Genomic approaches to understand the genetic response to Phytophthora cinnamomi Rands in Castanea spp.

Carmen Sofia Pedro dos Santos Dissertation presented to obtain the Ph.D degree in Biology Instituto de Tecnologia Química e Biológica António Xavier | Universidade Nova de Lisboa Oeiras, January, 2017 Genomic approaches to understand the genetic response to Phytophthora cinnamomi Rands in Castanea spp. II Work performed at: Molecular Biology Laboratory UEISSAFSV - Instituto Nacional de Investigação Agrária e Veterinária, I.P. Av. da República, 2780-157 Oeiras, Portugal Plant Cell Biotechnology Laboratory Instituto de Tecnologia Química e Biológica António Xavier Universidade Nova de Lisboa Av. da República, 2780-157 Oeiras, Portugal PhD Supervisors: Doctor Rita Lourenço Costa Head of laboratory, Molecular Biology Laboratory -UEIS Sistemas Agrários Florestais e Sanidade Vegetal, Instituto Nacional de Investigação Agrária e Veterinária, I.P. Professor Pedro Fevereiro Head of laboratory, Plant Cell Biotechnology Laboratory, Instituto de Tecnologia Química e Biológica António Xavier, Universidade Nova de Lisboa. Assistant Professor, Departamento de Biologia Vegetal, Faculdade de Ciências da Universidade de Lisboa II III ‘(...) o fruto dos frutos, o único que ao mesmo tempo alimenta e simboliza, cai de umas árvores altas, imensas, centenárias, que, puras como vestais, parecem encarnar a virgindade da própria paisagem. Só em Novembro as agita uma inquietação funda, dolorosa, que as faz lançar ao chão lágrimas (...). Abrindo-as, essas lágrimas eriçadas de espinhos deixam ver numa camada fofa a maravilha singular de que falo, tão desafectada que até no próprio nome é doce e modesta – a castanha.’ Miguel Torga (‘Reino Maravilhoso’) IV V Table of contents Acknowledgments/Agradecimentos VII List of abbreviations XI Summary XIII Sumário XVII Chapter I: General introduction 1 Chapter II: Phenotyping Castanea hybrids for Phytophthora cinnamomi resistance 43 Chapter III: Castanea root transcriptome in response to Phytophthora cinnamomi infection 73 Chapter IV: Expression analysis of genes associated with Castanea - Phytophthora cinnamomi interaction 139 Chapter V: Genetic mapping of resistance to Phytophthora cinnamomi in interspecific progenies of Castanea species 167 Chapter VI: Conclusions and future perspectives 203 Funding acknowledgment 211 VI Acknowledgments/Agradecimentos VII Acknowledgments/Agradecimentos This thesis represents not only my work at the keyboard, it is a milestone in these last years of work with chestnut. At the end of this journey, I would like to express my gratitude to the people who directly or indirectly contributed to this thesis: • Rita Lourenço Costa, my supervisor, first for giving me the opportunity to develop this PhD work in chestnut, the plant for which I am passionate. I am so proud to have worked under the breeding program that you initiated. I am truly grateful for all supervision, guidance, support, dedication and advices, which were precious in all stages of this work and allowed me to improve at both professional and personal level. Thank you so much for your continuous optimism and enthusiasm with this work, which encouraged me throughout the last years. I am also thankfull for careful and efficient way on revise this manuscript. • Pedro Fevereiro, my co-supervisor, for his pertinent questions and critical vision that helped me to look to the results in different perspectives, which made me grow scientifically. I am also thankful for rigorous reading and commenting this manuscript. • Helena Machado, the scientist that most taught me about Phytophthora and how to deal with this challenging pathogen. Thank you for all unconditional support and encouragement, you have been a true friend to me. • Dana Nelson, my American supervisor, for your knowledgeable advices on mapping approaches. I owe my deepest gratitude to you for having welcomed me so well in your daily life in USA. Thank you for our didactic and cultural conversations during the trips to the lab and back home, you were the best host ever! Summary XIV assays, was strongly and negatively correlated with the days of survival recorded after root inoculations. Therefore, the excised shoot inoculation test revealed be a reliable approach for screening the metrics of resistance of chestnut genotypes to P. cinnamomi. Moreover, a set of resistant genotypes was selected, constituting a valuable source of new genetic resources, essential to address the shortcomings of the Portuguese and European chestnut market. The association between genotype and phenotype enabled the identification of unique QTLs for P. cinnamomi resistance. Ten QTLs were mapped on five linkage groups of the European x Japanese chestnut map. The presence of QTLs on linkage group E was consistent with a previous pilot study for identification of QTLs in backcross families (Chinese chestnut x American chestnut hybrid), suggesting that different Castanea species might share resistant haplotypes, and therefore, common resistance mechanisms. Concerning the transcriptomic approach, candidate genes for P. cinnamomi resistance were identified from the root transcriptome of European and Japanese chestnut inoculated and non-inoculated with the pathogen. Those genes are involved, in both species, in the regulation of plant immune response and stress adaptation and recovery. The expression levels of eight of the candidate genes were quantified by digital PCR, using European and Japanese chestnut and four hybrid genotypes showing different levels of susceptibility to the disease. RNA-seq and gene expression analysis suggested that both species recognize the pathogen attack, which may trigger resistance signaling pathways and cell wall modification, as well as, the production of anti-fungal proteins. However, the resistant species may involve basal defense mechanisms, being protected in advance to the infection. Additionally, new molecular markers were developed from the sequences of candidate genes identified by transcriptome sequencing. Forty-one microsatellite showing polymorphism and high transferability within and Summary XV among chestnut species were used for genotyping European x Japanese chestnut populations. Two of them were mapped within the identified QTL intervals, being strong candidates for further validation and marker-assisted selection. The knowledge acquired in this project is a major breakthrough in understanding the Castanea-P. cinnamomi interactions and may contribute for the development of strategies to control ink disease. Furthermore, this project developed a crucial deliverable for farmers and society, since the genotypes with improved resistance to the pathogen have been propagated, to be released to the market as rootstocks, in the near future. XVI Sumário XVII Sumário O castanheiro é uma árvore polivalente, com importante impacto económico, ecológico e científico. O castanheiro Europeu (Castanea sativa) produz as castanhas mais apreciadas e valorizadas no mundo. No entanto, a área de soutos e castinçais está a diminuir na Europa devido a doenças e pragas, principalmente a doença da tinta. Esta doença altamente destrutiva é causada pelo oomiceta Phytophthora cinnamomi, difundido por todo o mundo. A infecção por P. cinnamomi ocorre nas raízes causando a sua podridão e levando à morte em espécies susceptíveis. No entanto, o nível susceptibilidade varia entre castanheiros, sendo as espécies asiáticas as mais resistentes ao patógeno. A investigação desenvolvida durante esta tese foi realizada em descendências que segregam para a característica de interesse, obtidas a partir do programa de melhoramento estabelecido há 10 anos. Este programa é baseado em cruzamentos controlados entre o castanheiro japonês resistente e o castanheiro europeu susceptível ao agente patogénico. Foram implementadas abordagens de mapeamento e de transcriptómica visando compreender os diferentes mecanismos de resposta do castanheiro à doença. Até ao momento foram obtidas 155 descendências, que foram genotipadas e fenotipadas de forma a mapear as regiões genómicas que controlam a resistência a P. cinnamomi (Quantitative Trait Loci-QTLs). Usando marcadores moleculares (microsatélites e SNPs) derivados de transcriptomas obtidos, previamente e durante este trabalho, foi realizada a genotipagem dos progenitores e respectivas descendências. Os dados de segregação obtidos foram analisados para a construção do primeiro mapa genético de castanheiro Europeu x castanheiro Japonês. O mapa genético interespecífico contém 283 marcadores moleculares, mapeados em 15 Sumário XVIII grupos de ligação e abrangendo um total de 714,8 cM, o que corresponde a cerca de 96% do mapa de referência de castanheiro Chinês. A fenotipagem foi realizada através da avaliação das métricas de resposta à doença, obtidas para todas as descendências, após a inoculação de raízes e/ou de estacas excisadas das plantas-mãe, com P. cinnamomi. A taxa de progressão da lesão observada nos ensaios de inoculação realizados em estacas foi fortemente e negativamente correlacionada com os dias de sobrevivência registados após a inoculação das raízes. Assim, o teste de inoculação em estaca demonstrou possuir rigor para avaliar a resistência à doença da tinta em diferentes genótipos de castanheiro. Além disso, foram selecionados um conjunto de genótipos com resistência melhorada a P. cinnamomi, constituindo novos recursos genéticos essenciais para colmatar o elevado défice de material vegetal melhorado no mercado tanto em Portugal e na Europa. A associação entre genótipo e fenótipo permitiu a identificação de QTLs relacionados com a resistência a P. cinnamomi pela primeira vez em castanheiro. Dez QTLs foram mapeados em cinco grupos de ligação do mapa genético de castanheiro Europeu x castanheiro Japonês. A presença de QTLs em determinados grupos de ligação foi consistente com um estudo piloto realizado anteriormente para a identificação de QTLs em famílias do programa de melhoramento Americano, sugerindo que as diferentes espécies do género Castanea podem partilhar haplótipos e mecanismos de resistência. Em relação à abordagem de transcriptómica, os genes candidatos para a resistência a P. cinnamomi foram seleccionados a partir do transcriptoma de raízes de castanheiro europeu e japonês respetivamente inoculadas e não inoculadas com o patogénio. Para ambas as espécies, estes genes estão envolvidos na regulação da resposta imune das plantas e na adaptação e recuperação do stress biótico. Os níveis de expressão de oito genes foram quantificados por PCR digital, em raízes de castanheiro Europeu e Japonês Sumário XIX e de quatro genótipos híbridos mostrando diferentes níveis de susceptibilidade à doença. A análise de expressão génica e da sequenciação do transcriptoma, sugere que ambas as espécies reconhecem o ataque de patogénio, podendo desencadear vias de sinalização de resistência que podem resultar na a modificação da parede celular e/ou na produção de proteínas antifúngicas. No entanto, as espécies resistentes parecem envolver mecanismos de defesa basal, encontrando-se protegidas antecipadamente à infecção. Adicionalmente, foram desenvolvidos novos marcadores moleculares a partir das sequências de genes candidatos, identificados na sequenciação dos transcriptomas. Quarenta e um microssatélites mostrando polimorfismo e alta transferibilidade, dentro e entre as diferentes espécies de castanheiro, foram utilizados para a genotipagem das populações híbridas de castanheiro Europeu x castanheiro Japonês. Dois dos marcadores desenvolvidos foram mapeados dentro de intervalos dos QTLs identificados, por isso constituem-se como fortes candidatos para validação adicional e seleção assistida por marcadores moleculares. O