1 UNIVERSIDAD NACIONAL AUTÓNOMA DE MÉXICO POSGRADO EN CIENCIAS BIOLÓGICAS INSTITUTO DE BIOLOGÍA ECOLOGÍA Efecto del huracán Patricia en el ensamble de las comunidades de hongos del suelo y de la red micorrízica en el bosque Neotro- pical caducifolio TESIS QUE PARA OPTAR POR EL GRADO DE: DOCTORA EN CIENCIAS BIOLÓGICAS PRESENTA: JULIETA ALVAREZ MANJARREZ TUTOR PRINCIPAL DE TESIS: DR. ROBERTO GARIBAY ORIJEL INSTITUTO DE BIOLOGÍA, UNAM COMITÉ TUTOR: DR. MIGUEL MARTÍNEZ RAMOS INSTITUTO DE INVESTIGACIONES EN ECOSISTEMAS Y SUS- TENTABILIDAD, UNAM DR. ALFONSO OCTAVIO DELGADO SALINAS INSTITUTO DE BIOLOGÍA, UNAM MÉXICO, CD. MX. DICIEMBRE, 2019 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. 2 POSGRADO DAME AOS * BIOLÓGICAS * COORDINACIÓN DEL POSGRADO EN CIENCIAS BIOLÓGICAS INSTITUTO DE BIOLOGÍA OFICIO CPCB/Folio de la Coord./2019 ASUNTO: Oficio de Jurado M. en C. Ivonne Ramírez Wence Directora General de Administración Escolar, UNAM Presente Me permito informar a usted que en la reunión ordinaria del Subcomité de Biología Experimental y Biomedicina del Posgrado en Ciencias Biológicas, celebrada el día 7 de octubre de 2019 se aprobó el siguiente jurado para el examen de grado de DOCTORA EN CIENCIAS de la estudiante ALVAREZ MANJARREZ JULIETA con número de cuenta 513014583 con la tesis titulada “Efecto del huracán Patricia en el ensamble de las comunidades de hongos del suelo y de la red micorrízica en el bosque Neotropical caducifolio”, realizada bajo la dirección del DR. ROBERTO GARIBAY ORIJEL, quedando integrado de la siguiente manera: Presidente: DR. ALFONSO VALIENTE BANUET Vocal: DRA. HERMELINDA MARGARITA VILLEGAS RIOS Secretario: DR. ALFONSO OCTAVIO DELGADO SALINAS Suplente: DRA. PATRICIA VELEZ AGUILAR Suplente DRA. MARIANA BENITEZ KEINRAD Sin otro particular, me es grato enviarle un cordial saludo. ATENTAMENTE , “POR MI RAZA HABLARA EL ESPIRITU” Ciudad Universitaria, Cd. Mx., a 14 de noviembre de 2019 COORDINADOR DEL PROGRAMA CAENCIA DR. ADOLFO GERARDO NAVARRO SIGUENZA c. Cc. p. Expediente del alumno COORDINACIÓN DEL POSGRADO EN CIENCIAS BIOLOGICAS UNIDAD DE POSGRADO Edificio D, 1? Piso. Circuito de Posgrados, Ciudad Universitaria Alcaldía Coyoacán. C. P. 04510 CDMX Tel. (+5255)5623 7002 http://pcbiol.posgrado.unam.mx/ 3 4 5 AGRADECIMIENTOS Agradezco al Posgrado en Ciencias Biológicas por permitirme empezar una maes- tría y concluir con un doctorado. Agradezco al Consejo Nacional de Ciencia y Tecnología (CONACYT) por el finan- ciamento del Proyecto de Ciencia Básica 239266, por la beca doctoral CONACYT 404413 y beca mixta CONACYT con número de convocatoria 291250. Hago un especial agradecimiento al Dr. Roberto Garibay Orijel por su apoyo in- condicional, su excelente dirección durante todos mis estudios de posgrado, por los comentarios oportunos y la confianza depositada. Agradezco a los doctores Alfonso Octavio Delgado Salinas y Miguel Martínez Ra- mos por ser parte de mi comité tutoral, sus ideas siempre fueron de gran apoyo y sus comentarios enmendaron esta tesis. 6 AGRADECIMIENTO A TÍTULO PERSONAL Agradezco al Dr. Matthew E. Smith por su amistad, apoyo en todas mis dudas y en la búsqueda bibliográfica. Agradezco al Departamento de Fitopatología de la Uni- versidad de Florida por permirtirme hacer ahí una estancia. Agradezco a los doctores Mohammad Bahram y Sergei Põlme por su enseñanza y tutoría para los análisis bioinformáticos y estadísticos. Agradezco al Museo de His- toria Natural de la Universidad de Tartu, Estonia por abrirme sus puertas. Agradezco al jurado de esta tesis: Alfonso Valiente Banuet, Margarita Villegas Ríos, Alfonso Delgado Salinas, Mariana Benitez Keinrad y Patricia Vélez Aguilar. Gracias por sus comentarios y correcciones. Mil gracias a mis ayudantes de campo a Rodolfo Ángeles Argáiz, Sergio Vidal, Abel Domínguez Pompa, Jorge Blanco Martínez, Lorenzo Vázquez Selem y Diego Flores López; sin ustedes esto no sería posible. Doy las gracias a Andrés Argüelles Moyao por su apoyo en el procesamiento mo- lecular de las muestras. De igual forma agradezco a Saleh Rahimlou por estar al tanto de mí durante mis numerosos errores bioinformáticos. Agradezco a la Estación de Biología de Chamela y a su personal por su excelente atención y servicio. 7 AGRADECIMIENTO A MIS SERES QUERIDOS Agradezco a mi familia por todo su apoyo en este camino que ha resultado ser muy largo. Innumberables veces los extrañé pero siempre estaban conmigo en cada paso, los amo. A Andrés Argüelles Moyao, compañero de felicidad, aventuras y sinsabores, gra- cias por haber estado a mi lado. A Haydée Hernánez Yáñez por compartir tus vivencias conmigo, por siempre invi- tarme a tus aventuras, por los chistes, por las horas de pláticas y risas. La distan- cia no puede con nosotras. Te quiero muchísimo Jaiba. A Dennis Bermúdez Díaz, César Pastor García Cruz y José Guadalupe Colmena- res Natharen por ser parte de mi familia. Agradezco a mis compañeros de laboratorio por hacer del lab mi segunda casa. Gracias por los seminarios, las discusiones, las salidas de campo, los hongos, los congresos, las bromas, los memes, las jorobas, las pizzas, las fiestas, las manda- rinas evitadas… todo fue importante para mí. A Eduardo Pérez Pazos por su amistad sin límites. Escribo estos agradecimientos después de escuchar que tu vida es tan feliz y plena que deseo que siga así por siempre. Te quiero mucho Lalín. A Abraham Solís Rodríguez y Oscar Zárate Martínez por depositar la confianza de dirigirles su tesis de licenciatura. Les deseo éxito y felicidad en toda su vida. A Margarita Villegas Ríos por ser mi mentora desde la maestría. Muchas gracias por todas sus enseñanzas, gran maestra. 8 A Lorenzo Vázquez Selem por creer en mí y siempre echarme porras, muchas gracias por todo tu apoyo. A Adriana Benítez Villaseñor por todos estos años de amistad. Te admiro mucho porque eres una gran investigadora, nunca dejes que nadie te diga lo contrario. A Rebeca Hernández Gutiérrez por tu apoyo, las pláticas, los helados. Gracias por escucharme. A Laura Alicia Rodríguez Bustos por tu buena vibra veracruzana, desde que dijiste que Xalapa es genial me conquistaste. Gracias por tus palabras y tu apoyo. A Juan Ramos Garza por ayudarme en todo lo que estaba a tu alcance, esta tesis se logró también gracias a ti. A Sandra Castro Santiusté, muchísimas gracias por tus palabras de aliento y apo- yo para lograr teminar esta tesis. A Ana Susana Estrada, mujer fuerte, inteligente y maravillosa, gracias por ser uno de mis ídolos. Te deseo toda la felicidad y que tu hija sea la siguiente generación de micólogas. A Leopoldo Andrade Gómez por siempre llegar con una sonrisa al laboratorio y un cálido abrazo, gracias Polín. A Olivia Ayala por ser una modelo de mujer trabajadora y dedicada con pasión a la micología. Maai mubuk! To Saleh Rahimlou, for his companion and endearing friendship. My days in Esto- nia would be empty without you. 9 DEDICATORIA A mis dos abuelas Engracia Gutiérrez Vázquez (1926 - 2018) y Haydeé Barradas Zárate (1925 - 2019), ustedes son el ejemplo de mujeres inteligentes, trabajadoras incansables y con gran pasión por la vida. Esta tesis es un homenaje a ustedes, las extraño mucho. A mi mamá, papá y hermanas. Ustedes saben del tiempo que me ausenté por en- tregarme a este doctorado, esta tesis también es de ustedes. A mi Chamela querida, el bosque que jamás me ha dejado de enseñar. 10 “Si la lluvia viene del cielo… entonces los hongos son divinos” José de Jesús Preciado de León. Yekwaate metá wixaritari Tateikietari. Hongos y Wixaritari de Tateikie 11 CONTENIDO RESUMEN ............................................................................................................ 14 ABSTRACT ........................................................................................................... 17 INTRODUCCIÓN .................................................................................................. 19 MARCO TEÓRICO ................................................................................................ 28 Los microorganismos del suelo y su papel ecológico ........................................ 28 Los hongos y sus gremios nutricionales ............................................................ 29 Los hongos micorrízicos .................................................................................... 31 La red micorrízica............................................................................................... 32 Propiedades de las redes ecológicas................................................................. 33 Perturbación y sus efectos en los microorganismos .......................................... 36 LITERATURA CITADA .......................................................................................... 39 CAPÍTULO 1. CARYOPHYLLALES ARE THE MAIN HOST OF A UNIQUE SET OF ECTOMYCORRHIZAL FUNGI IN A NEOTROPICAL DRY FOREST .................... 53 Resumen ............................................................................................................ 53 Abstract ............................................................................................................. 53 Key words ......................................................................................................... 53 Introduction ....................................................................................................... 53 Material & methods ........................................................................................... 54 Study site ...................................................................................................... 54 Root sampling ............................................................................................... 54 Plant identificaion and phylogenetic analysis ................................................ 55 Fungal identification and phylogenetic analysis ............................................ 55 Characterization of mycorrhizae .................................................................... 55 Results .............................................................................................................. 55 Ectomycorrhizal host plants .......................................................................... 55 Fungal diversity ............................................................................................. 56 Mycorrhizal morphotypes .............................................................................. 56 12 Fungal-host interactions ................................................................................ 57 Discussion .......................................................................................................... 58 The Caryophyllales are the main ectomycorrhizal hosts in the tripical dry forest ........................................................................................................................... 58 The hyperdiverse tropical dry forest harbors a unique set of ectomycorrhizal fungi .................................................................................................................. 60 Unusual morphotypes in the tropical dry forest ............................................. 61 Specific plant ectomycorrhizal fungal interactions in the tropical dry forest ... 63 Acknowledgments ............................................................................................. 64 Author contributions .......................................................................................... 64 References ........................................................................................................ 64 CAPÍTULO 2. SOIL FUNGAL COMMUNITY WAS PERSISTENT AND RESILIENT TO PATRICIA HURRICANE .................................................................................. 67 Resumen ............................................................................................................ 67 Abstract .............................................................................................................. 68 Keywords ........................................................................................................... 69 Introduction ........................................................................................................ 69 Methodology ...................................................................................................... 70 Study site ........................................................................................................ 70 Sampling ......................................................................................................... 71 Soil chemical analysis ..................................................................................... 71 Molecular biology and bioinformatics .............................................................. 71 Statistical analysis .......................................................................................... 72 Results ............................................................................................................... 72 Soil characteristics .......................................................................................... 72 Soil fungal diversity ......................................................................................... 73 Discussion .......................................................................................................... 78 Conclusions ....................................................................................................... 81 Acknowledgments .............................................................................................. 81 References ......................................................................................................... 82 Supplementary information ................................................................................ 90 CAPÍTULO 3. THE MYCORRHIZAL NETWORKS AND RHIZOSPHERIC FUNGAL COMMUNITIES IN A NEOTROPICAL DRY FOREST ARE RESILIENT .............. 96 13 Resumen ............................................................................................................ 96 Abstract .............................................................................................................. 97 Key words .......................................................................................................... 98 Introduction ........................................................................................................ 98 Materials and methods ..................................................................................... 100 Sampling ....................................................................................................... 100 Bioinformatics ............................................................................................... 101 Network analysis ........................................................................................... 101 Statistical analysis ........................................................................................ 102 Results ............................................................................................................. 103 Characteristics of plots.................................................................................. 103 Mycorrhizal network analysis ........................................................................ 103 Interspecific interactions of rhizospheric fungal community .......................... 105 Hurricane effect on rhizosphere fungal communities .................................... 109 Discussion ........................................................................................................ 110 Conclusions ..................................................................................................... 112 Acknowledgments ............................................................................................ 113 References ....................................................................................................... 113 Additional files .................................................................................................. 123 Characteristics of plots.................................................................................. 123 Hurricane effect on fungal community .......................................................... 134 DISCUSIÓN ........................................................................................................ 144 CONCLUSIONES ................................................................................................ 155 LITERATURA CITADA ........................................................................................ 157 ANEXO 1. Tomentella brunneoincrustata, the first described species of the Pisonieae-associated Neotropical Tomentella clade, and phylogenetic analysis of the genus in Mexico ............................................................................................ 164 Resumen .......................................................................................................... 164 ANEXO 2. Fungal diversity notes 1153–1267: taxonomic and phylogenetic contributions to fungal taxa.................................................................................. 176 Resumen .......................................................................................................... 176 14 RESUMEN Los escenarios del cambio climático predicen que habrá un aumento en frecuencia e intensidad en la incidencia de fenómenos extremos como los huracanes, por lo que estudiar sus efectos ecológicos se vuelve cada vez más relevante. En octubre del 2015, el huracán Patricia categoría 4 Saffir-Simpson, azotó el bosque tropical caducifolio (BTC) de la costa de Jalisco. Este evento catastrófico aportó 17.8 Me- gagramos por hectárea (Mg ha-1) de biomasa al suelo, lo que aumentó el C, N y P del suelo de la Estación de Biología de Chamela. Además, durante el 2015 au- mentaron la temperatura (de 25.1 a 25.6 ºC 1980-2015) y la precipitación media anual (de 765 mm a 800.4 mm 1983-2015). Los hongos y bacterias son los principales descomponedores de la materia orgá- nica y su actividad se ve favorecida por altas temperaturas y humedad. Debido a la materia orgánica, y a las variables ambientales que modificó el huracán en el bosque, se propuso estudiar las interacciones de los hongos ectomicorrízicos con las plantas del BTC, el efecto del huracán Patricia en el ensamble de las comuni- dades de hongos del suelo, de la rizósfera y la red micorrízica en el BTC. La tesis se dividió en tres capítulos y dos anexos: 1) la simbiosis ectomicorrízica del BTC previa al huracán; 2) efecto del huracán Patricia en la diversidad de hon- gos del suelo; 3) diversidad de hongos de la rizósfera y efecto del huracán en la red micorrízica. Adicionalmente, los dos anexos incluyen las descripciones de nuevas especies: el anexo 1 contiene la descripción de Tomentella brunneoincrus- tata, hongo ectomicorrízico asociado a Nyctaginaceae subfamilia Pisonieae; en el anexo 2 se encuentra la descripción de Scytinopogon minisporus, hongo saprótro- fro común de la hojarasca del BTC. En el capítulo 1 se realizaron muestreos de raíces con 91 núcleos de suelo del 2012 al 2014. Este muestreo fue realizado antes del huracán Patricia. Todas las raíces fueron revisadas con un microscopio estereoscópico para identificar morfo- tipos de ectomicorrizas. Las ectomicorrizas fueron separadas, descritas y se extra- jo su DNA. Se amplificó la región ITS rDNA y se secuenció con Sanger. Los resul- 15 tados arrojaron que 20 especies de plantas –no monodominantes, principalmente del orden Caryophyllales– son los hospederos ectomicorrízicos de 19 especies de hongos ectomicorrízicos (ECM). Achatocarpus y Guapira fueron los principales hospederos. Los resultados sentaron las bases para poder estudiar a los hospede- ros ectomicorrízicos y la red micorrízica después del huracán. En el capítulo 2 se estudiaron a las comunidades de hongos del suelo a través del tiempo. Se tomaron muestras de suelo en noviembre del 2014, mayo y octubre del 2016, y abril y septiembre del 2017 de un mismo sitio. Del suelo se extrajo el DNA y se realizó PCR multiplex para la región ITS2 con primers específicos para hon- gos; las librerías se secuenciaron usando Illumina Miseq. Se asignaron nombres taxonómicos a los OTUs comparando las secuencias contra UNITE y usamos FUNGuild para determinar los gremios. La riqueza y abundancia incrementaron después del huracán mientras la dominancia decrecía; esto se invirtió progresiva- mente con el paso del tiempo. Los hongos más comunes fueron basidiomicetos saprótrofos y formaron parte de la comunidad persistente. Además, se observó sucesión ecológica sin remplazo. Para el capítulo 3, a un año después del huracán (octubre del 2016), se montaron nueve cuadrantes en el BTC para estudiar el efecto del huracán en los hongos rizosféricos y específicamente en la red micorrízica. Dentro de las parcelas se marcó a todos los hospederos ectomicorrízicos, se tomaron raíces secundarias de cada individuo y se juntaron por especie; en el 2017 se re-muestrearon las plantas marcadas. Las raíces se revisaron con microscopio estereoscópico apartando úni- camente a las ectomicorrizas; se utilizó la misma metodología molecular del capí- tulo 1. La diversidad fue menor a un año después del huracán e incrementó al si- guiente año. Se encontró que la diversidad rizosférica tiene una correlación positi- va con la luz y negativa con la temperatura del suelo; la identidad del hospedero determina la comunidad rizosférica. Los Ascomycota fueron el grupo más diverso, y los Glomeromycota tuvieron baja diversidad en 2016. La red micorrízica perdió conectividad en 2016 produciendo alta modularidad, y las conexiones se recupera- ron en el 2017 en la mayoría de las parcelas. 16 La alta diversidad fúngica encontrada en el bosque ayuda a la rápida recuperación del ecosistema. Los datos de las comunidades fúngicas mostraron resiliencia, es decir, a pesar del disturbio los hongos tuvieron la capacidad de regresar al estado previo al huracán. La perturbación es un resultado no linear de naturaleza comple- ja, que tiene impacto a diferentes escalas. Los cambios en los ecosistemas ponen a prueba la capacidad inherente de los organismos a soportar nuevas condiciones y la diversidad fúngica colabora a que el BTC sea resiliente a los huracanes. 17 ABSTRACT Climate change scenarios predict that there will be an increase in frequency and intensity in hurricane formation, so studying their ecological effects becomes in- creasingly relevant. In October 2015, hurricane Patricia category 4 Saffir-Simpson made landfall in the tropical dry forest (TDF) of Jalisco’s coast. This catastrophic event contributed 17.8 Mg ha-1 of biomass to the soil, which increased the C, N and P of the soil of the Chamela Biology Station. In addition, during 2015 the tem- perature increased (25.1 to 25.6 ºC 1980-2015) and the average annual rainfall (765 mm to 800.4 mm 1983-2015). Fungi and bacteria are the main decomposers of organic matter and their activity is favored by high humidity and warm temperatures. Due to the organic matter, and environmental variables that modified the hurricane in the forest, it was proposed to study the effect of hurricane Patricia on the assembly of the fungal communities of the soil, the rhizosphere and the mycorrhizal network in the BTC. The thesis was divided into three chapters and two supplementary files: 1) the ec- tomycorrhizal symbiosis of the TDF prior to the hurricane; 2) effect of Hurricane Patricia on the diversity of soil fungi; 3) diversity of rhizosphere fungi and effect of the hurricane in the mycorrhizal network. Additionally, the two annexes include de- scriptions of new species: Supplementary 1 contains the description of Tomentella brunneoincrustata associated Nyctaginaceae subfamily Pisonieae; Supplementary 2 contains the description of Scytinopogon minisporus, saprotroph from TDF. In Chapter 1, root samples were sampled with 91 soil cores from 2012 to 201; Sampling done before Patricia. All roots were checked with stereoscopic micro- scope to identify morphotypes of ectomycorrhizae. The ectomycorrhizae were sep- arated, described and their DNA was extracted. The ITS rDNA region was ampli- fied and sequenced with Sanger. The results showed that 20 species of non- monodominant plants, mainly from order Caryophyllales, are the ectomycorrhizal hosts of 19 species of ectomycorrhizal fungi (ECM). Achatocarpus and Guapira 18 were the main hosts. The results laid the foundations for studying the ectomycor- rhizal hosts and the mycorrhizal network after the hurricane. In Chapter 2, soil fungal communities were studied over time. Soil samples were taken in November 2014, May and October 2016, and April and September 2017 from the same site. DNA was extracted from the soil and multiplex PCR was per- formed for the ITS2 region with fungal specific primers; the libraries were se- quenced using Illumina Miseq. Taxonomic names were assigned to the OTUs by comparing the sequences against UNITE and using FUNGuild to determine the guilds. Diversity increased after the hurricane while its dominance decreased; This was reversed over time. The most common fungi were saprotrophic basidiomy- cetes and were part of the resistant community. In addition, ecological succession was observed without replacement. For Chapter 3, we settled nine permanent plots in October 2016 were all ectomy- corhizal hosts were tagged. From there plants, we took secondary roots and the roots of each individual were pooled by species; In 2017, marked plants roots were re-sampled. We only processed roots with ectomycorrhizae and followed the same molecular methodology of Chapter 1. We found that rhizospheric diversity had a positive correlation with light and negative to soil temperature; also host identity determines the fungal community. Diversity was lesser one year after hurricane and increased two years after. Ascomycota was the most diverse group; Glomer- omycota had low diversity in 2016. Saprotrophs were the most common guild in high disturbance plots. ECM had higher richness in plots with low and high disturb- ance. The mycorrhizal network lost connectivity in 2016 producing high modularity, while connections recovered in 2017. The high fungal diversity found in the forest helps the rapid recovery of the ecosys- tem. Data from the fungal communities showed resilience, i.e., despite the disturb- ance, the fungi had the ability to return to the pre-hurricane state. The disturbance is a non-linear result of a complex nature, which has an impact on different scales. Changes in ecosystems test the inherent ability of organisms to withstand new conditions and fungal diversity helps to make the TDF resilient to hurricanes. 19 INTRODUCCIÓN Los eventos climáticos extremos son cada vez más frecuentes y se asume que tiene una conexión de ellos con el calentamiento global (Diffenbaugh et al., 2017). El aumento en la temperatura global está provocando el derretimiento del permafrost liberando reservorios de carbono y aumentando los gases invernadero (Schuur et al., 2015). El CO2 reacciona con el agua acidificando los océanos (Hoegh-Guldberg et al., 2007), las épocas de secas y lluvias se están viendo mag- nificadas, y en consecuencia los fenómenos del Niño y la Niña se presentan con mayor severidad (Cai et al., 2014; Miralles et al., 2014). Así mismo, los cambios en los ciclos biogeoquímicos están teniendo un efecto en los nutrientes, propiedades físicas y la humedad del suelo (Seidel et al., 2008), y el aumento en la temperatura oceánica está promoviendo la formación e intensidad de huracanes (Knutson et al., 2015). De manera particular, los huracanes también llamados ciclones o tifones, son eventos climáticos extremos formados por grandes cantidades de agua calien- te y su energía proviene de la evaporación de la superficie del océano. El agua condensada forma nubes y lluvia; las nubes se concentran alrededor de un núcleo caliente de baja presión, conocido como el ojo del huracán (Henderson-Sellers et al., 1998). Alrededor del ojo del huracán los vientos circulan como resultado del momento angular que le proporciona la rotación de la Tierra a medida que el aire fluye hacia el eje de rotación. Los huracanes se clasifican del 1 al 5, de acuerdo con la velocidad de sus vientos en la escala Saffir-Simpson, donde 1 va de 110- 153 km/h, hasta la catastrófica categoría 5 con ≥252 km/h (NOAA 2019). Después del impacto del huracán se presentan lluvias intensas durante varios días. La can- tidad de precipitación depende de la humedad del aire, el tamaño y la velocidad del huracán (Regmi et al., 2013). Para el 2075 se predice que la temperatura oceánica aumentará 4.5°C, y habrá 338% más de ciclones tropicales de catego- rías 4 y 5. También habrá un aumento del 17% en la intensidad de las tormentas tropicales y lluvia traída por huracanes (Knutson et al., 2015). La posición geográ- 20 fica de México vuelve vulnerable al país ante estos eventos, por lo que entender el efecto de los huracanes es cada vez más relevante. Los eventos climáticos extremos junto con las actividades antropogénicas son perturbaciones que modelan los sistemas forestales influyendo en su compo- sición, estructura y funcionamiento. Cada perturbación tiene efectos diferentes en los bosques, dependiendo de su severidad, frecuencia y magnitud (Holden & Tre- seder, 2013). Los eventos de mayor magnitud y severidad solían ser menos fre- cuentes antes del cambio climático (Figura 1); sin embargo, se pronostica un au- mento en frecuencia y magnitud de algunos de ellos (Knutson et al., 2015; Diffen- baugh et al., 2017). Los disturbios crean efectos en cascada en las propiedades abióticas y bióticas del sistema. Además, los ecosistemas pueden experimentar más de una perturbación, lo que tiene efectos compuestos que pueden llevar al ecosistema a nuevas composiciones o cambios sin precedentes (Dale et al., 2001). Figura 1. Modelo hipotético de la frecuencia con que suceden los eventos climáti- cos de gran magnitud (eventos extremos) a partir del cambio climático (idea modi- ficada de Selby 1982). Los cambios en los ecosistemas ponen a prueba la capacidad inherente de los organismos para soportar nuevas condiciones. Incluso hay especies que se ven beneficiadas por los cambios, al tener un ambiente más idóneo para produc- 21 ción de biomasa, por ejemplo el aumento de CO2 y de la temperatura beneficiará a algunas especies de plantas (Bellard et al., 2012). Las especies cuya capacidad intrínseca a variar (e.g. en su fisiología, fenología o distribución) y a persistir en la perturbación, confieren resistencia a las funciones del ecosistema (Bellard et al., 2012; Oliver et al., 2015). La respuesta plástica de las especies es importante para su persistencia a corto plazo pero esto también puede tener un costo y ser insufi- ciente para evadir su extinción (Gienapp et al., 2008; Moritz & Agudo, 2013). Los efectos de los eventos extremos que propicia el cambio climático pue- den medirse a diferentes escalas. Estas escalas pueden ser en espacio, tiempo o inherente a sus características biológicas-fisiológicas, y pueden tener efecto en los diferentes niveles de la biodiversidad (desde individuos hasta biomasa). En la es- cala de individuos un evento extremo puede hacer variar su fecundidad, suscepti- bilidad a enfermedades, sobrevivencia, tasa de actividad, crecimiento, etc; a nivel de poblaciones puede variar el reclutamiento, la estructura de edades, la abun- dancia, rango de su distribución, etc; en comunidades puede tener efecto en nue- vas interacciones interespecíficas, desequilibrio, desincronización, etc; en ecosis- temas puede variar el flujo de energía, la cantidad de biomasa, la producción de servicios ecosistémicos, etc. (Bellard et al., 2012). A escala de comunidad, si la composición biológica tiende a la redundancia ecológica –diferentes especies desempeñan el mismo papel ecológico– al experi- mentar perturbación, la comunidad tiene mayor propensión a mostrar resiliencia (Standish et al., 2014). La alta biodiversidad incrementa la resiliencia del ecosis- tema (Mori, 2016). Las especies que tienen alta tolerancia a la alteración de las condiciones (i.e. temperatura, pH, nutrientes, entre otros) son las especies que conforman las comunidades de ambientes con alto disturbio. Existe un debate sobre el concepto de resiliencia. Holling (1973) propuso el término para referirse a la habilidad del sistema para resistir el cambio y hacer frente al disturbio o para regresar después de la perturbación a un estado estable similar al incial. Con el tiempo, el concepto derivó en el proceso de recuperarse después del disturbio (Pimm 1994; Grimm & Wissel, 1997). Actualmente existen 22 términos que han sido derivados del estudio de la perturbación y que se han con- fundido con resiliencia. Por ejemplo, ‘resistencia’ se puede definir como el grado en el que una variable cambia después de un disturbio (Pimm 1994), como la in- mutación de la comunidad al disturbio (de Vries & Shade, 2013), o como la capa- cidad de regresar rápidamente al estado previo al disturbio (Oliver et al., 2015). Por otro lado, también el concepto ‘recuperación’, que invloucra los procesos en- dógenos que llevan al sistema al equilibrio, suele tomarse como resiliencia. Mien- tras que la ‘elasticidad’ es la tasa en la que un sistema se recupera del disturbio (Hodgson et al., 2015). En esta tesis se empleará el término resiliencia como la capacidad regresar a un estado estable similar al incial, de Holling (1973). Las comunidades microbianas, compuesta por archeas, bacterias, hongos, protozoarios, entre otros, habitan en prácticamente cualquier ecosistema o sustra- to del planeta. En el suelo las comunidades microbianas son altamente diversas y generalmente presentan redundancia ecológica. La alta diversidad del edafón (i.e. organismos que habitan en el suelo) puede llevar al ecosistema a la estabilidad. La estabilidad se manifiesta en la resistencia y de la resiliencia, por lo que la esta- bilidad de la comunidad microbiana serán resistentes y resilientes dependiendo de su diversidad y su fisiología (Cantrell et al., 2014). El edafón se compone de bacte- rias, artrópodos, nematodos, anélidos, mamíferos, reptiles, protozoos, hongos, etc. Particularmente los hongos, uno de los grupos más biodiversos del planeta (Bla- ckwell, 2011), tienen papeles ecosistémicos importantes como la descomposición de la materia orgánica, participando activamente en los ciclos biogeoquímicos del C, N y P (Bueé et al., 2009; Jaramillo et al., 2018). En el grupo de los hongos se encuentran especies con estilos de vida muy variados, como los saprótrofos, que son degradadores de la materia orgánica; los micorrízicos, que se establecen de manera mutualista con las raíces de las plantas; los parásitos y patógenos de animales, plantas y hongos; los endófitos, que pueden llegar a ser mutualistas o parásitos; los liquénicos, que tienen asociaciones mutualistas con cianobacterias y algas, etc. 23 Particularmente, los hongos micorrízicos tienen una relación mutualista con las plantas al haber intercambio de nutrimentos entre ambos (Trappe 2005). De acuerdo con su morfología y fitobionte participante se diferencían siete tipos de micorrizas (Smith & Read, 2008), dentro de las cuales las ectomicorrizas (ECM) son constituidas por hongos de los phyla Ascomycota y Basidiomycota, mientras que las micorrizas arbusculares (AM) están conformadas por hongos del subphy- lum Glomeromycotina. Éstas dos son las interacciones micorrízicas más comunes en los ecosistemas terrestres (Smith & Read, 2008). Los hongos ectomicorrízicos, tienen la capacidad de oxidar materia orgánica del suelo para obtener C, aunque en menor proporción que los hongos saprótrofos (Shah et al., 2016). Estos hongos reciben C de los fotosintatos de su planta hospedera y le transfieren hasta el 80% del N y el 70% del P adquirido del suelo. Los hongos micorrízicos arbusculares son simbiontes obligados pues dependen de su hospedero vegetal para la obten- ción del C; a cambio, estos hongos extraen del suelo N y P, y los transfieren a la planta (trasfieren a la planta hasta el 20% del N y el 90% del P adquirido del suelo; van der Heijden et al., 2015). Existen algunos trabajos donde refieren el efecto de las perturbaciones en las comunidades de hongos, como en incendios o deforestaciones (e.g. Holden et al., 2013) y en particular de los hongos micorrízicos (e.g. Rincón et al., 2014; Glassman et al., 2015). No obstante, poco se conoce sobre el efecto de los hura- canes en la estructura de las comunidades de hongos. Cantrell y colaboradores (2014) realizaron simulaciones de un huracán en parcelas experimentales en Puerto Rico, tratando de conocer el efecto en las comunidades de hongos y bacte- rias. Ellos encontraron que la apertura del dosel y la variación interanual de la pre- cipitación tienen un fuerte efecto en la actividad microbiana. Por otro lado, Vargas y colaboradores (2010) estudiaron el efecto del huracán Wilma en las comunida- des de hongos micorrízico arbusculares de Yucatán, donde encontraron que existe mayor porcentaje de colonización en las raíces de las plantas después del hura- cán; además de haber un enriquecimiento de N en el suelo, al aumentar la hume- dad y temperatura del suelo, lo que ayudó a la rápida descomposición. Ellos hipo- 24 tetizan que las plantas adoptan la estrategia de invertir más C en sus micorrizas, en vez de formar nuevas raíces, por lo que encuentran mayor colonización. Los hongos micorrízicos habitan en el suelo, donde llegan a completar su ciclo de vida formando propágulos sexuales o asexuales. Estos hongos necesa- riamente deben estar asociados a sus hospederos, ya que algunos pueden obte- ner C de la descomposición de la materia orgánica, pero necesitan el aporte de sus simbiontes vegetales (Shah et al., 2016). En algunos bosques tropicales se ha visto que los hongos ectomicorrízicos tienen alta especificidad hacia sus hospede- ros (Tedersoo et al., 2010). Por ejemplo, en el bosque tropical caducifolio, que es- tá compuesto principalmente de leguminosas, la mayoría de sus plantas no forma esta interacción (Alvarez-Manjarrez, 2014; Waring et al., 2016), por lo que las plantas ectomicorrízicas representan islas por colonizar. Es decir, a pesar de que en el suelo se encuentren todos los propágulos de estos hongos, sólo algunos de ellos llegarán a establecerse en las raíces. Cuando los hongos micorrízicos logran establecerse con su hospedero, co- lonizan todas las raíces posibles, ya sean de la misma planta, de la misma especie o de otras especies adyacentes. Por lo tanto, diferentes especies de plantas pue- den estar micorrizadas por el mismo hongo, interconectadas entre sí por micelio en común. A esta conexión planta-planta vía hongo se le conoce como red mico- rrízica. A través de la red micorrízica, se transfieren compuestos de planta a planta (e.g. agua, carbono, nitrógeno, fósforo, etc.) y se ayuda al establecimiento, creci- miento y sobrevivencia de plántulas (Simard et al., 2012). Las propiedades de las redes micorrízicas han sido analizadas desde la teoría de redes (Beiler et al., 2010; Montesinos-Navarro et al., 2012; Bahram et al., 2014; Põlme et al., 2018). La teoría de redes nos ayuda a comprender las interac- ciones ecológicas de las especies que componen una comunidad. Las propieda- des estadísticas y topológicas de las redes nos permiten entender la robustez (re- siliencia) de las comunidades ante la afectación de las comunidades por eventos climáticos extremos. Estas modificaciones de las comunidades son perturbaciones a la red de interacciones. Entiéndase perturbación de las redes como la pérdida de 25 nodos o enlaces entre nodos. No obstante, el estudio sobre la perturbación en las redes micorrízicas y su cambio en el tiempo todavía no ha sido explorada. En la presente tesis se realizó un estudio sobre el efecto del huracán Patri- cia en el bosque tropical caducifolio de Chamela, Jalisco. El huracán Patricia gol- peó las costas de Jalisco el 24 de octubre del 2015, alcanzó vientos máximos de 345 km/h y antes de tocar tierra redujo su velocidad a 240 km/h (categoría 4); im- puso un récord en el Pacífico, al alcanzar vientos de 190 km/h en 24 h. La fase cálida propiciada por el fenómeno de El Niño asociada a la Oscilación del Sur pro- vocó la formación de tan poderoso huracán. Este huracán causó daños en la ve- getación, provocando un aporte de troncos, ramas y árboles desenraizados; espe- cialmente de fracciones finas de lignina (Martínez-Yrízar et al., 2018). El dosel da- ñado aportó 17.8 Mg ha-1 de biomasa al suelo (Parker et al., 2018) engrosando el mantillo con hojas y ramas, lo cual generó una ganancia de nutrientes en el suelo (Gavito et al., 2018). La principal vegetación de la costa de Jalisco es el bosque tropical caducifo- lio (BTC), donde la familia Fabaceae tiene la mayor diversidad y densidad. Esta tesis hipotetizó que ya que algunas especies de leguminosas han sido reportadas como hospederos ectomicorrízicos en otras partes del mundo (e.g. Henkel et al., 2002) los hospederos ectomicorrízicos del BTC pertenecerán a la familia de las leguminosas. El BTC se caracteriza por un periodo de lluvias muy estacional; durante la época de secas, los árboles caducifolios reabsorben nutrientes antes de tirar sus hojas dando lugar a la acumulación de mantillo (Rentería et al., 2005). Mientras la sequía persiste, la actividad microbiana se ve reducida por la falta de agua, y la transformación y descomposición del N se detiene. La degradación del mantillo está en función de la disponibilidad y variación en la precipitación anual (Anaya et al., 2012). El huracán Patricia incrementó la cantidad de nutrientes en la materia orgánica, la temperatura (de 25.1 a 25.6 ºC 1980-2015) y la precipitación media anual (de 765 mm a 800.4 mm 1983-2015; Maass et al., 2018), la descomposición del mantillo (Gavito et al., 2018) pero se desconoce el efecto en las comunidades 26 fúngicas. Por lo que se propusieron las hipótesis: 1) el aumento de nutrientes en el suelo, principalmente C, N y P, incrementará la diversidad de hongos del suelo, principalmente saprótrofos del phylum Basidiomycota; 2) el aumento en la canti- dad total de nitrógeno y fósforo provocará la reducción de la interacción mico- rrízica, por lo que se reducirá la diversidad de hongos micorrízicos en el suelo. El suelo funciona como sustrato de muchos hongos, mas no explica toda la biología de los hongos micorrízicos, por lo que también se estudió la rizósfera. La rizósfera es la estrecha región del suelo en contacto con las raíces de las plantas. La diversidad fúngica de la rizósfera está fuertemente influenciada por su hospe- dero y el suelo que lo rodea (Grayston et al., 1998). Por lo que se esperó que: 1) el aumento de nutrientes en el suelo propiciará un aumento en la diversidad fúngica de la rizósfera; 2) los hongos rizosféricos compiten por la interacción con la raíz, por lo que un sistema perturbado generará un cambio en las interacciones interes- pecíficas; 3) las redes micorrízicas suelen tener alta modularidad –propiedad de la red que experimentalmente reduce el impacto de la perturbación– por lo que los sitios donde la perturbación del huracán Patricia fue mayor, se espera que la mo- dularidad sea más alta. La presente tesis estudió el efecto del huracán Patricia en las comunidades de hongos del suelo y, específicamente en las comunidades rizosféricas. Para en- tender cómo se modifica la comunidad de hongos del suelo tendiendo a volver a la estabilidad previa al huracán, además de modelar la red micorrízica; y con ello se contribuyó al conocimiento de la ecología de los hongos durante la perturbación. El documento está dividido en tres capítulos y dos anexos: Capítulo 1. Tuvo como objetivo deteminar a los simbiontes ectomicorrízicos, tanto plantas como hongos, del bosque tropical caducifolio antes del huracán Pa- tricia. Capítulo 2. Sus objetivos fueron conocer la comunidad de hongos saprótro- fos, micorrízicos, parásitos, patógenos y otros estilos de vida del suelo; y comparar el ensamble de las comunidades de hongos del suelo antes y después del hura- 27 cán para conocer el efecto de la perturbación de acuerdo con datos ambientales y cantidad de nutrientes en el suelo. Capítulo 3. Los objetivos fueron identificar las especies de hongos ectomi- corrízicos que se establecen en las rizósferas a pesar de la perturbación causada por el huracán; determinar cuáles son las variables ambientales o nutrimentales que benefician a los hongos ectomicorrízicos para que se establezcan en las raí- ces después de un huracán; comparar las interacciones interespecíficas de los diferentes gremios fúngicos en los dos años después del huracán; y evaluar el efecto de la perturbación del huracán Patricia en la estructura de la red mico- rrízica. Anexo 1. Descripción la nueva especie Tomentella brunneoincrustata, hon- go ectomicorrízico que se establece exclusivamente con raíces de Nyctaginaceae en el BTC de Chamela, Jalisco. Adicionalmente se realizó un análisis filogenético para todos los representantes del género Tomentella en México. Anexo 2. Descripción la nueva especie Scytinopogon minisporus, hongo saprótrofo común en los residuos del mantillo en el BTC de Chamela, Jalisco. 