Post on 27-Dec-2019
Centro de Investigación en Alimentación y Desarrollo, A.C.
ANÁLISIS TRANSCRIPTÓMICO DE EXOCARPO DE MANGO (Mangifera Indica L.) Y GENES QUE
PARTICIPAN EN LA BIOSÍNTESIS DE CUTÍCULA
Por:
Julio César Tafolla Arellano
TESIS APROBADA POR LA
COORDINACIÓN DE TECNOLOGÍA DE ALIMENTOS DE ORIGEN VEGETAL
Como requisito parcial para obtener el grado de:
DOCTOR EN CIENCIAS
Hermosillo, Sonora, México. Enero de 2015
iv
AGRADECIMIENTOS
Al Consejo Nacional de Ciencia y Tecnología (CONACYT) por la beca otorgada
para el doctorado y estancia en la Universidad Cornell.
Al Centro de Investigación en Alimentación y Desarrollo A.C. por todo el apoyo
y facilidades para llevar a cabo esta tesis doctoral.
Esta investigación fue financiada por el proyecto 20120 (P0045001):
Aseguramiento de Calidad De Frutas y Hortalizas del Centro de Investigación
en Alimentación y Desarrollo A.C.
A mi comité de tesis: Dr. Martín Ernesto Tiznado Hernández, Dr. Reginaldo
Baéz Sañudo, Dr. Alberto González León y Dr. Lorenzo Zacarías García por
aceptar mi propuesta de investigación, por su amistad y dirección de este
trabajo de investigación.
Al M.C. Javier Ojeda Contreras y M.C Alberto Sanchéz Estrada por su apoyo
incondicional para llevar a cabo esta investigación.
A Agrícola Daniella por facilitarnos sus instalaciones y materia prima para llevar
a cabo esta investigación.
A mis compañeros del laboratorio de Fisiología y Biología Molecular de Plantas:
Guillermo, Veronica, Rigel, Heriberto, Eduardo, Miguel y Alexel por su apoyo,
comentarios y observaciones para culminar esta tesis doctoral.
Al laboratorio de Fisiología de frutos y por todas la facilidades durante esta
investigación.
v
Al laboratorio de Nutrigenómica Molecular de la Dra. Silvia Moya, Dra. Maricela
Montalvo por todas la facilidades durante esta investigación.
Al Dr. Jesús Hernández y Q.B. Monica Resendiz por las facilidades y acceso a
sus equipos.
Al Dr. Jocelyn K. C. Rose de la Universidad Cornell por aceptarme en su
laboratorio, por el apoyo científico y de sus instalaciones para realizar el análisis
transcriptómico de mango.
A los miembros de Rose Lab: Sungjin Park, Eliel Ruiz May, Laetitia Martin, Eric
Fich, Iben Sørensen y Stephen Snyder por su asesoría y ayuda durante mi
estancia en la Universidad Cornell.
A Ricardo Morales y familia, Carlos Tzul por su gran amistad y apoyo durante
mi estancia en la Universidad Cornell.
Al Dr. Jesús Robles Parra, Dr. Martín Preciado y M.C Andrés Beltrán por su
asesoría y amistad durante esta investigación.
vi
DEDICATORIA
A mi esposa María Antonieta Rodríguez Ibarra e hija Elena por todo el sacrificio,
apoyo y comprensión durante esta etapa.
A mis padres René Tafolla y Ma. Félix Arellano y hermanos René, Raúl, María
Félix, Xochitl y Aurelia por todo el apoyo que me han brindado.
A mis hermanos Victor Hugo (†) y Arturo (†) que fueron una motivación para
seguir adelante con esta tesis doctoral.
A Silvia Elena Ibarra, Agustín, Silvia, Cristina, Francisco Rodriguez por
permitirme formar parte de su familia y todo el apoyo durante esta etapa.
vii
RESUMEN
El fruto de mango es altamente perecedero debido a su limitada vida de
anaquel, principalmente por la pérdida de peso y senescencia, lo cual reduce
la integridad del tejido e incrementa la infección microbiana; factores que limitan
su comercialización. Análisis recientes realizados en tomate sugieren que esas
características son influenciadas por la expresión de diferentes genes que están
relacionados con la biosíntesis de cutícula del fruto. Sin embargo, en el caso del
mango, el conocimiento del mecanismo molecular de biosíntesis de cutícula se
encuentra limitado por la falta de información de su genoma.
El objetivo de esta investigación fue correlacionar la expresión de genes que
participan en la biosíntesis con la formación de cutícula durante la ontogenia del
fruto de mango. Se realizó una investigación documental sobre la composición,
fisiología y biosíntesis de la cutícula en plantas que permitió proponer un
modelo del mecanismo molecular de biosíntesis. Se determinaron los cambios
en la cantidad de cutícula durante la ontogenia del fruto de mango. Asimismo,
se analizó el transcriptoma de exocarpo de mango maduro y senescente
mediante RNA-Seq y se realizó un análisis de los cambios en la expresión de
genes que fueron identificados que participan en la biosíntesis de cutícula. Se
generaron aproximadamente 400 millones de pares de lecturas y fueron
ensambladas de novo en 107,744 unigenes, con una longitud promedio de
1,717 bp. De estos, se identificaron 91,736 unigenes mostrando homología a
proteínas en la base de datos UniProt/TrEMBL. Estos participan principalmente
en el metabolismo de lípidos, cutina, metabolitos secundarios y polisacáridos de
la pared celular. Además, se identificó que la biosíntesis de monómeros de
cutina es una ruta metabólica enriquecida durante la maduración. La biosíntesis
viii
de cutícula mostró un patrón bifásico con una mayor acumulación durante la
maduración y senescencia del fruto. Este comportamiento correlaciona con la
expresión de los genes analizados. Con esta información, se propuso un
modelo molecular de biosíntesis de cutícula incluyendo todos los genes
analizados y un modelo donde se describe la función de un gen que codifica
para una proteína transportadora de lípidos.
Los genes analizados en esta investigación constituyen la primera evidencia
experimental que apoyará la elucidación del mecanismo molecular de la
biosíntesis de cutícula en mango. Los resultados de este estudio proporcionan
un recurso genómico de gran importancia que permitirá el diseño de estrategias
para aumentar la vida postcosecha del mango.
Palabras clave: Mango, exocarpo, cutícula, biosíntesis, RNA-Seq,
transcriptoma.
ix
ABSTRACT
Mango fruit is highly perishable due to a limited shelf life, mainly because of
desiccation and senescence, which leads to the loss of tissue integrity and
microbial infection; factors that limit its commercialization. Recent analyses in
tomato suggest that these traits are influenced by the expression of genes
playing role in the cuticle biosynthesis of fruit. . However, in mango, the
knowledge about the molecular mechanism of cuticle biosynthesis is rather
small due to the lack of genome data.
The objective of this research was to correlate the expression of genes playing a
role in the cuticle biosynthesis with the formation of cuticle during mango fruit
ontogeny. It was carried out a literature review on the composition, physiology
and cuticle biosynthesis in plants, which allowed the creation of a model about
the molecular mechanism of biosynthesis. It was evaluated the changes of
cuticle accumulation during mango fruit ontogeny. Besides, it was carried out the
analysis of ripe and overripe mango peels transcriptome using RNA-Seq and the
profile of cuticle-related gene expression. Approximately 400 million reads pairs
were generated and de novo assembled into 107,744 unigenes, with a mean
length of 1,717 bp. Out of these, a total of 91,736 unigenes showed homologous
to proteins in the UniProt/TrEMBL database. These unigenes are mainly playing
a role in the metabolism of lipids, cutin, secondary metabolites and cell wall
polysaccharides. Also, it was found that that cutin monomers biosynthesis
pathway is enriched during ripening. The cuticle biosynthesis showed a biphasic
pattern of cuticle deposition with an increased accumulation during fruit ripening
and senescence. This behavior correlates with the expression of analyzed
genes. With this information, it was proposed a molecular model of cuticle
x
biosynthesis involving all the genes analyzed and a model describing the role of
one gene encoding a lipid transfer protein. The genes analyzed in this research
constitute the first experimental evidence that will help in the elucidation of the
molecular mechanism of mango cuticle biosynthesis. The results of this study
provide a valuable genomic resource, which will help to design of strategies with
the goal to increase the postharvest shelf life of mango.
Keywords: Mango, fruit peel, cuticle, biosynthesis, RNA-Seq, transcriptome.
xi
CONTENIDO
Página
Resumen………………………………………………………..………....….. vii Abstract………………………………………………………………………… ix Sinopsis………………………………………………………………………… 1 Capítulo I. …………………………………………………………………….. 10 Composición, fisiología y biosíntesis de la cutícula en plantas. Publicado: Revista Fitotecnia Mexicana 2013, 36:3-12.
Capítulo II……………………………………………………………………... 22
Transcriptome Analysis of Mango (Mangifera indica L) Fruit Peel: First Insights Towards Understanding Cuticle Biosynthesis. En revisión de autores. Preparado para: Journal of Experimental Botany. Capítulo III…………………………………………………………………….. 68 Gene expression of a putative glycosylphosphatidylinositol-anchored lipid transfer protein 2 during cuticle biosynthesis in mango. Enviado: Revista Fitotecnia Mexicana.
1
SINOPSIS
El fruto de mango (Mangifera indica L.) es altamente perecedero y tiene
limitada vida de anaquel, lo cual aumenta las pérdidas postcosecha y reduce la
posibilidad de comercialización del producto en los mercados internacionales.
Este comportamiento depende de las condiciones ambientales a las que son
sometidos los frutos. La interfase entre el medio ambiente y el fruto es la
cutícula, que es la capa más externa de las células vegetales que interacciona
con el ambiente, la cual es una estructura producto de la evolución de las
plantas superiores que las aísla y protege de estreses bióticos y abióticos. La
cutícula cubre las partes aéreas de las plantas superiores, incluyendo hojas,
tallos, flores y frutos, es sintetizada por las células epidérmicas, y está
compuesta principalmente de cutina y ceras.
La cutícula tiene como principal función controlar la pérdida de agua y difusión
de gases, además, proporciona protección contra los insectos, patógenos, la
radiación UV, mantiene la palatabilidad y promueve la dispersión de semillas,
entre otras funciones. Los estudios cuticulares en frutas son importantes desde
la perspectiva fisiológica y económica debido a que el control de la calidad del
fruto y la vida postcosecha lo realiza reduciendo la pérdida de agua, la infección
microbiana y evitando desórdenes fisiológicos.
La importancia funcional de la cutícula es evidenciada por la gran energía que
utilizan las células epidérmicas para realizar la biosíntesis de cutícula. De
acuerdo a lo mencionado, más de la mitad de los ácidos grasos sintetizados por
las células epidérmicas durante la expansión del tallo en Arabidopsis son
utilizados en la formación de lípidos cuticulares. Además, las células
epidérmicas presentan un aumento en la expresión de genes que codifican
proteínas implicadas en el metabolismo de lípidos.
2
La mayoría de los estudios sobre la composición y ultraestructura de la cutícula
han sido descriptivos, comparativos, y es relativamente poco lo que se conoce
acerca de la biosíntesis, transporte e interacción de los compuestos cuticulares
para formar la cutícula.
La síntesis de cera requiere la coordinación de actividades de un gran número
de enzimas que están organizadas en complejos multienzimáticos en diferentes
compartimentos celulares (cloroplastos retículo endoplasmático y citoplasma)
donde se lleva a cabo la síntesis y elongación de los ácidos grasos, precursores
de las ceras y la formación de una multitud de compuestos alifáticos. En este
sentido se conocen diversas rutas metabólicas que participan en la biosíntesis
de los compuestos cuticulares, aunque los mecanismos de transporte siguen
siendo poco conocidos.
La biosíntesis de las ceras implica tres distintas etapas: primero, los ácidos
grasos de 16 y 18 carbonos son sintetizados de novo en los cloroplastos. Una
vez sintetizados en los cloroplastos, se transportan al retículo endoplasmático
para su elongación a ácidos grasos de cadenas muy largas como alcoholes,
ésteres, aldehídos, alcanos y cetonas mediante dos vías: reducción y
descarboxilación. Finalmente, la tercer etapa de la biosíntesis de la cutícula
requiere el transporte de los monómeros de cutícula de las células epidérmicas
al exterior de la pared celular.
Los resultados de las investigaciones revisadas concluyen que la cutícula es
una estructura heterogénea, cuya síntesis es controlada por factores genéticos,
fisiológicos, climatológicos y de manejo, tanto en campo como en postcosecha.
Estos factores influyen en su composición y ultraestructura, por lo que existe
mucha variación en su morfología y composición química. En el capítulo I se
describen los resultados del análisis documental acerca de la composición,
fisiología y biosíntesis de la cutícula, con la cual se propone un modelo de
biosíntesis que incluye los trabajos más recientes sobre el mecanismo de
transporte de los monómeros de cutícula a través de la pared celular, que es el
fenómeno menos conocido.
3
Las investigaciones para la identificación de genes que participan en la
biosíntesis de cutícula se han realizado principalmente en la planta modelo de
Arabidopsis, y tomate, donde han sido identificados varios mutantes con
diferentes fenotipos de cutícula. Recientemente, se han reportado estudios en
frutos de manzana y cereza. Los trabajos mencionados sugieren que la cutícula
es el componente de las frutas que controla el tiempo durante el cual, el fruto se
encontrará en condiciones óptimas de consumo, lo que se conoce como vida
postcosecha.
A pesar de la gran importancia económica y agronómica del fruto de mango, no
existe información acerca del mecanismo molecular de biosíntesis de cutícula.
Existen algunos estudios que han sido enfocados a los cambios de composición
y morfología de la cutícula durante el desarrollo y almacenamiento del mango o
en respuesta al tratamiento hidrotérmico.
Además de su importancia para México, el fruto del mango es un modelo
adecuado para el estudio de la cutícula ya que tiene gran cantidad de material
cuticular que puede ser aislado para los análisis químicos y biomecánicos
comparada con la cutícula de Arabidopsis que plantea algunas limitaciones
experimentales debido a que es relativamente delgada, frágil y difícil de aislar
en cantidades adecuadas. Por ejemplo, el mango ‘Keitt’ acumula hasta 193
µg/cm2 de ceras cuticulares, en comparación con el tallo de Arabidopsis, que
acumula 40 µg cm2.
El mayor obstáculo para realizar investigaciones moleculares en mango es la
limitada información genómica, sin embargo, con los nuevos métodos y
herramientas bioinformáticas para la secuenciación y análisis del ADN, se está
revolucionando la generación de información genómica y transcriptómica,
incluso en especies donde no existe información del genoma como es el caso
del mango. Se espera que la utilización de las herramientas mencionadas para
estudiar los genes que participan en la biosíntesis de cutícula, hará posible
elucidar varios aspectos de la biología de la cutícula como lo es el mecanismo
mediante el cual puede controlar la vida postcosecha de los frutos.
4
Con el objetivo de investigar el mecanismo molecular de la biosíntesis de
cutícula, analizamos el transcriptoma de exocarpo de mango maduro y
senescente usando secuenciación de ARN de alto rendimiento (RNA-Seq). En el capitulo II se describen los resultados del análisis transcriptómico de
exocarpo de mango. En este trabajo se generaron aproximadamente 400
millones de pares de lecturas de las cuales fue posible ensamblar de novo
107,744 unigenes. Los datos mostraron la presencia de unigenes que participan
principalmente en el metabolismo de lípidos, cutina, metabolitos secundarios y
los polisacáridos de la pared celular, entre otros. Los análisis funcionales de
RNA-Seq confirman que la ruta de biosíntesis de monómeros de cutina está
enriquecida durante la maduración.
Para validar los análisis bioinformáticos y analizar el mecanismo
molecular de biosíntesis de cutícula durante la ontogenia del fruto de mango se
seleccionaron 15 genes de mango con evidencias experimentales generadas en
estudios de genes ortólogos en Arabidopsis y tomate. Se ha demostrado que
estos genes participan en la biosíntesis, transporte y regulación de la cutícula,
descritos a continuación: MiSHN1/WIN1, MiCD2, MiCER1, MiCER2, MiCER3,
MiKCS2, MiKCS6, MiWBC11, MiLTP1, MiLTP2, MiLTP3, MiLTPG1, MiCUS1 y
MiCUS2. Se analizaron los perfiles de expresión de cada gen durante la
ontogenia del mango mediante transcriptasa reversa termocicladora cuantitativa
en tiempo real utilizando el gen MiPEL1 como control durante la maduración del
fruto y el gen MiActin1 como gen normalizador para el cálculo de la expresión
relativa.
Los resultados mostraron que la biosíntesis de cutícula en mango tiene
un patrón bifásico que se incrementa durante la maduración y senescencia, lo
cual correlaciona con los patrones de expresión de genes analizados. Estos
resultados diferentes a estudios previos en uva y tomate. Finalmente, basado
en los patrones de expresión se propone un modelo de la biosíntesis de la
cutícula. Los resultados de este estudio proporcionan un recurso genómico de
gran importancia para la futura investigación molecular de la biología y vida
postcosecha de mango.
5
Adicionalmente, se realizó un análisis de duplicación del genoma con la
finalidad de investigar posibles eventos de duplicación y especiación en el
mango. Este análisis indica que un evento reciente de duplicación del genoma
se llevó a cabo hace aproximadamente 14-16 millones de años durante la
evolución de mango, después de su divergencia de naranja, que se produjo
hace 57-62 millones de años.
La biosíntesis de cutícula requiere del transporte de lípidos desde las células
epidérmicas a través de la pared celular, función que realizan las proteínas de
transferencia de lípidos (LTPs). Recientemente, en Arabidopsis se reportó una
proteína de transferencia de lípidos 2 anclada a un dominio
glicosilfosfatidilinositol (LTPG2), y se demostró experimentalmente que está
involucrada en el transporte de lípidos durante la biosíntesis de cutícula. En el
capítulo III se presentan los resultados de la caracterización del gen ortólogo a
LTPG2 en mango (MiLTPG2) durante la ontogenia del fruto de mango. Se
demostró la presencia en la secuencia del gene MiLTPG2 de los tres dominios
característicos de las proteínas LTPG: un dominio péptido señal, un dominio de
proteína de transferencia de lípidos y un dominio transmembrana. El dominio de
proteína de transferencia de lípidos contiene los característicos ocho residuos
de cisteína altamente conservados. La acumulación de cutícula mostró un
patrón bifásico, caracterizado por una acumulación durante el crecimiento del
fruto, seguido de una segunda fase caracterizada por una gran deposición de
cutícula durante la maduración. MiLTPG2 mostró un incremento en su
expresión de 7.8 veces durante las etapas tardías de biosíntesis de cutícula que
corresponde a 153 días después de floración (DDF) comparado con 15 DDF.
Este aumento en la expresión correlaciona con el elevado incremento en la
acumulación de cutícula (2100 µg/cm2) observado en esta misma etapa. Con la
información generada en el análisis de expresión de este gene, se propone
modelo en el cual se describe la posible función del gen MiLTPG2 en el
mecanismo molecular de biosíntesis de cutícula en mango. Este estudio es el
primer esfuerzo que se realiza para elucidar la posible función del gen MiLTPG2
en la biosíntesis de la cutícula en frutos de mango.
6
HIPÓTESIS
La biosíntesis de cutícula de mango tiene un comportamiento bifásico que se
incrementa durante la maduración y la expresión de genes que participan en su
biosíntesis está correlacionada con este comportamiento.
7
OBJETIVO GENERAL
Correlacionar la expresión de genes que participan en la biosíntesis con la
formación de cutícula durante la ontogenia de los frutos de mango.
8
OBJETIVOS ESPECÍFICOS
1. Realizar un análisis transcriptómico de exocarpo de mango mediante RNA-
Seq.
2. Identificar y caracterizar mediante bioinformática los principales genes que
participan en la biosíntesis de cutícula de mango durante su ontogenia.
3. Analizar la regulación de la expresión de los genes identificados mediante
transcriptasa reversa termocicladora cuantitativa en tiempo real.
4. Cuantificar la deposición de cutícula durante la ontogenia del mango.
9
CONCLUSIONES GENERALES
La cutícula es una estructura heterogénea y de composición variable que
protege a la planta de diversos estreses bióticos y abióticos. El análisis transcriptómico identificó una gran cantidad de unigenes que
participan en diferentes procesos metabólicos de biosíntesis en el fruto de
mango como lípidos, cutina, metabolitos secundarios, polisacáridos de la pared
celular, entre otros.
Los análisis funcionales de RNA-Seq confirman que la ruta de biosíntesis
de monómeros de cutina está enriquecida durante la maduración.
La biosíntesis de cutícula en mango tiene un patrón bifásico que se
correlaciona con los patrones de expresión de los genes analizados,
principalmente en la etapa inicial y final durante la maduración y senescencia, el
cual es diferente a estudios previos realizados en uva y tomate.
Se encontró que posiblemente la proteína que es codificada por el gene
MiLTPG2 cumple una función en la biosíntesis de cutícula en el fruto de mango.
Los resultados de este estudio proporcionan un recurso genómico de
gran importancia para la futura investigación molecular de la biología y calidad
postcosecha de mango y otras frutas tropicales.
Con los estudios realizados en esta tesis, se crearon varios modelos del
mecanismo molecular de biosíntesis de cutícula. Un modelo teórico de la
biosíntesis de cutícula basado en la investigación documental y dos modelos
basados en datos experimentales.
CAPÍTULO I
COMPOSICIÓN, FISIOLOGÍA Y BIOSÍNTESIS DE LA CUTÍCULA EN PLANTAS.
Tafolla-Arellano JC, González-León A, Tiznado-Hernández ME, Zacarías García L, Báez-Sañudo R.
Revista Fitotecnia Mexicana (2013), 36:3-12.
11
Artículo de Revisión Rev. Fitotec. Mex. Vol. 36 (1): 3 - 12, 2013
Recibido: 11 de Mayo del 2012Aceptado: 21 de Noviembre del 2013
RESUMEN
La cutícula es la capa protectora que se encuentra en la superficie más externa de las plantas y que interacciona con el ambiente, la cual se encuentra en todas las partes aéreas de las plantas superiores. La cutícula está constituida principalmente de dos tipos de polímeros lipofílicos, cutina y ceras cuticulares, los cuales son alterados tanto en su composición como ultraestructura por factores genéticos, fisiológicos y ambientales, tanto durante el crecimiento y desarrollo como durante la postcosecha, por lo que no se debe generalizar sobre su morfología y composición química. La cutícula desempeña un papel importante al actuar como una barrera que reduce la pérdida de agua y difusión de gases, evita la acumulación de agua y polvo, participa en las interacciones planta-insecto, participa en la traducción de señales para la activación de genes específicos, controla los cambios de temperatura, y provee soporte mecánico. Aun cuando se conoce mucho sobre la composición y ultraestructura de la cutícula, es relativamente poco lo que se conoce acerca de su biosíntesis. En la presente revisión se compila y analiza la información científica actual referente a la biosíntesis de la cutícula, que incluye los trabajos más recientes sobre las vías de transporte de los polímeros cuticulares a través de la pared celular, que es el fenómeno menos conocido.
Palabras clave: Cutícula, ceras, cutina, biosíntesis.