conhecimento adquirido neste estudo constituiu um grande avanço na compreensão da interação entre Castanea e P. cinnamomi, podendo contribuir para o desenvolvimento de estratégias de controlo da doença da tinta. Além disso, este projecto desenvolveu um valioso produto para os agricultores e para a sociedade, uma vez que os genótipos com resistência melhorada ao patógenio que estão a ser propagados, serão lançados no mercado, como os porta-enxertos, num futuro próximo. XX General introduction 1 Chapter I General Introduction Chapter I 2 The chestnut: since ancient times to the present The chestnut is a multipurpose tree that has a very ancient history and tradition, as well as an important economic and environmental role. Chestnuts were historically distributed only throughout the northern hemisphere, but due to anthropogenic influences have been introduced into Chile, Argentina, Australia and New Zealand, being currently widely cultivated all over the temperate regions (Conedera and Krebs, 2008; Pereira-Lorenzo et al. 2012). Since the Middle Ages, the nuts of European chestnut, a noble hardwood, and also of Japanese and Chinese chestnuts, provided an essential food source, resulting in diverse types of use: fresh consumption, long-term storage, drying, flour and animal feed (Bounous and Marinoni 2005; Bounous 2009). In North America, the American chestnut, known as a forest giant, was a dominant species along the Appalachian range. The American chestnut trees played a dominant role for American people, since the wood was extensively used for building houses and furniture or used as fuelwood. Nuts were part of the human diet and also, they had an important role for wildlife in the forests (Anagnostakis, 2012; Jacobs et al. 2015). Nowadays, the chestnut continues to have an important role in many agroforestry systems. Although they are no longer a subsistence food, chestnuts are currently an appreciated product for an increasingly large market sector. The nuts, with both modern and traditional methods of storage and processing, meet the demand of consumers, who are progressively seeking for nutritious and healthy foods. Impact of chestnut in economy and environment The cultivation of chestnut has been mainly related with the abundant and versatile uses of the products that can be obtained: edible nuts, timber for building and for other woody products, associated mushrooms production and extracted tannins for tanning leather or pharmaceutical purposes. General introduction 3 Moreover, the chestnut ecosystems always contributed for biodiversity preservation, representing an important cultural heritage (Paillet 2002; Bounous 2005; Bounous and Marinoni 2005; Conedera and Krebs 2008; Bounous 2009). Chestnuts are one of the most important nut crops in the temperate zone. They have a delicious taste, being consumed in fresh, dried or processed. Processing is necessary to increase the available products and to extend the use of the product along the year (Bounous and Marinoni 2005). They have many culinary uses, ranging from first to main course dishes (used as side or served whole, boiled or roasted) as well as vegetable dishes (as soups, chestnut purées), desserts (marrons glacés, cakes, ice-creams and syrups) and pastries (as chestnut creams, mousse). It is also possible to prepare beverages from chestnuts such as liqueurs, beers and non-alcoholic drinks. Some examples of chestnut uses are shown in Figure 1. From a nutritional point of view, chestnuts are a very healthy, balanced and high-energy food. Fresh chestnuts are high in calorie content, low in fat and sodium, free of cholesterol and gluten, with a moderate but high-quality protein content and a favorable amino-acid ratio (Pereira-Lorenzo et al. 2006; De Vasconcelos et al. 2010). Many countries around the world have suitable edaphoclimatic conditions for chestnut plantation. The European, Chinese and Japanese chestnuts are the main species cultivated for fruit, due to their large nut size. Nevertheless, European chestnut (marrone types) are considered the most valuable for nut production (Pereira-Lorenzo et al. 2012). Interspecific hybrids which have emerged from disease resistance studies are also used for nut production directly or as rootstocks. Chapter I 10 The first unambiguous pollen data showing evidence of European chestnut trees spreading due to human activities date back to around 2100-2050 B.C. (Conedera et al. 2004). Currently, C. sativa is commonly found in Europe between 400 and 1 000 m above sea level depending on the latitude. The lowest elevations are recommended for the highest latitudes and vice versa (Mellano et al. 2012). The European chestnut germplasm is very extensive; there are hundreds of cultivar names used for chestnuts, many of which are synonyms or homonyms (Botta et al. 2001). In 2013, Italy was the largest chestnut producer in the European Union (FAOSTAT, 2016, faostat.fao.org), and led the world in producing processed chestnut products such as marron glacé (Bounous 2009). However, the introduction of gall wasp has decreased the nut production (Battisti et al. 2014). Italian marron type cultivars are ‘Chiusa Pesio’, ‘Luserna’, ‘Val Susa’, ‘Castel del Rio’, ‘Marradi’ and ‘Fiorentino’. Within Europe, France is the largest chestnut importer, mostly buying from Italy, but also from Spain and Portugal. Some C. sativa traditional French cultivars include ‘Bouche Rouge’, ‘Verdale’, ‘Arizinca’, ‘Toumive’, ‘Belle Epine’, ‘Savoye’, ‘Châtaigne de Laguepie’, ‘Sardonne’, ‘Rouse de Nay’, and ‘Dorée de Lyon’ (Mellano et al. 2012). In Iberian Peninsula, it seems that cultivar diversification was a result of distinct genotypes being related via hybridization and mutation, regardless of whether they shared the same name or not (Pereira-Lorenzo et al. 2011). In Spain, the main cultivars are ‘Loura’, ‘Garrida’, and ‘Parede’, being ‘Garrida’ more suited to industrial purposes, conservation and genetic studies and European breeding programs (Pereira-Lorenzo et al. 2001; Blanco Silva and Fernández-López, 2005; Pereira-Lorenzo et al. 2006; Míguez-Soto and Fernández-López 2012; Fernández-López and Fernández-Cruz, 2015; Míguez-Soto and Fernández-López 2015; Fernández-Cruz and Fernández-López, 2016). General introduction 11 In Portugal, more than 25 cultivars are known, they showed great genetic variability (each cultivar includes different genotypes) indicating their polyclonal origin (Costa et al. 2008). One of the most ancestral cultivar, that is distributed throughout the entire Iberian Peninsula, is ‘Longal’ that has been promoted as the best cultivar for industry (Pereira-Lorenzo et al. 2011). ‘Judia’ and ‘Martaínha’, due to their larger nut size, are usually preferred for the fresh market. In Portugal, chestnut is distributed mainly in the Northeast (Trás-os-Montes, Minho and Beira Litoral) but also is found in the center east, especially in Marvão region (Costa et al. 2008). Accordingly, four regions of Protected Designation of Origin (POD) were created to preserve the Portuguese cultivars: Castanha da Terra Fria, Castanha da Padrela, Castanha dos Soutos da Lapa and Castanha do Marvão. Reproductive biology and hybridization Castanea species is a monoecious species that generally flowers from June to July depending on the species; Asian species show precocious blossoming when compared to European species (Botta et al. 1995). C. sativa have been reported to begin flowering after 8-10 years, but flowering time can be shorted though grafting. Production is regular and high (GomesLaranjo et al. 2009). Female flowers are pollinated by wind (more usual in case of dry weather during flowering) or insects (dominating in wet weather conditions). Inflorescence male flowers are gathered in catkins that can occur in two types: bisexual catkins that bear one or more female flowers at the base and male flowers toward the tip; and unisexual male catkins, also called staminate catkins (Mert and Soylu 2006). Female inflorescence generally contains three flowers that are protected by a green, scaled wrapping that is destined to form the cupule that develops into the chestnut bur. Usually female inflorescences are positioned at the base of the male ones, in the Chapter I 12 upper part of the current year’s shoots (Hebard et al. 2014a). Fertilization produces typically three large and brownish nuts encapsulated in a spiny bur. The burs protect the seeds until they are ripe and then open widely, making the nuts readily available. Nuts ripen early in September to November (Hebard et al. 2014a). The nuts of Castanea genus vary greatly among species and cultivars. Nevertheless, in average European chestnuts show the largest size in the genus and can weight more than 30 g (Figure 3). Chinese and Japanese chestnuts have similar size and the American chestnuts are much smaller (Figure 3). Figure 3. Chestnuts morphology depending on the species. Photograph by Dr. Paul Sisco. The chestnut is very often self-incompatible, therefore cross-pollination is compulsory (Mert and Soylu 2006). Very little is still known about the genetic system controlling mating and the self-incompatibility system in chestnut, although it is considered to be of gametophytic type (Zou et al. 2014). Interspecific hybridizations between all Castanea species are possible. Main problems are related with the different flowering time between species. In all interspecies crosses, chromosome pairing exists among Castanea species, but the presence of segregation distortion in some mapping populations (Casasoli et al. 2001; Kubisiak et al. 1997; Kubisiak et al. 2013) suggest that significant chromosomal differences such as translocations and/or inversions may occur. For breeding purposes, artificial controlled crosses have been performed (Costa et al. 2011; Takada et al. 2012; Nishio et al. 2013; Nelson et al. 2014; Fernández-Cruz 2015): donor pollen is easily collected from the catkins; receptor female flowers are isolated avoiding unknown pollination; male flowers from the receptor tree must be also removed (emasculation); manual pollination is performed by placing donor pollen over the pistils; pollination General introduction 13 bags are used for covering female flowers avoiding pollen contamination; pollination bags are removed and replaced by net bags in order to collect the burs containing the nuts. Genomic resources Genomic research on forest trees has been motivated by the need to support genetic breeding programs and develop tools for conservation, restoration and management of natural populations (Neale and Kremer 2011). Important genomic resources such as ESTs molecular markers and genetic maps have been developed for chestnut and integrated in a Web-based resource for the Castanea genetics/genomics community (Fagaceae Genomic Database: www.fagaceae.org). Moreover, the whole genome sequencing project for C. mollissima is underway (www.hardwoodgenomics.org). Based on flow cytometric analysis, genome size appears to be fairly conserved among species: the estimated 1C genome size of C. sativa is 777Mb, of C. crenata and C. mollissima is 794 Mb and of C. dentata is 803 Mb (Kremer et al. 2007). Therefore, the genome size of Castanea species is only five times larger than Arabidopsis and less than twice the size of the poplar genome. The manageable genome size and abundant genetic and genomic resources make Castanea a good candidate as model for Fagaceae family in the near future. A large component of chestnut genomic resources is focused on the transcriptomes obtained for C. mollissima, C. dentata, C. sativa and C. crenata (Barakat et al. 2009; Sebastiana et al. 2009; Nishio et al. 2011; Barakat et al. 2012; Serrazina et al. 2015, Chapter III). Large EST databases are