28 MARCO TEÓRICO Los microorganismos del suelo y su papel ecológico El suelo, constructo de la alteración física y química de las rocas por la ac- tividad biológica, es la corteza terrestre más superficial. El suelo contiene materia orgánica, minerales, gases, líquidos y organismos; es un ecosistema hiperdiverso donde se establecen las plantas, habitan bacterias, hongos, artrópodos, mamífe- ros, reptiles, etc (Voroney & Heck, 2015). En una hectárea de suelo en buen esta- do de conservación hay 1000 kg de lombrices, artrópodos, 150 kg de protozoarios, algas, 1700 kg de bacterias y 2700 kg de hongos (Pimentel et al., 1980; Lee & Foster, 1991; Barget & ven der Putten, 2014). Esta cantidad de biomasa corres- ponde en gran medida a la materia orgánica encontrada en el suelo. La materia orgánica del suelo proviene de las plantas, animales y microor- ganismos que en él habitan. Esta materia orgánica es la que determina los nutrien- tes que tendrá el suelo, puesto que de la descomposición de la materia orgánica se reciclarán todos los elementos que la componían. La materia orgánica puede contener polisacáridos (celulosa, hemicelulosa, almidón, pectina), proteínas, quiti- na, lignina –compuesto más recalcitrante de la materia orgánica–, entre otras; de- pendiendo de su composición química podrán ser fácilmente adquiribles por los organismos del suelo o necesitarán de procesos enzimáticos u oxidativos para descomponerse (Chenu et al., 2015). Al finalizar la descomposición de la materia orgánica se pueden obtener diferentes elementos que son esenciales para todos los organismos: carbono, ni- trógeno, fósforo, potasio, calcio, azúfre, entre otros. Los nutrientes disponibles del suelo –iones asimilables por las plantas y microorganismos, por ejemplo el ion amonio (NH4)– se agotan rápidamente, por lo que el proceso de descomposición es constante. Los encargados de llevar a cabo la descomposición son los microor- ganismos del suelo, que incluye a los microartrópodos, nematodos, hongos y bac- terias. Los microartrópodos ayudan a romper mecánicamente el mantillo que apor- 29 tan las plantas en el suelo. Mientras que las bacterias y los hongos rompen las moléculas que conforman a la materia orgánica dejando nutrientes disponibles (Berg, 1999). Los hongos y las bacterias del suelo tienen un papel fundamental en la descomposición de la materia orgánica y el reciclaje de nutrientes por lo que regu- lan y equilibran a los ciclos biogeoquímicos (Bahram et al., 2018). Las bacterias son los organismos más abundantes del planeta y se encuetran habitando todo tipo de ambientes, incluso en condiciones extremas de temperatura o presión (Bargett & van der Putten, 2014). Tanto bacterias como hongos son diversos y sus roles ecológicos no se restringen a la descomposición de la materia orgánica. Por ejemplo, las bacterias del género Rhizobium pueden asociarse a las raíces de al- gunas leguminosas y fijar nitrógeno atomsférico gracias a sus enzimas nitrogena- sas (Zahran 2001); los hongos también pueden asociarse a las raíces del 96 % de las plantas y proveerlas de nitrógeno, fósforo y agua (van der Heidjen et al., 2015). Los hongos y sus gremios nutricionales El reino de los hongos se estima que es uno de los grupos más diversos del planeta, con una estimación aproximada de 1.5 - 5.1 millones de especies (Blackwell, 2011). Se han secuenciado 44,563 OTUs de hongos que habitan en los suelos de alrededor del mundo (Tedersoo et al., 2014b). Todos estos hongos pueden ser categorizados en los diferentes gremios, grupos de especies que usan estrategias ecológicas similares para explotar el mismo recurso (Peay et al., 2016). Estos gremios son los hongos saprótrofos, micorrízicos, liquénicos, endófi- tos, patógenos y parásitos de animales o plantas. Es importante mencionar que aunque los podemos clasificar de este forma, existen muchas especies que no pertenecen toda su vida a un gremio exclusivamente (e.g. hongos endófitos que pueden ser saprótrofos; hongos ectomicorrízicos que pueden ser saprótrofos; hongos saprótrofos que pueden ser parásitos, etc.; Thomas et al., 2016; Shah et al., 2016; Lanver et al., 2018). Los hongos saprótrofos obtienen su carbono de la descomposición de la materia orgánica inherte. Estos hongos pueden tener la capacidad enzimática para 30 descomponer la pared celular de las plantas (hongos de pudrición café y blanda) y lignina (hongos de pudrición blanca). Principalmente los hongos del pylum Basi- diomycota, y alguas especies de Ascomycota, contiene especies con capacidad de desarrollar pudrición blanca (Voriskova & Baldrian, 2013). Dependiendo de la composición de la materia orgánica es la sucesión de especies fúngicas; las espe- cies pioneras de hongos son los que tienen preferencia por moléculas sencillas de obtener, como glucosa y celulosa, hasta que llegan las especies con la capacidad de romper moléculas recalcitrantes, como la lignina (Talbot et al., 2015). Los hongos micorrízicos, endófitos y liquénicos mantienen una relación mutualista con sus simbiontes. Los hongos micorrízicos obtienen el carbono de sus simbiontes, mientras que pueden proporcionar nutrientes, metabolitos, protec- ción contra patógenos, humedad, etc. Los hongos endófitos tienen una relación estrecha con sus plantas hospederas, pues prácticamente pueden habitar toda su vida dentro de los tejidos vegetales. Estos hongos se transmiten de manera verti- cal a través de las semillas de las plantas, o de manera horizontal y al penetrar al hospedero suelen ser asintomáticos (Busby et al., 2016). Los hongos endófitos también tienen un papel importante en la descomposición de la materia orgánica, puesto que son los primeros hongos presentes en la materia en descomposición. La mayoría de los hongos endófitos son Ascomycota (Promputtha et al., 2007; Song et al., 2017). De igual forma, el 90% de los líquenes son asociaciones de ascomicetos (y 10% basidiomicetos) con cianobacterias o algas verdes. El hongo necesita de la interacción con los fotobiontes para poder formar un talo. Los líque- nes representan el 8% de la biomasa terrestre (Asplund & Wardle, 2017). En el siguiente apartado se hablará con más detalle de los hongos micorrízicos. Los hongos patógenos y parásitos de animales y de plantas son hongos antagónicos de sus hospderos, pues se alimentan de sus tejidos causándoles ne- crosis y otras enfermedades. Estos hongos fitopatógenos suelen ser agentes in- fecciosos que colonizan tallos, hojas, raíces, flores y frutos. Su infección se puede dar por los estomas de las hojas, flores o por daño directo al tejido infectado (Ga- rrett et al., 2015). Mientras que los hongos que atacan a los animales también 31 pueden atacar tejidos específicos, e incluso algunos de ellos son comensales, los cuales pueden ser patógenos oportunistas (Gauthier et al., 2015). Los hongos micorrízicos Los hongos micorrízicos son todos aquellos que tienen simbiosis con las raíces de las plantas, generalmente mutualistas. Cuando la asociación es mutual- sita, las plantas transfieren fotosintatos –principalmente glucosa– a sus simbiontes fúngicos, mientras que los hongos por medio de procesos enzimáticos u oxidativos (Shah et al., 2016), obtienen nitrógeno (NH4, NO3) y fósforo (PO4) del suelo y lo translocan a la raíz (van der Heidjen et al., 2015). Smith & Read (2008) propusieron clasificar a las micorrizas en siete tipos de acuerdo a su morfología y familia de planta que coloniza: ectomicorriza, endo- micorriza –también llamada micorriza arbuscular–, ectendomicorriza, micorriza ericoide, monotropoide, arbutoide y orquideoide. Las ectomicorrizas (ECM) y las micorrizas arbusculares (AM) son los dos tipos más estudiados por su importancia ecológica. Las ECM se forman con 2% de las plantas terrestres (Brundrett & Te- dersoo, 2018) principalmente coníferas aunque también se pueden encontrar en angiospermas (Corrales et al., 2018); fuera de los trópicos se estima que el 80% de los árboles forman ECM (Steidinger et al., 2019). Los phyla de hongos que for- man esta asociación son Ascomycota, Basidiomycota y Mucoromycota (Brundrett & Tedersoo, 2018). Mientras que las AM se con el 80% de las plantas terrestres, siendo más dominantes en los ecosistemas tropicales (van der Heijden et al., 2015; Steidinger et al., 2019). Esta asociación es formada por el phylum Muco- romycota subphylum Glomeromycotina (Spatafora et al., 2016). Los bosques tropicales cada vez han sido más explorados por los micólo- gos, lo cual ha generado el descubrimiento de plantas que forman la asociación ectomicorrízica, e.g., los bosques de leguminosas en Guyana (Smith et al., 2011). Adicionalmente se han descrito nuevas especies de hongos ectomicorrízicos (e.g. Ramírez-López et al., 2015; Sánchez-García et al., 2016; Sukarno et al., 2019) Estas exploraciones han ayudado a entender que la diversidad de hongos ectomi- corrízicos en los tropicos están fuertemente influenciados por la identidad y abun- 32 dancia de su hospedero puesto que son altamente específicos (Tedersoo et al., 2010; Alvarez-Manjarrez et al., 2018; Corrales et al., 2018). Los hospederos tropi- cales conocidos hasta el momento pertenecen a las familias Achatocarpaceae, Dipterocarpaceae, Fabaceae, Fagaceae, Junglandaceae, Myrtaceae, Nyctagina- ceae, Polygonaceae, Phyllantaceae, Pinaceae, Salicaceae, Sarcolanaceae (Alva- rez-Manjarrez 2014; Corrales et al., 2018) El patrón mundial indica que la diversi- dad de hongos ectomicorrízicos es menor en los trópicos, en comparación con las zonas templadas y boreales (Bahram et al., 2013), probablemente por la abundan- cia de sus hospederos (Steidinger et al., 2019). En los bosques donde la abun- dancia de los hospederos ectomicorrízicos es baja, los hongos ectomicorrízicos se encuentran restringidos en área; tal es el caso de los bosques tropicales caducifo- lios. Algunas plantas pueden presentar varios tipos de micorriza. Aunque poco se sabe de la importancia de la dualidad de las interacciones micorrízicas, es pro- bable que le confiera mayor beneficio (Teste et al., 2019). Por ejemplo, las conífe- ras presentan micorrizas arbusculares y ectendomicorrizas cuando están en esta- díos muy tempranos de desarrollo; conforme el árbol envejece, va asociándose mayoritariamente con hongos ectomicorrízicos. Que las plantas puedan tener en sus raíces a diferentes especies de hongos les confiere versatilidad en la adquisi- ción de nutrientes y agua, pero además pueden establecer conexión con otras plantas –de la misma o de diferente especie– a través de los hongos (Simard 2018). La red micorrízica Las conexiones que existen entre las raíces de las plantas a través de los hongos se conocen como redes micorrízicas. Estas conexiones les permiten com- partirse nutrientes y señales químicas entre plantas de la misma preferentemente a su descendencia, o de diferente especie (Simard et al., 2012; 2015; Simard 2018). Las redes micorrízicas se pueden dividir dependiendo del grupo de hon- gos que las forman: la red micorrízica arbuscular y la red ectomicorrízica. Esta di- 33 ferencia se basa en las diferentes estrategias que tiene cada grupo de micorriza, pero además por su ecología. Las redes micorrízicas arbusculares son formadas por un bajo número de especies fúngicas que pueden asociarse a un gran número de plantas (Montesinos-Navarro et al., 2012). Mientras que las redes ectomico- rrízicas, tienen alta especificidad hacia la identidad del hospedero, por lo cual pue- den llegar a ser más restringidas e.g. en los bosques tropicales (Bahram et al., 2014). En estas redes se ha comprobado el paso de carbohidratos entre plantas, lo cual se puede ver como un proceso de facilitación, lo que a su vez ayuda al es- tablecimiento del bosque. De igual forma, se ha descubierto que en los bosques caducifolios de Francia, los encinos no tienen suficientes reservas de carbono pa- ra rebrotar durante la primavera. Existe flujo de carbono de la red ectomicorrízica hacia sus hospederos, lo cual les ayuda a formar nuevo material fotosíntetico y volver a pagar el carbono a los hongos (Bernard et al., 2012). Adicionalmente, a través de la red micorrízico arbuscular, las plantas pueden alertar a las plantas aledañas de formar metabolitos secundarios así evitando el ataque por herbívoros (Babikova et al., 2013). Todas estas maravillas que pasan debajo de nuestros pies, han sido obje- to de estudio en modelos matemáticos, como lo son los modelos de redes (e.g. Simard 2018). Propiedades de las redes ecológicas Las redes ecológicas son modelos de interacciones de dos o más orga- nismos. Nos ayudan a representar, enumerar o catalogar las interacciones de una comunidad (Jordano et al., 2009). En estos modelos existen nodos que represen- tan a las especies interactuantes, y enlaces que son las interacciones que hay en- tre ambos. Estas interacciones pueden ser mutualistas o antagonistas. Un grupo de nodos conectados a cierto nodo constituye el vecindario, y el número de tales conexiones es el grado. El grado nos ayuda a saber la generalización- especialización de cada especie (Jordano et al., 2009). Las redes ecológicas se 34 han ayudado de la teoría de grafos para generar modelos de redes. Dichos mode- los pueden ayudar a entender estructura y propiedades de las redes. Cuando las redes se construyen con nodos de la misma naturaleza se le llama red unipartita (e.g., interacción entre ardillas, gatos, cacomixtles, perros de mi colonia; aunque son elementos heterogéneos todos son animales). Cuando las redes se contruyen con dos clases distintas de elementos, entonces son redes bipartitas (Figura 2), tal es el caso de la interacción de las plantas con los hongos micorrízicos (Mariani et al., 2019). Existen diferentes propiedades que se miden de las redes. A continuación se dará una breve explicación de cada una de las métricas de las redes. La conectividad es el número de interacciones con respecto al total posi- ble; generalmente es baja a medida que la riqueza de especies de una comunidad aumenta. Topología se refiere a distribución de los enlaces entre especies, es de- cir la distribución del grado o de conectividad. Saber la topología nos ayuda a en- tender cómo se entrelazan los nuevos nodos y qué tan sensible es la red cuando los nodos se pierden. En general, las redes reales tienen gran número de nodos con pocos enlaces, y nodos superenlazados –también llamados hubs– (Jordano et al., 2009). Los nodos que tienen pocas conexiones se les puede considerar espe- cialistas, mientras que los nodos más conectados serían los generalistas. Existen algunas ecuaciones para poder medir el grado de especialización de cada especie (d’) o de una red bipartita (H2’ ; Blüthgen et al., 2006). Al comparar las topologías de las redes ecológicas se pueden encontrar similitudes de cómo se distribuyen las conexiones. Uno de esos patrones es el anidamiento o encajamiento (nestedness). El anidamiento es la conexión de nodos i y j donde el grado de i es más pequeño que el grado de j, siempre y cuando ten- gan este conjunto de nodos se encuentren en el mismo vecindario (Figura 2; Ma- riani et al., 2019). Dicho de otro modo, las especies especialistas interactúan ex- clusivamente con las generalistas, y éstas a su vez interactúan entre ellas (Jor- dano et al., 2009). 35 Figura 2. Ejemplos de redes unipartitas y bipartitas con y sin anidamiento. Tomada de Mariani et al., 2019. Por otro lado, hay especies dentro de la red que interactúan con mayor in- tensidad y frecuencia entre sí, formando módulos o compartimentos, a esta carac- terística se le llama modularidad (Jordano et al., 2009). Estos subgrupos con ma- yor especificidad pueden funcionar como subredes (Lewinsohn et al., 2006). Las redes mutualistas y redes tróficas, pueden ser modulares (Raffaeli & Hall, 1992; Dicks et al., 2002); las redes con más de 50 polinizadores suelen ser significativamente modulares, y hay correlación entre el anidamiento y la modula- ridad. Las redes con menor número grado tienden a ser altamente modulares, mientras que al haber más conexiones tiende al anidamiento. Existen redes que pueden ser tanto anidadas como modulares, ya que se interconectan los módulos generando anidamiento (Fortuna et al., 2010). Las redes ectomicorrízicas presen- tan alta modularidad y no-anidamiento (Bahram et al., 2014). Mientras que las re- des micorrízicas arbusculares suelen ser anidadas (Montesinos-Navarro et al., 2012). Existen el debate acerca de que tanto anidamiento como la modularidad indican que la red al ser perturbada puede ser robusta y resiliente. El anidamiento, al estar compuesto por especies generalistas con alto grado, la pérdida de alguno de los nodos puede no colapsar la red; sin embargo, la consecuencia en la red depende del nodo que se pierda, pues al eliminarse un nodo con pocas conexio- 36 nes tiene menor impacto (Burgos et al., 2007). En contraparte, se sugiere que el anidamiento puede ser un indicador de comunidades menos estables en compa- ración con redes sin estructura; y al ser redes mutualistas puede demeritar su per- sistencia (Suweis et al., 2013). Mientras que la modularidad de la red también se ha observado que puede tener un efecto en el módulo afectado pero impide la afección a todos los demás miembros de la red (Gilarranz et al., 2017). Perturbación y sus efectos en los microorganismos Los ecosistemas se encuentran en constante perturbación, principalmente por actividades antropogénicas y por los eventos climáticos, cada vez más fre- cuentes. La perturbación crea heterogeneidad en el paisaje, así como cambio en las poblaciones de especies; lo cual a su vez tiene repercusión en el ensamble de las comunidades. Las especies que tengan mayor tolerancia al estrés o al ser mo- dificado su ambiente se vean favorecidas son las que formarán parte de la comu- nidad después de la perturbación (Oliver et al., 2015). La habilidad de una comunidad de sobrellevar el estrés ambiental o a la perturbación y tender al estado previo, se le conoce como estabilidad (Harrison, 1979). Orians (1975) menciona que la estabilidad lo conforman siete propiedades: constancia, persistencia, inercia, elasticidad, amplitud, estabilidad cíclica y trayec- toria de la estabilidad. A la constancia también se le puede llamar resistencia, pues es la falta de cambios de algún parámetro del sistema, número de especies, composición taxonómica, etc. después de la perturbación. La persistencia es la sobrevivencia en el tiempo de alguno de los elementos del sistema. La inercia es la habilidad del sistema a resistir perturbaciones, lo cual corresponde a la defini- ción de resiliencia de Holling (1973). La elasticidad es la velocidad a la que el sis- tema regresa al estado previo a la perturbación. La amplitud es el área en donde el sistema permanece estable. Actualmente la estabilidad se describe comúnmen- te por su resiliencia y resistencia; para esta tesis se utilizó el término resiliencia como la habilidad del sistema de absorber la perturbación mientras se mantienen la función y estructura (Meyer et al., 2016). 37 Los hongos y las bacterias presentan redundancia ecológica por ser prin- cipalmente descomponedores (Purahong et al., 2016). Sin embargo, tienen dife- rencias biológicas importantes que generan distintas reacciones de respuesta a los cambios ambientales. Una de las diferencias radica en el cociente C:N de am- bos organismos. El cociente C:N correlaciona con la tasa intrínseca de crecimiento de las comunidades microbianas. Los hongos tienen en promedio un cociente C:N de ~5-15 y las bacterias ~3-6. Por lo tanto, los hongos pueden crecer en sitios donde hay menor cantidad de N en comparación con el C, y en los sustratos ricos en N crecen rápidamente bacterias (Steren y Elser, 2002). Además, la tasa de crecimiento de las bacterias es exponencial en horas, mientras que los hongos crecen con menor velocidad. Los hongos, en general, pueden considerarse como una comunidad con estrategia K, en comparación con las bacterias (de Vries y Shade, 2013). Hemos aceptado ampliamanete que la riqueza y composición de las co- munindades vegetales afectan procesos ecosistémicos y su influencia es incorpo- rada en modelos de escalas globales. Los microorganismos, que desempeñan papeles ecológicos igualmente importantes, fueron la caja negra durante mucho tiempo (Allison & Martiny, 2008). El desarrollo de la biología molecular nos armó de herramientas para poder entender mejor la ecología de las comunidades mi- crobianas. Recientemente, la actividad de las comunidades microbianas, e.g. la descomposición ha sido modelada para escalas globales (Steidinger et al., 2019), y cada vez se sabe más de los efectos de la composición de las comunidades mi- crobianas. Al igual que en los macroorganismos, las comunidades microbianas pueden ser resistentes y resilientes después de la perturbación (Allison & Martiny, 2008). El impacto de la perturbación en las comunidades microbianas depende de la naturaleza del disturbio. Por ejemplo, la biomasa de los microorganismos se reduce después de un incendio hasta un 48.7%, después de la cosecha hasta 19.1% y de tormentas hasta 41.7%. Las diferencias en los efectos de las perturba- ciones abióticas y bióticas sobre las comunidades microbianas, varían dependien- 38 do de la disrupción del suelo y la remoción de la materia orgánica (Holden & Tre- seder, 2013). Generalmente la perturbación provoca sucesión de especies, es decir cambios locales en la composición de especies (Ellner & Fussmann, 2003). Ha habido gran seguimiento sobre las comunidades de hongos ectomicorrízicos des- pués de incendios (e.g. Stendell et al., 1999; Dahlberg, 2002; Rincón et al., 2014), donde encuentran que la riqueza decrece dependiendo del tiempo en que duró el incendio. Adicionalmente, hay hongos pirófilos que son propiciados por los incen- dios y colonizan las plántulas (Rincón et al., 2014). Así como este estudio, se han encontrado más sobre otros eventos climatológicos y actividades antropogénicas (e.g. Kalucka & Jagodzinski, 2017; Krug Vieira et al., 2018; Smith et al., 2018; Wang et al., 2019). Aunque cada vez toma más relevancia el estudio de las comu- nidades microbianas después de la perturbación, todavía falta mucho por saber. 39 LITERATURA CITADA Allison S, Martiny J. 2008. Resistance, resilience, and redundancy in microbial communi- ties. Proceedings of the National Academy of Sciences 105: 11512–11519. Alvarez-Manjarrez J. 2014. Diversidad de ectomicorrizas en el bosque tropical de Chamela, Jalisco. Tesis de Maestría. Instituto de Biología, UNAM, México. Anaya CA, Jaramillo VJ, Martínez-Yrízar A, García-Oliva F. 2012. Large rainfall pulses control litter decomposition in a tropical dry forest: evidence from an 8- year study. Ecosystems 15: 652-663. Asplund J, Wardle DA. 2017. How lichens impact on terrestrial community and ecosystem properties. Biological Reviews 92: 1720–1738. Babikova Z, Gilbert L, Bruce TJA, Birkett M, Caulfield JC, Woodcock C, Pickett JA, Johnson D. 2013. Underground signals carried through common mycelial networks warn neighbouring plants of aphid attack. Ecology Letters 16: 835– 843. Bahram M, Kõljalg U, Courty P-E, Diédhiou AG, Kjøller R, Põlme S, Ryberg M, Veldre V, Tedersoo L. 2013. The distance decay of similarity in communities of ectomycorrhizal fungi in different ecosystems and scales. Journal of Ecology 101: 1335–1344. Bahram M, Harend H, Tedersoo L. 2014. Network perspectives of ectomycorrhizal associations. Fungal Ecology 7: 70–77. Bahram M, Hildebrand F, Forslund SK, Anderson JL, Soudzilovskaia NA, Bodegom PM, Bengtsson-Palme J, Anslan S, Coelho LP, Harend H, et al. 2018. Structure and function of the global topsoil microbiome. Nature 560: 233–237. Bardgett RD, Van Der Putten WH. 2014. Belowground biodiversity and ecosystem functioning. Nature 515: 505–511. Beiler KJ, Durall DM, Simard SW, Maxwell SA, Kretzer AM. 2010. Architecture of the wood-wide web: Rhizopogon spp. genets link multiple Douglas-fir cohorts. New Phytologist 185: 543–553. 40 Bellard C, Bertelsmeier C, Leadley P, Thuiller W, Courchamp F. 2012. Impacts of climate change on the future of biodiversity. Ecology Letters 15: 365–377. Berg B. 2000. Litter decomposition and organic matter turnover in northern forest soils. Forest Ecology and Management 133: 13–22. Blackwell M. 2011. The fungi: 1, 2, 3 ... 5.1 million species? American Journal of Botany 98: 426–438. Blüthgen N, Menzel F, Blüthgen N. 2006. Measuring specialization in species interaction networks. BMC Ecology 6: 9 Brundrett MC, Tedersoo L. 2018. Evolutionary history of mycorrhizal symbioses and global host plant diversity. New Phytologist 220: 1108–1115. Buée M, Reich M, Murat C, Morin E, Nilsson RH, Uroz S et al. 2009. 454 pyrose- quenceing analyses of forest soils reveal an unexpectedly high fungal diversi- ty. New Phytologist 184: 449–456. Burgos E, Ceva H, Perazzo RPJ, Devoto M, Medan D, Zimmermann M, Delbue AM. 2007. Why nestedness in mutualistic networks? Journal of Theoretical Biology 249: 307–313. Busby PE, Ridout M, Newcombe G. 2016. Fungal endophytes: modifiers of plant disease. Plant Molecular Biology 90: 645–655. Cai W, Borlace S, Lengaigne M, Van Rensch P, Collins M, Vecchi G, Timmermann A, Santoso A, Mcphaden MJ, Wu L et al. 2014. Increasing frequency of ex- treme El Niño events due to greenhouse warming. Nature Climate Change 4: 111–116. Cantrell SA, Molina M, Jean Lodge D, Rivera-Figueroa FJ, Ortiz-Hernández ML, Marchetti AA, Cyterski MJ, Pérez-Jiménez JR. 2014. Effects of a simulated hurricane disturbance on forest floor microbial communities. Forest Ecology and Management 332: 22–31. Chenu C, Rumpel C, Lehmann J. 2015. Methods for studying soil organic matter: nature, dynamics, spatial accessibility, and interactions with minerals. In: Paul 41 EA (Ed.). Soil microbiology, ecology and biogeochemistry. Academic Press. 383-419 pp. Corrales A, Henkel TW, Smith ME. 2018. Ectomycorrhizal associations in the trop- ics – biogeography, diversity patterns and ecosystem roles. New Phytologist 220: 1076–1091. Dale VH, Joyce LA, McNulty S, Neilson RP, Ayres MP, Flannigan MD, Hanson PJ, Irland LC, Lugo AE, Peterson CJ, Simberloff D, Swanson FJ, Stocks BJ, Wot- ton MB. 2001. Climate change and forest disturbances: climate change can affect forests by altering the frequency, intensity, duration, and timing of fire, drought, introduces species, insect and pathogen outbreaks, hurricanes, windstorms, ice storms, or landslides. BioScience 51: 723–734. Dahlberg A. 2002. Effects of fire on ectomycorrhizal fungi in fennoscandian boreal forests. Silva Fennica 36: 69–80. De Vries FT, Shade A. 2013. Controls on soil microbial community stability under climate change. Frontiers in Microbiology 4: 1–16. Dicks LV, Corbet SA, Pywell RF. 2002. Compartmentalization in plant-insect flower visitor webs. Journal of Animal Ecology 71: 32–43. Diez JM, James TY, Mcmunn M, Ibáñez I. 2013. Predicting species-specific re- sponses of fungi to climatic variation using historical records. Global Change Biology 19: 3145–3154. Diffenbaugh NS, Singh D, Mankin JS, Horton DE, Swain DL, Touma D, Charland A, Liu Y, Haugen M, Tsiang M, et al. 2017. Quantifying the influence of global warming on unprecedented extreme climate events. Proceedings of the Na- tional Academy of Sciences 114: 4881–4886. Ellner SP, Fussmann G. 2003. Effects of successional dynamics on metapopula- tion persistence. Ecology 84: 882–889. 42 Fortuna MA, Stouffer DB, Olesen JM, Jordano P, Mouillot D, Krasnov BR, Poulin R, Bascompte J. 2010. Nestedness versus modularity in ecological networks: Two sides of the same coin? Journal of Animal Ecology 79: 811–817. Garrett KA, Nita M, De Wolf ED, Esker PD, Gomez-Montano L, Sparks AH. 2015. Plant Pathogens as Indicators of Climate Change. In: Letcher T, ed. Climate Change: Observed Impacts on Planet Earth: Second Edition. Elsevier B.V., 325–338. Gauthier GM. 2015. Dimorphism in Fungal Pathogens of Mammals, Plants, and Insects. PLoS Pathogens 11: 1–7. Gavito ME, Sandoval-Pérez AL, del Castillo K, Cohen-Salgado D, Colarte-Avilés ME, Mora F, Santibáñez-Rentería A, Siddique I, Urquijo-Ramos C. 2018. Re- silience of soil nutrient availability and organic matter decomposition to hurri- cane impact in a tropical dry forest ecosystem. Forest Ecology and Manage- ment 426: 81-90. Gienapp P, Teplitsky C, Alho JS, Mills JA, Merilä J. 2008. Climate change and evo- lution: disentangling environmental and genetic responses. Molecular ecology 17: 167–78. Gilarranz LJ, Rayfield B, Liñán-Cembrano G, Bascompte J, Gonzalez A. 2017. Ef- fects of network modularity on the spread of perturbation impact in experi- mental metapopulations. Science 357: 199–201. Glassman SI, Levine CR, DiRocco AM, Battles JJ. 2015. Ectomycorrhizal fungal spore bank recovery after a severe forest fire: some like it hot. The ISME Journal: 1–12. Grayston SJ, Wang S, Campbell CD, Edwards AC. 1998. Selective influence of plant species on microbial diversity in the rhizosphere. Soil Biology and Bio- chemistry 30: 369–378. 43 Grimm V, Wissel C. 1997. Babel, or the ecologial stability discussions: an inventory and analysis of terminology and a guide for avoiding confusion. Oecologia 109: 323–334. Harrison GW. 1979. Stability under environmental stress: resistance, resilience, persistence, and variability. The American Naturalist 113: 659–669. Henderson-Sellers A, Zhang H, Berz G, Emanuel K, Gray W, Landsea C, Holland G, Lighthill J, Shieh SL, Webster P, et al. 1998. Tropical Cyclones and global climate change: A post-IPCC assessment. Bulletin of the American Meteoro- logical Society 79: 19–38. Henkel TW, Terborgh J, Vilgalys RJ. 2002. Ectomycorrhizal fungi and their legumi- nous hosts in the Pakaraima mountains of Guyana. MycologicalResearch 106: 515–531. Hodgson D, McDonald JL, Hosken DJ. 2015. What do you mean, ‘resilient’? Trends in Ecology and Evolution 30: 503–506. Hoegh-Guldberg O, Mumby P, Hooten A, Steneck R, Greenfield P, Gomez E, Harvell C, Sale P, Edwards A, Caldeira K, et al. 2007. Coral reefs under rapid climate change and ocean acidification. Science 318: 1737–1743. Holden SR, Treseder KK. 2013. A meta-analysis of soil microbial biomass re- sponses to forest disturbances. Frontiers in Microbiology 4: 1–17. Holden SR, Gutiérrez A, Treseder KK. 2013. Changes in soil fungal communities, extracellular enzyme activities, and litter decomposition across a fire chrono- sequence in Alaskan boreal forest. Ecosystems 16: 34. Holling CS. 1973. Resilience and stability of ecological systems. Annual Review of Ecology, Evolution and Systematics 4: 1–23. Jaramillo VJ, Martínez-Yrízar A, Maass M, Nava-Mendoza M, Castañeda-Gómez L, Ahedo-Hernández R, Araiza S, Verduzco A. 2018. Hurricane impact on bi- ogeochemical processes in a tropical dry forest in western Mexico. Forest Ecology and Management 426: 72–80. 44 Jordano P, Vázquez D, Bascompte J. 2009. Redes complejas de interacciones mutualistas planta-animal. In: Medel R, Aizen M, Zamora R, eds. Ecología y evolución de interacciones planta-animal. Santiago de Chile: Editoral Univer- sitaria, 17–41. Kałucka IL, Jagodziński AM. 2017. Ectomycorrhizal fungi: a major player in early succession. In: Varma A, Prasad r, Tuteja N. Mycorrhiza – function, diversity, state of art. Springer, 187–229. Knutson TR, Sirutis JJ, Zhao M, Tuleya RE, Bender M, Vecchi GA, Villarini G, Chavas D. 2015. Global projections of intense tropical cyclone activity for the late twenty-first century from dynamical downscaling of CMIP5/RCP4.5 Sce- narios. American Meteorology Society 28: 7203–7224. Krug Vieira C, dos Anjos Borges LG, Marconatto L, Giongo A, Stürmer SL. 2018. Microbiome of a revegetated iron-mining site and pristine ecosystems from the Brazilian Cerrado. Applied Soil Ecology 131: 55–65. Lambers H, Raven JA, Shaver GR, Smith SE. 2008. Plant nutrient-acquisition strategies change with soil age. Trends in ecology & evolution 23: 95–103. Lanver D, Müller AN, Happel P, Schweizer G, Haas FB, Franitza M, Pellegrin C, Reissmann S, Altmüller J, Rensing SA, et al. 2018. The biotrophic develop- ment of Ustilago maydis studied by RNA-seq analysis. Plant Cell 30: 300– 323. Lee KE, Foster RC. 1991. Soil fauna and soil structure. Australian Journal of Soil Research 29: 745–775. Lewinsohn TM, Prado PI, Jordano P, Bascompte J, Olesen JM. 2006. Structure in plant-animal interaction assemblages. Oikos 113: 174–184. Martínez-Yrízar A, Jaramillo VJ, Maass M, Búrquez A, Parker G, Álvarez-Yépiz JC, Araiza S, Verduzco A, Sarukhán J. 2018. Resilience of tropical dry forest productivity to two hurricanes of different intensity in western Mexico. Forest Ecology and Management 426: 53–60. 45 Miralles DG, Van Den Berg MJ, Gash JH, Parinussa RM, De Jeu RAM, Beck HE, Holmes TRH, Jiménez C, Verhoest NEC, Dorigo WA, et al. 2014. El Niño-La Niña cycle and recent trends in continental evaporation. Nature Climate Change 4: 122–126. Mariani MS, Ren ZM, Bascompte J, Tessone CJ. 2019. Nestedness in complex networks: Observation, emergence, and implications. Physics Reports 813: 1–90. Maass M, Ahedo-Hernández R, Araiza S, Verduzco A, Martínez-Yrízar A, Jaramillo VJ, Parker G, Pascual F, García-Méndez G, Sarukhán J. 2018. Long-term (33 years) rainfall and runoff dynamics in a tropical dry forest ecosystem in west- ern Mexico: Management implications under extreme hydrometeorological events. Forest Ecology and Management 426: 7–17. Meyer K. 2016. A mathematical review of resilience in ecology. Natural Resource Modeling 29: 339–352. Montesinos-Navarro A, Segarra-Moragues J, Valiente-Banuet A, Verdu M. 2012. The network structure of plant – arbuscular mycorrhizal fungi. New Phytolo- gist 194: 536–547. Mori AS. 2016. Resilience in the Studies of Biodiversity-Ecosystem Functioning. Trends in Ecology and Evolution 31: 87–89. Moritz C, Agudo R. 2013. The future of species under climate change: Resilience or decline? Science 341: 504–508. National Oceanic and Atmospheric Administration. National Hurricane Center: Saf- fir-Simpson Hurricane Wind Scale. 2019. Última fecha de acceso: 09/04/2019 http://www.nhc.noaa.gov/aboutsshws.php Olesen JM, Bascompte J, Dupont YL, Jordano P. 2007. The modularity of pollina- tion networks. Proceedings of the National Academy of Sciences of the Unit- ed States of America 104: 19891–19896. 46 Oliver TH, Heard MS, Isaac NJB, Roy DB, Procter D, Eigenbrod F, Freckleton R, Hector A, Orme CDL, Petchey OL, et al. 2015. Biodiversity and resilience of ecosystem functions. Trends in ecology & evolution 30: 673–684. Orians G. 1975. Diversity, stability and maturity in natural ecosystems. In: van Dobben W, Lowe-McConnell R, eds. Unifying Concepts in Ecology. The Haugue, 139–150. Parker G, Martínez-Yrízar A, Álvarez-Yépiz JC, Maass M, Araiza S. 2018. Effects of hurricane disturbance on a tropical dry forest canopy in western Mexico. Forest Ecology and Management 426: 39–52. Peay KG, Kennedy PG, Talbot JM. 2016. Dimensions of biodiversity in the Earth mycobiome. Nature Reviews Microbiology 14: 434–447. Pimentel D, Garnick E, Berkowitz A, Jacobson S, Napolitano S, Black P, Valdes- Cogliano S, Vinzant B, Hudes E, Littman S. 1980. Environmental quality and natural biota. BioScience 30: 750–755. Pimm SL. 1994. Biodiversity and the Balance of Nature. In: Schulze ED, Mooney HA, eds. Biodiversity and Ecosystem Function. Springer, 347–359. Põlme S, Bahram M, Jacquemyn H, Kennedy P, Kohout P, Moora M, Oja J, Öpik M, Pecoraro L, Tedersoo L. 2018. Host preference and network properties in biotrophic plant–fungal associations. New Phytologist 217: 1230–1239. Promputtha I, Lumyong S, Dhanasekaran V, McKenzie EHC, Hyde KD, Jeewon R. 2007. A phylogenetic evaluation of whether endophytes become saprotrophs at host senescence. Microbial Ecology 53: 579–590. Purahong W, Krüger D, Buscot F, Wubet T. 2016. Correlations between the com- position of modular fungal communities and litter decomposition-associated ecosystem functions. Fungal Ecology 22: 106–114. Raffaelli D, Hall SJ. 1992. Compartments and Predation in an Estuarine Food Web. The Journal of Animal Ecology 61: 551. 47 Ramírez-López I, Villegas-Ríos M, Salas-Lizana R, Garibay-Orijel R, Alvarez- Manjarrez J. 2015. Thelephora versatilis and Thelephora pseudoversatilis: Two new cryptic species with polymorphic basidiomes inhabiting tropical de- ciduous and sub-perennial forests of the Mexican Pacific coast. Mycologia 107: 346–358. Regmi NR, Giardino JR, Vitek JD. 2013. Hazardousness of a place. In: Bobrowsky PT, ed. Encyclopedia of Natural Hazards. Dordrecht: Springer Netherlands, 435–446. Rentería LY, Jaramillo VJ, Martínez-Yrízar A, Pérez-Jiménez A. 2005. Nitrogen and phosphorus resorption in trees of a Mexican tropical dry forest. Trees - Structure and Function 19: 431–441. Rincón A, Santamaría BP, Ocaña L, Verdú M. 2014. Structure and phylogenetic diversity of post-fire ectomycorrhizal communities of maritime pine. Mycorrhi- za 24: 131–141. Sánchez-García M, Henkel TW, Aime MC, Smith ME, Matheny PB. 2016. Guy- anagarika, a new ectomycorrhizal genus of Agaricales from the Neotropics. Fungal Biology 120: 1540–1553. Schuur EAG, McGuire AD, Schädel C, Grosse G, Harden JW, Hayes DJ, Hugelius G, Koven CD, Kuhry P, Lawrence DM, et al. 2015. Climate change and the permafrost carbon feedback. Nature 520: 171–179. Seidel DJ, Fu Q, Randel WJ, Reichler TJ. 2008. Widening of the tropical belt in a changing climate. Nature geoscience 1: 21–24. Selby MJ. 1982. Magnitudes and frequencies of erosional events. In: Selby MJ (ed). Hillslope materials and processes. Oxford University Press, 222 pp. Smith JE, Cowan AD, Jumpponen A. 2018. The recovery of soil fungi following a fire. Science Findings 207. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 5 p. 207: 1–5. 48 Smith ME, Henkel TW, Catherine Aime M, Fremier AK, Vilgalys R. 2011. Ectomy- corrhizal fungal diversity and community structure on three co-occurring le- guminous canopy tree species in a Neotropical rainforest. New Phytologist 192: 699–712. Song Z, Kennedy PG, Liew FJ, Schilling JS. 2017. Fungal endophytes as priority colonizers initiating wood decomposition. Functional Ecology 31: 407–418. Standish RJ, Hobbs RJ, Mayfield MM, Bestelmeyer BT, Suding KN, Battaglia LL, Eviner V, Hawkes CV, Temperton VM, Cramer VA, et al. 2014. Resilience in ecology: Abstraction, distraction, or where the action is? Biological Conserva- tion 177: 43–51. Steidinger B, Crowther T, Liang J, Van Nuland M, Werner, GD Reich P, Nabuurs G, De-Miguel S, Zhou M, Picard N, Herault B, et al. 2019. Climatic controls of decomposition drive the global biogeography of forest tree symbioses. Nature 569: 404–408. Stendell ER, Horton TR, Bruns TD. 1999. Early effects of prescribed fire on the structure of the ectomycorrhizal fungus community in a Sierra Nevada pon- derosa pine forest. Mycological Research 103: 1353–1359. Shah F, Nicolás C, Bentzer J, Ellström M, Smits M, Rineau F, et al., 2016. Ectomy- corrhizal fungi decompose soil organic matter using oxidative mechanisms adapted from saprotrophic ancestors. New Phythologist 209: 1705–1719. Simard S, Beiler K, Bingham M. 2012. Mycorrhizal networks: mechanisms, ecology and modelling. Fungal Biology Reviews 26: 39–60. Simard S, Asay A, Beiler K, Bingham M, Deslippe J, He X, Philip L, Song Y. 2015. Resource transfer between plants throught ectomycorrhizal fungal networks. In: Horton TR, ed. Mycorrhizal Networks. Springer, 133–176. Simard S. 2018. Mycorrhizal networks facilitate tree communication, learning, and memory. In: Baluska F, Gagliano M, Witzany G, eds. Memory and Learning in plants. Springer, 191–213. 49 Smith SE, Read DJ. 2008. Mycorrhizal symbiosis. 3rd edition. Academic press. Smith ME, Henkel TW, Catherine Aime M, Fremier AK, Vilgalys R. 2011. Ectomy- corrhizal fungal diversity and community structure on three co-occurring le- guminous canopy tree species in a Neotropical rainforest. New Phytologist 192: 699–712. Spatafora JW, Chang Y, Benny GL, Lazarus K, Smith ME, Berbee ML, Bonito G, Corradi N, Grigoriev I, Gryganskyi A, et al. 2016. A phylum-level phylogenetic classification of zygomycete fungi based on genome-scale data. Mycologia 108: 1028–1046. Steidinger B, Crowther T, Liang J, Van Nuland M, Werner, GD Reich P, Nabuurs G, De-Miguel S, Zhou M, Picard N, Herault B, et al. 2019. Climatic controls of decomposition drive the global biogeography of forest tree symbioses. Nature 569: 404–408. Steren RW, Elser JJ. 2002. The stoichiometry of consumer-driven nutrient recy- cling. In: Steren RW, Elser JJ. Ecological Storichiometry: the biology of ele- ments from molecules to the biosphere. Princeton University Press, New Jer- sey, 231–260. Sukarno N, Listiyowati S, Rahayu N, Nara K. 2019. Elaphomyces tropicalis sp. nov.: A new ectomycorrhizal fungus associated with dipterocarps from tropical Indonesia. Mycoscience 60: 83–88. Suweis S, Simini F, Banavar JR, Maritan A. 2013. Emergence of structural and dy- namical properties of ecological mutualistic networks. Nature 500: 449–452. Talbot JM, Martin F, Kohler A, Henrissat B, Peay KG. 2015. Functional guild classi- fication predicts the enzymatic role of fungi in litter and soil biogeochemistry. Soil Biology and Biochemistry 88: 441–456. Tedersoo L, Bahram M, Dickie IA. 2014. Does host plant richness explain diversity of ectomycorrhizal fungi? Re-evaluation of Gao et al. (2013) data sets reveals sampling effects. Molecular Ecology 23: 992-995. 50 Tedersoo L, Bahram M, Polme S, Koljalg U, Yorou NS, Wijesundera R, Ruiz L V., Vasco-Palacios AM, Thu PQ, Suija A, et al. 2014b. Global diversity and geog- raphy of soil fungi. Science 346. Thomas DC, Vandegrift R, Ludden A, Carroll GC, Roy BA. 2016. Spatial ecology of the fungal genus Xylaria in a tropical cloud forest. Biotropica 48: 381–393. Trappe JM. 2005. A.B. Frank and mycorrhizae: is the challenge to evolutionary and ecology theory. Mycorrhiza 15: 277–281. van der Heijden MG, Martin FM, Sanders IR. 2015. Mycorrhizal ecology and evolu- tion: the past, the present, and the future. New Phytologist 205: 1406–1423. Vargas R, Hasselquist N, Allen EB, Allen MF. 2010. Effects of a hurricane disturb- ance on aboveground forest structure, arbuscular mycorrhizae and below- ground carbon in a restored tropical forest. Ecosystems 13: 118–128. Voriskova J, Baldrian P. 2013. Fungal community on decomposing leaf litter un- dergoes rapid successional changes. ISME Journal 7: 477–486. Voroney RP, Heck RJ. 2015. The soil habitat. In: Paul EA (Ed.). Soil microbiology, ecology and biogeochemistry. Academic Press. 15-39 pp. Walker LR, Wardle DA, Bardgett RD, Clarkson BD. 2010. The use of chronose- quences in studies of ecological succession and soil development. Journal of Ecology 98: 725–736. Walker LR, Wardle DA. 2014. Plant succession as an integrator of contrasting eco- logical time scales. Trends in Ecology and Evolution 29: 504–510. Wang J, Liu G, Zhang C, Wang G, Fang L, Cui Y. 2019. Higher temporal turnover of soil fungi than bacteria during long-term secondary succession in a semi- arid abandoned farmland. Soil and Tillage Research 194: 104305. Waring BG, Adams R, Branco S, Powers JS. 2016. Scale-dependent variation in nitrogen cycling and soil fungal communities along gradients of forest compo- 51 sition and age in regenerating tropical dry forests. New Phytologist 209: 845– 854. Zahran HH. 2001. Rhizobia from wild legumes: Diversity, taxonomy, ecology, nitro- gen fixation and biotechnology. Journal of Biotechnology 91: 143–153. 52 53 CAPÍTULO 1. CARYOPHYLLALES ARE THE MAIN HOST OF A UNIQUE SET OF ECTOMYCORRHIZAL FUNGI IN A NEOTROPICAL DRY FOREST Resumen Durante mucho tiempo se pensó que la simbiosis ectomicorrízica estaba restringi- da a los bosques templados. Sin embargo, a medida que se han explorado los bosques tropicales, ha quedado claro que estos hábitats albergan hongos ectomi- corrízicos (ECM) únicos. Hemos estado explorando los bosques tropicales caduci- folios, que son ecosistemas terrestres en peligro de extinción y puntos calientes de endemismo. Debido a que Fabaceae es la principal familia de plantas en este en- torno, planteamos la hipótesis de que los árboles en este linaje serían los principa- les huéspedes ectomicorrízicos. Hemos secuenciado la región de ITS rDNA de los hongos, y para plantas rbcL y trnL cDNA de para identificar a ambos simbiontes de las puntas micorrizadas. La posición sistemática de cada simbionte fue confirmada por inferencia filogenética bayesiana. Identificamos 20 especies de plantas perte- necientes a 10 familias que hospedaron 19 especies únicas de hongos ECM de 5 linajes. La mayoría de los hongos ECM se asociaron con plantas del orden Caryo- phyllales, no con Fabaceae. Achatocarpus y Guapira, los hospederos principales, están dispersos por todo el bosque y no están en parches monodominantes. La baja diversidad de hongos ECM puede explicarse por la baja densidad de las plan- tas hospederas y su alta especificidad. Nuestros resultados indican que Caryophy- llales es un orden importante de hospederos ectomicorrízicos tropicales con al menos cuatro linajes independientes que en su proceso evolutivo desarrollaron la capacidad de formar ectomicorrizas. Mycorrhiza https://doi.org/10.1007/500572-017-0807-7 M) CrossMark ORIGINAL ARTICLE Caryophyllales are the main hosts of a unique set of ectomycorrhizal fungi in a Neotropical dry forest Julieta Alvarez-Manjarrez'” . Roberto Garibay-Orijel' (3) - Matthew E. Smith? Received: 26 June 2017 /Accepted: 27 October 2017 CO) Springer-Verlag GmbH Germany 2017 Abstract The ectomycorrhizal symbiosis was long thought to be restricted to temperate forests. However, as tropical for- ests have been explored, it has become clear that these habitats host unique ectomycorrhizal (ECM) fungi. We have been ex- ploring tropical dry forests (TDF), which are endangered ter- restrial ecosystems and hotspots of endemism. Since Fabaceae is the main plant family in this environment, we hypothesized that trees in this lineage would be the main ECM hosts. We sequenced the ITS rDNA region from fungi and both rbeL and trnl cpDNA from plants to identify both symbiotic partners from root tips. The systematic position of each symbiont was confirmed by Bayesian phylogenetic inference. We identified 20 plant species belonging to 10 families that hosted 19 unique ECM fungal species from 5 lineages. Most ECM fungi were associated with Caryophyllales, not with Fabaceae. Achatocarpus and Guapira, the main hosts, are scattered throughout the forest and are not in monodominant patches. The low ECM fungal diversity can be explained by the low density of host plants and their high specificity. Our results Electronic supplementary material The online version of this article (https://doi.org/10.1007/500572-017-0807-7) contains supplementary material, which is available to authorized users. LJ Roberto Garibay-Orijel rearibay O ib.unam.mx Instituto de Biología, Universidad Nacional Autónoma de México, Tercer Circuito, Ciudad Universitaria. Del. Coyoacán, 04510 Ciudad de México, Cd Mx, Mexico ba Posgrado en Ciencias Biológicas, Edificio B, 1? Piso. Circuito de Posgrados, Ciudad Universitaria. Del. Coyoacán, 04510 Ciudad de México, Cd Mx, Mexico Department of Plant Pathology, University of Florida, Gainesville, FL 32611-0680, USA Published online: 27 November 2017 indicate that Caryophyllales is an important order of tropical ECM hosts with at least four independent evolutionary line- ages that have evolved the ability to form ectomycorrhizae. Keywords Achatocarpus - Caryophyllales - Chamela - ECM fungal community - ECM hosts - Guapira - Neotropical dry forest Introduction Nearly 90% of terrestrial plants acquire nutrients from soil via symbioses with mycorrhizal fungi. Mycorrhizal diversity 1s estimated to involve as many as 50,000 fungal species asso- ciated with 250,000 plant species (van der Heijden et al. 2015). Arbuscular mycorrhiza and ectomycorrhiza are the most common interactions. While much is known about taxa involved in arbuscular mycorrhizal relationships, there is less information concerning ectomycorrhizae, particularly in trop- ical ecosystems. Recently, there has been a growing interest in understanding both the biogeography and the diversity of the ectomycorrhizal (ECM) symbionts in temperate versus tropi- cal habitats (Bahram et al. 2013; Tedersoo et al. 2014a; Looney et al. 2016). It 15 known that ECM fungal diversity is higher in temperate forests compared to tropical forests, opposite to the trend for plant diversity (Tedersoo et al. 2014b) and other organisms. In those tropical forests where ECM plants are monodominant, ECM associations prevail (Fukami et al. 2017). Monodominance refers to a single plant species occu- pying large territories and allowing the establishment of high- ly diverse ECM fungi communities; this was observed in for- ests dominated by Dipterocarpaceae and some Fabaceae spe- cies. For example, the Dipterocarpus tropical rainforest of Thailand hosts 57 ECM fungal species (Dell et al. 2005), Y Springer 54 Mycorrhiza and the Malaysian dipterocarp rainforest hosts more than 90 species (Peay et al. 2010). Additionally, five leguminous spe- cies associated with 39 ECM fungi were found in the tropical rainforest in Guinea (Diédhiou et al. 2010). Similarly, in the Guiana Shield, species of Dicymbe and Aldina (Fabaceae) host a diverse community of more than 172 ECM fungal spe- cies (Smith et al. 2011, 2017). In contrast, most tropical forests are characterized by plant hyperdiversity and ECM hosts are scattered so that ECM associations are rare (Alexander 2006; Brundrett 2009; Peay 2016). In these ecosystems, few studies have been conducted on the ecology and diversity of ECM associations (e.g., Haug et al. 2005; Tedersoo et al. 2007, 2010). The ECM fungal richness in these forests is low com- pared to the richness of ECM monodominant tropical forests (Tedersoo et al. 2010). Accordingly, ECM fungal richness in the tropics can generally be predicted by the abundance (e.g., basal area) of ECM hosts and their diversity (Tedersoo et al. 2014b). The ECM symbiosis evolved independently in 145 genera from 24 families of plants (Wang and Qiu 2006; Brundrett 2009). Tropical ecosystems have revealed new evidence of phylogenetically disparate plants that form ECM associations. The symbiosis of Asteropeia (Asteropelaceae), Guapira, Neea and Pisonia (Nyctaginaceae), Coccoloba (Polygonaceae), Uapaca (Phyllantaceae), Gnetum (Gnetaceae), Aldina (Fabaceae subfamily Papilionoideae), and Oreomunnea (Junglandaceae) has only recently been documented or veri- fied (Wang and Qiu 2006; Tedersoo et al. 2007; Smith et al. 2011; Bá et al. 2012; Hayward and Horton 2012; Séne et al. 2015; Alvarez-Manjarrez et al. 2016; Corrales et al. 2016). Asteropeiaceae, Nyctaginaceae, and Polygonaceae belong to Caryophyllales, a plant order that is assumed to develop arbuscular mycorrhizas. Considering that tropical trees are hyperdiverse and that many tropical habitats remain unex- plored, it seems highly probable that many novel ECM asso- ciations still await discovery in the tropics. Neotropical forests may not only harbor unknown ECM hosts, but they also could be sites of origin and diversification for different ECM fungal lineages (e.g., Clavulina, Inocybaceae; Matheny et al. 2009; Kennedy et al. 2012). Most of this knowledge comes from rainforests; however, tropical dry forests (TDF) still remain largely unexplored. These forests are hyperdiverse (DRYFLOR 2016) and no monodominant plant taxa are known from TDFs. Fabaceae is one of the most extensive families in this ecosystem, with high richness and endemism (Linares-Palomino et al. 2006). TDFs are among the most imperiled ecosystems on the planet because land-use change has been accelerated by agricultural practices or grazing over large expanses of habitat (DRYFLOR 2016). TDFs are unique and globally important habitats, covering a total of 1,048,700 km? distributed throughout northern Australia, Southeast Asia, Africa, and the Americas (Miles et al. 2006). In the Neotropics, TDF 4 Springer comprise about 519,597 km”, 51% of them are in South America, 39% in North and Central America, and 9% in the Caribbean (Portillo-Quintero and Sánchez-Azofeifa 2010). Even when the aboveground plant ecology is well understood in TDFs, information of soil symbiotic interactions is scarce. This knowledge is urgently needed in order to understand the contribution of ECM fungi to the nutrient cycling and rhizo- sphere processes in this ecosystem. In this research, we examined the diversity and community structure of ECM fungi associated with several plant species in a Mexican TDF. Our main questions were as follows: (1) Is Fabaceae the main