SUMMARY
The cuticle is a protective layer located in the outermost surface of all aerial tissues of higher plants and therefore, interacts with the en-vironment. The cuticle is composed mainly of two types of lipophilic polymers, namely: cutin and cuticular waxes, which composition and ultrastructure can be altered by genetic, physiological and environ-mental factors, both during growth and development as well as dur-ing postharvest; its morphology and chemical composition cannot be generalized. The cuticle plays an important role acting as a barrier reducing water loss and gas diffusion, restraining water and dust accu-mulation, participating in the plant-insect interaction, as a component of the signal transduction leading to the activation of specific genes, controlling temperature fluctuations and providing mechanical sup-port. Although the cuticle composition and ultrastructure is fairly well understood, relatively little is known about its biosynthesis. This re-view compiles and analyzes the latest scientific information concerning the cuticle biosynthesis, including the most recent studies about the transport of cuticle polymers through the plant cell wall, which is the least understood phenomena.
Index words: Cuticle, waxes, cutin, biosynthesis.
INTRODUCCIÓN
Las partes aéreas de las plantas superiores, que incluyen hojas, tallos, flores y frutos, están cubiertas completamente, con excepción de la apertura estomática, de una membrana continua lipídica extracelular denominada cutícula (Pighin et al., 2004; Cameron et al., 2006; Jeffree, 2006), la cual es sintetizada por las células epidérmicas (Bargel et al., 2006; Yeats et al., 2010). La cutícula es una estructura producto de la evolución de las plantas superiores que las aísla y pro-tege del medio externo que les rodea (Shepherd y Griffiths, 2006; Reina-Pinto y Yephremov, 2009), que constituye un elemento estructural esencial, de importancia funcional y ecológica debido a que es la capa más externa de las cé-lulas vegetales que interacciona con el ambiente (Kunst y Samuels, 2003; Jeffree, 2006).
La ultraestructura de la cutícula varía ampliamente entre especies de plantas, tipos de órgano y su estado de desarro-llo, y está irreversiblemente asociada al crecimiento activo de los tejidos vegetales, ya que durante las etapas iniciales de desarrollo existe lo que se conoce como procutícula que luego origina a la cutícula madura durante las etapas fina-les de desarrollo (Petit-Jiménez et al., 2007; Isaacson et al., 2009). A pesar de esta variabilidad, todas las cutículas están constituidas principalmente de dos tipos de materiales li-pofílicos: cutina y ceras cuticulares (Leide et al., 2007; Do-mínguez et al., 2009).
Principales polímeros que conforman a la cutícula
Desde un punto de vista morfológico, en un corte trans-versal observado desde el exterior se aprecia que la cutícula cubre la pared celular de las células epidérmicas. Está com-puesta por una cubierta superior de ceras epicuticulares, seguida por otra capa inferior formada por cutina y ceras mezcladas con sustancias de la pared celular, pectinas, ce-lulosa y otros carbohidratos, los cuales constituyen la capa cuticular (Kunst y Samuels, 2003; Jetter et al., 2006; Domín-guez et al., 2011), como se ilustra en la Figura 1.
COMPOSICIÓN, FISIOLOGÍA Y BIOSÍNTESIS DE LA CUTÍCULA EN PLANTAS
COMPOSITION, PHYSIOLOGY AND BIOSYNTHESIS OF PLANT CUTICLE
Julio C. Tafolla-Arellano1, Alberto González-León1, Martín E. Tiznado-Hernández1, Lorenzo Zacarías García2 y Reginaldo Báez-Sañudo1*
1Coordinación de Tecnología de Alimentos de Origen Vegetal, Centro de Investigación en Alimentación y Desarrollo, A. C. Km 0.6 carretera a la Victoria, Apdo. Postal 1735. 83000, Hermosillo, Sonora, México. Tel.: +52 (662) 289 2421; Fax +52 (662) 289 2400 ext. 227. 2Instituto de Agroquímica y Tecnología de Alimentos, Consejo Superior de Investigaciones Científicas. Avenida Agustín Escardino, 7. 46980, Paterna. Valencia, España.
*Autor para correspondencia (rbaez@cascabel.ciad.mx, rbaez@ciad.mx)
12
4
COMPOSICIÓN, FISIOLOGÍA Y BIOSÍNTESIS DE LA CUTÍCULA Rev. Fitotec. Mex. Vol. 36 (1) 2013
Cutina
El principal componente de la cutícula es la cutina, que constituye una proporción que varía desde 40 a 80 % del peso. Según la especie, la cantidad de cutina puede variar de pocos microgramos a más de 1000 µg cm-2 y su grosor puede variar desde menos de 1 hasta 10 µm o más (Domín-guez et al., 2011; Yeats et al., 2012). La cutina es un políme-ro constituido principalmente por ácidos grasos de cade-na media, los cuales se encuentran formando enlaces tipo éster entre sí, así como también glicerol (Suh et al., 2005; Panikashvili et al., 2007; Lee et al., 2009). Debido a los enla-ces covalentes entre sus monómeros, la cutina resiste daños mecánicos y forma la estructura básica de la cutícula (Stark y Tian, 2006; Samuels et al., 2008). La cutina está formada casi exclusivamente por ácidos grasos de 16 carbonos, en-tre los cuales el ácido 10, 16-dihidroxihexadecanoico y su isómero posicional 9, 16-dihidroxihexadecanoico, consti-tuyen los principales componentes (Bessire et al., 2007). So-lamente una pequeña fracción de la cutina investigada está formada por ácidos grasos de 18 carbonos, entre ellos los ácidos 9, 10-epoxi-18-hidroxioctadecanoico y 9,10,18-tri-hidroxioctadecanoico, los más abundantes, aunque algunos derivados insaturados pueden estar presentes como com-ponentes minoritarios en algunas cutinas (Heredia, 2003).
La caracterización reciente de la cutícula en Arabidopsis thaliana ha revelado que la cutina también puede contener ácidos α, ω-dicarboxílicos, componentes característicos de suberina, otro polímero importante en las plantas (Franke et al., 2005; Reina-Pinto y Yephremov, 2009). En algunas cutículas de plantas (por ejemplo, en Agave americana L.) se encuentra presente otro polímero denominado cutan, fracción no hidrolizable de la cutícula, ya sea alternado o en combinación con cutina, con algunos polisacáridos de la pared celular y con compuestos aromáticos (Pollard et al., 2008); está constituido de ácidos grasos poliinsaturados que varían entre 22 y 34 átomos de carbono, en su mayoría unidos entre sí mediante enlaces éter (Bargel et al., 2006; Domínguez et al., 2011).
Ceras epicuticulares e intracuticulares
La función esencial de limitar la pérdida de agua por la cutícula puede deberse a que es un complejo poliéster con ceras asociadas de naturaleza hidrofóbica y muy escasa reactividad, porque la mayoría de los grupos carboxílicos presentes en la membrana están esterificados con grupos hidroxilos alifáticos de otros ácidos grasos (Riederer, 2006; Domínguez et al., 2011). La separación física mediante sol-ventes orgánicos y el análisis de sus componentes, han de-mostrado que las ceras intracuticulares están intercaladas
Figura 1. Ubicación de la cutícula con respecto a las células epidérmicas, y sección transversal de la misma que muestra la posición de los principales polímeros que la conforman.
13
5
Rev. Fitotec. Mex. Vol. 36 (1) 2013TAFOLLA, GONZÁLEZ, TIZNADO, ZACARÍAS Y BÁEZ
dentro del polímero de la cutina y tienen una composición química distinta de las ceras epicuticulares que se encuen-tran en la superficie exterior de la cutina, en forma de una capa más o menos uniforme y amorfa o como cristales dis-continuos (Bargel et al., 2006; Samuels et al., 2008; Domín-guez et al., 2011).
Los componentes de las ceras son muy variados y nor-malmente constituyen de 20 a 60 % de la masa de la cutícula (Heredia, 2003). La cera cuticular es una mezcla compleja de compuestos alifáticos de cadenas lineales que varían en-tre 20 y 40 carbonos de tamaño; sin embargo, también se han identificado ésteres de cera con cadenas que van desde 36 hasta 70 carbonos (Reina-Pinto y Yephremov, 2009). Los principales componentes químicos de las ceras son n-alca-nos, ésteres, alcoholes, aldehídos, cetonas y ácidos grasos de cadena larga en el caso de las epicuticulares, o de áci-dos grasos de cadena corta en las intracuticulares (Kunst y Samuels, 2003; Cameron et al., 2006; Leide et al., 2011).
Entre las ceras se han encontrado algunos metabolitos secundarios como los triterpenoides, compuestos fenólicos (ácido cumárico y ferúlico, flavonoides, fenilpropanoides), polisacáridos (principalmente celulosa y pectina) y algunos polipéptidos (Stark y Tian, 2006; Jeffree, 2006; Riederer, 2006; Kunst y Samuels, 2009).
Por su parte, las ceras epicuticulares por lo general tienen una estructura microcristalina, y que se visualizan como una capa subyacente amorfa. Varias de las estructuras mor-fológicas clasificadas por Barthlott et al. (1998) como héli-ces, túbulos, cintas, varillas o placas, pueden estar presentes. Algunas de éstas pueden estar relacionadas con la presencia de determinados componentes de la cera. Los compues-tos con cadena media tales como β-dicetonas, hidroxi-β-dicetonas, dioles y alcoholes secundarios, están asociados con los tubos, mientras que los alcoholes primarios se aso-cian con las placas. Los alcoholes primarios también están asociados con estructuras cristalinas (Shepherd y Griffiths, 2006). La importancia de la composición química de las ce-ras epicuticulares radica en la estrecha relación que existe con la morfología y ultraestructura de las mismas.
Cambios cuticulares durante el desarrollo vegetativo y periodo postcosecha de frutas y verduras
Existen varios estudios sobre la cutícula en tejidos vege-tativos y durante el desarrollo y vida postcosecha de frutas, los cuales se describen a continuación. En cuanto a com-posición, se ha reportado que la fracción mayoritaria de ceras cuticulares en hojas y tallos de Arabidopsis (Jenks et al., 2002) y en Kalanchoe daigremontiana (Van Maarseveen et al., 2009) son los alcanos. Con respecto a los cambios ontogénicos de la cutícula, Báez et al. (1993) reportaron
cambios fisiológicos y ultraestructurales durante la madu-ración y senescencia en mandarina (Citrus reticulata [Hort] Ex. Tanaka, cv Nules); por ejemplo, en frutos inmaduros la fracción de ácidos grasos fue la más abundante en ceras epicuticulares (50 a 55 %) e intracuticulares (70 a 35 %), y luego durante la maduración la proporción de ácidos gra-sos en ceras epicuticulares disminuyó y el contenido de al-canos con más de 26 carbonos aumentó considerablemente.
Asimismo, Petit-Jiménez et al. (2009), al analizar el efecto del tratamiento hidrotérmico sobre la ultraestructura de la cutícula de mango (Mangifera indica L.), observaron dife-rencias en el arreglo estructural de las ceras en la superficie cuticular entre los frutos con tratamiento hidrotérmico y el testigo sin tratar. En los frutos tratados se evidenció la for-mación tipo pergamino en la cutícula debido al efecto del calor, con placas alineadas en paralelo y en las ceras epicuti-culares se detectó la presencia de estructuras de cristales en transición con una distribución irregular; en cambio, en los frutos no tratados no se observó el efecto pergamino en la cutícula, se constató la formación de placas enteras y de ce-ras epicuticulares del tipo amorfo. Al correlacionar los cam-bios en la composición de la cutícula con la pérdida de agua durante postcosecha en pimiento (Capsicum annuum L.), Parsons et al. (2012) concluyeron que las cadenas alifáticas lineales forman barreras cuticulares más impermeables que los complejos basados en isoprenoides. En líneas mutantes de tomate (Lycopersicum esculentum Mill. o Solanum lyco-persicum), Kosma et al. (2010) correlacionaron los cambios cuticulares con producción de etileno, degradación de la pared celular y color. Los autores encontraron diferencias significativas entre frutos y etapas de desarrollo, por lo que concluyeron que la cutícula tiene una función importante en la vida de anaquel de los frutos.
En mango Petit-Jiménez et al. (2007) observaron cam-bios en la composición y ultraestructura de la cutícula du-rante el crecimiento, desarrollo y almacenamiento en tres variedades. En las ceras epicuticulares, la fracción de los alcanos fue la predominante durante el crecimiento (50 a 60 %), mientras que en la cosecha fue la de los ácidos grasos (38 a 46 %). Los alcoholes representaron la fracción minori-taria durante el crecimiento y almacenamiento de los frutos (2 a 4 %). Además, observaron diferencias significativas en-tre cultivares en la cantidad de cutícula por área ( ‘Tommy Atkins’ con 227 μg cm-2, ‘Keitt’ con 193 μg cm-2 y ‘Kent’ con 141 μg cm-2). La ultraestructura de las ceras mostró diferen-cias en la cosecha, ya que ‘Tommy Atkins’ y ‘Kent’ presen-taron 82.6 % de zonas cristalinas, mientras que en ‘Keitt’ hubo 74.1 % de zonas amorfas.
Durante el almacenamiento de los frutos de mango tam-bién hubo cambios cuticulares, pues al tercer día se observó una disminución en el contenido de las ceras intracuticulares
14
6
COMPOSICIÓN, FISIOLOGÍA Y BIOSÍNTESIS DE LA CUTÍCULA Rev. Fitotec. Mex. Vol. 36 (1) 2013
en todos los cultivares, seguida de un ligero incremento al sexto día, y luego de una nueva disminución en el noveno día. La masa de la cutícula se incrementó durante el cre-cimiento, con diferencias significativas entre cultivares ya que ‘Tommy Atkins’ alcanzó un valor máximo de 4513 μg a 45 días después de antesis (DDA), ‘Kent’ 2316 μg a 90 DDA y ‘Keitt’ 1609 μg a 135 DDA. Los autores concluyeron que la mayor eficiencia de la cutícula en regular la pérdida de agua ocurrió al momento de la cosecha y se relacionó con los cambios en la ultraestructura y contenido de las ceras cuticulares. Además, asociaron las diferencias con las ca-racterísticas genéticas de los cultivares, ya que éstos habían crecido en las mismas condiciones ambientales y de manejo del huerto, y tenían la misma edad. Con base en lo anterior, es posible afirmar que la composición y ultraestructura de la cutícula varía en respuesta a factores genéticos, fisiológi-cos y ambientales, tanto durante el crecimiento y desarrollo como durante la postcosecha de los frutos.
FISIOLOGÍA DE LA CUTÍCULA
A pesar de que el material cuticular aparece como un componente minoritario en el total de la masa de hojas y frutos, desempeña funciones importantes debido a sus pro-piedades físicas, químicas, mecánicas y morfológicas, que lleva a cabo a lo largo del desarrollo de la planta y son rele-vantes para la vida de las plantas y frutos.
Tales funciones se describen a continuación: (A) Como barrera que reduce la pérdida de agua y difusión de gases
(Riederer y Schreiber, 2001); (B) Induce desprendimiento de gotas de agua y partículas de polvo, así como de esporas, con la finalidad de mantener limpia y seca la superficie de la planta o del fruto (Jeffree, 2006; Samuels et al., 2008); (C) Por sus propiedades anti-adhesivas, influye en las inte-racciones planta-insecto (Müller, 2006), y ayuda a evitar la proliferación de microbios patógenos (Carver y Gurr, 2006; Reina-Pinto y Yephremov, 2009); (D) Involucrada en el re-conocimiento de señales de patógenos e insectos (Chassot et al., 2008); (E) Tiene un papel termorregulador importan-te en las interacciones de las plantas con el ambiente (Stark y Tian, 2006) y proteje contra los rayos UV (Pfündel et al., 2006); (F) Funciona como soporte mecánico (Domínguez et al., 2009) y participa de manera indirecta en la correcta formación de los órganos en las primeras fases de desarro-llo de la planta, ya que impide la adhesión incontrolada de las células epidérmicas de los órganos en formación (Riede-rer, 2006; Panikashvili et al., 2007; Leide et al., 2011). Tales funciones son esquematizadas en la Figura 2.
El rol de la cutícula en reducir la pérdida de agua parece ser su función primaria, ya que actúa como una eficaz ba-rrera hidrofóbica protectora para minimizar la pérdida de agua por evapotranspiración y también la pérdida de otros gases (CO2, O2), y de esta forma permite que los estomas puedan regular este proceso (Jeffree, 2006; Riederer, 2006; Panikashvili et al., 2007). Sin embargo, no es absolutamen-te impermeable (Burghardt y Riederer, 2006; Isaacson et al., 2009) ya que en forma lenta el agua traspasa la cutí-cula y del mismo modo la atraviesan en sentido contrario
Figura 2. Principales funciones de la cutícula en las plantas. A) Reducción de la pérdida de agua y difusión de gases. B) Evita acumulación de agua y polvo. C) Participa en las interacciones planta-insecto. D) Participa en la traducción de señales para la activación de genes específicos. E) Controla los cambios de temperatura. F) Provee soporte mecánico.
15
7
Rev. Fitotec. Mex. Vol. 36 (1) 2013TAFOLLA, GONZÁLEZ, TIZNADO, ZACARÍAS Y BÁEZ
las sustancias solubles que en ella se depositan (Lallana et al., 2006). La cutícula es una membrana permeable tanto a compuestos polares como no polares, donde las ceras cum-plen un papel clave en la reducción de la permeabilidad al agua, especialmente las ceras epicuticulares que regulan la capacidad de la superficie para la evapotranspiración.
Las funciones de la cutícula no están correlacionadas con su grosor sino con su estructura cuticular, con su composi-ción química y con las proporciones en que se encuentren sus componentes (Kerstiens, 2006; Leide et al., 2011; Yeats et al., 2012). El grosor de la cutícula varía entre 0.5 y 15 µm, lo que depende de la especie vegetal, la zona de la planta y su edad o estado de desarrollo, ya que aumenta durante el crecimiento y disminuye durante el proceso de maduración y senescencia (Jetter et al., 2000; Jetter et al., 2006; Stark y Tian, 2006). La composición química y la estructura cuti-cular son generadas por una red metabólica compleja, re-gulada por factores bióticos y abióticos, para proporcionar un mecanismo de adaptación durante la interacción planta-ambiente (Bernard y Joubès, 2012)
BIOSÍNTESIS DE LA CUTÍCULA
La mayoría de estudios sobre la composición y ultraes-tructura de la cutícula han sido descriptivos, comparati-vos, pero es relativamente poco lo que se conoce acerca de la biosíntesis, transporte y ensamblaje extracelular de los compuestos cuticulares para formar el biopolímero de la cutícula (Isaacson et al., 2009; DeBono et al., 2009; Yeats et al., 2010). Uno de los principales puntos de discusión sobre la biosíntesis de la cutina es el transporte de sus mo-nómeros desde el lugar de síntesis hasta el sitio donde son incorporados a la cutina en crecimiento (Pighin et al., 2004; DeBono et al., 2009).
En las plantas, las células epidérmicas emplean gran can-tidad de energía para producir cutícula. Por ejemplo, más de la mitad de los ácidos grasos sintetizados por las células epidérmicas durante la expansión del tallo en Arabidopsis son utilizados en la formación de lípidos cuticulares (Rei-na-Pinto y Yephremov, 2009). La síntesis de cera requiere la coordinación de actividades de numerosas enzimas organi-zadas en complejos multienzimáticos en varios organelos celulares (cloroplastos, citoplasma y retículo endoplasmáti-co), donde se lleva a cabo la síntesis y elongación de los áci-dos grasos, precursores de las ceras, y la formación de una multitud de compuestos alifáticos (Kunst y Samuels, 2003; Kunst et al., 2006).
Si bien se han propuesto diversas hipótesis sobre la bio-síntesis de los compuestos cuticulares, los mecanismos de transporte siguen siendo poco conocidos. Durante la década de los 70 quedó demostrado que la biosíntesis de
cutina está mediada por enzimas localizadas en las células epidérmicas o en la cara externa de la pared celular, y que tales enzimas requerían ATP y CoA (Samuels et al., 2008). La biosíntesis de ceras abarca tres distintas etapas: síntesis de novo de ácidos grasos, elongación de los ácidos grasos y transporte de monómeros hacia el exterior de la pared celular.
Síntesis de novo de ácidos grasos
Los ácidos grasos de 16 y 18 carbonos son sintetizados de novo en los cloroplastos (Kunst et al., 2006; Byers y Gong, 2007). En su biosíntesis, la cadena de grupos acilos de cre-cimiento es unida covalentemente a la proteína transpor-tadora de grupos acilo (ACP) mediante un enlace tioéster vinculado a un grupo prostético de fosfopanteteína, lo que resulta en la activación del carbono carboxilo del grupo aci-lo (Shepherd y Griffiths, 2006). La ACP es un componente de la enzima ácido graso sintasa (FAS), que participa como cofactor en por lo menos ocho reacciones de la síntesis de ácidos grasos y también puede funcionar como un donador de acilos para la biosíntesis de lípidos complejos (Kunst y Samuels, 2003; Byers y Gong, 2007). En este proceso se en-samblan largas cadenas de carbonos, ensamblaje que inicia con la condensación de acetil-CoA con una molécula de dos carbonos del malonil-ACP, los cuales se originan de acetil-CoA.
Después se produce el paso de la condensación, donde una secuencia de reacciones que incluyen la reducción de β-hidroxiacil-ACP, la deshidratación de β-hidroxiacil-ACP, y reducción de trans-∆2 –enoil-ACP, en la que se genera un acil-ACP con dos carbonos más que la molécula con la cual se inició el ciclo. Ciclos similares de elongación, que ahora empiezan con la condensación de malonil-ACP con una acil-ACP y terminan con la eliminación reductiva del gru-po β-ceto, se repiten de seis a siete veces (Harwood, 2005; Shepherd y Griffiths, 2006).
Dos o tres tipos de complejos de FAS son necesarios para la formación de un ácido graso de 16 ó 18 carbonos, res-pectivamente. Los complejos FAS difieren en sus enzimas condensadoras, las cuales tienen una estricta longitud espe-cífica de la cadena acilo: cetoacil ACP sintasa III (KAS III) (C2 a C4), KAS I (C4 a C16), y KAS II (C16 a C18) (Kunst et al., 2006). Una vez sintetizados en los cloroplastos, los ácidos grasos son transportados al retículo endoplasmático para su elongación, proceso para el que se han propuesto dos vías (Figura 4): 1) En muchas especies de plantas y tipos de células, el retículo endoplasmático se ha encontrado cerca de los cloroplastos, sin aparente fusión o mezcla de bicapas, proximidad que puede facilitar la transferencia de ácidos grasos al retículo endoplasmático mediante mecanismos no-vesiculares como la desorción espontánea, la difusión y
16
8
COMPOSICIÓN, FISIOLOGÍA Y BIOSÍNTESIS DE LA CUTÍCULA Rev. Fitotec. Mex. Vol. 36 (1) 2013
la absorción. 2) El transporte de lípidos de los cloroplastos al retículo endoplasmático podría verse facilitado por las proteínas acil-CoA “binding protein” (ACBPs), una clase de proteínas que ha sido descrita en una amplia variedad de células eucariotas (Schulz y Frommer, 2004; Kunst et al., 2006; Panikashvili y Aharoni, 2008).
Elongación de los ácidos grasos
La elongación de los ácidos grasos de 16 y 18 carbonos en el retículo endoplasmático (ER) genera ácidos grasos de cadenas muy largas (VLCFAs) de 20 a 34 carbonos. Esta extensión se lleva a cabo por complejos multienzimáticos que residen en la membrana del retículo endoplasmático, conocidas como elongasas de ácidos grasos (FAEs) (Kunst y Samuels, 2003; Shepherd y Griffiths, 2006). Análogo a la síntesis de ácidos grasos en los cloroplastos, la formación de VLCFAs implica cuatro reacciones enzimáticas consecu-tivas que resultan en una extensión de dos carbonos en la cadena de acilo por cada ciclo de elongación.