being created with significant numbers of sequence contigs showing similarity to predicted proteins in woody plants. Resistance candidate genes to chestnut blight and ink disease (Barakat et al. 2009; Barakat et al. 2012; Serrazina et al. 2015, Chapter III). have been identified in EST sequence data, as well as candidate genes for other traits (Sebastiana et al. 2009; Chapter I 14 Nishio et al. 2011). Furthermore, a great number of molecular markers have been developed from those sequences mainly Simple Sequence Repeats (SSRs) or microsatellites and Single Nucleotide Polymorphism (SNPs) (Nishio et al. 2011; Kubisiak et al. 2013; Santos et al. 2015, Chapter V). Beyond molecular markers derived from ESTs databases, smaller sets of SSR markers were earlier developed from enriched genomic libraries of European chestnut (Marinoni et al. 2003; Buck et al. 2003), Japanese chestnut (Yamamoto et al. 2003) and Chinese chestnut (Inoue et al. 2009). Microsatellite and SNP markers are highly informative, transferable across related taxa, having great prevalence in the genome and amenability to automated high-throughput analysis. Therefore, the molecular markers developed so far for chestnut are an invaluable resource for the scientific community interested in all aspects of the genetics, breeding and biotechnology. For breeding purposes, the molecular markers have been mapped on genetic maps constructed for the four main Castanea species (Kubisiak et al. 1997; Casasoli et al. 2001; Sisco et al. 2005; Kubisiak et al. 2013; Nishio et al. 2013 and Chapter V). The genetic map constructed for C. mollissima (Kubisiak et al. 2013) was accepted as the chestnut reference map and was integrated with the physical map obtained by sequencing of BAC libraries (Fang et al. 2013). Quantitative trait loci (QTLs) related with chestnut blight and adaptive traits have been identified and confirmed (Kubisiak et al. 1997; Casasoli et al. 2004; Kubisiak et al. 2013). Moreover, syntenic regions have been identified between the chestnut physical map and some genomes available of other related taxa, revealing syntenic regions between QTLs for reistance to chestnut blight diasese and QTLs for resistance to other fungal pathogens in Prunus spp. (Staton et al. 2015). However, only an exploratory study identified QTLs for P. cinnamomi resistance (Zhebentyayeva et al. 2014) until to the present study. General introduction 15 Major diseases affecting chestnut The most damaging diseases of chestnut are the ink disease, caused by the oomycete Phytophthora cinnamomi (and P. cambivora) and chestnut blight caused by the ascomycete fungus Chryphonectria parasitica. European and the American chestnut are highly susceptible to these pathogens whereas, Asian Castanea species show great resistance to the diseases (Crandall et al. 1945). Ink disease Ink disease, also known as root rot, is the most destructive disease affecting European chestnut. In most cases, Phytophthora cinnamomi is the causal agent of ink disease, whilst Phytophthora cambivora is less frequent and aggressive (Gouveia 2004). P. cinnamomi is a soilborne pathogen that parasites fine roots causing root and collar rot that extends to trunk and branches of young and mature trees, and consequently causes death. Typical symptoms include chlorosis and wilting of foliage, dieback of branches and crown (Robin et al. 2001; Vettraino et al. 2001; Hardham 2005; Kamoun et al. 2014). It is a silent disease since when first symptoms become visible in the crown the destruction of the fine root system is already in an advanced stage. The geographical origin of P. cinnamomi is not clearly established, however there are evidences for an Asian origin and it was spread across the Pacific to Latin America (Ko et al. 1978; Zentmyer 1988; Zhang et al. 1994). Ink disease on Castanea was first reported in Portugal in 1838 (Vettraino et al. 2001), and in the USA in 1825 (Rhoades et al. 2003). Since then, P. cinnamomi have spread over Europe and North America and their native plant species were not adapted and therefore they are often highly susceptible. Currently, P. cinnamomi is the most widely distributed Phytophthora species (Figure 4). Furthermore, climate change is predicted to have a significant impact on the intensity and distribution of P. cinnamomi Chapter I 16 (Thompson et al. 2014). The pathogen infects more than 3000 host species causing great economic impacts in forestry and horticulture, and in the nursery industry (Hardham 2005). Beyond Castanea species, P. cinnamomi affects most of the temperate of fruit trees such as Persea, Quercus, Ericaceae, Eucalyptus, Cinnamomum, Coniferales, Fagus, Juglans and many ornamental trees and shrubs (Hardham 2005; Robin et al. 2012). Like other Phytophthora spp., P. cinnamomi has a number of strategies for survival, propagation and dissemination. It is an oomycete and not a fungus, although everything about its biology and life cycle is fungus-like, such as mycelial growth habit. Features that differ oomycetes from fungi include the production of biflagellate heterokont zoospores, the occurrence of cellulose rather than chitin in the cell walls and diploid somatic cells (Hardham et al. 1994; Hardham 2005). Figure 4. Phytophthora cinnamomi distribution worldwide (in grey), adapted from European and Mediterranean Plant Protection Organization, Global Database (2016). P. cinnamomi is able to survive under unsuitable environmental conditions over several years in the soil or in infected root tissue, as dormant resting spores: chlamydospores, which are the most common or oospores, produced when different strains mate. Although sexual reproduction of P. cinnamomi is poorly understood, it is known that the pathogen is General introduction 17 heterothallic, requiring the presence of opposite mating types, designated A1 and A2, to form oospores (Hüberli et al. 1997; Hardham 2005). Nevertheless, in the most cases, P. cinnamomi has an asexual sporulation, through development of multinucleate sporangia (Hardham 2005) (Figure 5). When conditions favour growth prevails (high soil moisture, soil temperature superior to 10°C) the resting spores germinate and somatic hyphae form multinucleate sporangia that cleave and release motile, biflagellate and wallless zoospores into the soil water (Figure 5). These zoospores are chemotactically attracted by young fine root exudates, at the contact moment the zoospores encyst, forming walled cysts that germinate and penetrate the tissue. P. cinnamomi is able to grow interand intracellular showing typical coralloid to irregular and non-septate hyphae. Within 2-3 days in a susceptible host, sporangia will form on the plant surface. The asexual cycle may be repeated million of times in quick succession, rapidly amplifying the inoculum potential in the infected area (Hardham et al. 1994; Erwin and Ribeiro. 1996; Hardham 2005; Jung et al. 2013; Oßwald et al. 2014). Figure 5. Life cycle of soilborne Phytophthora cinnamomi (adapted from Hardham et al. 2005). Sexual and asexual sporulation are shown. Chapter I 18 On a local scale, the pathogen can be moved naturally by soil-splash, by wind-blown soil or debris, or by water movement and run-off in drainage/irrigation ditches. The most likely source of more distant movement is in contaminated soil or plant debris. Propagules can also be carried on machinery used for cultivation/harvesting (Hardham 2005; Robin et al. 2012). Cultural control measures include reliving of high soil moisture levels and improving aeration by increasing drainage, and attention to mineral nutrition. Chestnut blight Cryphonectria parasitica, a filamentous ascomycete fungus, is a necrotrophic pathogen that incites the chestnut blight disease. The destruction of the American chestnut by C. parasitica, was the greatest disaster in the history of forest pathology. It is thought to have been imported on seedlings from Asia and it was first discovered in 1904, on infected American chestnut trees at the Bronx Zoological Park in New York (Anagnostakis, 1987; Anagnostakis, 2001). By 1950, the disease had spread throughout its natural range, and by 1960 had killed an estimated 4 billion trees. In Europe, C. parasitica was first recorded in 1938 in Italy and was rapidly spread to the surrounding countries. Chestnut blight became one of the major pathogens that attacked chestnut trees and constituted a serious damage to European chestnut (Anagnostakis 1987; Robin and Heiniger 2001; Jacobs et al. 2015). Currently, C. parasitica is distributed along Europe, United States, west Asian and Australia (Figure 6). The pathogen infects primarily through wounds on stem tissues and kills the above ground portions of trees by girdling the cambium. Once established as germinating conidia (single-celled spores, produced asexually) or ascospores, the fungus grows rapidly through the bark and colonizes the cambial zone. Resistant reactions slow this growth, maintaining the fungus in a superficial canker, whereas susceptible reactions continue development General introduction 19 unimpeded, encircling the stem and causing vascular dysfunction, resulting in death of distal tissues and stem dieback (Anagnostakis, 2012). Figure 6. Cryphonectria parasitica distribution worldwide (in grey), adapted from European and Mediterranean Plant Protection Organization, Global Database (2016). Major pests affecting chestnut More than 50 species of insects are known to damage chestnut, including the Dryocosmus kuriphilus (Yasumatsu), which is the most severe insect pest worldwide affecting chestnut. D. kuriphilus attacks the vegetative buds of chestnuts and forms a gall, disrupting twig growth and reducing fruiting. Severe infestations may result in the decline and death of chestnut trees. This insect is endemic in China, and was accidentally introduced into Japan (1941), Korea (1963), and the USA (1974) (Abe et al. 2007). In 2002, gall wasp was reported for the first time in Europe in northwest Italy (Brussino et al. 2002). From then, chestnut gall wasp has been spread throughout Europe, being present in many countries (European and Mediterranean Plant Protection Organization, Global Database). Beyond European chestnut, D. kuriphilus attacks Asian chestnut species, the American chestnut and their hybrids. Chapter I 26 reprogramming the host cell to accommodate the needs of the pathogen. Intracellular disease resistance proteins mediate recognition of effectors entering the host cell and elicit effector-triggered immunity (ETI) (Jones and Dangl 2006; Stael et al. 2015). Besides local immune responses, PTI and ETI activate long-distance defense reactions, such as systemic acquired resistance (SAR) (Durrant and Dong 2004). Plants also have the ability to defend themselves against different pathogens also by regulating transcriptional activity, induction of tailored defense responses including callose deposition, cell wall thickening and production of reactive oxygen species (ROS) (Jones and Dangl 2006; Stael et al. 2015; Herrera-Vásquez et al. 2015). Plant hormone balance also play a key role in determining the outcome of plant–pathogen interactions. The best characterized defense hormones include salicylic acid (SA), jasmonic acid (JA), absisic acid (ABA) and ethylene (Spoel and Dong 2008; Bari and Jones 2009; Pieterse et al. 2009). If all defense responses fail to inhibit pathogen ingress, the plant cell under attack can undergo hypersensitive cell death (Kamoun et al. 1999; Mur et al. 2008; Choupina et al. 2014). Nevertheless, Phytophthora species have evolved a range of counter-defense mechanisms that can inhibit all host defense processes mentioned (Hardham and Blackman 2010). Despite all the knowledge acquired, molecular mechanisms involved in woody