during growth in complex TY medium at 30°C and separated using two-dimensional TLC. The phospholipids phosphatidylethanolamine (PE), phosphatidylcholine (PC), phosphatidylglycerol (PG), cardiolipin (CL), monomethyl PE (MMPE), dimethyl PE (DMPE) and the ornithine lipids (OLs) S1, S2, P1 and P2 are indicated. 1504 M. Á. Vences-Guzmán et al.  © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 79 Fig. 7. Growth of R. tropici mutants lacking olsC is affected under stress conditions. R. tropici wild-type CIAT899 and mutants were grown in complex TY medium adjusted to pH 7.0 (A), pH 4.5 (B) or pH 4.0 (C) at 30°C or in complex TY medium at 30°C (D), 37°C (E) or 42°C (F). The result of a typical experiment is shown. CIAT899 – closed circles, MAV04 – open circles, MAV05 – open triangles, 899-olsCD1 – closed triangles. Hydroxylated ornithine lipids in Rhizobium tropici 1505 © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 80 wild-type CIAT899 and mutant MAV04 (DolsE) grow slower than at the lower temperature and reach a final OD620 of only 0.65–0.68 (Fig. 7F). The mutants 899- olsCD1 and MAV05 (DolsCDolsE) divide distinctly slower at 42°C than the two former strains and reach a final OD620 of only 0.3. The lipid composition of the three mutants deficient in OL hydroxylation and the wild-type CIAT899 was also analysed at the different temperatures (Table 1). For each of the four strains the lipid compositions are very similar at 30°C and 37°C. At 42°C the amount of PG is increased by about 10–15% and also CL seems to be a bit more abun- dant at the higher temperature. The total of the four OLs is decreasing in all four strains. Whereas at 30°C and 37°C the sum of S1, S2, P1 and P2 is about 30%, at 42°C the strains contain only between 10% and 20% OLs. R. tropici mutants deficient in OlsC cause an increase in nodule number that is reverted by the deletion of olsE The R. tropici mutant deficient in OlsC (899-olsCD1) formed nodules on bean plants that were poorly devel- oped 21 days after inoculation with the bacteria, lacked lenticels and presented a twofold reduction in nitrogen fixation (Rojas-Jiménez et al., 2005). These results sug- gested that the R. tropici mutants MAV04 deficient in olsE and double mutants MAV05 deficient in olsC and olsE might also show nodulation phenotypes. Nodulation assays were performed in an agar-based medium in order to be able to observe the kinetics of nodule formation over time. While wild-type R. tropici CIAT899 and MAV04 (DolsE) produced reproducibly between 80 and 100 nodules per plant, the mutant 899-olsCD1 caused the formation of more than 160 nodules per plant. When the olsC-deficient mutant was complemented with the olsC gene, it again formed nodules in numbers similar to the wild-type (Fig. 8A). Surprisingly, the double mutant MAV05 (DolsCDolsE) formed a similar number of nodules as the wild-type. The nodules were sectioned and while the wild-type caused almost exclusively the formation of nodules that were red inside indicating the formation of leghaemoglobin, while the mutants caused formation of many small nodules that were whitish on the inside indicating the absence of leghaemoglobin (Fig. S2). Roots from plants infected with the olsC-deficient mutant 899-olsCD1 presented few red nodules and many whitish nodules and roots from plants infected with the olsE- deficient mutant MAV04 presented even less red and more white nodules. On roots infected with the olsC/olsE- deficient double mutant MAV05 almost no red nodules were formed (Fig. S2). Nitrogen fixation per hour and nodule fresh weight was affected in all three mutants in comparison with the wild-type (Fig. 8B). These results indicate that the absence of hydroxylated OLs strongly interferes with the development of functional nodules during R. tropici–bean symbiosis. Discussion Although OLs are widespread in eubacteria (López-Lara et al., 2003; Geiger et al., 2010) the genes olsB and olsA responsible for OL biosynthesis were only recently described in S. meliloti (Weissenmayer et al., 2002; Gao Fig. 8. Symbiotic phenotypes of R. tropici wild-type CIAT899 and strains deficient in OL modification on bean plants. A. Nodulation assay. Nodules were counted every second or third day. Plants were harvested 21 days post inoculation; nodules were assayed for nitrogen fixation activity. The experiment was repeated three times with five plants for each strain. The result of a typical experiment is shown. CIAT899 (closed circles), 899-olsCD1 (open circles), MAV04 (closed squares), MAV05 (open squares), 899-olsCD1.pERMAV05 (closed triangles), 899-olsCD1.pERMAV15 (open triangles). Uninoculated plants did not develop nodules. B. Mean acetylene reduction of nodulated bean roots inoculated with wild-type R. tropici CIAT899 and mutants 899-olsCD1, MAV04 and MAV05. Values are the mean  SD of three repetitions. 1506 M. Á. Vences-Guzmán et al.  © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 81 et al., 2004). In addition to the unmodified OL consisting of a 3-hydroxy fatty acid linked in an amide bond to the a-amino group of ornithine and a second fatty acid bound in an ester linkage to the first, several hydroxylated forms of OL have been described in organisms diverse as B. cepacia (Taylor et al., 1998), R. tropici (Rojas-Jiménez et al., 2005), Flavobacterium (Kawai et al., 1988), Thioba- cillus (Knoche and Shively, 1972), Streptomyces (Asse- lineau, 1991) and some Ralstonia (Galbraith et al., 1999) species. The fact that the genes coding for the enzymes responsible for the hydroxylation of OLs have not been identified except for the case of R. tropici where the gene olsC was described (Rojas-Jiménez et al., 2005) has made it difficult to study the function of these hydroxylated forms of OL. Apparently OL and especially their hydroxylated forms play a role in stress response as has been observed by Rojas-Jiménez et al. (2005) and Taylor et al. (1998). R. tropici mutants deficient in the formation of the hydroxy- lated OL P1 (899-olsCD1 and MAV05) are affected in growth at low pH and at high temperature in comparison with the wild-type. It has to be mentioned that in an earlier study the mutant 899-olsCD1 grew as well as the wild- type (Rojas-Jiménez et al., 2005). The explanation for this difference is unknown, but possibly slight differences in the pH of the medium cause drastic differences in the growth behaviour of the mutant. At pH 4.0 and 4.5 a drastic increase in the formation of OLs was observed when compared with growth at neutral pH. In the wild-type CIAT899 and in the mutant MAV04 (DolsE) especially P1 is increased, whereas in the olsC-deficient mutants unable to form P1 the substrate S1 is accumulating. This probably means that under acid growth conditions OL biosynthesis via OlsB and OlsA is induced. It is less clear what happens at the elevated growth temperature. Although the concentration of OLs is decreased during growth at 42°C in comparison with 30°C, again the pres- ence of P1 seems to be of importance as olsC-deficient mutants show a growth phenotype under this condition. The elevated temperature also seems to interfere with OlsE activity as S2 and P2 cannot be detected. Dees and Shively (1982) made the observation that in the extreme acid-tolerant bacterium Thiobacillus oxidans OLs are accumulated in the OM and therefore speculated about a role for OL in acid resistance in this organism (Dees and Shively, 1982). From the growth phenotype of the mutants unable to form P1 it is apparent that the hydroxylation at the 2 position of the secondary fatty acid is of importance under acid growth conditions. Our local- ization study confirms that although OLs seem to be present in both membranes, they show a higher relative abundance in the OM. Both studies therefore agree that OLs play a role in acid resistance, but it is not clear by which mechanism this effect of OLs is exerted. The hydroxyl group introduced by OlsC in the 2 position of the secondary fatty acid may increase hydrogen bonding between neighbouring OL molecules similarly as has been suggested for LpxO-hydroxylated lipid A in Salmo- nella and hydroxylated sphingolipids (Nikaido, 2003; Murata et al., 2007). These additional hydrogen bonds should result in bilayer stabilization and a decrease in membrane permeability which could explain the decrease in acid and temperature resistance of OlsC-deficient mutants. In this study we identified the OL hydroxylase OlsE using a functional expression screening. OlsE belongs to the fatty acyl hydroxylase superfamily, unlike the other OL hydroxylase OlsC from R. tropici which belongs to the aspartyl-/asparaginyl b-hydroxylase protein family to which also the lipid A-myristate b-hydroxylase LpxO from S. typhimurium belongs (Gibbons et al., 2000; 2008). The closest homologues to OlsC from R. tropici are present in the a-proteobacteria Agrobacterium radiobacter, Agro- bacterium vitis, Ochrobactrum anthropi, Brucella species, and in several cyanobacteria. Unlike other hydroxylations described in OL, the hydroxylation introduced by OlsE seems to be unique because it occurs in the ornithine moiety, but not in the fatty acid moieties as has been described for example in T. thiooxidans, B. cepacia or R. tropici (this study). Unrelated ornithine hydroxylases like for example PvdA from Pseudomonas aeruginosa have been described and studied in some detail (Visca et al., 1994; Meneely et al., 2009). PvdA is involved in pyoverdin biosynthesis and introduces a hydroxyl group in the d-amino group of ornithine but is unrelated on sequence level to OlsE. It is not clear yet in which position the OlsE-catalysed hydroxylation occurs, but apparently the newly introduced hydroxyl group is close enough to the d-amino group to change its reactivity with ninhydrin. As other members of the fatty acyl hydroxylase superfamily introduce hydroxyl groups at carbon atoms but not at nitrogen atoms OlsE possibly introduces a hydroxyl group at the d-carbon. It is not clear how the OlsE-dependent hydroxylation might affect membrane characteristics. Possibly the OlsE-dependent hydroxyla- tion enables the OLs S2 and P2 to form a lactone ring within the ornithine headgroup, the presence of which should change its biophysical properties drastically. The closest OlsE homologues are present in some a-proteobacteria and more distant homologues are present in several actinobacteria, a few g-proteobacteria and a few other a-proteobacteria. Possibly several of the closer homologues also function as OL hydroxylases. For the OlsE homologue Atu0318 from Agrobacterium tume- faciens we could show that is responsible for the forma- tion of the OL S2 (data not shown). Distant OlsE homologues such as the one in Bradyrhizobium japoni- cum may use distinct substrates. One example for Hydroxylated ornithine lipids in Rhizobium tropici 1507 © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 82 bacterial lipids that are frequently hydroxylated is the hopanoids. In B. japonicum, an a-proteobacteria that forms hopanoids but no OL (Perzl et al., 1998; López-Lara et al., 2003) the OlsE homologue might be responsible for the hydroxylation of hopanoids. The R. tropici mutants deficient in OL hydroxylation showed nodulation phenotypes, indicating that an adequate concentration of the correct OLs is required for the establishment of a successful symbiosis. It is possible that the nodulation phenotype is partly a consequence of the acid sensitivity phenotype, as during establishment of the symbiosis between rhizobia and legumes the bac- teria are exposed to low-pH conditions in the rhizosphere and later again inside symbiosomes (Udvardi and Day, 1997). Other aspects, however, seem to be important as well as the OlsE-deficient mutant grows like the wild-type in media at pH 4.0, but still presents a severe nodulation phenotype. Modification of OL might be also of impor- tance for the animal pathogen Brucella that has to survive acid pH conditions in the range of 4.0–4.5 inside phago- somes (Kohler et al., 2002). Brucella species form OL in a constitutive manner (Comerci et al., 2006; Bukata et al., 2008) and additionally have a close homologue to OlsC from R. tropici which makes it probable that they can form the hydroxylated OL P1. If hydroxylated OLs really play a role in conferring acid resistance then Brucella mutants deficient in their OlsC homologue might be affected in their survival inside phagosomes. The exact function of OL S1 and its hydroxylated forms is still not known, although our data argue for an important role in stress resistance. The knowledge of the complete scheme of OL biosynthesis in R. tropici should facilitate future functional studies on the role of OLs. In addition, the phenomenon of over-complementation described above allows the construction of R. tropici strains princi- pally accumulating one specific class of OL. Character- ization of these strains should make it possible to assign roles to the different forms of OL. Experimental procedures Bacterial strains, plasmids and growth conditions The bacterial strains and plasmids used in the present work and their relevant characteristics are shown in Table 4. R. tropici strains were grown in complex TY medium that con- tained 10 mM CaCl2 (Beringer, 1974) at 30°C, 37°C or 42°C. Acidic media at pH 4.0 and 4.5 were buffered with 25 mM Homopipes (Research Organics, Cleveland, OH, USA) adjusted to the respective pH with NaOH, and media at pH 7.0 were buffered with 25 mM HEPES (Sigma). E. coli strains were grown in Luria–Bertani (LB) medium at 37°C (Sambrook and Russell, 2001). When needed, antibiotics were added at the following final concentrations (mg ml-1): kanamycin (Km) 50; carbenicillin (Cb) 100; tetracycline (Tc) 10; nalidixic acid (Nal) 20; and chloramphenicol (Cm) 60. DNA manipulations Recombinant DNA techniques were performed according to standard protocols (Sambrook and Russell, 2001). The cosmid subclone containing olsE and PCR products were sequenced at Eurofins Medigenomix by the chain termination method. The DNA region containing olsE was analysed using the NCBI (National Center for Biotechnology Information) BLAST network server (Altschul et al., 1997). Oligonucleotide sequences are listed in Table S1. Expression cloning of the R. tropici OL hydroxylase gene olsE A cosmid library of R. tropici CIAT899 made in pVK102 using partially digested HindIII genomic DNA fragments (Vargas et al., 1990) was mobilized into S. meliloti CS111.pNG25 by triparental mating using pRK2013 as the helper plasmid (Fig- urski and Helinski, 1979). CS111.pNG25 was used to facili- tate the screening: CS111 is a phosphatidylserine synthase- deficient mutant (Sohlenkamp et al., 2004) derived from the wild-type 1021 which is constitutively expressing the gene olsB from Burkholderia cenocepacia. CS111.pNG25 will form increased amounts of S1 which is one of the suspected substrates of OlsE while lacking the ninhydrin-positive mem- brane lipid phosphatidylethanolamine. Plasmid pNG25 was constructed as follows: the oligonucleotide primers oLOP111 and oLOP112, introducing NdeI and HindIII sites, respec- tively, were used in the PCR to amplify the gene olsB from B. cenocepacia J2315 using genomic DNA as template. After digestion of the PCR product the obtained fragment was cloned into the plasmid pET17b previously digested with the same enzymes to yield the plasmid pNG23. To obtain plasmid pNG25, the BglII/HindIII fragment containing olsB of B. ceno- cepacia together with the T7 promoter of pET17b was sub- cloned from pNG23 and cloned into BamHI/HindIII-digested pBBR1-MCS. Via diparental mating using E. coli S17-1 as a donor strain, pNG25 was introduced into S. meliloti CS111 to obtain CS111.pNG25 which was used as a receptor strain for the cosmid bank. Cosmid transconjugants were selected on TY containing the following antibiotics: tetracycline 10 mg ml-1; nalidixic acid 20 mg ml-1; chloramphenicol 60 mg ml-1. Four hundred individual S. meliloti transconju- gants harbouring random fragments of the library were picked and streaked for subsequent lipid analysis in small patches (1 cm by 1 cm) on fresh plates. After growth for 3 days, cells from each patch were collected with a toothpick and swirled in 60 ml of chloroform–methanol (1:1, v/v) as described previously (Benning and Somerville, 1992). After the addition of 20 ml of 1 M KCl-0.2 N H3PO4, the tubes were vortexed and centrifuged to separate the organic and aqueous phases. A 10 ml aliquot from the lipid-containing lower phase was spotted on a HPTLC silica gel 60 plate (Merck). The TLC was developed in one dimension using the solvent system chloroform–methanol–glacial acetic acid (130:50:20, v/v). Under these conditions unmodified OL was readily separated from the modified OL we were looking for and from other polar lipids such as PC, PG and CL. Lipids were detected first with iodine and subsequently primary amine containing lipids were visualized by spraying the plates with a solution of 0.2% ninhydrin in acetone and heating the 1508 M. Á. Vences-Guzmán et al.  © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 83 plates at 120°C. A transconjugant containing a gene modify- ing S1 should have two ninhydrin-positive lipids being either S2 and S1 similar to the lipid profile of the R. tropici mutant 899-olsCD1 or S1 and P1. Once S. meliloti CS111.pNG25. pCos94 had been identified, cosmid pCos94 was isolated and re-introduced by conjugation into CS111.pNG25 to confirm that the lipid phenotype was caused by the presence of the cosmid and not by an independent mutation leading to the activation of an endogenous S. meliloti gene. In this independent transconjugant again the presence of S2 was observed. Next, the insert of pCos94 was digested with PstI. The resulting PstI/PstI fragments were subcloned into the broad host vector pRK404 and again mobilized into CS111. pNG25 repeating the lipid analysis described above. A pRK404-derived plasmid containing an approximately 3.5 kb insert was identified (pERMAV04) and its insert sequenced after subcloning into pUC18. Expression of the three candidate ORFs from R. tropici CIAT899 The three candidate ORFs from plasmid pERMAV04 were separately amplified using genomic DNA from R. tropici CIAT899 as a template and XL-PCR polymerase (Applied Biosystems). Specific oligonucleotide primers incorporating Table 4. Bacterial strains and plasmids used in this study. Strain or plasmid Relevant characteristics Reference Rhizobium tropici strains CIAT899 Wild-type; acid-tolerant, Nalr Martínez-Romero et al. (1991) 899-olsCD1 CIAT899 carrying a 211 bp non-polar deletion in olsC Rojas-Jiménez et al. (2005) MAV04 CIAT899 carrying a deletion in olsE This work MAV05 CIAT899 carrying deletions in olsC and olsE This work Sinorhizobium meliloti strains CS111 pssA-deficient mutant of wild-type 1021 Sohlenkamp et al. (2004) Burkholderia cenocepacia strains J2315 Wild-type Holden et al. (2009) Escherichia coli strains DH5a recA1, F80 lacZDM15; cloning strain Hanahan (1983) S17-1 thi pro recA hsdR- hsdM + RP4 integrated in the chromosome, 2-Tc::Mu, Km::Tn7(Tpr/Smr) Simon et al. (1983) Plasmids pET17b Expression vector, Cbr Studier (1991) pET9a Expression vector, Kanr Studier (1991) pRK404 Broad-host-range vector, tetracycline-resistant Ditta et al. (1985) pBBR1MCS Broad-host-range plasmid, chloramphenicol-resistant Kovach et al. (1994) pUC18 Cloning vector, ampicillin-resistant Yanisch-Perron et al. (1985) pRK2013 Helper plasmid; Kmr Ditta et al. (1985) pVK102 Cosmid vector Vargas et al. (1990) pK18mobsacB Conjugative suicide vector, kanamycin-resistant Schäfer et al. (1994) pNG23 olsB of B. cenocepacia cloned in pET17b This work pNG25 olsB of B. cenocepacia subcloned as a BglII/HindIII fragment from pNG23 into BamHI/HindIII-digested pBBR1MCS This work pCCS98 olsC of R. tropici in pET9a This work pCos94 pVK102 derivative containing the olsE gene This work pEMAV01 1 kb fragment upstream of olsE, cloned as SmaI/BamHI fragment in pUC18 This work pEMAV02 1 kb fragment downstream of olsE, cloned as BamHI/HindIII fragment in pUC18 This work pEMAV03 1 kb upstream and 1 kb downstream sequences flanking olsE, cloned into pUC18 This work pPMAV04 Suicide vector for construction of mutant MAV04 and MAV05 This work pURMAV03 olsE-containing 3.5 kb fragment of pCos94 cloned as PstI/PstI fragment in pUC18 This work pERMAV04 olsE-containing 3.5 kb fragment of pCos94 cloned as PstI/PstI fragment in pRK404 This work pERMAV06 pET9a cloned as a BamHI fragment into pRK404 This work pEMAV07 ORF1 in pET9a This work pEMAV08 ORF2 in pET9a This work pEMAV09 olsE in pET9a This work pERMAV11 pEMAV07 cloned as a BamHI fragment into pRK404 This work pERMAV12 pEMAV08 cloned as a BamHI fragment into pRK404 This work pERMAV13 pEMAV09 cloned as a BamHI fragment into pRK404 This work pERMAV15 pCCS98 cloned as a BamHI fragment into pRK404 This work pEMAV16 olsC cloned as a BamHI/BglII fragment into BamHI-digested pEMAV09 This work pERMAV17 pEMAV16 cloned as a BamHI fragment into pRK404 This work Hydroxylated ornithine lipids in Rhizobium tropici 1509 © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 84 NdeI and BamHI sites into the final PCR products were used (oORF1-01 and oORF1-02 for ORF1; oORF2-01 and oORF2-02 for ORF2; oORF3-01 and oORF3-02 for ORF3). After digestion with the respective enzymes, the PCR prod- ucts were cloned as NdeI/BamHI fragments into pET9a to yield the plasmids pEMAV07, pEMAV08 and pEMAV09 respectively. These three plasmids and pET9a were linear- ized with BamHI and were cloned into the BamHI site of pRK404, similarly to an earlier description (Gao et al., 2004) yielding the plasmids pERMAV11, pERMAV12, pERMAV13 and pERMAV06 respectively. Plasmids were mobilized into S. meliloti CS111.pNG25 and the lipids of the transconjugants were assayed as described above. Deletion of the olsE gene from R. tropici CIAT899 Oligonucleotide primers oOlsX899ar1 and oOlsX899ar2 were used in a PCR (XL-PCR kit; Applied Biosystems) to amplify about 1.0 kb of genomic DNA upstream of the puta- tive olsE gene from R. tropici CIAT899, introducing SmaI and BamHI sites into the PCR product. Similarly, primers oOlsX899ab1 and oOlsX899ab2 were used to amplify about 1.0 kb of genomic DNA downstream of the putative olsE gene from R. tropici CIAT899, introducing BamHI and HindIII sites into the PCR product. After digestion with the respective enzymes, PCR products were cloned as SmaI/BamHI or BamHI/HindIII fragments into pUC18 to yield the plasmids pUMAV01 and pUMAV02 respectively. Then, the BamHI/ HindIII fragment from pUMAV02 was subcloned into pUMAV01 to yield pUMAV03. Plasmid pUMAV03 was digested with SmaI and HindIII to subclone the regions usually flanking the rhizobial olsE gene into the suicide vector pK18mobsacB (Schäfer et al., 1994) to yield pPMAV04. Via diparental mating using E. coli S17-1 (Simon et al., 1983) as a mobilizing strain, pPMAV04 was introduced into the wild-type strain R. tropici CIAT899. Transconjugants were selected on TY medium containing neomycin to select for single recombinants in a first step. The plasmid pK18mobsacB contains the sacB gene (Selbitschka et al., 1993), which confers sucrose sensitivity to many bacteria. Growth of the single recombinants on high sucrose will there- fore select for double recombinants and the loss of the vector backbone of pK18mobsacB from the bacterial genome. Single recombinants were grown under non-selective condi- tions in complex medium for 1 day before being plated on TY medium containing 12% (w/v) sucrose. Several large and small colonies grew after 5 days, and the membrane lipids of eight candidates were analysed by in vivo labelling during growth on complex medium with [14C]acetate and subsequent TLC (data not shown). Four clones lacking S2 and P2 were identified. Southern blot analysis confirmed that the S2- and P2-deficient strains were indeed double recombinants in which the gene olsE was deleted (data not shown). Construction of a double mutant deficient in olsE and olsC To construct a R. tropici double mutant deficient in olsE and olsC, the suicide plasmid pPMAV04 was conjugated into the olsC-deficient mutant 899-olsCD1 (Rojas-Jiménez et al., 2005). The selection for double recombinants was performed in two steps as described above. Ten isolated colonies were chosen and their lipids were labelled with [14C]acetate (see below). We used R. tropici CIAT899 and the mutant 899- olsCD1 as control strains. The lipids were analysed by TLC. One strain presented the expected phenotype which is the absence of the OLs S2, P1 and P2. Therefore this colony was called MAV05. Southern blot analysis confirmed that MAV05 was indeed a double recombinant in which the genes olsC and olsE were deleted (data not shown). Complementation of the R. tropici mutants MAV04, MAV05 and 899-olsCD1 To show that the observed mutant phenotypes were caused by the introduced deletion and not by a secondary indepen- dent mutation, the mutants were complemented. The olsE- deficient mutant MAV04 was complemented with the plasmid pERMAV13. In this construct olsE is expressed under control of the T7 promoter. In earlier work we had observed constitutive expression from this promoter in different Rhizobiaceae. In the study published by Rojas-Jiménez et al. (2005) the mutant 899-olsCD1 was complemented by olsC under its endogenous promoter, but in order to be able to compare the results from the complementation of the olsC- deficient mutant with the complementations of the mutants MAV04 and MAV05 a new plasmid was constructed. The gene olsC was amplified using genomic DNA from R. tropici CIAT899 as a template and XL-PCR polymerase (Applied Biosystems). Specific oligonucleotide primers incor- porating NdeI and BamHI sites into the final PCR product were used (o5B_olsC and o3_olsC). The digested PCR product was cloned into pET9a to yield the plasmid pCCS98. Plasmid pCCS98 was linearized with BamHI and cloned into BamHI-digested pRK404 to yield pERMAV15. To comple- ment the double mutant MAV05 a plasmid containing both olsC and olsE under the control of the T7 promoter was constructed. A DNA fragment containing olsC under the control of the T7 promoter was subcloned from pCCS98 as BamHI/BglII fragment into the BamHI-digested pEMAV09 yielding plasmid pEMAV16. Plasmid pEMAV16 therefore con- tains the genes olsC and olsE, both under the control of separate T7 promoters. Subsequently, pEMAV16 was linear- ized with BamHI and cloned into BamHI-linearized pRK404 to yield pERMAV17. In vivo labelling of S. meliloti and R. tropici with [14C]acetate and quantitative analysis of lipid extracts The lipid compositions of bacterial strains were determined following labelling with [1-14C]acetate (Amersham Biosciences). Cultures (1 ml) of wild-type and mutant strains were inoculated from pre-cultures grown in the same medium. After addition of 0.5 mCi of [14C]acetate (60 mCi m- mol-1) to each culture, the cultures were incubated for 4 h. The cells were harvested by centrifugation, washed with 500 ml of water and resuspended in 100 ml of water, and lipid extracts were obtained according to Bligh and Dyer (1959). Aliquots of the lipid extracts were spotted on high- performance TLC silica gel 60 (Merck, Poole, UK) plates and 1510 M. Á. Vences-Guzmán et al.  © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 85 were separated in two dimensions using chloroform/ methanol/water (140:60:10, v/v) as a mobile phase for the first dimension and chloroform/methanol/glacial acetic acid (130:50:20, v/v) for the second dimension. Primary amine- containing lipids were visualized by spraying the plates with a solution of 0.2% ninhydrin in acetone and subsequent treat- ment at 120°C for 10 min. To visualize the membrane lipids, developed two-dimensional TLC plates were exposed to autoradiography film (Kodak) or to a PhosphorImager screen (Amersham Biosciences). The individual lipids were quanti- fied using ImageQuant software (Amersham Biosciences). Separation of IM and OM and determination of their respective lipid compositions Membrane separation was performed as described previously (de Maagd and Lugtenberg, 1986; Klüsener et al., 2009), with minor modifications. A 400 ml culture R. tropici CIAT899 was grown in TY medium at 30°C overnight to an OD600 of 0.5–0.6. Cells were harvested by centrifugation at 10 000 g, 4°C, for 10 min. The cells were resuspended in 24 ml of lysis buffer [50 mM Tris-HCl, pH 7.5, 20% (w/v) sucrose, 0.2 M KCl, 0.2 mM dithiothreitol (DTT), 0.2 mg ml-1 DNase I, 0.2 mg ml-1 RNaseA] and disrupted by two passages through a pre-chilled French pressure cell at 16 000 lb in-2. The lysate was treated with 0.5 mg ml-1 lysozyme for 1 h on ice and centrifuged at 10 000 g for 20 min, 4°C, to remove the unbroken cells. The supernatant was centrifuged at 150 000 g (SW40Ti), 4°C, for 1 h to collect the membranes. The resulting membrane pellet was carefully resuspended in 2 ml of 20% (w/v) sucrose containing 5 mM EDTA, pH 7.5, and 0.2 mM DTT. Material that was not completely suspended was removed by centrifugation for 5 min at 16 000 g. The gradient was prepared by layering 7.5 ml of 53% (w/v) sucrose over a cushion of 2.5 ml of 70% (w/v) sucrose. Both sucrose solutions contained 5 mM EDTA, pH 7.5. The membrane suspension was layered on the top of the gradient, and sucrose density gradient ultracentrifugation was carried out at 100 000 g (SW40Ti), 4°C, for 16 h. After ultracentrifugation, the separated membranes were fraction- ated in 500 ml aliquots. For each fraction the protein concen- tration was estimated, and the density, the NADH activity and the 2-keto-3-deoxyoctonate (KDO) content were determined. The protein distribution was estimated using absorption mea- surements at 280 nm (Scopes, 1987). The NADH oxidase activity was determined by the method of Osborn et al. (1972) and the KDO content was determined as described earlier after the fractions had been precipitated twice with 10% (w/v) TCA (Karkhanis et al., 1978). NADH oxidase activity and KDO content were used as marker for the IM and OM respec- tively. Fractions corresponding to the IM and the OM were pooled and the lipids were extracted with 1-butanol (Bremer, 1963). Lipids were analysed using two-dimensional TLC as described above and the lipids were detected by oxidative charring using ceric sulphate in sulphuric acid (Villaescusa and Pettit, 1972). The lipid spots were quantified using the program ImageQuant (Applied Biosystems). ESI-MS/MS analysis of lipids S1 and S2 In order to identify in which part of the OL S2 the modification is encountered, a 1 l culture of the mutant 899-olsCD1 (Rojas- Jiménez et al., 2005) was grown to an optical density at 620 nm of 1.0 in TY medium, and lipids were extracted accord- ing to a modified Bligh-and-Dyer procedure (Bligh and Dyer, 1959). Lipids were fractionated using a silica column and chloroform/methanol/water (140:60:8, v/v) as a mobile phase. Fractions were analysed by one-dimensional TLC using chloroform/methanol/water (140:60:8, v/v) as a mobile phase. Fractions containing OLs were identified by iodine and ninhy- drin staining as described above. OL-containing fractions were dried under N2 stream and re-dissolved in methanol/ chloroform (1:1, v/v). LC-ESI/MS of lipids was performed using an Agilent 1200 Quaternary LC system coupled to a QSTAR XL quadrupole time-of-flight tandem mass spectrometer (Applied Biosystems, Foster City, CA). An Ascentis® Si HPLC column (5 mm, 25 cm ¥ 2.1 mm) was used. Mobile phase A consisted of chloroform/methanol/aqueous ammonium hydroxide (800:195:5, v/v). Mobile phase B consisted of chloroform/methanol/water/aqueous ammonium hydroxide (600:340:50:5, v/v). Mobile phase C consisted of chloroform/methanol/water/aqueous ammonium hydroxide (450:450:95:5, v/v). The elution programme consisted of the following: 100% mobile phase A was held isocratically for 2 min and then linearly increased to 100% mobile phase B over 14 min and held at 100% B for 11 min. The LC gradient was then changed to 100% mobile phase C over 3 min and held at 100% C for 3 min, and finally returned to 100% A over 0.5 min and held at 100% A for 5 min. The total LC flow rate was 300 ml min-1. The post-column splitter diverted ~10% of the LC flow to the ESI source of the Q-Star XL mass spectrom- eter, with MS settings as follows: IS = -4500 V, CUR = 20 psi, GS1 = 20 psi, DP = -55 V and FP = -150 V. Nitrogen was used as the collision gas. Data acquisition and analysis were performed using Analyst QS software version 1.1. Determination of the position of the hydroxyl group introduced by OlsC into OLs Large cultures (4 l) of the R. tropici mutant MAV05 and the strain MAV05.pERMAV15 were grown to an OD620 of 0.9 in TY medium. MAV05 only forms S1 and MAV05.pERMAV15 forms preferentially P1. Cells were harvested and lipids were extracted from the cell pellets according to a modified Bligh- and-Dyer method. OLs S1 and P1 were purified using pre- parative TLC using Si500F plates (Baker) in two steps. First chloroform/methanol/water (140:60:10, v/v) was used as a mobile phase and the OLs were purified from the silica. Enriched OLs were further purified by a second preparative TLC using chloroform/methanol/glacial acetic acid (130:50:20, v/v) as mobile phase. ESI-MS/MS analysis of S1 and P1 was performed as described above. The derivatization of the lipids was performed essentially as described by Gibbons et al. (2008). Purified OLs S1 and P1 were hydrolysed in acidic methanol, and then converted to TMS ethers. Hydroxy fatty acid standards (a- and b-hydroxy palmitic acid, a- and b-hydroxy stearic acid) were processed and analysed in parallel with the samples. Typically, about 1 mg of sample was dried in a Reacti-vial and samples were hydrolysed by adding 300 ml of 1 M HCl in methanol and heated at 80°C for 16 h. The reactions were cooled and solvents were evaporated under a stream of nitrogen. Next, 200 ml Tri-Sil HTP reagent (Thermo) was added to the dried Hydroxylated ornithine lipids in Rhizobium tropici 1511 © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 86 samples. After incubation for 1 h at 25°C a 20 ml aliquot was diluted 1:6 in hexane and transferred to a new vial for GC/MS analysis. GC/MS was performed using a Clarus 600T MS instrument coupled to a Clarus 600 gas chromatography system (Perkin Elmer). The column was a Elite-5 MS (0.32 mm internal diam- eter and 0.25 mm phase thickness) from Perkin Elmer. The temperature programme of the GC was as follows: the column oven temperature was initially held at 140°C for 6 min, increased to 250°C at a rate of 4°C min-1 and finally held at 250°C for 5 min. The total run time was 38.5 min. The injector was operated in the split mode, and the temperature of the injector was kept at 250°C. Helium was the carrier gas at a constant pressure of 7 psi. The instrument was operated in the electron impact (EI) mode with the electron energy set at 70 eV. Plant tests Phaseolus vulgaris seeds were surface-sterilized with 1.2% sodium hypochlorite and were germinated on 1% agar-water plates as described (Vinuesa et al., 1999). Seedlings were transferred to 250 ml flasks filled 220 ml of nitrogen-free nutrient solution (Fahraeus, 1957) containing agar at 0.7% and were inoculated with about 50 000 cfu ml-1 per plant. Plants were grown in a controlled growth chamber at 28°C with a 15 h day/9 h night cycle and harvested 21 days after inoculation. Nitrogenase activity of nodulated roots was determined by the acetylene reduction assay as described previously (Martínez et al., 1985). Nitrogen fixation activity per plant was normalized with respect to the nodule fresh weight per plant. Acknowledgements This research was supported by grants from CONACyT- Mexico (46020-N) and DGAPA/UNAM (IN217907 and IN201310). Z.G. and the mass spectrometry facility in the Department of Biochemistry, Duke University Medical Center, were supported by a LIPID MAPS glue grant (GM-069338) from the National Institutes of Health. M.Á.V.-G. is a PhD student from the Programa de Doctorado en Ciencias Biomédicas, Universidad Nacional Autónoma de México and is a recipient of a scholarship from the Consejo Nacional de Ciencia y Tecnología, Mexico. We are grateful to Alfonso Leija Salas for his support in nodule analysis. References Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., Miller, W., and Lipman, D.J. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25: 3389–3402. Arthington, B.A., Bennett, L.G., Skatrud, P.L., Guynn, C.J., Barbuch, R.J., Ulbright, C.E., and Bard, M. (1991) Cloning, disruption and sequence of the gene encoding yeast C-5 sterol desaturase. Gene 102: 39–44. Asselineau, J. (1991) Bacterial lipids containing amino acids or peptides linked by amide bonds. Fortschr Chem Org Naturst 56: 1–85. Bard, M., Bruner, D.A., Pierson, C.A., Lees, N.D., Biermann, B., Frye, L., et al. (1996) Cloning and characterization of ERG25, the Saccharomyces cerevisiae gene encoding C-4 sterol methyl oxidase. Proc Natl Acad Sci USA 93: 186– 190. Benning, C., and Somerville, C.R. (1992) Isolation and genetic complementation of a sulfolipid-deficient mutant of Rhodobacter sphaeroides. J Bacteriol 174: 2352–2360. Benning, C., Huang, Z.H., and Gage, D.A. (1995) Accumu- lation of a novel glycolipid and a betaine lipid in cells of Rhodobacter sphaeroides grown under phosphate limitation. Arch Biochem Biophys 317: 103–111. Beringer, J.E. (1974) R factor transfer in Rhizobium leguminosarum. J Gen Microbiol 84: 188–198. Bligh, E.G., and Dyer, W.J. (1959) A rapid method of total lipid extraction and purification. Can J Biochem Physiol 37: 911–917. Bremer, J. (1963) Carnitine in intermediary metabolism. The biosynthesis of palmitylcarnitine by cell subfractions. J Biol Chem 238: 2774–2779. Bukata, L., Altabe, S., de Mendoza, D., Ugalde, R.A., and Comerci, D.J. (2008) Phosphatidylethanolamine synthesis is required for optimal virulence of Brucella abortus. J Bacteriol 190: 8197–8203. Comerci, D.J., Altabe, S., de Mendoza, D., and Ugalde, R.A. (2006) Brucella abortus synthesizes phosphatidylcholine from choline provided by the host. J Bacteriol 188: 1929– 1934. Dees, C., and Shively, J.M. (1982) Localization of quantita- tion of the ornithine lipid of Thiobacillus thiooxidans. J Bacteriol 149: 798–799. Ditta, G., Schmidhauser, T., Yakobson, E., Lu, P., Liang, X.W., Finlay, D.R., et al. (1985) Plasmids related to the broad host range vector, pRK290, useful for gene cloning and for monitoring gene expression. Plasmid 13: 149– 153. Fahraeus, G. (1957) The infection of clover root hairs by nodule bacteria studied by a simple glass slide technique. J Gen Microbiol 16: 374–381. Figurski, D.H., and Helinski, D.R. (1979) Replication of an origin-containing derivative of plasmid RK2 dependent on a plasmid function provided in trans. Proc Natl Acad Sci USA 76: 1648–1652. Fox, B.G., Shanklin, J., Somerville, C., and Munck, E. (1993) Stearoyl-acyl carrier protein delta 9 desaturase from Ricinus communis is a diiron-oxo protein. Proc Natl Acad Sci USA 90: 2486–2490. Galbraith, L., Jonsson, M.H., Rudhe, L.C., and Wilkinson, S.G. (1999) Lipids and fatty acids of Burkholderia and Ralstonia species. FEMS Microbiol Lett 173: 359–364. Gao, J.L., Weissenmayer, B., Taylor, A.M., Thomas-Oates, J., López-Lara, I.M., and Geiger, O. (2004) Identification of a gene required for the formation of lyso-ornithine lipid, an intermediate in the biosynthesis of ornithine-containing lipids. Mol Microbiol 53: 1757–1770. Geiger, O., Röhrs, V., Weissenmayer, B., Finan, T.M., and Thomas-Oates, J.E. (1999) The regulator gene phoB mediates phosphate stress-controlled synthesis of the membrane lipid diacylglyceryl-N,N,N-trimethylhomoserine in Rhizobium (Sinorhizobium) meliloti. Mol Microbiol 32: 63–73. 1512 M. Á. Vences-Guzmán et al.  © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 87 Geiger, O., González-Silva, N., López-Lara, I.M., and Sohlenkamp, C. (2010) Amino acid-containing membrane lipids in bacteria. Prog Lipid Res 49: 46–60. Gibbons, H.S., Lin, S., Cotter, R.J., and Raetz, C.R. (2000) Oxygen requirement for the biosynthesis of the S-2- hydroxymyristate moiety in Salmonella typhimurium lipid A. Function of LpxO, A new Fe2+/alpha-ketoglutarate- dependent dioxygenase homologue. J Biol Chem 275: 32940–32949. Gibbons, H.S., Reynolds, C.M., Guan, Z., and Raetz, C.R. (2008) An inner membrane dioxygenase that generates the 2-hydroxymyristate moiety of Salmonella lipid A. Biochem- istry 47: 2814–2825. Hanahan, D. (1983) Studies on transformation of Escherichia coli with plasmids. J Mol Biol 166: 557–580. Heath, R.J., Jackowski, S., and Rock, C.O. (2002) Fatty acid and phospholipid metabolism in prokaryotes. In Biochem- istry of Lipids, Lipoproteins and Membranes. Vance, D.E., and Vance, J.E. (eds). Amsterdam: Elsevier, pp. 55–92. Holden, M.T., Seth-Smith, H.M., Crossman, L.C., Sebaihia, M., Bentley, S.D., Cerdeno-Tarraga, A.M., et al. (2009) The genome of Burkholderia cenocepacia J2315, an epidemic pathogen of cystic fibrosis patients. J Bacteriol 191: 261– 277. Jackson, M., Crick, D.C., and Brennan, P.J. (2000) Phos- phatidylinositol is an essential phospholipid of mycobacteria. J Biol Chem 275: 30092–30099. Karkhanis, Y.D., Zeltner, J.Y., Jackson, J.J., and Carlo, D.J. (1978) A new and improved microassay to determine 2-keto-3-deoxyoctonate in lipopolysaccharide of gram- negative bacteria. Anal Biochem 85: 595–601. Kawai, Y., Yano, I., and Kaneda, K. (1988) Various kinds of lipoamino acids including a novel serine-containing lipid in an opportunistic pathogen Flavobacterium. Their structures and biological activities on erythrocytes. Eur J Biochem 171: 73–80. Kennedy, M.A., Johnson, T.A., Lees, N.D., Barbuch, R., Eck- stein, J.A., and Bard, M. (2000) Cloning and sequencing of the Candida albicans C-4 sterol methyl oxidase gene (ERG25) and expression of an ERG25 conditional lethal mutation in Saccharomyces cerevisiae. Lipids 35: 257– 262. Klüsener, S., Aktas, M., Thormann, K.M., Wessel, M., and Narberhaus, F. (2009) Expression and physiological rel- evance of Agrobacterium tumefaciens phosphatidylcholine biosynthesis genes. J Bacteriol 191: 365–374. Knoche, H.W., and Shively, J.M. (1972) The structure of an ornithine-containing lipid from Thiobacillus thiooxidans. J Biol Chem 247: 170–178. Kohler, S., Porte, F., Jubier-Maurin, V., Ouahrani-Bettache, S., Teyssier, J., and Liautard, J.P. (2002) The intramac- rophagic environment of Brucella suis and bacterial response. Vet Microbiol 90: 299–309. Kok, M., Oldenhuis, R., van der Linden, M.P., Raatjes, P., Kingma, J., van Lelyveld, P.H., and Witholt, B. (1989) The Pseudomonas oleovorans alkane hydroxylase gene. Sequence and expression. J Biol Chem 264: 5435– 5441. Kovach, M.E., Phillips, R.W., Elzer, P.H., Roop, R.M., 2nd, and Peterson, K.M. (1994) pBBR1MCS: a broad-host- range cloning vector. Biotechniques 16: 800–802. López-Lara, I.M., Sohlenkamp, C., and Geiger, O. (2003) Membrane lipids in plant-associated bacteria: their biosyn- theses and possible functions. Mol Plant Microbe Interact 16: 567–579. de Maagd, R.A., and Lugtenberg, B. (1986) Fractionation of Rhizobium leguminosarum cells into outer membrane, cytoplasmic membrane, periplasmic, and cytoplasmic components. J Bacteriol 167: 1083–1085. Martínez, E., Pardo, M.A., Palacios, R., and Cevallos, M.A. (1985) Reiteration of nitrogen fixation gene sequences and specificity of Rhizobium in nodulation and nitrogen fixation in Phaseolus vulgaris. J Gen Microbiol 131: 1779–1786. Martínez-Romero, E., Segovia, L., Mercante, F.M., Franco, A.A., Graham, P., and Pardo, M.A. (1991) Rhizobium tropici, a novel species nodulating Phaseolus vulgaris L. beans and Leucaena sp. trees. Int J Syst Bacteriol 41: 417–426. Meneely, K.M., Barr, E.W., Bollinger, J.M., Jr, and Lamb, A.L. (2009) Kinetic mechanism of ornithine hydroxylase (PvdA) from Pseudomonas aeruginosa: substrate triggering of O2 addition but not flavin reduction. Biochemistry 48: 4371– 4376. Mitchell, A.G., and Martin, C.E. (1997) Fah1p, a Saccharo- myces cerevisiae cytochrome b5 fusion protein, and its Arabidopsis thaliana homolog that lacks the cytochrome b5 domain both function in the alpha-hydroxylation of sphingolipid-associated very long chain fatty acids. J Biol Chem 272: 28281–28288. Murata, T., Tseng, W., Guina, T., Miller, S.I., and Nikaido, H. (2007) PhoPQ-mediated regulation produces a more robust permeability barrier in the outer membrane of Sal- monella enterica serovar typhimurium. J Bacteriol 189: 7213–7222. Nikaido, H. (2003) Molecular basis of bacterial outer mem- brane permeability revisited. Microbiol Mol Biol Rev 67: 593–656. Osborn, M.J., Gander, J.E., Parisi, E., and Carson, J. (1972) Mechanism of assembly of the outer membrane of Salmo- nella typhimurium. Isolation and characterization of cyto- plasmic and outer membrane. J Biol Chem 247: 3962– 3972. Perzl, M., Reipen, I.G., Schmitz, S., Poralla, K., Sahm, H., Sprenger, G.A., and Kannenberg, E.L. (1998) Cloning of conserved genes from Zymomonas mobilis and Bradyrhizobium japonicum that function in the biosynthesis of hopanoid lipids. Biochim Biophys Acta 1393: 108–118. Rojas-Jiménez, K., Sohlenkamp, C., Geiger, O., Martínez- Romero, E., Werner, D., and Vinuesa, P. (2005) A ClC chloride channel homolog and ornithine-containing mem- brane lipids of Rhizobium tropici CIAT899 are involved in symbiotic efficiency and acid tolerance. Mol Plant Microbe Interact 18: 1175–1185. Sambrook, J., and Russell, D.W. (2001) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Schäfer, A., Tauch, A., Jager, W., Kalinowski, J., Thierbach, G., and Puhler, A. (1994) Small mobilizable multi-purpose cloning vectors derived from the Escherichia coli plasmids pK18 and pK19: selection of defined deletions in the chro- mosome of Corynebacterium glutamicum. Gene 145: 69–73. Hydroxylated ornithine lipids in Rhizobium tropici 1513 © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 88 Scopes, R.K. (1987) Measurement of enzyme activity. In Protein Purification – Principles and Practice. Cantor, C.R. (ed.). New York: Springer, pp. 253–283. Selbitschka, W., Niemann, S., and Pühler, A. (1993) Con- struction of gene replacement vectors for gram-negative bacteria using a genetically modified sacRN as a positive selection marker. Appl Microbiol Biotechnol 38: 615–618. Shanklin, J., Whittle, E., and Fox, B.G. (1994) Eight histidine residues are catalytically essential in a membrane- associated iron enzyme, stearoyl-CoA desaturase, and are conserved in alkane hydroxylase and xylene monooxygenase. Biochemistry 33: 12787–12794. Simon, R., Priefer, U., and Pühler, A. (1983) A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram-negative bacteria. Bio- technology 1: 784–791. Sohlenkamp, C., López-Lara, I.M., and Geiger, O. (2003) Biosynthesis of phosphatidylcholine in bacteria. Prog Lipid Res 42: 115–162. Sohlenkamp, C., de Rudder, K.E., and Geiger, O. (2004) Phosphatidylethanolamine is not essential for growth of Sinorhizobium meliloti on complex culture media. J Bacteriol 186: 1667–1677. Sohlenkamp, C., Galindo-Lagunas, K.A., Guan, Z., Vinuesa, P., Robinson, S., Thomas-Oates, J., et al. (2007) The lipid lysyl-phosphatidylglycerol is present in membranes of Rhizobium tropici CIAT899 and confers increased resis- tance to polymyxin B under acidic growth conditions. Mol Plant Microbe Interact 20: 1421–1430. Studier, F.W. (1991) Use of bacteriophage T7 lysozyme to improve an inducible T7 expression system. J Mol Biol 219: 37–44. Suzuki, M., Hayakawa, T., Shaw, J.P., Rekik, M., and Harayama, S. (1991) Primary structure of xylene monooxy- genase: similarities to and differences from the alkane hydroxylation system. J Bacteriol 173: 1690–1695. Taylor, C.J., Anderson, A.J., and Wilkinson, S.G. (1998) Phe- notypic variation of lipid composition in Burkholderia cepacia: a response to increased growth temperature is a greater content of 2-hydroxy acids in phosphatidylethano- lamine and ornithine amide lipid. Microbiology 144: 1737– 1745. Udvardi, M.K., and Day, D.A. (1997) Metabolite transport across symbiotic membranes of legume nodules. Annu Rev Plant Physiol Plant Mol Biol 48: 493–523. Vargas, C., Martinez, L.J., Megias, M., and Quinto, C. (1990) Identification and cloning of nodulation genes and host specificity determinants of the broad host-range Rhizobium leguminosarum biovar phaseoli strain CIAT899. Mol Micro- biol 4: 1899–1910. Villaescusa, F.W., and Pettit, G.R. (1972) Steroids and related natural products. 69. Synthesis of 20(22)-Dihydro- 23-deoxodigitoxigenin. J Org Chem 37: 569–572. Vinuesa, P., Reuhs, B.L., Breton, C., and Werner, D. (1999) Identification of a plasmid-borne locus in Rhizobium etli KIM5s involved in lipopolysaccharide O-chain biosynthesis and nodulation of Phaseolus vulgaris. J Bacteriol 181: 5606–5614. Vinuesa, P., Neumann-Silkow, F., Pacios-Bras, C., Spaink, H.P., Martínez-Romero, E., and Werner, D. (2003) Genetic analysis of a pH-regulated operon from Rhizobium tropici CIAT899 involved in acid tolerance and nodulation competitiveness. Mol Plant Microbe Interact 16: 159–168. Visca, P., Ciervo, A., and Orsi, N. (1994) Cloning and nucle- otide sequence of the pvdA gene encoding the pyoverdin biosynthetic enzyme L-ornithine N5-oxygenase in Pseudomonas aeruginosa. J Bacteriol 176: 1128–1140. Weissenmayer, B., Gao, J.L., López-Lara, I.M., and Geiger, O. (2002) Identification of a gene required for the biosynthesis of ornithine-derived lipids. Mol Microbiol 45: 721–733. Yanisch-Perron, C., Vieira, J., and Messing, J. (1985) Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33: 103–119. Supporting information Additional supporting information may be found in the online version of this article. Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article. 1514 M. Á. Vences-Guzmán et al.  © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 79, 1496–1514 89 OCH3 O O Si CH3 CH3 CH3 OCH3 OO Si CH3 CH3 CH3 OCH3 O O Si CH3 CH3 CH3 371 327 OCH3 OO Si CH3 CH3 CH3 175 73 371 A B C D R e la ti v e A b u n d a n ce R e la ti v e A b u n d a n ce R e la ti v e A b u n d a n ce R e la ti v e A b u n d a n ce Time (min) Time (min) m/z m/z 90 OCH3 O O Si CH3 CH3 CH3 383 339 E F G C19:1c, methyl ester C18:0, 3-OH, methyl ester, TMS ether R e la ti v e A b u n d a n ce R e la ti v e A b u n d a n ce R e la ti v e A b u n d a n ce * * * Time (min) Time (min) m/z Figure S1. Determination of the position of the hydroxyl group introduced by OlsC into the esterified fatty acid of the ornithine lipid P1. 2-hydroxystearate and 3-hydroxystearate standards and ornithine lipids S1 and P1 were transmethylated, then converted to TMS derivatives and finally analyzed by GC/MS. (A) Total ion chromatogram (TIC) of the GC/MS analysis of derivatized 2-hydroxystearate. (B) EI/MS spectrum of the major peak in panel A showing a fragmentation pattern as expected for a TMS-2-hydroxystearoyl methyl ester. (C) TIC of the GC/MS analysis of derivatized 3-hydroxystearate. (D) EI/MS spectrum of the major peak in panel C showing a fragmentation pattern as expected for a TMS-3-hydroxystearoyl methylester. (E) TIC of the GC/MS analysis of the fatty acids derivatives obtained from the derivatization of ornithine lipid S1. (F) TIC of the GC/MS analysis of the fatty acid derivatives obtained from the derivatization of ornithine lipid P1. The asterics indicate peaks that are present in chromatogram F but absent in chromatogram E. (G) EI/MS spectrum of the major peak present in chromatogram F (P1) but absent from chromatogram E (S1) (retention time: 28.27 min). The observed fragmentation indicates that the spectrum is produced by TMS-2-hydroxy lactobacillic acid methyl ester. The proposed origins of the major fragments are indicated. The other two peaks labeled with asteriscs in panel F also fragment as expected for TMS-2-hydroxyl fatty acid methyl esters. 91 Figure S2. Bean plants were harvested 21 days after inoculation. A selection of nodules representing the different sizes and colors was harvested. Nodules were sectioned using a razorblade. Roots from bean plants were inoculated with the wild type R. tropici CIAT899 (A), OlsC-deficient mutant 899-olsC∆1, OlsE-deficient mutant MAV04 (C), and OlsC/OlsE-deficient double mutant MAV05 (D). 92 Name Sequence (5’-3’) Incorporated restriction site (underlined) oLOP111 AggAATACATATgCgAgAACTgCCgACgCC NdeI oLOP112 ACCCAAgCTTTCAgCCCAggAAgTggCggg HindIII oORF1_01 AACTgCATATggATCTgACgAATTggAAgggCgTC NdeI oORF1_02 ACTgggATCCTAgCCgAATgTCAATTTCgTCTTC BamHI oORF2_01 AACTgCATATgCgCATAgAgCAggCggCAgACgAAC NdeI oORF2_02 ACTgggATCCATCAgTgCATCCgAggCAAACTCAgTg BamHI oORF3_01 ACTgACATATgATATTTCAAAgCAgTTCgCgTC NdeI oORF3_02 ACTgggATCCTAgCCTgATTTgCggATTTTAgC BamHI oOlsX899ar1 ACTgCCCgggATCgCTTTCgCTATCTTTTCgAgg SmaI oOlsX899ar2 ACTgggATCCATCgTCCTCAgaTTCTgATCggCg BamHI oOlsX899ab1 ACTgggATCCggCAAgAAgCCgAAggAATTgATC BamHI oOlsX899ab2 ACTgAAgCTTggCTATggCgCACggCCTTgTggAA HindIII o5B_olsC ACgTCATATgACggAgAgTCCCTTgAgCgCACC NdeI o3_olsC ACgTggATCCTCAgggCTTCgggCggTTTCTgAACg BamHI Supplementary Table 1. Oligonucleotides used in this study 93 8.1.3. III Artículo  González‐Silva, N.,  López‐Lara,  I. M.,  Reyes‐Lamothe,  R.,  Taylor, A. M.,  Sumpton, D.,  Thomas‐Oates,  J.,  and  Geiger,  O.  (2011)  The  dioxygenase‐encoding  olsD  gene  from  Burkholderia  cenocepacia  causes  the  hydroxylation  of  the  amide‐linked  fatty  acyl  moiety of ornithine‐containing membrane lipids. Biochemistry 50, 6396‐6408.  Burkholderia  cenocepacia  es  un  patógeno  oportunista  importante,  y  una  de  sus  características más sobresalientes es el repertorio de lípidos polares de su membrana, que  incluye a  fosfatidiletanolamina  (PE) y  lípidos de ornitina  (OLs), así como sus derivados 2‐ hidroxilados  de  PE  y  OLs  (2‐OH‐PE  y  2‐OH‐OLs,  respectivamente).  Los  últimos  lípidos  difieren de las versiones estándar por la presencia de un grupo hidroxilo en el C2 (2‐OH) de  un residuo de acilo esterificado. Similarmente, un grupo miristato esterificado al  lípido A  de  Salmonella  typhimurium puede  tener una modificación  2‐hidroxilo que  se debe  a  la  enzima  LpxO.  Nosotros  postulamos  que  la  2‐hidroxilación  de  los  2‐OH‐OLs  podría  ser  catalizada por una nueva dioxigenasa homóloga a LpxO. En B. cenocepacia,  identificamos  dos  marcos  abiertos  de  lectura  (BCAM1214  and  BCAM2401)  homólogos  a  LpxO  de  S.  typhimurium.  La  introducción  de  bcam2401  (nombrado  olsD)  en  Sinorhizobium meliloti  provocó  la  formación  de  un  nuevo  lípido  y  en  B.  cenocepacia  de  dos  nuevos  lípidos,  Sorprendentemente, la modificación de los OLs debida a OlsD ocurre en la cadena de acilo  unida a la amida. Este es el primer reporte de una modificación hidroxilo de los OLs en el  residuo de acilo unido a la amida. La formación de los OLs hidroxilados ocurre solamente  cuando  la  ruta  de  biosíntesis  para  OLs  estándar  no  modificados  está  intacta.  La  modificación hidroxilo de  los OLs en el  residuo acilo unido a  la amida ocurre solamente  bajo condiciones de estrés ácido. Un ensayo ha sido desarrollado para la dioxigenasa OlsD,  y una caracterización inicial de la enzima es presentada.      94 Published: June 27, 2011 r 2011 American Chemical Society 6396 dx.doi.org/10.1021/bi200706v | Biochemistry 2011, 50, 6396–6408 ARTICLE pubs.acs.org/biochemistry The Dioxygenase-Encoding olsD Gene from Burkholderia cenocepacia Causes the Hydroxylation of the Amide-Linked Fatty Acyl Moiety of Ornithine-Containing Membrane Lipids Napoleon Gonzalez-Silva,† Isabel M. Lopez-Lara,† Rodrigo Reyes-Lamothe,† Adrian M. Taylor,‡ David Sumpton,‡ Jane Thomas-Oates,‡ and Otto Geiger*,† †Centro de Ciencias Genomicas, UniversidadNacional Autonoma deMexico, Apdo. Postal 565-A, Cuernavaca,Morelos CP62251,Mexico ‡Department of Chemistry, University of York, Heslington, York YO10 5DD, United Kingdom bS Supporting Information Burkholderia cenocepacia J2315 is a highly virulent member of the Burkholderia cepacia complex (BCC), a subgroup of important opportunistic pathogens of the Burkholderia genus that infects individuals with cystic fibrosis and chronic granulo- matous disease or the immunocompromised, and causes high mortality rates.1 Although a number of factors that contribute to BCC virulence are known, often BCC infections are not effi- ciently eliminated by treatment with common antibiotics.1 Among the most striking features of the Burkholderia genus are the polar lipids present in their membranes. B. cenocepacia J2315, like other members of its genus, possesses phosphatidylglycerol (PG), cardiolipin (CL), phosphatidylethanolamine (PE), and ornithine-containing lipids (OLs), as well as the 2-hydroxylated derivatives of PE and OLs (2-OH-PE and 2-OH-OLs, re- spectively), which differ from the standard versions by the presence a hydroxyl group at C2 (2-OH) of a fatty acyl residue.2,3 The standard OLs possess an ornithine residue, the R-amino group of which is amide-bound to a 3-hydroxy fatty acyl group. This amide-linked moiety is esterified through its hydroxyl to another nonhydroxylated fatty acyl residue.4,5 The biosynthesis pathway for such nonhydroxylated OLs has been resolved and consists of two steps. In the first step, theN-acyltransferase OlsB catalyzes the transfer of a 3-hydroxy fatty acyl group from 3-hydroxy fatty acylacyl carrier protein to the R-amino group of ornithine, forming lyso-ornithine lipid.6 In the second step, the O-acyltransferase OlsA catalyzes the transfer of a fatty acyl group from fatty acylacyl carrier protein to the hydroxy group of lyso- ornithine lipid, forming OL.7 In 2-OH-PE from B. cenocepacia J2315, the 2-hydroxy fatty acyl residue is exclusively linked to the sn-2 position of this glycerolipid, whereas in 2-OH-OLs, the 2-hydroxy fatty acyl chain is the ester- linked residue.2,3 The 2-hydroxy fatty acyl residues are not formed Received: May 6, 2011 Revised: June 24, 2011 ABSTRACT: Burkholderia cenocepacia is an important opportunistic pathogen, and one of the most striking features of the Burkholderia genus is the collection of polar lipids present in its membrane, including phosphatidylethanolamine (PE) and ornithine-containing lipids (OLs), as well as the 2-hydroxylated derivatives of PE and OLs (2-OH-PE and 2-OH-OLs, respectively), which differ from the standard versions by virtue of the presence of a hydroxyl group at C2 (2-OH) of an esterified fatty acyl residue. Similarly, a lipid A-esterified myristoyl group from Salmonella typhimurium can have a 2-hydroxy modification that is due to the LpxO enzyme. We thus postulated that 2-hydroxylation of 2-OH- OLs might be catalyzed by a novel dioxygenase homologue of LpxO. In B. cenocepacia, we have now identified two open reading frames (BCAM1214 and BCAM2401) homologous to LpxO from S. typhimur- ium. The introduction of bcam2401 (designated olsD) into Sinorhizo- biummeliloti leads to the formation of one new lipid and in B. cenocepacia of two new lipids. Surprisingly, the lipid modifications on OLs due to OlsD occur on the amide-linked fatty acyl chain. This is the first report of a hydroxyl modification of OLs on the amide-linked fatty acyl moiety. Formation of hydroxylated OLs occurs only when the biosynthesis pathway for nonmodified standard OLs is intact. The hydroxyl modification of OLs on the amide-linked fatty acyl moiety occurs only under acid stress conditions. An assay has been developed for the OlsD dioxygenase, and an initial characterization of the enzyme is presented. 95 6397 dx.doi.org/10.1021/bi200706v |Biochemistry 2011, 50, 6396–6408 Biochemistry ARTICLE during standard fatty acid biosynthesis, and specific enzymatic activities are required for the introduction of a hydroxyl group onto C2 of a fatty acyl residue. In Salmonella enterica serovar Typhimurium (S. typhimurium), a gene (lpxO) has been identified as being responsible for the introductionof a 2-hydroxy grouponto the lipidA-esterifiedmyristoyl group.8 The hydroxylation reaction is catalyzed by the Fe2+/O2/ R-ketoglutarate-dependent LpxO-encoded dioxygenase.9 It has been suggested that such extra hydroxyl groupsmight increase the extent of hydrogen bonding between adjacent lipidAmolecules, enhancing the outer membrane’s ability to resist penetration by certain compounds under some growth conditions.9 A homologue of LpxO, OlsC from Rhizobium tropici, is responsible for the formation of two of the four classes of OLs encountered in this organism.10 Recently, it has been shown that the expression of OlsC modifies OLs by hydroxylation at the 2 position of the esterified fatty acyl residue.11 As the functional roles associated with the 2-OHmodifications of PE andOLs are not known in B. cenocepacia, we embarked on a Table 1. Bacterial Strains and Plasmids Used in This Study strain or plasmid relevant characteristics ref or source S. meliloti 1021 SU47 str-21 12 S. meliloti AAK1 olsB::kan 6 S. meliloti CS111 pssA::gm 13 B. cenocepacia J2315 wild type LMG Bacteria Collection B. cenocepacia AQ3 olsD::cat this work B. cenocepacia NG1 olsB::cat this work E. coli DH5R recA1, Φ80 lacZΔM15 14 S17-1 modified RP4 plasmid integrated into the genome 15 pBluescriptSK+ cloning vector, CbR Stratagene pUC18 cloning vector, CbR 16 pRK404 broad host range vector, TcR 17 pBBR1MCS broad host range vector, CmR 18 pBBR1MCS-5 broad host range vector, GmR 19 pLysS production of lysozyme for repression of T7 polymerase, CmR 20 pET9a expression vector, KnR 20 pET16b expression vector, CbR 20 pET17b expression vector, ApR 21 pK18mobsacB suicide vector, KnR 22 pJG16 BamHI-restricted pET9a in pRK404 6 pJG20 olsBSm in pET9a 6 pILAS03 olsBSm in pBBR1MCS-5 this work pCAT pUC18 containing chloramphenicol resistance gene this work pSphx01 bcam1214 in pET9a this work pSphx02 olsD in pET9a this work pSphx03 BglII-restricted pSphx01 in pRK404 this work pSphx04 BglII-restricted pSphx02 in pRK404 this work pRRL01 1.1 kb downstream of olsD in pBluescriptSK+ this work pSphx06 flanking regions of olsD interrupted by a chloramphenicol resistance gene in pBluescriptSK+ this work pSphx08 flanking regions of olsD interrupted by a chloramphenicol resistance gene in pK18mobsacB this work pNG10 1.1 kb upstream of olsBBc in pBluescriptSK+ this work pNG11 1.2 kb downstream of olsBBc in pBluescriptSK+ this work pNG14 flanking regions of olsBBc in pBluescriptSK+ this work pNG15 flanking regions of olsBBc interrupted by a chloramphenicol resistance gene in pBluescriptSK+ this work pNG16 flanking regions of olsBBc interrupted by a chloramphenicol resistance gene in pK18mobsacB this work pNG23 olsBBc in pET17b 11 pNG24 BglII-restricted pNG23 in pRK404 this work pNG28 BamHI-restricted pET17b in pRK404 this work pNG40 olsD in pET16b this work 96 6398 dx.doi.org/10.1021/bi200706v |Biochemistry 2011, 50, 6396–6408 Biochemistry ARTICLE genomic approach in an attempt to identify potential genes that might be responsible for the introduction of the 2-OH modifica- tions. The B. cenocepacia J2315 genome contains two open reading frames (ORFs) [BCAM1214 and BCAM2401 (OlsD)] homologous to LpxO from S. typhimurium, which were cloned and introduced into Sinorhizobium meliloti 1021 and B. cenoce- pacia J2315. The introduction of olsD into S. meliloti leads to the formation of one new lipid and the introduction into B. cenoce- pacia to the formation of two new lipids. Mass spectrometric data suggest that OlsD can modify preexisting OLs of S. meliloti and B. cenocepacia by introducing a hydroxyl group. Surprisingly, the OlsD-derived modification does not occur on the ester-linked fatty acyl chain but on the amide-linked fatty acyl chain. The olsD gene is part of a biosynthetic pathway for OLs not previously described for any species. ’EXPERIMENTAL PROCEDURES Bacterial Strains, Plasmids, and Growth Conditions. All bacterial strains and plasmids used and their relevant character- istics are listed in Table 1.1122 Construction of burkholderial mutants is described in the Supporting Information. S. meliloti strains were grown either on complex tryptone/yeast extract (TY) medium that contained 4.5 mMCaCl2 23 or onminimal medium24 with succinate (8.3 mM) replacing mannitol as the carbon source at 29 C on a gyratory shaker. Growth in minimal medium that contained low concentrations (0.02 mM) of inorganic phosphate (Pi) was performed as previously described. 7 B. cenocepacia strains were grown at 29 C and Escherichia coli strains at 37 C in Luria- Bertani (LB)medium25on a gyratory shaker. To study the effect of acidic stress, B. cenocepaciawas grown in complex LBmedium that contained additionally a 50 mMHomopipes [homopiperazine-N, N0-bis(2-ethanesulfonic acid)]/NaOH buffer (pH 4) instead of the unbuffered, nearly neutral LBmedium. Antibiotics were added to the medium at the following concentrations when required: 70 μg/mL gentamicin, 40 μg/mL piperacillin, and 4 μg/mL tetracycline in the case of S. meliloti; 100 μg/mL carbenicillin, 20 μg/mL tetracycline, 10 μg/mL gentamicin, 50 μg/mL kana- mycin, and 20 μg/mL chloramphenicol in the case of E. coli; and 500 μg/mL tetracycline (for solid medium) or 300 μg/mL tetra- cycline (for liquid medium) and 300 μg/mL chloramphenicol for B. cenocepacia J2315. The pRK404-, pBBR1MCS-5-, or suicide-plasmid derivatives were mobilized into S. meliloti or B. cenocepacia strains by diparental mating using E. coli S17-1.15 DNA Manipulations. Recombinant DNA techniques were performed according to standard protocols.26 In all polymerase chain reactions, the XL-PCR kit from Applied Biosystems was used. To obtain plasmid pILAS03, the 987 bp BglIIBamHI fragment containing olsB of S. meliloti 1021 (olsBSm) and the regulating region was obtained from pJG206 and ligated into pBBR1MCS-519 that had been linearized with BamHI. Cloning of burkholderial genes is described in the Supporting Information. In Vivo Labeling of Bacterial Strains with [14C]Acetate, [14C]Ornithine, or 32Pi. The lipid compositions of S. meliloti 1021 and B. cenocepacia J2315 derivatives were analyzed follow- ing labeling with [14C]acetate, [14C]ornithine, or 32Pi. Cultures (1 mL) in minimal or complex medium were inoculated from precultures grown in the samemedium. After addition of 2 μCi of [1-14C]acetate (60 mCi/mmol), 1 μCi of DL-[1-14C]ornithine (56 mCi/mmol), or 1.3 μCi of 32Pi to each culture, the cultures were incubated for 24 h (on low Pi-containing minimal medium) or 4 h (on TY medium) in the case of S. meliloti and for 12 h in the case of B. cenocepacia. Cells were harvested by centrifugation and resuspended in 100 μL of water. The lipids were extracted using the method of Bligh and Dyer.27 The chloroform phase was used for lipid analysis on TLC plates after two-dimensional5 separation. The individual lipids were quantified as described previously28 or by using a Phosphor-Imager (Storm 820, Mol- ecular Dynamics). A chloroform/methanol/water mixture (130:50:8, v/v/v) was used as the solvent system for develop- ment in one dimension. Ninhydrin staining of lipids was per- formed as described previously.7 Preparation of the Radiolabeled OL Substrate for OlsD Enzyme Assays. PE-deficient strain S. meliloti CS111 pNG24 expressing OlsB from B. cenocepacia was cultivated in 100 mL of complex medium in the presence of 100 μCi of [1-14C]acetate (60 mCi/mmol) and harvested at an OD620 of 1. Cells were extracted using the method of Bligh and Dyer,27 and the chloro- form phase was separated by one-dimensional TLC using a chloroform/methanol/acetic acid mixture (130:50:20, v/v/v). Radiolabeled OL was localized by autoradiography and extracted from the OL-containing silica gel fraction, and the OL stock was stored in a chloroform/methanol mixture (1:1, v/v). Preparation of Cell-Free Extracts for Analysis of the OlsD Dioxygenase. Cultures (1 L) of exponentially growing E. coli BL21(DE3)  pLysS, harboring additionally pET16b or pNG40, were induced with 1 mM isopropyl β-D-thiogalactoside at a density of 5 108 cells/mL, and cells were incubated for an additional 3 h at 37 C. After cells had been harvested by low- speed centrifugation at 4 C, each cell pellet was washed once with 50 mM HEPES/KOH buffer (pH 7.5), and cells were resuspended in 3 mL of the same buffer. Cell suspensions were passed three times through a French pressure cell at 20000 lb/in.2 Unbroken cells and cell debris were removed by centrifugation at 4000g for 10 min, yielding cell-free extracts as supernatants. For some assays, cell-free extracts were centrifuged at 150000g for 1 h to separate soluble proteins and membranes with their associated proteins. Protein concentrations were determined by the bicinchoninic acid assay.29 In Vitro Assay for theOlsDDioxygenase.An in vitro assay for OlsD, usingOL as the putative acceptor substrate, was developed on the basis of the assay previously reported for the Fe2+/O2/ R-ketoglutarate-dependent LpxO hydroxylase.9 The reaction con- ditions, unless otherwise described, included 50mMHEPES/KOH buffer (pH 8.0), 1 mM R-ketoglutarate, 2 mM ascorbate, 10 μM Fe(NH4)2(SO4)2, 0.1%Triton X-100, 4mMdithiothreitol (DTT), and 169μMOL(25,960 cpm/reaction or 3072 cpm/nmol). Assays were conducted at 30 C in a final volume of 50 μL. For each assay, first the respective amount of Triton X-100 in water was mixed with radiolabeled OL in a chloroform/methanol mixture (1:1, v/v) and brought to dryness in an Eppendorf concentrator. Then, buffer and the remaining components were added; reactions were initiated via addition of E. coli crude extracts (final concentration usually of 1 mg/mL), and they were stopped via addition of 125 μL of methanol and 62.5μL of chloroform. After lipids had been extracted using the method of Bligh and Dyer,27 the chloroform phase was separated by one-dimensional TLC with a chloroform/methanol/ acetic acid mixture (130:50:20, v/v/v). In this chromatog- raphic system, the unmodified OL exhibited an Rf value of 0.37 whereas the OlsD-modified, hydroxylated OL exhibited an Rf value of 0.29. Preparation of Lipids forMass Spectrometric Analysis. For spectrometric analysis of lipids from S. meliloti strains carrying 97 6399 dx.doi.org/10.1021/bi200706v |Biochemistry 2011, 50, 6396–6408 Biochemistry ARTICLE pRK404 derivatives, after two passages in Sherwood minimal medium with a low phosphate concentration, 500 mL cultures were grown in the same medium until they reached the late exponential phase (OD620 = 0.570). Cells were harvested by centrifugation, and the lipids were extracted using the method of Bligh and Dyer.27 The chloroform phase was analyzed by mass spectrometry. Cultures of 1 L of B. cenocepacia were grown to an OD620 of 1. Cells were harvested by centrifugation, and the lipids were extracted using the method of Bligh and Dyer.27 The chloroform phase was either directly analyzed by mass spectro- metry or further fractionated as follows. Concentrated lipid preparations of B. cenocepacia strains were separated on high- performance TLC silica 60 plates by two-dimensional TLC.5 After the samples had been stained with iodine, the individual iodine-stained spots were scraped off, and lipids were repeatedly extracted from the silica gel as previously described28 and analyzed by mass spectrometry. Mass Spectrometric Analysis. The S. meliloti lipid extracts were analyzed as follows. ESI mass spectra were recorded on an Applied Biosystems (Warrington, U.K.) QSTAR pulsar i hybrid quadrupole time-of-flight tandem mass spectrometer. The ion source contained a microelectrospray arm mounted in an Applied Biosystems ion source housing. A 100 μLHamilton 1710N syringe with a 1.46 mm inside diameter was fitted to a fused silica capillary, which was used to deliver the sample, dissolved in methanol (Fischer Scientific, Loughborough, U.K.), to the ion source at a rate of 1 μL/min controlled by the integral syringe driver. The ion source gas reading was set to 4 and the curtain gas to 20, and the capillary was held at 5500 V. The instrument was operated in the positive mode with declustering and focusing potentials of 65 and 265 V, respectively. For Pulsar operation, the ion release delay was set to 6 and the ion releasewidth to 5. InMSmode, the collision gas was set to read 3 and increased to 6 during tandem MS. The “collision energy offset” for tandemMSwas varied between 25 and 70 V. Nitrogen was used for both the collision gas and the curtain gas. The data were recorded and analyzed using Applied Biosys- tems/MDS SCIEX Analyst QS. The B. cenocepacia lipid extracts were analyzed as follows. ES CID tandemmass spectra were recorded on anApplied Biosystems QSTAR pulsar i hybrid quadrupole time-of-flight tandem mass spectrometer with a nanoflow electrospray ion source. The sup- plied sample solutions were diluted in methanol (1:10, v/v) and were delivered to the ion source using a fused silica continuous flow sample introduction system and a syringe pump that delivered the sample solution at a rate of 0.2 μL/min. When the instrument was operated in positive mode, the ion spray voltage was set to 4800 V, declustering potential 1 to 50 V, declustering potential 2 to 10 V, and the focusing potential to 220V. In negativemode, the ion spray voltage was set to 4800 V, declustering potential 1 to 50 V, declustering potential 2 to 10 V, and the focusing potential to220 V. Ionization was assisted via the application of curtain gas (nitrogen) set at 25, and spectra were recorded over the m/z 1001000 range. Tandem mass spectra were recorded over the m/z 50800 range depending on the chosen precursor ion; nitrogen was used as the collision gas, and the spectra were recorded using “collision offsets” of 1045 “V”. Both MS and MS/MS data were recorded and processed using Analyst. ’RESULTS OlsD of B. cenocepacia, a Homologue of LpxO of S. typhimurium, Can Modify OL of S. meliloti.We have searched for homologues of LpxO from S. typhimurium (GenBank entry AAF87784) using the BLAST server of the genome sequencing project of B. cenocepacia J2315 at http://www.sanger.ac.uk/cgi- bin/blast/submitblast/b_cenocepacia. Searching with the LpxO protein (302 amino acids) in the B. cenocepacia genome, we found two homologues; BCAM1214 (299 amino acids) that is 57% identical and 72% similar to S. typhimuriumLpxO (E value = 1.7 1094) and BCAM2401 (OlsD, 249 amino acids) that is 28% identical and 46% similar in an overlap of 180 amino acids to S. typhimurium LpxO (E = 1.7 1018). The ORFs for these two proteins are located on chromosome 2, and in the recently published genome of B. cenocepacia J2315,30 putative β-hydro- xylase functions were suggested for both proteins. To determine whether any of these proteins is able tomodify OL and/or PE, the ORFs encoding each of the proteins were amplified from genomic DNA of B. cenocepacia, and each was cloned in a broad host-range vector (see Experimental Procedures) and designated pSphx03 for that carrying bcam1214 and pSphx04 for that carrying olsD. These vectors, as well as empty vector pJG16,6 were individually transferred into S. meliloti 1021 that produces only nonhydroxy- lated forms of OL and PE.5 The three different strains were grown in Sherwood minimal medium at phosphate-limiting concentrations (20 μM Pi) in the presence of [14C]acetate, and lipids were extracted and analyzed using one-dimensional thin-layer chromatography (TLC). Introduction of plasmid pSphx03 into S. meliloti gives the same lipid profile found in S. meliloti carrying empty vector pJG16 (Figure 1A, lanes 2 and 3). However, the introduction of pSphx04 leads to the formation of a new lipid (Figure 1A, lane 4). The new lipid (NL) reacts positively when stained with ninhydrin (data not shown), indicating that it is an amine-containing compound, but NL does Figure 1. Expression of BCAM2401 (OlsD) in S. meliloti causes the formation of a new ornithine-containing lipid (NL). (A) Lipid analysis of S. meliloti strains carrying different plasmids after growth on Sherwood minimal medium containing a low phosphate concentration and grown either in the presence of [14C]acetate (lanes 14) or in the presence of [14C]ornithine (lanes 5 and 6): lane 1, AKK1 pJG16, a strain deficient in OL production;6 lanes 2 and 5, S. meliloti 1021  pJG16; lane 3, S. meliloti 1021  pSphx03, expressing bcam1214; lanes 4 and 6, S. meliloti 1021 pSphx04, expressing bcam2401. (B) Separation of [14C]acetate- labeled lipids by two-dimensional TLC from S. meliloti harboring pILAS03 and pSphx04 (expressing olsBSm and bcam2401) after growth on complex TY medium. The lipids PC, dimethylphosphatidylethano- lamine (DMPE), PE, PG, CL, diacylglyceryl N,N,N-trimethylhomoser- ine (DGTS), OL, and new lipid (NL) are indicated. 98 6400 dx.doi.org/10.1021/bi200706v |Biochemistry 2011, 50, 6396–6408 Biochemistry ARTICLE not incorporate 32P-containing inorganic phosphate (Pi), sug- gesting that it might lack phosphorus in its structure. Lipid extracts obtained from S. meliloti 1021  pJG16 grown in the presence of [14C]ornithine show the specific incorporation of radioactivity into OL,6 and only the compound corresponding to OL can be detected6 (Figure 1A, lane 5). In contrast, lipid extracts obtained from S. meliloti 1021  pSphx04 contain a second lipidic compound into which [14C]ornithine is specifi- cally incorporated (Figure 1A, lane 6). Taken together, these data suggest that the new compound, due to olsD, was amodified form of the OL described previously.5 Previously, we have shown that in S. meliloti the biosynthesis of OL is regulated at the level of olsB gene expression because the introduction of olsB in a stable plasmid into S. meliloti leads to constitutive high-level production of OL, even inmedia with high concentrations of phosphate.6 Now, we have constructed a new broad host-range plasmid, pILAS03, that carries olsB from S. meliloti (olsBSm) and is compatible with pSphx04 (see Experi- mental Procedures). As in the case of pJG21,6 introduction of pILAS03 into S. meliloti provokes the formation of high levels of OL even when the S. meliloti is grown on TY medium (data not shown). When S. meliloti carries both plasmids (pSphx04 and pILAS03), the new lipid compound (NL) dependent on olsD is formed even in cultures grown on TY medium (Figure 1B). OlsB Is Required for the Formation of Unmodified OL and 2-OH-OL in B. cenocepacia. B. cenocepacia J2315 grown on LB medium forms four ninhydrin-positive spots (data not shown and Figure 2A) that were interpreted as PE, OL, and the respective 2-hydroxylated derivatives (2-OH-PE and 2-OH- OL), as described previously for members of the Burkholderia genus.2,3 Introduction of an empty broad host-range vector (pJG16) does not change the lipid profile of wild-type B. cenocepacia (Figure 2B). To determine if formation of standard OLs was essential for the biosynthesis of modified OLs, a mutant Figure 2. Membrane lipid profiles of B. cenocepacia strains defective with respect to OL biosynthesis genes (olsB or olsD) or overexpressing them. Separation of [14C]acetate-labeled lipids by two-dimensionalTLC fromwild-typeB. cenocepacia J2315 (A), wild-typeB. cenocepacia J2315 containing empty vector pJG16 (B), the olsB-deficient knockout mutant B. cenocepacia NG1 that lacks OLs (C), the olsB-deficient knockout mutant with olsB-expressing plasmid pNG24 (D), the olsD-deficient knockoutmutant B. cenocepaciaAQ3 (E), and wild-type B. cenocepacia J2315 with olsD-expressing plasmid pSphx04 (F), after growth on LB medium. The lipids PE, 2-OH-PE, OL, 2-OH-OL, PG, and CL and the two new lipids (NL1 and NL2) are indicated. 