Sin embargo, a diferencia de la FAS que utiliza malonil-ACP como donante de dos carbonos, la FAE utiliza unida-des de dos carbonos de malonil-CoA (Post-Beittenmiller, 1996; Kunst y Samuels, 2003; Shepherd y Griffiths, 2006; Samuels et al., 2008). Múltiples ciclos de elongación son necesarios para generar cadenas con longitudes de 24 a 34 carbonos para la producción de componentes alifáticos de ceras (Kunst et al., 2006). En la etapa final de la produc-ción de cera en el retículo endoplasmático, las VLCFAs son transformadas en alcoholes, ésteres, aldehídos, alcanos y cetonas, mediante reducción y descarboxilación (Samuels et al., 2008; Lee et al., 2009), como se ilustra en la Figura 3.
Síntesis de alcoholes primarios y ésteres de ceras
Una parte de la biosíntesis de cera, generalmente la lla-mada vía de la reducción de acil-CoA, es la responsable de la formación de componentes con predominante número par de carbonos (Figura 3) (Kunst et al., 2006; Shepherd y Griffiths, 2006). En diversas plantas y órganos, los com-puestos más importantes son los alcoholes primarios con cadena de 26 a 28 carbonos, aunque en algunos sistemas son de 30 a 32 carbonos. Los alcoholes se encuentran fre-cuentemente en forma libre o esterificada a diversos grupos acilos, e incluye los alcoholes aromáticos con número par de carbonos de cadena corta y cadena larga o ácidos alifáti-cos de cadenas muy largas.
Los alcoholes son generados por reducción de precur-sores de VLCFAs y se producen mediante aldehídos inter-mediarios. La biosíntesis de ésteres en plantas superiores es catalizada por enzimas, como la cera sintasa (WS), algunas de las cuales son capaces de usar una amplia gama de grasas
saturadas e insaturadas en forma de acil-CoA, que oscilan entre 14 y 24 carbonos, mientras que los alcoholes insatura-dos con 18 carbonos son el segundo sustrato más utilizado (Post-Beittenmiller, 1996; Kunst et al., 2006; Samuels et al., 2008).
Síntesis de alcanos, alcoholes secundarios y cetonas
La segunda parte de la ruta de la biosíntesis de cera es responsable de la formación de compuestos predominan-temente con número impar de carbonos. Entre éstos, los alcanos se han encontrado en las mezclas de ceras de varias plantas y órganos, donde frecuentemente se acumulan en altas concentraciones (Kunst et al., 2006). Los alcoholes se-cundarios y las cetonas con similar distribución de longitud de cadena, regularmente se encuentran con los alcanos, lo que sugiere una relación directa biosintética entre las tres clases de componentes. La reacción central de la vía de for-mación de alcanos (el paso que hace la transición de par a impar en las cadenas de carbono), implica la pérdida de un átomo de carbono de los precursores de grupos acilo, en lugar de la adición de un carbono. Los alcanos son lue-go convertidos en alcoholes secundarios y cetonas por dos reacciones consecutivas de oxidación (Kunst et al., 2006; Shepherd y Griffiths, 2006; Samuels et al., 2008). Está es-tablecido que la elongación procede a la descarboxilación y que, por tanto, ambas rutas de los alcoholes primarios y la de alcanos compiten por los precursores Acil-CoA de va-rias longitudes de cadena.
Transporte de monómeros a la cutícula
La tercera etapa de la biosíntesis de la cutícula requiere el transporte de los lípidos de las células epidérmicas al ex-terior de la pared celular (Pighin et al., 2004). Los monó-meros son transportados a través de ambientes hidrofílicos y membranas, es decir, de los cloroplastos, retículo endo-plasmático, citosol, membrana plasmática y finalmente la pared celular, por lo que el transporte de los componentes cuticulares es un proceso complejo y poco conocido (Post-Beittenmiller, 1996; Panikashvili y Aharoni, 2008; DeBono et al., 2009; Yeats et al., 2010; Beisson et al., 2012).
Un primer avance hacia la comprensión de la exportación de cera recientemente se realizó mediante el descubrimien-to de los transportadores tipo ABC (ATP binding cassette) ABCG12/CER5 y ABCG11/WBC11, cuya participación en el transporte de la cera fue reportada por Pighin et al. (2004). Estos autores fueron los primeros en demostrar me-diante pruebas moleculares el transporte activo de los com-ponentes de la cera a través de la membrana plasmática de las células de la epidermis (Kunst et al., 2006; Panikashvili et al., 2007; Panikashvili y Aharoni, 2008). Ambos trans-portadores fueron localizados en la membrana plasmática
17
9
Rev. Fitotec. Mex. Vol. 36 (1) 2013TAFOLLA, GONZÁLEZ, TIZNADO, ZACARÍAS Y BÁEZ
de células epidérmicas de tallos, mediante fusiones fluores-centes con las proteínas de transporte y utilización de mi-croscopía confocal.
Sin embargo, el mecanismo de exportación a partir de la membrana plasmática a través del medio hidrofílico de la pared celular a la cutina sigue siendo poco conocido y por ello constituye un fenómeno interesante a elucidar. Esto es debido a que una molécula de cera hidrofóbica exportada fuera de una célula epidérmica debe atravesar un medio extracelular hidrofílico para llegar a la cutícula, además de que los polisacáridos de la pared celular, como pectinas, hemicelulosas y celulosas, pueden representar un obstá-culo físico al transporte de la cera cuticular (Jeffree, 2006; Samuels et al., 2008; Yeats y Rose, 2008), como se ilustra en la Figura 4.
Las proteínas de transferencia de lípidos (LTPs) se han propuesto como candidatas para llevar a cabo la deposi-ción de los componentes de la cera durante el ensamblaje de la cutícula (Kunst et al., 2006; Lee et al., 2009; DeBo-no et al., 2009; Yeats et al., 2010). Se ha reportado tam-bién que las LTPs participan en la defensa de las plantas contra patógenos (Arondel et al., 2000). Las LTPs poseen características apropiadas para el transporte de cera hacia la cutícula: poseen una cavidad hidrofóbica, son capaces de unirse a los ácidos grasos in vitro, contienen un péptido señal, y son proteínas extracelulares situadas en la pared ce-lular (Kader, 1996; Li et al., 2008; Yeats y Rose, 2008). Han sido identificadas en hojas de tabaco (Nicotiana tabaccum L.) (Cameron et al., 2006), hojas de brócoli (Brassica ole-racea var. italica) (Pyee et al., 1994), en hojas de espinaca (Spinacia oleracea L.), en plántulas de maíz, en semillas de
Figura 3. Biosíntesis de cera. Una vez sintetizados en el cloroplasto, los ácidos grasos son transportados al retículo endoplasmático para su elongación. Me-diante reacciones de reducción se obtienen aldehídos, alcoholes primarios y és-teres derivados de la reacción entre un ácido carboxílico y un alcohol de alto peso molecular, lo que químicamente constituye una cera. A través de reacciones de descarboxilación se obtienen alcanos, aldehídos, alcoholes secundarios y ceto-nas en la vía de los alcanos. Fuentes: Millar et al. (1999) y Samuels et al. (2008).
18
10
COMPOSICIÓN, FISIOLOGÍA Y BIOSÍNTESIS DE LA CUTÍCULA Rev. Fitotec. Mex. Vol. 36 (1) 2013
cebada (Hordeum vulgare) y arroz (Oryza sativa L.) (Post-Beittenmiller, 1996; Ahn et al., 2009), en tallos, hojas y flo-res de Arabidopsis (Beisson et al., 2003; Suh et al., 2005; Lee et al., 2009; DeBono et al., 2009), y recientemente en tomate (Yeats et al., 2010).
Con base en la información revisada se diseñó la Figura 4 en la que se ilustran las diferentes vías propuestas para el transporte de los compuestos cuticulares. Las VLCFAs o sus derivados podrían ser transportados del retículo en-doplasmático a la membrana plasmática por dos posibles rutas: directamente del retículo endoplasmático a la mem-
brana plasmática (Figura 4, 2a), transportados mediante proteínas de unión a ácidos grasos (FABPs) y liberación de los lípidos directamente a un transportador ABC o en la bicapa de la membrana plasmática; los lípidos cuticulares podrían moverse a lo largo del sistema de endomembra-nas del retículo endoplasmático al aparato de Golgi y a la membrana plasmática, ya sea libres en la bicapa lipídica o a través de balsas lipídicas (Figura 4, 2b).
Una vez en la superficie celular, los componentes de cera podrían ser transferidos de la bicapa por transportadores ABC, o bien por dos mecanismos propuestos (Figura 4, 3a),
Figura 4. Mecanismos propuestos para la biosíntesis de la cutícula. En la primera etapa, la síntesis de novo de los ácidos grasos se lleva a cabo en los cloroplastos, con dos posibles vías de transporte hacia el retículo endoplasmático: (1a) La proximidad entre estos dos organelos puede facilitar la transferencia de los ácidos grasos mediante mecanismos no-vesiculares, como la desorción espontánea, la difusión y absorción. (1b) la transferencia se realizaría mediante las ACBPs (Acyl-CoA binding proteins). La segunda etapa consiste en la elongación de los ácidos grasos a VLCFAs (very long chain fatty acids) en el retículo endoplasmático, para posteriormente ser transportados a la membrana plasmática por dos posibles rutas: (2a) a través de las proteínas FABPs (fatty acid binding proteins) que liberan a los lípidos a un transportador tipo ABC o en la bicapa de la membrana plasmática. Alternativamente (2b), podrían desplazarse a lo largo del sistema de endomembranas al aparato de Golgi y enviados mediante vesículas a la membrana plasmática, y transferir las VLCFAs través de balsas lipídicas. La tercera etapa consiste en el transporte de los monómeros desde el exterior de la membrana plasmática hacia la cutícula, mediante dos mecanismos propuestos: (3a) transferidos directamente a través de la pared celular, o (3b) acarreados por proteínas de transferencia de lípidos (LTPs) hacia la cutícula.
19
11
Rev. Fitotec. Mex. Vol. 36 (1) 2013TAFOLLA, GONZÁLEZ, TIZNADO, ZACARÍAS Y BÁEZ
transferidos directamente a través de la pared celular, o aca-rreados por proteínas de transferencia de lípidos (LTPs) a la cutícula (Figura 4, 3b) (Schulz y Frommer, 2004; Kunst et al., 2006; Shepherd y Griffiths, 2006; Panikashvili y Aha-roni, 2008).
CONCLUSIÓN
La cutícula es una estructura heterogénea, cuya síntesis es controlada por factores genéticos, fisiológicos, climato-lógicos y de manejo, tanto en campo como en postcosecha. Estos factores influyen en su composición y ultraestructu-ra, por lo que no se debe generalizar sobre su morfología y composición química.
Además de ser una barrera física, la cutícula es una es-tructura que cumple funciones importantes en la fisiología de la planta, como: mantener limpia y seca la superficie de la planta o del fruto, y así evitar la acumulación de agua, partículas de polvo y esporas; influye en las interacciones planta-plaga, mediante el reconocimiento de señales de patógenos e insectos; termorreguladora importante en las interacciones de las plantas con el ambiente y sirve de pro-tección contra los rayos UV; soporte mecánico; y participa-ción indirecta en la correcta formación de los órganos en las primeras fases de desarrollo de la planta, ya que impide la adhesión incontrolada de las células epidérmicas de los órganos en formación.
La biosíntesis de la cera cuticular ha sido estudiada du-rante los últimos años, con enfoques bioquímicos y fisio-lógicos. A pesar de estos esfuerzos, todavía se conoce poco sobre los factores que regulan la localización de los precur-sores de los ácidos grasos y la regulación que existe entre la síntesis de ceras con la síntesis de cutina, así como los mecanismos de transporte y deposición de sus componen-tes. Los estudios recientes enfocados en el aislamiento y es-tudio de los genes que codifican proteínas de transferencia de lípidos, implicadas en el transporte de los monómeros, han contribuido a elucidar el fenómeno de la transferencia de los componentes cuticulares a través de la pared celular durante la biosíntesis de la cutícula
Estos conocimientos permiten comprender mejor la bio-síntesis y fisiología de la cutícula, y proporcionan las bases para llevar a cabo una modificación racional de las cutícu-las mediante ingeniería genética con el fin de mejorar la re-sistencia de productos agrícolas a diferentes tipos de estrés, tanto biótico como abiótico, y aumentar la vida postcose-cha de productos hortofrutícolas.
BIBLIOGRAFÍA
Arondel V, C Vergnolle, C Cantrel, J C Kader (2000) Lipid transfer pro-teins are encoded by a small multigene family in Arabidopsis thaliana. Plant Sci. 157:1-12.
Ahn S B, J Kim, J Pyee, J H Park (2009) Biochemical characterization of the lipid-binding properties of a broccoli cuticular wax-associated protein, WAX9D, and its application. BMB Rep. 42:367-372.
Báez R, F Tadeo, E Primo-Millo, L Zacarias (1993) Physiological and ultrastructural changes during the ripening and senescence of clementine mandarin. Acta Hort. 343:18-24.
Bargel H, K Koch, Z Cerman, C Neinhuis (2006) Structure–function re-lationships of the plant cuticle and cuticular waxes—a smart material? Funct. Plant Biol. 33:893-910.
Barthlott W, C Neinhuis, D Cutler, F Ditsch, I Meusel, I Theisen, H Wil-helmi (1998) Classification and terminology of plant epicuti-cular waxes. Bot. J. Linn. Soc. 126:237-260.
Beisson F, Y Li-Beisson, M Pollard (2012) Solving the puzzles of cu-tin and suberin polymer biosynthesis. Curr. Opin. Plant Biol. 15:329-337.
Beisson F, A J K Koo, S Ruuska, J Schwender, M Pollard, J J Thelen, T Paddock, J J Salas, L Savage, A Milcamps, V B Mhaske, Y Cho, J B Ohlrogge (2003) Arabidopsis genes involved in acyl lipid metabolism. A 2003 census of the candidates, a study of the distribution of expressed sequence tags in organs, and a web-based database. Plant Physiol. 132:681-697.
Bessire M, C Chassot, A C Jacquat, M Humphry, S Borel, J MacDonald-Comber, J Pierre, C Nawrath (2007) A permeable cuticle in Arabidopsis leads to a strong resistance to Botrytis cinerea. EMBO J. 26:2158-2168.
Bernard A, J Joubès (2012) Arabidopsis cuticular waxes: Advances in synthesis, export and regulation. Prog. Lipid Res. 52:110-129.
Burghardt M, M Riederer (2006) Cuticular transpiration: In: Biology of the Plant Cuticle. M Riederer, C Müller (eds). Julius-von-Sachs-Institut, für Biowissenschaften Universität Würzburg, Germany. pp:292-311.
Byers D M, H Gong (2007) Acyl carrier protein: structure–function re-lationships in a conserved multifunctional protein family. Re-view/Synthese. Biochem. Cell Biol. 85:649-662.
Cameron K D, M A Teece, L B Smart (2006) Increased accumulation of cuticular wax and expression of lipid transfer protein in res-ponse to periodic drying events in leaves of tree tobacco. Plant Physiol. 140:176-183.
Carver T L W, S J Gurr (2006) Filamentous fungi on plant surfaces: In: Biology of the Plant Cuticle. M Riederer, C Müller (eds). Julius-von-Sachs-Institut, für Biowissenschaften Universität Würz-burg, Germany. pp:368-392.
Chassot C, C Nawrath, J P Métraux (2008) The cuticle: Not only a barrier for plant defence. A novel defence syndrome in plants with cu-ticular defects. Plant Signal. Behav. 3:142-144.
DeBono A, T H Yeats, J K C Rose, D Bird, R Jetter, L Kunst, L Samuels (2009) Arabidopsis LTPG is a glycosylphosphatidylinositol-anchored lipid transfer protein required for export of lipids to the plant surface. Plant Cell 21:1230-1238.
Domínguez E, J A Heredia-Guerrero, A Heredia (2011) The biophysical design of plant cuticles: an overview. New Phytol. 189:938-949.
Domínguez E, L España, G López-Casado, J Cuartero, A Heredia (2009) Biomechanics of isolated tomato (Solanum lycopersicum) fruit cuticles during ripening: the role of flavonoids. Funct. Plant Biol. 36:613-620.
Franke R, I Briesen, T Wojciechowski, A Faust, A Yephremov, C Nawrath, L Schreiber (2005) Apoplastic polyesters in Arabi-dopsis surface tissues–a typical suberin and a particular cutin. Phytochemistry 66:2643-2658.
Harwood J L (2005) Fatty acid biosynthesis: In: Plant Lipids: Biology, Uti-lization and Manipulation. DJ Murphy (ed). Blackwell Publis-hing, Oxford. pp:27-66.
Heredia A (2003) Biophysical and biochemical characteristics of cutin, a plant barrier biopolymer. Biochim. Biophys. Acta 1620:1-7.
Isaacson T, D K Kosma, A J Matas, G J Buda, Y He, B Yu, A Pravita-sari, J D Batteas, R E Stark, M A Jenks, J K C Rose (2009) Cutin deficiency in the tomato fruit cuticle consistently
20
12
COMPOSICIÓN, FISIOLOGÍA Y BIOSÍNTESIS DE LA CUTÍCULA Rev. Fitotec. Mex. Vol. 36 (1) 2013
affects resistance to microbial infection and biomechanical properties, but not transpirational water loss. Plant J. 60:363-377.
Jeffree C E (2006) The fine structure of the plant cuticle. In: Biology of the Plant Cuticle. M Riederer, C Müller (eds). Julius-von-Sachs-Institut, für Biowissenschaften Universität Würzburg, Germany. pp:11-110.
Jenks M A, S D Eigenbrode, B Lemieux (2002) Cuticular Waxes of Arabi-dopsis. In: The Arabidopsis Book. C R Somerville, E M Meye-rowitz (eds.) American Society of Plant Biologists). Rockville, Maryland, USA. 22 p.
Jetter R, L Kunst, L Samuels (2006) Composition of plant cuticular waxes. In: Biology of the Plant Cuticle. M Riederer, C Müller (eds). Julius-von-Sachs-Institut, für Biowissenschaften Universität Würzburg, Germany. pp:145-175.
Jetter R, S Schäffer, M Riederer (2000) Leaf cuticular waxes are arranged in chemically and mechanically distinct layers: evidence from Prunus laurocerasus L. Plant Cell Environ. 23:619-28.
Kader J C (1996) Lipid transfer proteins in plants. Annu. Rev. Plant Phy-siol. Plant Mol. Biol. 47:627-54.
Kerstiens G (2006) Water transport in plant cuticles: an update. J. Exp. Bot. 57:2493-2499.
Kosma D K, E P Parsons, T Isaacson, S Lu , J K C Rose, Matthew A Jenks (2010) Fruit cuticle lipid composition during development in tomato ripening mutants. Physiol. Plant. 139:107-117.
Kunst L, A L Samuels (2003) Biosynthesis and secretion of plant cuticular wax. Prog. Lipid Res. 42:51-80.
Kunst L, L Samuels (2009) Plant cuticles shine: advances in wax biosynthesis and export. Curr. Opin. Plant Biol. 12:721-727.
Kunst L, R Jetter, L Samuels (2006) Biosynthesis and transport of plants cuticular waxes: In: Biology of the Plant Cuticle. M Riederer, C Müller (eds). Julius-von-Sachs-Institut, für Biowissenschaften Universität Würzburg, Germany. pp:182-207.
Lallana M, C E Billard, J H Elizalde, V H Lallana (2006) Breve revisión sobre características de la cutícula vegetal y penetración de herbicidas. Cien. Doc. Tecnol. XVII:229-241.
Lee S B, Y S Go, H J Bae, J H Park, S H Cho, H J Cho, D S Lee, O K Park, I Hwang, M C Suh (2009) Disruption of glycosylphosphati-dylinositol-anchored lipid transfer protein gene altered cuticu-lar lipid composition, increased plastoglobules, and enhanced susceptibility to infection by the fungal pathogen Alternaria brassicicola. Plant Physiol. 150:42-54.
Leide J, U Hildebrandt, G Vogg, M Riederer (2011) The positional sterile (ps) mutation affects cuticular transpiration and wax biosynthesis of tomato fruits. J. Plant Physiol. 168:871-877.
Leide J, U Hildebrandt, K Reussing, M Riederer, G Vogg (2007) The developmental pattern of tomato fruit wax accumulation and its impact on cuticular transpiration barrier properties: effects of a deficiency in a β-ketoacyl-coenzyme A synthase (LeCER6). Plant Physiol. 144:1667-1679.
Li C, W Xie, W Bai, Z Li, Y Zhao, H Liu (2008) Calmodulin binds to maize lipid transfer protein and modulates its lipids binding ability. FEBS J. 275:5298-5308.
Millar A A, S Clemens, S Zachgo, E M Giblin, D C Taylor, L Kunst (1999) CUT1, an Arabidopsis gene required for cuticular wax biosynthesis and pollen fertility, encodes a very-long-chain fatty acid condensing enzyme. Plant Cell 11:825-838.
Müller C (2006) Plant–insect interactions on cuticular surfaces: In: Biolo-gy of the Plant Cuticle. M Riederer, C Müller (eds). Julius-von-Sachs-Institut, für Biowissenschaften Universität Würzburg, Germany. pp:398-417.
Panikashvili D, A Aharoni (2008) ABC-type transporters and cuticle assembly linking function to polarity in epidermis cells. Plant Signal. Behav. 3:806-809.
Panikashvili D, S Savaldi-Goldstein, T Mandel, T Yifhar, R B Franke, R Hofer, L Schreiber, J Chory, A Aharoni (2007) The Arabidop-sis DESPERADO/AtWBC11 transporter is required for cutin and wax secretion. Plant Physiol. 145:1345-1360.
Parsons E P, S Popopvsky , G T Lohrey , S Lu, S Alkalai-Tuvia , Y Per-zelan , I Paran, E Fallik, M A Jenks (2012) Fruit cuticle lipid composition and fruit post-harvest water loss in an advanced backcross generation of pepper (Capsicum sp.). Physiol. Plant. 146:15-25.
Petit-Jiménez D, A González-León, G González-Aguilar, R Sotelo-Mundo, R Báez-Sañudo (2007) Cambios de la cutícula du-rante la ontogenia del fruto de Mangifera indica l. Rev. Fitotec. Mex. 30:51-60.
Petit-Jiménez D, E Bringas-Taddei, A González-León, J M García-Ro-bles, R Báez-Sañudo (2009) Efecto del tratamiento hidrotér-mico sobre la ultraestructura de la cutícula del fruto de mango. Rev. UDO Agríc. 9:96-102.
Pfündel E E, G Agati, Z G Cerovic (2006) Optical properties of plant surfaces: In: Biology of the Plant Cuticle. M Riederer, C Müller (eds). Julius-von-Sachs-Institut, für Biowissenschaften Univer-sität Würzburg, Germany. pp:216-239.
Pighin J A, H Zheng, L J Balakshin, I P Goodman, T L Western, R Jetter, L Kunst, L Samuels (2004) Plant cuticular lipid export requi-res an ABC transporter. Science 306:702-704.
Post-Beittenmiller D (1996) Biochemistry and molecular biology of wax production in plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 47:405-430.