plants resistance to Phytophthora species are poorly understood. This research is challenging because no genome data is available for many wood species and so, identification and characterization of Phytophthora resistance genes is necessary. Nevertheless, strategies of attack and defense in plantsoomycete interactions were recently reviewed (Oßwald et al. 2014; Fawke et al. 2015). Apart from the Castanea genus, the most characterized interactions in Fagaceae family are Quercus suber - P. cinnamomi (Coelho et al. 2006; Coelho et al. 2011; Ebadzad and Cravador 2014) and Fagus sylvatica - P. citricola (Portz et al. 2011; Schlink, 2010). Q. suber - P. General introduction 27 cinnamomi interactions have been studied by cloning and characterization of a set of candidate resistance genes (Coelho et al. 2006; Coelho et al. 2011; Ebadzad and Cravador 2014). Moreover, a hypothetical mechanisms model was proposed for five of those genes for which expression was increased 24hpi (Coelho et al. 2011). Molecular interactions observed between F. sylvatica and P. citricola was characterized by transcriptional changes after infection. Results indicated that P. citricola escapes the main recognition systems and/or suppresses the host's response (Schlink 2010). As a first step to identify transcripts involved in the Castanea - P. cinnamomi interaction, our research group identified and characterized root transcriptomes expressed sequence tags (ESTs) differentially expressed in European chestnut (Castanea sativa) and Japanese chestnut (Castanea crenata), in response to inoculation with P. cinnamomi (Serrazina et al. 2015, Chapter III). Nevertheless, the pathogenic process should comprise a network of molecular signaling and interaction events in different time points after P. cinnamomi infection that were not yet achieved in Castanea spp. Research objectives and thesis layout The general aim of the work here described was to provide new insights about the Castanea resistance mechanisms to P. cinnamomi infection using different approaches: genomics, phenomics and transcriptomics. The outcomes of this project constitute an essential contribution to the understanding of chestnut response to P. cinnamomi based on an elite plant material created from the breeding program, that segregates for the trait of resistance. The development of improved chestnut genotypes with increased resistance to pathogens and the production of genomic resources for future molecular assisted selection will also constitute an asset for the improvement and adaptation of woody plants, mainly belonging to Fagaceae family, to biotic stresses. Chapter I 28 The specific objectives of this work were: 1. Perform new controlled crosses between C. sativa and C. crenata in order to increase the hybrid chestnut population obtained in 2006 and 2009; 2. Phenotype C. sativa x C. crenata (SC) hybrid progenies obtained previously and obtained from new crosses. Phenotype a small population from C. sativa x C. mollissima (SM) controlled crosses to compare levels of resistance among progenies with different donors of resistance; 3. Construct the first interspecific genetic map for C. sativa x C. crenata population through genotyping of parents and progenies with molecular markers: microsatellites or Simple Sequence Repeats - SSRs and Single Nucleotide Polymorphism - SNPs; 4. Perform DNA marker:trait association analysis to identify genomic regions that explains the phenotypic variation in the SC population, by identification of Quantitative Trait Loci. 5. Identify candidate genes related with the resistance to P. cinnamomi by comparing the root transcript profiles of resistant and susceptible species, before and after inoculation; 6. Evaluate the expression of genes potentially involved in the resistance to P. cinnamomi in parental genotypes (C. sativa and a C. crenata), as well as, in hybrid genotypes with different responses to P. cinnamomi. 7. Localize in the genetic map the differential expressed genes by developing molecular markers on sequences obtained from the root transcriptomes. This thesis presents all the work organized in scientific articles, from Chapter II to V, and the work followed the steps described in Figure 7. Final conclusions and future perspectives are discussed in Chapter VI. General introduction 29 Figure 7. General organization of the research and thesis, highlighting the main approaches and techniques used during thesis studies. Chapter I 30 References Abe Y, Melika G, Stone G (2007) The diversity and phylogeography of cynipid gallwasps (Hymenoptera: Cynipidae) of the oriental and eastern Palearctic regions, and their associated communities. Orient Insects 41:169–212. doi: 10.1080/00305316.2007.10417504 Anagnostakis S (1987) Chestnut blight: the classical problem of an introduced pathogen. Mycologia 79:23–37. doi: 10.2307/3807741 Anagnostakis SL, Hillman B (1992) Evolution of the chestnut tree and its blight. Arnoldia (Jamaica Plain) 52:3–10. Anagnostakis SL (2001) American chestnut sprout survival with biological control of the chestnut-blight fungus population. For Ecol Manage 152:225–233. doi: 10.1016/S03781127(00)00605-8 Anagnostakis SL (2012) Chestnut Breeding in the United States for Disease and Insect Resistance. Plant Dis 96:1392–1403. doi: 10.1094/PDIS-04-12-0350-FE Andrade G, Nairn C, Le H, Merkle S (2009) Sexually mature transgenic American chestnut trees via embryogenic suspension-based transformation. Plant Cell Rep 28:1385–1397. doi: 10.1007/s00299-009-0738-7 Attard A, Gourgues M, Galiana E, et al (2008) Strategies of attack and defense in plantoomycete interactions, accentuated for Phytophthora parasitica Dastur (syn. P. Nicotianae Breda de Haan). J Plant Physiol 165:83–94. doi: 10.1016/j.jplph.2007.06.011 Barakat A, DiLoreto DS, Zhang Y, et al (2009) Comparison of the transcriptomes of American chestnut (Castanea dentata) and Chinese chestnut (Castanea mollissima) in response to the chestnut blight infection. BMC Plant Biol 9:51. doi: 10.1186/1471-2229-9-51 Barakat A, Staton M, Cheng C-H, et al (2012) Chestnut resistance to the blight disease: insights from transcriptome analysis. BMC Plant Biol 12:38. doi: 10.1186/1471-222912-38 Bari R, Jones JDG (2009) Role of plant hormones in plant defence responses. Plant Mol Biol 69:473–88. doi: 10.1007/s11103-008-9435-0 Battisti A, Benvegnù I, Colombari F, Haack RA (2014) Invasion by the chestnut gall wasp in Italy causes significant yield loss in Castanea sativa nut production. Agric For Entomol 16:75–79. doi: 10.1111/afe.12036 Blanco Silva R, Fernández-López J (2005) Analysis of genetic variation in Spanish chestnut populations for selecting seed stands. Acta Hortic 431–436. doi: 10.17660/ActaHortic.2005.693.53 Botta R, Vergano G, Me G, Vallania R (1995) Floral Biology and Embryo Development in General introduction 31 Chestnut (Castanea sativa Mill.). HortScience 30:1283–1286. Botta R, Marinoni D, Beccaro G, Akkak A (2001) Development of a DNA typing technique for the genetic certification of chestnut cultivars. Snow Landsc 76:425–248. Bounous G (2005) The chestnut: A multipurpose resource for the new millennium. In: Acta Horticulturae. pp 33–40 Bounous G, Marinoni D (2005) Chestnut: Botany, Horticulture, and Utilization. In: Horticultural Reviews. Willey, pp 291–348 Bounous G (2009) Chestnut industry development and quality of the productions. In: Acta Horticulturae. pp 21–26 Branzanti M, Rocca E, Pisi A (1999) Effect of ectomycorrhizal fungi on chestnut ink disease. Mycorrhiza 9:103–109. doi: 10.1007/s005720050007 Breisch H (1995) Châtaignes et marrons (Sweet chestnuts)., Centre Tec. Paris pp 1-239 Brussino G, Bosio G, Baudino M, Giordano R (2002) Pericoloso insetto esotico per il castagno europeo. Avversità delle piante 37:59–62. Buck EJ, Hadonou M, James CJ, et al (2003) Isolation and characterization of polymorphic microsatellites in European chestnut (Castanea sativa Mill.). Mol Ecol Notes 3:239–241. doi: 10.1046/j.1471-8286.2003.00410.x Carraway C, Wilde H, Merkle S, Warnel D (1994) Somatic embryogenesis and gene transfer in American chestnut. Am Chestnut Found J 8:22–25. Carraway D, Merkle S (1997) Plantlet regeneration from somatic embryos of American chestnut. Can J For 27:1805–1812. doi: 10.1139/x97-123 Carvalho J (2014) Métodos de luta alternativos contra a doença da tinta do castanheiro. Instituto Superior de Agronomia, Universidade de Lisboa Casasoli M, Mattioni C, Cherubini M, Villani F (2001) A genetic linkage map of European chestnut (Castanea sativa Mill.) based on RAPD, ISSR and isozyme markers. Theor Appl Genet 102:1190–1199. doi: 10.1007/s00122-001-0553-1 Casasoli M, Pot D, Plomion C, et al (2004) Identification of QTLs affecting adaptive traits in Castanea sativa Mill. Plant, Cell Environ 27:1088–1101. doi: 10.1111/j.13653040.2004.01214.x Choupina AB, Estevinho L, Martins IM (2014) Scientifically advanced solutions for chestnut ink disease. Appl Microbiol Biotechnol 98:3905–9. doi: 10.1007/s00253-014-5654-2 Coelho AC, Horta M, Neves D, Cravador a. (2006) Involvement of a cinnamyl alcohol dehydrogenase of Quercus suber in the defence response to infection by Phytophthora cinnamomi. Physiol Mol Plant Pathol 69:62–72. doi: 10.1016/j.pmpp.2007.01.001 Coelho AC, Horta M, Ebadzad G, Cravador A (2011) Quercus suber - Phytophthora cinnamomi interaction: A hypothetical molecular mechanism model. New Zeal J For Sci Chapter I 32 41:S143–S157. Conedera M, Manetti MC, Giudici F, Amorini E (2004) Distribution and economic potential of the Sweet chestnut (Castanea sativa Mill.) in Europe. Ecol. Mediterr. 30:179–193. Conedera M, Krebs P (2008) History, present situation and perspective of chestnut cultivation in Europe. In: Acta Horticulturae. pp 23–27 doi: 10.17660/ActaHortic.2008.784.1 Conedera M, Barthold F, Spinedi F, et al (2011) Climatic extremes: an additional threat for the chestnut tree? Sherwood - For ed Alberi Oggi 16–21. Costa R, Ribeiro C, Valdiviesso T, et al (2008) Variedades de Castanha das Regiões Centro e Norte de Portugal., INRB.I.P. Costa R, Santos C, Tavares F, et al (2011) Mapping and transcriptomic approches implemented for understanding disease resistance to Phytophthora cinammomi in Castanea sp. BMC Proc 5:O18. doi: 10.1186/1753-6561-5-S7-O18 Crandall BS, Gravatt GF, Ryan MM (1945) Root disease of Castanea species and some coniferous and broadleaf nursery stocks, caused by Phytophthora cinnamomi. Phytopathology 35:162–80. De Vasconcelos MC, Bennett RN, Rosa EA, Ferreira-Cardoso J V (2010) Composition of European chestnut (Castanea sativa Mill.) and association with health effects: fresh and processed products. J Sci Food Agric 90:1578–1589. doi: 10.1002/jsfa.4016 Dinis LT, Peixoto F, Pinto T, et al (2011) Study of morphological and phenological diversity in chestnut trees (‘Judia’ variety) as a function of temperature sum. Environ Exp Bot 70:110–120. doi: 10.1016/j.envexpbot.2010.08.003 Diskin M, Steiner KC, Hebard F V. (2006) Recovery of American chestnut characteristics following hybridization and backcross breeding to restore blight-ravaged Castanea dentata. For Ecol Manage 223:439–447. doi: 10.1016/j.foreco.2005.12.022 Durrant W, Dong X (2004) Systemic acquired resistance. Annu Rev Phytopathol 42:185–209. doi: 10.1146/annurev.phyto.42.040803.140421 Ebadzad G, Cravador A (2014) Quantitative RT-PCR analysis of differentially expressed genes in Quercus suber in response to Phytophthora cinnamomi infection. Springerplus 3:613. doi: 10.1186/2193-1801-3-613 Elorrieta J (1949) El castaño en España. Instituto Forestal de Investigaciones y Experiencias. Ministerio de Agricultura. Dirección General de Montes, Caza y Pesca Fluvial. Madrid, Ediciones Ares. pp 303 Erwin DC, Olaf KR (1996) Phytophthora diseases worldwide. American Phytopathological