99 6401 dx.doi.org/10.1021/bi200706v |Biochemistry 2011, 50, 6396–6408 Biochemistry ARTICLE of B. cenocepacia in the olsB gene was created and named NG1. We have searched for the homologue of the OlsB product from S. meliloti (OlsBSm) (GenBank entry NP_384499) using the BLAST server of the genome sequencing project of B. cenocoe- pacia J2315 at http://www.sanger.ac.uk/cgi-bin/blast/submit- blast/b_cenocepacia. Searching with the OlsBSm protein (296 amino acids) in the B. cenocepacia genome, we found a homo- logue, BCAL1281 (OlsBBc, 268 amino acids), that is 33% identical and 50% similar to OlsBSm (E = 6.7  1031). The olsB-deficient knockout mutant B. cenocepacia NG1 does not synthesize any form of OL (Figure 2C). Quantitative analysis of individual burkholderial lipids shows that when OLs are absent, the level of 2-OH-PE increases ∼2.5-fold to comprise 16% of total membrane lipids (Table 2). When we complemented the mutant NG1 in trans with plasmid pNG24 that contains the olsBBc gene, the levels of OL and 2-OH-OL increase dramatically, comprising 47 and 18% of total membrane lipids, respectively (Figure 2A,D and Table 2). In this OL-overproducing strain, levels of PE and 2-OH-PE are drastically reduced to 12 and 2% of the total level of membrane lipids, respectively (Table 2). It seems that the cells try to keep the amount of zwitterionic lipid in balance. This effect was not observed in the mutant containing empty vector pNG28 (Table 2). Mutation of olsD Does Not Lead to Any Change in the Membrane Lipid Profile of B. cenocepacia under Standard Growth Conditions; However, Its Expression Does. To study the function of olsD in B. cenocepacia, a knockout mutant in olsD was created and was named AQ3. Surprisingly, analysis via two- dimensional TLC (2D-TLC) of [14C]acetate-labeled lipids from the wild type and the AQ3 mutant after growth on LB medium does not show any detectable difference between them (Figure 2A,E and Table 2), and notably, olsD-deficient mutant AQ3 still forms the previously described 2-hydroxylated deriva- tives of PE and OL. Also, the B. cenocepacia strain that contained the bcam1214 gene showed a lipid profile similar to that of the wild type (data not shown). However, the constitutive expression of olsD in the AQ3 (Table 2) andB. cenocepacia J2315 (Figure 2F) backgrounds led to the formation of two new compounds in each strain, named NL1 and NL2, that could not be detected in a wild- type strain harboring an empty plasmid (Figure 2B and Table 2). The lipidsNL1 andNL2, like NL, stained positive with ninhydrin. Therefore, under our standard experimental conditions, the products dependent on olsD were not detected in wild-type B. cenocepacia. Spot NL1 produced in B. cenocepacia (Figure 2F) shows the same mobility in 2D-TLC as spot NL observed in S. meliloti after expression of olsD and olsB (Figure 1B). Quantitative analysis of the individual burkholderial lipids (Table 2) indicates that even if OlsD is expressed in B. cenocepacia  pSphx04, NL1 and NL2 are only minor lipids, each of them comprising slightly more than 1% of the total membrane lipids. Remarkably, the strain forming NL1 and NL2 contains reduced amounts of 2-OH-OL as if the organism tries tomaintain a constant level ofOLhydroxylation for a given physiological condition. When we expressed the olsD gene in OL-deficient mutant NG1, a membrane lipid profile similar to that of the mutant that contains empty vector pJG16 was observed (data not shown). Our results show that the formation of standard OLs is required for the formation of OLs hydroxylated on the external, ester-linked fatty acyl residue (2-OH-OL) as well as for the formation ofOlsD-derivedOLs in which the internal, amide- linked fatty acyl residue bears an extra hydroxyl group (N-acyl-OH- OL) as we show later. Modification of OLs by OlsD Occurs under Acidic Condi- tions. Distinct environmental stresses can provoke hydroxyla- tions of standard membrane lipids.3,8,11,31 An increase in the growth temperature of B. cenocepacia J2315 over the range of 2942 C led to increased proportions of 2-OH-OL and 2-OH-PE Table 2. Membrane Lipid Composition of Wild-Type B. cenocepacia J2315, olsD- or olsB-Deficient Mutants (AQ3 or NG1), the olsD-Deficient Mutant Complemented with the olsD Gene (AQ3 pSphx04) or Containing an Empty Broad Host-Range Vector (AQ3  pJG16), and the olsB-Deficient Mutant Complemented with the olsB Gene (NG1  pNG24) or Containing an Empty Broad Host-Range Vector (NG1  pNG28) after Growth on Complex LB Mediuma composition (% of total 14C) lipid wild type AQ3 AQ3  pSphx04 AQ3  pJG16 NG1 NG1  pNG24 NG1  pNG28 PG 12.2 ( 0.3 14.1 ( 1.9 15.0 ( 0.6 11.8 ( 0.9 11.2 ( 0.6 10.4 ( 0.8 14.2 ( 0.8 CL 9.9 ( 1.1 11.9 ( 1.9 10.2 ( 1.1 11.4 ( 1.2 12.6 ( 1.9 11.3 ( 1.6 11.1 ( 1.3 PE 64.9 ( 2.1 56.0 ( 1.1 61.2 ( 1.8 59.5 ( 0.5 60.3 ( 5.3 11.7 ( 0.9 61.5 ( 2.8 2-OH-PE 6.5 ( 1.2 11.4 ( 0.3 7.6 ( 1.2 9.2 ( 0.7 16.0 ( 6.6 1.8 ( 0.2 13.3 ( 2.0 OL 2.1 ( 0.1 1.8 ( 0.4 1.8 ( 0.4 2.3 ( 0.9 ndb 47.2 ( 4.6 ndb 2-OH-OL 4.4 ( 0.7 4.8 ( 0.7 1.9 ( 0.3 5.9 ( 0.4 ndb 17.8 ( 2.7 ndb NL1 ndb ndb 1.3 ( 0.1 ndb ndb ndb ndb NL2 ndb ndb 1.1 ( 0.3 ndb ndb ndb ndb aThe values shown are means ( the standard deviation derived from three independent experiments. bNot detected. Figure 3. OlsD-dependent modification of OLs in B. cenocepacia occurs under acidic growth conditions. Separation of [14C]acetate-labeled lipids by 2D-TLC from wild-type B. cenocepacia J2315 (A) or olsD- deficient knockoutmutant B. cenocepaciaAQ3 (B) after growth on acidic (pH 4) Homopipes/Na-buffered LB medium. The lipids PE, 2-OH-PE, OL, 2-OH-OL, PG, and CL and the two new lipids (NL1 and NL2) are indicated. 100 6402 dx.doi.org/10.1021/bi200706v |Biochemistry 2011, 50, 6396–6408 Biochemistry ARTICLE (data not shown), similar to those reported previously for B. cepacia NCTC 10661.3 When B. cenocepacia J2315 was grown in complex medium at pH 4 instead of in unbuffered LBmedium at neutral pH, two new minor compounds were formed that migrated like NL1 and NL2 on 2D-TLC (Figure 3A). In contrast, neither NL1 nor NL2 was formed when OlsD-deficient mutant AQ3 was grown in complex medium at pH 4 (Figure 3B). Therefore, in B. cenocepacia J2315, the formation of NL1 and NL2 under acidic conditions depends on an intact olsD gene, withNL1 andNL2 each amounting to ∼1% of the total membrane lipids (data not shown). OlsD in B. cenocepacia Produces OH-PE and OH-OL; PG and CL Are Not Hydroxylated. Individual lipids formed by B. cenocepacia J2315  pSphx04 were separated, extracted, and concentrated from preparative 2D-TLC plates (see Experimental Procedures), and fractions corresponding to lipids PE, 2-OH-PE, PG, CL, OL, and NL1, the mixture of 2-OH-OL and NL2, and the complete lipid extracts of some burkholderial or sinorhizobial strains were examined using mass spectrometry. Although the acyl chain substitutions of B. cenocepacia PG and CL were found to be similar to those of PE, clearly 2-OH-PE was substituted with hydroxylated fatty acyl chains at the sn-2 position (see Figure S1 of the Supporting Information). New MS Fragmentation of OLs Aids in Structural Assign- ment. The positive ESI mass spectrum of lipid spot OL (Figure S2A of the Supporting Information) contains ions for at least five distinct species (M +H+ atm/z 651, 665, 667, 625, and 615). The tandemmass spectrum of OL with an ion atm/z 651 (Figure 4A) contains fragment ions in the low-m/z region consistent with an OL lipid (m/z 115 = ornithine b1 ion, m/z 133 = ornithine protonatedmolecule,m/z 70 = ornithine immonium ionNH3). The fragment ions at m/z 369 and 387 correspond to the elimination and direct cleavage, respectively, of a C18:1 acyl chain, with m/z 351 corresponding to the elimination product minus H2O (Figure 4B). These data are characteristic of an N-acyl-OL in which a C16:0 hydroxy fatty acid is amide-bound to ornithine, which is ester-linked to a C18:1 fatty acid. Upon interpretation of a large number of different OL tandem mass spectra, an additional, characteristic, fragmentation of the amide-bound hydroxy fatty acid residue was consistently ob- served (regardless of the acyl chain structures) that has to the best of our knowledge not been described before. This ion is proposed to be formed following elimination of the esterified fatty acid side chain; this results in a new double bond within the amide-bound chain (Figure 4B). Upon further fragmentation (indicated in Figure 4B by the red arrow), this now unsaturated chain is lost from the ornithine as an acylium cation, producing a fragment ion that now has a double bond marking the position of the hydroxyl group of the original residue (Figure 4B). For example, in the fragmentation spectrum of OLwith an ion atm/z 651 (Figure 4A) (assigned a C16:0 hydroxyacyl amide-bound moiety on the basis of the other fragment ions observed), an ion at m/z 237 that is formed from the cleavage of the original amide- bound C16:0 hydroxy fatty acid as a C16:1 acylium cation is observed. This new fragmentation pathway has proved to be invaluable for supporting the assignment of OL structures from their product ion spectra, especially because many of the OL product ion spectra are more complicated than that of the ion at m/z 651, due to the presence of a mixture of isobaric species. In this way, the results of the tandem MS analyses of the OL lipids with M + H+ ions atm/z 651, 665, 667, 625, and 615 suggest that in all five lipids, a C16:0 hydroxy fatty acyl is the amide-linked chain, while the ester-linked chains are C18:1 (for m/z 651), C19:1 (form/z 665), C19:0 (m/z 667) (Figure 5A), C16:0 (m/z 625), and C13:0 (m/z 615), which seems to be anOLmethanol adduct (data not shown). It is worth noting that this fragmenta- tion can also be observed in spectra of other workers,33 although in that publication its origin was not recognized because only one OL was described in detail, and the fatty acids on that structure made it impossible to distinguish; our study in which so many different structures were analyzed made it possible to identify this useful diagnostic ion. OlsD Is Responsible for Hydroxylating the Fatty Acyl Residue That Is Amide-Bound to Ornithine in B. cenocepa- cia.During the 2D-TLC separation of the total lipid extract, new lipid 1 (NL1) migrated in a manner analogous to that of the altered OL (NL) observed in the study of S. meliloti carrying the olsD gene. NL1 isolated via TLC was analyzed using positive mode ESI MS (Figure S2B of the Supporting Information) and product ion experiments; ions at m/z 631, 667, 681, and 683 were identified as OLs, each of which is 16Da higher inmass than the OLs identified from the OL spot extract (Figure S2A of the Figure 4. Positive ion mode product ion mass spectrum of OL lipid 1 (m/z 651) (A) and proposed fragmentation scheme for OL (B). The step shown with the red arrow shows the loss of the newly formed unsaturated internal fatty acyl residue from ornithine as an acylium cation, producing the proposed diagnostic fragment ion that has one more degree of unsaturation than the original internal fatty acid residue. The cyclic structures and protonation sites shown are consistent with similar OL positive ion structures presented in the literature.32,33 Peak intensities are normalized to the most intense fragment ion. 101 6403 dx.doi.org/10.1021/bi200706v |Biochemistry 2011, 50, 6396–6408 Biochemistry ARTICLE Supporting Information), consistent with hydroxylated versions of the OLs characterized previously. The tandem mass spectrum of m/z 683 (Figure 5B) contains fragment ions in the low-m/z region (m/z 70, 115, and 133) for an OL lipid and ions at m/z 385 and 403 for the elimination and direct cleavage of the ester- bound C19:0 acyl chain. The ion atm/z 235 corresponds to the mass of a C16:2 acylium cation derived from the amide-bound residue by elimination of both its OH substituents. These results identify an N-acyl-OL in which a C16:0(OH)2 dihy- droxy fatty acid is amide-bound to ornithine, which in turn is ester-linked via one of its hydroxyl groups to a C19:0 fatty acid. This result is consistent with an OL lipid with an M + H+ ion at m/z 683 being the hydroxylated version of the previously characterized OL lipid with an M + H+ ion at m/z 667, with the additional hydroxylation on the amide-linked hydroxy fatty acid residue (compare the product ion spectra in panels A and B of Figure 5). Similar analysis of the product ion spectrum of the ion at m/z 681 (not shown) showed that it corresponds to a hydroxylated version of OL lipid 2 (m/z 665), again with the hydroxylation on the internal fatty acid residue. The product ion spectrum of the ion at m/z 667 (OL lipid 11) is consistent with a mixture of two isobaric OLs: OL [16:0(OH)2/18:1] corresponding to the hydroxylated version of OL lipid 1 (m/z 651) and an OL that is not additionally hydroxylated [16:0(OH)/19:0]. The signal in the spectrum of the NL1 spot extract atm/z 631 (OL lipid 8) corresponds to the hydroxylated version of OL lipid 7 (m/z 615). OL lipids 7 and 8 correspond to methanol adducts, presumably formed in the electrospray interface of protonated species with ions at m/z 599 and 583. Fragmentation is con- sistent with the hydroxylation being present on the amide-bound hydroxy fatty acid (data not shown). OlsD Also Hydroxylates the OL Amide-Bound Fatty Acyl Residue in S. meliloti. MS analysis of lipids from S. meliloti Figure 5. Positive mode product ion mass spectra of the ions atm/z 667 from OL spot extract (A),m/z 683 from NL1 spot extract (B),m/z 683 from 2-OH/NL2 spot extract (C), and m/z 699 from 2-OH-OL/NL2 spot extract (D). Peak intensities are normalized to the most intense fragment ion. 102 6404 dx.doi.org/10.1021/bi200706v |Biochemistry 2011, 50, 6396–6408 Biochemistry ARTICLE 1021  pJG16 yielded an intense [M + H]+ ion at m/z 693, together with a less intense ion at m/z 679. The product ion spectrum of the ion at m/z 693 yielded an intense ion at m/z 397, and a much less intense ion at m/z 415 for elimination and direct cleavage of a C19:1 fatty acid, and ornithine-derived fragment ions were observed at m/z 70, 155, and 133. Product ion analysis of the ion atm/z 679 also yielded an ion atm/z 155 for ornithine (data not shown). These results are identical to those reported previously.5 The generation of ions by the facile loss of only a C19:1 fatty acid is consistent with an N-acyl-OL in which a C18 hydroxy fatty acid is amide-bound to ornithine, and to which a C19:1 fatty acid is esterified to the hydroxyl group of theN-C18 fatty acid.5Analogous ionswere observed in theCIDmass spectrumobtained from the ion at m/z 679, consistent with it bearing an ester-linked C18:1 fatty acid on an N-C18 hydroxyornithine. Comparison of the mass spectrometric data of S. meliloti  pSphx03 lipids with those of the lipid extract from the control strain (S. meliloti pGJ16) shows the presence of ions similar to those mentioned above. This suggests that the B. cenocepacia J2315 bcam1214 gene that was more similar to the LpxO dioxygenase from S. typhimurium does not cause the formation of any additional extractable lipid by the method used.27 In contrast, when we analyzed the mass spectra of lipids extracted from S. meliloti  pSphx04, we detected the presence of an additional ion at m/z 709 (Figure S3A of the Supporting Information), which was not detected in S. meliloti strains harboring plasmid pJG16 or pSphx03 (data not shown). The ion at m/z 709 (m/z 693 + 16) is consistent with a hydroxylated version of the ion at m/z 693. Product ion analysis of the ion at m/z 709 yielded ions at m/z 413 and 431 for elimination and direct cleavage of a C19:1 fatty acid (Figure S3B of the Support- ing Information). Significantly, this analysis suggests that it is again the amide-bound C18 fatty acyl chain that is additionally hydroxylated in OLs of S. meliloti and not the ester-linked C19:1 residue. Therefore, the site of OlsD-derived hydroxylation of S. meliloti OL is analogous to that in OLs extracted from B. cenocepacia carrying the olsD gene. OlsD in B. cenocepacia Also Causes Hydroxylation of the Amide-Linked Fatty Acyl Chain in the Presence of a Hydro- xylated Esterified Fatty Acid. It proved to be impossible to extract the 2-OH-OL and NL2 spots independently. The ESI mass spectrum of the extract of the combined 2-OH-OL and NL2 spots (Figure S2C of the Supporting Information) appears to be very similar to that of the NL1 extract, with signals observed at m/z 631, 667, 681, and 683. Product ion analysis of the ion at m/z 683 (OL lipid 17) (Figure 5C) was consistent with an OL in which a C16:0 hydroxy fatty acid is amide-bound to ornithine (confirmed by the presence of the C16:1 acylium ion), which in turn is ester-linked via its hydroxyl group to a C19:0 (OH)-fatty acid, confirming that OL 17 (m/z 683) is the hydroxylated version of the previously characterized OL lipid 3 (m/z 667). However, unlike lipid 11 (Figure 5B), which is additionally hydroxylated on the amide-bound hydroxy fatty acid, the extra hydroxyl group in the B. cenocepacia lipid is now positioned on the external ester-linked fatty acyl residue. Product ion analysis of the ion at m/z 681 (OL lipid 16) shows that it corresponds to hydroxylated OL 2 (m/z 665), again with the hydroxylation group on the ester-bound residue. Similarly, the ion at m/z 641 corresponds to hydroxylated OL 5 (m/z 625), again with the hydroxyl group on the external residue, although the hydroxy- lated amide-bound fatty acid version of OL 5 was not detected in the NL1 extract, probably because of the very small amount of this lipid (apparent from the low abundance of this signal in the spectra of the different extracts). The mass spectrometric data of OL lipid 15 (m/z 667) are consistent with it being hydroxylated OL lipid 1 (m/z 651) with the hydroxyl group on the ester-linked fatty acid residue. However, because an 18:1(OH) and a 19:0 fatty acyl residue are isobaric, it is impossible to distinguish, solely on the basis of these MS data, between these two possible structures for OL 15. However, OL 15 migrates on TLC with OLs with hydroxylated ester-linked fatty acids, and no unhy- droxylated species were detected in the 2-OH-OL/NL2 extract. Intriguingly, ions at m/z 697 (OL lipid 18) and m/z 699 (OL lipid 19) in the spectrum of the 2-OH-OL/NL2 extract from B. cenocepacia are 16 m/z units higher than those for OLs 16 (m/z 681) and 17 (m/z 683), respectively (Figure S2C of the Sup- porting Information), and 32 m/z units higher than OLs 2 (m/z 665) and 3 (m/z 667), respectively, in the OL extract. OLs 16 and 17 have been shown to be hydroxylated on their ester-linked fatty acids. The observation of “+16”OLs 16 and 17 suggests the existence of hydroxylated versions of OLs 16 and 17 (i.e., doubly hydroxylated forms of OLs 2 and 3, respectively). Product ion analysis of OL 19 (m/z 699) (Figure 5D) shows that the lipid carries additional hydroxyl groups on both the amide- linked [C16:0(OH)2] and ester-linked [C19:0(OH)] fatty acids. Similarly, OL 18 (m/z 697) was shown to be a hydroxylated version of OL 16 and a doubly hydroxylated version of lipid 2. OlsD Dioxygenase Modifies the Amidified Fatty Acyl Residue of OL in Vitro. To obtain more OlsD-modified OL for structural analysis, we developed an enzymatic assay for the OlsD activity. Different E. coli cell-free extracts were incubated with ornithine-containing lipid, and the products were analyzed. When [14C]OL was incubated with a cell-free extract of E. coli Figure 6. Membrane-associated OlsD dioxygenase converts OL to OL that is hydroxylated in the amidified acyl residue. Cell-free extracts of E. coli BL21(DE3) pLysS pNG40 expressing OlsD (lane 1), BL21(DE3) pLysS containing the empty pET16b vector (lane 4), the soluble protein fraction from E. coli BL21(DE3) pLysS pNG40 (lane 2), the membrane fraction from E. coli BL21(DE3) pLysS pNG40 (lane 3), and buffer only (lane 5) were incubated with 8.45 nmol of [14C]OL (3072 cpm/nmol) and other soluble cofactors for 2 h as described in Experimental Procedures. OlsD enzyme assays usually contained a final protein concentration of 1 mg/mL. After extraction, radiolabeled lipids were analyzed by one-dimensional TLC using a chloroform/methanol/acetic acid mixture (130:50:20, v/v/v) as a solvent system and detected and quantified with a phosphorimager. Unmodified OL and OL that is hydroxylated at the amidified acyl residue (N-acyl-OH-OL) are indicated. 103 6405 dx.doi.org/10.1021/bi200706v |Biochemistry 2011, 50, 6396–6408 Biochemistry ARTICLE harboring OlsD-expressing pNG40, the formation of a com- pound (N-acyl-OH-OL) that migrated like NL or NL1 was observed (Figure 6, lane 1).When [14C]OLwas incubated with a cell-free extract of E. coli harboring the empty pET16b plasmid, no formation of N-acyl-OH-OL was observed (Figure 6, lane 4), as when the incubation was performed with buffer only (Figure 6, lane 5). Separation of the cell-free extract by ultracentrifugation showed that most of the OlsD dioxygenase was associated with the membrane fraction (Figure 6, lane 3) and minor activity with the soluble protein fraction (Figure 6, lane 2). Analysis of the OlsD protein sequence suggests that, in contrast to LpxO,9 the OlsD enzyme is devoid of membrane-spanning R-helices (data not shown) that might explain the only loose association of OlsD with the membrane fraction. The OlsD dioxygenase has a pH optimum of∼7.5 and works best at a detergent concentration of 0.04% Triton X-100 (Figure S4 of the Supporting Information). Even in the absence of detergent, significant OlsD activity is detectable (Figure S4B of the Supporting Information), again suggesting that a significant part of OlsD might be in the soluble protein fraction. The OlsD enzyme activity is strictly dependent on the addition of the second substrate R-ketoglutarate (Figure S5A of the Supporting Information), whereas in the absence of ascorbate, OlsD activity is only reduced (Figure S5A). Omission of DTT surprisingly causes an increase in OlsD activity (Figure S5A). Omission of Fe2+ from the complete assay does not cause a reduction in OlsD activity (Figure S5A), and the dependence on Fe2+ is detected only after this metal ion is chelated with 2,20- bipyridyl (Figure S5B of the Supporting Information). There- fore, we conclude that OlsD is a Fe2+/R-ketoglutarate-depen- dent mixed-functional dioxygenase. ’DISCUSSION Lipid A-containing lipopolysaccharides (LPS) usually occupy the outer surface of the outermembrane inGram-negative bacteria and pose a major permeability barrier for hydrophilic and hydro- phobic compounds. Hydrocarbon regions of the outer membrane are thought to be in a gel-like state of very low fluidity under physiological conditions31 because of strong interactions between the lipid molecules. Different environments and stresses require adjustments in the outer membrane that are realized by certain chemical modifications of LPS and/or other lipids composing the outer membrane,31,34 i.e., PE, sphingolipids, sulfonolipids,35 and OLs.11,34 Introduction of an additional hydroxyl group into fatty acyl residues of membrane lipids increases the likelihood of hydrogen bonding with neighboring molecules and thus reduces the fluidity of themembrane.With all of the outermembrane lipids mentioned above, the strategy of decorating their fatty acyl residues with hydroxyl groups at the 2 and/or 3 position, and thereby stabilizing the membrane, is used. In addition to these non-glycerol-based membrane lipids, the glycerophospholipid PE, which is enriched in the outer membrane, can be 2-hydroxylated on its sn-2-located fatty acyl residue in Burkholderia. In B. cepacia NCTC 10661, the levels of the 2-hydroxylated lipids 2-OH-OLs and 2-OH-PE are increased at high temper- atures,3 conditions that could be unfavorable for the formation and stabilization of monolayers of LPS, due to electrostatic repulsion between highly acidic molecules of LPS, probably altering the permeability of the external membrane.31 The hydroxylated OLs could substitute for LPS, with the hydroxylations allowing the establishment of lateral interactions such as hydrogen bond formation to stabilize the external membrane.31 There is evidence that the presence of OLs in membranes allows bacteria to survive in unfavorable environments. For example, OLs contribute to increased cell yields under phosphorus-limiting growth con- ditions36 and to increased resistance to acidic conditions.10,11 In Rhodobacter capsulatus, OLs are required for optimal steady-state amounts of c-type cytochromes.37 Although we found that the OlsD-dependent modification occurs under acidic growth condi- tions, the exact roles of the distinct OLs in B. cenocepacia are not clear. Transcriptome studies suggest that the olsB (bcal1281) transcript is upregulated 4.3-fold after growth of B. cenocepacia J2315 in cystic fibrosis sputum instead of in a minimal medium,38 whereas olsD (bcam2401) is downregulated inB. cenocepacia J2315 biofilms by oxidative stress (2.2-fold by H2O2 or 3.4-fold by NaOCl treatment).39 Another study reports 3-fold increased olsD and 4-fold increased olsB transcript levels after growth of B. cenocepacia J2315 in a soil extract-containing medium when compared to cystic fibrosis sputum-containing medium.40 Previous reports have indicated that members of the Bur- kholderia genus produce two different classes of OLs, standard nonhydroxylated OL and OL hydroxylated at the 2 position of their external ester-linked fatty acyl residue (2-OH-OL). Like Burkholderia, members of the genera Flavobacterium,41,42 Thiobacillus,4 Gluconobacter,43 Ralstonia,44 and Rhizobium11 produce 2-OH-OLs. We now report on an additional possible modification of OL that is caused by OlsD and to date has not been known to occur in nature. Expression of OlsD in S.meliloti or in B. cenocepacia causes hydroxylation of the amide- linked OL fatty acyl residue. Introduction of the olsD gene into S. meliloti 1021 caused the formation of one new lipid (NL) and introduction into B. cenocepacia J2315 two new lipids (NL1 and NL2). The results suggest that OlsD might use the preexisting OLs of S. meliloti 1021 and B. cenocepacia