Pollard M, F Beisson, Y H Li, J B Ohlrogge (2008) Building lipid barriers: biosynthesis of cutin and suberin. Trends Plant Sci. 13:236-246.
Pyee J, H Yu, P E Kolattukudy (1994) Identification of a lipid transfer protein as the major protein in the surface wax of broccoli (Brassica oleracea) leaves. Arch. Biochem. Biophys. 311:460-468.
Reina-Pinto J J, A Yephremov (2009) Surface lipids and plant defenses. Plant Physiol. Biochem. 47:540-549.
Riederer M (2006) Introduction: biology of the plant cuticle: In: Biology of the Plant Cuticle. M Riederer, C Müller (eds). Julius-von-Sachs-Institut, für Biowissenschaften Universität Würzburg, Germany. pp:1-8.
Riederer M, L Schreiber (2001) Protecting against water loss: analysis of the barrier properties of plant cuticles. J. Exp. Bot. 52:2023-2032.
Samuels L, L Kunst, R Jetter (2008) Sealing plant surfaces: cuticular wax formation by epidermal cells. Annu. Rev. Plant Biol. 59:683-707.
Schulz B, W B Frommer (2004) A plant ABC transporter takes the Lotus Seat. Science 306:622-625.
Shepherd T, D W Griffiths (2006) The effects of stress on plant cuticular waxes. New Phytol. 171:469-499.
Stark R, S Tian (2006) The cutin biopolymer matrix: In: Biology of the Plant Cuticle. M Riederer, C Müller (eds). Julius-von-Sachs-Institut, für Biowissenschaften Universität Würzburg, Ger-many. pp:126-141.
Suh M C, A L Samuels, R Jetter, L Kunst, M Pollard, J Ohlrogge, F Besis-son (2005) Cuticular lipid composition, surface structure, and gene expression in Arabidopsis stem epidermis. Plant Physiol. 139:1649-1665.
Van Maarseveen C, H Han, R Jetter (2009) Development of the cuticu-lar wax during growth of Kalanchoe daigremontiana (Hamet et Perr. de la Bathie) leaves. Plant Cell Environ. 32:73-81.
Yeats T H, K J Howe, A J Matas, G J Buda, T W Thannhauser, J K C Rose (2010) Mining the surface proteome of tomato (Solanum lyco-persicum) fruit for proteins associated with cuticle biogenesis. J. Exp. Bot. 61:3759-3771.
Yeats TH, G J Buda, Z Wang, N Chehanovsky, L C Moyle, R Jetter, A A Schaffer, J K C Rose (2012) The fruit cuticles of wild tomato species exhibit architectural and chemical diversity, providing a new model for studying the evolution of cuticle function. Plant J. 69:655-666.
Yeats TH, J K C Rose (2008) The biochemistry and biology of extracellu-lar plant lipid-transfer proteins (LTPs). Prot. Sci. 17:191-198.
21
CAPÍTULO II
RNA-Seq Analysis of the Mango (Mangifera indica L)
Fruit Peel: Towards the understanding of cuticle biosynthesis.
Tafolla-Arellano JC, Zheng Y, Sun H, Jiao C, Ruiz-May E, Hernández-Oñate M, González-León A, Báez-Sañudo
R, Fei Z, Rose JKC , Tiznado-Hernández ME. Preparado para el Journal of Experimental Botany.
23
Transcriptome Analysis of Mango (Mangifera indica L) 1
Fruit Peel: First Insights Towards Understanding 2
Cuticle Biosynthesis 3
4
Julio C. Tafolla-Arellano1,2,†, Yi Zheng3,†, Honghe Sun3, Chen Jiao3, Eliel 5
Ruiz-May4, Miguel Hernández-Oñate1, Alberto González-León1, Reginaldo 6
Báez-Sañudo1, Zhangjun Fei3,5, Jocelyn K.C. Rose2, Martín E. Tiznado-7
Hernández1* 8 9 1Coordinación de Tecnología de Alimentos de Origen Vegetal, Centro de 10
Investigación en Alimentación y Desarrollo, A. C. Km 0.6 carretera a la Victoria, 11
C.P. 83304, Hermosillo, Sonora, México. 12 2Plant Biology Section, School of Integrative Plant Sciences, Cornell University, 13
Ithaca, NY 14853, USA 14 3Boyce Thompson Institute for Plant Research, Cornell University, Ithaca, NY 15
14853, USA 16 4Red de Estudios Moleculares Avanzados, Instituto de Ecología A. C., Cluster 17
BioMimic®, Carretera Antigua a Coatepec 351, Congregación el Haya, C.P. 18
91070, Xalapa, Veracruz, México. 19 5 U.S. Department of Agriculture/Agriculture Research Service, Robert W. Holley 20
Center for Agriculture and Health, Ithaca, New York 14853, USA 21 †These authors contributed equally to this work 22
* To whom correspondence should be addressed: E-mail: tiznado@ciad.mx. Tel: 23
+52-662-2892400 ext. 346 Fax: +52-662-2800422 24 25 Email addresses: 26
JCTA: jctafare@gmail.com, YZ: yz357@cornell.edu, HS: hs738@cornell.edu 27
CJ: cj334@cornell.edu, ERM: eliel.ruiz@inecol.mx, MHO: 28
miguel.hernandez@ciad.mx, AGL: agonzalezl@ciad.mx, RBS: rbaez@ciad.mx, 29
ZF: zf25@cornell.edu, JKCR: jr286@cornell.edu. 30
Number of tables: 2, figures: 5 31
24
No one of color figures should be print 32
All color figures should be online-only. 33
Supplementary Data: 2 Figures; 3 Tables and 5 Datasets. 34
Running title: Mango fruit peel transcriptome 35
Highlight: The RNA-Seq transcriptome of mango peel helped to create a model 36
describing the cuticle biosynthesis phenomena and will assist with the 37
elucidation of other fruit physiological phenomena in the future. 38
Abstract 39
Mango fruit (Mangifera indica L.) is highly perishable with a limited shelf life, due 40
to postharvest desiccation and senescence, which leads to loss of tissue 41
integrity and microbial infection; factors that severely limit their global 42
distribution. Recent analyses in tomato suggest that these traits are influenced 43
by the expression of genes that are associated with cuticle metabolism in the 44
fruit epidermis. However, studies of these phenomena in mango fruit are 45
impeded by the lack of genome-scale data. In order to gain insight into the 46
cuticle biogenesis during mango fruit ontogeny, we analyzed the transcriptome 47
of ripe and overripe mango peels using RNA-Seq. 48
Approximately 400 million reads pairs were generated and de novo assembled 49
into 107,744 unigenes, with a mean length of 1,717 bp, grouped into 30,003 50
putative groups. A total of 91,736 (85.1%) unigenes showed homologous to 51
proteins in the UniProt/TrEMBL database. RNA-Seq analysis showed that cutin 52
monomers biosynthesis pathway is enriched during ripening. This was 53
confirmed by analysis of several cuticle-associated genes expression and 54
correlation with the rate of cuticle accumulation during the fruit ontogeny. The 55
present experiment uncovered the presence of a complex biphasic pattern of 56
cuticle deposition, and it provides data to propose a model for cuticle 57
biosynthesis in mango. The results of this present study provide a valuable 58
genomic resource for future molecular research into the biology and postharvest 59
quality traits of mango. 60
Keywords: Mango, fruit peel, cuticle, biosynthesis, RNA-Seq, transcriptome. 61
Background 62
25
Mango (Mangifera indica Linn.) is a large drupe and commercially important 63
tropical fruit known as “The king of fruits” (Mukherjee and Litz, 2009). Mango 64
fruits under tropical conditions ripen within 6 to 7 days and become overripe and 65
spoiled within 15 days after harvest (Vazquez-Salinas and Lakshminarayana, 66
1985). Postharvest desiccation leads to oversoftening, loss of tissue integrity 67
and microbial infection (Martin and Rose, 2014), which limits its availability in the 68
markets by causing postharvest losses. 69
The exocarp influences the outward appearance of the fruit (color, glossiness, 70
texture, and uniformity), and it appears to play an important role in the shelf life 71
(Mandel et al., 2007). The exocarp usually called ‘’peel’’ or “skin” is composed of 72
cuticle, epidermis, collenchyma, and even parenchyma tissues, depending on 73
how the peel was physically removed (Mintz-Oron et al., 2008). 74
All aerial plant organs, including the fleshy fruits, are covered by a hydrophobic 75
layer that is a barrier between the fruit mesocarp and its environment, composed 76
mostly of cutin and waxes, known as cuticle which is synthesized in epidermal 77
cells (Samuels et al., 2008). Cuticle acts mainly by reducing the water loss and 78
gas diffusion (Riederer and Schreiber, 2001), providing protection against 79
insects (Müller, 2006), pathogens attack (Carver y Gurr, 2006), UV radiation 80
(Pfündel et al., 2006), maintains palatability and promotes seed dispersal (Martin 81
and Rose, 2014), among other functions. Therefore, cuticle composition and 82
physical properties are suggested to play an important role in fruit quality and 83
postharvest shelf life (Saladié et al., 2007). Thus, an understanding of cuticle 84
formation at the molecular level is fundamental for designing of strategies to 85
improve fruit quality. 86
Cuticle-related genes studies have been carried out mainly on the vegetative 87
organs of the model plant Arabidopsis (Martin and Rose, 2014). However, 88
tomato fruit have become a model system for the study of the cuticle biology in 89
fleshy fruits (e.g., Isaacson et al., 2009; Matas et al., 2011; Yeats et al., 2012). 90
Also, additional information is available for other fruit such as apple (Albert et al., 91
2013) and sweet cherry (Alkio et al., 2012; Alkio et al., 2014; Balbontín et al., 92
2014). Despite the agronomic and economic importance of mango fruit, there is 93
26
no information about the molecular mechanism of cuticle biosynthesis, and only 94
few studies addressing the changes of fruit cuticle composition and morphology 95
during development and storage (Bally, 1999; Petit-Jiménez et al., 2007), or in 96
response to particular procedures such as kaolin treatment (Du Plooy et al., 97
2004;) and hydrothermal treatment (Jacobi and Gowanlock, 1995; Petit-Jiménez 98
et al., 2009) are available. The major obstacle to further progress in mango 99
genetic research is the limited availability of genomic data. However, the next 100
generation DNA sequencing methods and bioinformatics pipelines are 101
underpinning the generation of genomic and transcriptomic resources even 102
when the genome of the organisms is not available. The application of this tools 103
to study cuticle-associated genes will be crucial to elucidate many aspects of 104
cuticle biology that are not well understood (Martin and Rose, 2014). 105
The objective of our study was to gain insight into the cuticle biogenesis during 106
mango fruit ontogeny. To this aim, we applied High Throughput RNA 107
Sequencing in ripe and overripe fruit mango peels. We identified 5,349 108
differentially expressed unigenes by comparing the overripe and ripe fruit. 109
Moreover, the functional analysis indicates that the cutin monomers biosynthesis 110
pathway is enriched during ripening. Furthermore, the expression profile of 111
cuticle-associated genes were analyzed by real time quantitative reverse 112
transcription PCR (qRT-PCR) in mango peel and correlated with the rate of 113
cuticle accumulation during the fruit ontogeny. The cuticle deposition in mango 114
showed a biphasic pattern during fruit development, characterized by one phase 115
of accumulation during the early stages of fruit growth followed by a second 116
phase of maximal cuticle deposition during ripening and senescence. Besides, 117
we proposed a model of mango cuticle biosynthesis and discussed the putative 118
roles for these genes in the biosynthesis and transport of cutin and waxes during 119
the cuticle formation. Additionally, we carried out a genomic comparative 120
analysis suggesting that the mango experimented a recent Whole Genome 121
Duplications (WGD) event approximately 14.01-16.03 million years ago (MYA), 122
after the divergence of mango and orange, which occurred approximately 57-62 123
MYA. 124
27
Methods 125
Plant materials 126
Mangoes (Mangifera indica L) cultivar `Keitt´ used for RNA-Seq were obtained 127
from a commercial store at Ithaca, New York, USA. The label legend in the 128
mango was: Melissa’s tree ripened mango emp #020, PLU code #3365 (Mango 129
Ripe/Ready-to-Eat) from Mexico. We divided the mango fruits into two stages: 130
ripe and overripe (storing them at room temperature approximately 20°C and 60-131
65% relative humidity during 12 days). Mangoes cv. `Keitt´ used for qRT-PCR in 132
mango fruit ontogeny, were hand harvested every 15 days after flowering (DAF) 133
until ripening in a commercial orchard and packinghouse located at El Porvenir, 134
Ahome, Sinaloa, Mexico (www.agricoladaniella.com.mx). After ripening, 135
mangoes were stored at 20°C and 60-65% relative humidity during 18 days, with 136
sampling carried out every six days. 137
RNA-Seq library construction and sequencing 138
Peel tissue samples from three fruits were pooled together to create one 139
biological replicate and it was utilized three independent biological replicates 140
each for ripe and overripe mango. Total RNA was extracted using hot borate 141
method with minor modifications (Wan and Wilkins, 1994). 142
Strand-specific RNA-Seq libraries were constructed using the protocol described 143
in Zhong et al., (2011). The resulting six RNA-Seq libraries were sequenced on 144
an Illumina HiSeq 2500 system (Illumina Inc. San Diego, CA, USA) with the 145
paired-end mode and read length of 100 bp in the Institute of Biotechnology at 146
Cornell University (http://www.biotech.cornell.edu/biotechnology-resource-147
centerbrc). The raw sequencing reads were deposited in NCBI Sequence Read 148
Archive (SRA) under the accession number SRP043494. 149
RNA-Seq data processing, de novo assembly and annotation 150
RNA-Seq reads were first processed to trim adapter and low quality sequences 151
using Trimmomatic (Bolger et al., 2014). Reads shorter than 40 bp were 152
discarded. The resulting high-quality cleaned reads were assembled de novo 153
into contigs using Trinity with “min_kmer_cov” set to 10 (Grabherr et al., 2011). 154
28
Following assembly, the high-quality cleaned reads were aligned to assembled 155
contigs using Bowtie (Langmead et al., 2009) allowing 3 mismatches. Following 156
alignments, expression values (FPKM; fragments per kilobase of exon model 157
per million mapped reads) were derived for each contig. Low expressed contigs 158
(FPKM < 0.002), and contigs with low ratio number of sense to antisense reads 159
(< 0.1) were discarded as those sense reads may be derived from incomplete 160
digestion of 2nd-strand during the strand-specific RNA-Seq library construction. 161
The resulting assembled contigs were then blasted against GenBank Nucleotide 162
(nt) database and those having hits only to sequences from viruses, bacteria, 163
and archaea were discarded. Next, the rRNA, low-complexity, and polyA/T 164
sequences were removed or trimmed from the contigs using SeqClean 165
(https://sourceforge.net/projects/seqclean/). To remove redundancies in the 166
contigs, the remaining contigs were further de novo assembled using 167
iAssembler (Zheng et al., 2011) with 97% minimum percent identify. 168
The final assembled mango unigenes were blasted against the UniProt (Swiss-169
Prot and TrEMBL) and Arabidopsis protein databases with a cutoff E-value of 170
1e-5. Based on the blast results, precise functional descriptions (human 171
readable description) were assigned to each mango transcript using automated 172
assignment of human readable descriptions (AHRD: 173
https://github.com/groupschoof/AHRD). Gene ontology (GO) terms were 174
assigned to the mango assembled transcripts based on the GO terms annotated 175
to their corresponding homologues in the UniProt database (The gene ontology 176
consortium, 2015). Biochemical pathways were predicted from the mango 177
transcripts using the Pathway Tools (Karp et al., 2002). Transcription factors and 178
protein kinases were identified and classified into different families using the 179
iTAK pipeline (http://bioinfo.bti.cornell.edu/tool/itak). 180
Gene expression quantification and differential expression analysis 181
The high-quality cleaned RNA-Seq reads were aligned to the assembled mango 182
transcripts with the Bowtie program (Langmead et al., 2009) allowing 3 183
mismatches and only keeping the best alignments. Following alignments, raw 184
counts for each mango transcript and in each sample were derived and 185
29
normalized to FPKM. Differentially expressed genes (fold changes > 2 and 186
adjusted p-value <0.05) between ripen and overripe fruits were identified with 187
the DESeq package (Anders and Huber, 2010). GO terms enriched in the set of 188
differentially expressed genes and altered pathways were identified using the 189
Plant MetGenMAP system (Joung et al., 2009). 190
RNA isolation and cDNA synthesis for developmental time course 191
Peel tissue samples from three fruits were pooled together to create one 192
biological replicate and it was utilized three independent biological replicates for 193
each mango developmental stage. Total RNA was extracted using hot borate 194
method with minor modifications (Wan and Wilkins, 1994), and 2 µg of total 195
DNase-treated RNA was used for first strand cDNA synthesis using SuperScript 196
II reverse transcriptase and oligo(dT) primers (Invitrogen), according to the 197
manufacturer’s instructions. 198
Real-time quantitative reverse transcription PCR 199
DNA sequences for mango genes orthologues to Tomato and Arabidopsis 200
associated with cuticular biosynthesis, transport and regulation were obtained 201
from our Mango database using Basic Local Alignment Search Tool (Altschul et 202
al., 1997). The coding DNA sequence (CDS) and amino acid sequence deduced 203
were obtained using Open Reading Frame Finder 204
(http://www.ncbi.nlm.nih.gov/gorf/gorf.html), Signal-3L to identify the signal 205
peptide (Hong-Bin and Kuo-Chen, 2007) and PredGPI predictor for 206
glycosylphosphatidylinositol (GPI) domain (Pierleoni et al., 2008). The primer 207
designs were carried out according to the protocol described in Thornton and 208
Basu (2011). The sequences of oligonucleotide primers are listed in the 209
Supplementary Table S1. 210
Quantitative PCR experiments were performed using a StepOne™ Real-Time 211
PCR System (Applied Biosystems, Foster City, CA, USA). The cDNA samples 212
were diluted 5-fold with water and 1 µl was used as a template for each 20 µl 213
quantitative PCR, prepared using HotStart-IT SYBR Green qPCR Master Mix 214
(2X) (Affymetrix, Santa Clara, CA, USA) in biological triplicates. The thermal 215
cycling conditions consisted of 2 min at 95°C, followed by 40 cycles at 216
30
95°C for 15 s and 60°C for 1 min. Specificity of the PCR products was 217
determined by high-resolution melt curve analysis, gel electrophoresis and by 218
sequencing carried out at Macrogen (Macrogen Inc., Seoul, South Korea). 219
Statistical analysis of relative expression results was carried out using the 220
relative expression method (Livak and Schmittgen, 2001), with MiActin1 221
(GenBank accession No.JF737036) as reference gene (Luo et al., 2013), 222
assuming PCR efficiency of 1.0 for all genes. For each gene, expression was 223
linearly normalized, with a value of 0.0 assigned to the stage with lowest 224
expression and 1.0 to the stage showing the highest expression. Normalized 225
gene expression profile data were converted into a heat map using the gplots R 226
library (http://cran.r-project.org/web/packages/gplots/index.html). 227
Cuticle Microscopy 228
Mango peel from both ripe and overripe fruits was harvested, fixed, 229
cryoprotected, and embedded as outlined by Buda et al., (2009). Cryosections 230
of each mango sample were cut and melted at room temperature onto 231
VistaVision HistoBond slides (VWR), dried and stained with Oil Red O 232
(saturated in 60% isopropanol). Tissue sections were imaged with a Zeiss 233
AxioImager A1 microscope (Zeiss) equipped with a Zeiss EC-Plan NeoFluar 234
3100/1.3 oil immersion objective, a Zeiss AxioCam MRc color video camera, 235
and Zeiss AXIOVs40 4.6.3.0 software. Images were obtained using DIC optics 236
on an AxioImager A1 microscope equipped with an EC-Plan NeoFluar 40x/0.75 237
objective and an AxioCam Mrc color video camera (Zeiss, 238
http://www.zeiss.com/). 239
Quantification of mango fruit cuticle 240
The changes in cuticle deposition during mango ontogeny were quantified 241
following essentially a previously described protocol (Petit-Jiménez et al., 2007). 242
Whole-genome duplication (WGD) analysis 243
To investigate possible WGD and speciation events in mango, we carried out a 244
Whole-genome duplication (WGD) analysis. The results are included in 245
Supplementary Dataset S1. 246
Results 247
31
Cuticle accumulation during mango fruit ontogeny 248
The cuticular weight was 889 µg/cm2 at 15 DAF, and it was continued increasing 249
with the advance of fruit development (30-105 DAF) reaching 1763 µg/cm2. 250
Thereafter, it showed a decrease to 1632 µg/cm2 at 120 DAF. After that, 251
cuticular weight showed an increase of 1846 and 1920 µg/cm2 from 135 to 141 252
DAF, respectively. Unlike the foregoing pattern, a slight decrease to 1827 253
µg/cm2 was recorded at 147 DAF. Finally, the cuticle accumulation showed a 254
large increase by the end of the storage time to reach 2100 µg/cm2 at 153 DAF, 255
which is the maximum cuticular weight registered during mango fruit ontogeny 256
(Supplementary Figure S1). 257
In order to gain an overview of changes in cuticle deposition and structural 258
organization in mango fruit cuticle during ripe and overripe stage of development 259
(Fig. 1A-B), we used light microscopy to visualize Oil Red O-stained pericarp 260
sections (Fig. 1C-D) and SEMs microscopy (Fig. 1E-F). It was found a structural 261
variation in cuticle thickness during storage. In figures 1C and 1D it can be 262
clearly seen a slight reduction in the cuticle between the ripe and overripe 263
mango fruit. 264
De novo assembly, functional annotation and classification of mango 265
unigenes 266
We sequenced six cDNA libraries (three ripe and three overripe) of mango peel 267
cv. `Keitt´ using the Illumina HiSeq™ 2500 system with the paired-end mode. 268
We obtained an average of about 74 million reads pairs for each mango peel 269
ripe library and 57 million for each mango peel overripe library. After removing 270
low-quality and adaptor sequences, a total of around 681 million reads (62.5 271
Gbp high-quality sequences) were used to carry out the de novo assembly. A 272
total of 107,744 unigenes were assembled with a total length of 184,977,733 bp, 273
a mean length of 1,717 bp and a N50 of 2,235. Moreover, 3,243 (3%) unigenes 274
were <300 bp; 69,214 (64.23%) and 33,426 (31.02%) unigenes showed a length 275
in a range of 300 to 2,000 and 2,000 to 5000 bp, respectively. In addition, 2,235 276
(2.07%) unigenes were longer than 5,000 bp and the largest unigene was 277