ECM host lineage in TDF? (2) Who are the ECM fungi associated with the plant hosts and what are their phylogenetic relationships? (3) What is the morpho-anatomy of these associations? and (4) Are the fungal-plant interactions in the TDF dominated by generalists or specialists? Materials and methods Study site This study took place in the biological research station of Chamela, Jalisco, Mexico (N 19% 30', W 1059 03”. The site occupies 3319 ha of TDF and tropical sub-deciduous forest. The most dominant plant family is Fabaceae with 160 species, of which 57 species are trees. Besides legumes, other common families harboring trees are Euphorbiaceae (26 species), Anacardiaceae (3), Annonaceae (5), Nyctaginaceae (3), and Polygonaceae (9) (Lott and Atkinson 2002). The mean annual precipitation is 784 mm, approximately 85% of annual pre- cipitation is concentrated between July to October; the rest of the year is dry. However, precipitation is heterogeneous de- pending on the hurricane season. The mean annual tempera- ture is 24.6 *C with a maximum of 30.3 *C and minimum of 19.5 *C (García-Oliva et al. 1995). Soils of the region are young Entisols which are poorly developed with low organic matter and pH of 6-7 (Martínez-Yrízar and Sarukhán 1993). We performed four visits for sampling during the rainy sea- son: July-August 2012, October-November 2012, October 2013 and November 2014. Root sampling We used the published flora of the region (Lott 1993) to iden- tify putative ECM host plants in the Fabaceae, Nyctaginaceae, and Polygonaceae to conduct directed root sampling, focused on Fabaceae hosts. We collected both rhizosphere and soil core samples. Rhizosphere samples were extracted directly from suspected ECM hosts by tracking their roots. Soil sam- ples consisted of four pooled soil cores (2.5 cm diame- ter x 30 cm deep) taken under suspected ECM hosts or ECM sporocarps (Fig. Sl). In total, we sampled 13 55 Mycorrhiza rhizospheres and 91 soil cores (-53,508 em?), 61 under suspected ECM plants and 30 under putative ECM fruit bod- 1es. We sieved each sample to isolate fine roots and dissected roots that had a fungal mantle under a stereomicroscope. We further characterized clean root tips according to Agerer and Rambold (2004-2017) and preserved root tip vouchers in 96% ethanol at 4 *C until processed. Plant identification and phylogenetic analysis We identified the plant host of root tip samples using molec- ular techniques. We extracted DNA from leaves and roots with the REDExtract-N-Amp Plant kit (Sigma-Aldrich, St. Louis, MO, USA). We PCR-amplified the chloroplast rbcL region using rbcL-a_F and rbcL-a_R primers following Kress and Erickson (2007); we amplified the trnL region using trnC, trnD, trnE, and trnF primers following Taberlet et al. (1991). Amplicons were managed and sequenced as in Ángeles-Argáiz et al. (2016). We edited sequences, assem- bled, and clustered into MOTUSs at 100% similarity with Geneious v.6.1.4 (Biomatters Ltd.). We compared sequences with GenBank and/or the BOLD database (www. boldsystems.org/) using MEGABlast (Zhang et al. 2000). We produced sequences from MEXU herbarium vouchers (Instituto de Biología, UNAM) of the families that were most commonly associated with the root tip samples. A small leaf portion (2 mm) was taken from plant species of Achatocarpaceae, Nyctaginaceae, and Polygonaceae. From these tissues, we extracted DNA and sequenced the frnL and rbcL regions. Accession numbers from MEXU and GenBank of each species is presented in Table Sl. We generated a Bayesian phylogenetic analysis for the or- der Caryophyllales using the rbc£L region, including vouchers and root tip sequences, BOLD Systems, and GenBank se- quences. The alignment was performed using MAFFT v7 (http://mafft.cbcr.jp/alignment/server/) and revised manually with PhyDE. The substitution model was selected by jModelTest v2.1.3 (Darriba et al. 2012) using BIC calcula- tions. Mr. Bayes (Ronquist and Huelsenbeck 2003) was used for phylogenetic analyses. The model used was HKY +1 + G for 10,000,000 generations with 4 MCMC chains and a sam- pling frequency of 100 with a print frequency was 500. The branch support was computed by posterior probabilities and we generated a consensus majority rule tree. Fungal identification and phylogenetic analysis All terrestrial sporocarps in the study site were collected using plastic boxes and wax paper. From all materials, we made macroscopic descriptions including fresh colors coded with Kornerup and Wanscher (1978). A small piece of each fruit body was preserved in 96% ethanol for DNA extraction. Sporocarps were dehydrated with hot air (60 *C) and depos- ited in the MEXU herbarium. We extracted DNA from the fruit bodies with the REDExtract-N-Amp Plant kit (Sigma-Aldrich, St. Louis, MO, USA). The fungal internal transcribed spacer (ITS) re- gion was PCR-amplified with ITSIF and IT'S4 primers following Garibay Orijel et al. (2013) using the RubyTaq PCR master mix (Affymetrix, CA, USA). Amplicons were manipulated and sequenced with the same protocols we re- ported in the plant identification section. Sequences were edited, assembled, and clustered into MOTUSs with Geneious v.6.1.4 (Biomatters Ltd.). Fungal MOTUSs were assembled at 97% nucleotide similarity (Smith et al. 2007). Consensus se- quences were compared with GenBank and UNITE using MEGA Blast (Zhang et al. 2000). We conducted fungal identification (root tips and fruit bod- 1es) by phylogenetic analyses of the /clavulina, /inocybe, /Mlactarius-russula, /thelephora-tomentella and /sebacina line- ages. For this, we downloaded the close relatives that resulted from MEGABlast searches from GenBank and UNITE and aligned them with our sequences using MAFFT v7 (http:// mafft.cbcr.jp/alignment/server/) and PhyDE. Bayesian analysis was run in Mr. Bayes v3.2.2, over 5,000,000 generations with 4 MCMC chains and three partitions (ITS1, 5.85, ITS2). To generate a consensus majority rule tree with the posterior probabilities, we used the GTR + I + G substitution model, sampling frequency of 100, and printing frequency of 500. Characterization of mycorrhizae Samples with abundant root tips were fixed in FAA or Navashin for several hours and then later washed with tap water and preserved in 70% ethanol. We observed the fixed samples with a stereoscope microscope to characterize their morphology, using DEEMY as a reference (Agerer and Rambold 2004-2017). To characterize the mantle, Hartig net and external mycelium for each sample, we made anatomical slices, which were dyed with safranin and fast green FCF following the protocol from Sandoval-Zapotitla (2005). Results Ectomycorrhizal host plants We identified 98 root samples either by rbeL or trnl coDNA regions from 201 roots (48.75% success rate). The rbcL re- gion was inconclusive to approximate the genera of several ECM hosts. For instance, Guapira petenensis was grouped with Neea and Pisonia GenBank accessions. When we added the trnL of voucher sequences, the root tip sequences matched at 100% with G. petenensis. A Springer 56 Mycorrhiza Despite the fact that the sampling was targeted to Fabaceae (1.e., Caesalpinia spp., Cynometra oaxacana, Haematoxylum brasiletto, Lonchocarpus spp.), the most common plants am- plified from root tips with a fungal mantle were Achatocarpus gracilis (Achatocarpaceae) and G. petenensis (Nyctaginaceac), with 53 and 13% of the root tips, respectively. Achatocarpus gracilis shrubs were identified based on morphological com- parisons between field observations and herbarium specimens. Moreover, the rbcL sequences from root tips were identical with leaf voucher sequences and with GenBank sequences of Achatocarpus and Phaulothamnus species (Fig. 1). The main ECM host species belonged to Caryophyllales: A. gracilis, Coccoloba sp., G. petenensis, Pisonia Sp., Pisonieae sp., Ruprechtia sp. and five more unidentified Nyctaginaceae species. Despite the fact that Achatocarpus and Ruprechtia were commonly colonized by ECM fungi and were clearly covered by a mantle of ECM fungal hyphae, we were unable to observe the Hartig net in the mycorrhizas. We also found typically non-ECM plant lineages whose roots were colonized by ECM fungi but without Hartig net: Apocynaceae, Araliaceae (Aralia excelsa), Caesalpinoideae, Papilionoideae (Apoplanesia paniculata, Lonchocarpus sp.), Moraceae (Ficus sp.), Sapindaceae, Sapotaceae, and Surinaceae (Recchia mexicana). All of these had low coloni- zation by ECM fungi in their roots and they represent only 20.6% of the samples (Table 1). Fungal diversity In three sampling years, 26 ECM fruit bodies were found corresponding to: 18 of Thelephora versatilis, 6 of Tremelloscypha sp., 1 of Tomentella brunneoincrustata, and l of Thelephora pseudoversatilis. Similarly, ectomycorrhizae were seldom found in the soil cores: in c. 53,600 cm? of soil, we found 209 root tips with a fungal mantle. In fact, some of the soil cores did not have any ectomycorrhizae, whereas nearly all roots were ECM in a few soil cores. On average, there were 2.2 ECM root tips per soil core, with the maximum being 21 root tips in one soil core below Guapira. The ectomycorrhizae were located in restricted places, distributed in patches according to their host plant distribution. We successfully sequenced the ITS region of 115 from 209 roots samples, belonging to ECM and saprotrophic fungi. Most samples (109 root tips) belonged to ECM fungi and corresponded to 19 species (Table 2). Only ECM Basidiomycota species were found on the roots but one unidentified Ascomycota species (Dothideomycetes) was also sequenced (Table S2). ECM species belonged to five distinct lineages of fungi: /clavulina, /inocybe, /russula-lactarius, /sebacina, /thelephora-tomentella (Tedersoo and Smith 2013). The /sebacina lineage was present in 39% ofroot tips and was represented by Tremelloscypha sp. and Sebacina sp. (Figs. S2, S3). Tremelloscypha sp. was resolved in a Neotropical clade with A Springer samples from Mexico to Venezuela with posterior probability (PP) of 1; Sebacina sp. was related to Neartic species from Mexico with low support. The /thelephora-tomentella lineage corresponded to 32% of the total samples and was represented by two Thelephora species and six Tomentella species. All the Thelephoraceae belonged to tropical clades: 7h. versatilis, Th. pseudoversatilis, Tomentella brunneoincrustata, Tomentella sp. 1, Tomentella sp. 4, Tomentella sp. 5 belong to a clade supported by 0.98 PP; Tomentella sp. 2 and Tomentella sp. 3 belong to two different tropical clades with 0.88 and 1 PP, respectively (Fig. S4). The /clavulina lineage was also common and diverse (20% of root tips) with five different species, four Clavulina and one Membranomyces. The Membranomyces species was related with a clade from temperate forests with 0.98 PP. Clavulina spp. were closely related each other; there was no PP support for this relationship (Fig. S5). The lineages /inocybe and /russula- lactarius were each represented by just one or two species. The sequences from /russula-lactarius belonged to two independent tropical clades that also included Ecuadorian species, each with a PP of 1 (Fig. S6). Also /inocybe formed two independent clades with long branches, Inocybe sp. 2 was related with Australian species with low support (Fig. S7). Mycorrhizal morphotypes The mycorrhizal systems established with TDF plants were small in size (1.e., length of -1-3 mm). They usually were unbranched and showed different patterns of soil exploration depending on fungal species. Guapira morphotypes had all of the typical ectomycorrhiza characteristics including the mantle, Hartig net and external mycelia (Fig. 2). These morphotypes were evident, almost observable without magnification, and they varied depending on the mycobiont. Clavulina sp. 2 (Fig. 2a, b) presented a beige mantle formed by hyphae with refracting incrustations, with a mantle of 13-16 um, and the Hartig net was paraepidermal. The ECM of Membranomyces sp. (Fig. 2c, d) had a white mantle with a smooth surface contact explora- tion type, a paraepidermal Hartig net with infrequent lobules, and was present in the first cortical layer of the root cells. This species also had extensive intraradical hyphae (Fig. 2d). On the other hand, Tomentella sp. 1 associated with Pisonieae sp. (Fig. 2e, £) had a sinuous mycorrhiza with a dark brown mantle and a few emergent hyphae. Its mantle did not cover the tip and the Hartig net was paraepidermal with lobed septa. Meanwhile, when Tomentella sp.1 was associated with Polygonaceae, it had a copper colored mantle, with sinuous tips and a completely smooth surface. We found some morphotypes with very particular and un- usual characteristics. In the 7h. versatilis ectomycorrhizae, the hairy mantle with medium-distance exploration did not cover the root tip, with partial mantle zones on the roots. Also, Achatocarpus and /sebacina have a very thin, hyaline, and 57 Mycorrhiza %, Achatocarpaceae Didieracogo Portuaccaceae Cactaceae pizoaceae e ciseniace? 2e ayas Nyctaginace 0.2 Fig. 1 Bayesian phylogenetic analysis of the Caryophyllales based on rbcL. The four clades that are known to contain ectomycorrhizal plants are highlighted by blue boxes. The Achatocarpaceae clade includes almost imperceptible to white mantle, with short-distance ex- ploration type. The mantle was composed by 3-4 hyphae layers (< 12 um) with some emergent clamped hyphae that aggregate the nearby soil. Achatocarpus does not form an evident Hartig net with any of the ECM fungi that we examined (Fig. 2g, h). Fungal-host interaction We found 19 ECM fungal species associated in different pro- portions with 10 plant families (Table 1). In general, the Di on co ph yl la ce ae ygonaceae Y Y O, e % > 2 % O a RG 2 e o y o o 3 42 2 8 32%% >" oO <= .-—_ o Y D DS g 3. e 3 $ 3 $ o. sequences from the BOLD database as well as those from mycorrhizal roots (MR) and fresh leaves (voucher) collected at Chamela /sebacina lineage (Tremelloscypha sp. and Sebacina sp.) exhib- ited a preferential association with A. gracilis. Although Tremelloscypha sp. was found forming a fungal mantle on roots of Achatocarpus, Ficus sp. and Sapotaceae sp., a Hartig net was not observed. The /clavulina lineage was strongly associ- ated with Nyctaginaceae members, except for Clavulina sp. 1, which was associated with Achatocarpus (Fig. 3). Achatocarpus and Guapira, the two main ECM hosts, shared only two fungal species: 7. dichroa and Tomentella sp. 1. Some plant species were found hosting only one ECM fungal A Springer 58 Mycorrhiza Table 1 Ectomycorrhizal symbionts in the tropical dry forest of Chamela, Mexico Order Family Plant species No. RT Fungal species Hartig net Apiales Araliaceae Aralia excelsa 1 Thelephora versatilis 2 Caryophyllales Achatocarpaceae Achatocarpus gracilis 25 Tremelloscypha sp. No 15 Sebacina sp. No 3 Clavulina sp. 1 2 3 Tomentella sp. 2 2 l Inocybe sp. 2 ? 1 Russula sp. 1 2 1 Th. versatilis 2 1 Tomentella sp. 1 2 Caryophyllales Nyctaginaceac Guapira petenensis 6 Membranomyces sp. Yes > Tomentella sp. 1 2 l Clavulina sp. 2 ? l Tomentella brunneoincrustata Yes 1 Tremelloscypha sp. 2 Caryophyllales Nyctaginaceae Nyctaginaceae sp. 1 4 Membranomyces Sp. Yes 2 Tomentella sp. 1 2 Caryophyllales Nyctaginaceae subfam. Pisonicae Pisonicae sp. 3 To. brunneoincrustata Yes 1 Tomentella sp. 1 Yes Caryophyllales Nyctaginaceac Pisonia sp. l To. brunneoincrustata Yes 1 Inocybe sp. 1 ? Caryophyllales Nyctaginaceae Nyctaginaceae sp. 2 l Clavulina sp. 4 ? Caryophyllales Nyctaginaceac Nyctaginaceae sp. 3 l Clavulina sp. 3 2 Caryophyllales Nyctaginaceac Nyctaginaceac sp. 4 l Russula sp. 1 2 Caryophyllales Nyctaginaceae Nyctaginaceae sp. 5 l Tomentella sp. 4 ? Caryophyllales Polygonaceae Coccoloba sp. 3 Th. versatilis 2 Caryophyllales Polygonaceae Ruprechtia sp. 2 Th. versatilis 2 Ericales Sapotaceae Sapotaceae sp. l Tremelloscypha sp. ? Fabales Fabaceae subfam. Caesalpinoideae Caesalpinoideae sp. l Tomentella sp. 3 2 Fabales Fabaceae subfam. Papilionoideae Papilionoideae sp. 1 Tomentella sp. 5 1 Fabales Fabaceae subfam. Papilionoideae Lonchocarpus sp. l Tomentella sp. 1 ? Fabales Fabaceae subfam. Papilionoideac Apoplanesia paniculata 1 Russulaceace sp. 2 Fabales Surianaceae Recchia mexicana l Th. versatilis E Gentianales Apocynaceac Apocynaceae sp. l Tomentella sp. 1 2 Rosales Moraceae Ficus sp. 1 Tremelloscypha sp. 2 No. RT Number on root tips, ? unknown species that was not associated with any other tree. For instance, A. paniculata was associated with Russulaceae sp., Caesalpinioideae sp. with Tomentella sp. 3, and Papilionoideae sp. with Tomentella sp. 5. Certain species of Tomentella and Thelephora were the most generalist in this ecosystem. Thelephora versatilis was associ- Discussion ated with A. gracilis, A. excelsa, R. mexicana, and Ruprechtia sp. Similarly, Tomentella sp. 1 interacted with roots of A. gracilis, Coccoloba sp., G. petenensis, Lonchocarpus sp., and other Nyctaginaceae species. We also found saprotrophic fungi and species with un- known function on roots (Table S2), including Agaricales A Springer sp. on G. petenensis, Agaricales sp. on Caesalpinioideae, Dothidiomycetes sp. on Polygonaceae, and Entoloma sp. on Sapindaceae. The Caryophyllales as the main ectomycorrhizal hosts in the tropical dry forest Although Fabaceae is the most species-rich family in the study site (155 species; Lott 1993), our results demonstrate that 59 Mycorrhiza Table 2 Taxonomic identity and root tip abundance of ectomycorrhizal fungi Taxon name N - Accession number UNITE SH] DOI Best BLAST match Percent similarity Tremelloscypha sp. (KF061282) 98% (649/660) Clavulina cinerea (JIN228228) Sebacina sp. (FJ378741) Tomentella sp. (EU625861) 93% (641/689) Thelephora versatilis (KJ462494) 100% (665/665) Tomentella brunneoincrustata (KP896288) 100% (628/628) 83% (753/910) 97% (641/660) Uncultured Tomentella (EU625861) Clavulina sp. (3X287358) Russula nigricans (EU8 19428) Thelephora pseudoversatilis (3X075890) 93% (641/689) 93% (676/730) 84% (618/733) 99% (635/642) Inocybe sp. (GQ469526) Clavulina sp. (3X287358) Clavulina cf. amethystina (GUS50110) Clavulina sp. (KF359593) Inocybe calospora (HQ5S86852) Russula acrifolia (UDB002470) Tomentella sp. (KF472143) 81% (569/702) 93% (669/717) 90% (581/640) 92% (612/668) 81% (566/702) 91% (679/746) 97% (634/683) Tomentella ellisii (AB634268) 94% (541/570) Tremelloscypha sp. 29 KM269089 SH016792.07FU| 10.15156/BI10/SH016792.07FU Membranomyces sp. 15 KU1l75675 SH487941.07FU Sebacina sp. 15 KP896286 SH488205.07FU Tomentella sp. 1 12 KM269095 SH495677.07FU Thelephora versatilis* 9 KM269094 SH490448.07FU Tomentella 6 KP896288 SH489022.07FU brunneo-incrustata* Tomentella sp. 2 4 KM269090 SH496760.07FU Clavulina sp. 1 3 KM269091 SH493124.07FU Russulaceac sp. 3 KM269093 SH489887.07FU Thelephora 2 KUIl75679 SHO010135.07FU| pseudoversatilis* 10.15156/BI0/SH010135.07FU Inocybe sp. 1 2 KP896287 SH493665.07FU Clavulina sp. 2 1 KP896290 SH488204.07FU Clavulina sp. 3 Il KP896291 SH495009.07FU Clavulina sp. 4 1 KUI75676 SH496187.07FU Inocybe sp. 2 ]1 KM269092 - Russula sp. 1 1 KUI75680 SH491260.07FU Tomentella sp. 3 1] KUI75681 SH009960.07FU| 10.15156/B10/SH009960.07FU Tomentella sp. 4 Il KP896289 SH492685.07FU Tomentella sp. 5 | KUI75682 SH493257.07FU Uncultured Thelephoraceae (KF836018) 95% (599/630) N: Number of root tips. *Sequences are 100% similar to those of Thelephora versatilis, Thelephora pseudoversatilis, and Tomentella brunneoincrustata sporocarps collected at the same site (Ramirez-Lopez et al. 2015; Alvarez-Manjarrez et al. 2016) legumes were not frequent ECM hosts in the TDF (Table 1). This is in contrast to other tropical ecosystems around the world where leguminous plants are confirmed as common ECM hosts (e.g., Tedersoo et al. 2007; McGuire et al. 2008; Diédhiou et al. 2010; Smith et al. 2011). Achatocarpus gracilis, G. petenensis, Pisonia sp., Pisonieae sp., Nyctaginaceae spp., Coccoloba sp., and Ruprechtia sp., all from Caryophyllales, were the most im- portant ECM hosts in these TDF (16 of the 19 fungal species were associated with this group). Achatocarpus gracilis and G. petenensis are the host plants most frequently associated with ECM fungi. These species of trees (Coccoloba, Guapira and Ruprechtia) or shrubs (Achatocarpus and Pisonia) have low density at the study site and are also typically found at low density in other tropical forests (e.g., Tedersoo et al. 2010, 2012). Achatocarpus gracilis belongs to a monophyletic clade that includes Caryophyllaceae and Amaranthaceae (Scháferhoff etal. 2009; Crawley and Hilu 2012) but there are no members from this clade that have thus far been reported as ECM (Wang and Qiu 2006). This suggests that the symbiosis with ECM fungi probably evolved independently in the Achatocarpus lineage, and it may be found in other members of Achatocarpaceae. This host species forms a “pisonioid” mycorrhizal type whereby none of the ECM fungi associated with Achatocarpus roots forms a conspicuous Hartig net (Ashford and Allaway 1982; Imhof 2009). At the Chamela site, A. gracilis does not occupy an extensive basal area, but it is widely distributed and found in all successional stages of TDF (Alvarez-Añorve et al. 2012). To understand the impor- tance of the symbiosis with ECM fungi in this lineage, it would be necessary to answer the following questions: Do other species of the genus also develop this regular association with ECM fungi? Despite the lack of a Hartig net, is there nutrient transfer between A. gracilis and its associated ECM fungi? How do interactions with ECM fungi contribute to the fitness of this shrub in the successional forests? Nyctaginaceae was the family with the highest number of ECM host species. In this family, we were unable to identify 10.3% of the plants at the species level even after comparison to voucher sequences, because of the low resolution of the chloroplast markers. Within this family, putative ECM hosts present in the study site could be species of Abromia, Boerhavia, Commicarpus, Mirabilis, Okenia or Salpianthus (Lott 1993). However, none of these genera are known to form ECM associations. The Nyctaginaceae species with the most frequent ectomycorrhizae was G. petenensis, which is an abundant tree in the study site (almost 80 individuals per ha) occupying a basal area around 1 m”/ha (Durán et al. 2002). Guapira and Nyctaginaceae in general are known as ECM Y Springer 60 Mycorrhiza Fig. 2 a Ectomycorrhiza of Clavulina sp. 2 + Guapira petenensis. b Ectomycorrhiza of Clavulina sp. 2 + G. petenensis in longitudinal which shows the mantle and a detail of the Hartig net. e Ectomycorrhiza of Membranomyces sp. + Guapira pentensis. d Longitudinal view of Membranomyces sp. + G. petenensis ECM showing the mantle, the Hartig net, and numerous intraradical hyphae (arrows). e Ectomycorrhiza of Tomentella sp. 1 + Pisonieae sp. f Transversal slice that shows a dense mantle and a paraepidermal Hartig net (arrows). g Sebacina sp. + Achatocarpus gracilis root tips. h Sebacina sp. + Achatocarpus gracilis in transversal view, it shows the mantle and lack of Hartig net. Bars: a, ce, e 0.25 mm; b, £4.5 um; d 75 um; g 0.5 mm; h 6.6 um hosts throughout the tropics (Ashford and Allaway 1982; Chambers et al. 2005; Haug et al. 2005; Tedersoo et al. 2010; Hayward and Horton 2012). Another recognized ECM host family is Polygonaceae, especially the genus Coccoloba (e.g., Bandou et al. 2006; Séne et al. 2015). There have been recent new reports of Polygonaceae species with ECM associations, for example Gymnopodium floribundum (Bandala et al. 2011). Our results showed that Coccoloba sp. and Ruprechtia sp. were associat- ed with ECM fungi. We were not able to observe a Hartig net on Ruprechtia roots but this plant associated with Th. versatilis, a common ECM fungus in Chamela. Ruprechtia is a genus of Neotropical trees that primarily inhabits TDFs 4 Springer and is distributed from Mexico to Argentina and Uruguay (Pendry 2004). Ruprechtia (Triplarideae tribe), Coccoloba (Coccolobaeae), and Gymnopodium (Gymnopodieae) consti- tute the subfamily Eriogonoideae (Burke and Sanchez 2011), so it is likely that all of the plants in this subfamily may form ECM associations. The Eriogonoideae represents another dis- tinct ECM lineage within Caryophyllales. Interestingly, all the ECM Caryophyllales lineages except Bistorta are primarily found in tropical and subtropical forests. Caryophyllales is one of the orders with the highest speciation rate in angiosperms (Magallón and Castillo 2009). Most Caryophyllales species are assumed to develop arbuscular my- corrhizas or are non-mycorrhizal (Maherali et al. 2016). However, available data suggest that this group is an important lineage of ECM tropical hosts with at least four independent evolutionary lineages of ECM symbiosis in the families Achatocarpaceae (Achatocarpus), Asteropeiaceae (Asteropeia), Nyctaginaceae (Guapira, Neea, Pisonia), and Polygonaceae (Bistorta, Coccoloba, Gymnopodium, Ruprechtia) (Haug et al. 2005; Wang and Qiu 2006; Ducousso et al. 2008; Bandala et al. 2011; this study). The hyperdiverse tropical dry forest harbors a unique set of ectomycorrhizal fungi Previously the Neotropical dry forest in the Pacific coast was considered to be an exclusive arbuscular mycorrhizal ecosys- tem but our data suggest this habitat is home to a unique ECM fungal community. When including ectomycorrhizae and fruit bodies, we found 19 species of ECM fungi despite the fact that ectomycorrhizae were rare. We found 209 root tips in 91 soil samples (-53,600 cm of soil); this is 0.003 ECM root tips/cm? soil. Fruit bodies of ECM fungi were even less frequent; in approximately 60 field days, we obtained only 28 collections of ECM sporocarps. Although scarcely found, all the ECM fungal species, except Tremelloscypha sp., are unique lineages and are only known from the TDF on the Pacific coast of Mexico. The IT'S sequences of these taxa share only approxi- mately 90% nucleotide similarity with any of the records in public databases. Thus, ca. 95% of species we collected are likely new to science and our research group has recently de- scribed some of these taxa as new species (Tomentella brunneoincrustata, Thelephora pseudoversatilis, Th. versatilis) (Ramirez-Lopez et al. 2015; Alvarez-Manjarrez et al. 2016). In spite of the recent studies on soil diversity in the TDF, our identified fungal taxa did not match those from Costa Rica based on sequence comparisons (data not shown; Waring et al. 2016). These mismatches contrast with the high number of shared plant species between Mexico and Central America- northern South America (DRYFLOR 2016). Nonetheless, the fungal lineages found here are common lineages in many for- ests worldwide: /clavulina, /inocybe, /russula-lactarius, /sebacina, and /tomentella-thelephora. 61 Mycorrhiza ECM Fungi Tremelloscypha sp. Membranomyces sp. Sebacina sp. Thelephora versatilis Tomentella sp. 1 Tomentella brunneoincrustata Tomentella sp. 2 Clavulina sp. 1 Russula sp. 1 Inocybe sp. 1 Thelephora pseudoversatilis.. AAA > Clavulina sp. 2 Clavulina sp. 3 Clavulina sp. 4 Inocybe sp. 2 Russulaceae sp. Tomentella sp. 3 Tomentella sp. 4 Tomentella sp. 5 Fig. 3 Relationships between ectomycorrhizal fungi and plant hosts in the tropical dry forest of Chamela. The size of the square height represents the frequency of each species, the lines represent the interaction between The diversity of the ECM fungi at the local scale is related to the identity of the host plants and their density (Ishida et al. 2007; Tedersoo et al. 2010, 2012). The TDF on the Pacific coast of Mexico is considered a site with a distinctive and hyperdiverse vegetation (Lott and Atkinson 2002). Despite the fact that Chamela and nearby forests are among the driest known TDF habitats (Bullock 1986), the plant diversity is nonetheless similar to many tropical rainforests (Gentry 1988; Janzen 1988). The Mexican TDF is particularly rich in endemic species (Trejo and Dirzo 2002) and it has been estimated that 11% of the endemic species in Mexico are distributed in this ecosystem (Rzedowski 1991). Assuming that the TDFs have high plant endemism and that the high seasonality and drought stress are powerful physiological fil- ters, 1t is likely that the ECM fungi from TDFs have adapta- tions to the extreme conditions and that some fungi might be endemics as well. To test this hypothesis, a wider sampling in the TDFs of the Neotropical Pacific coast will be necessary. Hosts Achatocarpus gracilis Guapira petenensis Nyctaginaceae sp. 1 Pisonieae sp. Ruprechtia sp. Pisonia sp. Aralia excelsa Apoplanesia paniculata Apocynaceae sp. Caesalpinioideae sp. Coccoloba sp. Ficus sp. Lonchocarpus sp. Nyctaginaceae sp. 2 Nyctaginaceae sp. 3 Nyctaginaceae sp. 4 Nyctaginaceae sp. 5 Papilionideae sp. Recchia mexicana Sapotaceae sp. plant and fungi and the width of the line is the frequency of that symbiosis. All the hosts in bold belong to Caryophyllales The hyperdiverse TDF in Chamela is comparable in diver- sity and abundance of ECM hosts with those in the Ecuadorian Amazon, white-sand forests in Brazil, and French Guiana where the ECM hosts are dispersed across the forest (Tedersoo et al. 2010; Roy et al. 2016). The low ECM fungal diversity in tropical forests with hyperdiverse host plants (Table S3) can be explained by the strong host filter, widely dispersed and relatively small host plants, and also perhaps the microbial mineralization during the short rainy season in the TDFs (Anaya et al. 2007). Unusual morphotypes in the tropical dry forest We report here the first evidence of a common occurrence of ECM fungi on the roots of Achatocarpus gracilis. The roots of Achatocarpus formed typical ECM morphotypes with several hyphal layers forming a distinct mantle. All of the fungi that we sequenced from these ECM roots are typical ECM A Springer 62 Mycorrhiza lineages, including members of the /sebacina, /clavulina, /tomentella-thelephora, /inocybe, and /russula-lactarius line- ages (Table 1). The combination of the phylogenetic place- ment of A. gracilis along with the common occurrence of fungal mantles and the regular occurrence of a diverse array of ECM fungi suggests that this shrub species forms an ECM symbiosis. However, despite our morphological analysis, we did not find evidence of a Hartig net. This species can there- fore be assigned to the “Pisonioid” mycorrhiza type, which has a mantle and an underdeveloped or nonexistent Hartig net (Ashford and Allaway 1982; Imhof 2009). The Pisonioid type has been documented on the roots of Pisonia grandis, also an ECM-forming member of Caryophyllales. Guapira morphotypes from the /clavulina lineage present- ed typical characters of an ectomycorrhiza: mantle, Hartig net, extraradical mycelia, but they also have intraradical hyphae. The same observation has been made previously for Guapira and Neea, where the ECM fungi formed intraradical hyphae (Haug et al. 2005). These features suggest that Guapira may form a kind of ectendomycorrhiza instead of the typical ectomycorrhiza. The /tomentella-thelephora lineage ectomycorrhizae had a mantle zoned by discontinuous colonization along the roots with all of their hosts. This morphotype formed short explora- tion to medium-distance fringe exploration morphotypes. This was also reported by Haug et al. (2005) in the /tomentella- thelephora—Guapira ectomycorrhizas in Ecuador. Specific plant ectomycorrhizal fungal interactions in the tropical dry forest Until now, all the ECM fungal communities are characterized by a small number of plant host lineages and a large number of ECM fungal lineages. For instance, a mixed temperate forest in Japan showed that a large number of ECM fungal lineages were found on eight hosts from three different plant families: Betulaceae, Pinaceae, and Fagaceae (Ishida et al. 2007). The Fig. 4 Pattern between ectomycorrhizal fungal richness, plant richness, and ectomycorrhizal host density in different ecosystems. In the left side, there are some examples of studies about plant-ECM fungal diversity. Data from Ebenye et al. 2016; Argúelles-Moyao et al. 2017; Smith et al. 2011; Pólme et al. 2013; Ishida et al. 2007; Peay et al. 2010; Diédhiou et al. 2010; Tedersoo et al. 2010 This study Ecuadorian tropical rainforest (8) Guinean tropical rainforest (7) Dipterocarpaceae (6) Mixed temperate forest (5) Alnus (4) Dicymbe and Aldina (3) Abies (2) Gilbertiodendron (1) 3. Smith et al. 2011 4. Polme et al. 2013 5. Ishida et al, 2007 6. Peay et al. 2010 7. Diédhiou et al. 2010 8. Tedersoo et al. 2010 Y Springer 1. Ebenye et al, 2016 2. Argúelles-Moyao et al. 2017 TDF is a hyperdiverse forest where we found 20 hosts from 10 different families (1.e., Achatocarpaceae, Apocynaceae, Araliaceae, Fabaceae, Moraceae, Nytaginaceae, Polygonaceae, Sapotaceae, and Surianaceae). None of these plants form monodominant patches and their fungal symbionts showed strong host preferences. Although the ECM fungi from Chamela are associated with several host lineages, many of their hosts are still unknown. Some of the ECM fungi were found with low frequency on the roots of Aralia excelsa, Apoplanesia paniculata, Apocynaceae sp., Caesalpinioideae sp., Coccoloba sp., Ficus sp., Lonchocarpus sp., Papilionoideae sp., Recchia mexicana, and Sapotaceae sp. It is probable that these associa- tions are opportunistic and could be compared to Singer's con- cept of “cicatrizing mycorrhiza” (Singer 1988). In tropical forests where the ECM hosts are integrated among a matrix of arbuscular mycorrhizal trees, there is an inverse pattern between ECM fungal diversity and the plant diversity (e.g., Tedersoo et al. 2010). The monodominant for- ests, including temperate zones, have a high ECM fungal di- versity. This 1s in contrast with the hyperdiverse tropical for- ests where the ECM fungal diversity is lower (Fig. 4). In the TDF of Chamela, 80% of ECM fungal species are specialists (associated with one host lineage) but we did find four gener- alists (associated with more than one host lineage). For exam- ple, 7h. versatilis and Tomentella sp. 1 were able to interact with five and six plant lineages, respectively. Specifically, 7h. versatilis is associated with two endemic species, Recchia mexicana and Ruprechtia sp. These results confirmed the pat- tern that in sites with ECM hosts at low frequency, there is more specialization in ECM fungal species (Tedersoo et al. 2008, 2010). The host preferences in ECM fungal associations are de- fined by host selection and also by environmental factors (Dickie 2007); both factors together determine the mutualistic niche (Peay 2016). In an environment dominated by hyperdiverse non-ECM plants, ECM fungi face three chal- lenges: (1) to find their dispersed hosts, they must either have Tropical hyperdiverse forest Temperate 8 Tropical forests with monodominance A ECM fungal richness £% Plant richness A ECM host density + : 63 Mycorrhiza long-lived spores or have excellent wind or vegetative dispers- al; (2) to survive in such a restricted niche, they have to be highly competitive with saprotrophic, pathogenic, and other ECM fungl; (3) to widen their mutualistic niche by developing new opportunistic symbiosis or to specialize on only one plant lineage. The hyperdiverse TDFs represent a case of strong environmental filtering created by seasonality, intermittent drought stress, and pulses of high decomposition that result in edaphic heterogeneity (Campo et al. 1998, 2000), especially N mineralization (Anaya et al. 2007; Waring et al. 2016). All these represent biotic and abiotic filters that shape the ECM fungal community and also its specific interactions with plant hosts. In conclusion, in the hyperdiverse Neotropical dry forest the Caryophyllales is the most important ECM plant host lin- eage with four independent origins of this symbiosis. These forests have a unique set of specialized ECM fungi. In order to understand the ecological importance of these unique ECM interactions, it will be necessary to focus on the adaptive ad- vantages that the ECM symbiosis contfers to plant hosts. Acknowledgments This study was funded by CONACYT Ciencia Básica 239266 and PAPIIT IN223114. The MEXBOL network supported DNA sequencing through the CONACYT 1251085 grant. We thank to Posgrado en Ciencias Biológicas from Universidad Nacional Autónoma de México. We acknowledge the Academic Writing Team of the Centro de Estudios de Posgrado, UNAM, for their help with this manuscript. MES received support from the US National Science Foundation grant DEB 1354802. We thank Leho Tedersoo and Jeremy Hayward for their assistance in the methodology for the plant identification. We are also thankful to Mohammad Bahram for his helpful comments, and two anon- ymous reviewers that improved the article. We thank Estela Sandoval- Zapotitla for her help at the anatomical slices of ectomycorrhizae. We thank to the Biological Station of Chamela and its entire staff. Author contributions J.A.M. and R.G.O. were responsible for the ex- perimental design. J.A.M. made the field work, laboratory proceedings, and the data analysis. J.A.M., R.G.O., and M.E.S. wrote the manuscript. References Agerer R, Rambold G (2004-2017) DEEMY—an information system for characterization and determination of ectomycorrhizac. Miúnchen, Germany URL wwwdeemyde. Accessed 18 Mar 2017 Alexander I (2006) Ectomycorrhizas—out of Africa? New Phytol 172: 589-591 Alvarez-Añorve MY, Quesada M, Sanchez-Azofeifa GA et al (012) Functional regeneration and spectral reflectance of trees during suc- cession in a highly diverse tropical dry forest ecosystem. Am J Bot 99:816-826. https://doi.org/10.3732/ajb.1100200 Alvarez-Manjarrez J, Villegas-Ríos M, Garibay-Orijel R et al (2016) Tomentella brunneoincrustata, the first described species of the Pisonieae-associated Neotropical Tomentella clade, and phylogenet- ic analysis ofthe genus in Mexico. Mycol Prog 15:1-—11. https://doi. org/10.1007/511557-015-1152-x Anaya CA, García-Oliva F, Jaramillo VJ (2007) Rainfall and labile car- bon availability control litter nitrogen dynamics in a tropical dry forest, Oecologia 150:602—610. https://do1.org/10.1007/500442- 006-0564-3 Ángeles-Argáiz RE, Flores-García A, Ulloa M, Garibay-Orijel R (2016) Commercial Sphagnum peat moss is a vector for exotic ectomycorrhizal mushrooms. Biol Invasions 18:89-101, https:// doi.org/10.1007/s10530-015-0992-2 Argúelles-Moyao A, Garibay-Orijel R, Márquez-Valdelamar LM, Arellano-Torres E (2017) Clavulina-Membranomyces is the most important lineage within the highly diverse ectomycorrhizal fungal community of Abies religiosa. Mycorrhiza 27:53-67. https://doi. org/10.1007/500572-016-0724-1 Ashford A, Allaway W (1982) A sheathing mycorrhiza on Pisonia grandis R. Br. (Nyctaginaceae) with development of transfer cells rather than a Hartig net. New Phytol 90:511-519. https://doi.org/10. 1111/.1469-8137.1982.tb04483.x Bá AM, Duponnois R, Moyersoen B, Diédhiou AG (2012) Ectomycorrhizal symbiosis of tropical African trees. Mycorrhiza 22:1-29, https://doi.org/10.1007/500572-011-0415-x Bahram M, Kóljalg U, Courty P-E et al (2013) The distance decay of similarity in communities of ectomycorrhizal fungi in different eco- systems and scales. J Ecol 101:1335-1344. https://do1.org/10.1111/ 1365-2745.12120 Bandala VM, Montoya L, Villegas R (2011) Tremelloscypha gelatinosa (Sebacinales) from tropical deciduous Gymnopodium forests in south- em Mexico. Mycotaxon 118:147—157. https://doi.org/10.5248/118.147 Bandou E, Lebailly F, Muller F, Dulormne M (2006) The ectomycorrhizal fungus Seleroderma bermudense alleviates salt stress in seagrape (Coccoloba uvifera L.) seedlings. Mycorrhiza 16:559-565. https://doi.org/10.1007/500572-006-0073-6 Brundrett MC (2009) Mycorrhizal associations and other means of nutri- tion of vascular plants: understanding the global diversity of host plants by resolving conflicting information and developing reliable means of diagnosis. Plant Soil 320:37-—77. https://doi.org/10.1007/ $11104-008-9877-9 Bullock SH (1986) Climate of Chamela, Jalisco, and trends in the south coastal region of Mexico. Arch Meteor Geophy B 36:297-316. https://doi.org/10.1007/BF02263135 Burke JM, Sanchez A (2011) Revised subfamily classification for Polygonaceae, with a tribal classification for Eriogonoideae. Brittonia 63:510-—520. https://doi.org/10.1007/s12228-011-9197-x Campo J, Jaramillo VJ, Maass JM (1998) Pulses of soil phosphorus avail- ability in a tropical dry forest: effects of seasonally and level of wetting. Oecologia 115:167-172. https://doi.org/10.1007/5004420050504 Campo J, Maass JM, Jaramillo VJ, Martínez Yrizar A (2000) Calcium, potassium, and magnesium cycling in a Mexican tropical dry forest ecosystem. Biogeochemistry 49:21-36. https://doi.org/10.1023/A: 10062073 19622 Chambers SM, Hitchcock CJ, Cairney JWG (2005) Ectomycorrhizal mycobionts of Pisonia grandis on coral cays in the Capricorn- bunker group, great barrier reef, Australia. Mycol Res 109:1105— 1111. https://doi.org/10.1017/50953756205003576 Corrales A, Arnold AE, Ferrer A et al (2016) Variation in ectomycorrhizal fungal communities associated with Oreomunnea mexicana (Juelandaceae) in a Neotropical montane forest, Mycorrhiza 26:1— 17. https://doi.org/10.1007/500572-015-0641-8 Crawley SS, Hilu KW (2012) Impact of missing data, gene choice, and taxon sampling on phylogenetic reconstruction: the Caryophyllales (angiosperms). Plant Syst Evol 298:297-312. https://doi.org/10. 1007/500606-011-0544-x Darriba D, Taboada GL, Doallo R, Posada D (2012) ¡ModelTest 2: more models, new heuristics and parallel computing. Nat Methods 9:772 772. https://do1.org/10.1038/nmeth.2109 Dell B, Sanmee R, Lumyong P, Lumyong S (2005) Ectomycorrhizal fungi in dry and wet dipterocarp forests in northern Thailand —di- versity and use as food. Proceeding of the 8 th round-table confer- ence on dipterocarps. Ho chi Minh City, Vietnam: 1-11 A Springer 64 Mycorrhiza Dickie IA (2007) Host preference, niches and fungal diversity. New Phytol 174:230-233. https://doi.org/10.1111/j.1469-8137.2007.02055.x Diédhiou AG, Selosse M-A, Galiana A et al (2010) Multi-host ectomycorrhizal fungi are predominant in a Guinean tropical rainforest and shared between canopy trees and seedlings. Environ Microbiol 12: 2219-2232. https://doi.org/10.1111/;.1462-2920.2010.02183.x DRYFLOR (2016) Plant diversity patterns of the Neotropical dry forest and their conservation implicatons. Science 353:1383-—1387. https:// doi.org/10.1126/science.aaf5080 Ducousso M, Ramanankierana H, Duponnois R et al (2008) Mycorrhizal status of native trees and shrubs from castern Madagascar littoral forests with special emphasis on one new ectomycorrhizal endemic family, the Asteropeiaceac. New Phytol 178:233-238. https://doi. org/10.1111/,.1469-8137.2008.02389.x Durán E, Balvanera P, Lott E et al (2002) Estructura, composición y dinámica de la vegetación. In: Noguera FA, Vega Rivera JH, García Aldrete AN, Quesada Avendaño M (eds) Historia Natural de Chamela. Instituto de Biología, UNAM, México Ebenye HCM, Taudicre A, Niang N et al 016) Ectomycorrhizal fungi are shared between seedlings and adults in a monodominant Gilbertiodendron dewevrei rain forest in Cameroon. Biotropica: 1— 12. https://doi.org/10.1111/btp.12415 Fukami T, Nakajima M, Fortunel C et al (2017) Geographical variation in community divergence: insights from tropical forest monodominance by ectomycorrhizal trees. Am Nat 190:S105— S122. https://doi.org/10.1086/692439 García-Oliva F, Maass JM, Galicia L (1995) Rainstorm analysis and rainfall of a seasonal region with strong cyclonic influence on the Pacific Coast of Mexico. J) Appl Meteorol 34:2491-2498. https:// doi.org/10.1175/1520-0450(1995)034<2491:RAAREO>2.0.C0;2 Garibay Orijel R, Morales-Marañon E, Domínguez-Gutiérrez M, Flores- García A (2013) Caracterización morfológica y genética de las ectomicorrizas formadas entre Pinus montezumae y los hongos presentes en los bancos de esporas en la Faja Volcánica Transmexicana. Rev Mex Biodivers 84:153-169. https://doi.org/ 10.7550/rmb.29839 Gentry AH (1988) Changes in plant community diversity and floristic composition on environmental and geographical gradients. Ann Mo Bot Gard 75:1-34 Haug I, Weiss M, Homeier J et al (2005) Russulaceae and Thelephoraceae form ectomycorrhizas with members of the Nyctaginaceae (Caryophyllales) in the tropical mountain rain forest of southern Ecuador. New Phytol 165:923-936. https://doi.org/10. 1111/.1469-8137.2004.01284.x Hayward JA, Horton TR (2012) Edaphic factors do not govern the ectomycorrhizal specificity of Pisonia grandis (Nyctaginaceac). Mycorrhiza 22:647-6532. https://doi.org/10.1007/500572-012-0442-2 van der Heijden MG, Martin FM, Sanders IR (2015) Mycorrhizal ecology and evolution: the past, the present, and the future. New Phytol 205: 1406-1423. https://doi.org/10.1111/nph.13288 Imhof S (2009) Arbuscular, ecto-related, orchid mycorrhizas—three in- dependent structural lineages towards mycoheterotrophy: implica- tions for classification? Mycorrhiza 19:357-363. https://doi.org/1 0. 1007/500572-009-0240-7 Ishida TA, Nara K, Hogetsu T (2007) Host effects on ectomycorrhizal fungal communities: insight from eight host species in mixed conifer-broadleaf forests. New Phytol 174:430-440. https://doi. org/10.1111/;.1469-8137.2007.02016.x Janzen D (1988) Tropical dry forest. The most endangered major tropical ecosystems. In: Wilson EO (ed) Biodiversity. Nactional Academy Press, Washington, DC, pp 130-137 Kennedy PG, Matheny PB, Ryberg KM et al (2012) Scaling up: exam- ining the macroecology of ectomycorrhizal fungi. Mol Ecol 21: 4151-4154. https://doi.org/10.1111/1.1365-294X.2012.05703.x Kornerup A, Wanscher JH (1978) Methuen handbook of colour, 3rd edn. Eyre Methuen, London A Springer Kress WJ, Erickson DL (2007) A two-locus global DNA barcode for land plants: the coding rbcL gene complements the non-coding trnH- psbA spacer region. PLoS One 2:e508. https://doi.org/10.1371/ journal.pone.0000508 Linares-Palomino R, Oliveira-Filho AT, Pennington TR (2006) Neotropical seasonally dry forests: diversity, endemism, and bioge- ography of woody plants. In: Dirzo R, Young HS, Mooney HA, Ceballos G (eds) Seasonally dry tropical forests: ecology and con- servation. Island Press, Washington, D.C. Looney BP, Ryberg M, Hampe F et al (2016) Into and out of the tropics: global diversification patterns in a hyperdiverse clade of ectomycorrhizal fungi. Mol Ecol 25:630-647. https://doi.org/10. 1111/mec.13506 Lott EJ (1993) Annotated checklist of the vascular flora of the Chamela Bay region. Occas Pap Calif Acad Sci 148:1-60 Lott EJ, Atkinson TH (2002) Biodiversidad y fitogeografía de Chamela- Cuixmala, Jalisco. In: Noguera FA, Vega Rivera JH, García Aldrete AN, Quesada Avendaño M (eds) Historia Natural de Chamela. Instituto de Biología, UNAM, México Magallón S, Castillo A (2009) Angiosperm diversification through time. Am J Bot 96:349-365. https://doi.org/10.3732/ajb.0800060 Maherali H, Oberle B, Stevens PF et al (2016) Mutualism persistence and abandonment during the evolution of the mycorrhizal symbiosis. Am Nat 188:E113-—E125. https://doi.org/10.1086/688675 Martínez-Yrízar A, Sarukhán J (1993) Cambios estacionales del mantillo en el suelo de un bosque tropical caducifolio y subcaducifolio en Chamela, Jalisco, México. Acta Bot Mex 21:1-6 Matheny PB, Aime MC, Bougher NL et al (2009) Out of the Palacotropics? Historical biogeography and diversification of the cosmopolitan ectomycorrhizal mushroom family Inocybaceae. J Biogeogr 36:577-592. https://doi.org/10.1111/.1365-2699.2008. 02055.x McGuire KL, Henkel TW, Granzow De La Cerda I et al (2008) Dual mycorrhizal colonization of forest-dominating tropical trees and the mycorrhizal status of non-dominant tree and liana species. Mycorrhiza 18:217-—222. https://doi.org/10.1007/500572-008- 0170-9 Miles L, Newton AC, DeFries RS et al (2006) A global overview of the conservation status of tropical dry forests. J Biogeogr 33:491-505. https://doi.org/10.1111/3.1365-2699.2005.01424.x Peay KG (2016) The mutualistic niche: mycorrhizal symbiosis and com- munity dynamics. Annu Rev Ecol Evol Syst 47:143-164. https:// doi.org/10.1146/annurev-ecolsys-121415-032100 Peay K, Kennedy P, Davies S et al (2010) Potential link between plant and fungal distributions in a dipterocarp rainforest: community and phy- logenetic structure of tropical ectomycorrhizal fungi across a plant and soil ecotone. New Phytol 185:529-542. https://doi.org/10.1111/ 3.1469-8137.2009.03075.x Pendry CA (2004) Monograph of Ruprechtia (Polygonaceae). Syst Bot Monogr 67:1-113. https://doi.org/10.2307/25027911 Pólme S, Bahram M, Yamanaka T et al (2013) Biogeography of ectomycorrhizal fungi associated with alders (A/nus spp.) in relation to biotic and abiotic variables at the global scale. New Phytol 198: 1239-1249. https://doi.org/10.1111/mph.12170 Portillo-Quintero CA, Sánchez-Azofeifa GA (2010) Extent and conser- vation of tropical dry forests in the Americas. Biol Conserv 143: 144-155. https://doi.org/10.1016/¡.biocon.2009.09.020 Ramirez-Lopez I, Villegas-Rios M, Salas-Lizana R et al (2015) Thelephora versatilis and Thelephora pseudoversatilis: two new eryptic species with polymorphic basidiomes inhabiting tropical de- ciduous and sub-perennial forests of the Mexican Pacific coast. Mycologia 107:346-358. https://doi.org/10.3852/14-151 Ronquist F, Huelsenbeck JP (2003). MRBAYES 3: Bayesian phyloge- netic inference under mixed models. Bioinformatics 19: 1572-1574. https://doi.org/10.1093/bioinformatics/btg 180 65 66 67 CAPÍTULO 2. SOIL FUNGAL COMMUNITY WAS PERSIS-1 TENT AND RESILIENT TO PATRICIA HURRICANE 2 3 Julieta Alvarez-Manjarrez1,2 , Roberto Garibay-Orijel1* 4 1. Instituto de Biología, Universidad Nacional Autónoma de México. Tercer circuito 5 interior s/n, Ciudad Universitaria, Coyoacán, 04510, Mexico City, Mexico. 