J2315 as substrates. This is consistent with a model based on the results obtained via MS. The predicted dioxygenase encoded by olsD converts the standard OL form (protonated molecule at m/z 693) to NL (m/z 709) in S. meliloti 1021, and in B. cenocepacia J2315, OlsD converts the standard OL (m/z 667) to NL1 (m/z 683) and 2-OH-OL (m/z 683) to NL2 (m/z 699). The MS data also show that spot NL2 contains dihy- droxylated OLs with one additional hydroxyl group on the ester-linked fatty acid residue and the other on the amide- linked residue. The introduction of bcam2401 (olsD) into S. meliloti or B. cenocepacia causes hydroxylation of the amide-linked fatty acyl residue of OLs, and OlsD is not required for the formation of 2-OH-OL. Introduction of bcam1214 into S. meliloti or B. cenocepacia did not lead to the formation of any additional extractable lipid by the method used.27 This suggests that the B. cenocepacia J2315 BCAM1214 with more similarity to the LpxO-encoded dioxygenase from S. typhimurium does not con- duct the 2-hydroxylation on the ester-linked fatty acid residues of OLs to yield 2-OH-OLs. Although the reported structure of lipid A of B. cenocepacia J2315 does not indicate a 2-hydroxy mod- ification at any of its acyl residues,45 it seems likely that the bcam1214 gene is involved in modifying lipid A of B. cenocepacia J2315, because of its high degree of similarity to the LpxO enzyme from S. typhimurium. However, as in the case of LpxO- catalyzed hydroxylation in S. typhimurium, such a BCAM1214- catalyzed modification of burkholderial lipid A might occur only under certain stress conditions. Although our postgenomic approach of predicting biochemical functions based on protein similarities has led to the discovery of this new pathway for OL 104 6406 dx.doi.org/10.1021/bi200706v |Biochemistry 2011, 50, 6396–6408 Biochemistry ARTICLE modification (hydroxylation of the amidified acyl residue of OLs), we have not yet found the gene(s) or enzyme(s) responsible for the formation of 2-OH-OLs and 2-OH-PE. Therefore, at this point, the B. cenocepacia genes and/or enzymes that produce the 2-hydroxy group on the sn-2-linked acyl chain of PE or on the ester-linked fatty acyl chain of OL are still unknown; however, in B. cenocepacia, a close LpxOhomologue is probably not involved in this process. Notably, in R. tropici CIAT899, the enzyme generat- ing the 2-hydroxylation of 2-OH-OLs is the OlsC dioxygenase,11 but OlsC does not cause the formation of 2-OH-PE. Recently, a search with the iron-binding motif of LpxO-like dioxygenases led to the identification of the Kdo 3-hydroxylase KdoO that hydro- xylates the outer Kdo unit of lipopolysaccharide in Burkholderia ambifaria and Yersinia pestis.46However, the KdoO protein shows no overall sequence similarity to LpxO, OlsC, or OlsD and therefore is a distant relative within the Fe2+/O2/R-ketogluta- rate-dependent dioxygenase superfamily. Currently, it is not known how the different LpxO-like dioxygenases (LpxO, OlsC, and OlsD) recognize their substrate acyl chain on distinct membrane lipids. LpxO of S. typhimurium (LpxOSt) hydroxylates exclusively the myristate moiety at posi- tion 2 that is linked to the 30-(R)-3-hydroxymyristate chain of lipid A.9 Pseudomonas aeruginosa contains two LpxO orthologs (LpxO1PAO1 and LpxO2PAO1), and both secondary acyl chains of its lipid A are modified with 2-OH groups.47 Probably each of these orthologs is responsible for hydroxylating one of the two secondary acyl chains of P. aeruginosa lipid A.9 LpxOSt, LpxO1PAO1, LpxO2PAO1, BCAM1214 of B. cenocepacia, and many other LpxO-like dioxygenases form a clearly distinct subfamily (Figure S6 of the Supporting Information) of dioxy- genases that probably hydroxylate lipid A at the 2 position of distinct acyl chains. Two other LpxO-like dioxygenase subfami- lies exist (Figure S6 of the Supporting Information), one represented by OlsC and the other represented by OlsD. The OlsC-like dioxygenase subfamily exists in several R-proteobac- teria (Figure S6 of the Supporting Information) and modifies OLs by hydroxylation of the 2 position of the esterified fatty acyl residue.11 In contrast, the OlsD-like dioxygenase subfamily exists Figure 7. Biosynthesis of OLs in B. cenocepacia J2315. The genes encoding OlsB and OlsA have first been identified in S. meliloti, whereas the gene encoding theOL 2-hydroxylaseOlsC has been described inR. tropici. We show that bcal1281 fromB. cenocepacia encodes theOlsBN-acyltransferase that has an acyl chain preference for a 3-hydroxypalmitoyl. The O-acyltransferase OlsA is probably encoded by bcal3137 in B. cenocepacia. There is no homologue to the rhizobial OlsC inB. cenocepacia that hydroxylatesOL at the 2 position of the esterified acyl residue; however, anOlsC-like activitymust exist in B. cenocepacia. Here we describe that the hydroxylation introduced by the bcam2401-encoded OlsD occurs on the internal, amidified fatty acid. 105 6407 dx.doi.org/10.1021/bi200706v |Biochemistry 2011, 50, 6396–6408 Biochemistry ARTICLE exclusively in species of the genera Burkholderia and Serratia (Figure S6 of the Supporting Information) and was shown in this work to cause hydroxylation of the amide-linked fatty acyl residue of OLs. Spot NL1 produced by B. cenocepacia (Figure 2F) shows the same mobility on 2D-TLC as spot NL observed from S. meliloti after expression of olsD and olsBSm (Figure 1B).However, themass spectrometric data indicate that in the case of B. cenocepacia it is always a C16:0-amidified acyl residue that is hydroxylated byOlsD (Figure 7), whereas in the case of S. meliloti, the amidified acyl residue hydroxylated by OlsD is a C18:0. Analysis of the standard OL fractions from both organisms also indicated that standard OLs from B. cenocepacia have 3-hydroxypalmitate and S. meliloti 3-hydroxystearate as amidified acyl residues. In contrast, there is significant heterogeneity with respect to the esterified acyl residues in the two organisms. This observation might imply that OlsA- encodedO-acyltransferases7 are not highly selective with regard to the chain lengths of the acyl residues linked to acyl carrier proteins (ACPs) in acyl-ACP acyl donor substrates. OlsA fromRhodobacter capsulatus seems to be able to acylate 1-acyl-sn-glycerol-3-phos- phate in addition to lyso-ornithine lipid48 and therefore exhibits relaxed substrate specificity toward the acyl acceptor substrate, as well. In contrast, OlsB N-acyltransferases, which catalyze the first step in the biosynthesis of ornithine-containing lipids (Figure 7), might be highly selectivewith regard to their acyl residues linked to ACPs. We suggest that OlsB from B. cenocepacia might use predominatly 3-hydroxypalmitoyl-ACP as the acyl donor substrate whereas OlsB from S. melilotimay prefer 3-hydroxystearyl-ACP. A strict selectivity for chain length of the acyl residue has been reported for some LpxA O-acyltransferases performing the initial acylation step during lipid A biosynthesis.49 ’ASSOCIATED CONTENT bS Supporting Information. Cloning of burkholderial genes, construction of B. cenocepacia J2315 mutants, additional MS analysis of membrane lipids from B. cenocepacia J2315 and Sinorhizobium meliloti 1021, and enzymatic and phylogenetic characterization of the OlsD dioxygenase. This material is available free of charge via the Internet at http://pubs.acs.org. ’AUTHOR INFORMATION Corresponding Author *Telephone: +52-7773-131697. Fax: +52-7773-175581. E-mail: otto@ccg.unam.mx. Funding Sources N.G.-S. was supported during his Ph.D. studies (Programa de Doctorado en Ciencias Biomedicas, Universidad Nacional Autonoma de Mexico) by a scholarship from the Consejo Nacional de Ciencia y Tecnología (Mexico). This work was supported by grants from DGAPA/UNAM (200806 and 218009) and the Consejo Nacional de Ciencia y Tecnología de Mexico (CONACyT 82614). The York group gratefully ac- knowledges A.M.T.’s studentship provided by the Royal Society of Chemistry (RSC) and the Engineering and Physical Sciences Research Council (EPSRC) and D.S.’s studentship supported by the EPSRC and the European Union (ICA4-CT-2001- 10056), and J.T.-O. gratefully acknowledges funding from the Analytical Chemistry Trust Fund, the RSC Analytical Division, and EPSRC. ’ACKNOWLEDGMENT This article is dedicated to Franz Lingens on the occasion of his 86th birthday. We thank Alhondra Solares-Perez for help in the construction of OlsD-deficient mutant AQ3. ’ABBREVIATIONS BCC, B. cepacia complex;CL, cardiolipin;DGTS, diacylglyceryltri- methylhomoserine;HEPES, 4-(2-hydroxyethyl)-1-piperazineethane- sulfonic acid;Homopipes, homopiperazine-N,N0-bis(2-ethanesulfo- nic acid); LB, Luria-Bertani;MS, mass spectrometry;NL, new lipid; OD, optical density; OL, ornithine-containing lipid; PE, phosphatidylethanolamine; PG, phosphatidylglycerol; Pi, inor- ganic phosphate; TLC, thin layer chromatography; TY, tryp- tone/yeast extract; 2D, two-dimensional; 2-OH, hydroxylation at the 2 position of an acyl residue. ’REFERENCES (1) Mahenthiralingam, E., Urban, T. A., and Goldberg, J. B. (2005) The multifarious, multireplicon Burkholderia cepacia complex. Nat. Rev. Microbiol. 3, 144–156. (2) Phung, L. V., Chi, T. T. B., Hotta, H., Yabuchi, E., and Yano, I. (1995) Cellular lipid and fatty acid compositions of Burkholderia pseudomallei strains isolated from human and environment in Viet Nam. Microbiol. Immunol. 39, 105–116. (3) Taylor, C. J., Anderson, A. J., and Wilkinson, S. G. (1998) Phenotypic variation of lipid composition in Burkholderia cepacia: A response to increased growth temperature is a greater content of 2-hydroxy acids in phosphatidylethanolamine and ornithine amide lipid. Microbiology 144, 1737–1745. (4) Knoche, H. W., and Shively, J. M. (1972) The structure of an ornithine-containing lipid from Thiobacillus thiooxidans. J. Biol. Chem. 247, 170–178. (5) Geiger, O., R€ohrs, V., Weissenmayer, B., Finan, T. M., and Thomas-Oates, J. E. (1999) The regulator gene phoB mediates phos- phate stress-controlled synthesis of themembrane lipid diacylglyceryl-N, N,N-trimethylhomoserine in Rhizobium (Sinorhizobium) meliloti. Mol. Microbiol. 32, 63–73. (6) Gao, J.-L., Weissenmayer, B., Taylor, A. M., Thomas-Oates, J., Lopez-Lara, I. M., and Geiger, O. (2004) Identification of a gene required for the formation of lyso-ornithine lipid, an intermediate in the biosynthesis of ornithine-containing lipids. Mol. Microbiol. 53, 1757–1770. (7) Weissenmayer, B., Gao, J.-L., Lopez-Lara, I. M., and Geiger, O. (2002) Identification of a gene required for the biosynthesis of ornithine-derived lipids. Mol. Microbiol. 45, 721–733. (8) Gibbons, H. S., Lin, S., Cotter, R. J., and Raetz, C. R. (2000) Oxygen requirement for the biosynthesis of the S-2-hydroxymyristate moiety in Salmonella typhimurium lipid A. Function of LpxO, a newFe2+/ R-ketoglutarate-dependent dioxygenase homologue. J. Biol. Chem. 275, 32940–32949. (9) Gibbons, H. S., Reynolds, C. M., Guan, Z., and Raetz, C. R. H. (2008) An inner membrane dioxygenase that generates the 2-hydro- xymyristate moiety of Salmonella lipid A. Biochemistry 47, 2814–2825. (10) Rojas-Jimenez, K., Sohlenkamp, C., Geiger, O., Martínez- Romero, E., Werner, D., and Vinuesa, P. (2005) A CIC chloride channel homolog and ornithine-containingmembrane lipids of Rhizobium tropici CIAT899 are involved in symbiotic efficiency and acid tolerance. Mol. Plant-Microbe Interact. 18, 1175–1185. (11) Vences-Guzman, M. A., Guan, Z., Orme~no-Orrillo, E., Gonzalez-Silva, N., Lopez-Lara, I. M., Martínez-Romero, E., Geiger, O., and Sohlenkamp, C. (2011) Hydroxylated ornithine lipids increase stress tolerance in Rhizoboium tropici CIAT899. Mol. Microbiol. 79, 1496–1514. (12) Meade, H. M., Long, S. R., Ruvkun, G. B., Brown, S. E., and Ausubel, F. M. (1982) Physical and genetic characterization of symbiotic 106 6408 dx.doi.org/10.1021/bi200706v |Biochemistry 2011, 50, 6396–6408 Biochemistry ARTICLE and auxotrophic mutants of Rhizobium meliloti induced by transposon Tn5 mutagenesis. J. Bacteriol. 149, 114–122. (13) Sohlenkamp, C., de Rudder, K. E. E., and Geiger, O. (2004) Phosphatidylethanolamine is not essential for growth of Sinorhizobium meliloti in complex culture media. J. Bacteriol. 186, 1667–1677. (14) Hanahan, D. (1983) Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166, 557–580. (15) Simon, R., Priefer, U., and P€uhler, A. (1983) A broad host range mobilization system for in vivo genetic engineering: Transposon muta- genesis in Gram-negative bacteria. Nat. Biotechnol. 1, 784–791. (16) Yanisch-Perron, C., Vieira, J., and Messing, J. (1985) Improved M13 phage cloning vectors and host strains: Nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33, 103–119. (17) Ditta, G., Schmidhauser, T., Yacobson, E., Lu, P., Liang, X.-W., Finlay, D. R., Guiney, D., and Helinski, D. R. (1985) Plasmids related to the broad host range vector, pRK290, useful for gene cloning and for monitoring expression. Plasmid 13, 149–153. (18) Kovach, M. E., Phillips, R. W., Elzer, P. H., Roop, R. M., II, and Peterson, K.M. (1994) pBBR1MCS: A broad-host-range cloning vector. BioTechniques 16, 800–802. (19) Kovach, M. E., Elzer, P. H., Hill, D. S., Robertson, G. T., Farris, M. A., Roop, R. M., II, and Peterson, K. M. (1995) Four new derivatives of the broad-host-range cloning vector pBBR1MCS, carrying different antibiotic-resistance cassettes. Gene 166, 175–176. (20) Studier, F. W., Rosenberg, A. H., Dunn, J. J., and Dubendorff, J. W. (1990) Use of T7 RNA polymerase to direct expression of cloned genes. Methods Enzymol. 185, 60–89. (21) Seed, B. (1987) An LFA-3 cDNA encodes a phospholipids- linked membrane protein homologous to its receptor CD2. Nature 329, 840–842. (22) Sch€afer, A., Tauch, A., J€ager, W., Kalinowski, J., Thierbach, G., and P€uhler, A. (1994) Small mobilizable multi-purpose cloning vectors derived from the Escherichia coli plasmids pK18 and pK19: Selection of defined deletions in the chromosome of Corynebacterium glutamicum. Gene 145, 69–73. (23) Beringer, J. E. (1974) R factor transfer in Rhizobium legumino- sarum. J. Gen. Microbiol. 84, 188–198. (24) Sherwood, M. T. (1970) Improved synthetic medium for the growth of Rhizobium. J. Appl. Bacteriol. 33, 708–713. (25) Miller, J. H. (1972) Experiments in molecular genetics, Cold Spring Harbor Laboratory Press, Plainview, NY. (26) Sambrook, J., and Russell, D. W. (2001) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Plainview, NY. (27) Bligh, E. G., andDyer, J.W. (1959) A rapidmethod of total lipid extraction and purification. Can. J. Biochem. Physiol. 37, 911–917. (28) de Rudder, K. E. E., Thomas-Oates, J. E., and Geiger, O. (1997) Rhizobium meliloti mutants deficient in phospholipid N-methyltransfer- ase still contain phosphatidylcholine. J. Bacteriol. 179, 6921–6928. (29) Smith, P. K., Krohn, R. I., Hermanson, G. T., Mallia, A. K., Gartner, F. H., Provenzano,M. D., Fujimoto, E. K., Goeke, N.M., Olson, B. J., and Klenk, D. C. (1985) Measurement of protein using bicinch- oninic acid. Anal. Biochem. 150, 76–85. (30) Holden, M. T. G., Seth-Smith, H. M. B., Crossman, L. C., Sebaihia, M., Bentley, S. D., Cerdeno-Tarraga, A. M., Thomson, N. R., Bason, N., Quail,M. A., Sharp, S., Cherevach, I., Churcher, C., Goodhead, I., Hauser, H., Holroyd, N., Mungall, K., Scott, P., Walker, D., White, B., Rose, H., Iversen, P., Mil-Homens, D., Rocha, E. P. C., Fialho, A. M., Baldwin, A., Dowson, C., Barrell, B. G., Govan, J. R., Vandamme, P., Hart, C. A., Mahenthiralingam, E., and Parkhill, J. (2009) The genome of Burkholderia cenocepacia J2315, an epidemic pathogen of cystic fibrosis patients. J. Bacteriol. 191, 261–277. (31) Nikaido, H. (2003) Molecular basis of bacterial outer mem- brane permeability revisited. Microbiol. Mol. Biol. Rev. 67, 593–656. (32) Tomer, K. B., Crow, F. W., Knoche, H. W., and Gross, M. L. (1983) Fast atom bombardment and mass spectrometry/mass spectro- metry for analysis of a mixture of ornithine-containing lipids from Thiobacillus thiooxidans. Anal. Chem. 55, 1033–1036. (33) Zhang, X., Ferguson-Miller, S. M., and Reid, G. E. (2009) Characterization of ornithine and glutamine lipids extracted from cell membranes of Rhodobacter sphaeroides. J. Am. Soc. Mass Spectrom. 20, 198–212. (34) Dees, C., and Shively, J. M. (1982) Localization and quantita- tion of the ornithine lipid of Thiobacillus thiooxidans. J. Bacteriol. 149, 798–799. (35) Geiger,O.,Gonzalez-Silva,N., Lopez-Lara, I.M., and Sohlenkamp, C. (2010) Amino acid-containing membrane lipids in bacteria. Prog. Lipid Res. 49, 46–60. (36) Lopez-Lara, I. M., Gao, J. L., Soto, M. J., Solares-Perez, A., Weissenmayer, B., Sohlenkamp, C., Verroios, G. P., Thomas-Oates, J., and Geiger, O. (2005) Phosphorus-free membrane lipids of Sinorhizo- bium meliloti are not required for the symbiosis with alfalfa but contribute to increased cell yields under phosphorus-limiting conditions of growth. Mol. Plant-Microbe Interact. 18, 973–982. (37) Aygun-Sunar, S., Mandaci, S., Koch, H.-G., Murria, I. V. J., Goldfine, H., and Daldal, F. (2006) Ornithine lipid is required for optimal steady-state amounts of c-type cytochromes in Rhodobacter capsulatus. Mol. Microbiol. 61, 418–435. (38) Drevinek, P., Holden, M. T. G., Ge, Z., Jones, A. M., Ketchell, I., Gill, R. T., and Mahenthiralingam, E. (2008) Gene expression changes linked to antimicrobial resistance, oxidative stress, iron depletion and retained mobility are observed when Burkholderia cenocepacia grows in cystic fibrosis sputum. BMC Infect. Dis. 8, 121. (39) Peeters, E., Sass, A., Mahenthiralingam, E., Nelis, H., and Coenye, T. (2010) Transcriptional response of Burkholderia cenocepacia J2315 sessile cells to treatments with high doses of hydrogen peroxide and sodium hypochlorite. BMC Genomics 11, 90. (40) Yoder-Himes, D. R., Konstantinidis, K. T., and Tiedje, J. M. (2010) Identification of potential therapeutic targets for Burkholderia cenocepacia by comparative transcriptomics. PLoS One 5, e8724. (41) Kawai, Y., Yano, I., and Kaneda, K. (1988) Various kinds of lipoamino acids including a novel serine-containing lipid in an oppor- tunistic pathogen Flavobacterium. Their structures and biological activ- ities on erythrocytes. Eur. J. Biochem. 171, 73–80. (42) Asselineau, J. (1991) Bacterial lipids containing amino acids or peptides linked by amide bonds. Fortschr. Chem. Org. Naturst. 56, 1–85. (43) Tahara, Y., Kameda, M., Yamada, Y., and Kondo, K. (1976) An ornithine-containing lipid isolated from Gluconobacter cerinus. Biochim. Biophys. Acta 450, 225–230. (44) Galbraith, L., Jonsson, M.H., Rudhe, L. C., andWilkinson, S. G. (1999) Lipids and fatty acids of Burkholderia and Ralstonia species. FEMS Microbiol. Lett. 173, 359–364. (45) Silipo, A., Molinaro, A., Ierano, T., de Soyza, A., Sturiale, L., Garozzo, D., Aldridge, C., Corris, P. A., Khan, C. M. A., Lanzetta, R., and Parrilli, M. (2007) The complete structure and pro-inflammatory activity of the lipooligosaccharide of the highly epidemic and virulent Gram- negative bacterium Burkholderia cenocepacia ET-12 (strain J2315). Chem.—Eur. J. 13, 3501–3511. (46) Chung, H. S., and Raetz, C. R. H. (2010) Dioxygenases in Burkholderia ambifaria and Yersinina pestis that hydroxylate the outer Kdo unit of lipopolysaccharide. Proc. Natl. Acad. Sci. U.S.A. 108, 510–515. (47) Ernst, R. K., Yi, E. C., Guo, L., Lim, K. B., Burns, J. K., Hackett, M., and Miller, S. I. (1999) Specific lipopolysaccharide found in cystic fibrosis airway Pseudomonas aeruginosa. Science 286, 1561–1565. (48) Aygun-Sunar, S., Bilaloglu, R., Goldfine, H., and Daldal, F. (2007) Rhodobacter capsulatus OlsA is a bifunctional enzyme active in both ornithine lipid and phosphatidic acid biosynthesis. J. Bacteriol. 189, 8564–8574. (49) Wyckoff, T. J., Lin, S., Cotter, R. J., Dotson, G. D., and Raetz, C. R. (1999) Hydrocarbon rulers in UDP-N-acetylglucosamine acyl- transferases. J. Biol. Chem. 273, 32369–32372. 107 Supporting Information for: The dioxygenase-encoding olsD gene from Burkholderia cenocepacia causes the hydroxylation of the amide-linked fatty acyl moiety of ornithine-containing membrane lipids Napoleón González-Silva, Isabel M. López-Lara, Rodrigo Reyes-Lamothe, Adrian Taylor, David Sumpton, Jane Thomas-Oates, and Otto Geiger Supporting Experimental Procedures Cloning of genes from B. cenocepacia J2315 bcam1214, bcam2401 (olsD), and bcal1281 (olsBBc). The oligonucleotide primers oLOP23 (AGGAATACATATGCGCTGGGTCCTGCTG) and oLOP24 (AAAGGATCCTCAGATACAGAAGATCGCGAC), introducing NdeI and BamHI sites (underlined), respectively, were used in the PCR to amplify the gene bcam1214. The oligonucleotide primers oLOP25 (AGGAATACATATGCCCGTCGCGCTTCG) and oLOP26 (AAAAGATCTCTACACCGGAATCGACCG) introducing NdeI and BglII sites (underlined), respectively, were used in the PCR to amplify the olsD (bcam2401) gene. In all PCR reactions, genomic DNA from B. cenocepacia J2315 was used as a template. After digestion of the PCR products with the corresponding enzymes, the fragments obtained were cloned in the plasmids of the pET vector family that had been digested with the same enzymes. The gene bcam1214 was cloned in pET9a resulting in plasmids pSphx01 whereas olsD was cloned in pET9a and pET16b resulting in plasmids pSphx02 and pNG40, respectively. The gene olsBBc was previously cloned in pET17b to yield the plasmid pNG23 (1). Sequencing analysis confirmed that pSphx01 carries the DNA fragment coding for BCAM1214, pSphxO2 and pNG40 carry the DNA fragment coding for OlsD. The pET9a- and pET17b-derived plasmids were linearized with BglII and each was cloned in the broad- host-range plasmid pRK404 that had been digested with BamHI to obtain plasmids pSphx03, pSphx04, and pNG24 and they were verified to carry bcam1214, olsD, and olsBBc, respectively. A plasmid (pNG28) that was lacking the olsBBc gene was constructed by cloning BamHI-restricted pET17b into pRK404. Inactivation of olsD (bcam2401) by a chloramphenicol resistance-conferring cassette. The chloramphenicol acetyltransferase (CAT) gene of pBBR1MCS was amplified with the oligonucleotide primers oLOP27 (AAAGGATCCACGCGTATCCTGGTGTCCCTG) and oLOP28 (AAAGGATCCACGCGTCCACACAACATACGAG) that introduced BamHI 108 restriction sites and cloned into pUC18 to yield the plasmid pCAT. The oligonucleotide primers oLOP31 (AAATCTAGAAGATCGTCGGCAGCTG) and oLOP32 (AAAGGATCCGGATGAAGTGCTGGTC) were used in the PCR to amplify about 1.2 kb of genomic DNA upstream of olsD of B. cenocepacia J2315, introducing XbaI and BamHI sites (underlined) into the PCR product. Similarly, the primers oLOP33 (GCTGGGGATCCGGATCC) and oLOP34 (AAACTAAGCTTCCGCCGTGACGCGC) were used to amplify about 1.1 kb of genomic DNA downstream of the olsD gen, introducing a HindIII site (underlined) in oLOP34, while oLOP33 matches the BamHI sites (underlined) present in the genomic DNA. The PCR product amplified with oligonucleotides oLOP33 and oLOP34 was digested with BamHI/HindIII and cloned into pBluescriptSK+ that had been digested with the same two enzymes to yield the plasmid pRRL01. Then, plasmid pRRL01 was digested with BamHI and XbaI and was ligated together with the cat gene obtained with BamHI digestion from pCAT and with the PCR product resulting from the amplification with oLOP31 and oLOP32 that have been digested with XbaI and BamHI. The resulting plasmid containing the flanking regions of olsD interrupted by a chloramphenicol resistance gene was named pSphx06. The plasmid pSphx06 was digested with XbaI/HindIII to reclone the regions flanking olsD and the chloramphenicol resistance gene located between those regions as a XbaI/HindIII fragment into the suicide vector pK18mobsacB (2) to yield pSphx08. Via diparental mating using as donor strain E. coli S17-1, pSphx08 was introduced into the wild type B. cenocepacia J2315. Transconjugants were selected on LB medium containing chloramphenicol and piperacillin to counterselect against E. coli. Genetic analysis by PCR showed that the transconjugants were the result of a single event of recombination and in order to select for colonies in which the pK18mobsacB was lost, a selected transconjugant was grown in the presence of sucrose. A single recombinant was first grown under nonselective conditions in complex medium until an OD620 of 1.0 was achieved, and was then plated onto LB medium containing 5% (wt/vol) sucrose and chloramphenicol. Several large colonies among a background of small colonies appeared after 8 days and 25 of the biggest colonies were toothpicked on 5 and 10% sucrose and 7 of them grew well on 10% sucrose. The genetic organization of the 7 candidates was analyzed using the oligos oLOP25 and oLOP26 and only two of them were double recombinants. The clone AQ3 was selected for further studies. Southern hydridization analysis also showed that in the clone AQ3, olsD has been replaced by a chloramphenicol resistance cassette (data not shown). Inactivation of olsBBc (bcal1281) by a chloramphenicol resistance-conferring cassette. The oligonucleotide primers oLOP90 (ATGTTGATATCGCGTGTTCCAGCAAGTTTCG) and oLOP85 (AAAGGATCCTAGGCGTCGGCAGTTCTCG) were used in the PCR to amplify about 1.1 kb of genomic DNA upstream of olsBBc, introducing EcoRV and BamHI sites (underlined) into the PCR product. Similarly, the primers oLOP86 (AAGGGATCCCGACTTCAACTGC) and oLOP87 (ACTCTCTAGACTGTTCGCGCTCGTTTATTGG) were used to amplify about 1.2 kb of genomic DNA downstream of the olsBBc gen, introducing a BamHI site (underlined) in oLOP86 and a XbaI site (underlined) in oLOP87. The PCR product amplified with oligonucleotides oLOP90 and oLOP85 was digested with EcoRV/BamHI and cloned into pBluescriptSK+ that had been digested with the same two enzymes to yield the plasmid pNG10. The PCR product amplified with oligonucleotides oLOP86 and oLOP87 was digested with BamHI/XbaI and cloned into pBluescriptSK+ that had been digested with the same two enzymes to yield the plasmid pNG11. Then, plasmid pNG10 was digested with BamHI and XbaI and was ligated with the 1.2 kb DNA fragment obtained after BamHI/XbaI digestions of pNG11 to yield plasmid pNG14. Later, plasmid pNG14 was linearized with BamHI and was 109 ligated with the BamHI fragment from pCAT containing the cat gene. The resulting plasmid, containing the flanking regions of olsBBc interrupted by a chloramphenicol resistance gene, was named pNG15. Plasmid pNG15 was digested with EcoRV/XbaI to reclone the regions flanking olsBBc and the chloramphenicol resistance gene located between those regions as a EcoRV/XbaI fragment into the suicide vector pK18mobsacB (2) that had been digested with SmaI/XbaI to yield pNG16. Plasmid pNG16 was introduced into the wild type B. cenocepacia J2315 via diparental mating using as donor strain E. coli S17-1. Transconjugants were selected on LB medium containing chloramphenicol and piperacillin to counterselect against E. coli. A single recombinant was first grown under nonselective conditions in complex medium until the OD620 reached 1.0 and then was plated on LB medium containing 10% (wt/vol) sucrose and chloramphenicol. After 8 days of growth in selective medium, the three fastest growing colonies were chosen for genotypic and phenotypic analysis. Southern hydridization analysis showed that in the 3 strains olsBBc had been replaced by a chloramphenicol resistance cassette (data not shown) and one of them, NG1, was chosen for further studies. Supporting Results Glycerophospholipids of B. cenocepacia J2315. The B. cenocepacia wild type strain produces two forms of PE differing in the presence or absence of ester-linked 2-hydroxyfatty acids. Furthermore, for PE lipids, the incorporation of the 2-hydroxyfatty acid is known to be specific to the sn-2 position in B. cenocepacia strains (3, 4). In an effort to determine whether the presence of an extra copy of olsD alters the observed PE composition and the site of hydroxylation within these lipids, both forms of PE were extracted from the 2D-TLC plates obtained from the B. cenocepacia J2315 (pSphx04) lipids. Negative ion mass spectrometric data indicated the presence of a range of PE species (Figure S1A). The PE species giving more abundant MS signals were shown using tandem MS to carry C16:0, C17:1, C18:1, or C19:0 on sn-1 and C16:0, C16:1, C17:1, C18:1, or C19:1 on sn-2. In the fraction corresponding to 2-OH-PE (Figure S1B), the 2-OH-PE species (PE lipids 16, 19, 20, 21, and 22) giving the more abundant signals were similarly shown to carry C16:0, C17:1, C18:1, or C19:0 on sn-1 and C16:0(OH), C17:1(OH), C19:0(OH), or C19:1(OH) on sn-2. In some cases, hydroxylated versions of the corresponding PE lipids can be clearly assigned (Figure S1C) and as expected, hydroxylation is specific to the sn-2 fatty acyl moiety. However, the hydroxylated versions of PE found are not due to OlsD. Negative ESI mass spectra of PG and CL indicated that the acyl substitutions were similar to those on PE (data not shown). We did not detect any hydroxylated fatty acyl residues in PG or CL. Supporting References 1. Vences-Guzmán, M. A., Guan, Z., Ormeño-Orrillo, E., González-Silva, N., López-Lara, I. M., Martínez-Romero, E., Geiger, O., and Sohlenkamp, C. (2011) Hydroxylated ornithine lipids increase stress tolerance in Rhizoboium tropici CIAT899. Mol. Microbiol. 