12,271 bp (Fig. 2A). Then, unigenes were grouped into 30,003 putative groups 278
32
with a mean of 3 unigenes per group. A distribution analysis showed that 23,899 279
groups are constituted by 1 to 3 unigenes, 2,778 groups are composed by 4-6 280
unigenes and 474 groups are integrated by more than 30 unigenes (Fig. 2B). 281
With the objective to efficiently distribute our transcriptome sequences and the 282
associated analysis results to allow the research community to mine the mango 283
transcriptome dataset, we developed an online database called CIAD-Cornell 284
University Mango RNA-Seq Database, which can be accessed at 285
http://bioinfo.bti.cornell.edu/cgi-bin/mango/index.cgi. 286
The annotation of unigenes sequences was performed using a BLAST analysis 287
through homologous search against different protein databases. The numbers of 288
unigenes showing significant hits (E-value ≤ 1e-5) to Swiss-Prot, TrEMBL and 289
Arabidopsis protein databases were 68,649 (63.7%), 91,736 (85.1%) and 290
88,242 (81.9%), respectively. After removing 15,984 unigenes without 291
significantly hits to Swiss-Prot and TrEMBL, a total of 91,760 unigenes were 292
assigned to Human Readable Description, among which 5,705 were annotated 293
as unknown proteins. 294
Furthermore, the unigenes were classified into different Gene Ontology (GO) 295
terms. A total of 79,208 unigenes were assigned into 9,945 GO terms and out of 296
these, 67,003, 68,347 and 67,160 were assigned with at least one GO term in 297
the biological process, cellular component and molecular function categories, 298
respectively, while 53,585 were annotated with GO terms from all three 299
categories. Additionally, we identified the biochemical pathways represented in 300
the assembled mango transcriptome. Our analysis showed that a total of 7,740 301
unigenes were annotated in 461 pathways. Finally, we identified unigenes 302
encoding for transcription factors (TF) and protein kinases. A total of 5,342 TF 303
were classified into 55 different families and 3,662 protein kinases were 304
classified into 76 different families, respectively. A summary of mango unigenes 305
annotation is shown in Fig. 2C and total annotation inSupplementary Dataset 306
S1. 307
Differential expression analysis 308
33
Comparison of peel samples of overripe and ripe developmental stages of 309
mango using RNA-Seq showed that 1,616 and 3,733 unigenes were up-310
regulated and down-regulated, respectively. The enrichment analyses of the up-311
regulated unigenes showed about 215 biological processes GO terms enriched, 312
including those related to different stress responses, lipid metabolic process, 313
secondary metabolic process, cell wall metabolic process, and fruit ripening. For 314
down-regulated unigenes, the enrichment analysis identified 556 significantly 315
enriched biological processes including those related to the response to abiotic 316
and biotic stimulus such as water deprivation, cell wall polysaccharide metabolic 317
process, fruit development, response to ethylene, secondary metabolic process 318
and response to lipid. The top ten GO enrichment analysis is show in Figure S2 319
and the total analysis is shown in Supplementary Dataset S2). 320
In addition, the clustering of 5,349 differentially expressed unigenes by 321
expression levels grouped them in 7 clusters (Fig. 3A). Clusters I and II 322
comprise 66 and 213 induced unigenes, respectively. These are related to 323
glutathione-mediated detoxification II, homogalacturonan degradation, fatty acyl-324
CoA reductase and starch biosynthesis among others. Interestingly, cluster III 325
constituted by 1,318 induced unigenes related mainly to cutin monomers 326
biosynthesis, flavonoid biosynthesis, oleate biosynthesis and homogalacturonan 327
degradation, among others. Cluster IV encompases 19 strongly induced 328
unigenes with no clear function associated. On the other hand, the down-329
regulated unigenes were grouped into 3 clusters (V, VI and VII), which are 330
related mainly with cytochrome P450 activity, carbohydrates degradation, 331
glutathione transferase, chalcone synthase, chorismate biosynthesis and 332
ethylene biosynthesis among others (Fig. 3A and Supplementary Dataset S3). 333
Metabolic pathway enrichment analysis showed that the up-regulated unigenes 334
encoded enzymes in 106 metabolic pathways and the down-regulated unigenes 335
encoded enzymes in 131 pathways. The most up-regulated pathways included: 336
4-hydroxybenzoate biosynthesis V, phenylpropanoid biosynthesis, cutin 337
monomers biosynthesis, flavonoid biosynthesis, and homogalacturonan 338
degradation and the most down-regulated pathways included: glutathione-339
34
mediated detoxification II, chorismate biosynthesis I, chorismate biosynthesis 340
from 3-dehydroquinate, lactose degradation III and 1,4-dihydroxi-2-naphthoate 341
biosynthesis II (Fig. 3B and Supplementary Dataset S4). 342
Furthermore, we also identified the differentially expressed unigenes encoding 343
TF and protein kinases. A total of 342 and 92 transcription factors classified into 344
25 and 20 families were up-regulated and down-regulated, respectively. On the 345
other side, 119 and 54 proteins kinases classified into 17 and 19 families were 346
up-regulated and down-regulated, respectively (Supplementary Table S2-3). 347
qRT-PCR data validation and analysis of cuticle biosynthesis gene 348
expression during ontogeny 349
In order to corroborate the gene expression of our RNA-Seq data, we selected 350
fifteen candidates genes associated with cuticle biosynthesis, regulation and 351
transport (Table 1). To this end, we used qRT-PCR with cDNA isolated from ripe 352
and overripe mango peels. We compared the log2fold change (overripe vs ripe) 353
obtained by RNA-Seq and qRT-PCR analyzes (Fig. 3C). Linear regression 354
analysis showed a r2=0.798 and a Pearson correlation coefficient of 0.893, with 355
a p-value of 2.009e-5 indicating a large correlation between transcript abundance 356
quantified by qRT-PCR and the transcription profile obtained by RNA-Seq data 357
further supporting the accuracy of the data (Fig. 3D). 358
Furthermore, in order to get insight into the cuticle biogenesis we analyzed the 359
gene expression of fifteen candidate genes throughout mango fruit ontogeny 360
(Fig. 4). We found that MiCUS1, MiKCS2, MiLTP3, MiCER3, MiWBC11 and Mi 361
LTPG1 showed a similar expression patterns, characterized by medium 362
expression level during initial stages, low expression during intermediates 363
stages and an increased expression during storage (141-153 DAF), reaching the 364
maximum expression in final stage at 153 DAF. By other side, MiLTP1 and 365
MiCUS2 showed a similar gene expression, characterized by low gene 366
expression during initial and intermediate stages, increasing its expression at 367
147 DAF and reaching the maximum expression at 153 DAF during major 368
cuticle accumulation. Moreover, MiCER1, MiCD2, MiPEL1 and MiLTP2 had a 369
similar gene expression, characterized by a hardly detectable gene expression 370
35
during initial and intermediate stages (15-141 DAF), increasing its expression at 371
147 DAF and reaching the maximum expression at 153 DAF during major 372
cuticle accumulation. Moreover, MiCER2 and MiSHN1 had a similar gene 373
expression, characterized by low gene expression between 15 and 141 DAF, an 374
increased expression at 147 DAF, reaching the maximum expression at 153 375
DAF during major cuticle accumulation. Finally, MiKCS6 showed a high gene 376
expression during initial stages (15-60 DAF), low expression at 75 DAF, then 377
showed an increased expression between 90-147, reaching the maximum 378
expression during storage. However, this gene showed a hardly detectable 379
expression at 153 DAF during major cuticle accumulation. 380
All the RT-PCR products showed the predicted sizes after separation on 381
agarose gels. Further, the sequences of the amplified DNA fragments 382
correspond with the expected nucleotide sequences of the candidate genes 383
(data not shown). 384
385
Discussion 386
In recent years, we have been witnessing an increment in the international 387
commercialization of tropical fruits, including mangoes. However, several 388
important factors associated with postharvest shelf life and pathogen infection is 389
halting the presence of mango in the international market. Besides, there are 390
limited reports associated with the genomic information in mango. Only few 391
reports in the literature about de novo transcriptomic in mango are available. 392
Such as, in mango leaf (Azim et al., 2014), fruit pericarp and pulp (Wu et al., 393
2014); fruit response to hot water treatment (Luria et al., 2014), and in mango 394
mesocarp during ripening (Dautt-Castro et al., 2015). 395
Here we report a robust de novo transcriptome assembly of mango peel during 396
ripening, the robustness of our the transcriptome assembly is supported by the 397
N50 value of 2,235 bp and mean length of 1,717 bp, which are higher that those 398
previously reported for mango by Luria et al., (2014), who showed a N50 value 399
of 1,598 bp and a mean length of 863.3 bp. Furthermore, the transcriptome 400
assembly allowed us to identify 107,744 unigenes, almost twice of the 57,544 401
36
contigs reported by Luria et al., (2014). Our transcriptome summary and 402
comparison with previously reported is shown in Table 2. 403
Waxes, cutin, isoprenoids and phenylpropanoids are the four major metabolic 404
pathways involved in tomato cuticle biosynthesis (Mandel et al., 2007). Lipid 405
metabolism is essential for cuticle assembly because it plays an important role 406
supplying the precursors for the biosynthesis of wax and cutin. Indeed, Suh et 407
al., (2005) reported that over half of the fatty acids sinthesized in Arabidopsis 408
stem epidermis are exported into cuticle. By other side, Mintz-Oron et al., (2008) 409
reported that the 15% of genes associated with fatty acid metabolism, wax and 410
cutin were up-regulated in tomato peel during cuticle formation. Secondary 411
metabolites synthesized by the phenylpropanoid pathway are often constituents 412
of cuticular waxes (Mintz-Oron et al., 2008). In agreement with these data, our 413
transcriptome analysis allowed us to identify a total of 136 and 45 unigenes up-414
regulated involved in lipid metabolism process (GO:0006629) and cutin 415
biosynthetic process (GO:0010143), respectively. Furthermore, 53 and 4 416
unigenes involved in cutin biosynthesis pathway (PWY-321) were found to be 417
up-regulated and down-regulated, respectively, suggesting that an active cutin 418
biosynthesis was taking place. 419
In addition, We identified 123 and 110 unigenes up-regulated involved in 420
secondary metabolic process (GO:0019748) and phenylpropanoid metabolic 421
process (GO:0009698), indicating that the pathways related with the 422
biosynthesis of cuticle components were very active during mango fruit ripening. 423
Gene expression during cuticle biosynthesis 424
Regulation 425
The regulation of cuticle biosynthesis involves feedback from the cuticle 426
components and with interacting metabolic pathways playing a role in responses 427
to pathogen and environmental stress (Yeats and Rose, 2013). 428
In Arabidopsis, the first cuticle-associated transcription factor, the AP2-domain 429
super family member SHINE1/WAX INDUCER1 (AtSHN1/WIN1) induced the 430
expression of a large numbers of gene encoding enzymes that are involved in 431
fatty acid elongation and the formation of aliphatic compounds (Aharoni et al., 432
37
2004; Broun et al., 2004 Kannangara et al., 2007). We characterized a mango 433
homolog of AtWIN/SHN1, MiWIN/SHN1 (MIN047952), with a 67.68% of protein 434
identity. This gene showed a weak expression between 90 and 141 DAF, 435
increasing strongly its expression during the storage period (147-153 DAF), 436
which corresponds with the stage in which the fruits experienced a large cuticle 437
accumulation. 438
In agreement, the overexpression in tomato of SlSHN1 induced a higher 439
cuticular wax deposition as compared with the isogenic tomatoes lines (Al-440
Abdallat et al., 2014). Besides, in cherry sweet (Prunus avium), PaWIN/SHN1 441
gene showed a high expression at 21 days after full bloom (DAFB) during the 442
major cuticle accumulation (Alkio et al., 2012). 443
The transcription factor, CD2 (CUTIN DEFICIENT 2), an HD-Zip IV member was 444
proposed as a key regulator of cutin biosynthesis in tomato fruit. Indeed, the cd2 445
mutant showed a cuticle extremely thin, and an increased susceptibility to 446
microbial infection (Isaacson et al., 2009). Further, Matas et al., (2011) reported 447
that the CD2 was the most differentially expressed transcription factor in tomato. 448
In this sense, we characterized a mango homolog of LeCD2, MiCD2 449
(MIN074277), with 86% of protein identity. This gene showed a low expression 450
during the initial and intermediate stages of development (15-141 DAF), with a 451
strong increase during the storage time (147-153 DAF), in which it was taking 452
place a large accumulation of cuticle. The MiCD2 expression behavior correlates 453
with cuticle accumulation, suggesting that this gene plays an important role in 454
this phenomena. 455
Wax biosynthesis 456
The CER1 is a gene that encodes an aldehyde decarbonylase enzyme and 457
catalyzes the conversion of long chain aldehydes to alkanes, a key step in wax 458
biosynthesis (Aarts et al., 1995; Bernard et al., 2012). Bourdenx et al., (2011) 459
demonstrated that the overexpression of CER1 increased specifically the odd-460
carbon-numbered alkanes, mainly C27, C29, C31, and C33 alkanes. We 461
characterized a mango homolog of AtCER1, MiCER1 (MIN107433, 63.71% of 462
protein identity. This gene showed a very low expression during initial and 463
38
intermediate stages of development (15-141 DAF), and a strong increase in 464
expression during ripening and storage (147-153 DAF), the stage of a large 465
cuticle accumulation. Moreover, Albert et al., (2013) correlated the abundance of 466
alkanes in apple with the expression of the homolog MdCER1. By other side, 467
Broun et al., (2004) reported that CER1 gene expression was induced by 468
overexpression of AtWIN1/SHN1. In agreement with this, our data clearly 469
showed that the MiSHN1 transcripts are present before the expression increase 470
of the MiCER1 gene which suggest that this gene can be controlled by the 471
transcription factor MiSHN1, although more experimental evidences are needed 472
to probe this statement. 473
The CER2 gene encodes a putative BAHD acyltransferase involved in the 474
elongation of alkanes beyond C28. In Arabidopsis, cer2 mutant lacks waxes 475
longer than C28, suggest that CER2 plays a critical role in very long chain fatty 476
acids (VLCFA) synthesis (Xia et al., 1996; Haslam et al., 2012). We 477
characterized a mango homolog of AtCER2, MiCER2 (MIN052433), with 478
42.65% of protein identity. This gene showed a low expression during initial and 479
intermediate stages of development (15-141 DAF). Thereafter, this gene 480
showed an increased expression during ripening and storage time (147-153 481
DAF), during the major cuticle accumulation. The MiCER2 expression pattern 482
appears to correlate with the time of a large cuticle acumulation. It will be 483
interesting to test whether this expression increase with the accumulation of 484
VLCFA in cuticle that will give more insights to answer whether this gene plays 485
the same role as the homolog in Arabidopsis. 486
The CER3 (WAX2) was suggested to form an enzymatic complex catalyzing the 487
conversion of very long chain (VLC) acyl-CoAs to VLC alkanes (Bernard et al., 488
2012). The total wax amount on Arabidopsis cer3 mutant leaves and stems was 489
reduced by 78%. Also, it showed a reduction amount in aldehydes, alkanes and 490
secondary alcohols (Jenks et al., 1995; Chen et al., 2003; Rowland et al., 2007). 491
We characterized a mango homolog of AtCER3, MiCER3 (MIN064126), with 492
69.73% of protein identity. This gene showed low expression during initial and 493
intermediate stages of development (15-141 DAF), and high expression during 494
39
ripening and storage (147-153 DAF), a stage of the major cuticle accumulation. 495
The KCS2/Daysi gene encodes a 3-Ketoacyl-COA synthase 2 and it is 496
functionally redundant catalyzing the two-carbon elongation leading to C22 497
VLCFA that is required for cuticular wax and root suberin biosynthesis (Lee et 498
al., 2009a). We characterized a mango homolog of AtKCS2, MiKCS2 499
(MIN101804), with 78.25% of protein identity. This gene showed a high 500
expression during the initial stages of development (15-30 DAF). After that, it 501
exhibited a low expression during intermediate stages of development (75-141 502
DAF) and an increased expression during ripening and storage (147-153 DAF), 503
although much less as compared with the initial stages. The expression of 504
MiKCS2 does not correlate with the cuticle accumulation, and maybe it is more 505
related with changes in cuticle composition, although more experimental 506
evidences are needed to probe this statement. 507
The KCS6 (CER6 or CUT1) gene encodes a 3-Ketoacyl-CoA synthase 6 (Millar 508
et al., 1999; Fiebig et al., 2000; Costaglioli 2005; Leide et al., 2007). A loss-of-509
function tomato mutant showed a reduction in n-alkanes and aldehydes with 510
chain lengths beyond C30 in waxes of leaf and fruit (Vogg et al., 2004). We 511
characterized a mango homolog of AtKCS6, MiKCS6 (MIN040156), with 84.88% 512
of protein identity. This gene showed high expression during initial stages (15-45 513
DAF), reaching the maximum expression at 60 DAF. After that, it showed a low 514
expression during intermediate stages of development (75-141 DAF) and an 515
expression levels similar to the initial stages during ripening and storage time 516
(147-153 DAF). The MiKCS6 expression does not correlate with cuticle 517
accumulation. A similar behavior was observed by Alkio et al., (2012) in sweet 518
cherry, in which, PaKCS6 expression did not show an increase expression 519
during cuticle deposition at 30 and between 60-100 DAFB. Conversely, Yeats et 520
al., (2010) found a high expression of SlKCS6 (CER6) gene during the most 521
rapid phase of fruit expansion, and during 15-20 DPA, in which it is taking place 522
a large cuticle accumulation. The expression pattern of SlKCS6 is different as 523
compared with the expression of MiKCS6, however, both genes showed a 524
similar expression during the initial stages of fruit development. 525
40
Wax transport 526
Cuticle biosynthesis requires an extensive transport of lipids through the plasma 527
membrane and cell wall of the epidermal cells. ATP binding cassette (ABC) 528
transporters located in the plasma membrane of epidermal cells are required for 529
both cutin and wax deposition (Pighin et al., 2004). 530
In Arabidopsis, the WBC11 gene encodes an ABC transporter and it was 531
localized in the plasma membrane (Bird et al., 2007; Panikashvili et al., 2007). 532
WBC11 is involved in the export of both cutin precursors and wax during cuticle 533
development. The wbc11 mutant plant lines showed a 75–90% reduction of 534
alkanes, being the largest reduction recorded in the C29 alkane (Bird et al., 535
2007; Luo et al., 2007). We characterized a mango homolog of AtWBC11, 536
MiWBC11 (MIN106958), with 84.78% of protein identity. This gene showed low 537
expression during initial and intermediate stages (15-141 DAF), and a high 538
expression during ripening and storage time (147-153 DAF) during the major 539
cuticle accumulation. Bird et al., (2007) reported that alkane levels of Atwbc11 540
correlated with the levels of detectable WBC11 transcript. In sweet cherry, Alkio 541
et al., (2012) reported significant positive correlations between transcript levels 542
of the PaWBC11 and the rate of cuticle deposition between 20-30 DAFB. 543
However, PaWBC11 expression did not increase during cuticle deposition 544
between 55-100 DAFB. 545
The transport mechanism throug the cell wall is not well understood. Lipid 546
transfer proteins (LTPs) have been proposed to be involved in cuticular transport 547
across the cell wall, because they are abundantly expressed in the epidermis, 548
are secreted into the apoplast and are small enough to transverse the cell wall 549
(Kader, 1996; Yeats et al., 2008). 550
Yeats et al., (2010) reported changes in LTPs expression during cuticle 551
biosynthesis in tomato fruit ontogeny. We characterized three mango homologs 552
of tomato LTP SGN-U579687 (Yeats et al., 2010), 1) MiLTP1 (MIN026365), with 553
54.39% of protein identity; 2) MiLTP2 (MIN018326) with 47.79% of protein 554
identity; 3) MiLTP3 (MIN107167), with 36.84% of protein identity. 555
Our expression analysis showed that the MiLTP1 and MiLTP2 unigenes have a 556
41
low expression level during initial (15-141) and intermediate stages (15-135 557
DAF) of fruit development, respectively. In contrast, MiLTP3 showed medium 558
expression levels during initial stage (15-90 DAF) and low expression during 559
intermediate stage (105-141 DAF). Interestingly, the three unigenes showed a 560
high expression during ripening and storage time (147-153 DAF), which are the 561
period of major cuticle accumulation in fruit. These results suggest that these 562
genes could be participating in the cuticle accumulation during the late stages. 563
Conversely, Yeats et al., (2010), reported low expression at initial stages (5-20 564
DPA), during the major cuticle accumulation. The gene expression increased 565
during intermediates (30 DPA) and final stages (turning), which did not correlate 566
with the major cuticle accumulation. Mintz-Oron et al., (2008) reported high 567
expression of SGN-U579687 in tomato peel transcripts, which is homolog to 568
MiLTP1. In agreement, we recorded a higher MiLTP1 gene expression as 569
compared with all other genes analyzed. 570
In Arabidopsis, export of some wax compounds also appears to be facilitated by 571
glycosylphosphatidylinositol (GPI)-anchored lipid-transfer proteins known as 572
LTPG. LTPG1 was localized in the plasma membrane and ltpg1 mutant with 573