6 2. Posgrado en Ciencias Biológicas. Edificio B, 1º Piso. Circuito de Posgrados, 7 Ciudad Universitaria. Coyoacán, 04510, Mexico City, Mexico. 8 *correspondence author: rgaribay@ib.unam.mx 9 Resumen 10 El cambio climático producirá eventos catastróficos cada vez más comunes, en 11 particular los huracanes. Los huracanes incrementan la precipitación y el mantillo 12 en consecuencia de la destrucción de la vegetación. Estas condiciones son idea-13 les para los organismos saprótrofos, tales como hongos y bacterias, mas no para 14 los simbiontes obligados. Nuestro objetivo fue entender cómo el huracán Patricia 15 (categoría 4) cambió la diversidad y el ensamble de la comunidad de hongos del 16 suelo en un bosque tropical caducifolio. Las secuencias de ITS2 de DNA de hon-17 gos del suelo en una serie de tiempo de dos años (un año antes del huracán y 18 cuatro de después) mostraron que antes de Patricia había 519 OTUs, donde As-19 comycota era el phylum más diverso. Inmediatamente después del huracán, la 20 riqueza de incrementó a 1358 OTUs, los basidiomicetos saprótrofos fueron los 21 más abundantes, y la dominancia de la comunidad decayó. Después de uno y dos 22 años del paso de Patricia, la diversidad decreció y la dominancia se recuperó gra-23 dualmente a través del tiempo. La diversidad de los hongos micorrízicos fue más 24 alta después de Patricia y así permaneció durante dos años. Los hongos patóge-25 nos de plantas y parásitos incrementaron inmediatamente después de Patricia y 26 con el tiempo decrecieron incluso más que antes del huracán. La comunidad per-27 sistente de todos los muestreos se conformó por 105 OTUs, principalmente sapro-28 68 trofos. La heterogeneidad ambiental creada por el huracán explica el incremento 29 de la diversidad, y la pérdida de la dominancia; las condiciones estresantes pudie-30 ron haber filtrado la diversidad a través del tiempo. La comunidad de hongos del 31 suelo casi se recuperó después de dos años al estado previo al huracán; por lo 32 que la comunidad de hongos del suelo fue resiliente al huracán Patricia. La resi-33 liencia fúngica directamente aporta a la recuperación de las propiedades ecosis-34 témicas, tales como la descomposición de la materia orgánica que impacta en la 35 disponibilidad de nutrientes para la vegetación, la pedogénesis o la estabilidad de 36 agregados del suelo que tienen un efecto en la erosión del suelo. 37 Abstract 38 Climate change will make catastrophic events more common, in particular hurri-39 canes. Hurricanes increase rainfall, and litterfall through vegetation destruction. 40 These are ideal conditions for saprotrophs organisms –such as fungi and bacteria– 41 but inadeccuate for obligate symbionts. Our aim was to understand how hurricane 42 Patricia (category 4) changed the diversity and community assembly of soil fungi in 43 a tropical dry forest. ITS2 rDNA sequences from faungal soil of a two year-time 44 series (one sampling before hurricane and four samplings after it) showed that be-45 fore Patricia were 519 OTUs; Ascomycota was the most diverse phylum. Immedi-46 ately after the hurricane, fungal richness increased to 1358 OTUs, Basidiomycota 47 saprotrophs were more abundant; the community dominance decayed. After Patri-48 cia, fungal diversity decreased and dominance recovered gradually through time. 49 Mycorrhizal fungi diversity was higher after Patricia and kept high after two years. 50 Plant pathogens and parasites increased immediately after Patricia and decreased 51 through time even more than before the hurricane. We found 105 OTUs, mainly 52 saprobes, formed the persistent community during all the study. The diversity in-53 creased because hurricane created environmental heterogeneity, and the loss of 54 dominance match with intermediate-disturbance hypothesis; these new conditions 55 considered as harsh, could filter diversity through time. Soil fungal community al-56 most recovered the previous hurricane state two years after the hurricane; hence 57 soil fungal community was resilient to Patricia hurricane. Probably fungal resilience 58 69 directly helped to recover ecosystemic properties, such as decomposition of organ-59 ic matter. 60 Keywords 61 Chamela, disturbance, high dominance, fungi, recovery, resistant community, trop-62 ical dry forest 63 Introduction 64 Climate change produces catastrophic events in short-intervals of time, such as 65 hurricanes, even if they were considered infrequent events (Buma 2015). Cata-66 strophic events, along with anthropogenic disturbance, affect the environmental 67 conditions and create complexity in ecosystems (e.g. Hodgson et al., 2015; Riutta 68 et al., 2018; Martínez-Yrízar et al., 2018). Disturbance tests the survival capacity to 69 new conditions of each species, produces succession and loss of species in com-70 munities. After disturbance, some species resist and they could help to return 71 quickly to the state prior the disturbance, giving resistance to the community (de 72 Vries & Shade, 2013; Oliver et al., 2015). Whether the biological diversity is high it 73 produces ecological redundancy that mitigates the disturbance and this in turn 74 generates resilience (Standish et al., 2014; Mori 2016). Climate change will affect 75 biological communities (Moritz & Agudo, 2013); however, the question remains as 76 how species will change after the increasing disturbance. 77 Microbial communities, particularly fungi and bacteria, are the main soil decom-78 posers (Tedersoo et al., 2014; Bahram et al., 2018). Their enzymes break the re-79 calcitrant elements from the litter and soil, changing the availability of nutrients 80 (Berg 2000). Also, soil available nutrients or organic matter changes the microbial 81 communities through time (Grandy et al., 2009; Purahong et al., 2016). Their par-82 ticipation involves mineralization additionally with immobilization of elements in 83 their own biomass (Hobbie 2015). Despite the great role of soil microorganisms, 84 we continue to ignore the effect of catastrophic events in their communities. 85 Tropical forests are niche of hiperdiversity from several groups, and fungal com-86 munities are well represented (Tedersoo et al., 2014). In America, tropical dry for-87 70 est (TDF) is distributed from Mexico to Argentina, and the Caribbean. Nowadays it 88 has less than 10% of its original extent and is highly threatened by human activity 89 (DRYFLOR 2016) and catastrophic events, such as hurricanes and fires. TDF goes 90 through 3-6 months with less than 100 mm of rainfall (DRYFLOR 2016). During dry 91 season, plants reabsorb nutrients from photosynthetic material and throw the 92 leaves to prevent water loss (Rentería et al., 2005), and litterfall increases (Anaya 93 et al., 2012). Water availability controls decomposition rate and mineralization 94 (Anaya et al., 2012). Hurricanes in TDF increase soil organic matter, nutrients, and 95 annual average precipitation (Gavito et al., 2018; Jaramillo et al., 2018). Hurricanes 96 bring more rain and decomposition rate could modify depending on the hurricane 97 (Gavito et al., 2018). 98 High-throughput sequencing has helped to understand the complex response of 99 fungal communities’ assembly to disturbance (e.g. Lekberg et al., 2011; García de 100 León et al., 2018; Castillo et al., 2018). We used these methods to study the as-101 sembly of soil fungal communities before and after a hurricane. The studied hurri-102 cane was Patricia, category 4 Saffir-Simpson scale, made landfall in October 2015 103 in a tropical dry forest in the Mexican Pacific coast. Our aim was to understand 104 how soil fungal communities’ assembly was before and after hurricane in a time-105 lapse sampling. Our hypotheses were: 1) that due the increase of soil organic mat-106 ter nutrient enriched, mainly C, N and P, we would find more diversity of species 107 after hurricane, and an increase on the saprotroph guilds, mainly Basidiomycota –108 phylum with great enzymatic capacity– due the amount of lignin litter input; 2) the 109 increase of soil nutrients will affect considerably mycorrhizal fungal community. 110 Methodology 111 Study site 112 This study was conducted on the Biological Station of Chamela, Jalisco, Mexico 113 (19°30’ N, 105°03’ W) that has a warm subhumid weather (Aw0(x’)i) on summer 114 and warm dry (Bshw) in winter. This forest passes by a long dry period (8 months) 115 so 84% of mean annual precipitation (1007.9 mm; 2007-2017) happens between 116 July to October. Precipitation is variable depending on hurricane season in Sep-117 71 tember-October (e.g. in 2011 with Jova hurricane increased to 1215 mm). The 118 main vegetation type is tropical dry forest (TDF) and next to streams the vegetation 119 is tropical moist forest. 120 The major land relief that we sampled was summit surface with Regosol eutric soil 121 (Cotler et al., 2002) on granite (Martínez-Trinidad et al., 2008). This soil type along 122 with the land relief is the driest place of the forest, shallow, with low organic matter 123 accumulation, and without horizon B (Cotler et al., 2002). The organic phosphorus 124 is the main form of P in the TDF soil (Álvarez-Santiago 2002), however available 125 phosphorus is the most limited nutrient (Jaramillo & Sanford, 1995). 126 Sampling 127 We sampled two soil cores (10 x 5 cm diameter) in a 200 m transect, every 20 m in 128 both sides of transect; the 20 soil cores were pooled in a same sample. The soil 129 was dried with silica gel at environmental temperature, and then stored in a plastic 130 bag. The first sampling was done on the rainy season in November 2014 just be-131 fore Patricia hurricane. After the hurricane, we sampled on dry and rainy season of 132 two years: May and October 2016, April and September 2017. All the samplings 133 were in the same transect (19°30’19.6’’ N, 105°02’30.3’’ W). 134 Soil chemical analysis 135 Total N and P were obtained by Kjedahl method and quantified by molybdate yel-136 low colorimetric. Total C was determined by the Walkley-Black method. Electric 137 conductivity and pH were analyzed by a conductivity bridge and pHmeter, respec-138 tively. These analyses were done by "Laboratorio de Fertilidad de Suelos y Quími-139 ca" from Colegio de Posgraduados, Mexico. We calculated the C:N, C:P ratio per 140 sample. We plotted the results of soil characteristics with ggplot2 package in R 141 (Wickham et al., 2016). 142 Molecular biology and bioinformatics 143 We extracted total DNA from 2 g of the finest part of soil with PowerMax Soil DNA 144 isolation kit (MoBio; USA). After, ITS2 rDNA was amplified with the mix of fungal 145 primers reported in Tedersoo et al. (2014) coupled with NextEra adapters at 2.5 146 72 µM. We used the Taq Platinum Multiplex PCR Master Mix (Life technologies; USA) 147 protocol and made three multiplex PCR replicates per DNA extraction. The three 148 PCR products of each DNA sample were pooled and sequenced by Illumina Miseq 149 in "Instituto de Medicina Genómica”, Mexico. 150 We filtered sequences by quality with default values using vsearch (minovlen=15; 151 maxdiffs=99; minlength=150; maxee=1; truncqual=0; maxns=0). Sequences were 152 demultiplexed with MOTHUR v1.36.1 (Schloss et al., 2009). Chimera filtering was 153 done de novo using UNITE ITS2 v7.1 as a reference base filtering, cutting primers, 154 and removing primer artefacts by vsearch. ITS2 region was extracted with ITSx 155 v1.0.11 (Bengtsson-Palme et al., 2013). The OTU table was clustered using CD-156 HIT v4.6 (parameters: threshold similarity=0.97, min length=50, memory=400, min 157 size=2, length cutoff=0, threads=1, storing=0, algorithm=0). Taxonomy of the most 158 abundant read per cluster was assigned using BLAST+ v2.2.28 and UNITE v7.1 159 (Kõljalg et al. 2013). All these tools were performed in PipeCraft toolkit v1.0 160 (Anslan et al. 2017). We subtracted the number of sequences found in controls 161 from each sample (Nguyen et al., 2014). The fungal guilds of each species were 162 determined by FUNGuild (Nguyen et al., 2016). 163 Statistical analysis 164 We analyzed the diversity from all the soil samples determining richness and fre-165 quency. Matrix was rarefied; rarefaction curve was plotted by vegan package in R 166 (Oksanen et al., 2016). Then, the matrix was normalized by Hellinger transfor-167 mation with vegan package of R (Oksanen et al., 2016). All graphics were done in 168 ggplot2 package in R (Wickham et al., 2016). We compared diversity data per 169 each sampling with one-way ANOVA in PAST. We calculated a multiple rank-170 abundance curves to determine dominance with goeveg package in R. 171 Results 172 Soil characteristics 173 Soil analysis showed that electric conductivity (EC; Fig. S1a) increased during May 174 2016 (dry season) just after the hurricane, and in April 2017 decreased dramatical-175 73 ly. The same response was observed in total C (Fig. S1b) and again it increased 176 on September 2017 (rainy season). Organic matter increased after hurricane and it 177 was maintained high during the April 2017; in September 2017, organic matter de-178 creased (Fig. S1c). Besides, total N increased after hurricane and it was main-179 tained high during next seasons (Fig. S1d). After hurricane, soil pH was acid, with 180 the lowest pH in September 2017 (Fig. S1E). Total P had a peak on May 2016 af-181 ter hurricane and the next seasons decreased (Fig. S1f). Ratios C:N and C:P de-182 creased before hurricane, nonetheless they both reacted different on the next sea-183 sons: C:N was lower than before the hurricane; meanwhile C:P increased dramati-184 cally after hurricane having the maximum measure on April 2017 (Fig. S1g, h). 185 Soil fungal diversity 186 Rarefaction curves showed that immediately after the hurricane richness in-187 creased, and in the posterior samplings decreased (Figure 1a). When data was 188 normalized, Simpson diversity index (1-D) increased immediately after hurricane, 189 and in consecutive samplings diversity diminished; however, diversity was kept 190 higher than before hurricane (Figure 1b). Fungal diversity between samplings was 191 significantly different (F4,10225=274.5; P<0.001) but not between rainy and dry sea-192 sons. In November 2014, before hurricane three species (Acrocalymma vagum, 193 Calvatia fragilis and Tremellomycetes sp.) dominated with >500 sequences, and 194 community abundance changed after hurricane to became more evenly distributed. 195 Over time, the fungal community tended to recover dominance of few species (Fig-196 ure 1c). 197 Before Patricia hurricane, we sequenced 519 OTUs, and after Patricia fungal rich-198 ness increased to 1358 OTUs, in October 2016 (Table 1). This increment caused 199 that some taxonomic groups appeared, e.g. the phyla Calcarisporiellomycota and 200 Rozellomcyota (classification of Wiljayawardene et al., 2018) were not present be-201 fore hurricane (Fig. 2a-b). In general, Basidiomycota and Ascomycota were the 202 most diverse phyla of any of the seasons of sampling. However, Basidiomycota 203 abundance was higher immediately after Patricia (Figure 2b); Agaricomycetes 204 were the 27% of all Basidiomycota, and Agaricales order was the most diverse or-205 74 der in any sample. Most of Ascomycota belong to Sordariomycetes, Dothideomy-206 cetes, Eurotiomycetes, Leotiomycetes, Orbiliomycetes class, (Figure 2c-d). Pleo-207 sporales, Eurotiales, Xylariales, and Hypocreales were also diverse. In general, 208 several new orders appear after hurricane with low diversity, e.g. Archaeosporales, 209 Corticiales, Dothideales, Paraglomerales, and Diversisporales (Figure S2). 210 After hurricane 839 new OTUs appeared, and in consecutive samples the number 211 of exclusive OTUs in each sample decreased. For example, in May 2016 we found 212 425 OTUs not shared with another season, while in October 2016 we found 189 213 exclusive OTUs. All the species found in November 2014 were present in the pos-214 terior samples with different abundance (Figure 1d). Besides, we found that 105 215 OTUs that were present in all samples, thus they conform the resistant soil fungal 216 community. The resistant community is mostly integrated by fungi with unknown 217 taxonomical identity, so that unknown fungal guild (62.85%), and the rest were 218 saprotrophs (20%), plant pathogens (8.57%), and mycorrhizal fungi (5.71%) (Fig-219 ure S3). 220 221 Table 1. Richness and abundance of each sampling season. November 2014 is 222 the community before hurricane Patricia. 223 Sampling Phy Cla Ord Fam Gen OTUs Unk No. seq Nov 2014 6 17 40 76 101 519 233 17258 May 2016 7 29 68 147 204 1358 599 54576 Oct 2016 7 24 50 96 114 815 365 44424 Apr 2017 6 21 55 102 130 892 391 49608 Sep 2017 7 23 50 87 101 635 296 32959 Phy=Phylum, Cla=Class, Ord=Order, Fam=Family, Gen=Genus, Unk=Unkown, No. seq= 224 Number of sequences. 225 226 75 Before hurricane Patricia, arbuscular mycorrhizal fungi (AM) were rare and in the 227 posterior samples its diversity increased. Animal pathogens were less abundant 228 during May 2016 and more abundant in September 2017. Ectomycorrhizal fungi 229 had almost the same richness in each sample, however after Patricia their abun-230 dance increased, especially in dry seasons (April and May). Parasites had higher 231 richness before hurricane however their abundance was less compared with the 232 posterior seasons; plant pathogens and saprotrophs presented the same pattern. 233 However, in saprotrophs, we found an increase in abundance during dry seasons 234 (Figure 2e-f). 235 In the study site, the first five most abundant species through time were Euroti-236 omycetes sp. (OTU 2394), Membranomyces sp. (OTU 102; SH522097.07FU), 237 Latorua caligans (OTU 2724), Thelephoraceae sp. (OTU 516), Geastrum sp. (OTU 238 1901; SH025433.07FU); however, not all of them were present in all samplings, 239 for example, Membranomyces sp. was absent before Patricia (Figure S4). 240 76 241 Figure 1. A) Species rarefaction curves from each sampling, labels correspond to 242 sampling season. Vertical line belongs to sample before hurricane. B) Simpson 243 diversity index of each season of sampling. Red lines belong to sample before hur-244 ricane. C) Rank-abundance curves of each sample. D) Venn diagram show the 245 77 number of share species between seasons. Abbreviations: NOV_2014= November 246 2014, MAY_2016= May 2016, OCT_2016= October 2016, APR_2017= April 2017, 247 SEP_2017= September 2017. 248 249 250 78 Figure 2. a, b) Relative richness and relative abundance of phyla, c, d) class, and 251 e, f) guilds. Abbreviations: Sap= saprotrophs, Plant_path= plant pathogens, OM= 252 orchid mycorrhiza, Lich= lichenized, Endo= endophytes, EM= ericoid mycorrhiza, 253 ECM= ectomycorrhizal, Anim_path= animal pathogens, AM= arbuscular mycorrhi-254 za. 255 Discussion 256 Our results tracked the succession of the fungal community after hurricane Patricia 257 made landfall, and how soil fungi community was persistant and resilient in more 258 than two years. We observed the fungal succession through time, with the highest 259 diversity immediately after hurricane (1358 OTUs) and decreasing later. The high 260 dominance before Patricia decayed and recovered slowly, showing resilience. The 261 high diversity was a consequence of the appearance of 839 new OTUs after hurri-262 cane. Before Patricia hurricane, Ascomycota was the phylum more abundant, after 263 hurricane Basidiomycota saprotrophs was the guild with highest diversity, followed 264 by plant pathogens. 265 We highlight that before the disturbance there was high dominance of few species, 266 but immediately after Patricia it disappeared. The low dominance fostered high 267 fungal diversity after hurricane. This result is completely opposite to disturbed for-268 ests where some species become dominant over the community (Prieto et al., 269 2017). Disturbance can have a positive effect on richness, while undisturbed sites 270 limit the establishment of new species (Wohlgemuth et al., 2002; Farrior et al., 271 2016). Through time, dominance recovered in more than two years after hurricane. 272 Tropical dry forest had high turnover because on each sample we found exclusive 273 fungal species through time. We found no replacement from any of the initial spe-274 cies of 2014. The succession is non-directional replacement, i.e. harsh environ-275 ment is a strong filter for a successful establishment (Svoboda & Henry, 1987), and 276 diversity recovers slowly. However, fungi recover quicker compared to other organ-277 isms e.g. arbuscular mycorrhizal fungi have a quicker recovery than plant commu-278 nities (Mao et al., 2019). Despite fungi recovered slowly, it seems to be resilient. 279 Fungal resilience results from high fungal diversity which has ecological redundan-280 79 cy (Peršoh 2015). An evidence of fungal redundancy was decomposition rate in-281 creased after hurricane Patricia (Gavito et al., 2018). 282 Throughout two years, 105 fungal species persisted in all soil samples. We identi-283 fied them as the soil resistant community since these species persisted thru a hur-284 ricane category 4 and the seasonality of this ecosystem. These species are mainly 285 saprotrophs (e.g. Geastrum sp., Lepista sordida, Clitopilus sp., etc.), and given the 286 fluctuating environmental conditions, we hypothesize that they are highly plastic 287 with genetic adaptations that resist drought, high temperatures and nutrient fluctua-288 tion. These species had low sensitivity to disturbance because they are highly 289 adapted to the ecosystem, and they influenced in particular processes (Allison & 290 Martiny, 2008), such as decomposition. 291 Several ‘new’ species appeared in soil after hurricane. With all the vegetation 292 damage could be that endophytes that turn to saprotrophs (Promputtha et al., 293 2007) become abundant at soil. Also, water, wind, and animals, could be responsi-294 ble of the increase of richness after hurricane, they can transport spores through 295 long distances (e.g. Jumpponen 2003; Behzad et al., 2018; Correia et al., 2019). 296 Thus, hurricanes also can be agents of import fungal species as it happens with 297 plants (Bhattari & Cronin, 2014). After the peak of diversity brought by the hurri-298 cane, fungal diversity declined in each of the posterior samplings. Ecophysiological 299 characteristics of exotic species are tested when they arrive to a new environment 300 (Svoboda & Henry, 1987), and interspecific interactions, such as competition could 301 inhibit in the establishment of these (Koide et al., 2011). 302 Our results supported the hypothesis that predicted Basidiomycota saprotrophs 303 would increase after hurricane. After hurricane, deposition of 17.8 Mg ha-1 litterfall 304 (Parker et al., 2018) induced organic matter increment and C:N ratio reduction. 305 Basidiomycota saprotrophs –such as Lepista sordida or Geastrum sp.– abundance 306 increased after Patricia. Litterfall physico-chemical properties change through de-307 composition, and so fungal communities: Basidiomycota prevails in the latest stage 308 (Purahong et al., 2016) because they high enzymatic capacity to decompose ligno-309 cellulosic compounds (Voříšková & Baldrian, 2013). 310 80 Nutrient increment affects directly mycorrhizal establishment (Verlinden et al., 311 2018). Phosphorus input was notably higher after Patricia, and after a year de-312 creased to be lesser than before hurricane. Nitrogen remained higher than before 313 Patricia. Plants dispense with mycorrhizal interaction when available nutrients are 314 high (Verlinden et al., 2018). Nutrient stress-gradient reinforce mycorrhizal symbio-315 sis but hurricane nutrient-disturbance shaped mycorrhizal communities differential-316 ly. However, after hurricane, arbuscular mycorrhizal (AM) and ectomycorrhizal fun-317 gi (ECM) increased their abundance on soil (e.g. Membranomyces sp. increased 318 its abundance after Patricia). After hurricane –when ratio C:P increased– AM re-319 mained abundant. AM are more specialized on phosphorus acquisition while ECM 320 rather obtain nitrogen (van der Heijden et al., 2015). ECM abundance increased 321 immediately after Patricia and decreased in posterior samplings. 322 On the other hand, plant pathogens and parasites increased its diversity immedi-323 ately after the hurricane, probably due to damaged suffered by plants. A successful 324 infection of a plant pathogen depends on a susceptible host and an ideal environ-325 ment (Garrett et al., 2009). Our monitoring of soil fungal diversity was done on a 326 hilltop, where Patricia predominantly removed the canopy (Parker et al., 2018) 327 damaging strongly the vegetation. Many plant pathogens produce billions of mito-328 spores that travel aerially (Tedersoo et al., 2014) and they could find an establish-329 ment opportunity in damaged vegetation. One year after Patricia and thereafter, 330 pathogens diversity decreased, and their abundance became lesser than before 331 hurricane. 332 Dry and rainy season maintained the same fungal diversity; however, forest sea-333 sonality can be responsible of the slow recovery. In tropical dry forest rains less 334 than 1800 mm per year for 6-8 months (DRYFLOR 2016). Decomposition rate in-335 creases with water availability in this ecosystem, and stops during dry season 336 (Anaya et al., 2012). Fungi require water to break lignin and cellulose complex 337 molecules. Fungal spores can linger dry conditions waiting for humid conditions, 338 especially those who have cellular melanin wall (Fernandez et al., 2016). Soil fun-339 gal spores may be inactive with basal metabolism when water is no available. We 340 81 could have same diversity in any season because our sequence technique did not 341 differentiate active and inactive fungi. 342 Hurricanes will increase in frequency and severity while ocean water gets warmer 343 (Knutson et al., 2015), causing disturbance in tropical and subtropical ecosystems 344 around the globe. Climatic change scenarios predict desertification of the tropical 345 forests (Salazar et al., 2007). Fungal species with high plasticity and strategies to 346 survive droughts will continue as part of the soil community. While the fungal com-347 munity continue high diverse, the ecological redundancy will be also higher and will 348 drive the ecosystem to resilience. Same as other organisms (e.g. Ameca y Juárez 349 et al., 2013; Hogan et al., 2017; Lloyd et al., 2019; Paz et al., 2018; Temeles & 350 Bishop, 2019), fungi can vary between damage till beneficial impact. Hurricanes 351 disturbance, drought, land-use change, etc. jeopardize soil diversity that could 352 generate ‘defungation’, i.e. loss of fungal diversity. Proper soil management can 353 cause survival of the soil diversity; in the face of such a threat is a light of hope in 354 the face of the predicted environmental catastrophe. Understanding how communi-355 ties are affected by disturbance gives us the opportunity to make good decisions 356 about the management of resources. 357 Conclusions 358 Basidiomycota saprotrophs were the main fungal guild after hurricane due the in-359 crease of soil organic matter, same as mycorrhizal fungi. Plant pathogens in-360 creased on soil after Patricia probably due vegetation damage. After two years the 361 pattern of high dominance tended to recover, showing resilience in more than two 362 years. Some fungal species form a resistant community to disturbance, showing 363 plasticity to harsh environment. The response of each community depends on the 364 magnitude and severity of the hurricane and the previous state of the site. 365 Acknowledgments 366 We thank CONACYT for funding the project Ciencia Básica 239266, for the PhD 367 scholarship 404413 and for beca mixta to make a research stance at Natural Histo-368 ry Museum of Tartu Ülikool. We thank Posgrado de Ciencias Biológicas from Uni-369 versidad Nacional Autónoma de México. We thank Mohammad Bahram and Saleh 370 82 Rahimlou for the bioinformatic help. We thank the field workers Abel Domínguez, 371 Diego Flores, Jorge Blanco, Rodolfo Ángeles, Lorenzo Vázquez and Sergio Vidal. 372 Thanks to Andrés Argüelles for helping with the molecular protocols. We thank to 373 all the staff of the Biological Station of Chamela. We also thank to anonymous re-374 viewers. 375 References 376 Allison S, Martiny J. 2008. Resistance, resilience, and redundancy in microbial 377 communities. Proceedings of the National Academy of Sciences 105: 378 11512–11519. 379 Alvarez-Manjarrez J, Bahram M, Polme S, Garibay-Orijel R. in press. Hurricane 380 disturbance affects the structure of rhizosphere fungal communities and 381 mycorrhizal network in a Neotropical forest. 382 Álvarez-Santiago SA. 2002. Efecto de la perturbación en la interacción 383 micorrízica vesículo-arbuscular en un ecosistema tropical estacional. Tesis 384 de Maestría, Facultad de Ciencias, UNAM, México. 385 Ameca y Juárez EI, Mace GM, Cowlishaw G, Cornforth WA, Pettorelli N. 2013. 386 Assessing exposure to extreme climatic events for terrestrial mamamals. 387 Conservation Letters 6: 145-153 388 Anaya CA, Jaramillo VJ, Martínez-Yrízar A, García-Oliva F. 2012. Large rainfall 389 pulses control litter decomposition in a tropical dry forest: evidence from an 390 8-year study. Ecosystems 15: 652-663. 391 Anslan S, Bahram M, Hiiesalu I, Tedersoo L. 2017. Pipecraft: flexible open-392 source toolkit for bioinformatics analysis of custom high throughput am-393 plicon sequencing data. Molecular Ecology 17: e234-e240. 394 Bahram M, Hildebrand F, Forslund SK, Anderson JL, Soudzilovskaia NA, 395 Bodegom PM, Bengtsson-Palme J, Anslan S, Coelho LP, Harend H, et al. 396 2018. Structure and function of the global topsoil microbiome. Nature 560: 397 233–237. 398 83 Bhattarai GP, Cronin JT. 2014. Hurricane activity and the large-scale pattern of 399 spread of an invasive plant species. PLoS ONE 9(5): 400 https://doi.org/10.1371/journal.pone.0098478 401 Behzad H, Mineta K, Gojobori T. 2018. Global ramifications of dust and 402 sandstorm microbiota. Genome Biology and Evolution 10: 1970–1987. 403 Bengtsson-Palme J, Veldre V, Ryberg M, Hartmann M, Branco S, Wang Z, Godhe 404 A, Bertrand Y, De Wit P, Sanchez M, Ebersberger I, Sanli K, de Souza F, 405 Kristiansson E, Abarenkov K, Eriksson KM, Nilsson RH. 2013. ITSx: 406 Improved software detection and extraction of ITS1 and ITS2 from 407 ribosomal ITS sequences of fungi and other eukaryotes for use in 408 environmental sequencing. Methods in Ecology and Evolution 4: 914-919. 409 Berg B. 2000. Litter decomposition and organic matter turnover in northern forest 410 soils. Forest Ecology and Management 133: 13–22. 411 Buma B. 2015. Disturbance interactions: Characterization, prediction, and the po-412 tential for cascading effects. Ecosphere 6: 1–15. 413 Cantrell SA, Molina M, Jean Lodge D, Rivera-Figueroa FJ, Ortiz-Hernández ML, 414 Marchetti AA, Cyterski MJ, Pérez-Jiménez JR. 2014. Effects of a simulated 415 hurricane disturbance on forest floor microbial communities. Forest Ecolo-416 gy and Management 332: 22–31. 417 Castillo BT, Nave LE, Le Moine JM, James TY, Nadelhoffer KJ. 2018. Impacts of 418 experimentally accelerated forest succession on belowground plant and 419 fungal communities. Soil Biology and Biochemistry 125: 44-53. 420 Correia M, Heleno R, da Silva LP, Costa JM, Rodríguez-Echeverría S. 2019. First 421 evidence for the joint dispersal of mycorrhizal fungi and plant diaspores by 422 birds. New Phytologist. 423 Cotler H, Durán E, Siebe C. 2002. Caracterización morfo-edafológica y calidad de 424 sitio de un bosque tropical caducifolio. En: Noguera FA, Vega Rivera JH, 425 García Aldrete AN, Quesada Avendaño M. Historia Natural de Chamela. 426 Instituto de Biología, UNAM. México. 427 84 De Vries FT, Shade A. 2013. Controls on soil microbial community stability under 428 climate change. Frontiers in Microbiology 4: 1–16. 429 DRYFLOR. 2016. Plant diversity patterns in Neotropical dry forests and their con-430 servation implications. Science 353: 1383-1387. 431 Farrior CE, Bohlman SA, Hubbell SP, Pacala SW. 2016. Dominance of the sup-432 pressed: Power-law size structure in tropical forests. Science 351: 155–433 157. 434 Fernandez CW, Langley JA, Chapman S, McCormack ML, Koide RT. 2016. The 435 decomposition of ectomycorrhizal fungal necromass. Soil Biology and Bio-436 chemistry 93: 38–49. 437 García de León D, Neuenkamp L, Moora M, Öpik M, Davison J, Peña-Venegas 438 CP, Vasar M, Jairus T, Zobel M. 2018. Arbuscular mycorrhizal fungal 439 communities in tropical rain forest are resilient to slash-and-burn agricul-440 ture. Journal of Tropical Ecology 34: 186-199. 441 Garrett KA, Nita M, De Wolf ED, Esker PD, Gomez-Montano L, Sparks AH. 2015. 442 Plant Pathogens as Indicators of Climate Change. In: Letcher T, ed. Cli-443 mate Change: Observed Impacts on Planet Earth: Second Edition. Elsevier 444 B.V., 325–338. 445 Gavito ME, Sandoval-Pérez AL, del Castillo K, Cohen-Salgado D, Colarte-Avilés 446 ME, Mora F, Santibáñez-Rentería A, Siddique I, Urquijo-Ramos C. 2018. 447 Resilience of soil nutrient availability and organic matter decomposition to 448 hurricane impact in a tropical dry forest ecosystem. Forest Ecology and 449 Management 426: 81-90. 450 Grandy AS, Strickland MS, Lauber CL, Bradfor MA, Fierer N. 2009. The influence 451 of microbial communities, managemenet and soil texture on soil organic 452 matter chemistry. Geoderma 150: 278-286. 453 Hobbie SE. 2015. Plant species effects on nutrient cycling: revisiting litter feed-454 backs. Trends in Ecology & Evolution 30: 357–363. 455 85 Hogan JA, Mayorquín S, Rice K, Thompson J, Zimmerman JK, Brokaw N. 2017. 456 Liana dynamics reflect land-use history and hurricane response in a Puerto 457 Rican forest. Journal of Tropical Ecology 33: 155-164. 458 Hodgson D, McDonald JL, Hosken DJ. 2015. What do you mean, ‘resilient’? 459 Trends in Ecology and Evolution 30: 503–506. 460 Jaramillo JV, Sanford RL Jr. 1995. Nutrient cycling in tropical deciduous forests. 461 In: Bullock SH, Mooney HA, Medina E. Seasonally dry tropical forests. 462 Cambridge University Press, Great Britain. 463 Jaramillo VJ, Martínez-Yrízar A, Maass M, Nava-Mendoza M, Castañeda-Gómez 464 L, Ahedo-Hernández R, Araiza S, Verduzco A. 2018. Hurricane impact on 465 biogeochemical processes in a tropical dry forest in western Mexico. For-466 est Ecology and Management 426: 72–80. 467 Jumpponen A. 2003. Soil fungal community assembly in a primary successional 468 glacier forefront ecosystem as inferred from rDNA sequence analyses. 469 New Phytologist 158: 569–578. 470 Knutson TR, Sirutis JJ, Zhao M, Tuleya RE, Bender M, Vecchi GA, Villarini G, 471 Chavas D. 2015. Global projections of intense tropical cyclone activity for 472 the late twenty-first century from dynamical downscaling of CMIP5/RCP4.5 473 Scenarios. American Meteorology Society 28: 7203–7224. 474 Koide RT, Fernandez C, Petprakob K. 2011. General principles in the community 475 ecology of ectomycorrhizal fungi. Annals of Forest Science 68: 45–55. 476 Kõljalg U, Nilsson RH, Abarenkov K, Tedersoo L, Taylor AFS, Bahram M, Bates 477 ST, Bruns T, Bengtsson-palme J, Callaghan TM, et al. 2013. Towards a 478 unified paradigm for sequence-based identification of fungi. Molecular 479 ecology 22: 5271–5277. 480 Lekberg Y, Schnoor T, Kjoller R, Gibbons SM, Hansen LH, Al-Soud WA, 481 Sorensen SJ, Rosendahl S. 2011. 454-sequencing reveals stochastic local 482 86 reassembly and high disturbance tolerance within arbuscular mycorrhizal 483 fungal communities. Journal of Ecology 100: 151-160. 484 Lloyd JD, Rimmer CC, Salguero Faría JA. 2019. Short-term effects of hurricanes 485 Maria and Irma on forest birds of Puerto Rico. bioRxiv 578336. DOI: 486 https://doi.org/10.1101/578336 487 Mao L, Pan J, Jiang S, Shi G, Qin M, Zhao Z, Zhang Q, An L, Feng H, Liu Y. 488 2019. Arbuscular mycorrhizal fungal community recovers faster than plant 489 community in historically disturbed Tibetan grasslands. Soil Biology and 490 Biochemistry 134: 131–141. 491 Martínez-Trinidad S, Cotler H, Etchevers-Barra JD, Ordaz-Chaparro VM, De 492 León-González F. 2008. Efecto del manejo en la agregación del suelo en 493 un ecosistema tropical seco. Terra Latinoamericana 26: 299–307. 494 Martínez-Yrízar A, Jaramillo VJ, Maass M, Búrquez A, Parker G, Álvarez-Yépiz 495 JC, Araiza S, Verduzco A, Sarukhán J. 2018. Resilience of tropical dry for-496 est productivity to two hurricanes of different intensity in western Mexico. 497 Forest Ecology and Management 426: 53–60. 498 Mori AS. 2016. Resilience in the Studies of Biodiversity-Ecosystem Functioning. 499 Trends in Ecology and Evolution 31: 87–89. 500 Moritz C, Agudo R. 2013. The future of species under climate change: Resilience 501 or decline? Science 341: 504–508. 502 Nicolás C, Martin-Bertelsen T, Floudas D, Bentzer J, Smits M, Johansson T, et al. 503 2018. The soil organic matter decomposition mechanisms in ectomycorrhi-504 zal fungi are tuned for liberating soil organic nitrogen. The ISME Journal. 505 https://doi.org/10.1038/s41396-018-0331-6 506 Nguyen NH, Smith S, Peay K, Kennedy P. 2014. Parsing ecological signal from 507 noise in next generation amplicon sequencing. New phytologist 205: 1389-508 1393. 509 87 Nguyen NH, Song Z, Bates ST, Branco S, Tedersoo L, Menke J, Schilling JS, 510 Kennedy PG, 2016. FUNGuild: an open annotation tool for parsing fungal 511 community datasets by ecological guild. Fungal Ecology 20: 241–248. 512 Oksanen J, Guillaume F, Friendly M, Kindt R, Legendre P, McGlinn D, Minchin P 513 et al., 2016. Community ecology package: vegan v2.4-1. 514 Oliver TH, Heard MS, Isaac NJB, Roy DB, Procter D, Eigenbrod F, Freckleton R, 515 Hector A, Orme CDL, Petchey OL, et al. 2015. Biodiversity and resilience 516 of ecosystem functions. Trends in ecology & evolution 30: 673–684. 517 Parker G, Martínez-Yrízar A, Álvarez-Yépiz JC, Maass M, Araiza S. 2018. Effects 518 of hurricane disturbance on a tropical dry forest canopy in western Mexico. 519 Forest Ecology and Management 426: 39–52. 520 Paz H, Vega-Ramos F, Arreola-Villa F. 2018. Understanding hurricane resistance 521 and resilience in tropical dry forest trees: a functional traits approach. For-522 est Ecology and Management 426: 115-122. 523 Peršoh D. 2015. Plant-associated fungal communities in the light of meta’omics. 524 Fungal Diversity 75: 1–25. 525 Prieto P V., Seger GDS, Sánchez-Tapia A, Sansevero JBB, Braga JMA, Ro-526 drigues PJFP. 2017. Secondary succession and fire disturbance promote 527 dominance of a late-diverging tree lineage in a lowland Neotropical forest. 528 Plant Ecology and Diversity 10: 311–322. 529 Promputtha I, Lumyong S, Dhanasekaran V, McKenzie EHC, Hyde KD, Jeewon 530 R. 2007. A phylogenetic evaluation of whether endophytes become sapro-531 trophs at host senescence. Microbial Ecology 53: 579–590. 532 Purahong W, Wubet T, Lentendu G, Schloter M, Pecyna MJ, Kapturska D, 533 Hofrichter M, Krüger D, Buscot F. 2016. Life in leaf litter: novel insights into 534 community dynamics of bacteria and fungi during litter decomposition. Mo-535 lecular Ecology: 4059–4074. 536 88 Rentería LY, Jaramillo VJ, Martínez-Yrízar A, Pérez-Jiménez A. 2005. Nitrogen 537 and phosphorus resorption in trees of a Mexican tropical dry forest. Trees 538 19: 431–441. 539 Riutta T, Malhi Y, Kho L, Marthews T, Huaraca Huasco W, Khoo M, Tan S, 540 Turner E, Reynolds G, Both S, et al. 2018. Logging disturbance shifts net 541 primary productivity and its allocation in Bornean tropical forests. Global 542 Change Biology 24: 2913–2928. 543 Salazar LF, Nobre CA, Oyama MD. 2007. Climate change consequences on the 544 biome distribution in tropical South America. Geophysical Research Letters 545 34: 1–6. 546 Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, Hollister EB, et al. 547 2009. Introducing MOTHUR: Open-source, platform-independent, commu-548 nity-supported software for describing and comparing microbial communi-549 ties. Applied Environmental Microbiology 75: 7537–41. 550 Standish RJ, Hobbs RJ, Mayfield MM, Bestelmeyer BT, Suding KN, Battaglia LL, 551 Eviner V, Hawkes CV, Temperton VM, Cramer VA, et al. 2014. Resilience 552 in ecology: Abstraction, distraction, or where the action is? Biological Con-553 servation 177: 43–51. 554 Svoboda J, Henry GHR. 1987. Succession in Marginal Arctic Environments. Arc-555 tic and Alpine Research 19: 373–384. 556 Tedersoo L, Bahram M, Polme S, Koljalg U, Yorou NS, Wijesundera R, et al. 557 2014. Global diversity and geography of soil fungi. Science 346: DOI: 558 10.1126/science.1256688 559 Temeles EJ, Bishop GA. 2019. A hurricane alters pollinator relationships and nat-560 ural selection on an introduced island plant. Biotropica 51: 129-138. 561 Vargas R, Hasselquist N, Allen EB, Allen MF. 2010. Effects of a hurricane dis-562 turbance on aboveground forest structure, arbuscular mycorrhizae and be-563 lowground carbon in a restored tropical forest. Ecosystems 13: 118–128. 564 89 Voříšková J, Baldrian P. 2013. Fungal community on decomposing leaf litter un-565 dergoes rapid successional changes. The ISME journal 7: 477–86. 566 Wickham H, Chang W, RStudio. 2016. An implementation of the grammar of 567 frpahics, package ggplot2. 568 Wijayawardene NN, Pawłowska J, Letcher PM, Kirk PM, Humber RA, Schüßler A, 569 Wrzosek M, Muszewska A, Okrasińska A, Istel Ł, et al. 2018. Notes for 570 genera: basal clades of Fungi (including Aphelidiomycota, Basidiobolomy-571 cota, Blastocladiomycota, Calcarisporiellomycota, Caulochytriomycota, 572 Chytridiomycota, Entomophthoromycota, Glomeromycota, Kickxellomyco-573 ta, Monoblepharomycota, Mortierellomyc. Fungal Diversity 92: 43–129. 574 Wohlgemuth T, Bürgi M, Scheidegger C, Schütz M. 2002. Dominance reduction 575 of species through disturbance — a proposed management principle for 576 central European forests. Forest Ecology and Management 166: 1–15. 577 90 Supplementary information 578 579 Figure S1. Soil nutrients per sampling, rainny season 2014 correspond to the sampling 580 before the Patricia hurricane. The black vertical line indicates Patricia hurricane landfall.581 91 582 583 Figure S2. Relative richness (left) and relative abundance (right) of Order in each sample.584 92 585 586 Figure S3. Abundance of fungi belonging to the persistent community. This graph contains 587 just 39 taxa with taxonomical assignment from 105 species that shapes the resistant 588 community. 589 590 0% 25% 50% 75% 100% Paraconiothyrium_fungicola Trichoderma_sp Spiromastix_warcupii Aspergillus_tamarii Calvatia_fragilis Rhizophlyctis_rosea Penicillium_sp Auxarthron_chlamydosporum Spiromastix_asexualis Clitopilus_sp Acrocalymma_vagum Penicillium_sp Cristinia_sp Acrocalymma_fici Cristinia_sp Aspergillus_tubingensis Talaromyces_sp Geastrum_sp Lepista_sordida Agaricaceae_sp Geastrum_sp Chaetomium_sp Lasiodiplodia_sp Lasiodiplodia_theobromae Aspergillus_aculeatus Plectosphaerella_alismatis Ganoderma_sp Fusarium_chlamydosporum Knufia_sp Didymosphaeria_variabile Tetragoniomyces_uliginosus Botryosphaeria_sp Trichoderma_gamsii Ceratobasidium_sp Chloridium_sp Entolomataceae_sp Glomeraceae_sp Glomeraceae_sp GS24_sp NOV_2014 MAY_2016 OCT_2016 APR_2017 SEP_2017 93 591 Figure S4. Abundance fluctuation through time (plot includes just the 53 more abundant 592 fungi).593 Animal pathogens Simplicillium_obclavatum Medicopsis_romeroi Trichosporon_sp Nigrograna_sp Beauveria_sp Trichosporon_asahii Metarhizium_marquandii Nigrograna_sp Metarhizium_anisopliae Aschersonia_sp Septobasidium_wilsonianum Beauveria_sp Pneumocystis_wakefieldiae MH NOV_2014 MH MAY_2016 E OCT_2016 mM APR_2017 mSEP_2017 ldriella_lunata Clitopilus_passeckerianus Agaricaceae_s5p Acrocalymma_fici Clitopilus_hobsonii Geastrum_sp Agaricaceae_sp Leucoagaricus_paraplesius Marasmius_sp Podoscypha_cristata Clitopilus_sp Agaricaceae_sp Chaetomella_raphigera Conocybe_sp Cristinia_sp Phellinidium_sulphurascens Lepiota_flammeotincta Aspergillus_tubingensis Leucocoprinus_sp Geastrum_sp Vararia_ochroleuca Geastrum_sp Clitopilus_hobsonii Lepiota_sp Cystodermella_cristallifera Agaricaceae_sp Clitopilus_sp Geastrum_sp Agaricacege_5p Leucocoprinus_sp Talaromyces_sp Lepiota_tomentella Geastrum_sp Lepiota_flammeotincta Clitopilus_brunnescens Geastrum_sp Lepiota_sp Lepiota_sp Lepista_sordida Agaricaceae_sp Geastrum_sp mNOvV_2014 MIMAY_2016 mOcT_2016 mAPR_2017 HSEP_2017 Glomerales_sp Glomeromycota_sp Glomeraceae_sp Glomeraceae_sp Glomeraceae_sp Glomeraceae_sp Glomeromycota_sp Glomeraceae_sp Glomeraceae_sp Glomeraceae_sp Glomeraceae_sp Glomeraceae_sp Glomeraceae_sp Glomeraceae_sp Glomeraceae_sp Glomeraceae_sp Glomeraceae_sp Glomeraceae_sp Glomeraceae_sp Glomeraceae_sp Glomeromycota_sp Glomus_indicum Glomeromycota_sp Glomeraceae_sp Glomeraceae_sp Glomeraceae_sp GS24_sp Glomeraceae_sp G524_sp Glomeromycota_sp Glomeraceae_sp G524_sp Glomeraceae_sp Glomus_sp Glomeraceae_sp Glomeraceae_sp Arbuscular mycorrhiza 0% 20% 40% 60% 80% 100% mM NOV_2014 m MAY_2016 mOCT_2016 m APR_2017 mSEP_2017 Biatriospora_sp Biatriospora_mackinnonii Biatriospora_mackinnonii Xylaria_sp Beltrania_rhombica Beltrania_rhombica Xylaria_subtorulosa Cladosporium_sphaerospermum Biatriospora_sp Xylaria_feejeensis Biatriospora_sp Wiesneriomyces_laurinus Annulohypoxylon_sp Trichoderma_gamsii Capronia_acutiseta Phialocephala_humicola Wiesneriomyces_laurinus Wiesneriomyces_laurinus Xylaria_sp Wiesneriomyces_laurinus Xylaria_sp Wiesneriomyces_laurinus Endophytes m NOV_2014 MMAY_2016 mOocT_2016 m APR_2017 MSEP_2017 Entolomataceae_sp Entoloma_sp Sistotrema_sp Richoniella_sp Ceratobasidium_sp Tomentella_sp Entolomataceae_sp Entoloma_furfuraceum Inocybe_sp Entoloma_tenuissimum Entolomataceae_sp Tomentella_sp Chloridium_sp Lyophyllum_sp Ceratobasidium_sp Entoloma_sp Entoloma_caccabus Amanita_arenicola Clavulina_sp Ceratobasidium_sp Tomentella_sp Entolomataceae_sp Clavulina_sp Tricholoma_orirubens Lyophyllum_sp Russula_sp Entolomataceae_sp Lyophyllum_semitale Membranomyces_sp Ectomycorrhizal fungi Mm NOV_2014 MMAY_2016 mOCcT_2016 m APR_2017 MSEP_2017 — DO DI Didymellsceae_sp Fusicoccum_3p Lecanosiicta_acicola S5puelorraes_dolichespermus Curvulaña_banata Ceratobasidiaceat_5p Pseudocercasporella_bakerl Conisla_sp Ehirophydiumn_sp Aurecbasidium_sp Alternaria_alternata Phiyctochytriuen_reinboldtiar Laslodiplodia_theobromae Aapergilles_aculeatus Macophomina_phesealina Ceratobasidlacess_5p Plectospraerella_alsmats Coratobasidiacear_5p Ganoderma_sp Psuudocercospora_hurardi Meofusicoccunn_ribls Thanatephone_cucurmerls Cledosporkum_3p ¡Clsdosporium_sp Fusariurn_chlamydosporur dladosporiumn_herbarun Leptotrochila_sp £nuñia_sp Thanatephonia_cucumnerls Ceratobasidlacess_50 Pestalobopsis_ sp Leptatrochila_p Leptosphaeria_sp Spizellcernces_dolichs per Plant pathogens E 5 5 5 80% : mov_2014 máaY_2016 mOcT_2016 mAPA_2017 aASEP_2017 94 594 595 596 597 Figure S5. Abundance of the persistent community in each fungal guild. 598 LI A É - = E A de 95 96 CAPÍTULO 3. THE MYCORRHIZAL NETWORKS AND RHI-1 ZOSPHERIC FUNGAL COMMUNITIES IN A NEOTROPICAL 2 DRY FOREST ARE RESILIENT 3 4 Running title: Hurricane affects mycorrhizal networks 5 List of authors: Julieta Alvarez-Manjarrez1,5, Mohammad Bahram2,3, Sergei Põlme4, 6 Roberto Garibay-Orijel1 7 Institution affiliations: 1. Instituto de Biología, Universidad Nacional Autónoma de 8 México. Tercer circuito interior s/n, Ciudad Universitaria, 04510 Mexico City, Mexico. 9 2. Institute of Ecology and Earth Sciences, University of Tartu, 40 Lai St., 51005 Tar-10 tu, Estonia. 3. Department of Ecology, Swedish University of Agricultural Sciences, 11 Ulls väg 16, 756 51 Uppsala, Sweden. 4. Natural History Museum, Tartu University, 12 Ravila 12B, 51014 Tartu, Estonia. 5. Posgrado en Ciencias Biológicas. Edificio B, 1º 13 Piso. Circuito de Posgrados, Ciudad Universitaria, 04510, Mexico City, Mexico. 