Society (APS Press) Eshraghi L, Anderson JP, Aryamanesh N, et al (2014) Defence Signalling Pathways Involved in Plant Resistance and Phosphite-Mediated Control of Phytophthora cinnamomi. Plant General introduction 33 Mol Biol Report 32:342–356. doi: 10.1007/s11105-013-0645-5 Fang G-C, Blackmon BP, Staton ME, et al (2013) A physical map of the Chinese chestnut (Castanea mollissima) genome and its integration with the genetic map. Tree Genet Genomes 9:525–537. doi: 10.1007/s11295-012-0576-6 Fawke S, Doumane M, Schornack S (2015) Oomycete Interactions with Plants: Infection Strategies and Resistance Principles. Microbiol Mol Biol Rev 79:263–280. doi: 10.1128/MMBR.00010-15 Fernandes CT (1955) A luta contra a doença da tinta nos soutos do norte de Portugal e ensaios diversos para a sua maior eficiência e economia., Direcção-G. Publicações da Direcção-Geral dos Serviços Florestais e Aquícolas, Vol. XXII., Lisboa Fernández-Cruz J, Fernández-López J (2012) Morphological, molecular and statistical tools to identify Castanea species and their hybrids. Conserv Genet 13:1589–1600. doi: 10.1007/s10592-012-0408-0 Fernández-López J, Fernández-Cruz J (2015) Identification of traditional Galician sweet chestnut varieties using ethnographic and nuclear microsatellite data. Tree Genet Genomes 11:111. doi: 10.1007/s11295-015-0934-2 Fernández-Cruz J (2015) Identificación de especies e híbridos de castaño y descripción de la estructura genética de poblaciones de Castanea sativa mediante microsatélites. Universida de Vigo Fernández-Cruz J, Fernández-López J (2016) Genetic structure of wild sweet chestnut (Castanea sativa Mill.) populations in northwest of Spain and their differences with other European stands. Conserv Genet 17:949–967. doi: 10.1007/s10592-016-0835-4 Fleischmann F, Gottlein A, Rodenkirchen H, et al (2004) Biomass, nutrient and pigment content of beech (Fagus sylvatica) saplings infected with Phytophthora citricola, P. cambivora, P. pseudosyringae and P. undulata. For Pathol 34:79–92. doi: 10.1111/j.1439-0329.2004.00349.x Fleischmann F, Koehl J, Portz R, et al (2005) Physiological changes of Fagus sylvatica seedlings infected with Phytophthora citricola and the contribution of its elicitin ‘citricolin’ to pathogenesis. Plant Biol (Stuttg) 7:650–658. doi: 10.1055/s-2005-872891 Gentile S, Valentino D, Tamietti G (2009) Effectiveness of potassium phosphite in the control of chestnut ink disease. In: I European Congress on Chestnut. Gomes-Laranjo J, Peixoto F, Ferreira-Cardoso J, (2009) Castanheiros Técnicas e Práticas. Trás-os-Montes, Portugal: Pulido Consulting – Indústria Criativa & Universidade de Trás-osMontes e Alto Douro. González M, Cuenca B, López M (2011) Molecular characterization of chestnut plants selected for putative resistance to Phytophthora cinnamomi using SSR markers. Sci Chapter I 34 Hortic (Amsterdam) 130:459–497. doi: doi:10.1016/j.scienta.2011.07.020 Gouveia ME (2004) Métodos moleculares na identificação, caracterização e detecção de Phytophthora cambivora (Petri) Buisman e Phytophthora cinnamomi Rands associadas com a doença da tinta do castanheiro. UTAD. Vila Real Griffin BGJ (2000) Blight Control and Restoration of the American Chestnut. J For 98:22–27. Guerreiro M (1948) Alguns estudos do género Castanea. Alcobaça. Direcção Geral dos Serviços Florestais e Aquícolas. Guerreiro M (1957) Castanheiros. Instituto Superior de Agronomia. Instituto Superior de Agronomia Hardham AR, Cahill DM, Cope M, et al (1994) Cell surface antigens of Phytophthora spores: biological and taxonomic characterization. Protoplasma 181:213–232. doi: 10.1007/BF01666397 Hardham AR (2005) Phytophthora cinnamomi. Mol Plant Pathol 6:589–604. doi: 10.1111/j.1364-3703.2005.00308.x Hardham A, Blackman L (2010) Molecular cytology of Phytophthora-plant interactions. Australas Plant Pathol 39:29–35. doi: 10.1071/AP09062 Hebard F, Sisco P, Brinckman M, et al (2014a) How a flower becomes a chestnut: morphological development of Chinese chestnuts (Castanea mollissima). J. Am. Chestnut Found. 13–18. Hebard F V., Fitzsimmons SF, Gurney KM, Saielli TM (2014b) The breeding program of the American chestnut foundation. In: Acta Horticulturae. pp 135–140 Herrera-Vásquez A, Salinas P, Holuigue L (2015) Salicylic acid and reactive oxygen species interplay in the transcriptional control of defense genes expression. Front Plant Sci 6:171. doi: 10.3389/fpls.2015.00171 Hüberli D, Tommerup IC, Hardy GESJ (1997) The role of paragynous and amphigynous antheridia in sexual reproduction of Phytophthora cinnamomi. Mycol Res 101:1383– 1388. doi: 10.1017/S0953756296003413 Huitema E, Bos JIB, Tian M, et al (2004) Linking sequence to phenotype in Phytophthoraplant interactions. Trends Microbiol 12:193–200. doi: 10.1016/j.tim.2004.02.008 Inoue E, Ning L, Hara H, Agriculture C (2009) Development of simple sequence repeat markers in chinese chestnut and their characterization in diverse chestnut cultivars. 134:610–617. Jacobs DF, Dalgleish HJ, Nelson CD (2015) Synthesis of American chestnut (Castanea dentata) biological, ecological, and genetic attributes with application to forest restoration. Forest Health Initiative. pp:25 Jones J, Dangl J (2006) The plant immune system. Nature 444:323–329. doi: General introduction 35 10.1038/nature05286 Jung T, Colquhoun IJ, Hardy GESJ (2013) New insights into the survival strategy of the invasive soilborne pathogen Phytophthora cinnamomi in different natural ecosystems in Western Australia. For Pathol 43:266–288. doi: 10.1111/efp.12025 Kamoun S, Huitema E, Vleeshouwers V (1999) Resistance to oomycetes: a general role for the hypersensitive response? Trends Plant Sci 4:196–200. doi: 10.1016/S13601385(99)01404-1 Kamoun S, Furzer O, Jones JDG, et al (2014) The Top 10 oomycete pathogens in molecular plant pathology. Mol Plant Pathol 16:413–34. doi: 10.1111/mpp.12190 Ko WH, Chang HS, Su HJ (1978) Isolates of Phytophthora cinnamomi from Taiwan as evidence for an Asian origin of the species. Trans Br Mycol Soc 71:496–499. doi: 10.1016/S0007-1536(78)80080-1 Konstantinidis P, Tsiourlis G, Xofis P, Buckley GP (2008) Taxonomy and ecology of Castanea sativa Mill. forests in Greece. Plant Ecol 195:235–256. doi: 10.1007/s11258-007-93238 Kremer A, Abbott AG, Carlson JE, et al (2012) Genomics of Fagaceae. Tree Genet Genomes 8:583–610. doi: 10.1007/s11295-012-0498-3 Kremer A, Casasoli M, Barreneche T, et al (2007) Fagaceae Trees. In: Forest Trees. Springer Berlin Heidelberg, Berlin, Heidelberg, pp 161–187 Kubisiak T, Hebard F, Nelson C (1997) Molecular mapping of resistance to blight in an interspecific cross in the genus Castanea. Phytopathology 87:751–759. doi: 10.1094/PHYTO.1997.87.7.751 Kubisiak TL, Nelson CD, Staton ME, et al (2013) A transcriptome-based genetic map of Chinese chestnut (Castanea mollissima) and identification of regions of segmental homology with peach (Prunus persica). Tree Genet Genomes 9:557–571. doi: 10.1007/s11295-012-0579-3 Lang P, Dane F, Kubisiak TL, Huang H (2007) Molecular evidence for an Asian origin and a unique westward migration of species in the genus Castanea via Europe to North America. Mol Phylogenet Evol 43:49–59. doi: 10.1016/j.ympev.2006.07.022 Latijnhouwers M, de Wit PJGM, Govers F (2003) Oomycetes and fungi: similar weaponry to attack plants. Trends Microbiol 11:462–469. doi: 10.1016/j.tim.2003.08.002 Li AY, Crone M, Adams PJ, et al (2014) The Microscopic Examination of Phytophthora cinnamomi in Plant Tissues Using Fluorescent In Situ Hybridization. J Phytopathol 162:747–757. doi: 10.1111/jph.12257 Marinoni D, Akkak A, Bounous G, et al (2003) Development and characterization of microsatellite markers in Castanea sativa (Mill.). Mol Breed 11:127–136. doi: 42 Phenotyping Castanea hybrids for P. cinnamomi resistance 43 Chapter II Phenotyping Castanea hybrids for Phytophthora cinnamomi resistance Part of the work presented in this chapter was published in the following research publication: Santos C., Machado H., Correia I., Gomes F., Gomes-Laranjo J. and Costa R. (2015) Phenotyping Castanea hybrids for Phytophthora cinnamomi resistance. Plant Pathol. 64, 901–910. doi: 10.1111/ppa.12313 In this research paper Carmen Santos participated in the experimental design, laboratory experiments, results interpretation and paper writing. Chapter II 44 Abstract Castanea sativa is susceptible to Phytophthora spp., a serious root pathogen causing ink disease, while C. crenata and C. mollissima show resistance to infection. Interspecific controlled crosses were produced to introgress resistance genes from the resistant species into the susceptible C. sativa, and three mapping populations were created. Phytophthora cinnamomi resistance of the progenies C. sativa x C. crenata and C. sativa x C. molissima were evaluated by root and/or excised shoot inoculation tests. The number of days of survival after root inoculation was the best discriminator of resistance to P. cinnamomi while the percentage of shoots with internal lesions was the symptom most associated with survival. The lesion progression rate in the excised shoot inoculation test was strongly and negatively correlated with survival in the root inoculation test. The excised shoot inoculation test appears to be a reliable approach for screening the resistance of chestnut genotypes to P. cinnamomi. Therefore, a recently obtained progeny (in 2015) was phenotyped using the excised shoot inoculation test. Strong genetic correlations were obtained between survival and ink disease symptoms and among symptoms, indicating that common or linked genes might influence resistance to P. cinnamomi. The most resistant genotypes selected from this study will be tested for other commercial variables, such as ease of vegetative propagation and stock–scion compatibility. Keywords: Castanea hybrids, heritability, phenotypic and genetic correlations, Phytophthora cinnamomi Introduction The genus Castanea belongs to Fagaceae, a plant family that dominates much of the climax hardwood forests of the Northern Hemisphere (Manos et Phenotyping Castanea hybrids for P. cinnamomi resistance 45 al. 2008). The European chestnut (Castanea sativa) is considered to be the only native species in Europe. Chestnuts are multipurpose trees being used in the food industry, for its edible nuts, in the wood industry, as timber and also for ecological and landscaping purposes, having a major economic importance in the Mediterranean region. Chestnut fruit production has declined considerably in southwestern Europe due to social changes and cultural development, and particularly to the emergence of heavily damaging diseases. Ink disease, caused by Phytophthora spp. is one of the most destructive diseases affecting Castanea sativa. Phytophthora cinnamomi is an aggressive root pathogen, originally from the southeast Asian tropics (Hardham 2005). Nowadays, P. cinnamomi is widespread and continues to be destructive in forests of Mediterranean countries, Australia, southeast USA, southern California and more recently it was recognized as a danger to forests in western North America (Robin et al. 2012). Phytophthora cinnamomi has an exceptionally wide host range, being able to invade more than 3 000 plant species around the world (Hardham 2005; Cahill et al. 2008). Currently, it is the most important Phytophthora pathogen of forest trees; besides chestnut, P. cinnamomi causes root diseases in eucalyptus, oaks, pines and members of the Ericaceae family, as well as, several agricultural crops (Robin et al. 2012). Disease symptoms in chestnut are similar to other species: ink disease causes root rot, with necrosis of tap root, which extends to the lateral roots and the collar. P. cinnamomi infection induces necrosis of the cambial and xylem tissues, causing interference with