decreased expression showed a reduced wax load mainly in C29 alkane, and 574
higher susceptibility to infection by Alternaria brassicicola than the wild type 575
(DeBono et al., 2009; Lee et al., 2009). We characterized a mango homolog of 576
AtLTPG1, MiLTPG1 (MIN012243), with 48.68% of protein identity. This gene 577
showed medium expression during initial stages of development (15-60 DAF), 578
low expression in intermediates stages (75-135 DAF). Finally, it was recorded 579
the maximum expression level during ripe and storage time (153 DAF) during 580
the major cuticle accumulation. In sweet cherry, Alkio et al., (2012) reported 581
positive correlations between PaLTPG1 expression and the major cuticle 582
accumulation during initial stages (20-30 DAFB). However, PaLTPG1 583
expression did not correlate with cuticle deposition at 80 DAFB. 584
The polyesterification of cutin monomers is not well understood. However, first 585
insight had suggested that CUS1 (CD1), a cutin synthase, is an extracellular 586
enzyme located in the developing cuticle and it is required for polymerization 587
42
(Yeats et al., 2012). We characterized a mango homolog of SlCUS1, MiCUS1 588
(MIN010966), with 73.96% of protein identity, which will support that this family 589
protein is conserved among land plants as proposed Yeats et al., (2014). This 590
gene showed high expression during initial stage (15 DAF), medium expression 591
during intermediate stages (45-141 DAF). Finally, a high expression during 592
ripening and storage time (147-153 DAF) corresponding with the major cuticle 593
accumulation. In tomato, Yeats et al., (2010) reported that SlCUS1 was 594
maximally expressed during the most rapid phase of fruit expansion and cuticle 595
accumulation, peaking at 15-20 DPA. In agreement with these results, Girard et 596
al., (2012) reported in tomato that the highest SlCUS1 expression was found at 597
20 (DPA) and the level of cutin monomers per surface unit decreased 598
proportionally with the reduction in SlCUS1 expression. 599
Another cutin synthase, GDSL-motif lipase/hydrolase (SGN-U583101), it was 600
reported to show a high expression in the epidermis (Yeats et al., 2010). We 601
characterized a mango homolog of SGN-U583101 and we named MiCUS2 602
(MIN031338), with 40.06% of protein identity. This gene showed a low gene 603
expression during initial and intermediate stages, increasing its expression at 604
147 DAF and reaching the maximum expression at 153 DAF during major 605
cuticle accumulation. Conversely, Yeats et al., (2010) reported that SGN-606
U583101 was maximally expressed at 5 DPA which correspondes with a very 607
early stage of fruit development. 608
PEL1 (AY987389) encodes a pectate lyase in mango fruit. We used the PEL1 609
expression as a control to observe the advance of the fruit ripening phenomena. 610
This is because this gene is only induced during mango ripening and indeed it 611
had been associated with mango softening (Chourasia et al., 2006). We 612
characterized a MiPEL1 (MIN009006), with 99.08% of protein identity. This gene 613
showed almost undetectable levels expression during initial and intermediates 614
stages of development (15-135 DAF). After, the level of expression increased by 615
141-147 DAF, reaching the maximum expression level at 153 DAF during 616
ripening in storage. MiPEL1 expression increased during fruit ripening which 617
further supports the validity of the gene expression quantification data. 618
43
The cuticle arquitecture characterization is a helpful tool for major understanding 619
of cuticle biosynthesis. Buda et al., (2009) proposed that the cuticle arquitecture 620
characterization to detect subtle changes in cuticle deposition and structural 621
organization are needed to elucidate the underlying molecular pathways of 622
cuticle biosynthesis. However, the cuticle arquitecture cannot typically be 623
associated with particular compounds or structural features (Martin and Rose, 624
2014). In addition, researches on fruit cuticle during storage are limited (Lara et 625
al., 2014). In this context, in order to determine changes in cuticle arquitecture 626
during ripening of mango fruit we analyzed cuticle structural through of light and 627
SEMs microscopy. Interestingly, our results showed an evident structural 628
variation in cuticle thickness and shape when comparing ripe and overripe fruits, 629
suggesting that the cuticle arquitecture could play a important role in the 630
ripening phenomena of fruit. However, as mentioned by Martin and Rose, 631
(2014), to date nothing is known about its functional significance, or the factors 632
that determine the various patterns of cuticular deposition. 633
On the other hand, we also analyzed the cuticle deposition during fruit ontogeny. 634
Our analysis showed that the mango cuticle deposition follows a biphasic 635
specific temporal pattern during fruit development, including a large 636
accumulation during fruit growth followed by a second phase of maximal cuticle 637
deposition during ripe and overripe. These cuticle accumulation pattern agree 638
well with previously reported data by Petit-Jiménez et al., (2007) for the same 639
mango cultivar. However, in this last experiment the maximum cuticular weight 640
was 1609 µg recorded at 135 days post anthesis (DPA). Also, in apple fruits it 641
have been reported that the cuticular wax deposition increase significantly 642
during ripening, which was associated with a burst in ethylene production (Ju 643
and Bramlage, 2001). 644
In contrast, cuticle deposition has been reported to ceases or decrease during 645
early stage of fruit development, before the initiation of the ripening phenomena 646
which results in lower amount and thickness of cuticle at later stages of 647
development (Lara et al., 2014). In agreement, grape berry showed a cuticle 648
decrease after harvest (Commenil et al., 1997). Moreover, in tomato, Yeats et 649
44
al., (2010) reported the maximal cuticle accumulation at 15 DPA. Thereafter, 650
cuticle accumulation decreased until reaching the red ripe stage. In sweet 651
cherry, Alkio et al., (2012) reported a cuticle decrease after 30 DAFB. 652
The results of several studies suggests that cuticle biosynthesis depends on 653
species and stage of development and it is influenced by genetics, physiological, 654
environmental factors and postharvest handling (Tafolla-Arellano et al., 2013), 655
which can explain the differences in the studies above metioned. 656
Based in the expression profile of cuticle-associated genes analyzed, their 657
homology with others genes of known function and the cuticle accumulation 658
pattern, we had proposed a model placing the cell organelle and the putative 659
role that the different genes are most likely playing in the molecular mechanism 660
of mango cuticle biosynthesis (Fig. 5). 661
The evidence of profile gene expression, and consequently, cuticle composition 662
and accumulation are particular in each species and it will be interesting to 663
elucidate which mechanisms are conserved despite the variation in cuticle 664
composition and architecture across species. 665
Conclusions 666
Our transcriptome of mango peel had increased the genomics resources for 667
future molecular research in mango fruit biology. 668
Our study provides a large amount of unigenes involved in several metabolic 669
processes of mango fruits such as lipids, cutin, secondary metabolism, cell wall 670
polysaccharides, among others. Interestingly, the functional RNA-Seq analysis 671
indicates that the cutin monomers biosynthesis pathway is enriched during 672
ripening. Our WGD analysis is the first report for mango and it identified a recent 673
WGD event. The cuticle deposition in mango follows a biphasic pattern during 674
fruit development, characterized by one phase of accumulation during the early 675
stages of fruit growth followed by a second phase of maximal cuticle deposition 676
during ripening and senescence. The analysis of the cuticle-related genes 677
expression provides the first insight to understand cuticle biosynthesis in mango 678
fruit and indicate a closely correlation with cuticle accumulation. However, 679
further studies are required to confirm the role of these genes during fruit 680
45
development, particularly enzymatic activity and changes in cuticle composition 681
among the different fruit developmental stages. The cuticle-associated genes 682
identified in this study will help in the elucidation of the molecular mechanism 683
underlying cuticle biosynthesis and in the design of strategies to increase the 684
postharvest shelf life of mango fruits. 685
Conflict of interest 686
The authors declare that they have no conflict interests. 687
Supplementary material 688
Supplementary Figure S1. Cuticle accumulation. 689
Supplementary Figure S2. Gene Ontology enrichment analysis. 690
Supplementary Table S1. PCR primers used for gene expression analysis 691
Supplementary Table S2. List of transcription factors differentially expressed. 692
Supplementary Table S3. List of protein kinases differentially expressed. 693
Supplementary Dataset S1. Whole-Genome Duplication Analysis. 694
Supplementary Dataset S2.GO slims, biochemical pathways, TF and protein 695
kinases analysis. 696
Supplementary Dataset S3. GO enrichment analysis. 697
Supplementary Dataset S4. Differentially expressed unigenes. 698
Supplementary Dataset S5. Metabolic pathway enrichment analysis 699
Acknowledgements 700
The author Julio César Tafolla Arellano thanks Mexican Council of Science and 701
Technology (CONACyT) for the scholarship assigned. This work was supported 702
by Project 20120 (P0045001): Aseguramiento De Calidad De Frutas Y 703
Hortalizas from Research Center for Food and Development A.C 704
The authors gratefully acknowledge M.Sc. Javier Ojeda for the help during 705
mango sampling and storage. We also thank Agricola Daniella for helping in the 706
experiment to obtain mangoes during fruit ontogeny for the study. Eric Fich and 707
Laetitia Martin for advice in cuticle microscopy. Dra. Iben Sørensen for helpful in 708
SEMs microscopy analysis. 709
Dra. Silvia Moya and Dr. Jesus Hernández for access to qRT-PCR equipment 710
facilities. Dr. Marisela Montalvo and Monica Resendiz for helpful advice during 711
46
the qRT-PCR analysis. 712
References 713
Aarts MG, Keijzer CJ, Stiekema WJ, Pereira A. 1995. Molecular 714
characterization of the CER1 gene of arabidopsis involved in epicuticular wax 715
biosynthesis and pollen fertility. The Plant Cell Online 7:2115-27. 716
Aharoni A, Dixit S, Jetter R, Thoenes E, van Arkel G, Pereira A. 2004. The 717
SHINE Clade of AP2 Domain Transcription Factors Activates Wax Biosynthesis, 718
Alters Cuticle Properties, and Confers Drought Tolerance when Overexpressed 719
in Arabidopsis. The Plant Cell 16:2463-80. 720
Al-Abdallat AM, Al-Debei HS, Ayad JY, Hasan S. 2014. Over-Expression of 721
SlSHN1 Gene Improves Drought Tolerance by Increasing Cuticular Wax 722
Accumulation in Tomato. International journal of molecular sciences 15:19499-723
515. 724
Albert Z, Ivanics B, Molnár A, Miskó A, Tóth M, Papp I. 2013. Candidate 725
genes of cuticle formation show characteristic expression in the fruit skin of 726
apple. Plant Growth Regulation 70:71-8. 727
Alkio M, Jonas U, Declercq M, Van Nocker S, Knoche M. 2014. 728
Transcriptional dynamics of the developing sweet cherry (Prunus avium L.) fruit: 729
sequencing, annotation and expression profiling of exocarp-associated genes. 730
Horticulture Research 1:1-15. 731
Alkio M, Jonas U, Sprink T, Van Nocker S, Knoche M. 2012. Identification of 732
putative candidate genes involved in cuticle formation in Prunus avium (sweet 733
cherry) fruit. Annals of Botany 110:101-12. 734
Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W, Lipman 735
DJ. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein 736
database search programs. Nucleic Acids Research 25:3389-402. 737
Anders S, Huber W. 2010. Differential expression analysis for sequence count 738
data. Genome biology 11:R106. 739
Azim MK, Khan I, Zhang Y. 2014. Characterization of mango (Mangifera indica 740
L.) transcriptome and chloroplast genome. Plant Molecular Biology 85:193-208. 741
47
Balbontín C, Ayala H, Rubilar J, Cote J, Figueroa CR. 2014. Transcriptional 742
analysis of cell wall and cuticle related genes during fruit development of two 743
sweet cherry cultivars with contrasting levels of cracking tolerance. Chilean 744
journal of agricultural research 74:162-9. 745
Bally ISE. 1999. Changes in the cuticular surface during the development of 746
mango (Mangifera indica L.) cv. Kensington Pride. Scientia Horticulturae 79:13-747
22. 748
Bernard A, Domergue F, Pascal S, Jetter R, Renne C, Faure J-D, Haslam 749
RP, Napier JA, Lessire R, Joubès J. 2012. Reconstitution of Plant Alkane 750
Biosynthesis in Yeast Demonstrates That Arabidopsis ECERIFERUM1 and 751
ECERIFERUM3 Are Core Components of a Very-Long-Chain Alkane Synthesis 752
Complex. The Plant Cell 24:3106-18. 753
Bird D, Beisson F, Brigham A, Shin J, Greer S, Jetter R, Kunst L, Wu 754
X,Yephremov A, Samuels L. 2007. Characterization of Arabidopsis 755
ABCG11/WBC11, an ATP binding cassette (ABC) transporter that is required for 756
cuticular lipid secretion. The Plant Journal 52:485-98. 757
Bolger AM, Lohse M, Usadel B. 2014. Trimmomatic: a flexible trimmer for 758
Illumina sequence data. Bioinformatics 30 (15): 2114-2120 759
Bourdenx B, Bernard A, Domergue F, et al. 2011. Overexpression of 760
Arabidopsis ECERIFERUM1 Promotes Wax Very-Long-Chain Alkane 761
Biosynthesis and Influences Plant Response to Biotic and Abiotic Stresses. 762
Plant Physiology 156:29-45. 763
Broun P, Poindexter P, Osborne E, Jiang C-Z, Riechmann JL. 2004. WIN1, a 764
transcriptional activator of epidermal wax accumulation in Arabidopsis. 765
Proceedings of the National Academy of Sciences of the United States of 766
America 101:4706-11. 767
Buda GJ, Isaacson T, Matas AJ, Paolillo DJ, Rose JKC. 2009. Three-768
dimensional imaging of plant cuticle architecture using confocal scanning laser 769
microscopy. The Plant Journal 60:378-85. 770
Carver TLW, Gurr SJ. 2006. Filamentous fungi on plant surfaces. In Biology of 771
the Plant Cuticle, ed. M Riederer, C Müller. Julius-von- Sachs-Institut, für 772
48
Biowissenschaften Universität Würzburg, Germany: Annual Plant Reviews. 773
23:368-92. 774
Chen X, Goodwin SM, Boroff VL, Liu X, Jenks MA. 2003. Cloning and 775
Characterization of the WAX2 Gene of Arabidopsis Involved in Cuticle 776
Membrane and Wax Production. The Plant Cell Online 15:1170-85. 777
Chourasia A, Sane VA, Nath P. 2006. Differential expression of pectate lyase 778
during ethylene-induced postharvest softening of mango (Mangifera indica var. 779
Dashehari). Physiologia Plantarum 128:546-55. 780
Commenil P, Brunet L, Audran J-C. 1997. The development of the grape berry 781
cuticle in relation to susceptibility to bunch rot disease. Journal of Experimental 782
Botany 48:1599-607. 783
Costaglioli P, Joubès J, Garcia C, Stef M, Arveiler B, Lessire R, Garbay B. 784
2005. Profiling candidate genes involved in wax biosynthesis in Arabidopsis 785
thaliana by microarray analysis. Biochimica et Biophysica Acta (BBA) - 786
Molecular and Cell Biology of Lipids 1734:247-58. 787
Dautt-Castro M, Ochoa-Leyva A, Contreras-Vergara CA, Pacheco-Sanchez 788
MA, Casas-Flores S, Sanchez-Flores A, Kuhn DN, Islas-Osuna MA. 2015. 789
Mango (Mangifera indica L.) cv. Kent fruit mesocarp de novo transcriptome 790
assembly identifies gene families important for ripening. Frontiers in Plant 791
Science 6:62. 792
DeBono A, Yeats TH, Rose JKC, Bird D, Jetter R, Kunst L, Samuels L. 793
2009. Arabidopsis LTPG Is a Glycosylphosphatidylinositol-Anchored Lipid 794
Transfer Protein Required for Export of Lipids to the Plant Surface. The Plant 795
Cell Online 21:1230-8. 796
Fiebig A, Mayfield JA, Miley NL, Chau S, Fischer RL, Preuss D. 2000. 797
Alterations in CER6, a Gene Identical to CUT1, Differentially Affect Long-Chain 798
Lipid Content on the Surface of Pollen and Stems. The Plant Cell Online 799
12:2001-8. 800
Girard AL., F. Mounet., M. Lemaire-Chamley., C. Gaillard., K. Elmorjani., J. 801
Vivancos., B. Bakan 2012. Tomato GDSL1 is required for cutin deposition in 802
the fruit cuticle. The Plant Cell 24(7):3119-3134. 803
49
Grabherr MG, Haas BJ, Yassour M, et al. 2011. Full-length transcriptome 804
assembly from RNA-Seq data without a reference genome. Nature 805
biotechnology 29:644-52. 806
Haslam TM, Mañas-Fernández A, Zhao L, Kunst L. 2012. Arabidopsis 807
ECERIFERUM2 Is a Component of the Fatty Acid Elongation Machinery 808
Required for Fatty Acid Extension to Exceptional Lengths. Plant Physiology 809
160:1164-74. 810
Heredia A. 2003. Biophysical and biochemical characteristics of cutin, a plant 811
barrier biopolymer. Biochimica et Biophysica Acta (BBA) - General Subjects 812
1620:1-7. 813
Hong-Bin S, Kuo-Chen C. 2007. "Signal-3L: a 3-layer approach for predicting 814
signal peptides", Biochemical and Biophysical Research Communications 363: 815
297-303. 816
Holloway PJ, Baker EA. 1968. Isolation of Plant Cuticles With Zinc Chloride-817
hydrochloric Acid Solution. Plant Physiology 43:1878-9. 818
Hu X, Zhang Z, Li W, Fu Z, Zhang S, Xu P. 2009. cDNA cloning and 819
expression analysis of a putative decarbonylase TaCer1 from wheat (Triticum 820
aestivum L.). Acta Physiol Plant 31:1111-8. 821
Isaacson T, Kosma DK, Matas AJ, et al. 2009. Cutin deficiency in the tomato 822
fruit cuticle consistently affects resistance to microbial infection and 823
biomechanical properties, but not transpirational water loss. The Plant Journal 824
60:363-77. 825
Jacobi KK, Gowanlock D. 1995. Ultrastructural Studies of `Kensington' Mango 826
(Mangifera indica Linn.) Heat Injuries. HortScience 30:102-3. 827
Jenks MA, Tuttle HA, Eigenbrode SD, Feldmann KA. 1995. Leaf Epicuticular 828
Waxes of the Eceriferum Mutants in Arabidopsis. Plant Physiology 108:369-77. 829
Joung J-G, Corbett AM, Fellman SM, Tieman DM, Klee HJ, Giovannoni JJ, 830
Fei Z. 2009. Plant MetGenMAP: An Integrative Analysis System for Plant 831
Systems Biology. Plant Physiology 151:1758-68. 832
Kader J-C. 1996. Lipid-transfer proteins in plants. Annual review of plant biology 833
47:627-54. 834
50
Kannangara R, Branigan C, Liu Y, Penfield T, Rao V, Mouille G, Höfte H, 835
Karp PD, Paley S, Romero P. 2002. The Pathway Tools software. 836
Bioinformatics 18:S225-S32. 837
Kim H, Lee SB, Kim HJ, Min MK, Hwang I, Suh MC. 2012. Characterization of 838
Glycosylphosphatidylinositol-Anchored Lipid Transfer Protein 2 (LTPG2) and 839
Overlapping Function between LTPG/LTPG1 and LTPG2 in Cuticular Wax 840
Export or Accumulation in Arabidopsis thaliana. Plant and Cell Physiology 841
53:1391-403. 842
Kimbara J, Yoshida M, Ito H, et al. 2013. Inhibition of CUTIN DEFICIENT 2 843
Causes Defects in Cuticle Function and Structure and Metabolite Changes in 844
Tomato Fruit. Plant and Cell Physiology 1-14. 845
Langmead B, Trapnell C, Pop M, Salzberg SL. 2009. Ultrafast and memory-846
efficient alignment of short DNA sequences to the human genome. Genome 847
Biology 10:R25. 848
Lara I, Belge B, Goulao LF. 2014. The fruit cuticle as a modulator of 849
postharvest quality. Postharvest Biology and Technology 87:103-112. 850
Lee S-B, Go YS, Bae H-J, Park JH, Cho SH, Cho HJ, Lee DS, Park OK, 851
Hwang I, Suh M-C. 2009b. Disruption of Glycosylphosphatidylinositol-Anchored 852
Lipid Transfer Protein Gene Altered Cuticular Lipid Composition, Increased 853
Plastoglobules, and Enhanced Susceptibility to Infection by the Fungal 854
Pathogen Alternaria brassicicola. Plant Physiology 150:42-54 855
Lee S-B, Jung S-J, Go Y-S, Kim H-U, Kim J-K, Cho H-J, Park OK, Suh M-C. 856
2009a. Two Arabidopsis 3-ketoacyl CoA synthase genes, KCS20 and 857
KCS2/DAISY, are functionally redundant in cuticular wax and root suberin 858
biosynthesis, but differentially controlled by osmotic stress. The Plant Journal 859
60:462-75. 860
Leide J, Hildebrandt U, Reussing K, Riederer M, Vogg G. 2007. The 861
Developmental Pattern of Tomato Fruit Wax Accumulation and Its Impact on 862
Cuticular Transpiration Barrier Properties: Effects of a Deficiency in a β-863
Ketoacyl-Coenzyme A Synthase (LeCER6). Plant Physiology 144:1667-79. 864
51
Livak KJ, Schmittgen TD. 2001. Analysis of Relative Gene Expression Data 865
Using Real-Time Quantitative PCR and the 2−ΔΔCT Method. Methods 25:402-8. 866
Luo B, Xue X-Y, Hu W-L, Wang L-J, Chen X-Y. 2007. An ABC Transporter 867
Gene of Arabidopsis thaliana, AtWBC11, is Involved in Cuticle Development and 868
Prevention of Organ Fusion. Plant and Cell Physiology 48:1790-802. 869
Luo C, He XH, Chen H, Hu Y, Ou SJ. 2013. Molecular cloning and expression 870
analysis of four actin genes (MiACT) from mango. Biologia Plantarum 57:238-871
44. 872
Luria N, Sela N, Yaari M, Feygenberg O, Kobiler I, Lers A, Prusky D. 2014. 873
De-novo assembly of mango fruit peel transcriptome reveals mechanisms of 874
mango response to hot water treatment. BMC Genomics 15:957. 875
Mandel T, Rogachev I, Venger I, Mintz-Oron S, Aharoni A, Adato A. 2007. 876
The ins and outs of tomato fruit peel metabolome. Proc. XII EUCARPIA 877
Symposium on Fruit Breeding and Genetics 465-74. 878
Martin LBB, Rose JKC. 2014. There’s more than one way to skin a fruit: 879
formation and functions of fruit cuticles. Journal of Experimental Botany 880
65:4639-51. 881
Matas AJ, Yeats TH, Buda GJ, et al. 2011. Tissue- and Cell-Type Specific 882
Transcriptome Profiling of Expanding Tomato Fruit Provides Insights into 883
Metabolic and Regulatory Specialization and Cuticle Formation. The Plant Cell 884
Online 23:3893-910. 885
McFarlane HE, Shin JJH, Bird DA, Samuels AL. 2010. Arabidopsis ABCG 886
Transporters, Which Are Required for Export of Diverse Cuticular Lipids, 887
Dimerize in Different Combinations. The Plant Cell 22:3066-75. 888
Millar AA, Clemens S, Zachgo S, Giblin EM, Taylor DC, Kunst L. 1999. 889
CUT1, an Arabidopsis Gene Required for Cuticular Wax Biosynthesis and 890
Pollen Fertility, Encodes a Very-Long-Chain Fatty Acid Condensing Enzyme. 891
The Plant Cell Online 11:825-38. 892
Mintz-Oron S, Mandel T, Rogachev I, et al. 2008. Gene Expression and 893
Metabolism in Tomato Fruit Surface Tissues. Plant Physiology 147:823-51. 894
52