14 Contact information: (+52) 5556229250 ext. 47836, rgaribay@ib.unam.mx 15 Resumen 16 Los ecosistemas forestales están cada vez más sujetos a eventos climáticos extre-17 mos asociados al cambio climático, sin embargo, las consecuencias para la diversi-18 dad subterránea y sus interacciones bióticas siguen siendo poco exploradas. En este 19 trabajo se examinó el efecto del huracán en las redes micorrízicas y las comunida-20 des fúngicas de la rizósfera. Estudiamos los efectos del huracán Patricia (categoría 21 4) que azotó en el 2015 el bosque tropical caducifolio de la costa del Pacífico. Patri-22 cia dañó sustancialmente la vegetación y aumentó el mantillo, y por ende los nutrien-23 tes del suelo. Usamos la secuenciación Illumina MiSeq para caracterizar comunida-24 des fúngicas rizosféricas de plantas con colonización micorrízica dual. Se recolecta-25 ron raíces de nueve parcelas en 2016 y 2017, después de huracán. Se usaron ma-26 trices de presencia-abundancia para calcular las propiedades de las redes micorríci-27 cas –anidamiento, modularidad, especialización– y la concurrencia de especies fún-28 gicas. Inmediatamente después de Patricia, la pérdida de conectividad de la red mi-29 corrízica produjo una alta modularidad y recuperó las conexiones en 2017. Cuando 30 las conexiones se recuperaron, solo una parcela tenía anidamiento, pero la mayoría 31 97 de las redes eran modulares. La red micorrícica arbuscular fue la más dañada en 32 contraste con la red ectomicorrízica. Los gremios de hongos cambiaron sus interac-33 ciones interespecíficas entre años: todos los gremios se excluían mutuamente el 34 primer año de muestreo, y al segundo año, los patógenos de plantas, saprótrofos y 35 endófitos tuvieron una mayor co-ocurrencia. Encontramos una correlación negativa 36 significativa entre la cobertura del dosel (es decir, falta de luz), así como la tempera-37 tura del suelo con diversidad fúngica. La identidad del huésped y la perturbación de-38 terminan la comunidad fúngica. La comunidad fúngica de la rizosfera fue menos di-39 versa el año siguiente al huracán, pero aumentó en el año siguiente. En compara-40 ción con los grupos taxonómicos, la composición de los gremios funcionales respon-41 dió con mayor fuerza a la perturbación: los saprótrofos fueron los gremios de hongos 42 más abundantes en los sitios de alta perturbación. El cambio global puede aumentar 43 la frecuencia e intensidad de los huracanes; Nuestros resultados sugieren que las 44 comunidades fúngicas y sus interacciones son vulnerables a los huracanes pero re-45 silientes. 46 Abstract 47 Forest ecosystems are increasingly subject to climate change associated extreme 48 climatic events, yet the resulting consequences for belowground diversity and its bio-49 tic interactions remain little explored. Here we aimed to examine the effect of hurri-50 cane disturbance in mycorrhizal networks and rhizosphere fungal communities. We 51 studied the effects of Patricia hurricane (category 4) landfall in a tropical dry forest of 52 Pacific coast in 2015. Patricia substantially damaged vegetation and increased litter-53 fall and thus soil nutrients. We used Illumina MiSeq sequencing to characterize rhizo-54 spheric fungal communities of dual-mycorrhizal plants. Their roots were collected 55 from nine plots in 2016 and 2017. Presence-abundance matrixes were used to calcu-56 late properties of mycorrhizal networks –nestedness, modularity, specialization– and 57 co-occurrence of fungal species. Immediately after Patricia the mycorrhizal network 58 lost connectivity producing high modularity and it recovered connections in 2017. 59 When connections were recovered, just one plot was relatively nestedness but most-60 ly networks were modular. Arbuscular mycorrhizal network was the most damaged in 61 contrast with ectomycorrhizal network. Fungal guilds changed their interspecific in-62 teractions between years. The first year of sampling we found all fungal guilds had 63 98 mutual exclusion; in second-year, plant pathogens, saprotrophs and endophytes 64 switch to co-occurr. We found a significant negative correlation between canopy cov-65 erage (i.e. lack of light), as well as soil temperature with fungal diversity. Host identity 66 and disturbance category determined fungal community. Rhizosphere fungal com-67 munity was less diverse the year following hurricane but increased in the subsequent 68 year. Compared to taxonomic groups, the composition of functional guilds responded 69 more strongly to disturbance: saprotrophs were the more abundant fungal guild in the 70 high disturbance sites. Global change may increase frequency and intensity of hurri-71 canes; our results suggest that fungal communities and their interactions are vulner-72 able to hurricanes but resilient. 73 Key words 74 Arbuscular mycorrhiza, cyclone, ectomycorrhiza, extreme events, fungi, interspecific 75 interactions, Pacific coast, Patricia hurricane 76 Introduction 77 Fungi are a major component of topsoil microbial communities, with essential roles 78 for soil functioning as decomposers and plant mutualists. Recent advances in high-79 throughput-sequencing (HTS) have facilitated studying biogeographic patterns and 80 factors that affect fungal communities in natural preserve systems (Tedersoo et al., 81 2014; Peay et al., 2016; Bahram et al., 2018). Rapid advances in the field also allow 82 to classify HTS-based identification into functional guilds such as saprotrophic, my-83 corrhizal, parasitic/pathogen, endophytic, etc. (e.g. Tedersoo et al., 2014; Nguyen et 84 al., 2016; Nilsson et al., 2019). These methods allow the inference of biotic interac-85 tions between fungal species, including different guilds (Chen et al., 2019) as well as 86 those occurring inter-kingdom (Bahram et al., 2018). However, our knowledge about 87 how fungal communities change due extreme climatic events, and whether their inter-88 kingdom or interspecific interactions modified by the effect of them is little. 89 Global change is causing the disturbance to become increasingly common in ecosys-90 tems. Commonly, anthropogenic disturbance combines with natural disturbance such 91 as floods, drought and hurricanes (Banks et al., 2013). High temperature on the oce-92 anic surface has a direct effect on tropical storms and hurricane formation (Hender-93 son-Sellers et al., 1998). Hurricanes will be more frequent and stronger in the second 94 99 half of 21 century (Knudtson et al., 2015). As a result, there is a growing interest in 95 the effect of hurricanes on vegetation and their outcome in other organisms such as 96 birds, insects, mammals, etc. (e.g. Ameca et al., 2018; Bhattarai & Cronin, 2014; Ji-97 ménez-Rodríguez et al., 2018; Lloyd et al., 2019; Novais et al., 2018), with few stud-98 ies on fungi (Cantrell et al., 2014; Vargas et al., 2010). 99 Soil fungal communities, same as above-ground communities, can be affected by 100 disturbance (Banerjee et al., 2019) partly of a cascade effect. Further, environmental 101 changes can modify species interactions (Kennedy 2010; Mahmood 2003). Mycorrhi-102 zal fungi have mutualist interactions forming complex networks with plants (Bingham 103 & Simard, 2011; Toju et al., 2014). Arbuscular mycorrhiza is formed by 71% of terres-104 trial plants, and ectomycorrhizal is formed by 2% of plant (Brundrett & Tedersoo, 105 2018). Ectomycorrhizal fungi commonly habits in 30-60° latitudes (Steidinger et al., 106 2019) and some species in tropics (Corrales et al., 2018). Mycorrhizal fungi are af-107 fected differentially depending on the agent and intensity of disturbance (García de 108 León et al., 2018). 109 Network ecology can help to evaluate ecosystem resilience and stability (Ramirez et 110 al., 2018) or even interspecific interactions in microbial communities (Barberán et al., 111 2012). In general, mutualistic networks are nested, i.e. arranged in a cohesive nucle-112 us (Bascompte et al., 2009). For example, in plant-animal interactions, generalist 113 plants interact more with generalist animals; this result in functional redundancy on 114 the ecosystem (Bascompte et al., 2009). In contrast, modularity is the compartmen-115 talization of closely interacting species (Toju et al., 2014); in disturbed systems, high 116 modularity is proposed to retain the impact of the cascading effect to the neighbor 117 modules (Gilarranz et al., 2017), however still is a debatable topic. 118 Plant-fungal networks are more modular than expected by chance (Toju et al., 2014), 119 same as ectomycorrhizal networks in different ecosystems (Bahram et al., 2014). 120 Mycorrhizal networks found that orchid and ericoid mycorrhizal networks were more 121 modular than ectomycorrhizal, and arbuscular mycorrhizal were intermediate (Põlme 122 et al., 2018). The same properties previously mentioned could be used to study the 123 effect of disturbance. 124 In October 2015 Patricia hurricane category 4 was the strongest record (winds of 345 125 km/h) that made landfall in the Pacific Mexican coast. Our aim was to determine the 126 100 effect of disturbance caused by Patricia hurricane landfall on mycorrhizal networks 127 properties, on fungal rhizosphere communities and its interspecific interactions. Giv-128 en that Patricia hurricane modified drastically vegetation composition and structure 129 (Jiménez-Rodríguez et al., 2018; Parker et al., 2018), that the input of literfall resulted 130 in increased litter C, N, N:P and C:P ratio on soil (Gavito et al. 2018), and that there 131 was a high degree of short-term soil biogeochemical resilience (Jaramillo et al., 132 2018); we hypothesized that: 1) mycorrhizal networks will be highly modular because 133 a high disturbance, 2) increase of soil nutrients caused by hurricane Patricia’s dis-134 turbance has increased rhizosphere fungal diversity, 3) interspecific interactions will 135 be modified boosting competition between guilds. 136 Materials and methods 137 Sampling 138 The study site was the biological station of Chamela, Jalisco, Mexico where Patricia 139 hurricane made landfall in October 23rd, 2015. For sampling we established nine 140 plots of 20 x 20 m in October 2016, were randomly selected on same soil type (coor-141 dinates Table S4). For each plot, we identified, counted and measured diameter of 142 breast height (DBH) of all living trees and shrubs exceeding 3 cm of DBH (Table S1). 143 Light and temperature were measured on the forest floor with HOBO Pendant UA-144 002-08, and humidity with Kestrel three times per plot. We also included in our analy-145 sis the slope to calculate erosion (Stone & Hilborn 2000; table S2). All ectomycorrhi-146 zal hosts were identified and marked (Alvarez-Manjarrez et al., 2018; Figure S1), and 147 three root systems of 10 cm length were sampled per plant and pooled per host for 148 further analyses. In 2017, we re-sampled roots from the same host tree individuals as 149 in 2016. Additionally, we collected three soil cores (5 x 5 cm), they were pooled in a 150 same soil sample from each plot. Samples were stored upon collection at 4°C until 151 further processing. Ectomycorrhizal root-tips were extracted under stereo microscope 152 and preserved in 96% ethanol for further analyses. Soil samples were analyzed by 153 the "Laboratorio de Fertilidad de Suelos y Química Ambiental" from Colegio de Pos-154 graduados, Mexico. Total N and P were obtained by Kjedahl method and quantified 155 colorimetrically of molybdate yellow. Total C was determined by Warkley and Black 156 method. Electric conductivity and pH were analyzed by a conductivity bridge and 157 101 pHmeter, respectively. PO4 was extracted by blue molibdenum colorimetry, NO3 and 158 NO4 were obtained my steam distillation (Figure S2; Table S2). 159 DNA extraction from rhizosphere and soil was performed using PowerMax Soil DNA 160 isolation kit (MoBio; USA). ITS2 rDNA was amplified with five forward primers and 161 one reverse using NextEra adapters at 2.5 µM (Table S3): ITS3NGS1-ITS3NGS5 162 and ITS4NG (Tedersoo et al. 2014). We followed the Taq Platinum Multiplex PCR 163 Master Mix protocol (Life technologies; USA) to make three multiplex PCR replicates 164 per each sample. PCR replicates were pooled to be normalized, purified and se-165 quenced by Illumina Miseq in “Instituto de Medicina Genómica” (INMEGEN), Mexico. 166 Bioinformatics 167 Reads were quality filtered using vsearch software with the default parameters (mi-168 novlen=15; maxdiffs=99; minlength=150; maxee=1; truncqual=0; maxns=0). Se-169 quences were demultiplexed with MOTHUR (Schloss et al. 2009) with default param-170 eters (bdiffs=0). Chimera filtering was done de novo, with UNITE ITS2 v7.1 as a ref-171 erence base filtering, cutting primers, and removing primer artefacts. After, fungal 172 ITS2 was extracted with ITSx (Bengtsson-Palme et al., 2013). To generate the OTU 173 abundance table, resulting ITS regions were clustered using CD-HIT with parame-174 ters: threshold=0.97, min length=50, memory=400, min size=2, length cutoff=0, 175 threads=1, storing=0, algorithm=0. Taxonomy of the most abundance read per clus-176 ter was assigned using BLASTn comparing with UNITE ver. 7.1 (Kõljalg et al. 2013); 177 singleton, i.e. reads with one sequence, were avoided. All this tools were performed 178 in PipeCraft v1.0 toolkit (Anslan et al., 2017). Sequence counts found in controls 179 were subtracted per sample from the OTU table (Nguyen et al., 2014). Fungal guild 180 of each OTU was determined by FUNGuild software (Nguyen et al., 2016). 181 Network analysis 182 To analyze nestedness and modularity of mycorrhizal networks we used a co-183 occurrence matrix of plant-fungi taking into account all mycorrhizal fungal species 184 (i.e. AM, ECM, ericoid and orchid) to compare the variation of the network between 185 years. Separate calculations for AM and ECM networks were performed per each 186 year, per plot and per disturbance classification of plots. Nestedness index was cal-187 culated with NODF function in bipartite R package (Dormann et al., 2008) and its sig-188 102 nificance was tested based on comparing observed and 999 randomized matrices 189 using quantitative swap and shuffle methods ‘swsh_both’ in vegan R package. In ad-190 dition, modularity index and specialization (H2’; Blüthgen et al. 2006) were calculated 191 using netcarto in rnectarto R package (Doulcier & Stouffer, 2015) and bipartite R 192 package, respectively. Specialization measures the interaction strength or interaction 193 frequency between two organisms, making distinction between strong or occasional 194 interactions (Blüthgen et al., 2006). Nonparametric comparisons between network 195 properties were performed with gao function in nparcomp R package (Gao et al., 196 2008) using zero value instead missing data to run the test; missing data were com-197 mon in samples from 2016 due many mycorrhizal species were not colonized. Cen-198 trality was calculated with igraph in R to make the network plots and were visualized 199 using Kamada-Kawai algorithm with ggnetwork in ggplot2 R package (Wickham, 200 2016), coloring by fungal guilds. Bipartite network was plotted using function plotweb 201 in igraph R package (Csardi & Nepusz, 2006). 202 The co-occurrence network analysis was done in CoNet app of Cytoscape, using da-203 ta from rhizosphere and soil separated by years. The matrices contained only fungal 204 OTUs with known guilds, excluding lichenized and fungal parasite lichens because 205 they are not consider as common rhizosperic fungi. The parameters used were mini-206 mum row sum of 20, with Log standardization, using Pearson and Spearman correla-207 tion with Bray-Curtis dissimilarity distance. The analysis was performed using 100 208 bootstrap iterations with Benajmin-Hochberg test correction for the P-value threshold 209 0.05. The graph was generated using the mean of multi-graph, using union as net-210 work merge, adding the guilds as attributes to compare co-occurrence between them. 211 Statistical analysis 212 Sequences abundances bigger than median were rarefied with rrarefy in vegan 213 package of R (Oksanen et al., 2019). The rarefied matrix was used to calculate β di-214 versity turnover Simpson index, nestedness resultant from dissimilarity matrix 215 Sorensen and β diversity with Sorensen index with betapart in R package. 216 To compare the effect of disturbance on plot vegetation, soil and environmental 217 characteristics, a cluster analysis was used based on Ward distance with hclust in R 218 package (Figure S3). 219 103 The data were visualized in a NMDS ordination plot using Bray-Curtis dissimilarity 220 matrix (Bahram et al. 2018). Prior to that, data were normalized using Hellinger trans-221 formation. To determine the significance of the effect from abiotic characteristics a 222 permutational analysis of variance (PERMANOVA) was performed with adonis in ve-223 gan package of R (Oksanen et al., 2019). We calculate Shannon diversity index of 224 every sample and calculate a linear regression with lm in stats package of R. We 225 used a Student-t test to compare Shannon diversity index between 2016 and 2017. 226 Results 227 Characteristics of plots 228 Plant richness in plots varied between 28 to 59 species. The most abundant ectomy-229 corrhizal hosts were Achatocarpus gracilis, Apoplanesia paniculata, and Guapira pe-230 tenensis (Figure S1; Table S1); we also found Coccoloba liebmanii, Lonchocarpus 231 eriocarinalis, and Lonchocarpus spp. Cluster analysis grouped the samples accord-232 ing to the level of disturbance: 1) plots T2800 and A500 with low light at ground level, 233 lowest density of trees, tendency to be eroded, high content of PO4, and low number 234 of stand-up trees alive; 2) plots T450 and T1000 with high tree density, high number 235 of fallen trees alive, lowest slope and erosion, and high organic matter and total; 3) 236 the rest of plots with fewer dead trees, moderate slope, and lower NH4 (Figure S2; 237 Table S2). Hereafter, these groups are referred to as “high disturbance”, “low dis-238 turbance, and “recovery”, respectively (Figure S3). 239 Mycorrhizal network analysis 240 Mycorrhizal network properties did not show any difference according to level of dis-241 turbance. The comparison of mycorrhizal networks of all plots between one and two 242 years after the hurricane exhibited both anti-nestedness and low modularity: in 2016 243 the network had 78 fungal OTUs, NODF= 9.019 (z-value= -2.77, 2.5% CI=9.44, 244 97.5% CI=11.51, P=0.011, WNODF=10.093) and 0.41 modularity; and in 2017 the 245 network had 146 fungal OTUs, NODF= 5.496 (z-value= -3.257, 2.5% CI=5.82, 97.5% 246 CI=6.745, P=0.003, WNODF= 5.71) and 0.38 of modularity. 247 In 2016 mycorrhizal richness and abundance was low and increased in 2017 (Table 248 1). The first year after hurricane we found few mycorrhizal species shared between 249 hosts. Also, most of the shared fungal species were similar between same plant spe-250 104 cies. Tomentella sp. (SH006884.07FU) and Clavulina sp. (SH629574.07FU) were the 251 most generalist fungal species, meanwhile Guapira and Achatocarpus were the hosts 252 with more centrality. The second year after hurricane more and different fungal spe-253 cies connected the plants such as Russula sp. (SH526877.07FU), Clavulina spp. 254 (SH629574.07FU, SH220229.07FU), Ceratobasidium ramicola (SH218661.07FU), 255 Tomentella spp. (SH006884.07FU, SH489022.07FU), Inocybe sp. 256 (SH493665.07FU), Helvella sp. (SH492769.07FU), Chloridium sp. (KY88725), and 257 some Glomeraceae, sp. (SH001065.07FU). The plants with more centrality were 258 Guapira petenensis, Apoplanesia paniculata, and Achatocarpus gracilis (Figure 1). In 259 both years, Thelephoraceae species were common and abundant on host roots (Fig-260 ure 1a, d). 261 Analyzed rhizospheres harbored both arbuscular mycorrhizal (AM) and ectomycor-262 rhizal (ECM). Throughout both years following the hurricane AM networks showed 263 tendency for nested structure, while ECM communities exhibited anti-nested pat-264 terns; none of the observations were statistically significant (P= 0.06 in 2016; P=0.05 265 in 2017; Table 1). Modularity was always highest in AM networks. The specialization 266 of networks (H2’) increased during following years after the hurricane (Table 1). Ac-267 cording to nonparametric multiple comparison there were no significant differences in 268 mycorrhizal network properties between years (wNODF: F18, 15.68=0.63, P-value=0.53; 269 Modularity: F18, 14.42=1.63, P-value=0.12; Specialization: F18, 15.75=-0.63, P-270 value=0.53). 271 272 Table 1. Properties of mycorrhizal networks pooling all plots by years. 273 Year Myco- net* No. OTUs OTUs abun* NODF* z- value 2.5% CI 97.5% CI P va- lue wNODF * Mod* H2’* 2016 AM 31 172 7.889 0.509 5.759 9.085 0.627 8.185 0.504 0.758 ECM 46 25833 10.78 -1.74 10.712 13.431 0.063 12.039 0.358 0.791 2017 AM 51 1468 7.905 0.31 6.592 8.707 0.793 4.89 0.372 0.827 ECM 91 41373 7.088 -2.01 7.081 8.654 0.053 8.794 0.33 0.899 *Myconet=mycorrhizal network, AM= arbuscular mycorrhizal, ECM= ectomycorrhizal, 274 OTUs abun= OTUs abundance, Mod=modularity, NODF= nestedness, wNODF= 275 weigthed nestedness, H2’= specialization 276 277 105 When the same analyses were performed separately for ECM and AM communities, 278 the first year just two plots had at least one mycobiont shared between plants spe-279 cies, restoring mainly ectomycorrhizal fungi co-occurrence in 2017 (Table S5). One 280 year after hurricane it was impossible to calculate nestedness or modularity in most 281 of the plots because they were few mycorrhizal species shared by hosts (Table 2). 282 283 Table 2. Comparison of mycorrhizal network properties pooling all mycorrhizal fungi 284 in two different years in each plot 285 Year Plot No. OTUs OTUs abun* NODF z- value 2.5% CI 97.5% CI P va- lue wNODF Mod* H2’* 2016 A250 30 11425 6.316 0 6.31 6.316 1 6.24 NA 0.899 A500 5 11 0 0 0 0 1 0 NA 1 B200 14 303 0 0 0 0 1 0 NA 1 EC650 19 1284 14.91 -0.255 11.6 17.956 0.711 13.352 0.48 0.62 T450 27 7526 6.818 0 6.81 6.818 1 0.852 NA 0.993 T700 8 35 0 0 0 0 1 0 NA 1 T1000 5 16 0 0 0 0 1 0 NA 1 T2650 6 77 0 -2.621 0 16.667 0.185 5.55 0.625 0.789 T2800 13 5329 0 0 0 0 1 0 NA 1 2017 A250 59 26725 1.692 -3.012 1.99 4.132 0.011 4.307 NA 0.985 A500 42 1211 6.265 1.151 3.03 6.372 0.087 3.344 0.437 0.941 B200 20 213 13.6 -0.22 9.35 16.78 0.69 9.75 0.494 0.745 EC650 42 11648 5.717 -2.12 5.67 8.361 0.05 8.432 0.619 0.949 T450 25 218 3.921 -1.718 3.41 7.23 0.119 4.35 0.701 0.856 T700 9 28 0 0 0 0 1 0 NA 1 T1000 41 774 3.38 -2.02 3.17 6.159 0.069 5.34 0.553 0.916 T2650 15 128 5.66 0 5.66 5.66 1 0 NA 0.746 T2800 13 1924 0 0 0 0 1 0 NA 1 *OTUs abun= OTUs abundance, Mod=modularity, H2’= specialization, wNODF= weigthed nesteness. 286 NA= missing value. 287 Interspecific interactions of rhizospheric fungal community 288 Interactions between the rhizospheric fungal guilds changed in each year. During 289 2016, ECM fungi had negative occurrence interactions among saprotrophs, plant 290 pathogens, animal pathogens, endophytes, and other ECM species. Most of the 291 abundant Glomeraceae species co-occurred with ECM, saprotrophs, between other 292 106 arbuscular mycorrhizal fungi, and excluded plant pathogens (Figure S4a). In general, 293 plant pathogens co-occurred with saprotrophs and some endophytes species (Figure 294 2a, S4a). 295 AM abundance ECM abundance = 5 => E = L = 0 = = LY N O D o r o = < X W L J O O ..fn p r . . . J o s o o o . o 0 . - O e d 0 ' . . e . , . . o e 1 | a , . ¿ n e d M L A AS Ñ ? 2 . e s a! Ñ Ñ o ARA A . M A A e Al . e db : y e . le » o ú e " e - 0 p o ? == YX dl A A A e e N A L e $ + a . Y .e Y Y ¿ a C S 4 O +» » q ó z » z o a My a G z * y dá | . 7 : p ? e o 8zL m o ds sesoeJauoje 70 - OLOLMO' ds sesoejauo"o ¿ 9 6 0 0 d e s e s o e l s u o " o D L E O M O d s S e S I E l S U u O J o 6 2 5 . m o d s s e s d e l s u o o z o s m o d e S e S J e l S u o " o - 9 / 6 0 M 0 ' ds sesoejauo¡o - E L E F L M O d s S e s J E J a u o " e - D E z Z O M O d s s a / e J a u o ¡ D e l e g - +azimo d e 3 e S 0 B J a u o | o 25- 3000 20- B- 1 0- 1 oouepunqy 2000 auepunqy 1000- B- D- - 1 8 0 0 ds ejoDÁuOIauo"o - 1 S 9 0 m o d s sesoeJaulo"o - - 6 2 6 0 0 ds sesoeJauoje L - 9 0 $ 0 M O d s s e s o e j a u o o -££91 MO ds ejosÁwolsuo|o species s o r o m o ' ds sesoe/aulo|o 5991 m o ds aesoe/aulo|o 6801mo' ds eesoelaulo|o 8111 mods sesoeJauo]o 6 6 2 1 MoO ds $259 9501 m o ds aesoe/aulo|o g e g o m o d s aesdelaulo|o 50 D- 200 - 150 -| D- 3 100- aouepunqy N440'9101ZZ2HS B r d n u i a o u e u eL Y N W u n u l s e l y j u e e u o j o y u 3 M 4 2 0 ' S 8 9 6 L 1 H S ds elena W N ds ejnssny M420'ZZ9L£0HS d s e u I n a e ¡ o M320's99e6HHs ds eqÁsoul M 3 L 0 " 4 4 8 9 7 9 H s d s ejnssny MaAZO"pOZ8BYHs ds e u n a e ¡ o M420'50Z88vHs ds eulsegas N n 4 2 0 ' 8 0 6 6 4 1 H s ds s e s o e u i n a e l o M n 4 2 0 ' 2 2 0 6 8 b H s ds ejjejuelo 1 M420'y4S6Z9HS ds B u I n a e ¡ o m0 p 8 8 9 0 0 H s ds ejjejuanuo y species 500 - o- 1500 - E SouEpunqy e 6 s 0 m o d s s e s s e l s u o j s L L z O M O d s s e s d e l s u o o c 8 5 . m o d e S e S J e I S u o " o £osi mo ds s e s s e l s u o j s 0001. mMo d s " s n u o s - ¿ 2 0 1 0 s o 0 y s n 9 s n b e y d o z i y y + 2 mo ds s e s s e l s u o j o 6 s 0 0 m o ' d s snuwobojdas +92 m o d s S e d e s u o " o SL FL m o d s S e s d E l S u o j s 2 0 6 0 0 d s S e E S I E l S U O | S 2 2 6 0 0 d s S e s d e l S i o o species 3081 M0 ds epooÁuWIoJsulo [o ¿20.0 d s S e S I E l S U u O J o Z 8 0 . M O d s p e s o z o o m o d e S e S J e l S u o " o L L O L M O d s s e s d e l s u o J o £LLLMO d s S e s d e l S i o o 0841 mo ds p z 3 0 6 5 0 . 0 d s S s e s s E l S u o j s 2 L L L m o d s s e s d e l s u o o 6 2 6 0 0 d e S E S I E I S u o | o £soLmo ds s e s s e l s u o j s 2 6 9 0 0 d s S e s 9 E / S u o " o L e 6 0 m o O d e S e s d e l s u o " o £2£FL M O d s S e S I E l S u o O J S £ s L o n o ds e u I m a e o DESONO u n ussinua] BLIoo Ju s 8 0 . m o d e e u o ¡ o u y ó6sLOnoO ds E u m e Pe90nmo ISIStyr E N U e u + L 6 0 M 0 E]2908/ eueDJe7 6 s 9 L o m o ' ds e u n a e ¡ o + r L O m O d e enssny LESemoO ds u n i p u a y o 6 8 6 0 m o d s e u o o u zL0Lmo' a s u s b u o p u e y s Buojoyuy s e 2 0 m o d s enssmy 6 4 s 0 m o ' ds u n Aydo 47 2 t O S L MO de e l s a :ú Z E g o O m o O d s S e s o B u I n a e o 2 £ L s o m o d s e j a p u e u u o y 177] Z E Z O M O EJogILE L U N A S ego z z g o m o d s eydkssopaulal] £ z c o m o ' d s epausuwo! ¿ 9 1 0 ds eupegss B££1 M0 de s e 8 0 8 / 2 u o ¡ o J u £ L s e m o ds u n i p u a y o 9 0 8 1 0 ds sqásoul 2 8 L o n o d s euImaero e e v o m o d s e p s p u s u o y o r L o m o d s enssny L s s o m o ds eydissoauasL z 6 L o m o d e P u n a e ¡ o +1500]0 sipesisa eJoydejsyl L e s o m o d s e p a p u s u o y 107 296 108 Figure 1.Mycorrhizal fungal species abundance in 2016 and 2017, and the mycor-297 rhizal network, plotted by Kamada-Kawai algorithm. 298 299 In 2017, richness increased generating more significant interactions in a more com-300 plex network. Ectomycorrhizal fungi contained more OTUs but with lower abundance 301 compared to other guilds, showing mainly intra-guild and inter-guild negative co-302 occurrence, especially with saprotrophs. Arbuscular mycorrhizal species showed low 303 number of negative co-occurrences with saprotrophs and endophytes. Plant patho-304 gens had relations of positive correlations with most of the saprotrophs and arbuscu-305 lar mycorrhiza. Endophytes were co-present with arbuscular mycorrhiza, plant patho-306 gens, saprotrophs, however there were some species that excluded saprotrophs (e.g. 307 Xylaria can be classified also as saprotroph, exclude another saprotrophs) and ecto-308 mycorrhizal fungi (figure 2b). 309 310 Figure 2. Interspecific interactions between more abundant taxa of the rhizospheric 311 guilds one year (A) and two years (B) after hurricane Patricia. A) Build with the 31 312 OTUs with more than 100 sequences; B) Build with the 85 OTUs with more than 200 313 sequences. Size of the node is the frequency of negative interactions (mutual exclu-314 sion). Abbreviations: AM= Arbuscular mycorrhizal, ECM= ectomycorrhizal, OM= or-315 chid mycorrhizal. 316 109 Hurricane effect on rhizosphere fungal communities 317 PERMANOVA analysis revealed that significant predictors of rhizosphere community 318 were soil temperature (F1, 59= 2.858, R2= 0.04, P= 0.001), year (F1, 59=2.851, 319 R2=0.051, P=0.001), hosts (F9, 59=1.689, R2=0.214, P=0.001), disturbance in plots 320 (F2, 59=1.4, R2=0.039, P=0.018), and light at ground level (F1, 59=1.338, R2=0.018, 321 P=0.042) (Figure S2). Rhizospheric fungal community had a drastic replacement one 322 year after Patricia hurricane and recovered changed again two years after (Figures 323 S5-8). Ectomycorrhizal community was mainly determined by host identity (F8, 324 53=2.145, R2=0.238, P=0.001) followed by year (F1,53= 5.388, R2=0.075, P=0.01), soil 325 ammonium concentration (F1, 53=2.637, R2=0.036, P=0.028), soil temperature (F1,53= 326 3.15, R2=0.043, P=0.014), and plant richness (F1,53= 2.483, R2=0.034, P=0.037) 327 (Figure S9). 328 329 Figure 3. The significative predictors of the rhizosphere fungal community: A) Soil 330 temperature, B) year of sampling, C) plant species, D) light at ground level, E) plot 331 disturbance. Bold horizontal lines from boxplots represents mean values, and box 332 margins are variance. 333 110 Discussion 334 Hurricane Patricia affected mycorrhizal networks, interspecific interactions between 335 guilds and the structure of rhizospheric fungal communities. We provide the first 336 analysis from dual mycorrhizal networks in a tropical forest. Mycorrhizal networks lost 337 connectivity and specialization one year after hurricane and it recovered connections 338 in most of the plots in 2017. Fungal diversity was lower in 2016 than 2017; thus, in 339 2017 the increased diversity produced more complex networks and changed the in-340 teractions between species. Higher soil temperature was correlated negatively with 341 rhizospheric fungal diversity. Also we found an effect depending on year, host identi-342 ty, disturbance level and light at level ground. 343 Arbuscular mycorrhizal network was lost one year after hurricane; most of the mycor-344 rhizal species were restricted to colonize just one plant species. AM colonization can 345 increase after hurricane (Varga et al., 2010), but colonization loses the sight of con-346 nections between plants. We hypothesize that vegetation damage would produce a 347 bottom-up effect on rhizosphere communities in general, and mycorrhizal species 348 could be the most harmed due their need of photosynthetic carbon. 349 Results of ECM and AM networks showed no-nestedness and modularity in both my-350 corrhizal interactions, with higher modularity in ECM network. Additionally, we found 351 more specialization in 2017 in both interactions. Mycorrhizal species specificity to 352 their host could explain the high modularity (Bahram et al., 2014). Specificity together 353 with modularity, can change with disturbance, e.g. plant-herbivore network compari-354 son between before and after hurricane, found a decrease in specificity and number 355 of compartments (Luviano et al., 2018). Our data agree with the idea that more spe-356 cialized species are more common in seasons with high availability of resources (low 357 stressful abiotic conditions) and generalists prevail in any conditions (López-358 Carretero et al., 2014). Also, high modularity of mycorrhizal fungi could give resili-359 ence to the disturbance (Gilarranz et al., 2017). 360 Mycorrhizal communities showed host preference (Figure S7). The same pattern had 361 observed in tropical forests (Peay et al., 2013; Tedersoo et al., 2008; 2010). Before 362 and after hurricane, hosts associated with few common fungal species (Alvarez-363 Manjarrez et al., 2018), however successional changes occurred in the community. 364 The generalist species before hurricane were not resistant one year after hurricane 365 111 but were resilient two years after: Tremelloscypha sp. (SH016792.07FU), Membran-366 omyces sp. (SH1143177.08FU), Sebacina sp. (SH488205.07FU), Thelephora ver-367 satilis (SH490448.07FU) and Tomentella sp. (SH495677.07FU) connected plants 368 species before hurricane (Alvarez-Manjarrez et al., 2018). In 2016, these fungi were 369 replaced by Tomentella sp. OTU 537 (SH006884.07FU), Clavulina sp. OTU 169 370 (SH629574.07FU), Inocybe sp. OTU 1506 (SH493665.07FU), Tomentella sp. OTU 371 489 (SH489022.07FU) and Clavulinaceae, sp. OTU 332 (SH179908.07FU). Two 372 years after Patricia (2017), dominant species before hurricane replaced those that 373 appeared in 2016. 374 Besides abiotic factors, biotic interactions structure relationships between fun-375 gal species. Co-occurrence network analysis demonstrated significant correlations in 376 presence-absence patterns of fungal species. Abundance and rarity could be inter-377 preted as competition between fungi (Kennedy 2010). In 2016, ectomycorrhizal fungi 378 excluded arbuscular mycorrhizae, saprotrophic, pathogenic and other ECM species, 379 and same interactions were found in AM. Meanwhile, in 2017 when diversity in-380 creased, arbuscular mycorrhizal were more abundant than ECM, and co-occur more 381 with pathogens. ECM species can inhibit pathogens’ establishment (Mohan et al., 382 2015), mainly inhibit saprotrophs (‘Gadgil effect’; Gadgil & Gadgil, 1971; Fernandez & 383 Kennedy, 2015), and AM establishment (Chen et al., 2000). When hurricanes land-384 fall, the C allocation from plants to fungi change ebbs at root system, so competition 385 strengthened for the same resource. AM can exclude or improve establishment other 386 fungal species. AM species compete for root-resources, but also some AM species 387 promote the mycelial growth of other fungi (Bennett & Bever, 2009). Arbuscular colo-388 nization can be diminished by endophytes, while AM had not impact on endophytes 389 (Mack & Rudgers, 2008). In general, we found antagonistic relationships are unbal-390 anced (Mack & Rudgers, 2008) and tend to change with environmental alterations 391 (Kennedy 2010). 392 There is growing evidence that a regime of environmental disturbance has different 393 effects on plant communities depending on frequency, intensity, spatial and temporal 394 scale of the disturbance event. The main disturbance effect of a hurricane is the mor-395 tality and loss of vegetal biomass (e.g. Zimmerman et al., 1994; Parker et al., 2018), 396 which has cascading effects on the rest of the biotic communities. Our comparison 397 112 between years suggest that similarly to plants, rhizospheric fungal communities show 398 great changes following hurricane, pointing to their possible role in establishing the 399 recruitment of plants. 400 We refute our hypothesis about the soil nutrients having an effect on rhizosphere 401 fungal diversity, as it was demonstrated in different studies (Lilleskov et al., 2012; 402 Truong et al., 2019). Interestingly, diversity increased together with the light at level 403 ground, probably because the secondary succession of the forest. In 2016, light at 404 forest floor scarce on plots with few survival trees, because creeping pioneer herbs 405 (Mimosa pudica and others) covered completely the ground. Next year, lianas disap-406 peared, and new plant community grew up, they were herbs and recruits that allowed 407 more light on forest floor. 408 The fungal community composition did not differ between our three categorical dis-409 turbance levels, i.e. “low”, “recovery”, “high” disturbance (except when accounting for 410 fungal guilds). Fungal diversity was higher in plots categorized as ‘recovery’, though 411 we had more sampled plots in that category (five plots in comparison with two in oth-412 er categories). Diversity could be explained by the ‘intermediate disturbance hypoth-413 esis’, where disturbance produce variations in spatial-temporal resources developing 414 different coexistence mechanisms of species (Roxburgh et al., 2004). 415 Extreme climatic events, such as hurricanes, disturb more frequently, and will in-416 crease even more caused by global warming (Walsh et al., 2016); both, anthropogen-417 ic and extreme climatic events create disturbance and these in turn shape secondary 418 successional forests (Lewis et al., 2015; Salazar et al., 2015). Disturbance complexity 419 creates not a unidirectional effect on fungal communities, e.g. arbuscular mycorrhizal 420 communities (García de León et al., 2018). Hurricane Patricia harmed rhizospheric 421 communities and their interaction networks (i.e. mycorrhizal networks and interspecif-422 ic symbioses), however high fungal diversity contributed to their resilience. Whether 423 human activities preserve high fungal diversity, the future will be promising in the face 424 of global change. 425 Conclusions 426 This study was the first to examine the effect of hurricane on root associated fungal 427 communities and its interactions. It is important to consider that each hurricane event 428 vary the environmental conditions depending on its strength, vegetal community as-429 113 sembly and local conditions. Overall, our results indicate arbuscular mycorrhizal net-430 work is more sensible to disturbance than ectomycorrhizal network, and in general 431 rhizospheric fungal community are vulnerable to hurricane disturbance and able to 432 recover. Given the growing number of climate change associated disturbance, further 433 studies are needed to determine the functional implications of such changes. 434 Acknowledgments 435 We thank CONACYT for funding the project Ciencia Básica 239266, for the PhD 436 scholarship 404413 and for beca mixta. We thank the program ‘Posgrado en Cien-437 cias Biológicas’ of the National Autonomous University of Mexico for JAM postgradu-438 ate studies. We thank the field workers Abel Domínguez Pompa, Jorge Blanco, 439 Rodolfo Ángeles Argáiz, Sergio Vidal, and Diego Flores, and the staff at the Biologi-440 cal Station of Chamela. We also thank Andrés Argüelles-Moyao for helping at labora-441 tory and Saleh Rahimlou for his assistance in bioinformatics. We thank Alfonso Del-442 gado Salinas, Alfonso Valiente Banuet, Mariana Benítez Keinrad, Patricia Vélez, and 443 Margarita Villegas Ríos for their valuable comments and corrections. 444 References 445 Alvarez-Manjarrez, J., Garibay-Orijel, R., & Smith, M. E. (2018). Caryophyllales are 446 the main hosts of a unique set of ectomycorrhizal fungi in a Neotropical dry 447 forest. Mycorrhiza, 28(2), 103–115. https://doi.org/10.1007/s00572-017-0807-7 448 Ameca, E. I., Mace, G. M., Cowlishaw, G., & Pettorelli, N. (2018). Relative vulnera-449 bility to hurricane disturbance for endangered mammals in Mexico: a call for ad-450 aptation strategies under uncertainty. Animal Conservation, 22, 262–273. 451 https://doi.org/10.1111/acv.12461 452 Anslan, S., Bahram, M., Hiiesalu, I., & Tedersoo, L. (2017). PipeCraft: Flexible open-453 source toolkit for bioinformatics analysis of custom high-throughput amplicon se-454 quencing data. Molecular Ecology Resources, 17(6), e234–e240. 455 https://doi.org/10.1111/1755-0998.12692 456 114 Bahram, M., Harend, H., & Tedersoo, L. (2014). Network perspectives of 457 ectomycorrhizal associations. Fungal Ecology, 7(1), 70–77. 458 https://doi.org/10.1016/j.funeco.2013.10.003 459 Bahram, M., Hildebrand, F., Forslund, S. K., Anderson, J. L., Soudzilovskaia, N. A., 460 Bodegom, P. M., … Bork, P. (2018). Structure and function of the global topsoil 461 microbiome. Nature, 560(7717), 233–237. https://doi.org/10.1038/s41586-018-462 0386-6 463 Banerjee, S., Walder, F., Büchi, L., Meyer, M., Held, A. Y., Gattinger, A., … van der 464 Heijden, M. G. A. (2019). Agricultural intensification reduces microbial network 465 complexity and the abundance of keystone taxa in roots. ISME Journal, 13, 466 1722–1736. https://doi.org/10.1038/s41396-019-0383-2 467 Banks, S. C., Cary, G. J., Smith, A. L., Davies, I. D., Driscoll, D. A., Gill, A. M., … 468 Peakall, R. (2013). How does ecological disturbance influence genetic diversity? 469 Trends in Ecology and Evolution, 28(11), 670–679. 470 https://doi.org/10.1016/j.tree.2013.08.005 471 Barberán, A., Bates, S. T., Casamayor, E. O., & Fierer, N. (2012). Using network 472 analysis to explore co-occurrence patterns in soil microbial communities. ISME 473 Journal, 6(2), 343–351. https://doi.org/10.1038/ismej.2011.119 474 Bascompte, J. (2009). Mutualistic networks. Frontiers in Ecology and the 475 Environment, 7(8), 429–436. https://doi.org/10.1890/080026 476 Bennett, A. E., & Bever, J. D. (2009). Trade-offs between arbuscular mycorrhizal fun-477 gal competitive ability and host growth promotion in Plantago lanceolata. Oeco-478 logia, 160(4), 807–816. https://doi.org/10.1007/s00442-009-1345-6 479 Bengtsson-Palme, J., Ryberg, M., Hartmann, M., Branco, S., Wang, Z., Godhe, A., … 480 Nilsson, R. H. (2013). Improved software detection and extraction of ITS1 and 481 ITS2 from ribosomal ITS sequences of fungi and other eukaryotes for analysis of 482 environmental sequencing data. Methods in Ecology and Evolution, 4(10), 914–483 919. https://doi.org/10.1111/2041-210X.12073 484 115 Bhattarai, G. P., & Cronin, J. T. (2014). Hurricane activity and the large-scale pattern 485 of spread of an invasive plant species. PLoS ONE, 9(5). 486 https://doi.org/10.1371/journal.pone.0098478 487 Bingham, M. A., & Simard, S. W. (2011). Do mycorrhizal network benefits to survival 488 and growth of interior Douglas-fir seedlings increase with soil moisture stress? 489 Ecology and Evolution, 1(3), 306–316. https://doi.org/10.1002/ece3.24 490 Blüthgen, N., Menzel, F., & Blüthgen, N. (2006). Measuring specialization in species 491 interaction networks. BMC Ecology, 6. https://doi.org/10.1186/1472-6785-6-9 492 Brundrett, M. C., & Tedersoo, L. (2018). Evolutionary history of mycorrhizal symbio-493 ses and global host plant diversity. New Phytologist, 220(4), 1108–1115. 494 https://doi.org/10.1111/nph.14976 495 Cantrell, S. A., Molina, M., Jean Lodge, D., Rivera-Figueroa, F. J., Ortiz-Hernández, 496 M. L., Marchetti, A. A., … Pérez-Jiménez, J. R. (2014). Effects of a simulated 497 hurricane disturbance on forest floor microbial communities. Forest Ecology and 498 Management, 332, 22–31. https://doi.org/10.1016/j.foreco.2014.07.010 499 Chen, Y. L., Brundrett, M. C., & Dell, B. (2000). Effects of ectomycorrhizas and ve-500 sicular-arbuscular mycorrhizas , alone or in competition , on root colonization 501 and growth of Eucalyptus globulus and E . urophylla. New Phytologist, 146(3), 502 545–556. 503 Chen, J., Xu, H., He, D., Li, Y., Luo, T., Yang, H., & Lin, M. (2019). Historical logging 504 alters soil fungal community composition and network in a tropical rainforest. 505 Forest Ecology and Management, 433(October 2018), 228–239. 506 https://doi.org/10.1016/j.foreco.2018.11.005 507 Corrales, A., Henkel, T. W., & Smith, M. E. (2018). Ectomycorrhizal associations in 508 the tropics – biogeography, diversity patterns and ecosystem roles. New Phytol-509 ogist, 220(4), 1076–1091. https://doi.org/10.1111/nph.15151 510 116 Csardi, G. & Nepusz, T. (2006). The igraph software package for complex network 511 research, InterJournal, Complex Systems 1695. http://igraph.org 512 Dormann, C.F., Gruber B. & Fruend, J. (2008). Introducing the bipartite Package: 513 Analysing Ecological Networks. R news Vol 8/2, 8 - 11. 514 Doulcier, G. & Stouffer, D. (2015). Rnetcarto: Fast Network Modularity and Roles 515 Computation by Simulated Annealing. R package version 0.2.4. 516 DRYFLOR. (2016). Plant diversity patterns of the neotropical dry forest and their 517 conservation implicatons. Science, 353(6306), 1383–1387. 518 https://doi.org/10.1126/science.aaf5080 519 Fernandez, C. W., & Kennedy, P. G. (2016). Revisiting the “Gadgil effect”: Do inter-520 guild fungal interactions control carbon cycling in forest soils? New Phytologist, 521 209(4), 1382–1394. https://doi.org/10.1111/nph.13648 522 Gadgil, R., & Gadgil, P. (1971). Mycorrhiza and litter decomposition. Nature, 233, 523 133. https://doi.org/https://doi.org/10.1038/233133a0 524 Gao, X., Alvo, M., Chen, J., & Li, G. (2008). Nonparametric multiple comparison 525 procedures for unbalanced one-way factorial designs. Journal of Statistical 526 Planning and Inference, 138(8), 2574–2591. 527 https://doi.org/10.1016/j.jspi.2007.10.015 528 García de León, D., Davison, J., Moora, M., Öpik, M., Feng, H., Hiiesalu, I., … Zobel, 529 M. (2018). Anthropogenic disturbance equalizes diversity levels in arbuscular 530 mycorrhizal fungal communities. Global Change Biology, 24(6), 2649–2659. 531 https://doi.org/10.1111/gcb.14131 532 Gavito, M. E., Sandoval-Pérez, A. L., del Castillo, K., Cohen-Salgado, D., Colarte-533 Avilés, M. E., Mora, F., … Urquijo-Ramos, C. (2018). Resilience of soil nutrient 534 availability and organic matter decomposition to hurricane impact in a tropical dry 535 forest ecosystem. Forest Ecology and Management, 426(August), 81–90. 536 https://doi.org/10.1016/j.foreco.2017.08.041 537 117 Gilarranz, L. J., Rayfield, B., Liñán-Cembrano, G., Bascompte, J., & Gonzalez, A. 538 (2017). Effects of network modularity on the spread of perturbation impact in 539 experimental metapopulations. Science, 357(6347), 199–201. 540 https://doi.org/10.1126/science.aal4122 541 Goldbeg, D.E. (1990). Components of resource competition in plant communities. In: 542 Grace, J.B., & Tilman, D. (eds.) Perspective on plant competition. San Diego, 543 CA, USA: Academic Press, 27-49 544 Henderson-Sellers, A., Zhang, H., Berz, G., Emanuel, K., Gray, W., Landsea, C., … 545 McGuffie, K. (1998). Tropical Cyclones and Global Climate Change: A Post-IPCC 546 Assessment. Bulletin of the American Meteorological Society, 79(1), 19–38. 547 https://doi.org/10.1175/1520-0477(1998)079<0019:tcagcc>2.0.co;2 548 Jaramillo, V. J., Martínez-Yrízar, A., Maass, M., Nava-Mendoza, M., Castañeda-549 Gómez, L., Ahedo-Hernández, R., … Verduzco, A. (2018). Hurricane impact on 550 biogeochemical processes in a tropical dry forest in western Mexico. Forest 551 Ecology and Management, 426(October), 72–80. 552 https://doi.org/10.1016/j.foreco.2017.12.031 553 Jiménez-Rodríguez, D. L., Alvarez-Añorve, M. Y., Flores-Puerto, J. I., Oyama, K., 554 Avila-Cabadilla, L. D., Pineda-Cortes, M., & Benítez-Malvido, J. (2018). Structur-555 al and functional traits predict short term response of tropical dry forests to a 556 high intensity hurricane. Forest Ecology and Management, 426(April), 101–114. 557 https://doi.org/10.1016/j.foreco.2018.04.009 558 Kennedy, P. (2010). Ectomycorrhizal fungi and interspecific competition: Species 559 interactions, community structure, coexistence mechanisms, and future research 560 directions. New Phytologist, 187(4), 895–910. https://doi.org/10.1111/j.1469-561 8137.2010.03399.x 562 Knutson, T. R., Sirutis, J. J., Zhao, M., Tuleya, R. E., Bender, M., Vecchi, G. A., … 563 Chavas, D. (2015). Global projections of intense tropical cyclone activity for the 564 late twenty-first century from dynamical downscaling of CMIP5/RCP4.5 scenari-565 118 os. Journal of Climate, 28(18), 7203–7224. https://doi.org/10.1175/JCLI-D-15-566 0129.1 567 Kõljalg, U., Nilsson, R. H., Abarenkov, K., Tedersoo, L., Taylor, A. F. S., Bahram, M., 568 … Larsson, K.-H. (2013). Towards a unified paradigm for sequence-based iden-569 tification of fungi. Molecular Ecology, 22, 5271–5277. 570 Lewis, L., Edwards, D., & Galbraith, D. (2015). Increasing human dominance of tropi-571 cal forests. Science, 349(6250), 827–832. 572 https://doi.org/10.1126/science.aaa9932 573 Lilleskov, E. A., Fahey, T. J., Horton, T. R., Lovett, G. M., Ecology, S., & Jan, N. 574 (2012). Belowground ectomycorrhizal fungal community change over a nitrogen 575 deposition gradient in Alaska. Ecological Society of America, 83(1), 104–115. 576 https://doi.org/10.2307/2680124 577 López-Carretero, A., Díaz-Castelazo, C., Boege, K., & Rico-Gray, V. (2014). Evaluat-578 ing the spatio-temporal factors that structure network parameters of plant-579 herbivore interactions. PLoS ONE, 9(10). 580 https://doi.org/10.1371/journal.pone.0110430 581 Lloyd, J. D., Rimmer, C. C., & Salguero Faría, J. A. (2019). Short-term effects of hur-582 ricanes Maria and Irma on forest birds of Puerto Rico. BioRxiv, 578336. 583 https://doi.org/10.1101/578336 584 Luviano, N., Villa-Galaviz, E., Boege, K., Zaldívar-Riverón, A., & del-Val, E. (2018). 585 Hurricane impacts on plant-herbivore networks along a successional chronose-586 quence in a tropical dry forest. Forest Ecology and Management, 587 426(September 2017), 158–163. https://doi.org/10.1016/j.foreco.2017.09.011 588 Mack, K. M. L., & Rudgers, J. A. (2008). Balancing multiple mutualists: Asymmetric 589 interactions among plants, arbuscular mycorrhizal fungi, and fungal endophytes. 