79, 1496- 1514. 110 2. Schäfer, A., Tauch, A., Jäger, W., Kalinowski, J., Thierbach, G., and Pühler, A. (1994) Small mobilizable multi-purpose cloning vectors derived from the Escherichia coli plasmids pK18 and pK19: selection of defined deletions in the chromosome of Corynebacterium glutamicum. Gene 145, 69-73. 3. Phung, L. V., Chi, T. T. B., Hotta, H., Yabuchi, E., and Yano, I. (1995) Cellular lipid and fatty acid compositions of Burkholderia pseudomallei strains isolated from human and environment in Viet Nam. Microbiol. Immunol. 39, 105-116. 4. Taylor, C. J., Anderson, A. J., and Wilkinson, S. G. (1998) Phenotypic variation of lipid composition in Burkholderia cepacia: a response to increased growth temperature is a greater content of 2-hydroxy acids in phosphatidylethanolamine and ornithine amide lipid. Microbiology 144, 1737-1745. 111 Figure S1. Negative ion mass spectra of the extracts of the PE spot (A) or the 2-OH-PE spot (B) from B. cenocepacia J2315 x pSphx04. Peak intensity is normalized to the most intense signal in the region displayed. Comparison of some PE and 2-OH-PE structures (C). 702 704 706 708 710 712 714 716 718 720 722 724 726 728 730 732 734 736 738 740 742 744 746 748 750 752 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 In te n si ty ( % ) 700 754 m/z 702 706 714 716 720 732 746 748 660 670 680 690 700 710 720 730 740 750 760 770 780 790 800 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 78 m/z In te n si ty ( % ) 850 650 662 674 688 690 702 714 716 728 730 733 742 747 756 719 732 PE 2 B A P E 7 P E 8 P E 1 P E 9 P E 4 P E 3 P E 1 0 P E 5 P E 1 3 PE 11 PE 12 P E 6 P E 1 4 P E 1 5 P E 1 6 P E 1 7 P E 1 8 P E 1 9 P E 2 0 P E 2 1 PE 22 112 Figure S1 continued: PE lipid # [M-H] - Structure sn-2 sn-1 9 690 16:0 16:0 16 706 16:0(OH) 16:0 4 716 16:0 18:1 20 732 16:0(OH) 18:1 11 730 19:1 16:0 21 746 19:1(OH) 16:0 12 732 16:0 19:0 22 748 16:0(OH) 19:0 C 113 Figure S2. Positive ESI mass spectra of the extracted spots assigned OL (A), NL1 (B), 2-OH- OL/NL2 (C). Peak intensity is normalized to the most intense signal in the region displayed. 6 0 0 6 0 5 6 1 0 6 1 5 6 2 0 6 2 5 6 3 0 6 3 5 6 4 0 6 4 5 6 5 0 6 5 5 6 6 0 6 6 5 6 7 0 6 7 5 6 8 0 6 8 5 6 9 0 6 9 5 7 0 0 m / z , a m u 0 2 0 4 0 6 0 8 0 0 0 2 0 4 0 6 0 8 0 0 0 2 0 4 0 6 0 8 0 0 0 2 0 4 0 6 0 8 0 0 0 2 0 4 0 6 0 8 0 0 0 2 0 4 0 6 0 8 0 0 0 1 5 615 In te n si ty ( % ) m/z 700 625 651 665 667 600 610 615 620 625 630 635 640 645 650 655 660 665 670 675 680 685 690 695 700 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 227 In te n si ty ( % ) m/z 683 700 600 681 667 697 699 631 641 645 629 O L 1 2 OL 17 6 0 0 6 1 0 6 2 0 6 3 0 6 4 0 6 5 0 6 6 0 6 7 0 6 8 0 6 9 0 7 0 0 m / z , a 0 5 1 0 1 5 2 0 2 5 3 0 3 5 4 0 4 5 5 0 5 5 6 0 6 5 7 0 7 5 681 683 631 667 In te n si ty ( % ) m/z 600 700 A B C O L 1 3 O L 1 4 O L 1 5 O L 1 6 O L 1 8 O L 1 9 O L 8 O L 9 O L 1 0 OL 11 O L 7 O L 5 O L 1 O L 2 OL 3 114 Figure S3. (A) Electrospray mass spectrum of the total lipids from Sinorhizobium meliloti x pSphx04 that contains the olsD gene from Burkholderia cenocepacia J2315. Peak intensity is normalized to the most intense signal in the region displayed. (B) Positive ion mode CID spectrum of the hydroxylated OL at m/z 709 from S. meliloti x pSphx04. Peak intensities are normalized to the most intense fragment ion. 115 Figure S4. pH and detergent dependence of OlsD activity. (A) OlsD-catalyzed conversion of OL to its hydroxylated derivative N-acyl-OH-OL was monitored after incubation in 50 mM Hepes/KOH buffer at the pH values 7.0, 7.5, 8.0, and 8.5. (B) OlsD-catalyzed conversion of OL to its hydroxylated derivative N-acyl-OH-OL was monitored in the presence of increasing amounts of Triton X-100 (% v/v). OlsD enzyme assays were performed at a final protein concentration of 1 mg/mL and for 120 min. The values shown are mean values ± standard deviation derived from three independent experiments. A B 116 Figure S5. Dependence of OlsD activity on soluble cofactors. (A) A standard OlsD assay mixture containing 1 mg/mL of protein was prepared in which either α-ketoglutarate (a), ascorbate (b), iron (c), or dithiothreitol (d) was omitted. The complete assay (e) is described in Experimental Procedures and was also performed with cell-free extract of E. coli BL21 (DE3) x pET16b in which no OlsD had been expressed (f) or with buffer only (g). The values shown are mean values derived from three independent experiments. (B) Inhibition of OlsD activity by increasing concentrations of 2,2’-bipyridyl, an iron chelator. The standard incubation mixture containing 1 mg/mL protein was supplemented with the indicated concentrations of 2,2’-bipyridyl. [ 14 C]OL and the OlsD product N-acyl-OH-[ 14 C]OL are indicated. NE indicates the no-enzyme control. A Time (min) % C o n ve rs io n B 0 0.005 0.01 0.05 0.1 0.3 0.5 1.0 2.0 NE [2,2’-bipyridyl] mM OL N-acyl- OH-OL 117 Figure S6. Unrooted phylogenetic tree of LpxO-like bacterial dioxygenases. The tree was constructed using the program CLUSTAL W (http://www.expasy.ch/). Distances between the sequences are expressed as 10 changes per amino acid residue. Accession numbers and locus tags are as follows: ORFBm (Brucella melitensis 16M; NP_539381,BMEI0464), OlsCRt(Rhizobium tropici CIAT899; AAY28727, OlsC), ORFMsp (Mesorhizobium sp. BNC1; YP_674289, Meso_1730), ORFSpr (Serratia proteomaculans 568; YP_001479424), ORFBt (Burkholderia thailandensis MSMB43; ZP_02466722, Bpse38_010100025414), ORFBma (Burkholderia mallei ATCC23344; YP_105332, BMAA0571 ), ORFBam (Burkholderia ambifaria IOP40-10; ZP_02888624, BamIOP4010DRAFT_0686), OlsD (Burkholderia cenocepacia J2315; BCAM2401 [OlsD]), LpxO2PAO1 (Pseudomonas aeruginosa PAO1; NP_249627, PA0936), BCAM1214 (Burkholderia cenocepacia J2315; BCAM1214), LpxOSt (Salmonella typhimurium LT2; NP_463151, STM4286 [LpxO]), ORFXoo (Xanthomonas oryzae pv. oryzae; AAW73486, XOO0232), LpxO1PAO1 (Pseudomonas aeruginosa PAO1; NP_253202,PA4512). 118 8.2. Resultados adicionales    Tabla 2. Cepas y plásmidos, y sus características relevantes. En la sección de procedimientos experimentales  de resultados adicionales se describe detalladamente su construcción.  Capas  Características relevantes Fuente o referencia S.  meliloti  1021  str‐21  Meade et. al., 1982 S.  meliloti  AAK1  olsB::kan  Gao et. al., 2004  S.  meliloti  CS111  pssA::gm  Sohlenkamp et. al., 2004 S. meliloti NG8  Smc02490::cat  Este trabajo  Rhizobium  tropici  899‐ olsCΔ1  Tiene una deleción no polar de 211 bp en el gen olsC Rojas‐Jiménez et al., 2005 B. cenocepacia  J2315  Wild type   LMG Bacteria Collection B. cenocepacia  AQ3  olsD::cat  González‐Silva et al., 2011 B. cenocepacia  NG1  olsB::cat     González‐Silva et al., 2011 B. cenocepacia  NG4  bcam1214::cat  Este trabajo  E. coli      JM110  rpsL (Strr) thr leu thi‐l lacY galK galT ara tonA tsx dam dmc supE44 Δ(lac‐ proAB) F'[traD36proAB+ laclq lacZiΔM15], cepa deficinete en las  metilasas de DNA Dam y Dcm  Yanisch‐Perron et al.,  1985  DH5α  recA1, Φ80 lacZΔM15, cepa usada para clonar  Hanahan, 1983  S17‐1  Contiene el plásmido RP4 modificado integrado en su genoma  Simon et al., 1983 BL21 (DE3)  F– dcm ompT hsdS(rB – mB –) gal λ(DE3), cepa de expresión  Studier et al., 1990 pLysS  Contiene un gen que codifica para la lisozima T7, inhibidor de ARN  polimerasa T7, CmR  Studier et al., 1990 Plásmidos      pET9a  Vector de expresión, KnR  Studier et al., 1990 pET16b  Vector de expresión, ApR Studier et al., 1990 pET17b  Vector de expresión, ApR Seed, 1987  pET3d  Vector de expresión, ApR Novagen  119 pUC18  Vector de clonación, ApR  Yanisch‐Perron et al.,  1985  pBluescriptSK+  Vector de clonación, ApR  Stratagene  pCR®2.1‐TOPO  Vector de clonación, ApR KnR Invitrogen  pBBR1MCS  Vector de amplio rango de hospedero, CmR Kovach et al., 1994 pBBR1MCS‐5  Plásmido de amplio rango de hospedero, GmR Kovach et al., 1995 pRK404  Vector de amplio rango de hospedero, TcR   Ditta et. al., 1985 pK18mobsacB  Vector suicida, KnR  Schäfer et. al., 1994 pIML21  acpP de S. meliloti clonado en pRK404 como BamHI‐BglII, TcR Isabel M. López Lara pJG21  olsB de S. meliloti clonado en pRK404, KnR TcR Gao et al,.2004  pILAS03  olsB de S. meliloti clonado en pBBR1MCS‐5  González‐Silva et al., 2011 pERMAV1  olsB de R. tropici clonado en pET9a como NdeI‐BamHI, KnR Miguel A. Vences Guzmán pERMAV32  pERMAV1 clonado en pRK404 como BamHI, KnR TcR Miguel A. Vences Guzmán pCCS98  olsC corto de R. tropici clonado en pET9a como NdeI‐BamHI, KnR Christian Sohlenkamp pCCS82  olsC largo de R. tropici clonado en pET9a como NdeI‐BamHI, KnR Christian Sohlenkamp pNG12  F1 de bcam1214 clonado en pBluescriptSK+ como EcoRV‐BamHI, ApR Este trabajo  pNG18  bcam2775 clonado en pET9a como NdeI‐BglII, KnR Este trabajo  pNG20  pNG18 clonado en pRK404 como BglII, KnR TcR Este trabajo  pNG23  olsBBc clonado en pET17b, ApR Vences‐Guzmán et al.,  2011  pNG24  pNG23 clonado en pRK404 como BglII, ApR TcR González‐Silva et al., 2011 pNG25  olsBBc clonado en pBBR1MCS, CmR Vences‐Guzmán et al.,  2011  pNG26  bcal0511 clonado en pET9a, KnR Este trabajo  pNG27  pNG26 clonado en pRK404 como BamHI, KnR TcR Este trabajo  pNG28  pET17b clonado en pRK404 como BamHI, ApR TcR González‐Silva et al., 2011 pNG29  olsCcorto clonado en pNG23 como BamHI‐BglII, ApR Este trabajo  pNG30  olsClargo clonado en pNG23 como BamHI‐BglII, ApR Este trabajo  pNG31  pNG29 clonado en pRK404 como BamHI, ApR TcR Este trabajo  pNG32  pNG30 clonado en pRK404 como BamHI, ApR TcR Este trabajo  pNG33  olsBSm clonado en pIML21 como BamHI‐BglII, TcR Este trabajo  120 pNG39  pET3d clonado en pRK404 como BamHI, ApR TcR Este trabajo  pNG40  olsD clonado en pET16b, ApR González‐Silva et al., 2011 pNG42  olsCRt corto clonado en pET16b como NdeI‐BamHI, ApR Este trabajo  pNG43  pNG42 clonado en pRK404 como BamHI, ApR TcR Este trabajo  pNG45  F2 de smc02490 clonado en pBluescriptSK+ como BamHI‐XbaI, ApR Este trabajo  pNG49  F2 de bcam1214 clonado en pUC18 como BamHI‐XbaI, ApR Este trabajo  pNG50  F1 de smc02490 clonado en pCR®2.1‐TOPO®, ApR KnR Este trabajo  pNG51  F2 de bcam1214 clonado en pNG12 como BamHI‐XbaI, ApR Este trabajo  pNG52  F1 de smc02490 clonado en pNG45 como EcoRV‐BamHI, ApR Este trabajo  pNG53  cat clonado en pNG51 como BamHI, ApR CmR Este trabajo  pNG54  cat clonado en pNG52 como BamHI, ApR CmR Este trabajo  pNG55  F1 de bcam1214+ cat + F2 de bcam1214 clonado en pK18mobsacB  como EcoRV‐XbaI, KnR CmR  Este trabajo  pNG56  F1 de Smc02490 + cat + F2 de Smc02490 clonado en pK18mobsacB  como EcoRV‐XbaI, KnR CmR  Este trabajo  a: KnR resistencia a kanamicina, TcR resistencia a tetraciclina, CmR resistencia a cloranfenicol, ApR resistencia a ampicilina,  GmR resistencia a gentamicina, F1 y F2, región rio arriba y rio abajo de un gen, respectivamente.    8.2.1.  Los  lisolípidos  (LOLs)  derivados  de  los OLs modificados  por OlsD  tienen menor  movilidad en TLC con respecto a los LOLs derivados de OLs no modificados  De acuerdo a  los datos de espectrometría de masas, OlsD modifica con un grupo  hidroxilo el ácido graso amidificado a la ornitina de los OLs. Para confirmar este resultado  se analizó la movilidad en TLC de los OLs no modificados y modificados por OlsD, así como  la  de  sus  respectivos  LOLs  generados  después  de  deacilarlos  químicamente  por  un  tratamiento alcalino suave. Se observó que  los OLs modificados  tienen menor movilidad  en TLC posiblemente debido a que son más polares porque presumiblemente poseen un  grupo hidroxilo adicional respecto a  los OLs no modificados (fig. 11). También se observó  que los LOLs derivados de los OLs modificados por OlsD tienen una menor movilidad en la  TLC respecto a los LOLs derivados de los OLs no modificados (fig. 11). El resultado anterior  sugiere que el grupo hidroxilo adicional que supuestamente poseen los OLs modificados se  121 localiza  en  el  ácido  graso  amidificado  o  en  el  residuo  de  ornitina.  Por  lo  tanto,  este  resultado  es  consistente  con  los  datos  obtenidos  por  el  análisis  de  espectrometría  de  masas de  los OLs no modificados y modificados que sugieren que OlsD modifica con un  grupo hidroxilo el ácido graso amidificado a la ornitina de los OLS.    Figura.  11.  Separación  por  TLC  de  lípidos  de  ornitina  no  modificados  y modificados por OlsD,  así  como  sus  respectivos  lisolípidos (LOLs). OLs no modificados  (línea 1), OLs modificados  por OlsD (línea 2), LOLs derivados de OLs no modificados (línea 3)  y LOLs derivados de los OLs modificados por OlsD (línea 4). Ácidos  grasos (AGs).      8.2.2. OlsD también modifica in vitro a los 2‐OH‐OL  Al realizar el ensayo enzimático con OlsD usando  los [14C]2‐OH‐OLs como sustrato  se observó que OlsD también es capaz de convertir 33 % del sustrato al producto NL2 (fig.  12). Lo que sugiere que esta enzima tiene la flexibilidad para reconocer como sustratos a  OLs  hidroxilados  en  el  C2  de  su  acilo  esterificado.  El  uso  de  2‐OH‐OLs  como  sustrato  también se observó in vivo [González‐Silva et al., 2011].    Figura 12. Modificación in vitro de [14C]2‐OH‐OL por OlsD. 2‐OH‐ OL incubado con: extractos crudos libres de células de E. coli BL21  (DE3) x pLysS x pET16b  (vector vacío)  (línea 1), extractos crudos  libres de células de E. coli BL21 (DE3) x pLysS x pNG40 (línea 2). 2‐ OH‐OL y NL2 se indican.    122 8.2.3.  La  mutante  en  bcam1214  derivada  de  B.  cenocepacia  J2315  aún  forma  los  derivados 2‐hidroxilados de los OLs y PE  El análisis de hibridación tipo Southern de varias colonias candidatas a mutantes en  el gen bcam1214 mostró que bcam1214 se había remplazado en su mayor parte por el gen  cat (fig. 13), como se esperaba que ocurriera. Lo anterior quedó demostrado debido a que  la  sonda hibridó  en  las  candidatas  a mutantes  con un  fragmento de 2130 pb,  como  se  esperaba. Aunque  también  hibridó  en  las  candidatas  a mutantes  con  un  fragmento  de  alrededor de 6.5 kb, que no se esperaba. En  la capa silvestre  la sonda hibridó solamente  con un fragmento de 1654 pb, como se esperaba. Lo anterior sugiere que la construcción  que se uso para mutar a bcam1214 provocó rearreglos en los genomas de las células en las  cuales  se  introdujo.  Probablemente  por  eso  la  sonda  hibridó  en  las  mutantes  con  un  fragmento  de  alrededor  de  6.5  kb,  que  no  esperaba  que  hibridara.  Una  mutante  se  seleccionó  (se nombró NG4) para analizar su perfil de  lípidos de membrana. El perfil de  lípidos de membrana de la cepa NG4 (datos no mostrados) fue igual al de la cepa silvestre  y aún forma los derivados 2‐OH de los OLs y de PE. Lo anterior sugiere que bcam1214 no  está involucrado en modificar a los OLs ni a PE en B. cenocepacia J2315 como lo indicaban  los  datos  de  su  expresión  heteróloga.  Posiblemente  BCAM1214  esté  involucrado  en  hidroxilar al  lípido A de B. cenocepacia  J2315 en ciertas condiciones de crecimiento o a  alguna  proteína  de  esa  bacteria  como  lo  hacen  proteínas  similares  a  LpxO  como  las  aspartil/asparaginil β‐hidroxilasas de mamíferos.        Figura  13.  Perfil  de  hibridación  tipo  Southern  de  candidatas  a  mutantes en el gen bcam1214. Marcador molecular de 1 kilobase  (kb) (Invitrogen) (línea M2), DNA genómico digerido con NcoI y PstI  de:  candidata  a mutante en bcam1214 NG3  (línea 1),  candidata  a  mutante  en  bcam1214  NG4  (línea  2),  candidata  a  mutante  en  bcam1214 NG5  (línea  3),  candidata  a mutante en  bcam1214 NG6  (línea  4),  candidata  a  mutante  en  bcam1214  NG7  (línea  5),  marcador molecular de 1 kb  (Fermentas)  (línea M1), cepa silvestre  B. cenocepacia J2315 (línea S).  123 8.2.4.  La  expresión  de  los  genes  bcal0511  y  bcam2775  de  B.  cenocepacia  J2315  que  codifican  para  productos  homólogos  a  PHYH  sugiere  que  estos  genes  no  están  involucrados en modificar a los OLs ni a PE  La  enzima  PHYH  de  humano  cataliza  la  2‐hidroxilación  del  ácido  fitánico  proveniente del consumo de ciertos alimentos [Jansen et al., 1997; Jansen et al., 2000]. La  secuencia de PHYH  (338 aminoácidos) se obtuvo del banco de genes  (geneBank),  la cual  está  depositada  con  el  número  de  acceso  CAG46852.  Posteriormente  realizamos  una  búsqueda  en  el  genoma  de  B.  cenocepacia  J2315  (http://www.sanger.ac.uk/cgi‐ bin/blast/submitblast/b_cenocepacia) utilizando la secuencia de PHYH. En el genoma de B.  cenocepacia  J2315  encontramos  dos  ORFs  homólogos  a  PHYH:  BCAL0511  (257  aminoácidos)  con 26 % de  identidad, 42 % de  similitud  y un  valor E de 1.7  x 10‐15  con  respecto PHYH de humano y BCAM2775 (291 aminoácidos) con 28 % de  identidad, 43 %  de similitud con un traslape de 145 aminoácidos con PHYH de humano y un valor E de 2.5  x 10‐05.  El gen bcal0511 se localiza en el cromosoma 1 y bcam2775 en el cromosoma 2. Los  productos de ambos genes están anotados como dioxigenasas putativas. Para conocer si  uno  de  estos  productos  tenía  la  capacidad  de  modificar  a  PE,  los  plásmidos  pNG26  (contiene a bcam2775) y pNG18 (contiene a bcal0511) se introdujeron a la cepa de E. coli  BL21 (DE3) x pLysS, se indujo su expresión con IPTG y se verificó en geles desnaturalizantes  de poliacrilamida al 10 % (fig. 14). En el extracto total de proteínas de la cepa BL21 (DE3) x  pLysS  x pNG26  se observó una proteína  con una abundancia  relativa mayor a  todas  las  demás proteínas de aproximadamente 29 kDa, que probablemente se trata de BCAL0511.  Las  predicciones  indican  que  BCAL0511  tiene  un  tamaño  de  28.7  kDa.  También,  en  el  extracto  total de proteínas de  la cepa BL21  (DE3) pLysS pNG18 se observó una proteína  con una abundancia relativa mayor a todas  las demás proteínas de aproximadamente 32  kDa,  que  probablemente  se  trata  BCAM2775.  Las  predicciones  indican  que  BCAM2775  tiene  un  tamaño  de  32.6  kDa.  El  resultado  anterior  confirmó  la  expresión  de  ambas  proteínas. Sin embargo, los perfiles lipídicos de los cultivos paralelos marcados con acetato  de  sodio  [1‐14C]  derivados  de  las  cepas  que  contenían  los  plásmidos  pNG26  y  pNG18  124 fueron  similares  al  perfil  de  la  cepa  que  contenía  el  plásmido  vacío  (pET9a),  en  la  que  solamente  se  observó  la  presencia  de  PE,  PG  y  CL  (datos  no  mostrados).  El  resultado  anterior  sugiere  que  ninguna  de  las  dos  proteínas  BCAL0511  ni  BCAM2775  están  involucradas en hidroxilar a PE de E. coli. Por eso se espera que BCAL0511 o BCAM2775  tampoco estén involucradas en la 2‐hidroxilación de PE de B. cenocepacia J2315.    Fig.  14. Análisis de  extractos  crudos  libres de  células  por  SDS‐PAGE.  Marcador  de  tamaño  molecular  de  proteínas  (LabAid™,  PageRuler™  Plus  Prestained  Protein  Ladder,  #SM1811)  (línea 1). Extractos de E. coli BL21 (DE3) x pLysS  x pET16b el vector de expresión vacío (línea 2),  de E. coli BL21 (DE3) x pLysS x pNG26 expresa a  BCAL0511 (línea 3), de E. coli BL21 (DE3) x pLysS  x  pNG18  expresa  a  BCAM2775  (línea  4).  Las  bandas más  prominentes  de  los  carriles  3  y  4  probablemente  corresponden  a  las  proteínas  BCAL0511 y BCAM2775, respectivamente.      El  plásmido  pNG27  (contiene  a  bcal0511)  se  movilizó  a  las  cepas  de  S.  meliloti  CS111 pNG25 (la cual, no forma PE ni sus derivados metilados y sobreproduce OL) y a B.  cenocepacia  J2315.  La  cepa  CS111  x  pNG25  x  pNG27  mostró  un  perfil  de  lípidos  de  membrana similar a la cepa CS111 x pNG25. De igual manera la cepa B. cenocepacia J2315  x pNG27 (contiene a bcam2775) tuvo un perfil cualitativo y cuantitativo similar a la cepa B.  cenocepacia J2315 pJG16 (datos no mostrados). El plásmido pNG20 (contiene bcam2775)  se movilizó  a  la  cepa  S. meliloti  1021  x  pILAS03  (sobreproduce OL)  y  a  B.  cenocepacia  J2315.  La  cepa  S.  meliloti  1021  x  pILAS03  x  pNG20  mostró  un  perfil  de  lípidos  de  membrana similar a la cepa S. meliloti 1021 x pILAS03 x pJG16, de igual manera la cepa B.  cenocepacia  J2315  pNG20  tuvó  un  perfil  cualitativo  y  cuantitativo  similar  a  la  cepa  B.  cenocepacia  J2315  pJG16.  Estos  resultados  sugieren  que  ninguno  de  los  dos  productos  BCAL0511 ni BCAM2775 están involucrados en hidroxilar a los OLs ni a ningún otro  lípido  de membrana de  las  cepas derivadas de  S. meliloti 1021 o de B.  cenocepacia  J2315.  Si  BCAL0511 o BCAM2775 estuvieran involucrados en modificar algún lípido se esperaba que  125 al introducir a pNG27 o a pNG20 en las cepas derivadas de S. meliloti 1021 se formara un  nuevo  lípido y al  introducirlos en B. cenocepacia J2315 se esperaba que se  incrementara  considerablemente  la  producción  de  algún  lípido  de  los  preexistentes.  Lo  anterior  se  esperaba  debido  a  que  tenemos  el  antecedente  de  que  al  introducir  un  gen  bajo  la  regulación del promotor T7 en  S. meliloti 1021 o en B.  cenocepacia  J2315 es  suficiente  para que el gen se exprese constitutivamente.  8.2.5. Los genes olsBs de rizobiales no complementan a  la mutante NG1 derivada de B.  cenocepacia J2315  Los genes olsB de S. meliloti 1021 y de R. tropici CIAT899 no complementaron a la  mutante NG1  (datos no mostrados). Lo anterior se demostró  fehacientemente porque el  pJG21 obtenido de la cepa NG1 si complementó a la mutante AAK1 (fig. 15). No sabemos  porque  los  olsB  rizobiales  no  complementaron  a  la  mutante  NG1.  Nuestra  hipótesis  a  comprobar fue que el gen olsBSm no interacciona correctamente con la AcpPBc involucrada  en el primer paso de  la  síntesis de  los OLs, por  lo  tanto, expresando  a olsBSm  y acpPSm  posiblemente  se  sintetizarían  los  OLs  en  la  mutante  NG1.  Para  comprobar  la  hipótesis  anterior  los plásmidos pIML21 y pNG33 se movilizaron a  las mutantes AAK1 y NG1, y se  analizó  su  perfil  de  lípidos  de membrana,  la  cepa  AAK1  x  pIML21 mostró  un  perfil  de  lípidos  similar  a  la  cepa  AAK1,  como  se  esperaba;  y  la  cepa  AAK1  x  pNG33  formó  OL  también como se esperaba  (datos no mostrados). Sin embargo,  la cepa NG1 x pIML21 y  NG1 x pNG33 no formaron OLs (datos no mostrados).  Los  resultados  anteriores  sugieren  que  el  gen  olsB  de  S.  meliloti  1021  no  complementa  a  la mutante NG1  aún  cuando  se  coexpresa  con el acpPSm. Posiblemente  AcpPSm no se convirtió a la forma holo en la cepa NG1, para lo cual se requiere la actividad  de AcpSSm (enzima que transfiere el residuo 4‐fosfopanteteína de coenzima A a la AcpPSm,  para  convertirla a  su  forma activa). Sin el grupo prostético  la AcpPSm no portará grupos  acilos para  ser  transferidos por OlsBSm a  la ornitina  y debido a eso posiblemente no  se  formó  el  lisolípido  de  ornitina  y  por  consecuencia  tampoco  OL.  En  contraste,  olsBBc  sí  complementó  a  la  cepa mutante AAK1  (fig.  15). Un  fenómeno  reportado  en  S. meliloti  1021  y  en  B.  cenocepacia  J2315  es  que  la  introducción  de  sus  respectivos  olsB  bajo  la  126 regulación del promotor T7, promueve  la sobreproducción de  los OLs. Este fenómeno no  se observó  cuando  se  introdujeron  los  olsB  rizobiales  en  la  cepa B.  cenocepacia  J2315,  pero sí cuando se introdujo el olsBBc en la cepa AAK1. Lo que reafirma y es consistente con  nuestros resultados de que los olsB rizobiales no son funcionales en B. cenocepacia J2315.      Figura 15. Separación de  lípidos  totales de membrana por 2D‐TLC marcados  con acetato de  sodio  [1‐14C] de  cepas  derivadas de  la complementación con olsBSm y olsBBc de  la mutante en olsB de S. meliloti 1021  (AAK1).  (A) AAK1 x  pJG16 el plásmido está vacío, (B) AAK1 x pJG21 el plásmido contiene al gen olsBSm, el plásmido pJG21 se extrajo de  la  cepa NG1 x pNG21 y (C) AAK1 x pNG24 el plásmido contiene al gen olsBBc. Los lípidos cardiolipina (CL), fosfatidilglicerol  (PG),  lípido  de  ornitina  (OL),  fosfatidiletanolamina  (PE),  dimetilfosfatidiletanolamina  (DMPE)  y  fosfatidilcolina  (PC)  se  indican.    8.2.6. Sensibilidad a antibióticos de cepas derivadas de B. cenocepacia J2315  B. cenocepacia J2315 es un patógeno oportunista resistente a muchos antibióticos  y  a  altas  concentraciones  de  ellos.  Para  saber  sí  los  OLs  están  involucrados  directa  o  indirectamente en el mecanismo de resistencia a antibióticos se probó  la sensibilidad de  las  cepas  J2315,  AQ3,  AQ3  pJG16,  AQ3  pSphx4,  NG1,  NG1  pNG28  y  NG1  pNG24  a  18  antibióticos con diferentes blancos de acción (tabla 1, pag. 48), entre ellos polimixina B y  bacitracina  que  actúan  sobre  la  membrana.  La  polimixina  B  es  un  péptido  catiónico  antimicrobiano que se une a los grupos cargados negativamente de las membranas de las  bacterias,  los agregados de péptidos en  la membrana externa rompen  la  integridad de  la  membrana, lo que provoca la muerte celular [Lewenza et al., 2011]. La bacitracina es una  127 mezcla  de  polipéptidos  cíclicos  que  interfiere  con  la  desfosforilación  del  isoprenilo  C55‐ pirofosfato, una molécula que  transporta  los elementos estructurales del peptidoglicano  en  la membrana  celular  bacteriana,  inhibiendo  la  formación  de  la  pared  celular  de  las  bacterias  [Stone  y  Strominger,  1971].  En  el  ensayo  de  sensibilidad  a  antibióticos  no  se  observaron halos de  inhibición, excepto en el ácido nalidíxico que  inhibió el crecimiento  uniformemente en  todas  las cepas probadas. Los halos de  inhibición que  se observaron  fueron de pocos milímetros.  8.2.7.  La  coexpresión  en  la  cepa  CS111  de  la  versión  corta  de  olsC  y  olsB  de  B.  cenocepacia J2315 formaron un nuevo lípido  Para determinar cuál de  las dos versiones (corta o  larga) de olsC es funcional o si  ambas  lo  son  se clonaron  las dos versiones  independientemente en el plásmido pNG23  que  contiene  al  gen  olsB  de  B.  cenocepacia  J2315.  Los  plásmidos  resultantes  pNG29  y  pNG30 se  ligaron al pRK404 resultando  los plásmidos pNG31 y pNG32, respectivamente,  los cuales se movilizaron a la cepa CS111, así como el pNG24. La cepa CS111 no forma PE,  MMPE ni DMPE. El gen olsB promoverá la sobreproducción de los OLs no modificados. Por  lo tanto, si ambas versiones fueran funcionales se esperaba que las cepas CS111 x pNG31  y CS111 x pNG32  formaran adicionalmente OLs  como  los OLs P1. El perfil de  lípidos de  membrana de  la cepa CS111 x pNG24 consistió de CL, PG, PC y una alta cantidad de OLs  (fig.  16A),  como  se  esperaba.  El  perfil  de  lípidos  de  la  cepa  CS111  x  pNG32  (datos  no  mostrados) fue similar al perfil de la cepa CS111 pNG24, lo anterior sugiere que la versión  larga  de  olsC  no  es  funcional.  En  contraste,  la  cepa  CS111  x  pNG31  si  formó  un  lípido  adicional que tiene una movilidad en 2D‐TLC similar a los OLs P1 de R. tropici CIAT899 (fig.  16B).  Lo  anterior  sugiere  que  la  versión  corta  de  olsC  sí  es  funcional.  Desconocemos  porque  no  fue  funcional  la  versión  larga  de  OlsC  en  la  cepa  CS111  x  pNG32  pero  posiblemente se debe a que esta versión se pliega mal y por lo tanto es degradada.    128   Figura 16. Separación de  lípidos  totales de membrana por 2D‐TLC marcados  con acetato de  sodio  [1‐14C] de  cepas  derivadas de  la cepa CS111.  (A) CS111 x pNG24 el plásmido contiene el gen olsBBc y  (B) CS111 x pNG31 el plásmido  contiene el gen olsBBc y la versión corta del gen olsC de R. tropici CIAT899. Los lípidos cardiolipina (CL), fosfatidilglicerol  (PG), lípido de ornitina (OL), lípido de ornitina modificado dependiente de olsCc (OLM) y fosfatidilcolina (PC) se indican.    8.2.8. OlsC con cola de histidinas es funcional y aparentemente es una proteína soluble  Los  genes  que  se  clonan  en  el  vector  pET16b  como  NdeI/BamHI  producen  proteínas con una cola de 10 histidinas adicionales en el extremo N‐terminal. El gen olsC  se clonó en el pET16b produciendo el plásmido pNG42, el cual subsecuentemente se ligó  al pRK404, produciendo el plásmido pNG43. El pNG43 se introdujo en la mutante en olsC  (899‐olsCΔ1) derivada de R.  tropici CIAT899  y después  se analizó  su perfil de  lípidos de  membrana. OlsC es  la proteína reponsable de hidroxilar  los dos OLs menos polares (S1 y  S2) en R. tropici CIAT899 para convertirlos en otros dos más polares (P1 y P2), por lo tanto,  una mutante en dicho gen no forma los OLs P1 ni P2 [Rojas‐Jimenez et al., 2005; Vences‐ Guzmán et al., 2011]. Nosotros observamos que  la mutante en olsC  (899‐olsCΔ1) con el  plásmido pNG43  restauró  la producción de  los OLs P1 y P2. Eso sugiere que OlsC con  la  cola de histidinas es funcional (fig. 17).    129 Figura  17.  Separación  de  lípidos  totales  de  membrana por 2D‐TLC marcados  con  acetato  de  sodio  [1‐14C]  de  la  cepa  899‐olsCΔ1  x  pNG43, el plásmido  contiene el gen olsC.  Los  lípidos  cardiolipina  (CL),  fosfatidilglicerol  (PG),  lípido  de  ornitina  estándar  (S1),  lípido  de  ornitina  hidroxilado  en  la  ornitina  (S2),  lípido  de ornitina hidroxilado en el acilo esterificado  (P1),  lípido  de  ornitina  hidroxilado  en  la  ornitina  y  en  el  acilo  esterificado  (P2),  fosfatidiletanolamina  (PE),  dimetilfosfatidiletanolamina  (DMPE)  y  fosfatidilcolina (PC) se indican.                   Por otro lado, los extractos crudos líbres de células de las cepas BL21 (DE3) x pLysS  x  pET16b  y  BL21  (DE3)  x  pLysS  x  pNG42,  así  como  las  fracciones  soluble  e  insoluble  (membranas) provenientes de los extractos libres de células de la última cepa se utilizaron  para hacer  el  ensayo  enzimático.  En  el  ensayo  se observó un porcentaje de  conversión  mayor  (57 %)  cuando  se usó  la  fracción de proteína  soluble  (fig. 18A). El porcentaje de  conversión cuando se utilizaron extractos crudos y la fracción de membranas es de 38 % y  23 %, respectivamente. El resultado anterior sugiere que la proteína olsC es una proteína  soluble que no  está  asociada  a  las membranas de R.  tropici CIAT899,  como  también  lo  predice un análisis con el algoritmo TMHMM [Krogh et al., 2001] (fig. 18B).      Figura 18. OlsC es una proteína soluble que introduce un grupo  hidroxilo  en  el  C2  del  acilo  esterificado  de  los  OLs.  A.  OL  incubado  con extractos  libre de  células de: E.  coli BL21  (DE3) x  pLysS  x  pET16b  (vector  vacío)  (línea  1),  BL21  (DE3)  x  pLysS  x  pNG42 (expresa a OlsC) (línea 3), la fracción de proteína soluble  de  E.  coli BL21  (DE3)  x pLysS  x pNG42  (línea 4),  la  fracción de  membrana  de  E.  coli  BL21  (DE3)  x  pLysS  x  pNG42  (línea  5)  y  solamente  bufer  (línea  2),  se  incubaron  con  [14C]OL  a  una  concentración final de 169 µM por 2 h. Los lípidos de ornitina no  modificados (OLs) y  los OL 2‐hidroxilados (2‐OH‐OLs) se  indican.  B.  Predicción  de  hélices  transmembranales  de  OlsC  con  el  algoritmo TMHMM [Krogh et al., 2001].  A  130   B    8.2.9. Los homólogos a SMc02490 se encuentran  fusionados con homólogos a OlsB en  muchas bacterias  Recordemos  que  olsB  y  smc02490  en  algunas  bacterias  coexisten  o  se  encuentran  fusionados [Geiger et al., 2010]. En un análisis reciente la fusión de homólogos a OlsB con  homólogos  de  SMc02490  se  detectó  en  50  especies  de  bacterias  distribuidas  en  los  géneros  Klebsiella  (3),  Dickeya  (3),  Pectobacterium  (3),  Cronobacter  (2),  Erwinia  (1),  Citrobacter (1), Serratia (1), Enterobacter (1), Cellvibrio (1), Shewanella (13), Alteromonas  (1),  Pseudoalteromonas  (1),  Teredinibacter  (1),  Idiomarina  (1),  Psychromonas  (1),  Stenotrophomonas (2), Thiomicrospira (1), Kangiella (1), Marinomonas (1), Aeromonas (2),  Tolumonas (1), Desulfovibrio (4), Arcobacter (1), Sulfurospirillum (1), Magnetococcus (1) y  Akkermansia (1).  SMc02490  es  similar  a  miembros  de  la  superfamilia  N‐aciltransferasa  (NAT).  Los  miembros  esta  superfamilia  de  enzimas  catalizan  la  transferencia  en  su mayoría  de  un  grupo acilo a un sustrato y están implicados en una variedad de funciones, que van desde  la resistencia bacteriana a  los antibióticos a  los ritmos circadianos en  los mamíferos. Los  miembros  incluyen  N‐acetiltransferasas  relacionadas  con  GCN5  (GNAT),  tales  como  las  enzimas aminoglucósidos N‐acetiltransferasas, histona N‐acetiltransferasas, y la serotonina  N‐acetiltransferasa, que catalizan la transferencia de un grupo acetilo a un sustrato. Otros  miembros de  la superfamilia  incluyen  la arginina/ornitina N‐succiniltransferasa, miristoil‐ CoA: proteína N‐miristoiltransferasa, y la sintasa de acil homoserina lactona que tienen un  mecanismo  catalítico  similar  pero  difieren  en  los  tipos  de  grupos  acilo  transferidos.  131 Leucil/fenilalanil‐tRNA‐proteína  transferasa  y  las  peptidiltransferasas  no  ribosomales  FemXAB que catalizan reacciones similares a peptidiltransferasas también están incluidas.  Se  ha  reportado  que  en  medio  mínimo  limitado  en  fosfatos  se  incrementa  la  transcripción  del  smc02490  como  también  lo  hacen  otros  genes  involucrados  en  la  biosíntesis de lípidos de membrana sin fósforo en cepas de S. meliloti [Krol y Becker, 2004].  También en medio mínimo  limitado en  fosfatos se han detectado cambios en  los ácidos  grasos provenientes de los lípidos de membrana de S. meliloti. La composición porcentual  relativa  del  ácido  lactobacilico  (ácido  cis‐11,12‐metilene  octadecanoico)  este  ácido  contiene  un  grupo  ciclopropano  y  aumentó  2  veces  en  la  condición  de  crecimiento  limitada  en  fosfatos,  el aumento  produjo  una  disminución modesta  en  todos  los  otros  ácidos  grasos,  especialmente  el  cis‐vaccénico,  este  es  precursor  del  ácido  graso  con  ciclopropano  [Basconcillo  y  McCarry,  2008].  Aunque  no  se  estudió  de  que  lípidos  de  membrana provenía el ácido  lactobacilico, éste ha sido detectado como acilo secundario  de  los  OLs,  los  cuales  se  incrementan  en  S.  meliloti  cuando  se  crece  en  limitación  de  fosfatos  [Geiger  et  al.,  1999].  Los  datos  anteriores  sugieren  que  OlsB  y  SMc02490  sí  podrían  estar  involucrados  en  la  biosíntesis  de  los  OLs  en  condiciones  de  crecimiento  limitadas en fosfatos.  8.2.10. Las mutantes en el gen smc02490 de S. meliloti 1021  Mediante  el  análisis  de  hibridación  tipo  Southern  (fig.  19)  se  demostró  que  5  candidatas a mutantes en el gen smc02490, tenían remplazado el gen smc02490 por el gen  cat que confiere  resistencia a cloranfenicol. Debido a que  tenían el perfil de hibridación  esperado descrito en procedimientos experimentales de  los  resultados  adicionales. Una  mutante, NG8, se seleccionó para estudios posteriores. El plásmido pNG39 se transfirió a  la cepa NG8, así como a  la cepa silvestre para posteriores estudios que se  realizaran en  medio mínimo limitado en fosfatos, debido a que se ha reportado que en esa condición se  incrementa  la  transcripción  del  gen  smc02490.  Aunque  para  verificar  si  NG8  todavia  produce  OL  también  podríamos  introducirle  el  plásmido  pJG21  (expresa  a  olsBSm)  y  crecerla en un medio de cultivo con concentraciones de fosfato no limitantes.    132   Figura  19.  Perfil  de  hibridación  tipo  Southern  de  candidatas  a  mutantes  en  el  gen  smc02490.  Marcador  molecular  de  1  kb  (Fermentas)  (línea  M1),  DNA  genómico  digerido  con  EcoRI  de:  candidata  a  mutante  en  smc02490  NG8  (línea  1),  candidata  a  mutante  en  smc02490  NG9  (línea  2),  candidata  a  mutante  en  smc02490 NG10  (línea 3), candidata a mutante en smc02490 NG11  (línea  4),  candidata  a  mutante  en  smc02490  NG12  (línea  5),  marcador molecular de 1 kb (Invitrogen) (línea M2), cepa silvestre S.  meliloti 1021 (línea S).       133 9. Discusión  Los  lípidos de ornitina estándar consisten de un residuo del aminoácido ornitina y  de  dos  ácidos  grasos.  Un  ácido  graso  está  amidificado  a  la  ornitina  y  el  otro  está  esterificado  al  C3  del  acilo  amidificado.  En  algunas  bacterias,  los  OLs  estándar  (sin  modificaciones  con  grupos  polares)  se  incrementan  cuando  se  crecen  en  condiciones  limitadas en nutrimentos como el fosfato [Geiger et al., 1999], cationes divalentes [Wee y  Wilkinson,  1988]  o  cuando  se  interrumpe  la  ruta  biosintética  de  otros  lípidos  de  membrana  sin  fósforo  como el  sulfonolípido  [Pitta et  al., 1989]  y el diacilgliceril N,N,N‐ trimetilhomoserina [López‐Lara et al., 2005].  Sin embargo, muchas bacterias, entre ellas nuestro modelo de estudio, el patógeno  oportunista B. cenocepacia  J2315, sintetizan OLs aún cuando son crecidas en medios de  cultivo con abundante fósforo y cationes divalentes [Palacios‐Chaves et al., 2011; Vences‐ Guzman  et  al.,  2011; González‐Silva  et  al.,  2011].  Esto  sugiere  que  la  regulación de  los  genes  involucrados en  la biosíntesis de  los OLs es diferente entre  los distintos grupos de  bacterias que los poseen. La biosíntesis de los OLs estándar se elucidó en S. meliloti 1021 y  las enzimas OlsA, OlsB y la acil‐ACP están involucradas [Weissenmayer et al., 2002; Gao et  al., 2004]. Adicionalmente, algunos grupos de bacterias producen enzimas que  tienen  la  capacidad de hidroxilar  los OLs estándar en distintas partes de  la molécula produciendo  nuevas  clases más polares de OLs.  La biosíntesis de  las  clases hidroxiladas  conocidas  se  dilucidó en R.  tropici CIAT899 e  incluye a  las enzimas OlsC  [Rojas‐Jiménez et al., 2005] y  OlsE [Vences‐Guzmán et al., 2011] además de OlsA, OlsB y la acil‐ACP.  Se  ha  observado  que  en  R.  tropici  CIAT899  los  lípidos  de  ornitina  estándar  y  hidroxilados  incrementan  la  tolerancia  de  la  bacteria  a  estreses  como  pH  ácido  y  altas  temperaturas, además de estar involucrados en la eficiencia simbiótica [Vences‐Guzmán et  al.,  2011].  En  un  estudio  anterior  también  se  reportó  que  Burkholderia  cepacia  NCTC  10661  forma  2‐OH‐PE  además  de  los  2‐OH‐OLs  y  que  ambas  clases  de  lípidos  2‐ hidroxilados  se  incrementaban  cuando  esta  bacteria  se  crecía  en  altas  temperaturas  [Taylor  et  al.,  1998].  Un  fenómeno  similar  ocurre  en  B.  cenocepacia  J2315  (datos  no  mostrados).  134 Al  inicio  de  este  trabajo  nos  planteamos  identificar  el  gen  que  codifica  para  la  enzima que hidroxila a los OLs estándar de B. cenocepacia J2315 para convertirlos en los 2‐ OH‐OLs,  los  cuales  son  análogos  a  la  clase  de  OLs  P1  de  R.  tropici  CIAT899.  Nosotros  buscamos e identificamos en el genoma de B. cenocepacia J2315 dos genes (bcam1214 y  olsD) que codificaban para proteínas similares a LpxO de S. typhimurium, se clonaron en el  vector  de  expresión  pET9a  y  subsecuentemente  en  el  pRK404.  La  transcripción  de  los  genes quedó regulada por el promotor T7. Después los plásmidos que contenían los genes  se introdujeron en S. meliloti 1021 y en B. cenocepacia J2315. Los genes regulados por el  promotor T7 al  introducirlos en S. meliloti 1021 y en B. cenocepacia J2315 se transcriben  eficientemente. Al  introducir el plásmido que contenía al gen olsD en S. meliloti 1021 se  formó un nuevo lípido (NL) positivo a la tinción con ninhidrina. Sin embargo, la mutante en  olsD  (AQ3)  derivada  de  B.  cenocepacia  J2315  aún  formó  las  dos  clases  de  lípidos  de  ornitina  (OLs  y  2‐OH‐OLs)  y  no  mostró  ninguna  alteración  en  el  perfil  de  lípidos  de  membrana respecto a la cepa parental cuando se crecieron en condiciones normales (LB a  30  ⁰C en agitación). Eso sugería que OlsD estaba  involucrada en biosintetizar otra nueva  clase de lípidos, y que no se expresa en las condiciones normales de crecimiento.  Además, al introducir el plásmido que contenía al gen olsD en B. cenocepacia J2315  se  formaron  dos  nuevos  lípidos  adicionales  (NL1  y  NL2)  positivos  a  la  tinción  con  ninhidrina. Los nuevos  lípidos se movieron diferentemente en 2D‐TLC a  todos  los  lípidos  de  membrana  previamente  caracterizados  de  esa  bacteria.  Eso  era  consistente  con  la  presunción  de  que  los  lípidos  dependientes  de  olsD  eran  diferentes  a  los  lípidos  de  membrana conocidos de B. cenocepacia J2315.  Por análisis de espectrometría de masas se mostró que los lípidos NL1 y NL2 que se  formaron en B.  cenocepacia  J2315  y NL en  S. meliloti 1021 dependientes de OlsD eran  versiones de OLs hidroxilados en un carbono del acilo amidificado de OLs preexistentes en  ambos organismos. Lo anterior, también se confirmó al analizar la movilidad en TLC de los  lisolípidos  (LOLs) derivados de  los OLs estándar  y de  los OLs modificados por OlsD.  Los  lisolípidos  derivados  de  los  OLs  modificados  por  OlsD  se  movieron  menos  en  la  TLC  respecto a  los LOLs derivados de  los OLs estándar. Esto sugería que  los LOL derivados de  135 los OLs modificados por OlsD eran más polares, probablemente porque OlsD introduce el  grupo hidroxilo en el acilo amidificado o en el  residuo de ornitina,  los cuales,  forman el  LOL.  La  existencia  en  la  naturaleza  de  OLs  hidroxilados  en  un  carbono  de  su  acilo  amidificado no había sido  reportada previamente. Probablemente debido a que el perfil  de lípidos de membrana de la cepa silvestre no había sido analizado en la condición que se  expresa  el  gen  olsD.  Nosotros  observamos  pequeñas  cantidades  de  los  nuevos  lípidos  dependientes OlsD (NL1 y NL2) cuando la cepa silvestre B. cenocepacia J2315 se creció en  LB a pH 4 a 30 ⁰C en agitación, los cuales, no fueron observados en la mutante AQ3 crecida  en las mismas condiciones. Una baja producción (menor a 1.5 %) de cada uno los OLs NL1  y NL2 también se observó cuando se introdujo el plásmido que contenía el gen olsD en la  cepa mutante en OlsD (AQ3) [González‐Silva et al., 2011]. El dato anterior es sorprendente  debido  a  que  existe  el  antecedente  de  que  los  genes  regulados  por  el  promotor  T7  se  transcriben eficientemente en B. cenocepacia  J2315 y  la cantidad de  sustratos  (OLs y 2‐ OH‐OLs)  in vivo parece no ser  limitante. Nosotros pensamos que existe  la posibilidad de  que una fracción considerable de los OLs NL1 y NL2, producidos por la célula tal vez están  unidos a una molécula hidrofílica y por esa razón el método de extracción usado puede ser  ineficiente y que a eso se debería que observamos pequeñas cantidades de estos lípidos.  Recientemente  se  ha  reportado  que  los  OLs  S1  (estándar)  y  P1  (2‐OH‐OL,  hidroxilado en el C2 del acilo esterificado) se incrementan considerablemente en R. tropici  CIAT899 cuando  se crece en pHs ácidos de 4.5 y 4.0  [Vences‐Guzmán et al., 2011]. Esto  sugiere que los genes olsD de B. cenocepacia J2315 y olsC de R. tropici CIAT899 que están  involucrados en  síntesis de OLs hidroxilados en el acilo amidificado y en el esterificado,  respectivamente, se expresan en condiciones ácidas de crecimiento [González‐Silva et al.,  2011; Vences‐Guzmán et al., 2011]. Estos resultados son consistentes con la presunción de  que  en  ciertas  condiciones  de  crecimiento  en  pH  ácido,  altas  temperaturas  y  bajas  concentraciones de cationes divalentes, los OLs estándar o modificados podrían remplazar  al  LPS,  porque  en  esas  condiciones  la  estabilidad  del  LPS  en  la  membrana  no  estaría  garantizada, debido a  la  repulsión electrostática entre  las moléculas adyacentes de LPSs  136 con cargas negativas [Nikaido, 2003]. En esas condiciones  los OLs hidroxilados en el acilo  amidificado  y  en  el  esterificado  podrían  estar  formando  enlaces  de  hidrógeno  entre  moléculas  adyacentes.  Por  lo  tanto,  los  OLs  hidroxilados  podrían  incrementar  la  impermeabilidad de las membranas a ciertos iones o moléculas.  Al  invadir macrófagos de humanos y sobrevivir en su  interior B. cenocepacia J2315 se  podría  enfrentar  a  condiciones  adversas  que  presumiblemente  existen  en  los  fagolisosomas de los macrófagos como pH ácido y limitación de nutrimentos entre ellos el  Mg+  [Guo  et  al.,  1997]  además  de  temperaturas  mayores  a  37  ⁰C.  Condiciones  que  favorecen  la  síntesis  de  los  OLs  estándar  o  sus  formas  modificadas  en  diferentes  organismos.  En  la  cepa  de  Burkholderia  cepacia  NCTC  10661  solamente  las  clases  de  lípidos  de  ornitina  (OLs  y  2‐OH‐OLs)  se  detectaron  cuando  esta  cepa  se  creció  en  condiciones  limitadas  en  fósforo  [Taylor  et  al.,  1998].  Lo  que  sugiere  que  en  esas  condiciones  los OLs y  sus  formas modificadas de B. cenocepacia  J2315 podrían  jugar un  papel  protagónico  e  indispensable  para  la  sobrevivencia  de  esta  bacteria  en  esas  condiciones.  Los  OLs  de  muchas  bacterias  tienen  la  propiedad  de  activar  el  sistema  inmune  de  mamíferos porque son reconocidos como marcadores moleculares asociados a patógenos  [Kawai  y  Yano,  1983;  Kawai  et  al.,  1988;  Kawai  et  al.,  1988  b;  Kawai  et  al.,  1999].  Contradictoriamente, Palacios‐Chaves et al. (2011) reportaron recientemente que  los OLs  de B. abortus carecen de un marcador de patrón molecular asociado a patógeno y que no  otorgan ventaja a B. abortus para multiplicarse  intracelularmente  [Palacios‐Chaves et al.,  2011]. Sin embargo, los OLs de B. cenocepacia J2315 no son idénticos a los de B. abortus y  no se ha estudiado si  los OLs B. cenocepacia J2315 están  involucrados de alguna manera  en  inducir  o  en  resistir  la  respuesta  inmune  de  mamíferos.  Por  lo  tanto,  se  necesita  evidencia  experimental  para  determinar  si  los  OLs  o  sus  formas  modificadas  están  involucrados en la patogenicidad de esta bacteria.  Además  de  la  diversidad  de  especies  moleculares  de  OLs  generada  por  las  hidroxilaciones,  los  OLs  también  son  heterogéneos  en  sus  ácidos  grasos.  Un  resultado  consistente de este trabajo con otros estudios previos es que el ácido graso amidificado de  137 los OLs de B. cenocepacia J2315 es de C16:0, como también  lo es en otros miembros del  género  Burkholderia  [Phung  et  al.,  1995;  Taylor  et  al.,  1998].  Los  ácidos  grasos  esterificados de los OLs de B. cenocepacia J2315 son muy variables e incluyen tamaños de  entre C16 a C19 y algunos de ellos tienen insaturaciones que también han sido reportados  en  otros  miembros  del  género  Burkholderia  [Phung  et  al.,  1995;  Taylor  et  al.,  1998].  Basado en nuestros resultados  la enzima OlsB de B. cenocepacia J2315 pareciera ser una  N‐aciltransferasa específica para  transferir acilos de C16 y  la OlsB de S. meliloti 1021 de  C18.  Para  obtener  más  OL  modificado  por  OlsD  para  usarlo  para  análisis  estructural  y  determinar la posición de la hidroxilación, así como para comprobar que OlsD depende de  los mismos componentes (cofactores) para su actividad que LpxO, nosotros desarrollamos  un  ensayo  asumiendo  que  OlsD  era  una  dioxigenasa  dependiente  de  Fe2+/O2/α‐ cetoglutarato.  [14C]OL  se  usó  como  sustrato  aceptor  en  presencia  de  los  cofactores  apropiados y diferentes extractos  libres de células provenientes de cepas de E. coli y  los  productos se analizaron por TLC. Al  incubar [14C]OL con el bufer y extractos crudos  libres  de células de E. coli que contenían a pNG40 que expresaba OlsD, se observó que se formó  el compuesto dependiente de OlsD  (N‐acyl‐OH‐OL) y que se movió menos en TLC que el  sustrato. El compuesto N‐acyl‐OH‐OL no se formó cuando [14C]OL y el bufer se incubó con  extractos crudos libres de células de E. coli que contenían el plásmido vacío. Al separar por  ultracentrifugación  la  fracción  soluble  de  proteínas  y  la  insoluble  (membranas)  de  los  extractos  crudos  libres  de  células  y  evaluar  su  actividad  se  mostró  que  mucha  de  la  proteína OlsD está asociada  con  la  fracción de membrana y una menor  cantidad  con  la  fracción soluble de proteínas. Análisis de  la secuencia de OlsD con el algoritmo TMHMM  [Krogh  et  al.,  2001]  que  predice  hélices  transmembranales  muestra  que  esta  proteína  carece de hélices α (fig. 20) que atraviesan la membrana y probablemente a eso se debe  que parte de  la proteína se disocie de  la fracción de membrana. Aunque para OlsD si se  predice  una  pequeña  región  hidrofóbica  en  el  extremo  C‐terminal  que  no  atraviesa  la  membrana (fig. 20), la cual probablemente determina la asociación mayoritaria de OlsD a  la membrana.  138 En el ensayo de LpxO se adicionaron fosfolípidos (0.5 mg/ml) de E. coli además de los  contenidos  en  las  membranas  [Gibbons  et  al.,  2008].  Nosotros  para  OlsD  solamente  adicionamos  los  fosfolípidos  (alrededor  de  10  µg)  que  contenían  los  extractos  crudos  libres de células provenientes de E. coli. La actividad de OlsD sobre los OLs parece ocurrir  después de  las actividad de OlsA y también de  la actividad como OlsC en B. cenocepacia  J2315 ya que OlsD fue capaz de modificar in vitro y en vivo a los OLs [González‐Silva et al.,  2011] y a  los 2‐OH‐OLs  (fig. 12). Además, OlsD no es específica para hidroxilar OLs de B.  cenocepacia  J2315  en  el  que  acilo  amidificado  es  de  C16,  sino  que  también  muestra  flexibilidad para usar como sustrato  in vivo a  los OLs de S. meliloti 1021 en  los cuales el  acilo amidificado es de C18 [González‐Silva et al., 2011].      Figura. 20. Predicción de hélices transmembranales de OlsD con el algoritmo TMHMM [Krogh et al., 2001].    Se observó que la actividad de OlsD es dependiente de α‐cetoglutarato, la omisión de  Fe2+ adicional no afecta la actividad de OlsD, y la dependencia de este ion metálico solo se  observó cuando se adicionó el quelante de fierro (2,2’‐bipyridyl). La omisión de ascorbato  disminuyó considerablemente la actividad de OlsD. Gibbons et al. (2008) reportaron que la  inclusión  de  catalasa  en  el  ensayo  in  vitro  para  LpxO  eliminaba  parcialmente  el  requerimiento de ascorbato con lo que demostraron que ascorbato no es necesario para la  actividad de LpxO. Lo anterior es consistente con la actividad in vivo de LpxO en E. coli la  cual  no  forma  ascorbato  [Gibbons  et  al.,  2008].  