Mukherjee S, Litz RE. 2009. Introduction: botany and importance. In The 895
mango: Botany, production and uses, ed. RE Litz, 23:1-18. 896
Müller C. 2006. Plant–insect interactions on cuticular surfaces. In Biology of the 897
Plant Cuticle, ed. M Riederer, C Müller. Julius-von- Sachs-Institut, für 898
Biowissenschaften Universität Würzburg, Germany: Annual Plant Reviews. 899
23:398-417. 900
Panikashvili D, Savaldi-Goldstein S, Mandel T, Yifhar T, Franke RB, Höfer 901
R,Schreiber L, Chory J, Aharoni A. 2007. The Arabidopsis 902
DESPERADO/AtWBC11 Transporter Is Required for Cutin and Wax Secretion. 903
Plant Physiology 145:1345-60. 904
Pauly M, Riechmann JL, Broun P. 2007. The Transcription Factor WIN1/SHN1 905
Regulates Cutin Biosynthesis in Arabidopsis thaliana. The Plant Cell Online 906
19:1278-94. 907
Petit-Jiménez D, Bringas Taddei E, González León A, García Robles J, 908
Báez Sañudo R. 2009. Effect of hydrothermal treatment on the ultrastructure of 909
cuticle of mango fruit. Revista Cientifica UDO Agricola 9:96-102. 910
Petit-Jiménez D, González-León A, González-Aguilar G, Sotelo-Mundo R, 911
Báez-Sañudo R. 2007. Cambios de la cutícula durante la ontogenia del fruto de 912
Mangifera indica l. Revista Fitotecnia Mexicana 30:51-60. 913
Pfündel EE, Agati G, Cerovic CG. 2006. Optical properties of plant surfaces. In 914
Biology of the Plant Cuticle, ed. M Riederer, C Müller. Julius-von- Sachs-Institut, 915
für Biowissenschaften Universität Würzburg, Germany: Annual Plant Reviews. 916
23:216-39. 917
Pighin JA, Zheng H, Balakshin LJ, Goodman IP, Western TL, Jetter R, 918
Kunst L, Samuels AL. 2004. Plant Cuticular Lipid Export Requires an ABC 919
Transporter. Science 306:702-4. 920
Pierleoni A, Martelli PL, Casadio R. 2008. -PredGPI: a GPI anchor predictor- 921
BMC Bioinformatics 9:392. 922
Plooy Wd, Merwe Cvd, Korsten L. 2004. Differences in the surface structures 923
of three mango cultivars and the effect of kaolin on these structures. Research 924
Journal - South African Mango Growers' Association 24:29-36. 925
53
Riederer M, Schreiber L. 2001. Protecting against water loss: analysis of the 926
barrier properties of plant cuticles. Journal of Experimental Botany 52:2023-32. 927
Rowland O, Lee R, Franke R, Schreiber L, Kunst L. 2007. The CER3 wax 928
biosynthetic gene from Arabidopsis thaliana is allelic to WAX2/YRE/FLP1. FEBS 929
Letters 581:3538-44. 930
Saladié M, Matas AJ, Isaacson T, et al. 2007. A Reevaluation of the Key 931
Factors That Influence Tomato Fruit Softening and Integrity. Plant Physiology 932
144:1012-28. 933
Samuels L, Kunst L, Jetter R. 2008. Sealing plant surfaces: cuticular wax 934
formation by epidermal cells. Plant Biology 59:683. 935
Shi JX, Adato A, Alkan N, He Y, et al. 2013. The tomato SlSHINE3 936
transcription factor regulates fruit cuticle formation and epidermal patterning. 937
New Phytologist 197:468-80. 938
Suh MC, Samuels AL, Jetter R, Kunst L, Pollard M, Ohlrogge J, Beisson F. 939
2005. Cuticular Lipid Composition, Surface Structure, and Gene Expression in 940
Arabidopsis Stem Epidermis. Plant Physiology 139:1649-65. 941
Tafolla-Arellano JC, González-León A, Tiznado-Hernández ME, Zacarías 942
García L, Báez-Sañudo R. 2013. Composición, fisiología y biosíntesis de la 943
cutícula en plantas. Revista Fitotecnia Mexicana 36:3-12. 944
Tharanathan RN, Yashoda HM, Prabha TN. 2006. Mango (Mangifera indica 945
L.), “The King of Fruits”—An Overview. Food Reviews International 22:95-123. 946
The Gene Ontology Consortium. 2015. Gene Ontology Consortium: going 947
forward. Nucleic Acids Research 43:D1049–D1056. 948
Thornton B, Basu C. 2011. Real-time PCR (qPCR) primer design using free 949
online software. Biochemistry and Molecular Biology Education 39:145-54. 950
Vazquez‐Salinas C, Lakshminarayana S. 1985. Compositional changes in 951
mango fruit during ripening at different storage temperatures. Journal of Food 952
Science 50:1646-8. 953
Vogg G, Fischer S, Leide J, Emmanuel E, Jetter R, Levy AA, Riederer M. 954
2004. Tomato fruit cuticular waxes and their effects on transpiration barrier 955
54
properties: functional characterization of a mutant deficient in a very‐long‐chain 956
fatty acid β‐ketoacyl‐CoA synthase. Journal of Experimental Botany 55:1401-10. 957
Wan CY, Wilkins TA. 1994. A Modified Hot Borate Method Significantly 958
Enhances the Yield of High-Quality RNA from Cotton (Gossypium hirsutum L.). 959
Analytical Biochemistry 223:7-12. 960
Wu H-x, Jia H-m, Ma X-w, Wang S-b, Yao Q-s, Xu W-t, Zhou Y-g, Gao Z-s, 961
Zhan R-l. 2014. Transcriptome and proteomic analysis of mango (Mangifera 962
indica Linn) fruits. Journal of Proteomics 105:19-30. 963
Xia Y, Nikolau BJ, Schnable PS. 1996. Cloning and characterization of CER2, 964
an Arabidopsis gene that affects cuticular wax accumulation. The Plant Cell 965
Online 8:1291-304. 966
Yang Z. 1997. PAML: a program package for phylogenetic analysis by 967
maximum likelihood. Computer applications in the biosciences 13:555-6. 968
Yeats TH, Howe KJ, Matas AJ, Buda GJ, Thannhauser TW, Rose JKC. 969
2010. Mining the surface proteome of tomato (Solanum lycopersicum) fruit for 970
proteins associated with cuticle biogenesis. Journal of Experimental Botany 971
61:3759-71. 972
Yeats TH, Martin LB, Viart HM, et al. 2012. The identification of cutin synthase: 973
formation of the plant polyester cutin. Nature chemical biology 8:609-11. 974
Yeats TH, Rose JK. 2008. The biochemistry and biology of extracellular plant 975
lipid‐transfer proteins (LTPs). Protein Science 17:191-8. 976
Yeats TH, Rose JKC. 2013. The Formation and Function of Plant Cuticles. 977
Plant Physiology 163:5-20. 978
Yeats TH, Huang W, Chatterjee S, Viart HMF, Clausen MH, Stark RE, Rose 979
JK. 2014. Tomato Cutin Deficient 1 (CD1) and putative orthologs comprise an 980
ancient family of cutin synthase‐like (CUS) proteins that are conserved among 981
land plants. The Plant Journal 77(5): 667-675. 982
Zheng Y, Zhao L, Gao J, Fei Z. 2011. iAssembler: a package for de novo 983
assembly of Roche-454/Sanger transcriptome sequences. BMC Bioinformatics 984
12:453. 985
55
Zhong S, Joung JG, Zheng Y, Chen YR, Liu B, Shao Y, Xiang YZ, Fei Z, 986
Giovannoni JJ. 2011. High-Throughput Illumina Strand-Specific RNA 987
Sequencing Library Preparation. Cold Spring Harbor Protocols 2011:940-9. 988
doi:10.1101/pdb.prot5652. 989
Ju Z, Bramlage WJ. 2001. Developmental changes of cuticular constituents 990
and their association with ethylene during fruit ripening in ‘Delicious’ apples. 991
Postharvest biology and technology 21(3): 257-263. 992
Tables 993
Table 1. Characterization of cuticle-associated genes in mango. 994
Table 2. Mango transcriptome summary. 995
Figures legends 996
Fig. 1. Fruit cuticle structural diversity. Mangoes used for RNA-Seq, ripe (A) 997
and overripe (B) mango fruit scale bar=1 inch. Cuticle stained with Oil red in ripe 998
(C) and overripe (D) mango fruit. Scale bar=20 µm SEMs images of cuticle Oil 999
Red Ripe (E), Overripe (F). Scale bar=50 µm. 1000
Fig. 2. Mango RNA-Seq assembly annotation. A, Length distribution of mango 1001
unigenes. B. Unigenes distribution of the 30,003 assembled groups. C, 1002
Summary of mango unigenes annotation. The numbers indicate the annotated 1003
unigenes. D, Mango whole genome duplication analysis (WGD). Distribution of 1004
synonymous nucleotide substitution (Ks) rates between homologous gene pairs 1005
within mango (light blue), sweet orange (green) and between mango and 1006
Arabidopsis (orange), mango and orange (grey), and orange and papaya (dark 1007
blue). 1008
Fig. 3. Differential expression analysis. A. Heat-map of 5349 differentially 1009
expressed unigenes. The roman number in the right side and the left column 1010
colors indicate independent clusters and the color key indicates the log2 fold 1011
change of Overripe vs Ripe samples. B. Graph shows the pathways enriched in 1012
the overripe vs ripe comparison. Right and left side indicates the metabolic 1013
pathways enriched in the up-regulated and down-regulated unigenes, 1014
respectively. Asterisks indicate the statistical significance level, corrected p-1015
value ≤0.001 (***), corrected p-value ≤0.01 (**), corrected p-value ≤0.05 (*). C. 1016
56
Graph shows the expression ratio (Log2) obtained by qRT-PCR and RNA-Seq 1017
of 15 selected genes involved in cuticle biosynthesis. D. Graph shows the 1018
Pearson Correlation value between the gene expression ratios obtained from 1019
RNA-Seq data and qRT-PCR. 1020
Fig. 4. Profile expression of cuticle-associated genes and cuticle 1021
accumulation during mango fruit ontogeny. Time course expression of 1022
selected genes during fruit growth and ripening. The two phases of cuticle 1023
deposition are indicated above the fruit development stages considered. 1024
Fig. 5. Cuticle biosynthesis model proposed for mango fruit. Biosynthesis of 1025
fatty acids are synthesized in the plastids (1) and modified to cutin monomers 1026
and wax constituents in the endoplasmatic reticulum by a fatty acid elongases 1027
complex (2). After, these monomers are transported through the plasma 1028
membrane (3) and the cell wall (4). Also, it is shown the cuticle polymerization in 1029
the developing cuticle (5). Genes are described in the text. For further 1030
explanations see text. The cuticle roles are: (A) Controlling the water loss and 1031
gas diffusion; (B) Prevents water and dust accumulation; (C) Protection against 1032
insects; (D) Pathogens attack; (E) Signal transduction; (F) Thermoregulatory 1033
role and protection of UV radiation; and (G) Mechanical support. 1034
1035
Fig.1. Fruit cuticle structural diversity. 1036
1037
57
1038 1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
Fig. 2. Mango RNA-Seq assembly annotation. 1050
58
1051 1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
59
1064
Fig. 3. Differential expression analysis. 1065
1066
1067 1068
1069
1070
1071
1072
1073
60
1074
1075
1076
Fig. 4. Profile expression of cuticle-associated genes and cuticle 1077
accumulation during mango fruit ontogeny. 1078
1079
1080 1081
1082
1083
1084
1085
1086
1087
61
1088
1089
1090
1091
1092
Fig. 5. Cuticle biosynthesis model proposed for mango fruit. 1093
1094
1095 1096
1097
1098
1099
62
1100
1101
1102
1103
1104
1105
1106
1107
Table 1. Characterization of cuticle-associated genes in mango. 1108 1109
1110
1111 Mango genes analyzed in this study were named after the most similar Tomato and Arabidopsis genes and including a 1112 prefix “Mi” for Mangifera indica. 1113 The raw sequencing reads were deposited in NCBI Sequence Read Archive (SRA) under the accession number 1114 SRP043494. 1115 *MiKCS6 no RT-PCR products were generated in the RNA-Seq validation. 1116 **MiCUS2 was named according homolog tomato protein identity. 1117 LTPs were numbered according homolog tomato protein identity. 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131
Gene ID
Mi ID Best hit ID Protein identity (%)
Gene product Role Reference
MiSHN1 MIN047952 AT1G15360.1 67.68 AP2/EREBP-type transcription factor
SHINE 1
Transcription Factor
Aharoni et al., 2004; Broun et al., 2004; Kannangara et al., 2007
MiCD2 MIN074277 Solyc01g091630.2 86 Homeobox-leucine zipper protein ATHB-9
Transcription Factor
Isaacson et al., 2009
MiCER1 MIN107433 AT1G02205.2 63.71 Aldehyde decarbonylase
VLCFA elongation
Aarts et al., 1995; Bourdenx et al., 2011; Bernard et al., 2012
MiCER2 MIN052433 AT4G24510.1 42.65 BAHD acyltransferase VLCFA elongation
Xia et al., 1996; Haslam et al., 2012
MiCER3 MIN064126 AT5G57800.1 69.73 Fatty acid reductase VLCFA elongation
Jenks et al., 1995; Chen et al., 2003; Rowland et al., 2007
MiKCS2 MIN101804 AT1G04220.1 78.25 3-Ketoacyl-COA Synthase 2
VLCFA elongation
Lee et al., 2009a
MiKCS6* MIN040156 AT1G68530.1 84.88 3-Ketoacyl -COA Synthase 6
VLCFA elongation
Millar et al., 1999; Fiebig et al., 2000; Vogg et al., 2004; Leide et al.,
2007 MiWBC11 MIN106958 AT1G17840.1 84.78 ABC Transporter
White-Brown Complex Homolog Protein 11
Transport of wax Bird et al., 2007; Panikashvili et al., 2007; Luo et al., 2007
MiLTP1 MIN026365
SGN-U579687
54.39 Lipid transfer protein
Wax deposition Yeats et al., 2010
MiLTP2 MIN018326
SGN-U579687
47.79 Lipid transfer protein
Wax deposition Yeats et al., 2010
MiLTP3 MIN107167
SGN-U579687
36.84
Lipid transfer protein
Wax deposition Yeats et al., 2010
MiLTPG1 MIN012243
AT1G27950.1
48.68 Lipid transfer protein
Wax deposition DeBono et al., 2009, Lee et al., 2009
MiCUS1 MIN010966 SGN-U585129 73.96 Cutin synthase Cuticle
polymerization
Yeats et al., 2010; Yeats et al., 2012; Yeats et al., 2014
MiCUS2** MIN031338 SGN-U583101 40.06 Cutin synthase Cuticle polymerization
Yeats et al., 2010
MiPEL1 MIN009006
AY987389
99.08 Pectate lyase
Control for ripening
Chourasia et al. 2006
63
1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 Table 2. Summary of mango transcriptomes studies. We compared our 1152 transcriptome analysis with previous studies directed to different tissues. 1153 1154
In this study
Azim et al., 2014
Wu et al., 2014
Luria et al., 2014
Castro et al., 2015 Tissue Analyzed Peel Leaf Pericarp and pulp Peel Mesocarp Mango Cultivar Keitt Langra Zill Shelly Kent
Total sequenced bases 62.5 Gb >1 Gb 6.1 Gb 8.6 GB 4.8 Gb Total unigenes 107,744 30,509 54,207 57,544 52,948
Total assembled bases 184,977,733 16,354,267 - - - Average length unigenes 1,717 bp 536 838 863.3 836
Largest unigenes 12,271 bp - - - 8,713 N50 2,235 bp 687 1,328 1,598 1456
Sequencing system HiSeq 2500 HiSeq 2000 HiSeq 2000 HiSeq 2000 Genome Analyzer GAIIx II Sequence Read Archive
Accession Number SRP043494 SRR947746 SRP035450 SRX375390 SRP045880 Swiss-Prot 68,649 (63.7%) 14,447 (47.5%) 26,380 (48.67%) - 25,154
TrEMBL 91,736 (85.1%) - - - - Arabidopsis Database 88,242 (81.9%) - - - -
NCBI non-redundant (NR) - 24,593 (80 %) 42,515 (78.43%) 35,719 (62.07%) 32,560 GO terms 79,208 (73.51%) 21,054 (69%) 35,198 (64.93%) 28,317 (49.2%) 29,844
Unigenes/Pathways 17,686/461 Pathway Tools 13,561/293
KEGG pathways 23,741/128 KEGG pathways - 7,458/327
KEGG pathways Transcription
Factor/Families 5,432/55 - - - - Protein kinases/Families
3,662/76
- - - -
1155 1156 1157
1158
1159
1160
1161
1162
1163
64
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
Supplementary Figure S1. Cuticle accumulation. 1176
1177
1178
65
Supplementary Figure S2. Gene Ontology enrichment analysis. (A)Top ten 1179
of GO terms enriched in the up-regulated unigenes. Blue bars indicate the 1180
biological process enriched, orange bars indicate the cellular components 1181
enriched and green bars indicate the molecular functions enriched. Asterisks 1182
indicate the statistical significance level, corrected p-value≤0.001 (***), corrected 1183
p-value ≤0.01 (**), corrected p-value≤0.05 (*). 1184
1185
1186 1187
1188
1189
1190
1191
1192
1193
1194
1195
66
1196
1197
Supplementary Figure S2. Gene Ontology enrichment analysis. (B) Top ten 1198
of GO terms enriched in the down-regulated unigenes. Blue bars indicate the 1199
biological process enriched, orange bars indicate the cellular components 1200
enriched and green bars indicate the molecular functions enriched. Asterisks 1201
indicate the statistical significance level, corrected p-value≤0.001 (***), corrected 1202
p-value ≤0.01 (**), corrected p-value≤0.05 (*). 1203
1204
1205 1206
1207
1208
1209
1210
1211
1212
67
1213
1214
1215
1216
1217
Supplementary Table S1. DNA sequence of the primers used for the analysis 1218
of gene expresion by real time quantitative reverse transcription PCR. 1219
Gene ID Forward primer 5´à3´sequence Reverse primer 5´à3´ sequence Amplicon lenght from cDNA (bp)
MiSHN1 GGCTCTTGGGTCTCTGAG CCTCTTCAGCCGTCTCAA 79
MiCUS1 GACAAGGACCCTACAATGGAATTGG GATCTGTTGTACGATAACTCTGCCG 131
MiCD2 TAATGGACCCACAAACGGAAACAAT TAGCTGTAGGAAGACTGTTCACCAA 110
MiCUS2 TCTACTGGACAGTCCTTGTGTTTCA AGTTGTTGTTTCCCACATCAACCAA 111
MiCER1 GATTGTTTCTACCACTTAACACC CACCCTTCTTGGAAGCCAATTC 90
MiCER2 GGAGGAAGAAAGTGAAAG TTCAACCCATAAACATCCG 90
MiCER3 GAGGAGCCAAGAATTGAAT GCATGTTGCTGTAGGAGTT 97
MiKCS2 GAATCTGGAGCTGAGTGA CGATCACCTCTCTTTATCCT 135
MiKCS6 TCTTCTTCGTCCCTCTGGTA CGTCAACTGGTGTCTTGA 154
MiWBC11 GAGATAGAGACGAGCAAG CTCCCACAAGTTCTGTATTAG 106
MiLTP1 CATCCATCTCAGGCATCAACTA ATGGGCTGATCTTGTAAGGG 84
MiLTP2 GGCATTCATAGCTGTGCT GGTCACTTGTTCGCATGTTAT 82
MiLTP3 TGCAAAATGCAGCTAAAGGA GTTGGTGGAGGTGCTGATCT 107
MiLTPG1 CTTCCCACTGCCTGTCAAAT GAAGATAGCCGCATCTGGAG 93
MiPEL1 ATGGCGGTTTCTCCTAGA TCACTGTCGATGCTTTAACG 85
MiActin1 CGTTCTGTCCCTCTATGCCA AGATCACGGCCAGCAAGATC 141
CAPÍTULO III
Gene expression of a putative
glycosylphosphatidylinositol-anchored lipid transfer protein 2 during cuticle biosynthesis in mango.
Tafolla-Arellano JC, Ojeda-Contreras AJ, Báez-Sañudo R, Tiznado-Hernández ME.
Enviado a Revista Fitotecnia Mexicana.
69
GENE EXPRESSION OF A PUTATIVE
GLYCOSYLPHOSPHATIDYLINOSITOL-ANCHORED LIPID
TRANSFER PROTEIN 2 DURING CUTICLE BIOSYNTHESIS
IN
MANGO
GENE EXPRESSION OF A PUTATIVE MiLTPG2
EXPRESIÓN GÉNICA DE UNA PROTEÍNA PUTATIVA DE
TRANSFERENCIA DE LÍPIDOS 2 ANCLADA A
GLICOSILFOSFATIDILINOSITOL DURANTE LA
BIOSÍNTESIS DE CUTÍCULA EN MANGO
Julio C. Tafolla-Arellano, Angel J. Ojeda-Contreras, Reginaldo Báez-
Sañudo, Martín E. Tiznado-Hernández*
Laboratorio de Fisiología y Biología Molecular de Plantas. Coordinación de
Tecnología de Alimentos de Origen Vegetal, Centro de Investigación en
Alimentación y Desarrollo, A. C. Km 0.6 carretera a la Victoria, C.P. 83304,
Hermosillo, Sonora, México.
* To whom correspondence should be addressed: e-mail: tiznado@ciad.mx
70
Abstract
Mango fruit (Mangifera indica L.) is highly perishable, mainly due to
desiccation, which increases the postharvest losses. The cuticle plays a role
controlling the water loss and gas diffusion, among other functions. Cuticle
biosynthesis requires the transport of lipids from epidermal cells through the
plant cell wall, function carried out by lipid transfer proteins (LTPs). Recently, it
was reported a glycosylphosphatidylinositol-anchored lipid transfer protein 2
(LTPG2) in Arabidopsis, and experimentally demonstrated to be involved in
lipids transport during cuticle biosynthesis. The objective of this work was to
characterize LTPG2 of mango (MiLTPG2) and correlate its expression with the
pattern of biosynthesis and cuticle accumulation during mango fruit ontogeny.
Mango flowers were tagged and fruits sampled every 15 days after flowering
(DAF) until ripe. After that, mangoes were stored during 18 days with sampling
every 6 days. mRNA was isolated from mango peel and utilized to quantify gene
expression by real-time quantitative reverse transcription PCR. MiLTPG2
contains the three different domains characteristic of the LTPG proteins: a signal
peptide domain, a lipid transfer domain and a transmembrane domain. The lipid
transfer domain contains the characteristic eight highly conserved cysteine
residues. The cuticle accumulation showed a biphasic pattern, characterized by
an accumulation during fruit growth followed by a second phase of maximal
cuticle deposition during ripening. The MiLTPG2 gene showed a 7.48 fold
change increase in expression during late stages of cuticle biosynthesis (153
DAF) as compared with 15 DAF. This increase correlated with the highest
amount of cuticle accumulation (2100 µg/cm2) observed in this same stage. A
71
model is proposed describing the putative role of the MiLTPG2 gene in the
molecular mechanism of cuticle biosynthesis in mango fruit. This study provides
the first insight to understand the putative role of MiLTPG2 gene in cuticle
biosynthesis in mango fruit.
Keywords: Mango, Ontogeny, Cuticle, Gene Expression, MiLTPG2.
Resumen
La fruta de mango es altamente perecedera principalmente debido a la
deshidratación, lo cual incrementa pérdidas postcosecha. La cutícula tiene la
función de controlar la pérdida de agua y difusión de gases, entre otras
funciones. La biosíntesis de cutícula requiere del transporte de lípidos desde las
células epidérmicas a través de la pared celular, función que realizan las
proteínas de transferencia de lípidos (LTPs). Recientemente, en Arabidopsis se
reportó una proteína de transferencia de lípidos 2 anclada a un dominio
glicosilfosfatidilinositol (LTPG2), y se demostró experimentalmente que está
involucrada en el transporte de lípidos durante la biosíntesis de cutícula. El
objetivo de este trabajo fue caracterizar el gen LTPG2 en mango (MiLTPG2) y
correlacionar su expresión con los patrones de síntesis y acumulación de
cutícula durante la ontogenia del fruto de mango. Se marcaron las flores de
mango y los frutos fueron muestreados cada 15 días después de floración
(DAF). Posteriormente, los mangos fueron almacenados durante 18 días con
muestreos cada 6 días. Se obtuvo mRNA de exocarpo de mango y se cuantificó
la expresión mediante PCR cuantitativa en tiempo real. MiLTPG2 contiene los
tres dominios característicos de las proteínas LTPG: un dominio péptido señal,
un dominio de proteína de transferencia de lípidos y un dominio
72
transmembrana. El dominio de proteína de transferencia de lípidos contiene los
característicos ocho residuos de cisteína altamente conservados. La
acumulación de cutícula mostró un patrón bifásico, caracterizado por una
acumulación durante el crecimiento del fruto, seguido de una segunda fase
caracterizada por una gran deposición de cutícula durante la maduración.
MiLTPG2 mostró un incremento en su expresión 7.8 veces durante las etapas
tardías de biosíntesis de cutícula (153 DAF) comparado con 15 DAF. Este
incremento correlaciona con el elevado incremento en la acumulación de
cutícula (2100 µg/cm2) observado en esta misma etapa. Proponemos un modelo
en el cual se describe la posible función del gen MiLTPG2 en el mecanismo
molecular de biosíntesis de cutícula en mango. El presente estudio es el primer
esfuerzo que se realiza para elucidar la posible función del gen MiLTPG2 en la
biosíntesis de la cutícula en frutos de mango.
Palabras clave: Mango, Ontogenia, Cutícula, Expresión génica, MiLTPG2.
Introduction
Mango fruit (Mangifera Indica L.) is an ideal model system to study cuticle
biosynthesis because it is a highly perishable fleshy fruit, tropical, economical
crop in the world and contains a large cuticular mass (Petit-Jiménez et al. 2007).
However, the scarce genomics resources are the principal limitant to carry out
research at molecular level in the case of many biological processes in mango.