590 Oikos, 117(2), 310–320. https://doi.org/10.1111/j.2007.0030-1299.15973.x 591 119 Mahmood, S. (2003). Colonisation of spruce roots by two interacting ectomycorrhizal 592 fungi in wood ash amended substrates. FEMS Microbiology Letters, 221(1), 81–593 87. https://doi.org/10.1016/S0378-1097(03)00166-6 594 Mao, L., Pan, J., Jiang, S., Shi, G., Qin, M., Zhao, Z., … Liu, Y. (2019). Arbuscular 595 mycorrhizal fungal community recovers faster than plant community in historical-596 ly disturbed Tibetan grasslands. Soil Biology and Biochemistry, 134(March), 597 131–141. https://doi.org/10.1016/j.soilbio.2019.03.026 598 Martínez-Yrízar, A., Jaramillo, V. J., Maass, M., Búrquez, A., Parker, G., Álvarez-599 Yépiz, J. C., … Sarukhán, J. (2018). Resilience of tropical dry forest productivity 600 to two hurricanes of different intensity in western Mexico. Forest Ecology and 601 Management, 426(February), 53–60. 602 https://doi.org/10.1016/j.foreco.2018.02.024 603 Mohan, V., Nivea, R., & Menon, S. (2015). Evaluation of ectomycorrhizal fungi as 604 potential cio-control agents against selected plant pathogenic fungi. Journal of 605 Academia and Industrial Research, 3(9), 408–412. Retrieved from 606 http://jairjp.com/FEBRUARY 2015/03 MOHAN.pdf 607 Nguyen, N.H., Smith, S., Peay, K., Kennedy, P. (2014). Parsing ecological signal 608 from noise in next generation amplicon sequencing. New Phytologist, 205, 1389–609 1393. https://doi.org/10.1111/nph.12923 610 Nguyen, N. H., Song, Z., Bates, S. T., Branco, S., Tedersoo, L., Menke, J., … Ken-611 nedy, P. G. (2016). FUNGuild: An open annotation tool for parsing fungal com-612 munity datasets by ecological guild. Fungal Ecology, 20, 241–248. 613 https://doi.org/10.1016/j.funeco.2015.06.006 614 Nilsson, R. H., Anslan, S., Bahram, M., Wurzbacher, C., Baldrian, P., & Tedersoo, L. 615 (2019). Mycobiome diversity: high-throughput sequencing and identification of 616 fungi. Nature Reviews Microbiology, 17(2), 95–109. 617 https://doi.org/10.1038/s41579-018-0116-y 618 120 Novais, S., Macedo-Reis, L. E., Cristobal-Peréz, E. J., Sánchez-Montoya, G., Janda, 619 M., Neves, F., & Quesada, M. (2018). Positive effects of the catastrophic Hurri-620 cane Patricia on insect communities. Scientific Reports, 8(1), 1–9. 621 https://doi.org/10.1038/s41598-018-33210-7 622 Oksanen, J.F., Blanchet, G., Friendly, M., Kindt, R., Legendre, R., McGlinn, D., 623 Minchin, P.R, O'Hara, R. B., Simpson, G.L., Solymos, P., Stevens, M.H.H., 624 Szoecs, E., Wagner, H. (2019). vegan: Community Ecology Package. R package 625 version 2.5-4. https://CRAN.R-project.org/package=vegan 626 Parker, G., Martínez-Yrízar, A., Álvarez-Yépiz, J. C., Maass, M., & Araiza, S. (2018). 627 Effects of hurricane disturbance on a tropical dry forest canopy in western 628 Mexico. Forest Ecology and Management, 426(December 2017), 39–52. 629 https://doi.org/10.1016/j.foreco.2017.11.037 630 Peay, K. G., Baraloto, C., & Fine, P. V. A. (2013). Strong coupling of plant and fungal 631 community structure across western Amazonian rainforests. The ISME Journal, 632 7(9), 1852–1861. https://doi.org/10.1038/ismej.2013.66 633 Peay, K. G., Kennedy, P. G., & Talbot, J. M. (2016). Dimensions of biodiversity in the 634 Earth mycobiome. Nature Reviews Microbiology, 14(7), 434–447. 635 https://doi.org/10.1038/nrmicro.2016.59 636 Põlme, S., Bahram, M., Jacquemyn, H., Kennedy, P., Kohout, P., Moora, M., … 637 Tedersoo, L. (2018). Host preference and network properties in biotrophic plant–638 fungal associations. New Phytologist, 217(3), 1230–1239. 639 https://doi.org/10.1111/nph.14895 640 Purahong, W., Krüger, D., Buscot, F., & Wubet, T. (2016). Correlations between the 641 composition of modular fungal communities and litter decomposition-associated 642 ecosystem functions. Fungal Ecology, 22, 106–114. 643 https://doi.org/10.1016/j.funeco.2016.04.009 644 121 Ramirez, K. S., Geisen, S., Morriën, E., Snoek, B. L., & van der Putten, W. H. (2018). 645 Network Analyses Can Advance Above-Belowground Ecology. Trends in Plant 646 Science, 23(9), 759–768. https://doi.org/10.1016/j.tplants.2018.06.009 647 Roxburgh, Stephen; Shea, Katriona; Wilson, B. (2004). The intermediate disturbance 648 hypothesis: patch dynamics and mechanisms of species coexistence. Ecology, 649 85(2), 359–371. 650 Salazar, L. F., Nobre, C. A., & Oyama, M. D. (2007). Climate change consequences 651 on the biome distribution in tropical South America. Geophysical Research Let-652 ters, 34(9), 1–6. https://doi.org/10.1029/2007GL029695 653 Salazar, A., Baldi, G., Hirota, M., Syktus, J., & McAlpine, C. (2015). Land use and 654 land cover change impacts on the regional climate of non-Amazonian South 655 America: A review. Global and Planetary Change. Elsevier B.V. 656 https://doi.org/10.1016/j.gloplacha.2015.02.009 657 Schloss, P. D., Westcott, S. L., Ryabin, T., Hall, J. R., Hartmann, M., Hollister, E. B., 658 … Weber, C. F. (2009). Introducing mothur: Open-source, platform-independent, 659 community-supported software for describing and comparing microbial communi-660 ties. Applied and Environmental Microbiology, 75(23), 7537–7541. 661 https://doi.org/10.1128/AEM.01541-09 662 Steidinger, B., Crowther, T., Liang, J., Van Nuland, M., Werner, GD Reich, P., 663 Nabuurs, G., … Peay, K. (2019). Climatic controls of decomposition drive the 664 global biogeography of forest tree symbioses. Nature, 569, 404–408. 665 https://doi.org/10.1038/s41586-019-1128-0 666 Stone, R.P., & Hilborn, H. (2000). Universal soil loss equation (USLE). Agricultural 667 engineering. 668 Tedersoo, L., Jairus, T., Horton, B. M., Abarenkov, K., Suvi, T., Saar, I., & Koljalg, U. 669 (2008). Strong host preference of ectomycorrhizal fungi in a Tasmanian wet scle-670 rophyll forest as revealed by DNA barcoding and taxon-specific primers. New 671 122 Phytologist, 180(November), 479–490. https://doi.org/10.1111/j.1469-672 8137.2008.02561.x 673 Tedersoo, L., Sadam, A., Zambrano, M., Valencia, R., & Bahram, M. (2010). Low 674 diversity and high host preference of ectomycorrhizal fungi in Western Amazo-675 nia, a neotropical biodiversity hotspot. ISME Journal, 4(4), 465–471. 676 https://doi.org/10.1038/ismej.2009.131 677 Tedersoo, L., Bahram, M., Polme, S., Koljalg, U., Yorou, N. S., Wijesundera, R., … 678 Abarenkov, K. (2014). Global diversity and geography of soil fungi. Science, 679 346(6213). https://doi.org/10.1126/science.1256688 680 Toju, H., Guimarães, P. R., Olesen, J. M., & Thompson, J. N. (2014). Assembly of 681 complex plant-fungus networks. Nature Communications, 5(May), 1–7. 682 https://doi.org/10.1038/ncomms6273 683 Truong, C., Gabbarini, L. A., Corrales, A., Mujic, A. B., Escobar, J. M., Moretto, A., & 684 Smith, M. E. (2019). Ectomycorrhizal fungi and soil enzymes exhibit contrasting 685 patterns along elevation gradients in southern Patagonia. New Phytologist, 686 222(4), 1936–1950. https://doi.org/10.1111/nph.15714 687 Vargas, R., Hasselquist, N., Allen, E. B., & Allen, M. F. (2010). Effects of a hurricane 688 disturbance on aboveground forest structure, arbuscular mycorrhizae and 689 belowground carbon in a restored tropical forest. Ecosystems, 13(1), 118–128. 690 https://doi.org/10.1007/s10021-009-9305-x 691 Walsh, K. J. E., Mcbride, J. L., Klotzbach, P. J., Balachandran, S., Camargo, S. J., 692 Holland, G., … Sugi, M. (2016). Tropical cyclones and climate change. Wiley In-693 terdisciplinary Reviews: Climate Change, 7(1), 65–89. 694 https://doi.org/10.1002/wcc.371 695 Wickham, H. (2016). ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag 696 New York. 697 123 Zimmerman, J. K., Everham, E. M., Waide, R. B., Lodge, D. J., Taylor, C. M., & 698 Brokaw, N. V. L. (1994). Responses of tree species to hurricane winds in sub-699 tropical wet forest in Puerto-Rico: Implications for tropical tree life histories. 700 Journal of Ecology, 82(4), 911–922. https://doi.org/10.2307/2261454 701 Additional files 702 Characteristics of plots 703 The most common species were Achatocarpus gracilis (Achatocarpaceae), 704 Apoplanesia paniculata (Fabaceae subfam. Papilionoideae), Bonelia macrocarpa 705 subsp. pungens (Primulaceae), Caesalpinia eryostachis (Fabaceae subfam. Caesal-706 pinioideae), Cordia alliodora (Boraginaceae), Guapira petenensis (Nyctaginaceae), 707 Heliocarpus pallidus (Tiliaceae), Piptadenia obliqua (Fabaceae), and Thouinia pauci-708 dentata (Sapindaceae). 709 124 710 Figure S1. Abundance from the most common plant species in each plot. Ectomycor-711 rhizal (ECM) hosts are indicated in bold. 712 125 713 Figure S2. Soil characteristics in each plot. A) Electric conductivity (EC); B) Percent-714 age of organic matter (OM); C) Percentage of total carbon; D) Percentage of total 715 and estimated nitrogen; E) Percentage of total phosphorus; F) Percentage of mineral 716 nutrients: NH4, NO3, PO4; G) Potential of hydrogen (pH). 717 718 126 719 Figure S3. Cluster analysis by Ward distance from soil, environmental and vegetation 720 characteristics of plots. Rectangles separate plots in three main groups. 721 722 Table S1. Trees and shrubs alive from each plot 723 SPECIES A250 A500 B250 EC650 T450 T700 T1000 T2600 T2800 Total Guapira petenen- sis 13 2 12 15 37 35 7 31 18 170 Thouinia pauciden- tata 9 5 7 16 30 19 24 8 7 125 Piptadenia obliqua 2 18 14 2 0 9 25 9 0 79 Bonellia macrocar- pa subsp. pungens 13 1 1 4 3 25 19 5 2 73 Cordia alliodora 11 12 4 7 3 14 11 2 7 71 Heliocarpus pa- llidus 8 5 7 2 13 6 12 7 6 66 Achatocarpus gra- cilis 7 8 0 8 14 9 4 5 7 62 Apoplanesia pani- culata 0 0 0 1 38 9 4 8 1 61 Caesalpinia erios- tachys 5 10 2 8 11 2 13 4 5 60 Lonchocarpus mu- tans 6 8 7 2 5 5 9 11 0 53 Croton niveus 0 2 6 7 12 0 3 1 3 34 Caesalpinia pul- cherrima 0 5 18 2 0 0 0 7 0 32 Casearia tremula 1 1 6 8 7 2 5 1 0 31 127 SPECIES A250 A500 B250 EC650 T450 T700 T1000 T2600 T2800 Total Lonchocarpus eriocarinalis 3 2 6 6 0 5 7 2 0 31 Ruprechtia fusca 2 2 9 4 1 6 2 2 0 28 Caesalpinia platy- loba 13 1 10 0 2 0 0 1 0 27 Croton pseudo- niveus 0 1 0 8 6 0 12 0 0 27 Lonchocarpus mi- nor 0 0 0 1 11 7 8 0 0 27 Lysiloma microp- hylla 1 2 1 0 11 4 0 2 6 27 Maclura tinctoria 2 2 0 7 1 13 0 0 2 27 Cordia elaeagnoi- des 1 0 1 0 2 1 11 9 1 26 Trichilia trifolia 1 0 1 8 0 0 5 4 3 22 Caesalpinia scle- rocarpa 1 1 0 2 1 2 11 0 2 20 Spondias purpurea 0 6 2 1 0 0 4 0 2 15 Lonchocarpus ma- gallanesii 0 0 0 0 0 5 9 0 0 14 Recchia mexicana 0 0 0 2 0 7 2 0 3 14 Urera caracasana 0 1 6 0 4 0 3 0 0 14 Bursera simaruba 0 1 0 1 4 1 1 3 2 13 Chloroleucon mangense 0 2 4 0 1 2 1 3 0 13 Ipomoea wolcot- tiana 0 3 2 0 1 2 5 0 0 13 Unknown1 0 0 0 9 0 0 0 0 0 9 Guettarda elliptica 0 0 0 0 1 1 7 0 0 9 Mimosa albida 0 2 1 0 0 0 0 6 0 9 Vitex hemsleyi 0 0 0 0 3 1 5 0 0 9 Cochlospermum vitifolium 0 0 4 0 0 0 3 0 1 8 Leucaena lanceo- lata 3 0 1 0 2 0 1 0 1 8 Brosimum alicas- trum 0 0 0 0 3 2 2 0 0 7 Celosia monos- perma 1 0 0 2 0 4 0 0 0 7 Lonchocarpus sp1 0 4 0 0 0 3 0 0 0 7 128 SPECIES A250 A500 B250 EC650 T450 T700 T1000 T2600 T2800 Total Cordia alba 0 0 0 0 1 2 3 0 0 6 Cordia sp. 0 6 0 0 0 0 0 0 0 6 Jatropha sympeta- la 1 0 0 1 0 0 0 2 2 6 Lonchocarpus sp.2 0 0 0 0 0 6 0 0 0 6 Vitex mollis 0 0 0 0 0 6 0 0 0 6 Unknown2 0 0 0 0 2 0 3 0 0 5 Unknown3 0 0 0 0 0 0 5 0 0 5 Unknown4 5 0 0 0 0 0 0 0 0 5 Caesalpinia co- riaria 0 0 0 0 3 0 1 1 0 5 Erythrina lanata 0 0 1 0 2 2 0 0 0 5 Lonchocarpus sp.3 0 4 0 0 0 1 0 0 0 5 Psidium sartoria- num 0 0 0 1 4 0 0 0 0 5 Sinclairia caducifo- lia 1 1 1 0 2 0 0 0 0 5 Tabebuia rosea 0 0 0 3 0 0 2 0 0 5 Astronium graveo- lens 0 0 0 0 2 0 0 1 1 4 Bursera instabilis 0 0 1 0 2 1 0 0 0 4 Bursera sp. 0 0 4 0 0 0 0 0 0 4 Coccoloba lie- bmannii 0 0 1 2 0 0 1 0 0 4 Styphnolobium protantherum 0 0 0 0 2 1 0 0 1 4 Unknown5 1 0 0 0 0 0 0 0 2 3 Unknown6 0 0 3 0 0 0 0 0 0 3 Amphipterygium adstringens 0 0 3 0 0 0 0 0 0 3 Bursera heterest- hes 0 0 0 0 3 0 0 0 0 3 Capparisdastrum frondosum 0 0 0 0 1 0 2 0 0 3 Cascabela ovata 2 0 0 0 0 0 0 1 0 3 Colubrina hetero- neura 0 0 1 0 2 0 0 0 0 3 Colubrina triflora 0 0 0 0 0 2 0 1 0 3 129 SPECIES A250 A500 B250 EC650 T450 T700 T1000 T2600 T2800 Total Croton suberosus 0 0 0 0 3 0 0 0 0 3 Forchhammeria pallida 0 0 0 0 1 0 2 0 0 3 Guazuma ulmifolia 0 0 0 1 1 0 1 0 0 3 Jatropha malacop- hylla 0 0 1 0 2 0 0 0 0 3 Randia aculeata 0 0 0 0 3 0 0 0 0 3 Bursera excelsa 0 0 2 0 0 0 0 0 0 2 Casearia nitida 0 0 1 0 0 0 0 1 0 2 Ceiba aesculifolia 0 0 0 1 0 0 0 1 0 2 Coccoloba barba- densis 0 0 0 2 0 0 0 0 0 2 Coccoloba sp. 0 0 0 1 1 0 0 0 0 2 Comocladia ma- crophylla 0 0 0 0 2 0 0 0 0 2 Unknown 7 0 0 0 0 0 2 0 0 0 2 Jacaratia mexica- na 0 0 0 0 0 0 2 0 0 2 Lagrezia monos- perma 0 0 0 0 0 0 2 0 0 2 Lonchocarpus constrictus 0 0 0 2 0 0 0 0 0 2 Lonchocarpus sp.3 0 0 0 0 0 0 0 0 2 2 Lonchocarpus sp.4 0 0 0 0 0 0 0 0 2 2 Machaonia acumi- nata 0 0 0 0 1 1 0 0 0 2 Piranhea mexicana 0 1 0 0 0 0 0 0 1 2 Sideroxylon ste- nospermum 2 0 0 0 0 0 0 0 0 2 Unknown8 0 0 0 0 0 0 0 0 1 1 Unknown9 0 0 1 0 0 0 0 0 0 1 Unknown10 0 1 0 0 0 0 0 0 0 1 Unknown11 0 0 1 0 0 0 0 0 0 1 Unknown12 0 0 0 1 0 0 0 0 0 1 Unknown13 0 0 1 0 0 0 0 0 0 1 Unknown14 0 0 1 0 0 0 0 0 0 1 130 SPECIES A250 A500 B250 EC650 T450 T700 T1000 T2600 T2800 Total Unknown15 0 0 0 1 0 0 0 0 0 1 Unknown16 0 0 0 0 1 0 0 0 0 1 Aralia excelsa 0 0 0 1 0 0 0 0 0 1 Bourreria sp 0 0 0 0 0 0 1 0 0 1 Unknown 17 0 1 0 0 0 0 0 0 0 1 Brongniartia sp. 1 0 0 0 0 0 0 0 0 1 Capparis sp. 0 0 0 1 0 0 0 0 0 1 Capsicum annuum 0 1 0 0 0 0 0 0 0 1 Croton sp. 0 1 0 0 0 0 0 0 0 1 Euphorbia tan- quahuete 1 0 0 0 0 0 0 0 0 1 Ficus cotinifolia 0 0 0 0 0 0 0 0 1 1 Heliocarpus sp. 0 0 0 0 0 0 0 1 0 1 Jatropha chame- lensis 0 0 1 0 0 0 0 0 0 1 Lonchocarpus sp.5 0 0 0 0 0 1 0 0 0 1 Opuntia excelsa 0 0 1 0 0 0 0 0 0 1 Pachycereus pec- ten-alboriginum 0 0 0 1 0 0 0 0 0 1 Pterocarpus orbi- culatus 0 0 0 0 0 0 1 0 0 1 Roseodendron donell-smithii 0 0 0 0 0 1 0 0 0 1 Senna atomaria 0 0 0 0 0 0 0 1 0 1 Tabernaemontana donnell-smithii 1 0 0 0 0 0 0 0 0 1 Unknown18 0 0 0 0 0 0 0 0 0 0 Cordia gerascant- hus 0 0 0 0 0 0 0 0 0 0 724 Table S2. Biotic and abiotic variables taken in each plot 725 Variable A250 A500 B200 EC650 T450 T700 T1000 T2600 T2800 Dead trees 13 20 13 13 19 24 31 12 16 Fall alive trees 17 21 26 31 30 27 47 20 11 131 Variable A250 A500 B200 EC650 T450 T700 T1000 T2600 T2800 Stand up alive trees 102 114 142 120 275 246 217 121 81 T (°C) 31.8 30.2 32.6 32.5 35.6 36 33.6 32.9 32.5 Humidity 88.4 87.4 70.43 71 65.4 66.5 66.6 80.9 85.2 Light (lux) 2139.23 808.44 5947.60 9546.31 10052.72 9814.43 8085.28 2420.87 1511.18 Litter (cm) 7.5 6 11.25 4 7.5 3.85 6.5 4 3.5 Total plant richness 52 36 43 43 59 47 45 33 28 ECM host richness 4 4 4 8 5 5 6 5 3 ECM host abundance 25 14 28 39 91 64 25 48 26 Slope (°) 26 32 32 28 18 35 17 24 45 Tree den- sity 0.29 0.33 0.42 0.37 0.76 0.68 0.66 0.35 0.23 Erosion (Kg*ha-) 3.44 4.97 3.55 2.80 1.31 4.18 1.19 3.95 12.25 pH 7.5 6.7 7.1 7.4 6.9 7.2 6.6 7 7 CE (dS*m- ) 0.19 0.1 0.15 0.12 0.1 0.19 0.07 0.09 0.08 OM (%) 12.3 6.2 10.5 8 11.5 12.9 12.4 7.5 12.2 Nt (%) 0.5 0.3 0.4 0.4 0.4 0.6 0.5 0.4 0.5 PO4 (ppm) 18 105 25 25 19 27 22 29 64 Pt (%) 0.07 0.05 0.06 0.07 0.06 0.11 0.P09 0.14 0.11 NO3 (ppm) 71 59 70 45 45 99 29 45 83 NH4 (ppm) 19 24 19 17 18 19 16 14 24 Ct (%) 6.9 3.6 5.5 4.8 6.3 7.5 7.3 4.2 6.5 Abbreviations: Ct= total carbon; EC=electric conductivity; ECM=ectomycorrhizal; Nt= 726 total nitrogen; OM=organic matter; Pt=total phosphorus; T=temperature. 727 728 Table S3. NextEra adapters with primers sequences used in the PCR multiplex 729 132 Sequence Name Target fungal group tcgtcggcagcgtcagatgtgtataagagacagcatcgatgaagaa- cgcag ITS3NGS 1 Universal tcgtcggcagcgtcagatgtgtataagagacagcaacgatgaagaa- cgcag ITS3NGS 2 Phylum Chytri- diomycota cgtcggcagcgtcagatgtgtataagagacagcaccgatgaagaa- cgcag ITS3NGS 3 Order Sebacina- les tcgtcggcagcgtcagatgtgtataagagacagcatcgatgaagaa- cgtag ITS3NGS 4 Subphylum Glomeromycoti- na tcgtcggcagcgtcagatgtgtataagagacagcatcgatgaagaa- cgtgg ITS3NGS 5 Order Sordaria- les gtctcgtgggctcggagatgtgtataagagacagtcctscgcttattgata- tgc ITS4NG Reverse Primers sequences are bold, the rest correspond to NextEra sequences. 730 731 Table S4. Coordinates of the plots 732 Latitude Longitude 19.504935° N -105.03988° W 19.505019° N -105.040835° W 19.498762° N -105.041002° W 19.502078° N -105.040591° W 19.501029° N -105.044208° W 19.503951° N -105.044596° W 19.504644° N -105.047465° W 19.50953° N -105.040987° W 19.509748° N -105.039097° W 733 734 133 Table S5. Characteristics of the arbuscular and ectomycorrhizal networks in same plots from two years (2016 and 2017). 735 2016 2017 Nestedness Modularity Nestedness Modularity Plot Net NODF Z-value Mean P-value WNODF NODF Z-value Mean P-value WNODF A250 ECM 5.341 0 5.341 1 8.76 NA 2.204 -2.84 4.161 0.02 5.239 0.483 AM 22.58 0 22.58 1 0 NA 0 0 0 1 0 NA A500 ECM 0 0 0 1 0 NA 9.909 0.804 8.463 0.5 11.261 0.792 AM NA NA NA NA NA NA 5.438 0.44 5.109 0.69 0.906 0.813 B200 ECM 0 0 0 1 0 NA 9.23 -0.73 10.82 0.38 8.461 0.765 AM 0 0 0 1 0 NA 0 0 0 1 0 NA T1000 ECM NA NA NA NA NA NA 7.911 0.884 6.923 0.36 8.104 0.566 AM NA NA NA NA NA NA 0 0 0 1 0 NA T450 ECM 14.28 0 14.28 1 5.357 NA 5.031 -1.583 7.109 0.15 5.87 0.641 AM 0 0 0 1 0 NA 0 0 0 1 0 NA EC650 ECM 12.87 0.273 12.501 1 17.82 0.401 10.484 0.419 10.03 0.79 15.46 0.374 AM 20 0.261 20 1 0 NA NA NA NA NA NA NA T2800 ECM 0 0 0 1 0 NA 0 0 0 1 0 NA AM NA NA NA NA NA NA NA NA NA NA NA NA T2650 ECM 0 0 0 1 7.692 NA 5.66 0 5.66 1 0 NA AM NA NA NA NA NA NA NA NA NA NA NA NA T700 ECM 0 0 0 1 0 NA 0 0 0 1 0 NA AM 0 0 0 1 NA NA NA NA NA NA NA NA 134 736 Figure S4. Co-occurrrence network with of all known fungal guilds with minimum 737 abundance of 20 sequences using Pearson and Spearman correlation, 100 bootstrap 738 iterations with Benajmin-Hochberg test correction for the P-value threshold 0.05. A) in 739 2016, N=284 OTUs. B) in 2017, N=1029 OTUs. 740 Hurricane effect on fungal community 741 We obtained 2,003,944 (9.82%) quality-filtered sequences from initial 20,402,884 742 reads; after chimeras filtering 1,268,036 sequences were left that clustered in to 3763 743 OTUs; after the subtraction from control sequences, we obtained 3625 OTUs. Eight 744 135 samples contained less than 200 sequences (i.e. 28-193 sequences), four of them 745 belong to Ruprechtia fusca rhizospheres, and the rest were from 2016, following the 746 hurricane. From 3625 OTUs, 996 belonged to Ascomycota, 704 to Basidiomycota, 747 251 to Glomeromycota, 69 to Chytridiomycota, 34 to Mucoromycota, and 16 to Mor-748 tierellomycota. They belong to 36 class, 91 orders, 224 families, and 424 genera. The 749 first three OTU richest classes were Agaricomycetes (626 OTUs), Dothideomycetes 750 (345), and Eurotiomycetes (262); the OTU richest orders were Agaricales (350), 751 Glomerales (182), and Pleosporales (164); the richest families were Glomeraceae 752 (175), Aspergillaceae (124), and Agaricaceae (121); the richest genera were Asper-753 gillus (72), Geastrum (46), and Penicillium (45). 754 High disturbance plots were dominated by Ascomycota. Glomeromycota were domi-755 nant in recovery disturbance plots followed by low plots but remained rare in high 756 disturbance plots (Figure S5). Agaricomycetes was the most OTU rich class across 757 all plots; Dothideomycetes, Eurotiomycetes, and Sordariomycetes were common in 758 sites with high disturbance and with less richness in low disturbance; Glomeromy-759 cetes were absent in plots with high disturbance in both years, and were absent in 760 low disturbance in 2016, recovering diversity in 2017 (Figure S6). Agaricales, Botry-761 osphaeriales, Cantharellales, Eurotiales, Hypocreales, and Pleosporales were the 762 common orders in all plots and years; Thelephorales was abundant in low and high 763 disturbance but not in recovery, in constrast with Glomerales that was found in low 764 and recovery plots (Figure S7). The most common fungal guild in the high disturb-765 ance sites were saprotrophs, whereas AM was absent in plots with high disturbance. 766 Ectomycorrhizal fungi had more richness in plots with low and high disturbance, but 767 not in recovery plots (Figure S5). We found significant abundance differences in 768 guilds between plots and years (chi-squared= 38, P= 3.7e-7, df=5). Ordination analy-769 sis showed communities are arranged by year and with some hosts (Figure S10). 770 In both years, Ascomycota was the Phylum with more species, and in 2016 Basidio-771 mycota had more abundance (Figure S8). 772 136 773 Figure S5. Fungal community in different disturbance level in 2016 and 2017. A) 774 Richness and relative richness of each phylum, B) Abundance and relative abun-775 dance of each phylum, C) Richness and relative richness of each guild, D) Abun-776 dance and relative abundance of each guild. 777 778 137 779 Figure S6. A) Class richness and relative richness in each year and disturbance plots, B) Class abundance and relative abun-780 dance in each year and disturbance plots 781 138 782 783 Figure S7. A) Order richness and relative richness in each year and disturbance plots, B) Order abundance and relative abun-784 dance in each year and disturbance plots 785 786 139 Figure S8. Boxplot showed no differences in richness and abundance in different classifica- tion level depending on disturbance degree and year. 140 Figure S9. Ectomycorrhizal fungal diversity with the significative predictors: A) plant species, B) year of sampling, C) soil ammonium, D) soil temperature, and E) plant richness. Abbrevia- tions: Ach_gracilis= Achatocarpus gracilis, Apo_paniculata= Apoplanesia paniculate, Coc_liebmanii= Coccoloba liebmanii, Gua_petenensis= Guapira petenensis, Lon_eriocarinalis= Lonchocarpus eriocarinalis, Lon_sp= Lonchocarpus sp, Rup_fusca= Ruprechtia fusca, Sid_excelsium= Syderoxylon excelsium, Tho_paucidentata= Thouinia paucidentata. 141 Figure S10. NMDS ordination plot of the fungal communities. Open figures correspond to samples of 2016 and solid figures belong to 2017. Each figure belongs to the three plot groups and colors represent host species. Black figures were soil samples. Abbreviations: Abu_H=abundance of ectomycorrhizal hosts, CE=electric conductivity, Ct= total Carbon, Densitree= tree density, MO=organic matter, Nt= total Nitrogen, Pt= total Phosphorus, Rich= plant richness, Rich_H=richness of ectomycorrhizal hosts, Temp= temperature. Ach_gracilis=Achatocarpus gracilis, Apo_paniculata=Apoplanesia paniculata, Coc_liebmanii=Cocoloba liebmanii, Gua_petenensis=Guapira petenensis, Lon_constrictus=Lonchocarpus constrictus, Lon_eriocarinalis=L. eriocarinalis, Lon_sp= Lon- chocarpus spp. Rup_fusca=Ruprechtia fusca, Sid_excelsium= Sideroxylon excelsium, Tho_paucidentata=Thouinia paucidentata. 142 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Mycorrhizal community in each host (2016) Glomeraceae_spOtu0633 Tomentella_spOtu0579 Tomentella_spOtu0578 Tomentella_lateritiaOtu0562 Chloridium_spOtu3537 Chloridium_spOtu3513 Cortinarius_spOtu1850 Glomeromycota_spOtu1633 Glomeraceae_spOtu0929 Glomeraceae_spOtu0657 Tomentella_spOtu0534 Glomeraceae_spOtu0506 Glomeraceae_spOtu0405 Clavulina_spOtu0221 Cortinariaceae_spOtu1100 Tomentella_spOtu0564 Glomeraceae_spOtu1118 GS24_spOtu1799 Tomentella_spOtu0577 Glomeraceae_spOtu1089 Tomentella_spOtu0580 Cenococcum_geophilumOtu2754 Glomeraceae_spOtu0412 Glomeraceae_spOtu1049 Glomeraceae_spOtu0807 Tomentella_spOtu0535 Entoloma_spOtu0655 Tomentella_spOtu0533 Helvella_spOtu1623 Entoloma_tenuissimumOtu0590 Tomentella_spOtu0523 Glomeraceae_spOtu1585 Glomeraceae_spOtu1056 Glomeraceae_spOtu0698 Tomentella_spOtu0585 Russula_spOtu0246 Russula_spOtu0389 Tomentella_spOtu0532 Russula_spOtu0144 Sebacina_spOtu1167 Tomentella_spOtu0573 Clavulina_spOtu0163 Russula_spOtu0140 Clavulina_spOtu0188 Clavulina_spOtu0231 Clavulinaceae_spOtu0332 Tomentella_spOtu0489 Inocybe_spOtu1506 Clavulina_spOtu0169 Tomentella_spOtu0537 A 143 Figure S11. A) Mycorrhizal community in each host in 2016 and B) 2017. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Mycorrhizal community in each host (2017) Glomeraceae_spOtu0506 Clavulina_spOtu0163 Tomentella_spOtu0594 Glomeraceae_spOtu1475 Laccaria_laccataOtu0914 Entolomataceae_spOtu0898 Entolomataceae_spOtu1404 Glomeraceae_spOtu1364 Glomeraceae_spOtu1052 Entoloma_tenuissimumOtu0590 Glomeraceae_spOtu1011 Glomeromycota_spOtu1806 Glomeraceae_spOtu1173 Entoloma_spOtu1085 Clavulina_spOtu0169 Helvella_spOtu1304 Tomentella_spOtu0532 Russula_spOtu0144 Glomeraceae_spOtu1059 Glomeraceae_spOtu0929 Cortinariaceae_spOtu0045 Glomeraceae_spOtu1053 Glomeraceae_spOtu0941 Amanita_spOtu1151 Chloridium_spOtu3513 Glomeraceae_spOtu1118 Tomentella_spOtu0585 Entoloma_spOtu0989 Glomeraceae_spOtu0698 Tomentella_spOtu0523 Russula_spOtu0389 GS24_spOtu1780 Clavulina_spOtu0159 Lyophyllum_spOtu0979 Sebacina_spOtu1167 Ceratobasidium_ramicolaOtu02 92 B 144 DISCUSIÓN La presente tesis es la primera aportación para conocer a los hongos ecto- micorrízicos junto con sus hospederos del bosque tropical caducifolio (BTC); ade- más de entender el efecto de los huracanes sobre las comunidades de hongos del suelo y sobre la red micorrízica en el BTC. Nuestros resultados indican que el hu- racán Patricia no sólo afectó a la vegetación generando un considerable aporte de materia orgánica y nutrimentos al suelo, sino también a las comunidades fúngicas. En el primer capítulo, se publicó qué plantas del orden Caryophyllales son los principales hospederos ectomicorrízicos en el bosque tropical cafucifolio; estos son datos previos al huracán. En el segundo capítulo se analizó la sucesión, resi- liencia y resistencia de las comunidades de hongos en el suelo tras el paso del huracán. Mientras que, en el tercer capítulo encontramos que la diversidad de los hongos en la rizósfera se vio reducida por el paso del huracán Patricia. Dos años después del evento climático, la diversidad rizosférica, así como la conectividad de la red micorrízica fueron altamente resilientes. Los capítulos 1 y 2 presentan datos previos al huracán: en el primero se descifró la simbiosis ectomicorrízica del BTC y en el segundo se compararon las comunidades del suelo a través del tiempo. En el capítulo 1 se muestreó del 2012 al 2015 bajo el supuesto de que las plantas de la familia Fabaceae –familia que presenta la mayor diversidad en el BTC– serían las principales hospederas ecto- micorrízicas tal como ha sido evidenciado en otros ecosistemas tropicales (Henkel et al., 2002; 2012; Smith et al., 2011). Los resultados rechazaron esta hipótesis puesto que los principales hospederos ectomicorrízicos del BTC pertenecieron al orden Caryophyllales: Achatocarpaceae, Nyctaginaceae y Polygonaceae. Achato- carpus gracilis es la primera especie de Achatocarpaceae que se conoce como asociada a hongos ectomicorrízicos. En general, estas plantas tienen baja abun- dancia en el bosque, con excepción de Guapira petenensis, por lo que los hospe- deros ectomicorrízicos se encuentran inmersos en una matriz de plantas con in- teracción micorrízica arbuscular. Esto genera que los hongos ectomicorrízicos pa- ra encontrar sus hospederos dispersos, 1) deben tener esporas de larga vida, dis- 145 persión por viento o vegetativa, para sobrevivir en un nicho tan restringido; 2) además tienen que ser altamente competitivos con hongos saprotrofos, patógenos y otros ectomicorrízicos; y 3) deben ampliar su nicho mutualista desarrollando nuevas simbiosis oportunistas o especializarse en un solo linaje de plantas (Alva- rez-Manjarrez et al., 2018). En general, los hongos ectomicorrízicos presentan una correlación negativa a la distancia al ecuador (Bahram et al., 2013), donde la mayoría de los organis- mos tienen mayor diversidad. No obstante, existen bosques tropicales donde hay una alta diversidad de hongos ectomicorrízicos (e.g. Smith et al., 2011; Peay et al., 2015). La baja dominancia de los hospederos ectomicorrízicos en los bosques tro- picales caducifolios explica la baja diversidad de hongos ectomicorrízicos. En otros bosques tropicales donde también se presenta baja dominancia de hospederos, han llegado a la misma conclusión (Tedersoo et al., 2010; Roy et al. 2016). Esto podría deberse en parte a una alta especificidad hacia su hospedero, por lo que su nicho se ve reducido al haber baja densidad de hospederos. Además, se identificaron 19 especies de hongos que se encuentran for- mando ectomicorrizas con 19 especies vegetales. La mayoría de estos hongos no habían sido secuenciados previamente por lo que la similitud deposistadas en GenBank y UNITE era menor al 90% en casi todas las especies –aspecto que po- ne en evidencia lo novedoso del trabajo, y en particular, la diversidad del ecosis- tema–. Esta tesis tiene la hipótesis de que el bosque tropical caducifolio es un sitio de alta diversidad fúngica, donde un gran número de especies no han sido descri- tas. Para la descripción de las nuevas especies realizamos muestreo de esporo- mas durante del 2012 al 2017 en la época de lluvias. Los esporomas recolectados nos llevaron a describir al hongo ectomico- rrízico Tomentella brunneoincrustata M. Villegas & Contreras-Pachecho 2016. És- ta es la primera especie descrita para el clado encontrado en el caribe y la costa del Pacífico mexicano asociado a Nyctaginaceae subfamilia Pisonieae (Anexo 1; Alvarez-Manjarrez et al., 2016). El color de esta Tomentella es más oscuro que el resto de los miembros del género y forma ectomicorrizas con Pisonia y otros 146 miembros de la misma subfamilia. De igual forma, durante el muestreo se recolec- taron esporomas de Scytinopogon, los datos morfológicos y filogenéticos nos indi- caron que se trata de una especie nueva, a la cual se nombró como Scytinopogon minisporus J. Alvarez-Manjarrez, M. Villegas & R. Garibay-Orijel 2019. Adicional- mente los análisis filogenéticos indican que, tanto Scytinopogon como su grupo hermano Trechispora, son parafiléticos y se encuentran mal clasificados en Clava- riaceae, cuando deberían ser reacomodados en Hydnodontaceae (Trechisporales; Desjardin & Perry 2015). Tomentella y Scytinopogon son géneros que necesitan una corrección taxonómica, ya que ambos son grupos parafiléticos: Tomentella quien es grupo hermano de Thelephora, debería ser unificado en Thelephora pues los caracteres morfológicos –Tomentella es resupinado y Thelephora es teleforoi- de o clavarioide– no se sostienen con las filogenias. Este es el mismo caso para Scytinpogon (clavarioide) y Trechispora (resupinado), que tendrían que ser sino- nimizados en Trechispora. El carácter resupinado es un carácter ancestral en los Agaricomycetes (Varga et al., 2019), por lo que es altamente probable que en otros clados también haya este mismo problema taxonómico. También se describieron dos especies del género Thelephora: T. versatilis Ramírez-López & M. Villegas 2015 y T. pseudoversatilis Ramírez-López & M. Vi- llegas 2015 (Ramírez-López et al., 2015), del cual se determinó que Guapira pete- nensis es hospedero de T. versatilis. Adicionalmente, las secuencias de las ecto- micorrizas coincidieron con esporomas de Clavulina muestreados antes del 2012, lo cual determinó que Achatocarpus gracilis es el hospedero de la nueva especie Clavulina subtilis M. Villegas, Garibay-Orijel & Ramírez-López (Villegas et al., en preparción). Por otro lado, en el segundo capítulo abordamos la hipótesis que establecía que al aumentar los nutrientes en el suelo, la diversidad de hongos incrementaría. El suelo sí aumentó en C, N y P, inmediatamente después del huracán lo cual ge- neró que la tasa C:N y C:P decreciera; estos datos corresponden con los resulta- dos de Gavito y colaboradores (2018) para el mismo huracán. Para la siguiente temporada de lluvias, donde la tasa C:N había vuelto a aumentar, hubo una dismi- 147 nución en la diversidad fúngica; y los posteriores muestreos siguieron el mismo patrón. La baja tasa de C:N afecta la mineralización de C, N y P en el suelo (Mooshammer et al., 2012) y la descomposición incrementa (Britton et al., 2018). Además, el cambio en la composición de la comunidad pueden alterar la este- quiometría de C:N:P (Heuck et al., 2015). Que la tasa de C:P después del huracán bajara drásticamente (< 37.5) y alcanzara meses después niveles más altos (113- 130) que previo al huracán (< 75) puede explicar el decremento de diversidad de hongos que hubo con el tiempo. Después del huracán Patricia se depositaron 17.8 Mg ha-1 de biomasa, cuando de manera natural se aporta de 3.2 - 4.2 Megagramos por hectárea (Mg ha-1; Parker et al., 2018). Durante el 2016 se presentaron lluvias atípicas prolon- gando la época de lluvias, mientras que en 2017 la precipitación volvió a su tem- poralidad normal, i.e. 730 mm anuales (valores de estación meteorológica). En los BTC la descomposición de la materia orgánica está mediada por la cantidad de agua disponible (Anaya et al., 2007, 2012). Gavito et al. (2018) encontraron que la tasa de descomposición aumentó durante el huracán Patricia, lo cual coincide con la alta diversidad de hongos. Durante la descomposición de la materia orgánica se conoce que existe una sucesión de especies. Antes de Patricia nosotros encontramos mayor diversidad del phylum Ascomycota, pero posteriormente encontramos mayor diversidad de Basidiomycota. Al inicio de la descomposición, las especies más abundantes son aquellas que pueden aprovechar las moléculas de baja complejidad, tales como la glucosa y otros carbohidratos. Las especies fúngicas que suelen estar en estas primeras etapas son algunos Ascomycota, mientras que en la descomposición tardía aparecen los Basidiomycota (Purahong et al., 2016). Esto se debe al poder enzimático que tienen estos últimos con capacidad para degradar compuestos difíciles de despolimerizar –como la lignina– (Voříšková & Baldrian, 2013) y que muchos microorganismos no pueden llevar a cabo. Los huracanes promueven ma- teria orgánica rica en nutrientes, ya que las plantas no reabsorben N o P antes del huracán. Esto confiere que la materia orgánica sea un sustrato del cual muchos 148 organismos se puedan alimentar. Nuestro muestreo demostró no sólo que hubo sucesión de especies durante la descomposición de la materia orgánica, siendo más abundantes los hongos con gran poder enzimático justo después del huracán; sino que también había alta dominancia de tres especies antes del huracán, y con el paso del huracán la dominancia cayó, siendo resiliente en los consecuentes muestreos. Antes del huracán encontramos 519 OTUs, después del huracán cada muestreo secuenció especies que no se compartieron con los demás muestreos. Esto indicó que hubo sucesión sin recambio de especies; es decir, a pesar de que se encontraron especies que no se detectaron antes de Patricia, los 519 OTUs iniciales seguían siendo parte de la comunidad después del huracán (aunque no estuvieran todas presentes en todos los muestreos). La recuperación del sistema es lenta puesto que según la hipótesis de Svoboda y Henry (1987): la incorpora- ción de nuevas especies no desplazan a la comunidad inicial. Esto predice un sitio inhóspito para el establecimiento, por lo que no hay remplazo ni competencia. Efectivamente nuestro muestreo fue en la cumbre de un lomerío –las cuales Par- ker y colaboradores (2017) reportan como las zonas con mayor afección por el huracán–. El daño en la vegetación generó que el dosel se perdiera, lo cual invo- lucra mayor temperatura y cantidad de luz en el suelo, además de pérdida de hu- medad. Estos factores representan un ambiente hostil para el establecimiento de los hongos del suelo. Por otro lado, se encontraron 105 especies de hongos del suelo que esta- ban presentes antes del huracán y que fueron capaces de sobrevivir, fluctuando su abundancia, al impacto de Patricia. Se considera que la comunidad es persis- tente cuando hay sobrevivencia en el tiempo de alguno de los elementos del sis- tema. En este caso, estas 105 especies se encontraron todo el tiempo en el suelo, demostrando que una parte de la comunidad de hongos es presistente y proba- blemente resistente a los huracanes. Los datos del capítulo 3 nos muestran como el huracán Patricia redujo la di- versidad de las comunidades de hongos rizosféricos, y generó la desconexión de 149 la red micorrízica. Al hablar de la rizosfera nos referimos a las inmediaciones de las raíces en el suelo, donde se lleva a cabo la rizodepositación de carbohidratos y proteínas, así como compuestos alelopáticos, que propician el establecimiento de diferentes microorganismos (Hinsinger et al., 2005). La rizodepositación depende en gran medida de la intensidad de la fotosíntesis y factores que la afectan como la concentración de CO2, contenido de N en el suelo, intensidad de luz y humedad del suelo (Pausch & Kuzyakov, 2018). El huracán Patricia hizo que la fotosíntesis declinara (Parker et al., 2018) y el N mineralizado aumentara, lo cual pudo generar una disminución considerable de la rizodepositación. En parte, la reducción de carbohidratos en la rizósfera, junto con las nuevas condiciones inhóspitas pudieron generar la afección a la comunidad rizosférica. En contraste con las comunidades del suelo, los hongos rizosféricos se vie- ron menguados después del huracán y su recuperación fue mayor a dos años. Cabe mencionar que el suelo es el hábitat tanto de hongos de vida libre como de simbiontes obligados, por lo que el suelo es un reservorio de la comunidad poten- cial que colonizará las raíces. Las comunidades rizosféricas pasan por el filtro del hospedero, por lo que la comunidad es principalmente determinada por la identi- dad del hospedero (Tedersoo et al., 2010; Koide et al., 2011). Los resultados muestran que la diversidad rizosférica es explicada por la temperatura, la luz a nivel del suelo y la identidad del hospedero, y no por las ca- racterísticas del suelo. La diversidad rizosférica tiene una correlación negativa con la temperatura y una correlación positiva con la luz. Particularmente, la diversidad de los hongos ectomicorrízicos (ECM) fue explicada por la identidad de la planta hospedera (i.e. Guapira petenensis presentó la mayor diversidad de todas las plantas); además la diversidad tuvo una correlación negativa al amonio al igual que a la temperatura del suelo y una correlación positiva a la riqueza vegetal. Mu- chos otros estudios han determinado que el NH4 en el suelo determina las comu- nidades ectomicorrízicas (e.g. Corrales et al., 2016, 2017; Truong et al., 2019). Estos resultados coinciden con bosques tropicales monodominantes, donde la po- 150 ca disponibilidad de N se debe a los hongos ECM y la necesidad de asociarse a ellos para obtener N (Corrales et al., 2016). Por otro lado, la correlación positiva a la riqueza vegetal nos podría estar indicando la presencia de otros hospederos ectomicorrízicos que todavía se des- conocen. Además, probamos que la abundancia de hospederos en cada parcela no explicaba la diversidad de ECM. En otros estudios se ha corroborado que la composición de la comunidad vegetal es un buen predictor de la comunidad fúngi- ca (Leff et al., 2018; van der Linde et al., 2018). Después del huracán revisamos las raíces de todas las plantas reportadas en el capítulo 1 sin embargo, muchas de ellas no presentaban ectomicorrizas en el 2016. En el 2017, por considerar que no volveríamos a encontrar ectomicorrizas, ya no repetimos el muestreo en esas plantas, lo cual podría ser un error metodológico. Cada hospedero tuvo asociada una diversidad específica con unas pocas especies generalistas. Después del huracán estos generalistas no fueron resisten- tes al disturbio, lo cual generó la pérdida de conexiones entre diferentes especies. En los muestreos del 2012-2014 encontramos que algunas especies como Treme- lloscypha dichroa, Membranomyces sp., Sebacina sp., Thelephora versatilis y To- mentella sp. eran los hongos que conectaban a plantas de diferentes especies. Para el muestreo del 2016, prácticamente estos hongos estuvieron ausentes y fueron remplazados por Tomentella sp. (OTU 537), Clavulina sp. (OTU 169), Inocybe sp. (OTU 1506), Tomentella sp. (OTU 489) y Clavulinaceae, sp. (OTU 332). Dos años después, la comunidad volvió a tener como hongos frecuentes a Tomentella sp. (OTU 537), Thelephora versatilis (OTU 514), Clavulina sp. (OTU 192), Tremelloscypha sp. (OTU 851) y Tomentella sp. (OTU 489). El recambio a las especies que se conocían antes del disturbio nos podría hablar de la resiliencia del ecosistema. Por otro lado, el BTC está dominado por hongos micorrízico arbusculares, siendo los ectomicorrízicos los más focalizados a ciertas plantas. Nuestro mues- treo encontró que las plantas que tienen asociación ectomicorrízica también for- man micorriza arbuscular (AM). Los hongos ectomicorrízicos fueron los más diver- 151 sos en las raíces de las plantas muestreadas, lo cual era esperable al ser todos hospederos ectomicorrízicos. Sin embargo, la riqueza de AM fue alta (i.e. 251 OTUs), considerando que su riqueza mundial se estima en ~1500 OTUs, y la ri- queza reportada para México es de 235 especies morfológicas (Montaño et al., 2012). El único estudio de AM después de un huracán, cuantificó el porcentaje de colonización después del paso del Wilma (Vargas et al., 2010); ellos reportaron un aumento en la colonización de las plantas. Este estudio no midió colonización de ningún tipo de micorrizas, pero la presencia de DNA en la rizósfera fue suficiente para aceverar la interacción. Según los resultados de esta tesis, el porcentaje de colonización no es una medida que nos ayude a entender del todo la interacción. En todos los muestreos se encontraron AM pero de habernos quedado exclusiva- mente con esa información, habría pasado por alto que la conexión entre plantas de diferentes especies se perdió. La predicción del aumento en la modularidad de la red micorrízica en sitios con mayor perturbación fue rechazada. A pesar de la clasificación de las parcelas según el nivel de perturbación, encontramos que el huracán tuvo un efecto devas- tador en todas; después del paso de un huracán tan intenso como Patricia (vientos de 265 km/h) las conexiones de la red micorrízica se perdieron, i.e. no encontra- mos especies micorrízicas compartidas entre los diferentes hospederos. Al respec- to, nuestro muestreo no corrobora si la red micorrízica se rompió entre individuos de la misma especie vegetal. Para el 2017 la red micorrízica se reestableció en- contrando diferentes especies de hongos compartidos entre más de dos especies de plantas. Las redes calculadas en esta tesis encuentran que no hay anidamiento, pe- ro los valores NODF y wNODF fueron más altos en el 2016 que en 2017, al igual que la modularidad; esto coincide con el menor grado de especialización del 2016 comparado contra el 2017 (no hay diferencias significativas en todos los datos). Nuestros datos encuentran que las especies inmediatamente después de Patricia fueron generalistas formando módulos, y con el paso del tiempo las especies es- pecialistas volvieron a establecerse en las raíces. Corroboramos que las especies 152 especialistas son las más vulnerables a la perturbación y su recuperación puede ser lenta (Devictor et al., 2008; VanWallendael 2019). Estas propiedades de las redes nos pueden ayudar a entender su reacción hacia la perturbación. Por ejem- plo, lo que sugiere que las redes modulares son resilientes al disturbio, ya que los cambios sólo afectan dentro del módulo donde ocurrió el disturbio (Gilarranz et al., 2017). Mientras que el anidamiento, al medir la cantidad de especies generalistas que hay, nos indica de forma indirecta la redundancia ecológica de las redes. Esta propiedad también tiene una respuesta a los disturbios, contribuyendo a que el sistema sea resiliente (Bascompte 2009). La comparación entre redes micorrízicas arbusculares y ectomicorrízicas muestran que las redes ectomicorrízicas fueron menos afectadas por el huracán, pues sus conexiones siempre se mantuvieron. Ambos gremios son simbiontes obligados con diferencias fisiológicas importantes. Los hongos ECM tienen la ca- pacidad de obtener C tanto de su hospedero como de la descomposición de la materia orgánica por medio de oxidación (Tunlid et al., 2016); tienen micelio sep- tado con mayor capacidad de recuperación, y sólo tienen asociación con el 2% de las plantas terrestres (Brundrett & Tedersoo, 2018). Los hongos AM obtienen C exclusivamente de sus fitobiontes, tienen micelio cenocítico el cual es más vulne- rable, y se asocian con el 71% de las plantas terrestres (Brundrett & Tedersoo, 2018). A pesar de que los AM se consideran generalistas encontramos que fueron los hongos más afectados. Nosotros consideramos que la afección a la tasa foto- sintética después del huracán, el daño físico a las plantas, y las características biológicas de los hongos AM, pudieron tener mayor efecto en los hongos AM, quienes dependen completamente del C de sus hospederos. Los huracanes son agentes de disturbio que afectan la sobrevivencia de las comunidades que habitan por encima del suelo –animales y plantas (i.e. las plan- tas de nuestras parcelas tuvieron 5.8-14.8% de mortalidad, mientras que del 9.0- 14.3% fueron tumbados y arrancados de raíz pero rebrotaron)– y a la composi- ción de las comunidades fúngicas. Ya sea que los vientos del huracán exporten especies o los animales también dispersen especies (Jumponnen 2003; Behzad et 153 al., 2018; Correia et al., 2019), los nuevos hongos deben pasar por todos los filtros que generan la estructura y composición de las comunidades. Los factores bióti- cos y abióticos representan filtros para el desarrollo exitoso de esas especies re- cién llegadas. Después de un huracán, el primer filtro por pasar serían las condi- ciones de alta radiación solar y temperatura en el suelo. Si la fisiología de la espe- cie permite la sobrevivencia en ese ambiente, después debe pasar por las condi- ciones bióticas. Estas condiciones podrían ser la presencia del hospedero y las interacciones interespecíficas, tales como la competencia, la facilitación o el para- sitismo (Svoboda y Henry, 1987; Koide et al., 2011). Los procesos de ensamble que se conoce para hongos son el ‘priority ef- fect’, la hipótesis de la lotería, los ‘storage effect’, el ‘competition-colonization tra- de-off’ (Kennedy 2010; Kennedy et al., 2011), partición de nicho (Lindahl et al., 2007; Peay et al., 2008; Mujic et al., 2016), dinámica de parches, mortalidad de- pendiente de la densidad (Bruns, 1995) donde todos ellos asumen competencia entre todos los hongos. La sucesión de especies sin reemplazamiento en el suelo sugiere que la competencia no es el proceso de ensamble de esta comunidad. Aunque la competencia es el proceso de ensamble más comúnmente estudiado (e.g. Johnson et al., 2013; Kunstler et al., 2016; Mills et al., 2019) se ha dejado de lado la coexistencia por facilitación o comensalismo. Ambos procesos de ensam- blaje se han observado en diferentes comunidades (e.g. en comunidades baceria- nas, redes de polinización o comunidades vegetales; Mittelbach et al., 2015; Losa- pio et al., 2017; Montesinos-Navarro et al., 2019) La heterogeneidad del ambiente y el enriquecimiento de nutrientes del suelo pudieron haber generado nichos don- de se permitiera la coexistencia de todas las especies sin competir por los mismos recursos, sugiriendo la partición de nicho (Mittelbach et al., 2015) y la dinámica de parches (Bruns 1995) inmediatamente después del huracán. Para la comunidad de hongos rizosféricos, la historia es distinta puesto que son hongos en asociación obligada u oportunista con la planta huésped. El en- samble de las comunidades puede ser explicado por el ‘storage effect’, donde in- volucra el reclutamiento de especies a través del tiempo y los modelos de lotería, 154 donde se compite por un recurso limitante (Kennedy 2010). Es decir, las especies eran parte de la comunidad del suelo y con la heterogeneidad del ambiente provo- cado por el huracán, éstas encontraron las condiciones idóneas para competir por el recurso limitante que eran las raíces vivas. Por lo que las condiciones abióticas, pero también las condiciones bióticas de su hospedero (la pérdida de biomasa que tuvieron las plantas) pudieron afectar la interacción de los simbiontes obligados e.g. los hongos micorrízico arbusculares o ectomicorrízicos. Las comunidades de hongos micorrízicos se encuentran en constante com- petencia por el recurso de las raíces y el C que las plantas translocan a ellas. Nuestros análisis de co-ocurrencia nos indican que los hongos ectomicorrízicos compiten con prácticamente el resto de los hongos encontrados en las raíces. Las interacciones de exclusión o co-ocurrencia pueden generalmente no son recípro- cas (Mack & Rudgers, 2008). Por ejemplo, los hongos micorrízico arbusculares co- habitan con los ECM, patógenos, endófitos y saprótrofos. Los ECM mostraron el mismo patrón durante los dos años muestreados, sin embargo los AM cambiaron de tener mutua exclusión con varios grupos a co-habitar con ellos. Kennedy (2010) y Mahmood (2003) mencionan que el ambiente puede hacer cambios en la inter- acción y nuestros resultados sugieren que el ambiente de un año a otro sí modeló el tipo de simbiosis entre especies. Mientras la temperatura del océano siga incrementando la probabilidad de formación de huracanes sigue aumentando. Los huracanes Jova y Patricia han demostrado que el bosque no es resistente pero sí resiliente a los huracanes (Par- ker et al., 2017; Gavito et al., 2018; Jaramillo et al., 2018; Martínez-Yrízar et al., 2018; Paz et al., 2018), al igual que sus comunidades fúngicas e interacciones con las plantas. Aunado a esto, el aumento de temperatura en el continente generará la desertificación de las zonas tropicales y pérdida de hospederos micorrízicos (Salazar et al., 2007; Setidinger et al., 2019). A pesar de que las interacciones puedan modificarse por el disturbio, la resiliencia de todo el ecosistema siempre dependerá de la alta diversidad. La conservación de la diversidad del suelo redi- tuará en la resiliencia de los ecosistemas ante los disturbios que se prevén por el 155 calentamiento global. La conservación de la diversidad del suelo del bosque tropi- cal caducifolio es un seguro de vida ante las perturbaciones. CONCLUSIONES Los daños que generó el huracán Patricia incrementaron la cantidad de nu- trimentos del suelo. Estos nutrimentos a su vez generaron un cambio en las co- munidades de hongos del suelo; mientras que el daño a la vegetación afectó direc- tamente a los simbiontes rizosféricos. Las respuestas entre suelo y rizósfera fue- ron contrastantes: en el suelo la diversidad incrementó después del huracán y disminuyó dos años después; mientras que en las comunidades rizosféricas justo después del huracán la diversidad disminuyó e incrementó dos años después. El ensamble de las comunidades de hongos del suelo y rizosféricos siguen diferentes procesos; los hongos rizosféricos son simbiontes facultativos u obligados de sus hospederos vegetales, mientras que los hongos del suelo pueden ser organismos de vida libre o cualquier otro estilo de vida. Las comunidades de hongos del suelo fueron resilientes puesto que la do- minancia de especies tendió a aumentar con el tiempo, tal y como estaba antes del huracán. Por otro lado, las comunidades rizosféricas fueron reducidas por la alta temperatura del suelo, lo cual a su vez podría ser explicado por la baja tasa de productividad primaria neta que experimentaron las plantas después del huracán. La red micorrízica recobró conectividad conforme la diversidad rizosférica aumen- tó. Tanto la comunidad rizosférica, como sus interacciones con las plantas, fueron resilientes después de dos años después del huracán Patricia. Sin embargo, es probable que la resiliencia de las comunidades pueda reducirse conforme se vuel- van más frecuentes e intensos los huracanes. Las investigaciones por largos periodos de tiempo en el sitio de estudio han demostrado que la variación interanual es alta, al igual que la variación para even- tos catastróficos, como lo fueron los huracanes Jova y Patricia. Esta tesis asienta los primeros antecedentes de los efectos que tiene el huracán más fuerte que se 156 haya registrado en el Pacífico en las comunidades de hongos. Sin embargo, se debe considerar que cada huracán puede generar diferencias importantes en las comunidades, ya sea por el estado de conservación del bosque, así como de la cantidad de materia orgánica aportada al suelo. 