transpiration from roots to shoots, and consequently causes wilting of leaves and dieback of young shoots (Marçais and Dupuis 1996; Robin et al. 2001; Vannini and Vettraino 2001; Hardham 2005; Gomes-Laranjo et al. 2009). The pathogen spreads slowly through root-to-root contact and more rapidly in presence of water. Human activities that move soil and the planting of Chapter II 46 infested nursery stock intensify pathogen spread (Robin et al. 2012). With changing climates, P. cinnamomi is expected to expand its area of destruction, mainly in Europe and North America (Robin et al. 2012; Thompson et al. 2014). In Portugal, ink disease has become widespread, since P. cinnamomi was first recorded, in 1838. Despite the protection measures taken, it is still a great threat to chestnut orchards, as fruit and timber production is negatively impacted. Chestnut production is an important source of income for rural populations and so new plantings have been carried out while old orchards are being restored. In Europe, chestnut breeding for ink resistance began with the introduction of the Asian chestnut germplasm, which is resistant to the main diseases: ink and blight (caused by Cryphonectria parasitica). Japanese and Chinese species (Castanea crenata and Castanea mollissima, respectively) were introduced since 1917 in several southern European countries (Elorrieta, 1949). However, the low value of the Asian species as timber and fruit producers was notable. They also presented low compatibility for grafting with local sweet chestnut varieties (Elorrieta, 1949). In Portugal, the first interspecific hybridizations were initiated in 1948 by Bernardino Barros Gomes to introduce resistance to ink disease in C. sativa (Guerreiro, 1948; Guerreiro, 1957). The objectives of these programs were to breed for resistance to ink disease, as well as to produce rootstock or varieties selected for early nut production or better wood production as compared with Asian species (Fernández-López 2011). In 2006, interspecific controlled crosses were performed between C. sativa and C. crenata (SC) and between C. sativa and C. mollissima (SM) in order to introgress the resistance from Asian species into the European (Costa et al. 2011). The main goal of our ongoing program was to produce a hybrid segregant population to perform DNA marker-phenotype association analysis to identify genomic regions related with the ink disease resistance (Quantitative Trait Loci, QTL). For this purpose, it is crucial to determine Phenotyping Castanea hybrids for P. cinnamomi resistance 47 accurately both genotype and phenotype of each hybrid progeny. However, there are some limitations for the determination of the resistance of Castanea spp. to P. cinnamomi. Different methods have been used used for screening the resistance of chestnut to Phytophthora spp. by different authors.: i) root inoculation using seedlings (Vettraino et al. 2001; Santini et al. 2003; Robin et al. 2006; Jeffers et al. 2009); ii) root inoculation using cuttings (Miranda-Fontaiña et al. 2007) or iii) plantlets from micropropagation (Cuenca et al. 2009) and iv) direct inoculation on the top of excised or intact stem/shoot from seedlings or clones selected in the field (Guedes-Lafargue & Salesses 1999; FernándezLópez et al. 2001; Vettraino et al. 2001b; Robin et al. 2006; Miranda-Fontaíña et al. 2007; Cuenca et al. 2009). There are advantages and drawbacks to each method; in particular, shoot inoculation is easy to achieve and enables the screening of a high number of individuals at low cost (Fernández-López 2011), but a criticism of this method is that P. cinnamomi is a root pathogen. The objectives of the present study were (i) to select the best resistance discriminators from root and excised shoots inoculation tests and clarify their correlations; and (ii) to assess the resistance to P. cinnamomi and evaluate its heritability in progenies of three Castanea sp. segreganting populations. Material and Methods Plant material Three full-sib progenies were obtained from artificial controlled crosses: C. sativa (cultivar Aveleira) x C. crenata2 (SC), C. sativa (cultivar Aveleira) x C. mollissima (SM) and C. sativa (cultivar Bária) x C. crenata1 (BC) in 2006, 2009, 2012 and 2015. C. sativa female flowers were isolated before pollination season by placing pollination bags on the branches and cutting off the closer catkins. The parental line of the C. sativa (cultivar Aveleira) was the same for crosses with C. crenata2 and C. mollissima. Catkins from C. crenata and C. mollissima were collected and dried the day before Chapter II 48 pollination. On the day of pollination, pollen was removed from catkins and filtered. Then, pollen was placed on the stigmas using a paintbrush or a piece of glass (only the pollen grains are adhered to the glass). Pollinated flowers were covered with paper or polyester pollination bags or and kept until the end of pollination season. Subsequenlty, pollination bags were replaced by net bags in order to collect the seeds. The crosses were performed at the germplasm bank of Universidade de Trás-os-Montes e Alto Douro, Vila Real for SC and SM crosses, and in a private orchard in Marvão for BC crosses, both in Portugal. A total of 142 F1 genotypes were tested for P. cinnamomi infection by either root inoculation test or excised shoot inoculation tests or by both (Table 1). For root inoculation test, 137 plantlets were produced from 20 genotypes by in vitro propagation from buds of mother plants. At the time of inoculation, plantlets were different ages, as determined by the number of days after acclimatization, but were most frequently 80 days old; aerial parts were 16.66 cm on average. Table 1. Number of individuals from Castanea sativa (cultivar Aveleira) x C. crenata2 (SC), C. sativa (cultivar Aveleira) x C. mollissima (SM) and C. sativa (cultivar Bária) x C. crenata1 (BC) crosses tested by root inoculation and excised shoot inoculation. Test method SC SM BC Total Root inoculation 16 4 0 20 Excised shoot inoculation (total) 45 18 76 139 Excised shoot inoculation (spring) 30 17 0 47 Excised shoot inoculation (autumn) 42 18 76 136 Total 48 18 76 142 The excised shoot inoculation tests were carried out in spring and autumn of 2012 for a total of 63 SC and SM progenies. For BC progenies, the excised shoot inoculation tests was performed in autumn of 2016 (Table 1). A total number of 1034 shoots were collected from the mother plants. Both Phenotyping Castanea hybrids for P. cinnamomi resistance 49 experiments were performed in a controlled chamber with temperatures ranging between 18 and 22ºC, photoperiod 16h light/8h dark and 65% of relative humidity. Inoculum of P. cinnamomi In all experiments, the same isolate of P. cinnamomi was used (IMI 340340), which was selected, as the most virulent, following tests using several isolates (Abreu et al. 1999). The high pathogenicity of this isolate in European chestnuts was also confirmed by Dinis et al. (2011). For the root inoculation test, the P. cinnamomi inoculum was prepared by growing mycelia on sterilized millet seeds (Ponicium mileaceum), which were thoroughly moistened with V8 medium broth [20% (v/v) with 3 g/L of CaCO3]. Afterwards, this mixture was incubated for 3 weeks in darkness at 24°C. For the excised shoot inoculation test, P. cinnamomi was grown on potato dextrose agar for 6 days in darkness at 22ºC. Root inoculation test Four experiments of root inoculation were carried out using clonal plantlets placed in sterile substrate. For each experiment, one or two plantlets of each genotype, were used as a control, without inoculation. For root inoculation, P. cinnamomi-infected millet seed inoculum was carefully placed into the substrate (600 mL) of each pot, at a concentration of 5% (v/v). Mostly, eight replicates per genotype were inoculated, but, due to limitations of in vitro propagation, this was not alwalys possible. Therefore, the mean number of plants per genotype was 6.85. Inoculated plants and controls were placed separately in different trays and each pot was flooded for 1h, three times a week, to stimulate zoospore release and to promote disease development. P. cinnamomi was recovered from water collected from flooding process, using a modified baiting technique adapted from Jung et al. (1996). Chapter II 50 The experimental design was adapted from Miranda-Fontaíña et al. (2007). For each individual, the days of survival after inoculation were recorded, until 100 days after inoculation (dai). After death, plantlets were removed from the soil and the roots were gently washed to observe and record ink disease symptoms.The level of root rot was assessed on a scale from 1 to 6, according to Miranda-Fontaíña et al. (2007), where 1 indicates the least severe level of root rot and 6 indicates the most severe level. The percentage was then used to rate the level of root collar rot on a scale of 1-6 (1, no rot; 2, 0.1-9.9% rot; 3, 10-19.9% rot; 4, 20-29.9% rot; 5, 30-49.9% rot and 6, >50% rot). Shoot internal and external lesions were recorded as the percentage length of internal and external lesion of the longest shoot, respectively. Biomass parameters were also evaluated for each plantlet: leaf and shoot dry weight (g) and root dry weight (g). In order to determine the dry weight, leaves and shoots were separated from roots and both parts were dried at 60ºC for 2 days. At the end of each experiment, plantlets that did not die during the test were analysed without destruction and were transplanted to new pots that were placed in a greenhouse with controlled conditions (20-25ºC). In the spring of the next year, the number of plantlets that showed budburst was recorded. Excised shoots inoculation test Excised shoot inoculation test for SC and SM populations took place in spring, using 47 genotypes (eight genotypes were common to the root inoculation test), and in autumn, using 60 genotypes (16 common to the root inoculation test). The 76 progenies from BC cross were screnned in autumn, 2015 (Table 1). Excised shoots, the majority with a length of 15 cm, were collected from each mother plant. The mean number of excised shoots inoculated per genotype was 7.94 in spring and 4.86 in autumn. All apart from two upper leaves were removed to reduce evapotranspiration. The diameter of the top of each excised shoot was recorded before inoculation. Phenotyping Castanea hybrids for P. cinnamomi resistance 51 Mycelial plugs of P. cinnamomi were then placed on the top of the shoots and were covered with an aluminium sheet to avoid desiccation. Replicates were distributed randomly in three trays with perlite and water, in an environmental controlled chamber. Five days after inoculation, the aluminum sheets were removed from each shoot, when colonization by the pathogen had occurred. Resistance to P. cinnamomi was evaluated by measuring the visible external lesion length (LL) at 5, 7, 9, 12 and 14 dai. The lesion progression rate (cm/day) was calculated for each genotype, using the following formula: ( 𝐿𝐿5 𝑑𝑎𝑖 5)+( 𝐿𝐿7 𝑑𝑎𝑖 − 𝐿𝐿5 𝑑𝑎𝑖 2)+( 𝐿𝐿9 𝑑𝑎𝑖 − 𝐿𝐿7 𝑑𝑎𝑖 2)+( 𝐿𝐿12 𝑑𝑎𝑖 − 𝐿𝐿9 𝑑𝑎𝑖 3)+( 𝐿𝐿14 𝑑𝑎𝑖 − 𝐿𝐿12 𝑑𝑎𝑖 2) 5 Statistical analysis Analysis of variance (ANOVA) was conducted for root and shoot variables using linear mixed effects models of the general form y = Xβ + Zγ + ε, where y is the vector of observations; X and Z are design matrices of the parameters associated to fixed and random effects, respectively; β and γ are vectors of fixed effects (including the general mean) and random effects, respectively; and ε is the vector of residual errors. In preliminary data analyses, resistance from SC and SM crosses was evaluated separately for each cross by specifying a two-level ‘Family’ effect (SC and SM). Because no