En  ausencia  de  ditiotreitol  (DTT)  la  actividad  OlsD  se  incrementó  más  del  50  %.  Esto  sugiere  que  el  ambiente  químico  139 generado por la presencia de ese compuesto afecta negativamente la actividad de OlsD, lo  que posiblemente se debe a que el DTT tiene la capacidad de reducir parte del O2 disuelto  en el bufer a H2O [Usha y Ramasarma, 1982], y probablemente a que el O2 se requiere en  la reacción de hidroxilación. Consecuentemente el DTT podría provocar que el O2 disuelto  sea un componente  limitante para  la velocidad de  la  reacción, posiblemente haciéndola  más  lenta.  Aunque  nosotros  no  evaluamos  si  OlsD  requiere  de  O2  para  realizar  su  actividad, debido a  la  similitud que comparte con LpxO y a que al menos demostramos  que  dos  componentes  de  la  reacción  (Fe2+/α‐cetoglutarato),  al  igual  que  para  LpxO,  también  son necesarios en el proceso  catalítico de OlsD,  creemos que OlsD  también es  dependiente de Fe2+/O2/α‐cetoglutarato al igual que LpxO.  Comparando  el  porcentaje  de  conversión  de  sustrato  a  producto  se  observó  que  la  enzima OlsC es más eficiente para hidroxilar el C2 del acilo esterificado de  los OLs que  la  enzima  OlsD  que  hidroxila  un  carbono  del  acilo  amidificado  también  de  los  OLs.  Cabe  recordar que  los  ensayos  enzimáticos de OlsC  se  realizaron  con  0.4 mg/ml de proteína  total  y  los  ensayos  de  OlsD  con  1  mg/ml.  Cuando  nosotros  usamos  1  mg/ml  para  los  ensayos  de  OlsC  obtuvimos  un  porcentaje  de  conversión  mayor  a  80  %  (datos  no  mostrados). Otra diferencia de  los ensayos presentados es que con OlsD se usó 0.1 % de  Triton X‐100 y con OlsC 0.2 %.  Recientemente se han identificado algunos de los genes que codifican para las enzimas  involucradas  en  la  biosíntesis  de  los OLs.  Sin  embargo,  no  se  ha  reportado  ningún  gen  involucrado en la biodegradación de esos lípidos. Este tema espera ser estudiado.      140 10. Conclusiones  En esta  tesis doctoral  se  identificó el  gen olsD  y  se  caracterizó  su producto.  Los  resultados presentados nos permiten concluir:  1. El gen olsD codifica para una enzima que está  involucrada en hidroxilar  los  lípidos de  ornitina de B. cenocepacia J2315.  2. La enzima OlsD hidroxila el acilo amidificado de los lípidos de ornitina (OLs y 2‐OH‐OLs)  de  B.  cenocepacia  J2315  para  formar  otros  dos  lípidos  adicionales más  polares  (NL1  y  NL2). La presencia de los genes olsB, olsA, un gen que codifica para una enzima que hace  una actividad como OlsC y olsD en B. cenocepacia J2315  la capacitan para formar cuatro  clases diferentes de lípidos ornitina.  3.  La hidroxilación por  la enzima OlsD ocurre  cuando B.  cenocepacia  J2315  se  crece en  condiciones ácidas y OlsD requiere de Fe2+/α‐cetoglutarato para realizar su actividad.  4. Para que se produzca cualquier clase de lípidos de ornitina en B. cenocepacia J2315 se  requiere que el gen olsB esté integro.  5.  La  hipótesis  que  nos  planteamos  es  falsa,  debido  a  que  ninguno  de  los  dos  genes  (bcam1214 o olsD) de B. cenocepacia  J2315 homólogos a LpxO de S.  typhimurium están  involucrados en hidroxilar a los OLs estándar para convertirlos en los 2‐OH‐OLs.    11. Perspectivas  Los resultados y conclusiones generados en esta tesis doctoral, dan lugar a estudiar  los siguientes aspectos:  • Determinar el carbono del acilo amidificado de los OLs que es hidroxilado por OlsD.  • Comprobar si OlsD requiere de O2 para realizar su actividad.  • Determinar si los lípidos de ornitina en B. cenocepacia J2315 están involucrados en la  patogenicidad de esta bacteria.  • Estudiar la función que desempeñan los lípidos de ornitina en la membrana de B.  cenocepacia J2315.  141 12. Referencias bibliográficas  Asselineau,  J.  (1991) Bacterial  lipids containing amino acids or peptides  linked by amide  bonds. Chem. Org. Naturst. 56, 1‐85.  Aygun‐Sunar, S., Bilaloglu, R., Goldfine, H., and Daldal, F.  (2007) Rhodobacter capsulatus  OlsA  is  a  bifunctional  enyzme  active  in  both  ornithine  lipid  and  phosphatidic  acid  biosynthesis. J. Bacteriol. 189, 8564‐8574.  Aygun‐Sunar, S., Mandaci, S., Koch, H.‐G., Murria, I. V. J., Goldfine, H., and Daldal, F. (2006)  Ornithine  lipid  is  required  for  optimal  steady‐state  amounts  of  c‐type  cytochromes  in  Rhodobacter capsulatus. Mol. Microbiol. 61, 418‐435.  Basconcillo, L. S., and McCarry, B. E. (2008) Comparison of three GC/MS methodologies for  the analysis of fatty acids in Sinorhizobium meliloti: Development of a micro‐scale, one‐vial  method. J. Chromatogr. B. 871, 22‐31.  Basconcillo, L. S., Zaheer, R., Finan, T. M., and McCarry, B. E. (2009) A shotgun  lipidomics  study of a putative lysophosphatidic acid acyl transferase (PlsC) in Sinorhizobium meliloti. J.  Chromatogr. B Analyt. Technol. Biomed. Life Sci. 877, 2873‐2882.  Batrakov,  S.  G.,  Nikitin,  D.  I.,  Mosezhnyi,  A.  E.,  and  Ruzhitsky,  A.  O.  (1999)  A  glycine‐ containing  phosphorus‐free  lipoaminoacid  from  the  gram‐negative  marine  bacterium  Cyclobacterium marinus WH. Chem. Phys. Lipids 99, 139‐143.  Batrakov, S. G., Nikitin, D.  I., Sheichenko, VI, and Ruzhitsky, A. O. (1998) A novel sulfonic‐ acid  analogue  of  ceramide  is  the  major  extractable  lipid  of  the  gram‐negative  marine  bacterium Cyclobacterium marinus WH. Biochim. Biophys. Acta 1391, 79‐91.  Beringer, J. E. (1974) R factor transfer in Rhizobium leguminosarum. J. Gen. Microbiol. 84,  188‐198.  Bligh, E. G., and Dyer, J. W. (1959) A rapid method of total lipid extraction and purification.  Can. J. Biochem. Physiol. 37, 911‐917.  Brady,  S.  F.,  and Clardy,  J.  (2005) N‐acyl derivatives of  arginine  and  tryptophan  isolated  from environmental DNA expressed in Escherichia coli. Org. Lett. 7, 3613‐3616.  Brady,  S.  F.,  Chao,  C.  J.,  and  Clardy,  J.  (2002)  New  natural  product  families  from  an  environmental DNA (eDNA) gene cluster. J. Am. Chem. Soc. 124, 9968‐9969.  Clardy, J., and Brady, S. F. (2007) Cyclic AMP directly activates NasP, an N‐acyl amino acid  antibiotic biosynthetic enzyme cloned from an uncultured β‐proteobacterium. J. Bacteriol.  189, 6487‐6489.  Cullinane,  M.,  Baysse,  C.,  Morrissey,  J.  P.,  and  O’Gara,  F.  (2005)  Identification  of  two  lysophosphatidic  acid  acyl  transferase  genes with  overlapping  function  in  Pseudomonas  fluorescens. Microbiology 151, 3071‐3080.  142 Dees, C., and Shively,  J. M.  (1982) Localization and quantitation of  the ornithine  lipid of  Thiobacillus thiooxidans. J. Bacteriol. 149, 798‐799.  Donovan, C., and Bramkamp, M.  (2009) Characterization and subcellular  localization of a  bacterial flotillin homologue. Microbiology 155, 1786‐1799.  Dorrer  E.,  and  Teuber,  M.  (1977)  Induction  of  polymyxin  resistance  in  Pseudomonas  fluorescens by phosphate limitation. Arch. Microbiol. 114, 87‐89.  Dowhan,  W.,  Bogdanov,  M.,  and  Mileykovskaya,  E.  (2008)  Functional  roles  of  lipids  in  membranes. In: Vance, D. E., and Vance, J. E., editors. Biochemistry of Lipids, Lipoproteins  and Membranes (5th Edition). Amsterdam, The Netherlands: Elsevier. p. 1‐37.  Facey, S. J., and Kuhn, A. (2010) Biogenesis of bacterial inner‐membrane proteins. Cell Mol.  Life Sci. 67, 2343‐2362.  Galbraith, L., Jonsson, M. H., Rudhe, L. C., and Wilkinson, S. G. (1999) Lipids and fatty acids  of Burkholderia and Ralstonia species. FEMS Microbiol. Lett. 173, 359‐364.  Gao, J.‐L., Weissenmayer, B., Taylor, A. M., Thomas‐Oates, J., López‐Lara, I. M., and Geiger,  O.  (2004)  Identification  of  a  gene  required  for  the  formation  of  lyso‐ornithine  lipid,  an  intermediate  in  the biosynthesis of ornithine‐containing  lipids. Mol. Microbiol. 53, 1757‐ 1770.  Geiger, O., González‐Silva N.,  López‐Lara,  I. M.,  and  Sohlenkamp, C.  (2010) Amino  acid‐ containing membrane lipids in bacteria. Prog. Lipid Res. 49, 46‐60.  Geiger, O., Röhrs, V., Weissenmayer, B., Finan, T. M., and Thomas‐Oates,  J. E.  (1999) The  regulator  gene  phoB  mediates  phosphate  stress‐controlled  synthesis  of  the  membrane  lipid diacylglyceryl‐N,N,N‐trimethylhomoserine in Rhizobium (Sinorhizobium) meliloti. Mol.  Microbiol. 32, 63‐73.  Gibbons, H. S., Kalb, S. R., Cotter, R. J., and Raetz, C. R. H. (2005) Role of Mg2+ and pH in the  modification  of  Salmonella  lipid A  after  endocytosis  by macrophage  tumour  cells. Mol.  Microbiol. 55, 425‐440.  Gibbons, H. S., Lin, S., Cotter, R. J., and Raetz C. R. H. (2000) Oxygen requirement for the  biosynthesis  of  the  S‐2‐hydroxymyristate  moiety  in  Salmonella  typhimurium  lipid  A.  Function of LpxO, a new Fe2+/α‐ketoglutarate‐dependent dioxygenase homologue. J. Biol.  Chem. 275, 32940‐32949.  Gibbons, H. S., Reynolds, C. M., Guan, Z., and Raetz, C. R. H.  (2008) An  inner membrane  dioxygenase  that  generates  the  2‐hydroxymyristate  moiety  of  Salmonella  lipid  A.  Biochemistry 47, 2814‐2825.  Gomi, K., Kawasaki, K., Kawai, Y., Shiozaki, M., and Nishijima, M. (2002) Toll‐like receptor 4‐ MD‐2  complex  mediates  the  signal  transduction  induced  by  flavolipin,  an  amino  acid‐ containing lipid unique to Flavobacterium meningosepticum. J. Immunol. 168, 2939‐2943.  143 González‐Silva,  N.,  López‐Lara,  I.  M.,  Reyes‐Lamothe,  R.,  Taylor,  A.  M.,  Sumpton,  D.,  Thomas‐Oates,  J.,  and  Geiger,  O.  (2011)  The  dioxygenase‐encoding  olsD  gene  from  Burkholderia cenocepacia causes the hydroxylation of the amide‐linked fatty acyl moiety of  ornithine‐containing membrane lipids. Biochemistry 50, 6396‐6408.  Gough,  J.,  Karplus,  K.,  Hughey,  R.,  and  Chothia,  C.  (2001)  Assignment  of  homology  to  genome sequences using a library of hidden Markov models that represent all proteins of  known structure. J. Mol. Biol. 313, 903‐919.  Gould, T. A., Schweizer, H. P., and Churchill, M. E. A. (2004) Structure of the Pseudomonas  aeruginosa acylhomoserine lactone synthase LasI. Mol. Microbiol. 53, 1135‐1146.  Guo, L., Lim, K. B., Gunn, J. S., Bainbridge, B., Darveau, R. P., Hackett, M., and Miller, S.  I.  (1997)  Regulation  of  lipid  A  modifications  by  Salmonella  typhimurium  virulence  genes  phoP‐phoQ. Science 276, 250‐253.  Hanahan, D.  (1983)  Studies on  transformation of  Escherichia  coli with plasmids.  J. Mol.  Biol. 166, 557‐580.   Heath,  R.  J.,  and  Rock,  C.  O.  (1998)  A  conserved  histidine  is  essential  for  glycerolipid  acyltransferase catalysis. J. Bacteriol. 180, 1425‐1430.  Hoischen, C., Gura, K., Luge, C., and Gumpert, J. (1997) Lipid and fatty acid composition of  cytoplasmic membranes from Streptomyces hygroscopicus and its stable protoplast‐type L  form. J. Bacteriol. 179, 3430‐3436.  Jansen, G. A., Hogenhout, E. M., Ferdinandusse, S., Waterham, H. R., Ofman, R., Jakobs, C.,  Skjeldal, O. H., and Wanders, R. J. (2000) Human phytanoyl‐CoA hydroxylase: resolution of  the gene structure and the molecular basis of Refsum's disease. Hum. Mol. Genet. 9, 1195‐ 1200.  Jansen, G. A., Ofman, R., Ferdinandusse, S., Ijlst, L., Muijsers, A. O., Skjeldal, O. H., Stokke,  O., Jakobs, C., Besley, G. T., Wraith, J. E. and Wanders, R. J. (1997) Refsum disease is caused  by mutations in the phytanoyl‐CoA hydroxylase gene. Nat. Genet. 17, 190‐193.  Kadner, R.  J.  (1996) Cytoplasmic membrane.  In: Escherichia coli and Salmonella. Cellular  and Molecular Biology, 2nd ed.  In: Neidhardt, F. C., editor. ASM press; Washington, DC. p.  58‐87.  Kato,  H.,  and  Goto,  N.  (1997)  Adjuvanticity  of  an  ornithine‐containing  lipid  of  Flavobacterium meningosepticum  as  a  candidate  vaccine  adjuvant. Microbiol.  Immunol.  41, 101‐106.  Kawahara,  K.,  Seydel, U., Matsuura, M., Danbara, H.,  Rietschel,  E.  T.,  and  Zähringer, U.  (1991) Chemical structure of glycosphingolipids isolated from Sphingomonas paucimobilis.  FEBS Lett. 292, 107‐110.  144 Kawai, Y, and Akagawa, K. (1989) Macrophage activation by an ornithine‐containing lipid or  a serine‐containing lipid. Infect. Immun. 57, 2086‐2091.  Kawai, Y. and Yano, I. (1983) Ornithine‐containing lipid of Bordetella pertussis, a new type  of hemagglutinin. Eur. J. Biochem. 136, 531‐538.  Kawai, Y., Kaneda, K., Morisawa, Y., and Akagawa, K. (1992) Protection of mice from lethal  endotoxemia  by  use  of  an  ornithine‐containing  lipid  or  a  serine‐containing  lipid.  Infect  Immun. 1991 59, 2560‐6. Erratum in: Infect. Immun. 60, 320.  Kawai, Y., Nakagawa, Y., Matuyama, T., Akagawa, K., Itagawa, K., Fukase, K., Kusumoto, S.,  Nishijima, M., and Yano,  I.  (1999) A  typical bacterial ornithine‐containing  lipid Nα‐(D)‐[3‐ (hexadecanoyloxy)hexadecanoyl]‐ornithine  is  a  strong  stimulant  for  macrophages  and  a  useful adjuvant. FEMS Immunol. Med. Microbiol. 23, 67‐73.  Kawai,  Y.,  Watanabe,  M.,  Matsuura,  M.,  Nishijima,  M.,  and  Kawahara,  K.  (2002)  The  partially  degraded  lipopolysaccharide  of  Burkholderia  cepacia  and  ornithine‐containing  lipids derived from some Gram‐negative bacteria are useful complex lipid adjuvants. FEMS  Immunol. Med. Microbiol. 34, 173‐179.  Kawai, Y., Yano, I., and Kaneda, K. (1988) Various kinds of lipoamino acids including a novel  serine‐containing  lipid  in an opportunistic pathogen Flavobacterium. Their structures and  biological activities on erythrocytes. Eur. J. Biochem. 171, 73‐80.  Kawai, Y., Yano, I., Kaneda, K. and Yabuuchi, E. (1988b) Ornithinecontaining lipids of some  Pseudomonas species. Eur. J. Biochem. 175, 633‐641  Kawasaki, K., Gomi, K., Kawai, Y., Shiozaki, M., and Nishijima, M. (2003) Molecular basis for  lipopolysaccharide mimetic action of Taxol™ and flavolipin. J. Endotoxin Res. 9, 301‐307.  Kawazoe,  R.,  Monde,  K.,  Reichardt,  W.,  and  Okuyama,  H.  (1992)  Lipoamino  acids  and  sulfonolipids  in  Cytophaga  johnsonae  Stanier  strain  C21  and  six  related  species  of  Cytophaga. Arch. Microbiol. 158, 171‐175.  Kawazoe, R., Okuyama, H., Reichardt, W., and Sasaki, S. (1991) Phospholipids and a novel  glycine‐containing  lipoamino acid  in Cytophaga  johnsonae Stanier strain C21. J. Bacteriol.  173, 5470‐5475.  Keck, M., Gisch, N., Moll, H., Vorhölter, F. J., Gerth, K., Kahmann, U., Lissel, M., Lindner, B.,  Niehaus, K., and Holst, O. (2011) Unusual outer membrane lipid composition of the gram‐ negative, lipopolysaccharide‐lacking myxobacterium Sorangium cellulosum So ce56. J. Biol.  Chem. 286, 12850‐12859.  Knoche, H. W., and Shively, J. M. (1972) The structure of an ornithine‐containing lipid from  Thiobacillus thiooxidans. J. Biol. Chem. 247, 170‐178.  145 Kovach, M. E., Elzer, P., Hill, D., Robertson, T., Farris, M., Roop, M., and Peterson, K. M.  (1995) Four new derivatives of  the broad‐host‐range cloning vector pBBR1MCS, carrying  different antibiotic‐resistance cassettes. Gene 166, 175‐176.  Kovach, M.  E.,  Phillips,  R. W.,  Elzer,  P. H.,  Roop,  R. M.,  II,  and  Peterson,  K. M.  (1994)  pBBR1MCS: A broad‐host‐range cloning vector. BioTechniques 16, 800–802.  Krogh, A., Larsson, B., von Heijne, G., and Sonnhammer, E. L. L.  (2001) Predicting  trans‐ membrane  protein  topology  with  a  hidden  Markov  model:  application  to  complete  genomes. J. Mol. Biol. 305, 567‐580.  Krol, E., and Becker, A.  (2004) Global transcriptional analysis of the phosphate starvation  response in Sinorhizobium meliloti strains 1021 and 2011. Mol. Genet. Genomics 272, 1‐17.  Lewenza,  S.,  Falsafi, R., Bains, M., Rohs, P.,  Stupak,  J.,  Sprott, G. D.,  and Hancock. R. E.  (2011)  The  olsA  gene  mediates  the  synthesis  of  an  ornithine  lipid  in  Pseudomonas  aeruginosa  during  growth  under  phosphate‐limiting  conditions,  but  is  not  involved  in  antimicrobial peptide susceptibility. FEMS Microbiol. Lett. 320, 95‐102.  López, D., and Kolter, R. (2011) Functional microdomains in bacterial membranes. Genes &  Development 24, 1893‐1902.  López‐Lara, I. M., Gao, J.‐L., Soto, M. J., Solares‐Pérez, A., Weissenmayer, B., Sohlenkamp,  C., Verroios, G. P., Thomas‐Oates, J., Geiger, O. (2005) Phosphorus‐free membrane lipids of  Sinorhizobium meliloti  are  not  required  for  the  symbiosis with  alfalfa  but  contribute  to  increased cell yields under phosphorus‐limiting conditions of growth. Mol. Plant‐Microbe  Interact. 18, 973‐982.  López‐Lara,  I.  M.,  Sohlenkamp,  C.,  and  Geiger,  O.  (2003)  Membrane  lipids  in  plant‐ associated  bacteria:  their  biosyntheses  and  possible  functions.  Mol.  Plant‐Microbe  Interact. 16, 567‐579.  Madigan  M.  T.,  and  Martinko  J.  M.  (2006)  Brock  biology  of  microorganisms.  Pearson  Prentice Hall, 11th ed. U. S. A.  Mahenthiralingam,  E.,  Urban,  T.  A.,  and  Goldberg,  J.  B.  (2005)  The  multifarious,  multireplicon Burkholderia cepacia complex. Nat. Rev. Microbiol. 3, 144‐156.  Maneerat,  S.,  Bamba,  T., Harada,  K.,  Kobayashi, A.,  Yamada, H.,  and  Kawai,  F.  (2006) A  novel  crude  oil  emulsifier  excreted  in  the  culture  supernatant  of  a  marine  bacterium,  Myroides sp. strain SM1. Appl. Microbiol. Biotechnol. 70, 254‐259.  Martínez‐Romero, E., Segovia, L., Mercante, F. M., Franco, A. A., Graham, P., and Pardo, M.  A.  (1991) Rhizobium  tropici,  a novel  species nodulating Phaseolus  vulgaris  L. beans  and  Leucaena sp. trees. Int. J. Syst. Bacteriol. 41, 417‐4126.  146 Meade, H. M., Long, S. R., Ruvkun, G. B., Brown, S. E., and Ausubel, F. M. (1982) Physical  and genetic characterization of symbiotic and auxotrophic mutants of Rhizobium meliloti  induced by transposon Tn5 mutagenesis. J. Bacteriol. 149, 114‐122.  Minnikin,  D.  E.,  and  Abdolrahimzadeh,  H.  (1974)  The  replacement  of  phosphatidylethanolamine  and  acidic  phospholipids  by  an  ornithine‐amide  lipid  and  a  minor phosphorus‐free  lipid  in Pseudomonas  fluorescens NCMB 129. FEBS Lett. 43, 257‐ 260.  Nelson, D. L. and Cox, M. M. (2010) Lehninger principles of Biochemistry. Freeman, 4th ed.  U. S. A.  Nikaido,  H.  (1996)  Outer  membrane.  In:  Escherichia  coli  and  Salmonella.  Cellular  and  Molecular Biology, 2nd ed. In: Neidhardt, F. C., editor. ASM press; Washington, DC. p. 29‐47.  Nikaido, H.  (2003) Molecular  basis  of  bacterial  outer membrane  permeability  revisited.  Microbiol. Mol. Biol. Rev. 67, 593‐656.  Nikitin, D. I., Oranskaya, M. S., Pitryuk, I. A., Chernykh, N. A., and Lysenko, A. M. (1994) A  new ring‐forming bacterium Arcocella aquatica gen. et sp. nov. Microbiology (Moscow) 63,  87‐90 (Engl. Transl.)  Okemoto, K., Hanada, K., Nishijima, M., and Kawasaki, K. (2008) The preparation of a lipidic  endotoxin affects its biological activities. Biol. Pharm. Bull. 31, 1952‐1954.  Okuyama, H.,  and Monde,  K.  (1996)  Identification  of  an  ornithine‐containing  lipid  from  Cytophaga johnsonae Stanier strain C21 by 1H‐NMR. Chem. Phys. Lipids 83, 169‐173.  Palacios‐Chaves, L., Conde‐Álvarez, R., Gil‐Ramírez, Y., Zúñiga‐Ripa, A., Barquero‐Calvo, E.,  Chacón‐Díaz, C., Chaves‐Olarte, E., Arce‐Gorvel, V., Gorvel, J. P., Moreno, E., de Miguel, M.  J.,  Grilló,  M.  J.,  Moriyón,  I.,  and  Iriarte  M.  (2011)  Brucella  abortus  ornithine  lipids  are  dispensable  outer  membrane  components  devoid  of  a  marked  pathogen‐associated  molecular pattern. PLoS ONE 6, e16030.  Phung, L. V., Chi, T. T. B., Hotta, H., Yabuchi, E., and Yano, I. (1995) Cellular  lipid and fatty  acid  compositions  of  Burkholderia  pseudomallei  strains  isolated  from  human  and  environment in Viet Nam. Microbiol. Immunol. 39, 105‐116.  Pitta, T. P., Leadbetter, E. R., and Godchaux, III, W. (1989) Increase of ornithine amino lipid  content in a sulfonolipid‐deficient mutant of Cytophaga johnsonae. J. Bacteriol. 171, 952‐ 957.  Raetz,  C.  R.  H.,  and  Whitfield,  C.  (2002)  Lipopolysaccharide  endotoxins.  Annu.  Rev.  Biochem. 71, 635‐700.  Raetz, C. R. H., Reynolds C. M., Trent M.  S., and Bishop, R. E.  (2007)  Lipid modification  systems in Gram‐negative bacteria. Annu. Rev. Biochem. 76, 295‐329.  147 Raj, H. D., and Maloy, S. R.  (1990) Proposal of Cyclobacterium marinus gen. nov., comb.  nov.  for a marine bacterium previously assigned  to  the genus  Flectobacillus.  Int.  J.  Syst.  Bacteriol. 40, 337‐347.  Rock, C. O. (2008) Fatty acid and phospholipid metabolism in prokaryotes. In: Vance, D. E.,  and Vance, J. E., editors. Biochemistry of Lipids, Lipoproteins and Membranes (5th Edition).  Amsterdam, The Netherlands: Elsevier. p. 59‐96.  Rojas‐Jiménez,  K.,  Sohlenkamp,  C.,  Geiger,  O.,  Martínez‐Romero,  E.,  Werner,  D.,  and  Vinuesa, P.  (2005) A CIC  chloride  channel homolog  and ornithine‐containing membrane  lipids of Rhizobium tropici CIAT899 are involved in symbiotic efficiency and acid tolerance.  Mol. Plant‐Microbe Interact. 18, 1175‐1185.  Rumley, M. K., Therisod, H., Weissborn, A. C., and Kennedy, E. G.  (1992) Mechanism of  regulation  of  biosynthesis  of  membrane‐derived  oligosaccharides  in  Escherichia  coli.  J.  Biol. Chem. 267, 11806‐11810.  Sambrook,  J.,  and  Russell,  D.  W.  (2001)  Molecular  Cloning:  A  Laboratory  Manual.  Cold  Spring Harbor Laboratory Press, Cold Spring Harbor, NY.  Sandoval‐Calderón, M., Geiger, O., Guan, Z., Barona‐Gómez, F., and Sohlenkamp, C. (2009)  A  eukaryote‐like  cardiolipin  synthase  is  present  in  Streptomyces  coelicolor  and  in  most  Actinobacteria. J. Biol. Chem. 284, 17383‐17390.  Schäfer, A., Tauch, A.,  Jäger, W., Kalinowski,  J., Thierbach, G., and Pühler, A.  (1994) Small  mobilizable multi‐purpose cloning vectors derived from the Escherichia coli plasmids pK18  and  pK19:  selection  of  defined  deletions  in  the  chromosome  of  Corynebacterium  glutamicum. Gene 145, 69‐73.  Schüler,  D.  (2004)  Molecular  analysis  of  a  subcellular  compartment:  the  magnetosome  membrane in Magnetospirillum gryphiswaldense. Arch. Microbiol. 181, 1‐7.  Shih, G. C., Kahler, C. M., Swartley, J. S., Rahman, M. M., Coleman, J., Carlson, R. W., and  Stephens,  D.  S.  (1999)  Multiple  lysophosphatidic  acid  acyltransferases  in  Neisseria  meningitidis. Mol. Microbiol. 32, 942‐952.  Shimazu,  R., Akashi,  S., Ogata, H., Nagai,  Y.,  Fukudome,  K., Miyake,  K.,  and  Kimoto, M.  (1999)  MD‐2,  a  molecule  that  confers  lipopolysaccharide  responsiveness  on  Toll‐like  receptor 4. J. Exp. Med. 189, 1777‐1782.  Shiozaki, M., Deguchi, N., Ishikawa, T., Haruyama, H., Kawai, Y., and Nishijima, M. (1998a)  Revised  structure  of  flavolipin  and  synthesis  of  its  isomers.  Tetrahedron  Lett.  39,  4497‐ 4500.  Shiozaki, M., Deguchi, N., Mochizuki, T., Wakabayashi, T., Ishikawa, T., Haruyama, H., Kawai,  Y., and Nishijima, M. (1998b) Revised structure and synthesis of flavolipin. Tetrahedron 54,  11861‐11876.  148 Simon, R., Priefer, U., and Pühler, A. (1983) A broad host range mobilization system for  in  vivo  genetic  engineering:  Transposon  mutagenesis  in  Gram‐negative  bacteria.  Nat.  Biotechnol. 1, 784–791.  Sohlenkamp, C., de Rudder, K. E. E., and Geiger, O. (2004) Phosphatidylethanolamine is not  essential for growth of Sinorhizobium meliloti  in complex culture media. J. Bacteriol. 186,  1667‐1677.  Stone, K. J., and Strominger, J. L. (1971) Mechanism of action of bacitracin: complexation  with metal ion and c55‐isoprenyl pyrophosphate. Proc. Nat. Acad. Sci. 68, 3223‐3227.  Studier, F. W., Rosenberg, A. H., Dunn,  J.  J., and Dubendorff,  J. W.  (1990) Use of T7 RNA  polymerase to direct expression of cloned genes. Methods Enzymol. 185, 60‐89.  Tahara, Y., Kameda, M., Yamada, Y., and Kondo, K.  (1976b) An ornithine‐containing  lipid  isolated from Gluconobacter cerinus. Biochim. Biophys. Acta 450, 225‐230.  Tahara, Y., Shinmoto, K., Yamada, Y., and Kondo, K.  (1978) Enzymatic  synthesis of  tauro‐ ornithine lipid in Gluconobacter cerinus. Agr. Biol. Chem. 42, 205‐206.  Tahara, Y., Yamada, Y., and Kondo, K.  (1976a) A new  lysine‐containing  lipid  isolated  from  Agrobacterium tumefaciens. Agr. Biol. Chem. 40, 1449‐1450.  Taylor,  C.  J.,  Anderson,  A.  J.,  and  Wilkinson,  S.  G.  (1998)  Phenotypic  variation  of  lipid  composition  in  Burkholderia  cepacia:  a  response  to  increased  growth  temperature  is  a  greater content of 2‐hydroxy acids in phosphatidylethanolamine and ornithine amide lipid.  Microbiology 144, 1737‐1745.  Usha, D., S., and Ramasarma T. (1987) Hemin‐mediated oxidation of dithiothreitol reduces  oxygen to H20. Mol. Cell. Biochem. 77, 111‐120.  Van Wagoner, R. M., and Clardy,  J.  (2006) FeeM, an N‐acyl amino acid synthase  from an  uncultured soil microbe: Structure, mechanism, and acyl carrier protein binding. Structure  14, 1425‐1435.  Vences‐Guzmán, M. A., Guan, Z., Ormeño‐Orrillo, E., González‐Silva, N., López‐Lara,  I. M.,  Martínez‐Romero, E., Geiger, O., and Sohlenkamp, C. (2011) Hydroxylated ornithine  lipids  increase stress tolerance in Rhizobium tropici CIAT899. Mol. Microbiol. 79, 1496‐1514.  Vollmer, W., and Bertsche, U.  (2008) Murein  (peptidoglycan)  structure, architecture and  biosynthesis in Escherichia coli. Biochim. Biophys. Acta 1778, 1714‐1734.  Watson, W. T., Minogue, T. D., Val, D. L., von Bodman S. B., and Churchill, M. E. A. (2002)  Structural basis and  specificity of acyl‐homoserine  lactone  signal production  in bacterial  quorum sensing. Mol. Cell. 9, 685‐694.  149 Wee, S., and Wilkinson, B. J. (1988)  Increased outer membrane ornithine‐containing  lipid  and  lysozyme  penetrability  of  Paracoccus  denitrificans  grown  in  a  complex  medium  deficient in divalent cations. J. bacterial. 170, 3283‐3286.  Weissenmayer, B., Gao, J.‐L., López‐Lara I. M., and Geiger O. (2002) Identification of a gene  required for the biosynthesis of ornithine‐derived lipids. Mol. Microbiol. 45, 721‐733.  West,  J., Tompkins, C. K., Balantac, N., Nudelman, E., Meengs, B., White, T., Bursten, S.,  Coleman, J., Kumar, A., Singer, J. W., and Leung, D. W. (1997) Cloning and expression of two  human  lysophosphatidic  acid  acyltransferase  cDNAs  that  enhance  cytokine‐induced  signaling responses in cells. DNA Cell. Biol. 16, 691‐701.  Zhang, H. M.,  Li, Z., Tsudome, M.,  Ito, S., Takami, H., and Horikoshi, K.  (2005) An alkali‐ inducible flotillin‐like protein from Bacillus halodurans C‐125. Protein J. 24, 125‐131.  Zhang, X., Ferguson‐Miller, S. M., and Reid, G. E. (2009) Characterization of ornithine and  glutamine  lipids extracted  from cell membranes of Rhodobacter sphaeroides.  J. Am. Soc.  Mass. Spectrom. 20, 198‐212.  150