157 LITERATURA CITADA Alvarez-Manjarrez J, Villegas-Ríos M, Garibay-Orijel R, Contreras-Pacheco M, Kõl- jalg U. 2016. Tomentella brunneoincrustata, the first described species of the Pisonieae-associated Neotropical Tomentella clade, and phylogenetic analysis of the genus in Mexico. Mycological Progress 15: 1–11. Alvarez-Manjarrez J, Garibay-Orijel R, Smith ME. 2018. Caryophyllales are the main hosts of a unique set of ectomycorrhizal fungi in a Neotropical dry forest. Mycorrhiza 28: 103–115. Anaya CA, García-Oliva F, Jaramillo VJ. 2007. Rainfall and labile carbon availabil- ity control litter nitrogen dynamics in a tropical dry forest. Oecologia 150: 602–610. Anaya CA, Jaramillo VJ, Martínez-Yrízar A, García-Oliva F. 2012. Large rainfall pulses control litter decomposition in a tropical dry forest: evidence from an 8-year study. Ecosystems 15: 652-663. Bahram M, Kõljalg U, Courty PE, Diédhiou AG, Kjøller R, Põlme S, Ryberg M, Veldre V, Tedersoo L. 2013. The distance decay of similarity in communities of ectomycorrhizal fungi in different ecosystems and scales. Journal of Ecology 101: 1335–1344. Balvanera P, Aguirre E. 2006. Tree diversity, environmental heterogeneity, and productivity in a Mexican tropical dry forest. Biotropica 38: 479–491. Behzad H, Mineta K, Gojobori T. 2018. Global ramifications of dust and sandstorm microbiota. Genome Biology and Evolution 10: 1970–1987. Britton AJ, Mitchell RJ, Fisher JM, Riach DJ, Taylor AFS. 2018. Nitrogen deposi- tion drives loss of moss cover in alpine moss–sedge heath via lowered C:N ratio and accelerated decomposition. New Phytologist 218: 470–478. Brundrett MC, Tedersoo L. 2018. Evolutionary history of mycorrhizal symbioses and global host plant diversity. New Phytologist 220: 1108–1115. 158 Bruns D. T. 1995. Thoughts on the processes that maintain local species diversity of ectomycorrhizal fungi. Plant and Soil 170: 63–73. Corrales A, Mangan SA, Turner BL, Dalling JW. 2016. An ectomycorrhizal nitrogen economy facilitates monodominance in a Neotropical forest. Ecology Let- ters 19: 383–392. Corrales A, Turner BL, Tedersoo L, Anslan S, Dalling JW. 2017. Nitrogen addition alters ectomycorrhizal fungal communities and soil enzyme activities in a tropical montane forest. Fungal Ecology 27: 14–23. Desjardin DE, Perry BA. 2015. A new species of Scytinopogon from the island of Príncipe, Republic of São Tomé and Príncipe, West Africa. Mycosphere 6: 433–440. Gavito ME, Sandoval-Pérez AL, del Castillo K, Cohen-Salgado D, Colarte-Avilés ME, Mora F, Santibáñez-Rentería A, Siddique I, Urquijo-Ramos C. 2018. Resilience of soil nutrient availability and organic matter decomposition to hurricane impact in a tropical dry forest ecosystem. Forest Ecology and Management 426: 81–90. Gessner MO, Swan CM, Dang CK, McKie BG, Bardgett RD, Wall DH, Hät- tenschwiler S. 2010. Diversity meets decomposition. Trends in Ecology and Evolution 25: 372–380. Gilarranz LJ, Rayfield B, Liñán-Cembrano G, Bascompte J, Gonzalez A. 2017. Ef- fects of network modularity on the spread of perturbation impact in experi- mental metapopulations. Science 357: 199–201. Henkel TW, Terborgh J, Vilgalys RJ. 2002. Ectomycorrhizal fungi and their legumi- nous hosts in the Pakaraima mountains of Guyana. Mycological Research 106: 515–531. Henkel TW, Aime MC, Chin MML, Miller SL, Vylgalys R, Smith ME. 2012. Ectomy- corrhizal fungal sporocarp diversity discover of new taxa in Dicymbe 159 monodominant forests of the Guiana Shield. Biodiversity and Conservation 21: 2195-2220. Heuck C, Weig A, Spohn M. 2015. Soil microbial biomass C: N: P stoichiometry and microbial use oforganic phosphorus. Soil Biology and Biochemistry 85: 119–129. Hinsinger P, Gobran GR, Gregory PJ, Wenzel WW. 2005. Rhizosphere geometry and heterogeneity arising from root-mediated physical and chemical pro- cesses. New Phytologist 168: 293–303. Jaramillo VJ, Martínez-Yrízar A, Maass M, Nava-Mendoza M, Castañeda-Gómez L, Ahedo-Hernández R, Araiza S, Verduzco A. 2018. Hurricane impact on biogeochemical processes in a tropical dry forest in western Mexico. For- est Ecology and Management 426: 72–80. Johnson PTJ, Preston DL, Hoverman JT, Richgels KLD. 2013. Biodiversity de- creases disease through predictable changes in host community compe- tence. Nature 494: 230–233. Kennedy P. 2010. Ectomycorrhizal fungi and interspecific competition: Species interactions, community structure, coexistence mechanisms, and future re- search directions. New Phytologist 187: 895–910. Kennedy PG, Higgins LM, Rogers RH, Weber MG. 2011. Colonization-competition tradeoffs as a mechanism driving successional dynamics in ectomycorrhi- zal fungal communities. PLoS ONE 6. Koide RT, Fernandez C, Petprakob K. 2011. General principles in the community ecology of ectomycorrhizal fungi. Annals of Forest Science 68: 45–55. Kunstler G, Falster D, Coomes DA, Hui F, Kooyman RM, Laughlin DC, Poorter L, Vanderwel M, Vieilledent G, Wright SJ, et al. 2016. Plant functional traits have globally consistent effects on competition. Nature 529: 204–207. Leff JW, Bardgett RD, Wilkinson A, Jackson BG, Pritchard WJ, De Long JR, Oak- ley S, Mason KE, Ostle NJ, Johnson D, Baggs EM, Fierer N. 2018. Predict- 160 ing the structure of soil communities from plant community taxonomy, phy- logeny, and traits. The ISME journal 12: 1794–1805. Lindahl BD, Ihrmark K, Boberg J, Trumbore SE, Högberg P, Stenlid J, Finlay RD. 2007. Spatial separation of litter decomposition and mycorrhizal nitrogen uptake in a boreal forest. New Phytologist 173: 611–620. Losapio G, Fortuna MA, Bascompte J, Michalet R, Neumeyer R, Castro L, Cerretti P, Germann C, Haenni J-P, Klopfstein S, et al. 2017. Facilitation between plants shapes pollination networks. bioRxiv: 1–22. Mahmood S. 2003. Colonization of spruce roots by two interacting ectomycorrhizal fungi in wood ash amended substrates. FEMS Microbiology Letters 221: 81–87. Mack KML, Rudgers JA. 2008. Balancing multiple mutualists: Asymmetric interac- tions among plants, arbuscular mycorrhizal fungi, and fungal endophytes. Oikos 117: 310–320. Martínez-Yrízar A, Jaramillo VJ, Maass M, Búrquez A, Parker G, Álvarez-Yépiz JC, Araiza S, Verduzco A, Sarukhán J. 2018. Resilience of tropical dry forest productivity to two hurricanes of different intensity in western Mexico. For- est Ecology and Management 426: 53–60. Mills DR, Do Linh San E, Robinson H, Isoke S, Slotow R, Hunter L. 2019. Competi- tion and specialization in an African forest carnivore community. Ecology and Evolution 9: 10092–10108. Mittelbach GG, Schemske DW. 2015. Ecological and evolutionary perspectives on community assembly. Trends in Ecology and Evolution 30: 241–247. Montaño NM, Alarcón A, Camargo-Ricalde SL, Hernández-Cuevas LV, Álvarez- Sánchez J, González-Chávez MDCA, Gavito ME, Sánchez-Gallen I, Ra- mos-Zapata J, Guadarrama P, et al. 2012. Research on arbuscular mycor- rhizae in Mexico: An historical synthesis and future prospects. Symbiosis 57: 111–126. 161 Montesinos-Navarro A, Valiente-Banuet A, Verdú M. 2019. Plant facilitation through mycorrhizal symbiosis is stronger between distantly related plant species. New Phytologist 224: 928–935. Mooshammer M, Wanek W, Schnecker J, Wild B, Leitner S, Hofhansl F, Blöchl A, Hämmerle I, Frank AH, Fuchslueger L, et al. 2012. Stoichiometric controls of nitrogen and phosphorus cycling in decomposing beech leaf litter. Ecol- ogy 93: 770–782. Mujic AB, Durall DM, Spatafora JW, Kennedy PG. 2016. Competitive avoidance not edaphic specialization drives vertical niche partitioning among sister species of ectomycorrhizal fungi. New Phytologist 209: 1174–1183. Parker G, Martínez-Yrízar A, Álvarez-Yépiz JC, Maass M, Araiza S. 2017. Effects of hurricane disturbance on a tropical dry forest canopy in western Mexico. Forest Ecology and Management 426: 39–52. Pausch J, Kuzyakov Y. 2018. Carbon input by roots into the soil: Quantification of rhizodeposition from root to ecosystem scale. Global Change Biology 24: 1–12. Peay KG, Bruns TD, Kennedy PG, Bergemann SE, Garbelotto M. 2007. A strong species-area relationship for eukaryotic soil microbes: Island size matters for ectomycorrhizal fungi. Ecology Letters 10: 470–480. Peay KG, Kennedy PG, Bruns TD. 2008. Fungal community ecology: a hybrid beast with a molecular master. Biosciece 58: 799–810. PeayKG, Russo SE, Mcguire KL, Lim ZY, Chan JP, Tan S, Davies SJ. 2015. Lack ofhost specificity leads to independent assortment ofdipterocarps and ec- tomycorrhizal fungi across a soil fertility gradient. Ecology Letters 18: 807– 816. Purahong W, Krüger D, Buscot F, Wubet T. 2016. Correlations between the com- position of modular fungal communities and litter decomposition- associated ecosystem functions. Fungal Ecology 22: 106–114. 162 Ramírez-López I, Villegas-Ríos M, Salas-Lizana R, Garibay-Orijel R, Alvarez- Manjarrez J. 2015. Thelephora versatilis and Thelephora pseudoversatilis: Two new cryptic species with polymorphic basidiomes inhabiting tropical deciduous and sub-perennial forests of the Mexican Pacific coast. Mycolo- gia 107: 346–358. Roy M, Schimann H, Braga-Neto R, Da Silva RAE, Duque J, Frame D, Wartchow F, Neves MA. 2016. Diversity and distribution of ectomycorrhizal fungi in Amazonian lowland white sand forests. Biotropica 48: 90–100. Salazar LF, Nobre CA, Oyama MD. 2007. Climate change consequences on the biome distribution in tropical South America. Geophysical Research Letters 34: 1–6. Smith ME, Henkel TW, Aime MC, Fremier AK, Vilgalys R. 2011. Ectomycorrhizal fungal diversity and community structure on three co-occurring leguminous canopy tree species in a Neotropical rainforest. New Phytologist 192: 699– 712. Steidinger B, Crowther T, Liang J, Van Nuland M, Werner, GD Reich P, Nabuurs G, De-Miguel S, Zhou M, Picard N, Herault B, et al. 2019. Climatic controls of decomposition drive the global biogeography of forest tree symbioses. Nature 569: 404–408. Tedersoo L, Sadam A, Zambrano M, Valencia R, Bahram M. 2010. Low diversity and high host preference of ectomycorrhizal fungi in Western Amazonia, a neotropical biodiversity hotspot. The ISME Journal 4:465–471. Tunlid A, Floudas D, Koide R, Rineau F. 2016. Soil organic matter decomposition mechanisms in ectomycorrhizal fungi. In: Martin FM, ed. Molecular Mycor- rhizal Symbiosis. John Wiley & Sons, Inc., 257–275. Truong C, Gabbarini LA, Corrales A, Mujic AB, Escobar JM, Moretto A, Smith ME. 2019. Ectomycorrhizal fungi and soil enzymes exhibit contrasting patterns along elevation gradients in southern Patagonia. New Phytologist. 163 van der Linde S, Suz LM, Orme CDL, Cox F, Andreae H, Asi E, Atkinson B, Ben- ham S, Carroll C, Cools N, et al. 2018. Environment and host as large- scale controls of ectomycorrhizal fungi. Nature 558: 243–248. VanWallendael A. 2019. Digest: Species collapse from disturbance occurs quickly, and recovery is slow. Evolution: 1–2. Vargas R, Hasselquist N, Allen EB, Allen MF. 2010. Effects of a hurricane disturb- ance on aboveground forest structure, arbuscular mycorrhizae and below- ground carbon in a restored tropical forest. Ecosystems 13: 118–128. Varga T, Krizsán K, Földi C, Dima B, Sánchez-García M, Sánchez-Ramírez S, Szöllősi GJ, Szarkándi JG, Papp V, Albert L, et al. 2019. Megaphylogeny resolves global patterns of mushroom evolution. Nature Ecology and Evo- lution 3: 668–678. Villegas Ríos M, Salas-Lizana R, Garibay-Orijel R, Alvarez-Manjarrez J, Pérez- Pazos E, Matías Ferrer N. en preparación. New Mesoamerican lineages of Clavulina reveal a biogeographic link between tropical and temperate re- gions. Mycologia. 164 ANEXO 1. Tomentella brunneoincrustata, the first de- scribed species of the Pisonieae-associated Neotropical Tomentella clade, and phylogenetic analysis of the genus in Mexico Resumen El linaje /tomentella- thelephora es uno de los clados más dominantes entre las comunidades de hongos ectomicorrízicos en todo el mundo. A pesar de la impor- tancia de estos hongos como simbiontes de raíces, sus esporomas son inconspi- cuos y raros de encontrar. El conocimiento de la diversidad de Tomentella en el Neotrópico es escaso, y está basado en secuencias ambientales. Aquí describios la nueva especie Tomentella brunneoincrustata, incluyendo la morfología de los basidiomas, anatomía de las micorrizas y su ecología. Ya que el conocimiento de México sobre Tomentella es poco, nosotros realizamos el primer análisis filogené- tico del género para este país. Nosotros secuenciamos la región nrITS de las muestras fúngicas, y secuencias las regiones rbcL y trnL para identificar a las plantas hospederas. Llos análisis filogenéticos se realizaron con inferencia baye- siana. Los análisis bayesianos mostraron que muchos clados parafiléticos dentro del linaje /tomentella-thelephora están asociados a Pisonieae, presentes a través de áreas tropicales del mundo. Sin embargo, las secuencias de las ectomicorrizas de Puerto Rico, Florida, Dominica y México constituyeron un clado monofilético bien soportado, el cual denominamos el “clado de Tomentella Neotropical asocia- do a Pisonieae”. Dentro de este clado, T. brunneoincrustata fue descrita como: basidioma finamente costroso, fuertemente adherido al sustrato; subículo del mis- mo color, indiferenciado de margen estéril; con dos tipos de hifas en el subículo; y esporas globosas a elipsoides de tamaño pequeño (<8 µm). Esta especie se desa- rrolla en los bosques tropicales caducifolios, donde se asocia con hospederos de la tribu Pisonieae dentro de Nyctaginaceae. Los otros esporomas de Tomentella fueron recolectados en bosques templados de México y pertenecen a los clados de T. atramentaria, T. pilosa, T. muricata, T. fuscocinerea, T. stuposa, T. punicea, T. artroarenicolor, T. bryophylia y T. lateritia. Cinco basidiomas tuvieron secuen- cias con clados independientes y previamente desconocidos de Tomentella. Mycol Progress (2016) 15:10 DOI 10.1007/511557-015-1152-x Q CrossMark ORIGINAL ARTICLE Tomentella brunneoincrustata, the first described species of the Pisonieae-associated Neotropical Tomentella clade, and phylogenetic analysis of the genus in Mexico Julieta Alvarez-Manjarrez'” . Margarita Villegas-Ríos? - Roberto Garibay-Orijel' - Magdalena Contreras-Pacheco? - Urmas Kóljalg* Received: 15 October 2015 /Revised: 10 December 2015 / Accepted: 15 December 2015 () German Mycological Society and Springer-Verlag Berlin Heidelberg 2015 Abstract The /tomentella-thelephora lineage is one of the most highly dominant clades among ectomycorrhizal commu- nities worldwide. Despite its importance as a root symbiont, its fruit bodies are inconspicuous and rarely found. Knowl- edge regarding the diversity of Tomentella in the Neotropics is scarce, and is based largely on environmental samples. Here, we describe a new species, Tomentella brunneoincrustata, including its basidiocarp morphology, mycorrhizal anatomy, and ecology. Because knowledge of Tomentella in Mexico is scarce, we provide the first phylogenetic analysis of this genus in the country. We sequenced the nrIT'S region of the fungal Section Editor: Franz Oberwinkler Electronic supplementary material The online version of this article (doi:10.1007/s11557-015-1152-x) contains supplementary material, which is available to authorized users. L< Roberto Garibay-Orijel rgaribay(vib.unam.mx Laboratorio de Sistemática y Ecología de Micorrizas, Instituto de Biología, Universidad Nacional Autónoma de México, Tercer circuito s/n, Ciudad Universitaria, Delegación Coyoacán 04510, D.F., México ha Laboratorios de Micología, Facultad de Ciencias, Universidad Nacional Autónoma de México, Circuito exterior s/n, Ciudad Universitaria, Delegación Coyoacán 04510, D.F., México Laboratorio de Micología, Escuela Nacional de Ciencias Biológicas, Instituto Politécnico Nacional, Plan de Ayala y Carpio s/n, Col. Santo Tomás, Delegación Coyoacán 11340, MéxicoD.F. Institute of Botany and Ecology, University of Tartu, 40 Lai St., EE-51005 Tartu, Estonia Posgrado en Ciencias Biológicas, Universidad Nacional Autónoma de México, Av. Ciudad Universitaria 3000, C.P. 04360 Delegación Coyoacán, D.F., México Published online: 28 December 2015 samples, and sequenced the rbcL and trnL regions to identify the host plant. The phylogenetic analyses were conducted by Bayesian inference. The Bayesian analysis showed that sev- eral paraphyletic clades within the lineage /tomentella- thelephora are associated with Pisonieae present across tropi- cal regions of the world. However, the ectomycorrhizae se- quences from Puerto Rico, Florida, Dominica, and Mexico constituted a well-supported monophyletic clade that we de- note as the “Pisonieae-associated Neotropical Tomentella clade”. Within this clade, 7. brunneoincrustata was character- ized as follows: a thin crustose, strongly attached to the sub- strate basidiome; concolorous subiculum, undifferentiated and sterile margin; two types of subiculum hyphae; and small (<8 um) globose to ellipsoid spores. This species develops in tropical dry forests, where it associates with hosts in the Pisonieae tribe within the Nyctaginaceae. The remaining Tomentella fruit body vouchers collected in temperate forests of Mexico belonged to clades related to 7. atramentaria, T. pilosa, T. muricata, T. fuscocinerea, T. stuposa, T. punicea, T. atroarenicolor, T. bryophila, and T. lateritia. Five fruit body vouchers had unique sequences forming independent and un- known clades of Tomentella. Keywords Ectomycorrhizal - Neotropics - Pisonieae - Phylogeny - Tomentella - Tropical dry forest Introduction The Thelephoraceae family comprises the genera Amaurodon, Thelephora, Pseudotomentella, Tomentella (Larsson et al. 2004; Agerer 2006), and Odontia (Tedersoo et al. 2014). This family presents clavarioid, effused, flabelliform, pileate or re- supinate basidiocarps (Agerer 2006). A characteristic apomorphy of the family is the irregular-shaped, non-amyloid, A Springer 165 10 Page 2 of 11 Mycol Progress (2016) 15:10 ornamented, and often dark basidiospore with a large apiculus (Larsson et al. 2004). Tomentella has inconspicuous resupi- nate fruit bodies formed by several layers of loose hyphae on soil, wood, twigs or rock surfaces (Koljalg 1996). This genus is paraphyletic, and 1t comprises species that are divided into two lineages: an ectomycorrhizal (/tomentella-thelephora) and a saprotrophic (Tomentella p. parte) lineage. The /tomentella- thelephora lineage has a pan-global distribution and is one of the most species-rich and abundant ectomycorrhizal (ECM) clades associated with all major plant host taxa in a variety of ecosystems (Tedersoo et al. 2010a). The mycorrhizae of Tomentella are morphologically diverse (Jakucs et al. 2015), but share more than three of the following features: black- brown to brown mycorrhiza; clamped hyphae; an angular out- er mantle layer; mantle cells that are organized in a star-like pattern; a mantle surface network composed of hyphae or angular-triangular, horn-shaped cells; groups of globular cells on the mantle surface; rhizomorphs with bilateral, nodal ram- ifications and a rind formed by thin, clamped, densely entwined, multi-branched marginal hyphae; and clamped cystidia (Jakucs and Erós-Hont1, 2008). The /tomentella-thelephora lineage has the following bio- logical and ecological traits. In almost any ECM fungal com- munity (based on mycorrhizal DNA), it is among the three most dominant groups, based on either the number of MOTUs (molecular operational taxonomical units) or the frequency of its DNA sequences (e.g. Dahlberg et al. 1997; Kóljalg et al. 2001; Trowbridge and Jumpponen 2004; Haug et al. 2005; Peay et al. 2007; Smith et al. 2007; Morris et al. 2008; Hynes etal. 2010; Suvi et al. 2010; Tedersoo et al. 201 0b; Smith et al. 2011; Bonito et al. 2012; Brown et al. 2013; Wu et al. 2013). Despite its importance as a root symbiont, its fruit bodies are inconspicuous and rarely found (Jakucs and Erós-Hont1, 2008; Bá et al. 2012). Most of the lineage appears to be ECM (Tedersoo et al. 2010a), while its sister genus Odontia has a stable isotope pattern with an intermediate position be- tween ECM fungi and saprotrophs. The 'C pattern of this genus suggests that 1t does not obtain carbon from its fruiting substratum, although its C source is unknown (Tedersoo et al. 2014). As a consequence of the morphological plasticity of their ectomycorrhizae, the species of /tomentella-thelephora can be distributed either in the mineral soil horizon (Harrington and Mitchell 2005; Batier et al. 2006) developing a “contact exploration type” ECM, or in the organic horizon of broad-leaved forests (Tedersoo et al. 2003), in which they are often attached to plant foliar debris. In the latter case, they develop slightly or highly differentiated rhizomorphs, indicat- ing that these morphotypes belong to the “medium-distance exploration type” (Jakucs and Erós-Honti, 2008). While the /tomentella-thelephora is dominant in boreal and temperate forests in the Northern Hemisphere, 1t has also been identified in the Southern Hemisphere and in tropical and sub- tropical ecosystems such as those in India (Thind and Rattan Y Springer 1971), Korea (Jung 1994), and the Canary Islands (Larsen 1994). It was recently found to be dominant in the following tropical areas: subtropical broadleaf mixed forests in China (Gao et al. 2015); Coccoloba uvifera coastal forests in the Gua- deloupe island in the Lesser Antilles (Séne et al. 2015); African tropical forests containing Caesalpinioideae (Fabaceae), Sarcolaenaceae, Dipterocarpaceae, Asteropelaceas, Phyllanthaceae, Sapotaceae, Papilionoideae (Fabaceae), Gnetaceae and Proteaceae, distributed in open, gallery and rainforests of the Guineo-Congolian basin; Zambezian Miombo woodlands of East and South-Central Africa; and Sudanian savanna woodlands of the sub-Saharan region (Bá et al. 2012). Despite their importance in tropical ecosystems, most Tomentella species have been identified in temperate regions (Larsen 1974, Júlich and Stalpers 1980, Stalpers 1993, Kóljalg 1996). Several new tropical species were recently described from Africa (Yorou et al. 2007; Yorou and Agerer 2007; Yorou and Agerer 2008; Yorou et al. 2011; Yorou et al. 2012a; Yorou et al. 2012b) and the Seychelles (Suvi et al. 2010). However, knowledge regarding the diversity of Tomentella in the Neotropics is scarce, and based only on en- vironmental samples from Ecuador (Tedersoo et al. 2010b), Dominica, Puerto Rico, and Vieques (Hayward and Horton 2014). Similar to those from other regions worldwide, environ- mental DNA sequences in the Mexican Neotropics indicate that the /tomentella-thelephora lineage is dominant in the ECM roots of several ecosystems including subtropical pine-oak for- ests (Garibay-Orijel 2008), cloud oak forests (Morris et al. 2009), alpine conifer forests (Reverchon et al. 2010), and A/nus temperate and tropical forests (Kennedy et al. 2011). However, based on basidiocarp collections, only 7. chlorine (Massee) G. Cunmn., 7. ferruginea (Pers.) Pat., T. griseoumbrina Litsch., T. pilosa (Burt) Bourdot $: Galzin, 7. subsaccicola M.J. Larsen, and 7: umbrinospora M.J. Larsen have been detected in Mexico (Welden et al. 1979; Urbizu et al. 2004; Contreras-Pacheco 2008; Contreras-Pacheco et al. 2014). In our laboratory, we study the diversity, ecology, and as- sociations of ECM fungi residing in Neotropical dry forests along the Pacific coast of Mexico. In this seasonal ecosystem, the /tomentella-thelephora lineage has been shown to be dom- inant in the ECM community, consisting of species new to science (Ramírez-López et al. 2015). Here, we describe a new species, Tomentella brunneoincrustata, including its basidiocarp morphology, mycorrhizal anatomy, ecology, and host associations. Because knowledge regarding Tomentella in Mexico is scarce, we also provide the first phylogenetic analysis of the diversity of this genus in this country. Materials and methods Study site The study was conducted at the Chamela- Cuixmala Biosphere Reserve (N 19930”, W 10503”) in 166 Mycol Progress (2016) 15:10 Page 3 of 11 10 Jalisco, Mexico (Fig. 1), where the principal type of vegeta- tion is tropical dry forest, and the tropical sub-deciduous forest is restricted to creeks and streams. During the summer, the weather 1s sub-humid and warm, whereas it is dry in the win- ter. The tropical dry forest exhibits water stress for 8 months, and the rainy season usually extends from July to October, which coincides with hurricane season. The average annual precipitation is 784.8 mm (1977-2011), and the average an- nual temperature is 24.6 *C, with an average maximum and minimum of 30.3 *C and 19.5 *C, respectively. The atmo- spheric humidity is >65 % during the rainy season (Bullock 1986; García-Oliva et al. 1995). Sampling The reserve was accessed during the rainy season each year from 2012 through 2014, and opportunistic sporo- carp sampling of ectomycorrhizal species was conducted ac- cording to O”Dell et al. (2004). Root tips were sampled with soil cores (PVC tubes 30x 5 cm; -589 cm? of soil) under suspected ectomycorrhizal hosts. The ECM were separated from the roots by carefully washing of the soil with tap water into a sieve. All the ECM were then isolated using a stereo- microscope. The ECM were fixed in 96 % ethanol and stored at 4 *C for a maximum of 2 weeks until further processing. All of the morphotypes were photographed prior to the an- atomical analysis. The root tips were mounted in Paraplast (Leica Biosystems, Buffalo Grove, IL, USA ); the anatomical slices were performed with a rotation microtome, and then mounted and stained in permanent preparations according to Sandoval-Zapotitla (2005). The ECM morphotypes were described after fixation, based on morphological and ana- tomical characteristics according to Agerer and Rambold (2004-2015). Morphological data The macroscopic characteristics of the sporocarps were determined based on fresh material, and the color was determined according to the Munsell soil color charts (Munsell Color Company 1954). The microscopic char- acteristics of the fruit body vouchers were observed using tissue rehydrated in 2.5 % KOH by Nomarski Interference Contrast with an Olympus BX51 microscope. All of the mea- surements of basidia (n= 10), basidiospores (n=30), and hy- phae (n= 30) were performed using 1000x KOH preparations. We calculated the length/width ratio (Q), average (Q), average length (L) and average width (W) of the spores. The spore ornamentation was observed using a scanning electron micro- scope (JEOL JSM-5310LV). Molecular procedures When sufficient material was collect- ed from a given ECM morphotype, a 1-2 mm section was used to extract DNA with the XNAP kit (Sigma-Aldrich Corp., St. Louis, MO, USA). DNA was extracted from the sporocarps using the same protocol as that used for the ECM. We amplified the nuclear ribosomal internal transcribed spacer (nrITS) region by polymerase chain reaction (PCR) with the ITS1F/1TS4 primer pair (Gardes and Bruns 1993) using RubyTaq PCR Master Mix (Affymetrix, Inc., Santa Clara, CA, USA). DNA extraction and PCR were performed as described by Garibay-Orijel et al. (2013). To identify the host plant from the root tips, we amplified the rbc£L and trnL regions using the rbcL-aF/rbcL-aR and trnC/tnD primer pairs (Kress and Erickson 2007). All of the PCR products were observed in 1 % agarose gels stained with GelRed (Biotium, Hayward, CA, USA). Amplicons of the appropriate size were cleaned with ExoSAP-IT (Affymetrix, Inc.). DNA sequences were generated in both directions using PCR primers and > ; AS 5 ] le a 4 / y / ' A A A Y a Y / SA | kr108747% SN S SÁ > ey Ma A a, 5 A KP8062 sh, e Legends === Trails 0 Biological Station XK Sporocarp 20 Ectomycorrhiza KT1 98742 : Ve KT 198744 ñ q _ E E M KT198746 * y, y KTLO874S y e - => dé 0 A ._ * --- os , q , 9 Ma 100 K x m IS E - > 3d E E E t md! = TT $ KT198743 ; ? Fig. 1 Location of the Chamela-Cuixmala Biological Reserve, and distribution of the holotype fruit body and ectomycorrhizae of Tomentella brunneoincrustata. Samples are indicated by their GenBank accession numbers A Springer 167 10 Page 4 of 11 Mycol Progress (2016) 15:10 BigDye Terminator v3.1 chemistry at the “Laboratorio de Secuenciación Genómica de la Biodiversidad y de la Salud” at the UNAM Biology Institute with an ABI PRISM 3100 genetic analyzer (Applied Biosystems, Foster City, CA, USA). Bioinformatics The DNA sequences were edited and assem- bled using Geneious 6.1.4 software (Biomatters Ltd., Auck- land, New Zealand). The plant hosts were identified by com- paring the DNA sequences with those available in the BOLD Systems genetic barcode database. The identity of the fungal DNA sequences was assessed by phylogenetic analysis. First, we compared the sequences obtained in the present study against those in the GenBank and UNITE databases and downloaded all of the best matches (290 % similarity). We included all the tropical /tomentella-thelephora sequences from fruit body vouchers like those from the Seychelles (Suvi et al. 2010) and Benin (Yorou et al. 2011). We also selected environmental samples of /tomentella-thelephora from the Neotropics in GenBank and included Tomentella fruit body voucher sequences collected throughout Mexico in recent years by our laboratory (Table S1). The alignment was performed using MAFFT v7 (http://mafft.cbrc.jp/ alignment/server/) and revised it manually with Mesquite v2. 75. The molecular phylogenetic analyses included a Bayesian analysis that was performed using MrBayes v3.2.2 with 4 MCMC, 5 million generations, and three partitions (ITS1, 5. 8S, ITS2). To select the best substitution model for each partition, we performed a reversible-jump Markov chain Monte Carlo computation (Pagel and Meade 2006) with Thelephora terrestris as the external group. We generated the consensus tree, adding posterior probabilities on the branches (20.75), and the nodes were depicted in decreasing order with FigTree v1.0.4. Results The IT'S sequences of the six selected root tips and one spo- rocarp had an overall nucleotide sequence similarity of 98.3 %. The collection sites of these samples were widely distributed across the tropical dry forest of Chamela (Fig. 1). The Bayesian analysis grouped these sequences into a clade together with an ECM sequence from Dominica (JX548248), with high support (BPP=1). The sequence from Dominica demonstrated an overall nucleotide sequence similarity of 96.8 % with the samples representing 7. brunneoincrustata, and it contained 12 unique single-nucleotide polymorphisms (SNPs) (Table S2). This clade, together with two clades consisting of environmental samples of ECM from subtropical forests in Florida and tropical dry forests in Puerto Rico, made up the “Pisonieae-associated Neotropical Tomentella clade” (Fig. 2). Analysis of the rbcL and trnL sequences from the Y Springer ECM revealed that five of the ectomycorrhizae were associat- ed with a member of the Pisonieae tribe in the Nyctaginaceae and that one was associated with Pisonia sp. (Table 1). The sister clades from Puerto Rico and Florida were also associat- ed with hosts in the Nyctaginaceae family (Fig. 2). Our sam- ples belong to an undescribed clade with unique morpholog- ical and ecological characteristics, which is described here as the new species Tomentella brunneoincrustata. The Bayesian analysis revealed that the Tomentella fruit body vouchers collected in temperate forests of Mexico be- long to clades with species inhabiting temperate forests world- wide. KC 152246 and KT353045 are related to 7. atramentaria from the USA, Estonia, and Austria; KT353044 and KC152245 are a sister group of 7. pilosa from Estonia and Sweden; KT353054 and UDBO018512 formed a group with T. muricata from Estonia and Finland; KT353055 is related to T. fuscocinerea from Iran; KT353058 is a sister group of T. stuposa from Austria; KC152248 is related to 7. punicea from China; and KT353052 is similar to 7. atroarenicolor from Russia. We also found that KT353049, KT353048, and KT353047 are sister groups of 7. bryophila from Scotland and Canada; however, this species is paraphyletic. The same case was found for KT35305 1, which is related to 7. lateritia from Italy. Five fruit body vouchers had unique sequences (i.e. KC152247, KT353046, KT353056, KT353057, and KT353050) that formed independent and unknown clades of Tomentella. Taxonomy Tomentella brunneoincrustata M. Villegas k Contreras-Pacheco, sp. nov. MycoBank: MB814303 Diagnosis Basidiome resupinate, crustose, thin, adherent to the substrate, dark brown, undifferentiated sterile margin, without rhizomorphs. Subicular hyphae dimitic, dark brown or purple brown; basidia subclavate, tetrasporic, clamped at base, rarely with transverse septa. Basidiospores subglobose to ellipsoid, dark brown, (6) 6.0—7.5 (8)x 5.5—6.5 um; orna- mentation echinulate, frequently bi- or trifurcate. Inhabiting soil and dead wood on tropical dry forests, forming ectomycorrhizae with different members of the Nyctaginaceae family. HOLOTYPE: Álvarez-Manjarrez 152b, (MEXU 27661). Basidiome resupinate, thin, less than 1 mm thick, crustose, mostly continuous, indeterminate edges with patches around, strongly adherent to the substrate; hymenium dark brown (2.5/ 2-3/7.5 YR Munsell), smooth to the naked eye, densely to- mentose and iridescent when seen under a dissection micro- scope, turns darker in 2.5 % KOH; subiculum concolorous 168 Mycol Progress (2016) 15:10 Page5 of 11 10 Fig. 2 Phylogenetic Bayesian analysis of vouchers and environmental samples of Tomentella and its host preferences. The sequences from Tomentella brunneoincrustata, including the ectomycorrhizae and the holotype, are shown in bold in a green square. The terminals indicate the regions where they were collected; sequences from environmental samples are labeled as “ectomycorrhiza” and sequences of Tomentella vouchers are labeled with the species names. The symbols indicate the host family: Aceraceac (circle), Betulaceae (half*round), Dipterocarpaceae (diamond), Fabaceae (square), Fagaceae (oval), Myrtaceae (spiral), Nyctaginaceae (star), Pinaceae (triangle), Polygonaceae (pentagon) and Salicaceae (bold line) HM189960_Thelephora_terrestris HM189958_Thelephora_terrestris GQ267490_Thelephora_terrestris_NewZeland HM189965_Thelephora_terrestris JQ711777_Thelephora_terrestris_Canada AY667418_Ectomycorhiza_Ecuador MY 1 1 UDB003343_Tomentella_sp_Tanzania A KA4/A< A FM955847_Tomentella_parmastoana_Seychelles Mi UDB003341_Tomentella_sp_Australia AM412303_Tomentella_hjortstamiana_Seychelles Mi FM244908_Tomentella_pisoniae_Seychelles fir DQ974775_Tomentella_ellisii_ USA «> KT353046_Tomentella_sp_Mexico AY687424_Ectomycorrhiza_Ecuador lr AY667422_Ectomycorhiza_Ecuador lr AY667423_Ectomycorrhiza_Ecuador ly G0Q268671_Ectomycorrhiza_Malaysia UDB004274_Tomentella_sp_Ecuador (Y JX548277_Ectomycorrhiza_PuertoRico JX548274_Ectomycorrhiza_PuertoRico JX548276_Ectomycorrhiza_PuertoRico AM412299 Tomentella_tenuis_: KC152247_Tomentella_sp. ) Mexico KC155401_Tomentella_sp_Guyana Ml 085 AM412290_Ectomycorrhiza_Seychelles dy 1 AM412292 Ectomycorrhiza_Seychelles [3 FN396396_Ectomycorrhiza_Seychelles 4) 1 FR682090_Ectomycorrhiza_Guadeloupe 1 KF472143_Ectomycorrhiza_Guadeloupe UY KF472145_Ectomycorhiza_Guadeloupe (Y FM955845_Tomentella_pileocystidiata_Seychelles Mi A 570_Ectomycorrhiza_Australia UDB003335_Tomentella_cf_clavigera_CostaRica > 1 EFS38421_Tomentella_sp S “1———————— FM244909_Tomentella_tedersooi_Seychelles fly ————————— JN 168773_Tomentella_sp_Guyana Ml á JN168772_Tomentella_sp [8 E UDB016439 Tomentella_lateritia_Estonia DQ974777_Tomentella_lateritia_USA UDBO016493_Tomentella_coerulea_ltalia UDB016705_Tomentella_lateritia_Australia 1 orion 7_Tomentella_bryophila_ ECM_Canada Ah KT353047_Tomentella_sp_Mexico A UDB016193_Tomentella_cinerascens_Estonia UDB000274_Tomentella_brunneorufa_Austrañia Q) A ——— 1UDB003300_Tomentella_fuscocinerea_lran KT353055_Tomentella_sp_Mexico 1 KT353057_Tomentella_sp_Mexico KT353050_Tomentella_sp_Mexico 1 AF272926_Tomentella_lateritia_ Sweden UDB000267_Tomentella_lateritia_Estonia 9 UDB000268_Tomentella_lateritia_Australia 096 JN168763_| iza_ Guyana Ml 1 KT353058_Tomentella_sp_Mexico EF644117_Tomentella_stuposa_Austria | KF472158_1 AM412296_Tomentella_intsiae_Seychelles Mi UDB018564_Tomentella_sp_Estonia UDB016307_Tomentella_stuposa_Estonia UDB001655_Tomentella_bryophila_Scotland AF272907_Thelephora_pseudoterrestris_Sweden KT353056_Tomentella_sp_Mexico UDB011602_Tomentella_cinereoumbrina_Fintand [| UDB000238_Tomentella_badia_Russia UDB000955_Tomentella_atramentaria_Estonia DQ974772_Tomentella_atramentaria_USA KC152246_Tomentella_sp_Mexico KT353045_Tomentella_sp_Mexico EF507253_Tomentella_africana_Benin ES EF507257_Tomentella_agbassaensis_Benin UDB018677_Tomentella_sp_India 1 KT353051_Tomentella_sp_Mexico AA KF275145_Tomentella_lateritia_ltaly 1 UDB000255_Tomentella_botryoides_Sweden A o KC152248_Tomentella_sp_Mexico UDB018440_Tomentella_punicea_China UDB003324_Tomentella_pilosa_Estonia AF272925_Tomentella_pilosa_Sweden 12298_Tomentella_beaverae_Seychelles Mi HOSTS Angiosperms e Acer e Alnus h% Betula «5 Castanea % Coccoloba M Dicymbo a Eucalyptus % Fagus Guapira El Intsia Neea *k Pisonia Il Populus % Quercus 4) Vateriopsis Gymnosperms A Pinus A Tsuga J3Q814474_Tomentella_minispora_Guinea KF836015_Ectomycorrhiza_Florida Nyctaginacese KF472141_Ectomycorrhiza_Guadeloupe (Y A Springer 169 10 Page 6 of 11 Mycol Progress (2016) 15:10 Table 1 BLAST identification of the ECM host rbcL and trnL regions Sample type Accession number GenBank rbcL BLAST results trnL BLAST results Host % Id Match % Id Match ECM KT906429 100 Pisonia aculeata (KJ594427) - - Pisonicae sp. ECM KT906430 Neea psychotrioides (JQ592987) - - ECM KT906431 Guapira standleyana (GQ981748) _ ECM KT906428 — A ECM KT906427 ECM KT906432 95 Pisonia albida (JX8444286) Pisonia sp. with the hymenium, undifferentiated sterile margin, rising slightly from the substrate; rhizomorphs absent (Fig. 3a). Subicular hyphae consisting of two types: a) very com- mon generative hyphae, dark brown in 2.5 % KOH, 3.2— 5.1 (6.3) um wide, thick-walled (up to 1 tum), clamped, branched mostly at right angles, with irregular swellings of up to 20 um in some hyphae, anastomoses not ob- served, hyphae not congophilous, not cyanophilous and not amyloid (Fig. 3b); b) infrequent hyphae with simple septa, purple-brown, thin-walled, sometimes dichoto- mously branched, 1.8-3.3 um wide, very ornamented on Fig. 3 Tomentella brunneoincrustata holotype (Alvarez-Manjarrez 152b). a Resupinate basidiome; b generative hyphae of subiculum with irregular swellings; e omnamented hyphae of subiculum; d sub-hymenium hyphae, immature basidia (arrow) and tetrasporic basidia; e SEM of young basidia with clamp at the base; f, g SEM of basidiospores in lateral and basal view showing obtuse hilar appendix and bi- or trifurcate omnmamentation. Scale bars: b, c=20 um; d=15 um; e=3 um;f. g=1 um Y Springer the surface with fine crystals insoluble in 2.5 % KOH, and not cyanophilous (Fig. 3c). Subhymenial hyphae consisting of swollen cells with ir- regular forms, 4.2-11.1 ¡um wide, thick-walled (up to 1 um), clamped, dark brown to light brown in 2.5 % KOH, and not congophilous or cyanophilous. Immature basidia dark brown in 2.5 % KOH, clavate, sphaeropedunculate or napiform, clamped and thick-walled; mature basidia 29.1-37.5x 9.26— 15.8 um, subclavate, four sterigmata (5-7 um), slightly thickened wall at the base and thin wall at the apex, light brown in 2.5 % KOH, clamped at the base, rarely exhibiting 170 Mycol Progress (2016) 15:10 Page 7 of 11 10 transverse septa, and most septa collapsed (Fig. 3d, e). Basidiospores (6) 6.0-7.5 (8)x5.5-6.5 um (Q=1.1- 1.3 um, Q=1.1 um, L=6.8 um, W=6.1), in front view, subglobose to ellipsoid, some slightly lobed, dark brown in 2.5 % KOH, slightly thickened wall and echinulate, not congophilous, not cyanophilous, not amyloid. In SEM, spores showed an obtuse hilar appendix, 1-1.5x1.2- 1.5 um; echinulate ornamentation frequently bi- or trifurcate, 1-1.2x0.5-1 um, with rounded or sub-rounded tips (Fig. 3f, g). Remarks This species is characterized by a thin basidiome that is crustose and strongly attached to the substrate; subiculum concolorous with the hymenium, undifferentiated and sterile margin; two types of subiculum hyphae, of which the ornamented one does not present clamps; and small glo- bose to ellipsoid spores (<8 um). Among the tropical species described in the literature, this species is similar only to Tomentella minispora Yorou et al. (2012a) from Guinea, which also possess basidiomes, strongly attached to the sub- strate basidiomes, no rhizomorphs, clamps on both hyphae and basidia, ornamentation on the surface of some hyphae, and has a similar spore size. Despite this apparent similarity, T. minispora displays important differences, such as the pres- ence of a differentiated sterile margin with clearer pigmenta- tion, hyphae from the subiculum that are thin-walled or slight- ly thickened, hyphae ornamentation that is present only on the subhymenium, and spore ornaments that are never bi- or trifurcated. Fig. 4 a-b Ectomycorrhiza of Tomentella brunneoincrustata associated with Pisonicae sp. e Transversal section of an ectomycorrhizal tip showing the hyphal layer of the mantle and the peri-epidermal Hartig net. d Detail of the Hartig net, with arrows indicating the peri- epidermal hyphac. Scale bars: a = 0.5 mm; b=0.25 mm; c=75 um; d = 4.5 um Etymology From the Latin brunneus and incrustata, in refer- ence to the brown color of the basidiome and extracellular incrustations on the hyphae of the subiculum. Habit, habitat, and distribution This species develops in tropical dry forests in which it associates with hosts in the Pisonieae tribe within the Nyctaginaceae. Specimens examined HOLOTYPE: Mexico, Jalisco, La Huerta municipality, Estación de Biología de Chamela, Tejón sidewalk, 19230” N, 105239” W, 26 Nov 2014, Álvarez- Manjarrez 152b, (MEXU 27661). Anatomical description of the ectomycorrhizae Tomentella brunneoincrustata + Pisonieae sp. Ectomycorrhiza sinuous with monopodial ramifications and rounded tips. Completely black with emanating black and erect hyphae (Fig. 4a, b). Mantle thick and partially shiny, with 12-16 hyphal layers consisting of three different confor- mational structures. External mantle black, emanating hy- phae septate with clamps, thick walls (>1 um), and rounded terminations. Internal mantle has clearer hyphae in compar- ison with the remaining mantle, hyphae epidermoid or irreg- ular (4-11 x4-13 um). Hartig net is prominent, peri-epider- mal, enclosing the epidermal and the first cortical cell layer, infrequently lobulated (Fig. 4c, d). A Springer 171 10 Page 8 of 11 Mycol Progress (2016) 15:10 Tomentella brunneoincrustata + Pisonia sp. Ectomycorrhiza sinuous with monopodial ramifications and rounded tips. Mantle black and extremely dense, tomentose- granulose surface and emanating hyphae dark in color. Hy- phae more abundant and larger at the base of the ECM (Fig. Sa, b). Mantle with 10-17 hyphal layers (46-72 um), resembling divergent lamellar trama. External mantle pre- sents cylindrical, emanating straight hyphae (2-4 x 7— 19 um) with dark septa and a wide wall (<1 um). Internal mantle has epidermoid lighter-coloured hyphae (4-8 x 3— 7 um). Hartig net hyaline, infrequently lobed, penetrating more than 1 cortical cell (Fig. Sc, d). Considerations This species forms very similar morphotypes with different Nyctaginaceae hosts, consisting of a black, dense mantle with short exploration type (Agerer 2001) and with monopodial ramifications. The Hartig net is prominent, peri-epidermal, and infrequently lobulated. Discussion Tomentella brunneoincrustata produces dark brown fruit bod- les that are somewhat similar to those of 7. agbassaensis Yorou, 7. amyloapiculata Yorou, T. guineensis Yorou, T. guinkoi Yorou, T. minispora Yorou, T. afrostuposa Yorou, and 7. intsiae Suvi $ Koljalg. Another important characteris- tic of Tomentella brunneoincrustata is its adherence to the Fig. 5 a-b Ectomycorrhiza of Tomentella brunneoincrustata associated with Pisonia sp. € Transversal section of the ectomycorrhizal tip (10%). d Detail of the dark mantle. Scale bars: a= 1.0 mm; b= 0.5 mm: d - 4.5 um Y Springer substrate and absence of rhizomorphs, both of which are ob- served in 7. amyloapiculata, T. guineensis, T. guinkoi, T. minispora, and T. intsiae. This new species exhibits greater similarity to 7. minispora and T. afrostuposa due to a common arachnoid subiculum, hymenia exhibiting the same color, and some sub-hymenial hyphae with incrustations. The size of the spores coincides with that of 7. minispora. Nonetheless, T. brunneoincrustata presents unique characteristics: a diffuse concolorous margin, non-cyanescent subiculum hyphae, hy- phal ornamentation that is present only on the subiculum, and spore ornaments that are bifurcate or trifurcated. The fruit body of the holotype was found on the underside ofa piece of wood without evident rotting. Odontia, the sister genus of Tomentella, has been reported to be saprotrophic (Tedersoo et al. 2014). However, 7. brunneoincrustata forms ECM and belongs to an ectomycorrhizal clade that is associ- ated with the Pisonieae tribe from the Nyctaginaceae. This is the first study to describe a Tomentella species from the Neo- tropics, including its ectomycorrhizae. The ECM of this spe- cies displayed a dense, dark brown mantle; the Hartig net was found to be peri-epidermal and very prominent in both morphotypes. This species shares only the dark mantle with the Tomentella EMC morphotypes described by Jakucs and Erós-Honti (2008) and Jakucs et al. (2015). The ECM of this species exhibits greater similarity to the one described for the Guapira ECM from Ecuador (Haug et al. 2005), which shares the prominent Hartig net. However, 7. brunneoincrustata de- velops a mantle wrapping the root tips completely, with the Hartig net penetrating two cell layers. 