significant differences were found between the two full-sib families, the family effect was dropped from the model. For the root inoculation test data, Age, Genotype, Inoculation, Inoculation Date and Genotype x Inoculation Date were fitted as fixed effects (β vector); for the excised shoot inoculation test data, Length, Genotype, Season and Genotype x Season were treated as fixed effects. Ftests were used to test the significance of the fixed effects and of genotype mean comparisons, the latter implementing the Tukey adjustment and a matching letter display (adapted from Piepho, 2012). If significant, the Chapter II 58 lesion was lower than shoot internal lesion, ranging from 4.16% (SM901) to 26.63% (SC918) (Table 3). The phenotypic and genetic correlation coefficients were estimated for all the variables recorded on inoculated plantlets. Both correlation coefficients showed a great similarity among pairs of variables, in regard to direction and to magnitude (Table 4). The phenotypic correlations showed that survival had highly significant negative correlations with three of the four ink disease symptoms analysed: level of root collar rot, shoot internal lesion and shoot external lesion. Both phenotypic and genetic correlations showed that shoot internal lesion was the main symptom negatively associated to survival. The level of root rot was the least important symptom associated to survival, with nonsignificant phenotypic correlation (Table 4). The phenotypic correlations evaluated between survival and the biomass parameters were positive and highly significant (Table 4). Correlations were lower for leaves and shoots than for roots, while the symptoms and biomass parameters were not strongly correlated. The phenotypic correlation coefficients observed between symptoms were positive and highly significant, especially level of root collar rot with shoot external lesion, followed by shoot external lesion with shoot internal lesion. The weakest correlated symptoms were level of root rot with shoot external lesion. The highest genetic correlations were found among symptoms and among biomass parameters. The genetic correlation coefficients among symptoms ranged from 0.95 to 1.00 (level of root collar rot with shoot external lesion), in agreement with phenotypic correlations. Phenotyping Castanea hybrids for P. cinnamomi resistance 59 Days of survival Leaf and shoots dry weight Root dry weight Level of root collar rot Level of root rot Shoot external lesion Shoot internal lesion Age 0.31*** 0.38*** 0.60*** -0.01 -0.02 -0.05 -0.22* Days of survival 0.43*** 0.50*** -0.42*** -0.16 -0.44*** -0.69*** Leaf and shoots dry weight 0.51 0.75*** 0.02 0.18 -0.13 -0.26* Root dry weight 0.64 0.99 0.02 0.18 -0.10 -0.24* Level of root collar rot -0.88 -0.08 -0.23 0.49*** 0.76*** 0.61*** Level of root rot -0.86 -0.05 -0.21 1.00 0.40*** 0.44*** Shoot external lesion -0.90 -0.13 -0.29 1.00 1.00 0.66*** Shoot internal lesion -0.97 -0.36 -0.50 0.96 0.95 0.97 Table 4. Phenotypic (above the diagonal) and genetic (below the diagonal) correlation coefficients determined between pairs of the variables measured in root inoculation test (N=108 to 137). Phenotypic correlations were analysed with the Spearman’s correlation coefficient. Genetic correlations were adjusted for the covariate Age effect (Age=80 days after acclimatization). Significance is indicated by asterisks: *, P < 0.05; **, P < 0.01; and ***, P < 0.001. Chapter II 60 The heritability values for chestnut resistance to P. cinnamomi varied between 0.34 and 0.90, with low standard errors. Survival showed the highest heritability (0.90 ± 0.04) with low residual variance and thus the highest potential to be inherited. Among symptoms, the highest heritability value was obtained for shoot internal lesion whereas shoot external lesion showed the lowest value, with the lowest variance explained by both components, genetic and residual (Table 5). Table 5. Narrow-sense heritabilities (h2) and their standard errors (in parentheses) estimated for the variables (root inoculation test). Variables Variance components Narrow-sense heritabilities (h2) Additive genetic (σa 2) Residual (σe 2) Days of survival 0.92 (0.34) 0.10 (0.01) 0.90 (0.04) Level of root rot 0.03 (0.02) 0.04 (0.01) 0.46 (0.16) Level of root collar rot 0.25 (0.11) 0.22 (0.03) 0.54 (0.11) Shoot internal lesion 0.45 (0.19) 0.15 (0.02) 0.75 (0.09) Shoot external lesion 0.01 (0.01) 0.03 (0.00) 0.34 (0.14) n=108-137. Excised shoots inoculation At 5 dai in both seasons, P. cinnamomi had induced visible necrotic lesions of varying length, depending on the genotype. At 14 dai a very low percentage (0.36%) of shoots did not show any lesion and 8.80% of shoots showed total necrosis (100% of lesion length). Analysis of variance revealed that lesion length was significantly affected by Genotype, which had the strongest effect, and also by Season and the interaction of both. In contrast to diameter, the Initial Shoot Length had a significant effect on the lesion length, and therefore was used as a covariate effect in the analysis of SC and SM data. Further analysis showed that differences observed between genotypes for lesion length were very high for all time points of measurements and in both Phenotyping Castanea hybrids for P. cinnamomi resistance 61 seasons, when applicable. Moreover, differences observed between genotypes (SC and SM progenies) for lesion length were maximal at 5 dai (F=19.80, P < 0.001) in spring (F=33.11, P < 0.001) after bud burst. Different responses to P. cinnamomi were observed in the progenies: a continuous range of resistance-susceptibility levels among genotypes was observed. For the majority of genotypes, the lesion length in the shoots increased over time. In addition, for the most resistant genotypes the lesion length stopped at a given time point, until the end of the experiment. Therefore, the lesion progression rate (cm/day) was calculated for each genotype. Genotype mean values (SC and SM), estimated across the two seasons and adjusted for the covariate Initial Length, are shown in Figure 2. The lesion progression rate varied from 0.15 to 1.13 cm per day across genotypes and seasons; SC57 was the most resistant genotype while SC915 the genotype with the greatest lesion progression rate (i.e. most susceptible). Concerning BC progenies, lesion progression rates obtained in autumn 2015 were also ranged from the most susceptible (BCC01) to the most resistant (BDC40) (Supplementary material 1). This population seems to be more resistant to P. cinnamomi than SC and SM, since lesion progression rate varied from 0.11 to 0.89 cm per day. Correlations between inoculation tests The estimation of phenotypic and genetic correlations between the root inoculation test and excised shoots inoculation test was possible using the mean response of common genotypes in both tests (n=17). In this way, some differences were obtained in the phenotypic and genetic correlations between survival and symptoms (Table 4 and 6). Chapter II 62 Figure 2. Mean values in lesion progression rate, adjusted for the covariate Initial Length, for 63 genotypes tested with excised shoot inoculation. The genotypes were ranked by lesion progression rate, from the most resistant to the most susceptible. SC915 genotype showed the highest susceptibility to Phytophthora cinnamomi whereas SC57 was the most resistant. Bars represent standard errors. Phenotyping Castanea hybrids for P. cinnamomi resistance 63 Phenotypic correlations obtained among all variables from both inoculation tests showed that lesion progression rate was the parameter with the strongest correlation with survival (r= -0.85, P < 0.001). Shoot internal lesion was correlated moderately with lesion progression rate, as well as level of root collar rot (Table 6). Genetic correlation coefficients were in agreement with phenotypic correlations and the heritability for lesion progression rate was 0.67 ± 0.04. Table 6. Phenotypic (above the diagonal) and genetic (below the diagonal) correlation coefficients determined between pairs of the variables measured in root inoculation test: Days of survival, Level of root collar rot, Level of root rot, Shoot external lesion and Shoot internal lesion; and between the variables measured in root inoculation test and Lesion progression rate from excised shoot inoculation test (n=17). Days of survival Level of root collar rot Level of root rot Shoot external lesion Shoot internal lesion Lesion progression rate Days of survival -0.63** -0.36 -0.47 -0.74** -0.85*** Level of root collar rot -0.67 0.59* 0.58* 0.56* 0.62* Level of root rot -0.56 0.82 0.09 0.36 0.44 Shoot external lesion -0.45 0.62 -0.05 0.75*** 0.45 Shoot internal lesion -0.87 0.44 0.39 0.66 0.67** Lesion progression rate -0.83 0.62 0.73 0.34 0.75 Significance is indicated by asterisks: *, P < 0.05; **, P < 0.01; and ***, P < 0.001. Discussion This study addresses two types of inoculation test used to determine the response of individual chestnut plantlets from three mapping populations to inoculation with P. cinnamomi. Lesion progression rate was the variable selected to perform DNA marker–variable association, for QTL identification (Chapter V). Chapter II 64 Although more than one isolate of P. cinnamomi is commonly used in this type of study, only one isolate was used in the present investigation in order to maximize the number of replicates of each chestnut genotype screened, thus making the analysis more robust and accurate. Several previous studies of P. cinnamomi on chestnut (Abreu et al. 1999; Dinis et al. 2011) enabled the most virulent isolate to be selected for the present investigation. Frampton et al. (2013) also used a single isolate of P. cinnamomi in soil inoculation of Abies spp. seedlings. Moreover, Fernández-Lóopez et al. (2001) and Miranda-Fontaiña et al. (2007) observed that there was no significant interaction between isolates and genotypes in the rot symptoms evaluated, indicating no specificity of those isolates in chestnut. In previous studies, the origin and physiological conditions of plant material, replicate number, time point of lesion measurements and test conditions varied for both inoculation tests (Vettraino et al. 2001a,b; Robin et al. 2006; Miranda-Fontaiña et al. 2007; Cuenca et al. 2009). Therefore, in the present study, the experiments were designed in order to obtain the most reliable results possible: a high number of clonal plantlets per genotype and the use of the same controlled environmental conditions for both inoculation tests. Clonal testing of progeny from mapping populations is the most efficient way to minimize the effect of environmental variation and obtain better estimates of the phenotypic value (Bradshaw & Foster 1992). Minimizing environmental variation and therefore increasing heritability, increases the robustness and the ability for QTL detection. In the root inoculation test, control plantlets grew more than inoculated plantlets, as expected. The lesions in roots and shoots caused by P. cinnamomi may hinder water and nutrient absorption, and as a consequence, cause a reduction in the photosynthesis rate and growth. Robin et al. (2006) and Miranda-Fontaiña et al. (2007) also reported a reduction in growth of chestnut plants inoculated with P. cinnamomi. Cahill et al. (1989) observed Phenotyping Castanea hybrids for P. cinnamomi resistance 65 that P. cinnamomi inoculation stopped root growth in a group of plant species within 24–48 h. Previous studies have indicated that days of survival should be the main discriminator of Phytophthora spp. resistance in chestnut (Vettraino et al. 2001a), in Abies spp. (Frampton et al. 2013) and in Eucalyptus spp. (Stukely & Crane, 1994). However, other authors have considered