The fruit exocarp provides mechanical strength and it appears to play an
important role in the shelf life. Furthermore, it is composed of the cuticle, the
single-cell epidermal layer and several collenchymatous cell layers (Mintz-Oron
et al. 2008). The cuticle is a hydrophobic layer that acts as an important surface
73
barrier between the fruit and its environment, composed mostly of cutin and wax
lipids (Kunts and Samuels, 2003). The principal function of cuticle to is control
the water loss and gas diffusion (Riederer and Schreiber, 2001). Besides, cuticle
plays an important role in fruit quality and postharvest shelf life by controlling
desiccation, microbial infection and physiological disorders (Martin and Rose,
2014; Tafolla-Arellano et al. 2013). Therefore, the knowledge regarding the
cuticle biosynthesis is fundamental for the improvement of fruit quality.
During cuticle biosynthesis, after the different wax components have been
exported from the epidermal cell, they must cross the hydrophilic cell wall to reach
the cuticle, where the cell wall polysaccharides, may represent a physical barrier
to the transport of cuticular wax. The phenomenon less understood is the
transport mechanism throughout of the cell wall. Lipid transfer proteins (LTPs)
have been proposed to be involved in cuticular transport across the cell wall,
because they are abundantly expressed in the epidermis, are secreted into the
apoplast and are small enough to transverse the cell wall (Kader, 1996). LTPs are
characterized by a high pI (~9) and a characteristic eight- cysteine motif: C. . .C. .
.CC. . .CXC. . .C. . .C (Yeats and Rose, 2008), which is essential for the lipid-
binding cavity formation (Kader, 1996).
The LTPs had been proposed to play a role in cuticle biosynthesis in different
species, e.g., carrot (Sterk, 1991), broccoli (Pyee and Kolattukudy, 1994), tobacco
(Cameron et al. 2006), etc. Yeats et al. (2010) reported changes in the regulation
of genes encoding LTPs correlating with the cuticle biosynthesis phenomena
during tomato fruit development. However, Borner et al. (2002) identified in
Arabidopsis 18 glycosylphosphatidylinositol (GPI)-anchored lipid-transfer proteins
74
known as LTPG. These proteins are composed of a range between 140 to 204
amino acids and therefore are somewhat larger than LTPs, which are less than
120 amino acids. The Cys residues are also conserved in the 18 LTPGs,
suggesting that they have a similar fold as the LTPs and an extracellular location.
Also, Gene Ontology analyses strongly suggest that the LTPGs play to role in the
synthesis or deposition of cuticular waxes (Edstam et al. 2013).
Kim et al. (2012) showed that the protein that encodes AtLTPG2 was
targeted to the plasma membrane via the vesicular trafficking system and mainly
expressed in stem epidermal peels. In the ltpg2 mutant the composition of the
cuticular wax was significantly altered in the stems and siliques. Also, it was
recorded a reduction in the amount of the C29 alkane, which is the major
component of cuticular waxes. Besides, it was reported that LTPG2 is
functionally overlapped with LTPG/LTPG1 during cuticular wax export (Kim et al.
2012).
Based of the above mentioned the objective of the present study was to
characterize LTPG2 of mango (MiLTPG2) and correlate its expression with the
pattern of biosynthesis and cuticle accumulation during mango fruit ontogeny
Methods
Plant materials
Mangoes cv. `Keitt´ used for qRT-PCR analysis were hand harvested every
15 days after flowering (DAF) until ripening at a commercial orchard and
packinghouse located at El Porvenir, Ahome, Sinaloa, Mexico
(www.agricoladaniella.com.mx). Also, mangoes at fully mature-green stage were
stored at 20°C and 60-65% relative humidity during 18 days and sampled every
75
six days. In each harvesting point, samples of mango peel tissue were isolated
and immediately frozen with liquid nitrogen and stored at -80 oC for qRT-PCR
analysis.
RNA isolation and cDNA synthesis for developmental time course
Peel tissue samples from three fruits were pooled together to create one
biological replicate and the extraction was carried out in three independent
biological replicates for each mango developmental stage. Total RNA was
extracted from frozen mango fruit peel using hot borate method with minor
modifications (Wan and Wilkins, 1994), 2 µg of total DNase-treated RNA was
used for first strand cDNA synthesis using SuperScript II reverse transcriptase
and oligo(dT) primers (Invitrogen. Carlsbad, CA 92008. USA), according to the
manufacturer’s instructions.
Real-time quantitative reverse transcription PCR
MiLTPG2 sequence was obtained from the transcriptomic analysis data
generated from mango peel (Tafolla-Arellano et al. 2014) using Basic Local
Alignment Search Tool (Altschul et al. 1997). The coding DNA sequence (CDS)
and amino acid sequence deduced were obtained using Open Reading Frame
Finder (http://www.ncbi.nlm.nih.gov/gorf/gorf.html), Signal-3L for predicted signal
peptide (Hong-Bin and Kuo-Chen, 2007), PredGPI predictor for GPI (Pierleoni et
al., 2008) and ProtParam tool for amino acid protein composition (Gasteiger et
al. 2005).
Quantitative PCR experiments were performed using a StepOne™ Real-
Time PCR System (Applied Biosystems, Foster City, CA, USA). The primers
utilized were designed according to the protocol described in Thornton and Basu
76
(2011). The sequence of the MiLTPG2 primer forward and reverse are: 5'à3'
TTAAGCTCCCTTCTGTCTGC and 5'à3' GCAGCTAAACCAGGCACG,
respectively.
The cDNA samples were diluted 5-fold with water and 1 µl was used as a
template for each 20 µl quantitative PCR, prepared using HotStart-IT SYBR
Green qPCR Master Mix (2X) (Affymetrix, Santa Clara, CA, USA) in biological
triplicates. The thermal cycling conditions were as follows: 2 min at 95°C,
followed by 40 cycles at 95°C for 15 s and 60°C for 1 min. Specificity of the PCR
products was determined by high-resolution melt curve analysis, gel
electrophoresis and by DNA sequencing carried out at Macrogen (Macrogen
Inc., Seoul, South Korea).
Statistical analysis of relative expression results was carried out using the
relative expression method (Livak and Schmittgen, 2001), with MiActin1
(GenBank accession Number JF737036) as the reference gene. The sequences
for amplification of MiACtin1 gene forward and reverse are 5'à3'
CGTTCTGTCCCTCTATGCCA and 5'à3' AGATCACGGCCAGCAAGATC,
respectively. For time course experiments, fold change of MiLTPG2 were
calculated.
Isolation of mango fruit cuticle
The changes in cuticle deposition amount during mango ontogeny were
quantified following essentially a previously described protocol (Petit-Jiménez et
al. 2007).
77
Results and Discussion
Characterization of MiLTPG2 gene encoding GPI-anchored LTP
There is no published data about the role of LTPG2 like genes in cuticular
wax transport, with the exception of Arabidopsis. Therefore, we will compare our
findings with the information available about in the Arabidopsis gene (AtLTPG2). It
is important to mention that this is the first report of a LTPG2 gene isolated from a
fleshy fruit.
MiLTPG2 (MIN029156) ortholog to AtLTPG2 (AT3G43720) showed a 43.20%
of protein identity. Furthermore, MiLTPG2 was predicted to have a composition of
190 amino acids (Figure 1A), with a molecular weight of 18,951 Da and a
theoretical isoelectric point of 4.52. Conversely, LTPs are characterized for having
an isoelectric point ~9 (Yeats and Rose, 2008). Also, this protein does not contain
any tryptophan residues in agreement with the amino acid composition of LTPs
previously reported (Kader, 1996).
MiLTPG2 contains the three different domains characteristic of the LTPG
proteins: a signal peptide domain, a lipid transfer domain and a transmembrane
domain (Fig. 1B). Analysis with the Signal-3L engine showed that the signal
peptide domain is encoded by the first 26 amino acids. The lipid transfer domain
contains the characteristic eight highly conserved cysteine residues at positions
59, 69, 86, 87, 100, 101, 126 and 136, which agrees with the characteristic eight-
cysteine motif (C. . .C. . .CC. . .CXC. . .C. . .C) reported in LTPs (Yeats and Rose,
2008). Cysteine residues contribute to the four disulfide bonds, which are known
to be essential for the formation of the lipid-binding cavity (Kader, 1996). Finally,
the C-terminal end (168-189 amino acid residues) was predicted with highly
78
probability to encode a GPI transmembrane domain in serine residue located at
position 168 of the omega site with a specificity of 100% (Figure 1C). Similar
results were reported by Kim et al. (2012) for AtLTPG2, which was predicted to
have a signal peptide, required for protein secretion, composed of 18 amino acid
residues located within the positions 5–22 at the N-terminal end. Also, this protein
showed the characteristic eight highly conserved cysteine residues at positions
38, 48, 67, 68, 81, 83, 110 and 120. Finally, it was identified a GPI
transmembrane domain in serine residue at position 170 of the omega site.
Studies in Arabidopsis showed that the AtLTPG1 and AtLTPG2 are targeted
to the plasma membrane via the vesicular trafficking system (Lee et al. 2009,
DeBono et al. 2009, Kim et al. 2012). Based in GPI domain identified and those
findings, we had hypothesized that MiLTPG2 is also localized to the plasma
membrane, although, further experimental evidences are needed to probe this
statement.
MiLTPG2 expression and cuticle accumulation
We evaluated the MiLTPG2 gene expression by real time quantitative
reverse transcription PCR (qRT-PCR) during mango fruit ontogeny. The qRT-PCR
products showed the predicted sizes after separation by agarose gels
electrophoresis. Further, the DNA fragment sequence is the expected nucleotide
compositions of the MiLTPG2 amplicon (Figure 2).
The relative expression of MiLTPG2 from 30 to 140 DAF was less a 1-fold
change as compared with 15 DAF. Then, showed a slight increase of 1.31 fold
change at 147 DAF, and finally, MiLTPG2 expression showed a high increase of
7.48 fold change at 153 DAF as compared with 15 DAF.
79
The initial cuticular weight at 15 DAF was 889 µg/cm2 and continued
increasing until 105 DAF reaching 1763 µg/cm2. Thereafter, it showed a decrease
to 1632 µg/cm2 at 120 DAF. After that, cuticular weight showed an increase of
1846 and 1920 µg/cm2 from 135 to 141 DAF, respectively. Moreover, a slight
decrease to 1827 µg/cm2 was recorded at 147 DAF. Finally, the cuticle
accumulation showed a large increase by the end of the storage time to reach
2100 µg/cm2 at 153 DAF, which is the maximum cuticular weight registered during
mango fruit ontogeny (Figure 3). As it can be observed, the cuticle deposition
follows a biphasic specific temporal pattern during fruit development, including a
large accumulation during fruit growth followed by a second phase of maximal
cuticle deposition during ripening.
These cuticle accumulation patterns agree with previously reported data by
Petit-Jiménez et al., (2007) for the same mango cultivar. However, in this last
experiment the maximum cuticular weight was 1609 µg recorded at 135 days post
anthesis (DPA). Many studies had been reported that the cuticle biosynthesis is
influenced by genetics, physiological, environmental factors and postharvest
handling (Tafolla-Arellano et al. 2013), which can explain the differences in both
studies.
The MiLTPG2 expression profile showed low expression during the initial
stages of development (15-60 DAF), almost undetectable levels of expression in
intermediates stages of development (75-135 DAF) and a high expression during
141-147 DAF (Figure 3). Furthermore, the maximum expression level was
recorded during the ripening and storage time (153 DAF), which is the
developmental stage in which it is taking a place a major cuticle accumulation. In
80
general, the MiLTPG2 gene expression does not correlate with cuticle
accumulation during mango fruit ontogeny, exception at initial (15 DAF) and final
stages (153 DAF).
In mango, Petit-Jiménez et al. (2007) reported that epicuticular wax content
increased during fruit growth. Also, they recorded changes during the storage
time. Waxes ultrastructure showed varietal differences among mango varieties
at harvest time. Indeed, cultivars `Tommy Atkins´ and `Kent´ showed a major
proportion of crystalline zones (82.6 %), whereas cultivar `Keitt´ showed a large
percentage of amorphous zones (74.1 %). In other study, Petit-Jiménez et al.
2009 analyzed the relative composition of the fractions in epicuticular wax at 60
DPA in mango `Keitt´. It was found that the largest proportion was alkanes (50-
60%), followed by fatty acids (38-46%) and fatty alcohols (2-4%). In the case of
intracuticular wax, the predominant fraction was found composed of fatty acid
(62-77%), followed by alkanes (21-35%) and alcohol (2-6%) during ontogeny of
fruit. Interestingly, in mango cv. `Keitt´ the alkane fraction at 60 DPA, after
harvest and after storage was 49, 28 and 24 µg/cm2, respectively.
In the Arabidopsis ltpg2 mutant, the composition of the cuticular wax was
significantly altered in the stems and siliques. Indeed, it was recorded a
reduction in level of the C29 alkane, which is the major component of cuticular
waxes. Conversely, studies in mango cv. `Keitt´, it was reported that the alkane
C29 was almost undetectable (traces), although alkanes C21-C32 were reported
at 60 DPA, after harvest and after 18 days of storage (Petit-Jiménez et al.,
2009). Because of this, further experimental evidences are needed to probe that
81
the MiLTPG2 gene studied in this work plays a role transporting the alkane C29
likewise its homolog gene AtLTPG2 from Arabidopsis.
Based in the MiLTPG2 expression profile found in this study, along with
previous data demonstrating that the ABC transporter proteins located in the
plasma membrane of epidermal cells are required for both cutin and wax
deposition (Pighin et al., 2004) and our own published model (Tafolla-Arellano et
al. 2013), we would like to suggest a model showing the putative role of
MiLTPG2 transporting the cuticle wax through the cell wall into the cuticle
(Figure 4).
Conclusions
The MiLTPG2 gene shown the three different domains characteristic of the
LTPG proteins: a signal peptide domain, a lipid transfer domain and a
transmembrane domain, similar to AtLTPG2. The analysis of MiLTPG2 gene
expression provides the first insight to understand the putative role of this gene
in wax transport during cuticle biosynthesis in mango fruit. Further studies are
required to confirm the role of MiLTPG2, particularly enzymatic activity and
changes in cuticle composition among fruit developmental stages. These studies
will help in the elucidation of the molecular mechanism underlying cuticle
biosynthesis of mango fruits.
Acknowledgements
The author Julio César Tafolla Arellano thanks Mexican Council of Science
and Technology (CONACyT) for the doctoral scholarship assigned. This work was
supported by the Project 20120 (P0045001): “Aseguramiento De Calidad De
Frutas Y Hortalizas” from Research Center for Food and Development A.C. The
82
authors gratefully acknowledge Agricola Daniella for helping in the experiment to
obtain mangoes during fruit ontogeny for the study.
References
Altschul S.F., T.L. Madden., A.A. Schäffer., J. Zhang., Z. Zhang., et al.
(1997) Gapped BLAST and PSI-BLAST: a new generation of protein database
search programs. Nucleic Acids Research 25:3389-402
Borner, G. H., D.J. Sherrier., T.J. Stevens., I.T. Arkin., P. Dupree (2002)
Prediction of glycosylphosphatidylinositol-anchored proteins in Arabidopsis. A
genomic analysis. Plant Physiology 129(2): 486-499.
Cameron K.D., M.A. Teece., L.B. Smart (2006) Increased Accumulation of
Cuticular Wax and Expression of Lipid Transfer Protein in Response to Periodic
Drying Events in Leaves of Tree Tobacco. Plant Physiology 140:176-83
DeBono A., T.H. Yeats., J.K. Rose., D. Bird., R. Jetter., L. Kunst., L.
Samuels (2009) Arabidopsis LTPG is a glycosylphosphatidylinositol-anchored
lipid transfer protein required for export of lipids to the plant surface. The Plant
Cell 21(4):1230-1238.
Edstam M.M., K. Blomqvist., A. Eklöf., U. Wennergren., J. Edqvist (2013)
Coexpression patterns indicate that GPI-anchored non-specific lipid transfer
proteins are involved in accumulation of cuticular wax, suberin and
sporopollenin. Plant molecular biology 83(6): 625-649.
Gasteiger E., C. Hoogland., A. Gattiker., S. Duvaud., M.R. Wilkins., R.D.
Appel., A. Bairoch (2005) Protein Identification and Analysis Tools on the
83
ExPASy Server. (In) John M. Walker (ed): The Proteomics Protocols Handbook,
Humana Press (2005). pp. 571-607.
Hong-Bin S., C. Kuo-Chen (2007) "Signal-3L: a 3-layer approach for predicting
signal peptides", Biochemical and Biophysical Research Communications 363:
297-303.
Kader J. C. (1996) Lipid-transfer proteins in plants. Annual review of plant
biology 47(1): 627-654.
Kim, H., S.B Lee., H.J. Kim., M.K. Min., I. Hwang., M.C. Suh (2012)
Characterization of glycosylphosphatidylinositol-anchored lipid transfer protein 2
(LTPG2) and overlapping function between LTPG/LTPG1 and LTPG2 in
cuticular wax export or accumulation in Arabidopsis thaliana. Plant and Cell
Physiology 53(8):1391-1403.
Kunst L., A.L. Samuels (2003) Biosynthesis and secretion of plant cuticular
wax. Progress in Lipid Research 42:51-80.
Lee S. B., Y.S. Go., H.J. Bae., J.H. Park., S.H. Cho., H.J. Cho., ... M.C. Suh
(2009) Disruption of glycosylphosphatidylinositol-anchored lipid transfer protein
gene altered cuticular lipid composition, increased plastoglobules, and
enhanced susceptibility to infection by the fungal pathogen Alternaria
brassicicola. Plant Physiology 150(1): 42-54.
Livak K.J., T.D. Schmittgen (2001) Analysis of Relative Gene Expression Data
Using Real-Time Quantitative PCR and the 2−ΔΔCT Method. Methods 25:402-
8.
84
Martin L.B.B., J.K.C. Rose (2014) There’s more than one way to skin a fruit:
formation and functions of fruit cuticles. Journal of Experimental Botany
65:4639-51.
Mintz-Oron S., T. Mandel., I. Rogachev., et al (2008) Gene Expression and
Metabolism in Tomato Fruit Surface Tissues. Plant Physiology 147:823-51.
Petit-Jiménez D., A. González-León., G. González-Aguilar., R. Sotelo-
Mundo., R. Báez-Sañudo (2007) Cambios de la cutícula durante la ontogenia
del fruto de Mangifera indica L. Revista Fitotecnia Mexicana 30(1):51-60.
Petit-Jiménez D., T. Carvallo-Ruíz, A. González-León., R. Sotelo-Mundo., R.
Báez-Sañudo (2009) Composición química de las ceras cuticulares del fruto de
Mangifera indica L. Revista Iberoamericana de Tecnología Postcosecha
10(1):14-25.
Pierleoni A., P.L. Martelli., R. Casadio (2008) -PredGPI: a GPI anchor
predictor- BMC Bioinformatics 9:392.
Pighin J.A., H. Zheng., L.J. Balakshin., I.P. Goodman., T.L. Western., R.
Jetter., L. Kunst., A.L. Samuels (2004) Plant Cuticular Lipid Export Requires
an ABC Transporter. Science 306:702-4.
Pyee J., H.S Yu., P.E Kolattukudy (1994) Identification of a Lipid Transfer
Protein as the Major Protein in the Surface Wax of Broccoli (Brassica oleracea)
Leaves. Archives of Biochemistry and Biophysics 311:460-8.
Riederer M., L. Schreiber (2001) Protecting against water loss: analysis of
the barrier properties of plant cuticles. Journal of Experimental Botany
52:2023-32.
85
Saladié M, A.J Matas., T. Isaacson., et al. (2007) A Reevaluation of the Key
Factors That Influence Tomato Fruit Softening and Integrity. Plant Physiology
144:1012-28.
Samuels L., L. Kunst., R. Jetter (2008) Sealing plant surfaces: cuticular wax
formation by epidermal cells. Plant Biology 59(1):683.
Sterk P., H. Booij., G.A. Schellekens., A. Van Kammen., S.C. De Vries
(1991) Cell-specific expression of the carrot EP2 lipid transfer protein gene. The
Plant Cell, 3(9), 907-921.
Tafolla-Arellano J.C., A. González-León., M.E. Tiznado-Hernández., L.
Zacarías García., R. Báez-Sañudo (2013) Composición, fisiología y biosíntesis
de la cutícula en plantas. Revista Fitotecnia Mexicana 36:3-12.
Tafolla-Arellano J.C., Y. Zheng., H. Sun., C. Jiao., E. Ruiz-May., R. Báez-
Sañudo., Z. Fei., J.C.K Rose., M.E. Tiznado-Hernández (2014) RNA-Seq
Analysis of the Mango (Mangifera indica L) Fruit Epidermis: Elucidating Factors
that Determine the Postharvest Quality of Tropical Fruits. The 2nd Plant
Genomics Congress USA. Saint Louis, Missouri, USA.
Thornton B., C. Basu (2011) Real-time PCR (qPCR) primer design using free
online software. Biochemistry and Molecular Biology Education 39:145-54
Yeats T. H., J.K.C. Rose (2008) The biochemistry and biology of extracellular
plant lipid‐transfer proteins (LTPs). Protein Science 17(2):191-198.
Yeats T.H., K.J. Howe., A.J. Matas., G.J. Buda., T.W. Thannhauser., J.K.C.
Rose (2010) Mining the surface proteome of tomato (Solanum lycopersicum)
fruit for proteins associated with cuticle biogenesis. Journal of Experimental
Botany 61:3759-71.
86
Wan C.Y., T.A. Wilkins (1994) A Modified Hot Borate Method Significantly
Enhances the Yield of High-Quality RNA from Cotton (Gossypium hirsutum
L.). Analytical Biochemistry 223:7-12.
Figures
Figure 1. Characterization of MiLTPG2 protein. (A) MiLTPG2 amino acid
protein composition. (B) Schematic representation of the different MiLTPG2
domains. SP, signal peptide. GPID, Glycosylphosphatidylinositol-Anchored
domain. The numbers are indicating the location of the amino acid in the
sequence of the LTPG2 protein. (C) Alignment of AtLTPG2 with MiLTPG2. The
single and double underlines are indicating the signal peptide domain and the
GPI transmembrane domain, respectively. The conserved Cys residues are
highlighted in grey color.
87
Figure 2. Real time quantitative reverse transcription PCR analysis of MiLTPG2 transcripts in mango fruit peel. (A) Alignment of the MiLTPG2
sequence (MiLTPG2 1) with the DNA sequence obtained from the fragment
(MiLTPG2) amplified in the qRT-PCR. (B) Agarose gel electrophoresis of the PCR
amplified products. Lane 1: 50 bp DNA ladder, lane 2 MiLTPG2 (size band: 115
bp), and lane 3: MiActin1 (size band: 141 bp).
88
Figure 3. Profile expression of MiLTPG2 and cuticle accumulation during mango fruit ontogeny. Regulation of the MiLTPG2 gene expression is shown in
bars and the line included is indicating the cuticle accumulation during mango fruit
ontogeny.
89
90
Figure 4. Theoretical model explaining the putative role of MiLTPG2 in
cuticular wax transport. The ABC transporter is carrying out the monomer
transport through the plasma membrane and the MiLTPG2 protein is
transporting the cuticular wax through the cell wall and into the cuticle. This is a
model describing the role of MiLTPG2 in the molecular mechanism of cuticle
biosynthesis based in data generated in this work and the current knowledge.