172 Mycol Progress (2016) 15:10 Page9 of 11 10 In phylogenetic analysis, the sequence of the ECM from Dominica exhibited the closest similarity to those from T. brunneoincrustata (96.3 % similarity). However, according to the 97 % similarity consensus to form MOTUs of ECM fungi (Nilsson et al. 2008; Peay et al. 2008; Setaro et al. 2012) and the UNITE species hypothesis of 98 % (Koóljalg et al. 2013), this sequence is not included with 7. brunneovincrustata. Although genetic markers that are recognized as genetic barcodes for plants were used for host identification, we were able to identify only one host to genus level, Pisonia sp., according to the list of plants from Chamela (Lott 1993). The reserve contains 13 species from 8 genera of Nyctaginaceae; Guapira petenensis 1s the unique species in this genus, while Pisonia has two species, P aculeata and P. macranthocarpa. The association of /tomentella-thelephora with Pisonieae has been reported in several regions throughout the world: Dominica, Ecuador, Florida, Hawaii, Puerto Rico, Rota, the Seychelles, and Vieques (Haug et al. 2005; Hayward and Horton 2012, 2014; Suvi et al. 2010, Tedersoo et al. 2010b). Bayesian analysis showed that several paraphyletic clades within the lineage /tomentella-thelephora are associated with Pisonieae across the tropical regions of the world. However, the ECM sequences from Puerto Rico, Florida, Dominica, and Mexico constitute the “Pisonieae-associated Neotropical Tomentella clade”, which is monophyletic and inhabits tropi- cal dry and subtropical forests of the Neotropics, especially the Mesoamerican and Caribbean regions. The specificity of this fungal clade to the Pisonieae supports the hypothesis of partner choice phylogenetic trait conservation proposed by Hayward and Horton (2014). The Pisonieae tribe includes three ectomycorrhizal genera: Guapira, Neea, and Pisonia. Neea and Guapira are paraphyletic groups (Hayward and Horton 2014), both of which are exclusive to tropical forests in Mexico, Central America, and South America (Douglas and Manos 2007). There are three Guapira species, eight Neea species, and five Pisonia species in Mexico. These species are distributed in 25 of the 32 Mexican states, among which Chiapas exhibits the greatest diversity, with 13 spp., followed by the Yucatan Pen- insula with 11 spp. Given that 7. brunneoincrustata 18 associ- ated with two of these genera, there is a high probability that this species, or other undescribed species within the Pisonieae-associated Neotropical clade, has a wider distribu- tion within the Neotropics. More systematic sampling of the entire area is needed to understand the biology, ecology, and diversity of Tomentella in the Neotropics. Given the distribution and ecosystem preferences of the Pisonieae-associated Neotropical Tomentella clade, it 1s likely that this clade is associated with water stress conditions, such as in those present in the Chamela tropical dry forest. The samples from Puerto Rico and Dominica were also obtained from tropical dry forests (Hayward and Horton 2014). In Puerto Rico, the mean temperature is 29.7 *C, with a maxi- mum of 32.4 *C, a minimum of 14.6 *C, and mean annual precipitation of 1687 mm. The distribution of water resources is critical in the Caribbean islands, and similar patterns are observed in different islands (Daly et al. 2003), such as Dom- inica. Even the samples from Florida inhabited a subtropical region with an average temperature of 23.8 *C and average rainfall of approximately 1524 mm per year, 75 % of which occurred from June through October, coinciding with hurri- cane season (Multer and Hoffmeister 1968). Thus, all of the members of this clade seem to develop in (sub)tropical areas with high temperatures and heterogeneous rainfall regimes that are unevenly distributed throughout the year. Six species of Tomentella have been reported in Mex- ico; however, the Bayesian analysis revealed a large di- versity of Tomentella species, some of which are related to known taxa, and many others which are likely new species. The Tomentella fruit body vouchers from temper- ate forests in Mexico that were included in the analysis showed greater genetic similarity with species from tem- perate climates than those from tropical climates. These results are consistent with the biology of the species and its host associations. Hayward and Horton (2014) noted that when Neea buxifolia and Pisonia aculeata were planted in New York soil, in which local thelephoroids were available, the plants failed to form ECM with the local species. The Nearctic and Neotropical biotas coin- cide in Mexico (Estrada-Contreras et al. 2015), and even if the vegetation types are similar (e.g., the transitions of pine-oak forest, montane cloud forest, tropical dry forest, and sand dunes), temperate and tropical tree species do not intermix, enabling a high diversity of many biological groups. This is the first study to analyze the diversity of Tomentella in Mexico. The phylogenetic analysis presented here will help to guide future investigations designed to identify and de- scribe the Tomentella species in this region. However, given the vast diversity and complexity of the genus in this country, a complete knowledge of its diversity and ecology is a long- term task that would require the participation of several re- search groups. Acknowledgments This research was funded by PAPIT IN218210 and PAPIUT 1N223114. The MEXBOL network supported DNA se- quencing thru CONACYT grant 1251085. We thank the Posgrado en Ciencias Biológicas, Universidad Nacional Autónoma de Mexico; this work was conducted in fulfillment of requirements within the JAM PhD program. We would like to thank José Luis Villaseñor-Ríos and his laboratory for the Nyctaginaceae distribution data. We also thank Silvia Espinoza-Matías for her support with the SEM photographs, and Estela Sandoval-Zapotitla for her help at the anatomical slices of ectomycorrhizae. We thank to the Biological Station of Chamela and its entire stafF. A Springer 173 10 Page 10 of 11 Mycol Progress (2016) 15:10 References Agerer R (2001) Exploration types of ectomycorrhizae: a proposal to classify ectomycorrhizal mycelial systems according to their pat- terns of differentiation and putative ecological importance. Mycorrhiza 11:107-114 Agerer R (2006) Fungal relationships and structural identity of their ectomycorrhizae. Mycol Prog 5:7-107. doi:10.1007/s11557-006- 0505-x Agerer R, Rambold G (2004-2015) DEEMY Information system for characterization and determination of ectomycorrhizae. www. deemy.de. Accessed 28 April 2015 Bá AM, Duponnois R, Moyersoen B, Diédhiou AG (2012) Ectomycorrhizal symbiosis of tropical African trees. Mycorrhiza 22:1-29. doi:10.1007/500572-011-0415-x Baier R, Ingenhaag J, Blaschke H, Góttlein A, Agerer R (2006) Vertical distribution ofan ectomycorrhizal community in upper soil horizons of a young Norway spruce (Picea abies [L.] Karst.) stand of the Bavarian Limestone Alps. Mycorrhiza 16:197-206. doi:10.1007/ $00572-006-0035-z Bonito G, Smith ME, Brenneman T, Vilgalys R (2012) Assessing ectomycorrhizal fungal spore banks of truffle producing soils with pecan seedling trap-plants. Plant Soil 356:357-366. doi:10.1007/ s11104-012-1127-5 Brown SP, Callaham MA, Oliver AK, Jumpponen A (2013) Deep lon Torrent sequencing identifies soil fungal community shifts after fre- quent prescribed fires in a southeastern US forest ecosystem. FEMS Microbiol Ecol 86:557-566. doi:10.1111/1574-6941.12181 Bullock SH (1986) Climate of Chamela, Jalisco, and trends in the south coastal region of Mexico. Arch Meteor Geophys B 36:297-316. doi: 10.1007/BF02263135 Contreras-Pacheco M (2008) Estudio taxonómico de Hymenomycetes resupinados lisos de los estados del Golfo y Chiapas, México. Master Dissertation, Universidad Nacional Autonoma de Mexico Contreras-Pacheco M, Raymundo T, Bautista-Hernández S, Díaz- Moreno R, Valenzuela R (2014) Hongos corticioides del Bosque Las Bayas, municipio de Pueblo Nuevo, Durango, México. Bol Soc Mic Madrid 38:33-40 Dahlberg A, Jonsson L, Nylund J-E (1997) Species diversity and distri- bution of biomass above and below ground among ectomycorrhizal fungi in an old-growth Norway spruce forest in south Sweden. Can J Bot 75:1323-1335. doi:10.1139/b97-844 Daly C, Helmer EH, Quiñones M (2003) Mapping the climate of Puerco Rico, Vieques and Culebra. Int J Climatol 23:1359-1381. doi:10. 1002/j0c.937 Douglas N, Manos P (2007) Molecular phylogeny of Nyctaginaceac: taxonomy, biogeography, and characters associated with a radiation of xerophytic genera in North America. Am J Bot 94:856-872. doi: 10.3732/ajb.94.5.856 Estrada-Contreras 1, Equihua M, Castillo-Campos G, Rojas-Soto O (2015) Climate change and effects vegetation in Veracruz, Mexico: an approach using ecological niche modelling. Acta Bot Mex 112: 73-93 Gao C, Zhang Y, Shi NN, Zheng Y, Chen L, Wubet T, Bruelheide H, Both S, Buscot F, Ding Q, Erfmeier A, Kuhn P, Nadrowski K, Scholten T, Guo LD (2015) Community assembly of ectomycorrhizal fungi along a subtropical secondary forest succession. New Phytol 205: 771-785. doi:10.1111/nph.13068 García-Oliva F, Maass JM, Galicia L (1995) Rainstorm analysis and rainfall of a seasonal region with strong cyclonic influence on the Pacific Coast of Mexico. J Appl Meteorol 34:2491-2498. doi:10. 1175/1520-0450 Gardes M, Bruns TD (1993) ITS primers with enhanced specificity for basidiomycetes - application to the identification of mycorrhizas and rusts. Mol Ecol 2:113-118 Y Springer Garibay-Orijel R (2008) Importancia funcional de los hongos ectomicorrizógenos. In: Soberón J, Halffter G, Llorente-Bousquets J (eds) Capital natural de México, vol I, Conocimiento actual de la biodiversidad. CONABIO, México, pp 373-375 Garibay-Orijel R, Morales-Marañón E, Domínguez-Gutiérrez M, Flores- García A (2013) Caracterización morfológica y genética de las ectomicorrizas formadas entre Pinus montezumae y los hongos presentes en los bancos de esporas en la Faja Volcánica Transmexicana. Rev Mex Biodivers 84:153-169. doi:10.7550/ rmb.29839 Harrington TJ, Mitchell DT (2005) Ectomycorrhizas associated with a relict population of Dryas octopetala in the Burren, western Ireland 1 Distribution of ectomycorrhizas in relation to vegetation and soil characteristics. Mycorrhiza 15:425-433. doi:10.1007/500572-005- 0347-4 Haug I, Weiss M, Homeier J, Oberwinkler F, Kottke 1 (2005) Russulaceae and Thelephoraceae form ectomycorrhizas with members of the Nyctaginaceae (Caryophyllales) in the tropical mountain rain forest of southern Ecuador. New Phytol 165:923-936. doi:10.1111/5.1469- 8137.2004.01284.x Hayward JA, Horton TR (2012) Edaphic factors do not govern the ectomycorrhizal specificity of Pisonia grandis (Nyctaginaceae). Mycorrhiza 22:647-652. doi:10.1007/500572-012-0442-2 Hayward J, Horton TR (2014) Phylogenetic trait conservation in the partner choice of a group of ectomycorrhizal trees. Mol Ecol 23: 4886-4898. doi:10.1111/mec.12903 Hynes MH, Smith ME, Zasoski RJ, Bledsoe CS (2010) A molecular survey of ectomycorrhizal hyphae in a California Ouercus-Pinus woodland. Mycorrhiza 20:265-274. doi:10.1007/500572-009- 0281-y Jakucs E, Erós-Honti Z (2008) Morphological-anatomical characteriza- tion and identification of Tomentella ectomycorrhizas. Mycorrhiza 18:277285. doi:10.1007/500572-008-0183-4 Jakucs E, Erós-Honti Z, Seress D, Kovács GM (2015) Enhancing our understanding of anatomical diversity in Tomentella ectomycorrhizas: characterization of six new morphotypes. Mycorrhiza 25:419-429. doi:10.1007/500572-014-0622-3 Jillich W, Stalpers JA (1980) The resupinate non-poroid Aphyllophorales of the temperate Northern Hemisphere. North Holland Publishing Co., Amsterdam Jung HS (1994) Floral studies on Korean wood-rotting fungi II: on the flora of the Aphyllophorales (Basidiomycotina). Korean J Mycology 22:62-99 Kennedy PG, Garibay-Orijel R, Higgins LM, Angeles-Arguiz R (2011) Ectomycorrhizal fungi in Mexican A/nus forests support the host co- migration hypothesis and continental-scale patterns in phylogeography. Mycorrhiza 21:559-568. doi:10.1007/500572- 011-0366-2 Koóljalg U (1996) Tomentella (Basidiomycota) and related genera in Temperate Eurasia. Synopsis Fungorum, Fungiflora Koóljalg U, Jakucs E, Boka K, Agerer R (2001) Three ectomycorrhizae with cystidia formed by different Tomentella species as revealed by rDNA-ITS sequences and anatomical characteristics. Folia Cryptog Estonia 38:27-39 Koóljalg U, Nilsson RH, Abarenkov K, Tedersoo L, Taylor AFS, Bahram M, Bates ST, Bruns TD, Bengtsson-Palme J, Callaghan TM, Douglas B, Drenkhan T, Eberhardt U, Dueñas M, Grebenc T et al (2013) Towards a unified paradigm for sequence-based identifica- tion of fungi. Mol Ecol 22:5271-5277. doi:10.1111/mec.12481 Kress WJ, Erickson DL (2007) A two-locus global DNA barcode for land plants: the coding rbcL gene complements the non-coding trnH- psbA spacer region. PLoS One 2:e508. doi:10.1371/journal.pone. 0000508 Larsen MJ (1974) A contribution to the taxonomy of genus Tomentella. Mycologia Memoirs 4:1-145 174 Mycol Progress (2016) 15:10 Page 11 of 11 10 Larsen MJ (1994) Tomentella oligofibula sp. nov. (Aphyllophorales, Thelephoraceae s. str.) from the Canary Islands. Mycotaxon 63:1-8 Larsson KH, Larsson E, Koóljalg U (2004) High phylogenetic diversity among corticioid Homobasidiomycetes. Mycol Res 108:983-1002. doi:10.1017/50953756204000851 Lott EJ (1993) Annotated checklist of vascular flora of the Chamela Bay region, Jalisco, Mexico. Occasional Paper Calif Acad Sci 148:1-60 Morris MH, Pérez-Pérez MA, Smith ME, Bledsoe CS (2008) Multiple species of ectomycorrhizal fungi are frequently detected on individ- ual oak root tips in a tropical cloud forest. Mycorrhiza 18:375-383. doi:10.1007/500572-008-0186-1 Morris MH, Pérez-Pérez MA, Smith ME, y Bledsoe CS (2009) Influence of host species on ectomycorrhizal communities associated with two co- occurring oaks (Quercus spp.) in a tropical cloud forest. FEMS Microbiol Ecol 69:274-287. doi:10.1111/.1574-6941.2009.00704.x Multer HG, Hoffimeister JE (1968) Subaerial laminated crusts of the Florida Keys. Geol Soc Am Bull 79:183-192 Munsell Color Company (1954) Munsell soil color charts. Munsell Color Company, Baltimore Nilsson R, Kristiansson E, Ryberg M, Hallenberg N, Larsson K-H (2008) Intraspecific IT'S variability in the kingdom Fungi as expressed in the international sequence databases and its implications for molec- ular species identification. Evol Bioinform 4:193-—201 O”Dell TE, Lodge DJ, Mueller GM (2004) Approaches to sampling macrofungi. In: Mueller GM, Bills GF, Foster MS (eds) Biodiversity of Fungi. Inventory and monitoring methods. Elsevier Academic Press, San Diego, pp 163-168 Pagel M, Meade A (2006) Bayesian analysis of correlated evolution of discrete characters by reversible-jump Markov chain Monte Carlo. Am Nat 167:808-825. doi:10.1086/503444 Peay KG, Bruns TD, Kennedy PG, Bergemann SE, Garbelotto M (2007) A strong species-area relationship for eukaryotic soil microbes: is- land size matters for ectomycorrhizal fungi. Ecol Lett 10:470-480. doi:10.1111/;.1461-0248.2007.01035.x Peay KG, Kennedy PG, Bruns TD (2008) Fungal community ecology: a hybrid beast with a molecular master. Bioscience 58:799-810. doi: 10.1641/B580907 Ramírez-López 1, Villegas-Ríos M, Salas-Lizana R, Garibay-Orijel R, Alvarez-Manjarrez J (2015) Thelephora versatilis and Thelephora pseudoversatilis: two new cryptic species with polymorphic basidiomes inhabiting tropical deciduous and sub-perennial forests of the Mexican Pacific coast. Mycologia 107:346-358. doi:10.3852/ 14-151 Reverchon F, Ortega-Larrocea P, Pérez-Moreno J, Peña-Ramírez VM, Siebe C (2010) Changes in community structure of ectomycorrhizal fungi associated with Pinus montezumae across a volcanic soil chronosequence at Sierra Chichinautzin, México. Can J For Res 40:1165-1174. doi:10.1139/X 10-062 Sandoval-Zapotitla E (2005) Técnicas aplicadas al estudio de la anatomía vegetal. Cuadernos 38. Instituto de Biología, Universidad Nacional Autónoma de México, Mexico Séne S, Avril R, Chaintreuil C, Geoffroy A, Ndiaye C, Diédhiou A, Sadio O, Courtecuisse R, Sylla S, Selosse MA, Bá A (2015) Ectomycorrhizal fungal communities of Coccoloba uvifera (L.) L. mature trees and seedlings in the neotropical coastal forests of Guadeloupe (Lesser Antilles). Mycorrhiza. doi:10.1007/500572- 015-0633-8 Setaro SD, Gamica S, Herrera PI, Suárez JP, Góker M (2012) A clustering optimization strategy to estimate species richness of Sebacinales in the tropical Andes based on molecular sequences from distinct DNA regions. Biodivers Conserv 21:2269-2285. doi:10.1007/s10531- 011-0205-y Smith ME, Douhan GW, Rizzo DM (2007) Ectomycorrhizal community structure in a xeric Quercus woodland based on rDNA sequence analysis of sporocarps and pooled roots. New Phytol 174:847— 863. doi:10.1111/.1469-8137.2007.02040.x Smith ME, Henkel TW, Catherine Aime M, Fremier AK, Vilgalys R (2011) Ectomycorrhizal fungal diversity and community structure on three co-occurring leguminous canopy tree species in a Neotropical rainforest. New Phytol 192:699-—712. doi:10.1111/. 1469-8137.2011.03844.x Stalpers JA (1993) The Aphyllophoraceous fungi I. Keys to the species of the Thelephorales. Stud Mycol 35:1-168 Suvi T, Tedersoo L, Abarenkov K, Beaver K, Gerlach J, Kóljalg U (2010) Mycorrhizal symbionts of Pisonia grandis and P. sechellarum in Seychelles: identification of mycorrhizal fungi and description of new Tomentella species. Mycologia 102:522-533. doi:10.3852/09-147 Tedersoo L, Kóljalg U, Hallenberg N, Larsson K-H (2003) Fine scale distribution of ectomycorrhizal fungi and roots across substrate layers including coarse woody debris in a mixed forest. New Phytol 159:153-165. doi:10.1046/¡.0028-646x.2003.00792.x Tedersoo L, May TW, Smith ME (2010a) Ectomycorrhizal lifestyle in fungi: global diversity, distribution, and evolution of phylogenetic lineages. Mycorrhiza 20:217-263. doi:10.1007/500572-009-0274-x Tedersoo L, Sadam A, Zambrano M, Valencia R, Bahram M (2010b) Low diversity and high host preference of ectomycorrhizal fungi in western Amazonia, a neotropical biodiversity hotspot. ISME Journal 4:465-471. doi:10.1038/ismej.2009.131 Tedersoo L, Harend H, Buegger F, Pritsch K, Saar 1, Kóljalg U (2014) Stable isotope analysis, field observations and synthesis experiments suggest that Odontia is a non-mycorrhizal sister genus of Tomentella and Thelephora. Fungal Ecol 11:80-90 Thind KS, Rattan SS (1971) The Thelephoraceae of India. Part 4. The genus Tomentella. Indian Phytopathol 24:32-42 Trowbridge J, Jumpponen A (2004) Fungal colonization of shrub willow roots at the forefront of a receding glacier. Mycorrhiza 14:283-293. doi:10.1007/500572-003-0264-3 Urbizu M, Siqueiros M, Abrego N, Salcedo 1 (2004) Nuevos registros de hongos afiloforoides de Aguascalientes, México y una aproximación a sus preferencias ecológicas. Rev Mex Mic 85: 107-118. doi:10.7550/rmb.35264 Welden AL, Dávalos L, Guzmán G (1979) Segunda lista de los hongos, líquenes y mixomicetos en las regiones de Uxpanapa, Coatzacoalcos, Los Tuxtlas, Papaloapan y Xalapa. Bol Soc Mex Mic 13:151-161 Wu YT, Wubet T, Trogisch S, Both S, Scholten T, Bruelheide H, Buscot F (2013) Forest age and plant species composition determine the soil fungal community composition in a Chinese subtropical forest. PLoS ONE 8, e66829. doi:10.1371/jo0umnal.pone.0066829 Yorou NS, Agerer R (2007) Tomentella furcata, a new species from Benin (West Africa) with basidia forming internal hyphac. Mycol Prog 6:239-247. doi:10.1007/s11557-007-0543-z Yorou NS, Agerer R (2008) Tomentella africana, a new species from Benin (West Africa) identified by morphological and molecular da- ta. Mycologia 100:68-80. doi:10.3852/mycologia.100.1.68 Yorou NS, Koóljalg U, Sinsin B, Agerer R (2007) Studies in African thelephoroid fungi: 1. Tomentella capitata and Tomentella brunneocystidia, two new species from Benin (West Africa) with capitate cystidia. Mycol Prog 6:7-18. doi:10.1007/s11557-006- 0519-4 Yorou NS, Diabate M, Agerer R (2011) Phylogenetic placement and anatomical characterization of two new West African Tomentella (Basidiomycota, Fungi) species. Mycol Prog 11:171-180. doi:10. 1007/511557-011-0739-0 Yorou NS, Diabaté M, Agerer R (2012a) Two new resupinate Thelephorales (Basidiomycota, Agaricomycetes) from Guinea (West Africa). Nova Hedwigia 96:167-180. doi:10.1127/0029- 5035/2012/0056 Yorou NS, Gardt S, Guissou ML, Diabaté M, Agerer R (2012b) Three new Tomentella species from West Africa identified by anatomical and molecular data. Mycol Prog 11:449-462. doi:10.1007/s11557- 011-0760-3 A Springer 175 176 ANEXO 2. Fungal diversity notes 1153–1267: taxonomic and phylogenetic contributions to fungal taxa Resumen Esta es la onceava contribución en la serie de Notas de Diversidad Fúngica sobre la taxonomía fúngica, donde se recolectaron materiales de muchos países, se examinaron y describieron utilizando los métodos de morfología, anatomía, cultivo de cepas, combinados con análisis de secuencias de DNA. Se describen nuevos taxones, que incluyen seis nuevos géneros, 97 nuevas especies, seis nuevas combinaciones, un sinónimo, un nuevo huésped, una nueva especie secuenciada y tres nuevos registros que se acomodaron en 41 familias y 1 incertae sedis en Dothideomycetes. Los nuevos géneros son Amyloceraceomyces, Catenuliconidia, Hansenopezia, Ionopezia, Magnopulchromyces y Pseudothaxteriellopsis. Las nue- vas especies son Amyloceraceomyces angustisporus, Amylocorticium ellipsospo- rum, Arthrinium sorghi, Catenuliconidia uniseptata, Clavulina sphaeropedunculata, Colletotrichum mangiferae-indicae, C. parthenocissicola, Coniothyrium triseptais cusp. subsanguineus, C. xiaojinensis, Diaporthe pimpinellae, Dictyosporella guiz- houensis, Diplodia torilicola, Fuscoporia marquesiana, F. semiarida, Hansenopezia decora, Helicoarctatus thailandicus, Hirsutella hongheensis, Humidicutis brunneo- vinacea, L. variabilis, Lycoperdon lahorense, L. pseudocurtisii, Magnopulchromy- ces scorpiophorus, Moelleriella gracilispora, Neodevriesia manglicola, Neodidyme- lliopsis salvia, N. urticae, Neoroussoella magnoliae, Neottiella Gigaspora, Nigro- grana thailandica, Ophiosphaerella chiangraiensis, Phaeotremella yunnanensis, Podosphaera yulii, Preussia cucurbitae, Pseudothaxteriellopsis obliqus, Rigidopo- rus juniperinus, Rhodofomitopsis pse udofeei, Russula benghalensis, Scleroramu- laria pauciseptata, S. sanyaensis, S. vermispora, Scytinopogon minisporus, Sporo- rmurispora paulsenii, Tomentella asiae-orientalis, T. atrobadia, T. atrocastanea, T. aureo-marginata, T. brevis, T. brevis, T. brevis, T. brevis . brunneoflava, T. brun- neogrisea, T. capitatocystidiata, T. changbaiensis, T. citrinocystidiata, T. coffeae, T. conclusa, T. cystidiata, T. dimidiata, T. duplexa, T. efibulata, T. efibulis, T . fari- nosa, T. flavidobadia, T. fuscocrustosa, T. fuscofarinosa, T. fuscogranulosa, T. 177 fuscopelliculosa, T. globospora, T. gloeocystidiata, T. griseocastanea, T. gri- seofusca, T. griseomarginata, T. inconspicua, T. incrustata , T. interrupta, T. liao- ningen-sis, T. longiaculeifera, T. longiechinuli, T. megaspora, T. olivacea, T. oliva- ceobrunnea, T. pallidobrunnea, T. pallidomarginata, T. parvispora, T. pertenuis, T. qingyuanensis , T. segregata, T. separata, T. stipitata, T. storea, Trichoderma ce- ratophylletum, Tyromyces minu-tulus, Umbelopsis heterosporus y Xylolentia reni- formis. Las nuevas combinaciones son Antrodiella descendena, Rhodofomitopsis monomitica, Rh. oleracea, Fuscoporia licnoides, F. scruposa, Ionopezia gerardii. Se reporta un sinónimo, Chloridium macrocladum (= Gonytrichum mac-rocladum), un nuevo hospedero, Aplosporella prunicola, una nueva especie secuenciada Graphis supracola y tres nuevos registros, Paradictyoarthrinium diffractum, Prost- hemium betulinum y Golovinomyces monardae. 178 Fungal diversity notes 1153–1267: taxonomic and phylogenetic contribu- tions to fungal taxa Hai-Sheng Yuan1,2* · Xu Lu1,2 · Yu-Cheng Dai3* · Kevin D. Hyde4,5,6,7,8* · Yu-He Kan1,2 · Chuan-Gen Lin5,6,7 · Ivana Kusan9 · Ningguo Liu10 · V. Venkateswara Sarma11 · Shuang-Hui He3 · Chang-Lin Zhao12 · Bao-Kai Cui3 · Nousheen Yousaf13 · Fang Wu3 · Milan C. Samarakoon5 · Monika Dayarathne5 · Guang- yu Sun14 · Shu-Yan Liu15 · Tatiana Baptista Gibertoni16 · Lucas B. Conceição17 · Roberto Garibay-Orijel18 · Margarita Villegas-Ríos19 · Rodolfo Salas-Lizana 19 · Tie-Zheng Wei20 · Jun-Zhi Qiu21 · Ze-Fen Yu22 · Rungtiwa Phookamsak4,5,8 · Zeng Ming4,5 · Soumitra Paloi23 · Dan-Feng Bao5,24,25 · Pranami Abeywick- rama5,26 · Yang Jing5 · Ishara Manawasinghe5,26 · Harishchandra5,6,26 · Brah- manage RS5,6,27 · Nimali Indeewari de Silva4,8,28,29 · Danushka S. Tennakoon4,5,6 · Anuruddha Karunarathna5,30 · Gladstone Alves da Silva31 · Yusufjon Gaf- forov1,32,33 · Dhandevi Pem5 · Shengnan Zhang5,30 · Dulanjalee Harishchandra5 · André L. C. M. de Azevedo Santiago34 · Jadson Diogo Pereira Bezerra35 · Bálint Dima36 · Krishnendu Acharya23 · Julieta Alvarez-Manjarrez18, 37 · Ali H. Bahkali37,38 · Vinod K. Bhatt39 · Tor Erik Brandrud40 · Timur S. Bulgakov41 · E. Camporesi42,43,44 · Ting Cao1,2 · Yu-Xi Chen21 · Yuan-Yuan Chen45 · Abdallah M. Elgorban37,38 · Long-Fei Fan3 · Xing Du22 ·Liu Gao14 · Camila Melo Gon- çalves35 · Luis F. P. Gusmão17 · Naruemon Huanraluek5 · Margita Jadan9 · Ru- vishika S. Jayawardena5 · Abdul Nasir Khalid46 · Ewald Langer47 · Diogo X. Lima34 · Nelson Correia de Lima-Júnior48 · Carla Rejane Sousa de Lira16 · Jian-Kui (Jack) Liu49 · Shun Liu3 · Saisamorn Lumyong28,29,50 · Zong-Long Luo5,24,25 · Neven Matočec9 · José Ribamar Costa Oliveira-Filho16 · Viktor Papp51 · Eduardo Pérez-Pazos 19, 52 · Alan J. L. Phillips53 · Peng-Lei Qiu15 · Yihua Ren14 · Rafael F. Castañeda Ruiz54 · Kamal C. Semwal55 · Rejane Maria Ferreira da Silva31 · Karl Soop56 · Carlos A. F. de Souza34 · Cristina Maria Souza-Motta35 · Li-Hua Sun57 · Meng-Le Xie58 · Yi-Jian Yao20 · Li-Wei Zhou1 *Corresponding authors. E-mail: hsyuan@iae.ac.cn; yuchengd@yahoo.com; kdhyde3@gmail.com; Extended author information available on the last page of the article Abstract This is the eleventh contribution in the Fungal Diversity Notes series on the fungal taxon- omy, where materials were collected from many countries, examined and described using the methods of morphology, anatomy, strain culture, combined with DNA sequence anal- yses. Novel taxa are described, including six new genera, 97 new species, six new combina- tions, one synonym, one new host, one new sequenced species and three new records which accommodated in 41 families and 1 incertae sedis in Dothideomycetes. The new genera are 179 Amyloceraceomyces, Catenuliconidia, Hansenopezia, Ionopezia, Magnopulchromyces and Pseudothaxteriellopsis. The new species are Amyloceraceomyces angustisporus, Amylocor- ticium ellipsosporum, Arthrinium sorghi, Catenuliconidia uniseptata, Clavulina sphaero- pedunculata, Colletotrichum mangiferae-indicae, C. parthenocissicola, Coniothyrium tri- septatum, Cortinarius indorusseus, C. paurigarhwalensis, C. sinensis, C. subsanguineus, C. xiaojinensis, Diaporthe pimpinellae, Dictyosporella guizhouensis, Diplodia torilicola, Fuscoporia marquesiana, F. semiarida, Hansenopezia decora, Helicoarctatus thailandicus, Hirsutella hongheensis, Humidicutis brunneovinacea, Lentaria gossypina, L. variabilis, Lycoperdon lahorense, L. pseudocurtisii, Magnopulchromyces scorpiophorus, Moelleriella gracilispora, Neodevriesia manglicola, Neodidymelliopsis salvia, N. urticae, Neoroussoella magnoliae, Neottiella gigaspora, Nigrograna thailandica, Ophiosphaerella chiangraiensis, Phaeotremella yunnanensis, Podosphaera yulii, Preussia cucurbitae, Pseudothaxteriellop- sis obliqus, Rigidoporus juniperinus, Rhodofomitopsis pseudofeei, Russula benghalensis, Scleroramularia pauciseptata, S. sanyaensis, S. vermispora, Scytinopogon minisporus, Sporormurispora paulsenii, Tomentella asiae-orientalis, T. atrobadia, T. atrocastanea, T. aureomarginata, T. brevis, T. brunneoflava, T. brunneogrisea, T. capitatocystidiata, T. changbaiensis, T. citrinocystidiata, T. coffeae, T. conclusa, T. cystidiata, T. dimidiata, T. duplexa, T. efibulata, T. efibulis, T. farinosa, T. flavidobadia, T. fuscocrustosa, T. fuscofar- inosa, T. fuscogranulosa, T. fuscopelliculosa, T. globospora, T. gloeocystidiata, T. griseo- castanea, T. griseofusca, T. griseomarginata, T. inconspicua, T. incrustata, T. interrupta, T. liaoningensis, T. longiaculeifera, T. longiechinuli, T. megaspora, T. olivacea, T. oliva- ceobrunnea, T. pallidobrunnea, T. pallidomarginata, T. parvispora, T. pertenuis, T. qing- yuanensis, T. segregata, T. separata, T. stipitata, T. storea, Trichoderma ceratophylletum, Tyromyces minutulus, Umbelopsis heterosporus and Xylolentia reniformis. The new com- binations are Antrodiella descendena, Rhodofomitopsis monomitica, Rh. oleracea, Fusco- poria licnoides, F. scruposa, Ionopezia gerardii. A synonym, Chloridium macrocladum (= Gonytrichum macrocladum), a new host, Aplosporella prunicola, a new sequenced species Graphis supracola and three new records, Paradictyoarthrinium diffractum, Prosthemium betulinum and Golovinomyces monardae, are reported. Keywords Agaricomycetes · Ascomycota · Basidiomycota · Mucoromycota · New combi- nation · New genus · New species · Phylogeny · Taxonomy Table of contents Mucoromycota Doweld Umbelopsidales Spatafora, Stajich & Bonito Umbelopsidaceae W. Gams & W. Mey. 1153. Umbelopsis heterosporus C.A. de Souza, D.X. Lima & A.L. Santiago, sp. nov Ascomycota Dothideomycetes Botryosphaeriales C.L. Schoch et al. Aplosporellaceae Slippers, Boissin & Crous, Stud. Mycol. 76(1): 41 (2013) 180 1154. Aplosporella prunicola Damm & Crous, Fungal Diversity 27(1): 39 (2007), new host record Botryosphaeriaceae Theiss. & H. Syd., [as 'Botryosphaeriacae'], Annls Mycol. 16(1/2): 16 (1918) 1155. Diplodia torilicola Harishchandra, Camporesi & K.D. Hyde, sp. nov. Capnodiales Woron. Neodevriesiaceae Quaedvl. & Crous, in Quaedvlieg, Binder, Groenewald, Summerell, Car- negie, Burgess & Crous, Persoonia 33: 24 (2014) 1156. Neodevriesia manglicola Devadatha, V.V. Sarma & E.B.G. Jones, sp. nov. Pleosporales Luttr. ex M.E. Barr Coniothyriaceae W.B. Cooke, Revta Biol., Lisb. 12: 289 (1983) 1157. Coniothyrium triseptatum Dayarathne, Thyagaraja & K.D. Hyde, sp. nov. Didymellaceae Gruyter, Aveskamp & Verkley, Mycol. Res. 113(4): 516 (2009) 1158. Neodidymelliopsis salviae Brahmanage, Camporesi & K.D. Hyde, sp. nov. 1159. Neodidymelliopsis urticae Manawasinghe, Camporesi & K.D. Hyde, sp. nov. Lophiostomataceae Sacc. [as 'Lophiostomaceae'], Syll. Fung. (Abellini) 2: 672 (1883) 1160. Magnopulchromyces L.B. Conc., Gusmão & R.F. Castañeda, gen. nov. 1161. Magnopulchromyces scorpiophorus L.B. Conc., Gusmão & R.F. Castañeda, sp. nov. Nigrogranaceae Jaklitsch & Voglmayr, Stud. Mycol. 85: 54 (2016) 1162. Nigrograna thailandica Samarak. & K.D. Hyde, sp. nov. Paradictyoarthriniaceae Doilom, Jian K. Liu & K.D. Hyde, in Liu et al., Fungal Diversity: 10.1007/s13225-015-0324-y, [133] (2015) 1163. Paradictyoarthrinium diffractum Matsush., Matsush. Mycol. Mem. 9:18 (1996) new host and record from India Phaeosphaeriaceae M.E. Barr, Mycologia 71(5): 948 (1979) 1164. Ophiosphaerella chiangraiensis Phookamsak & K.D. Hyde, sp. nov. Pleomassariaceae M.E. Barr, Mycologia 71(5): 949 (1979) 1165. Prosthemium betulinum Kunze, in Kunze & Schmidt, Mykologische Hefte (Leipzig) 1: 18 (1817) new record from Italy Roussoellaceae J.K. Liu, R. Phookamsak, D.Q. Dai & K.D. Hyde, Phytotaxa 181: 7 (2014) 1166. Neoroussoella magnoliae N.I. de Silva & K.D. Hyde, sp. nov. Sporormiaceae Munk, Dansk Botanisk Arkiv 17 (1): 450 (1957) 1167. Preussia cucurbitae R.M.F. Silva & G.A. Silva, sp. nov. 1168. Sporormurispora paulsenii D.Pem, Y. Gafforov & K.D. Hyde, sp. nov. Tubeufiales Boonmee & K.D. Hyde 181 Tubeufiaceae M.E. Barr, Mycologia 71(5): 948 (1979) 1169. Helicoarctatus thailandicus D.F. Bao, Z.L. Luo, K.D. Hyde & H.Y. Su, sp. nov. 1170. Pseudothaxteriellopsis M. Niranjan & V.V. Sarma, gen. nov. 1171. Pseudothaxteriellopsis obliqus M. Niranjan and V.V. Sarma, sp. nov. Incertae sedis in Dothideomycetes 1172. Scleroramularia pauciseptata G.Y. Sun & L. Gao, sp. nov. 1173. Scleroramularia sanyaensis G.Y. Sun & L. Gao, sp. nov. 1174. Scleroramularia vermispora G.Y. Sun & L. Gao, sp. nov. Lecanoromycetes O.E. Erikss. & Winka, Myconet 1(1): 7 (1997) Ostropales Nannf. Graphidaceae Dumort., [as 'Graphineae'], Comment. Bot. (Tournay): 69 (1822) 1175. Graphis supracola A.W. Archer, Aust. Syst. Bot. 14(2): 267 (2001), new sequenced species from Thailand Leotiomycetes Erysiphales Gwynne-Vaughan & Barnes Erysiphaceae Tul. & C. Tul., Select. Fung. Carpol. 1: 191 (1861) 1176. Podosphaera yulii S.-Y. Liu & P.-L. Qiu, sp. nov., sp. nov. 1177. Golovinomyces monardae (G.S. Nagy) M. Scholler, U. Braun & Anke Schmidt, in Scholler, Schmidt, Siahaan, Takamatsu & Braun, Mycol. Progr. 15(no. 56): 4 (2016) new record from China Pezizomycetes O.E. Erikss. & Winka Pezizales J. Schröt. Pezizaceae Dumort., Syst. Mycol. (Lundae) 3(1): 72 (1829) 1178. Ionopezia Matočec, I. Kušan & Jadan, gen. nov. 1179. Ionopezia gerardii (Cooke) Matočec, I. Kušan & Jadan, comb. nov. 1180. Hansenopezia Matočec, I. Kušan & Jadan, gen. nov. 1181. Hansenopezia decora Matočec, I. Kušan & Jadan, sp. nov. Pyronemataceae Corda, [as 'Pyronemaceae'], Anleit. Stud. Mykol., Prag: 149 (1842) 1182. Neottiella gigaspora M. Zeng, Q. Zhao & K.D. Hyde, sp. nov. Sordariomycetes Amphisphaeriales D. Hawksw. & O.E. Erikss. Apiosporaceae K.D. Hyde, J. Fröhl., Joanne E. Taylor & M.E. Barr, Sydowia 50 (1): 23 (1998) 1183. Arthrinium sorghi J.D.P. Bezerra, C.M Gonçalves & C.M. Souza-Motta, sp. nov. 182 Chaetosphaeriales Huhndorf, A.N. Mill. & F.A. Fernández Chaetosphaeriaceae Réblová, M.E. Barr & Samuels, Sydowia 51(1): 56 (1999) 1184. Chloridium macrocladum (Sacc.) S. Hughes, Trans. Br. Mycol. Soc. 34(4): 565 (1952) [1951], synonym Diaporthales Nannf. Diaporthaceae Höhn. ex Wehm., Am. J. Bot. 13: 638 (1926) 1185. Diaporthe pimpinellae Abeywickrama, Camporesi, Dissanayeke & K.D. Hyde, sp. nov. Hypocreales Lindau Clavicipitaceae Kirk, Cannon, Minter & Stalpers, Dictionary of the fungi (2008) 1186. Moelleriella gracilispora Jun Z. Qiu, Y.X. Chen & Q.L. Xue, sp. nov. Hypocreaceae De Not., G. Bot. Ital 2(1): 48 (1844) 1187. Trichoderma ceratophylletum Z.F. Yu & X. Du, sp. nov. Ophiocordycipitaceae G.H. Sung, J.M. Sung, Hywel-Jones & Spatafora, in Sung, Hywel- Jones, Sung, Luangsa-ard, Shrestha & Spatafora, Stud. Mycol. 57: 35 (2007) 1188. Hirsutella hongheensis D.P. Wei & K.D. Hyde, sp. nov. Glomerellales Chadef. ex Réblová et al. Glomerellaceae Locq. ex Seifert & W. Gams, in Zhang, Castlebury, Miller, Huhndorf, Schoch, Seifert, Rossman, Rogers, Kohlmeyer, Volkmann-Kohlmeyer & Sung, Mycologia 98(6): 1083 (2007) [2006] 1189. Colletotrichum mangiferae-indicae Jayawardena, Huanraleuk & K.D. Hyde, sp. nov. 1190. Colletotrichum parthenocissicola Jayawardena, Bulgakov, Huanraleuk & K.D. Hyde, sp. nov. Junewangiaceae J.W. Xia & X.G. Zhang, Scientific Reports 7: 7888 (2017) 1191. Dictyosporella guizhouensis J. Yang & K.D. Hyde, sp. nov. Rhamphoriaceae Réblová, in Réblová & Štěpánek, Mycologia: 10.1080/00275514.2018.1475164, 5 (2018) 1192. Xylolentia reniformis C.G. Lin, K.D. Hyde & J.K. Liu, sp. nov. Xylariales Nannf. Xylariaceae Tul. & C. Tul. [as 'Xylariei'], Select. Fung. Carpol. (Paris) 2: 3 (1863) 1193. Catenuliconidia N.G. Liu & K.D. Hyde, gen. nov. 1194. Catenuliconidia uniseptata N.G. Liu & K.D. Hyde, sp. nov. Basidiomycota Agaricomycetes Agaricales Underw. Agaricaceae Chevall. , Fl. Gén. Env. Paris (Paris) 1: 121 (1826) 183 1195. Lycoperdon lahorense Yousaf & Khalid, sp. nov. 1196. Lycoperdon pseudocurtisii Yousaf & Khalid, sp. nov. Cortinariaceae R. Heim ex Pouzar, Česká Mykol. 37(3): 174 (1983) 1197. Cortinarius indorusseus Dima, Semwal, V.K. Bhatt & Brandrud, sp. nov. 1198. Cortinarius paurigarhwalensis Semwal, Dima & Soop, sp. nov. 1199. Cortinarius sinensis L.H. Sun, T.Z. Wei & Y.J. Yao, sp. nov. 1200. Cortinarius subsanguineus T.Z. Wei, M.L. Xie & Y.J. Yao, sp. nov. 1201. Cortinarius xiaojinensis T.Z. Wei, M.L. Xie & Y.J. Yao, sp. nov. Hygrophoraceae Lotsy, Vortr. Bot. Stammesgesch. 1: 705 (1907) Humidicuteae Padamsee & Lodge, Fungal Diversity 64: 38 (2014) 1202. Humidicutis brunneovinacea R. Garibay-Orijel, sp. nov. Amylocorticiales K.H. Larss., Manfr. Binder & Hibbett Amylocorticiaceae Jülich, Biblthca Mycol. 85: 354 (1982) 1203. Amyloceraceomyces S.H. He, gen. nov. 1204. Amyloceraceomyces angustisporus S.H. He, sp. nov. 1205. Amylocorticium ellipsosporum S.H. He, sp. nov. Cantharellales Gäum Clavulinaceae Donk, Beih. Nova Hedwigia 1(4): 407 (1970) 1206. Clavulina sphaeropedunculata E. Pérez-Pazos, M. Villegas-Ríos & R. Garibay- Orijel, sp. nov. Gomphales Jülich Lentariaceae Jülich, Bibliotheca Mycologica 85: 375 (1982) 1207. Lentaria gossypina R. Salas-Lizana, M. Villegas-Ríos & E. Pérez-Pazos, sp. nov. 1208. Lentaria variabilis M. Villegas-Ríos , R. Garibay-Orijel & N. Matías-Ferrer, sp. nov. Hymenochaetales Oberw. Hymenochaetaceae Donk, Bull. Bot. Gdns Buitenz. 17(4): 474 (1948) 1209. Fuscoporia licnoides (Mont.) JRC Oliveira-Filho & Gibertoni, comb. nov. 1210. Fuscoporia marquesiana Gibertoni & Lira, sp. nov. 1211. Fuscoporia scruposa (Mont.) Gibertoni & JRC Oliveira-Filho, comb. nov. 1212. Fuscoporia semiarida Lima-Júnior, Lira & Gibertoni, sp. nov. 1213. Rigidoporus juniperinus Gafforov, L.W. Zhou, E. Langer, sp. nov. Polyporales Gäum. Fomitopsidaceae Jülich, Biblthca Mycol. 85: 367 (1981) 1214. Rhodofomitopsis pseudofeei B.K. Cui & Shun Liu, sp. nov. 1215. Rhodofomitopsis monomitica (Yuan Y. Chen) B.K. Cui, Yuan Y. Chen & Shun Liu, comb. nov. 1216. Rhodofomitopsis oleracea (R.W. Davidson & Lombard) B.K. Cui, Yuan Y. Chen & Shun Liu, comb. nov. 184 Polyporaceae Fr. ex Corda [as 'Polyporei'], Icon. Fung. (Prague) 3: 49 (1839) 1217. Antrodiella descendena (Corner) C.L. Zhao & Y.C. Dai, comb. nov. 1218. Tyromyces minutulus Y.C. Dai & C.L. Zhao, sp. nov. Russulales Kreisel ex P.M. Kirk, P.F. Cannon & J.C. David Russulaceae Lotsy 1219. Russula benghalensis Paloi & K. Acharya, sp. nov. Thelephorales Corner ex Oberw. Thelephoraceae Chevall. [as 'Thelephoreae'], Fl. gén. env. Paris (Paris) 1: 84 (1826) 1220. Tomentella asiae-orientalis H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1221. Tomentella atrobadia H.S. Yuan & Y.C. Dai, sp. nov. 1222. Tomentella atrocastanea H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1223. Tomentella aureomarginata H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1224. Tomentella brevis H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1225. Tomentella brunneoflava H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1226. Tomentella brunneogrisea H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1227. Tomentella capitatocystidiata H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1228. Tomentella changbaiensis H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1229. Tomentella citrinocystidiata H.S. Yuan & Y.C. Dai, sp. nov. 1230. Tomentella coffeae H.S. Yuan & Y.C. Dai, sp. nov. 1231. Tomentella conclusa H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1232. Tomentella cystidiata H.S. Yuan & Y.C. Dai, sp. nov. 1233. Tomentella dimidiata H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1234. Tomentella duplexa H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1235. Tomentella efibulata H.S. Yuan & Y.C. Dai, sp. nov. 1236. Tomentella efibulis H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1237. Tomentella farinosa H.S. Yuan & Y.C. Dai, sp. nov. 1238. Tomentella flavidobadia H.S. Yuan & Y.C. Dai, sp. nov. 1239. Tomentella fuscocrustosa H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1240. Tomentella fuscofarinosa H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1241. Tomentella fuscogranulosa H.S. Yuan & Y.C. Dai, sp. nov. 1242. Tomentella fuscopelliculosa H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1243. Tomentella globospora H.S. Yuan & Y.C. Dai, sp. nov. 1244. Tomentella gloeocystidiata H.S. Yuan & Y.C. Dai, sp. nov. 1245. Tomentella griseocastanea H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1246. Tomentella griseofusca H.S. Yuan & Y.C. Dai, sp. nov. 1247. Tomentella griseomarginata H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1248. Tomentella inconspicua H.S. Yuan & Y.C. Dai, sp. nov. 1149. Tomentella incrustata H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1250. Tomentella interrupta H.S. Yuan & Y.C. Dai, sp. nov. 1251. Tomentella liaoningensis H.S. Yuan & Y.C. Dai, sp. nov. 1252. Tomentella longiaculeifera H.S. Yuan & Y.C. Dai, sp. nov. 1253. Tomentella longiechinuli H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1254. Tomentella megaspora H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1255. Tomentella olivacea H.S. Yuan & Y.C. Dai, sp. nov. 1256. Tomentella olivaceobrunnea H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 185 1257. Tomentella pallidobrunnea H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1258. Tomentella pallidomarginata H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1259. Tomentella parvispora H.S. Yuan & Y.C. Dai, sp. nov. 1260. Tomentella pertenuis H.S. Yuan & Y.C. Dai, sp. nov. 1261. Tomentella qingyuanensis H.S. Yuan & Y.C. Dai, sp. nov. 1262. Tomentella segregata H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1263. Tomentella separata H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1264. Tomentella stipitata H.S. Yuan, X. Lu & Y.C. Dai, sp. nov. 1265. Tomentella storea H.S. Yuan & Y.C. Dai, sp. nov. Trechisporales K.H. Larsson Clavariaceae Chevallier, Flore Générale des Environs de Paris: 102 (1826) 1266. Scytinopogon minisporus J. Alvarez-Manjarrez, M. Villegas-Ríos & R. Garibay- Orijel, sp. nov. Tremellomycetes Tremellales Fr. Phaeotremellaceae A.M. Yurkov & Boekhout, Stud. Mycol. 81: 137 (2015) 1267. Phaeotremella yunnanensis L.F. Fan, F. Wu & Y.C. Dai, sp. nov. … Trechisporales K.H. Larsson Notes: The order Trechisporales was proposed by Larsson (2007) and accepted in Agaricomycetes based on phylogenetic analysis (Binder et al. 2005; Larsson et al. 2004; Matheny et al. 2007). It includes stipitate, clavariod and resupinate genera with smooth, hydnoid, grandiniod or poroid hymenia. Hyphae with clamps, forming a monomitic system, and the subicular hyphae with or without ampullate septa. Basidia have four to six sterig- mata, and their basidiospores are smooth or ornamented. Some species have cystidia. They grow mostly on wood or soil (Hibbett et al. 2007). Clavariaceae Chevallier, Flore Générale des Environs de Paris: 102 (1826) Notes: The family Clavariaceae was proposed by Chevallier (1826), and phylogenetic analysis included a variety of fruitbody shapes: species with simple clubs, coralloid, hyd- noid, stipitate with lamella and resupinate. The genus type is Clavaria, recently discovered as paraphyletic (Birkebak et al. 2013). Other genera included in this family are Ca- marophyllopsis, Clavulinopsis, Clavicorona, Hyphodontiella, Mucronella, Ramariopsis, Scytinopogon and Setigeroclavula. However, Scytinopogon is related to Trechispora which is included in Hydnodontaceae (Birkebak et al. 2013; Desjardin & Perry, 2015). Jülich (1981) proposed its own family Scytinopogonaceae Jülich, and segregated Trechispora and Hydnodon to Hydnodontaceae Jülich, nonetheless Scytinopogon is still considered within Clavariaceae. 186 Scytinopogon Singer, New genera of fungi. Lloydia 8:139-144 (1945) Notes: The genus Scytinopogon was proposed by Singer in 1945, with pallid, cream, alutaceous, tan, tinged pink or purple, or white fruit bodies with flat branches in one plane, dilatating before branching, polychotomous in the first divisions and dichotomous near to the tips, also it can be dichotomous in slim specimens. This genus includes 10 species (www.indexfungorum.org). Scytinopogon minisporus J. Alvarez-Manjarrez, M. Villegas & R. Garibay-Orijel, sp. nov. MycoBank number: MB829209; Facesoffungi number: FoF 05672; Fig. 258 Etimology: Refers to small-sized basidiospores. Holotype: MEXU 28300. Basidiocarps clavarioid 15–40 mm height, branches are 15–17 mm, stipe 5–15 × 1–3 mm, acute to sub-rounded, flat, and whitish or even gray (5A2–5B3), axils rounded. Branching near to the stipe is polytomical and near to the tips can be dichotomical to poly- tomical. The stipe is cylindrical slightly flat, whitish to pale orange brown (5A2–5A3). The base of the stipe is covered by numerous short hyphae looking plushy. Surface of all the fruit body, except the base of the stipe, looks smooth at naked eye, and dusty under stereo- scopic microscope. Consistency is cartilaginous to alutaceous. Flesh of middle portion has the same color as the surface. Odor indistinguishable, taste slightly astringent. Basidio- spores 4.2–5(–5.6) × 2.1–2.8(–3) µm [xm = 4.6 ± 0.4 × 2.4 ± 0.3 µm, Q = 1.3–1.6, n = 30], dacryoid with lateral view slightly elliptical, hyaline, thin wall, verrucose where sometimes the warts merge, without conical spines; plage has not ornamentation, and lateral hilar ap- pendix, dimly cyanophylic. Basidia mostly tetrasporic, scarce, (10–)20.3–28 × 4.9–5.6 µm, cylindrical to subcylindrical, hyaline, thick and smooth wall, base with clamp connection. Sterigmata 2.8–4.2 × 1.4 µm, hyaline, straight, acute apex. Cystidia clavate incrusted on the tip, they do not dispel with KOH, very scarce. Subhymenial with monomitic hyphae, gen- erative hyphae of 1.4–3.5 × 50–56 µm, with thickened wall (< 1 µm), septae with clamp connections, and H connections between hyphae. Tramal hyphae strictly parallel, (2.1– )2.8–3.5 µm, thickened wall (<1 µm), septate with clamp connections, and hyphae with H connections. Stipe with generative hyphae, 1.4–2.1 µm width, hyaline, wall slightly thick- ened, frequent septa with clamp connections, and abundant crystals on their surface form- ing irregular plates. Habitat and known distribution: Gregarious or solitary, growing on soil or debris of tropical dry forest from the Pacific coast of Jalisco, Mexico. Material examined: MEXICO, Jalisco, municipality La Huerta, Estación de Biología de Chamela (EBCH), pathway Camino Antiguo, October 9 th 2005, Villegas Ríos M. 2630 (FCME 26014); pathway Chachalaca, August 11 th 2006, Villegas Ríos M. 2672 (FCME 26015); pathway Camino Antiguo Sur, August 11 th 2006, Aguirre, Bautista and Pulido II- 40 (MEXU 26345); October 1 st 1977, A. Pérez J. and A. Solís M. (MEXU 11923); pathway Buho, October 10 th 2015, Alvarez-Manjarrez AM170 (MEXU 28300, holotype) (GenBank ITS: MK328885; LSU: MK328894); pathway Buho, October 18 th 2015, collected by Alva- rez-Manjarrez AM176 (MEXU 28301, isotype) (GenBank ITS: MK328886; LSU: MK328895). Notes: Scytinopogon minisporus differs from other species of Scytinopogon in having clavate cystidia with crystals on the tip. Microscopically it is very similar to S. scaber (Berk. & M.A. Curtis) D.A. Reid 1962, however S. minisporus has verrucose spores, bigger 187 basidia and encrusted cystidia. Scytinopogon papillosus Corner 1970 also coincides in spore size, but has minute papillae in the surface of the fruitbody. Phylogenetic analyses of the combined ITS and LSU dataset reveal S. minisporus as an independent branch within Trechispora sister of Trechispora bispora (Fig. 259). It is important to remark that Trechis- pora and Scytinopogon, both are paraphyletic, belonging to the same clade (Birkebak et al. 2013; Desjardin and Perry 2015). The type sequences of both genera are necessary to de- termine if the resupinate Trechispora and coralloid Scytinopogon are synonyms (Desjardin and Perry 2015). Based on ITS and LSU phylogenies Scytinopogon should be transferred to Hydnodontaceae together with Trechispora and Hydnodon. Fig. 258 Scytinopogon minisporus (MEXU 28300, holotype). a–b Basidiomes of Scytino- pogon minisporus. c Basidiospores with verrucose ornamentation, showing the plage with- out ornamentation (spores from the holotype). d Ornamentation of spores (MEXU 28301, 188 isotype). e Immature basidiospores attached to a tetrasterigmata basidia. f Clavate cystidia with crystals on the tip. Scale bars: b = 5 mm, c–e = 1.5 µm, f = 4 µm Fig. 259 Bayesian analysis combined dataset of ITS and LSU sequence data of the Trechis- pora-Scytinopogon clade, with bipartitions and mixed substitution model. The node support is indicated as ML/PP. The tree is rooted with Brevicellicium atlanticum (NR_119820 and HE963773) and B. olivascens (JN649327). The type sequences are indicated in blue bold. Bayesian analysis (Mr. Bayes) of the Trechispora-Scytinopogon clade showing the phylo- genetic position of Scytinopogon minisporus inferred from ITS and LSU sequences. Indi- vidual alignments were done in MAFFT (Katoh et al. 2017); concatenated alignment was assembled in Mesquite. The Maximum Likelihood (ML) analysis was performed using the GTR+ gamma substitution model, with 1000 bootstrap replicates. Bayesian analysis ap- plied mix models for two partitions (ITS and LSU) with 5,000,000 iterations. The node support is indicated as ML bootstrap / Bayesian posterior probabilities respectively. New sequences are indicated in blue bold. …