the level of root or collar rot as the main indicator of resistance to Phytophthora spp. in chestnut (Robin et al. 2006; Miranda-Fontaiña et al. 2007; Cuenca et al. 2009). The present study showed that variable ‘Days of survival’ was the most important indicator of resistance to P. cinnamomi because differences in response between genotypes were maximized; the presence of high levels of root and collar rot in almost all inoculated plantlets indicated that these symptoms were not good discriminators for resistance. Miranda-Fontaiña et al. (2007) also reported high levels of root and collar rot in a high percentage of chestnut plants. Likewise, Cuenca et al. (2009) observed root rot in 60% of the resistant C. crenata plants. Survival has not been considered as the main descriptor of resistance to P. cinnamomi in chestnut, mainly because of the high mortality of control chestnut plants before and during the experiment, due to biotic or abiotic factors or cross contamination (Miranda-Fontaiña et al. 2007; Cuenca et al. 2009). In this study, all control plantlets survived until the end of the experiment, indicating that P. cinnamomi cross contamination and other biotic and abiotic stresses, such as flooding, did not occur. In future, similar studies should include preventive measures, such as the use of sterile substrates and avoiding excess flooding time during the experiments. Twenty to thirty minutes flooding is sufficient for P. cinnamomi release zoospores and cysts to germinate (Hardham, 2005). In summary, long survival was considered evidence of high resistance and thus seven genotypes (35%) were selected as the most resistant. Chapter II 66 In this study, shoot internal lesion was evaluated for the first time as a parameter to assess chestnut resistance to P. cinnamomi. It was chosen because it indicates the spread of the pathogen from the roots and collar to the aerial vascular system. This is important for determining the degree of plant resistance, as the rapid invasion of the pathogen into the phloem and xylem may affect water and nutrient movement through the shoots, causing death. With regard to biomass parameters, the phenotypic correlation was stronger between survival and root dry weight than with leaf and shoot dry weight. Cuenca et al. (2009) also observed a good correlation between fresh root weight and survival. A healthy and developed root system is an important factor for resistance to P. cinnamomi. The lesion caused by the inoculation of excised shoots is considered to be an indirect measure of Phytophthora spp. resistance. The length of the lesion is negatively proportional to resistance to the pathogen (Fernández-López et al. 2001). The results showed that, similar to root-inoculated plants, the resistance to P. cinnamomi in the shoots is related to the confinement of the lesion to point of inoculation. For the most resistant genotypes, the surrounding tissues dried, limiting the progression of the lesion. In the present study, differences observed between genotypes for lesion length were very high for all time points of measurement and in both seasons. Nevertheless, the results revealed that spring was the better season to perform excised shoot inoculation tests. After budburst (in spring), plants have good physiological conditions that may allow a better resistance response. It was found that the best time to take measurements in future investigations would be 5 dai, when differences in lesion lengths between genotypes were maximized. The present study has shown that the strongest and most significant phenotypic and genetic correlations were obtained for lesion progression Phenotyping Castanea hybrids for P. cinnamomi resistance 67 rate and survival; therefore, these would be the best variables to measure in future investigations. The estimation of heritabilities and genetic correlations (genetic parameters) is an important strategy for plant breeding. Phenotypic variables with higher heritabilities (in this study: survival, shoot internal lesion and lesion progression rate) have the potential to be inherited to varying degrees in populations exposed to differential natural selection pressures in distinct environments (White et al. 2007). Resistance to P. cinnamomi is a polygenic and quantitative trait (Irwin et al. 1995) that was here evaluated by measuring several variables. The strong genetic correlations observed between survival and symptoms suggest common genetic determinants. Similarly, survival had a strong genetic correlation with lesion progression rate in the excised shoot inoculation test. The assessment of survival by root inoculation testing is expensive and laborious and cannot always be determined in a population. The strong favourable genetic correlation observed between the two variables suggests that indirect selection could be made by lesion progression rate, which is easily measured. Therefore, the C. sativa x C. crenata population obtained in 2015 (BC) was phenotyped using the excised shoot inoculation test in autumn. Preliminary results indicate a higher level of resistance when compared with SC and SM populations. However, excised shoot inoculation test shall be repeated in spring to obtain lesion progession rates across both seasons. Then, mother plants will be root-inoculated with P. cinnamomi to identify the most resistant genotypes (survivors to the inoculation). Finally, those genotypes will be established for in vitro culture for further phenotype validation by root inoculation test. Acknowledgments We acknowledge Helen Beltrame and Patrícia Figueiredo for the establishment, multiplication, rooting and acclimatization of chestnut Chapter III 74 Abstract The European chestnut, an important forest species for the economy of Southern Europe, covers an area of 2.53 million hectares, including almost 110 000 hectares devoted to fruit production. Castanea sativa is declining due to ink disease caused by Phytophthora cinnamomi. To elucidate chestnut defense mechanisms to ink disease we compared the root transcriptome of the susceptible species C. sativa and the resistant species C. crenata after P. cinnamomi inoculation. Four cDNA libraries were constructed, two of them included root samples from C. sativa, inoculated and non-inoculated and the other two libraries comprised samples from C. crenata at identical conditions. Pyrosequencing produced 771 030 reads and assembly set up 15 683 contigs for C. sativa and 16 828 for C. crenata. GO annotation revealed terms related to stress as ‘response to stimulus’, ‘transcription factor activity’ or ‘signaling’ for both transcriptomes. Differential gene expression analysis revealed that C. crenata involved more genes related with biotic stress upon pathogen inoculation than C. sativa. Those genes for both species are involved in regulation of plant immune response and stress adaptation and recovery. Furthermore, it is suggested that both species recognize the pathogen attack; however, the resistant species may involve more genes in the defense response than the susceptible species. RNA-seq enabled the selection of candidate genes for ink disease resistance in Castanea. The present data is a valuable contribution to the available Castanea genomic resources and constitutes the basis for further studies. Keywords: Castanea sativa; Castanea crenata; Phytophthora cinnamomi; RNA-seq; differentially expressed genes; biotic stress. Castanea root transcriptome in response to P. cinnamomi 75 Introduction European chestnut (sweet chestnut, Castanea sativa Miller) has great economic value due to fruit production, and ecological value including forest diversity and soil stability. In the last 100 years, ink disease caused by the soil oomycete Phytophthora cinnamomi Rands has contributed to a drastic reduction of C. sativa distribution area in Europe. Oomycetes (eukaryotic heterokonts) show strategies of plant infection that are similar to many fungal pathogens (Latijnhouwers et al. 2003). Sweet chestnut groves have also been affected since the 1930’s by the chestnut blight fungus [Cryphonectria parasitica (Murril) Barr], causal agent of the American chestnut [Castanea dentata (Marshall) Borkh.] decimation. Ink disease was introduced to Europe from the USA through the Azores islands (Fernandes 1955; Anagnostakis 2001). The first records on its appearance in northern Portugal date from 1838. It has since been reported in many European countries, including Spain, Italy, France and the United Kingdom. The progression of the disease in grove areas with high humidity has limited the establishment of new groves and impeded the conservation of old ones (Vannini and Vettraino 2001). Presently, the greatest impact of ink disease is limited to the warm southwestern and southern regions of central Europe [reviewed by (Brasier and Jung 2006)]. In the 19th century, ink disease was partially responsible for a decline of C. dentata in the Southeastern USA (Anagnostakis 2001), prior to its broad decimation by chestnut blight. Ink disease is currently re-emerging in the USA and constitutes a serious threat to the American chestnut reintroduction (Jacobs et al. 2013). Common woody hosts of P. cinnamomi include Eucalyptus, Quercus, Juglans, Betula and Castanea, and the mycelia also persist saprophytically in soil. In the presence of water, oospores and chlamydospores differentiate sporangia that form and release zoospores. Zoospores are motile and are able to penetrate non-lignified root tissue and the base of stems or trunks: Chapter III 76 both scenarios result in local tissue rot. Growth, reproduction and dissemination of the pathogen are favored under compacted and water saturated soils with poor aeration. Symptoms on the adult trees include leaf chlorosis, thinning of the crown and the persistence of immature fruits on the trees after leaf-fall. Larger roots are mainly affected, producing a black exudate which increases during spring and fall. Infected seedlings undergo a rapid or gradual leaf wilting, depending on the severity of the infection. The root system suffers extensive necrosis of the tap root that extends to the lateral roots and up the lower stem (Vannini and Vettraino 2001). Oßwald et al (Oßwald et al. 2014) explain the primary physiological, biochemical and molecular reactions described on infected roots of susceptible Phytophthorahost interaction, summarized as follows: 1) The pathogen releases elicitins into the rhizosphere, facilitating root penetration; 2) Down-regulation of defense genes in the host, facilitating pathogen growth; 3) Destruction of roots and impairment of water and nutrient uptake; 4) Increase of the abscisic acid phytohormone in roots; 5) Decrease in leaf water potential; 6) Stomata closure and decrease in photosynthesis; 7) Probable release of toxins and effectors into the host tissue during biotrophic growth of the pathogen and transport into the canopy via xylem sap flow; 8) Up-regulation of genes of the ethylene pathway and release of the phytohormone by leaves; 9) Decrease in cytokinin content in roots during the necrotrophic growth of the pathogen; 10) Chlorosis and wilting of leaves resulting from the changed water and hormonal status of the host caused by root infection. Progression of ink disease depends on environmental conditions, pathogen virulence and plant susceptibility. One strategy to control the disease is through breeding with resistant species. Soon after the introduction of Asian chestnuts to Europe it was verified that C. crenata (Castanea crenata Siebold & Zucc., the Japanese chestnut) has a high level of resistance to Phytophthora (Vannini and Vettraino 2001). Since the 1950’s, breeding programs with the European and Japanese chestnut were established in Castanea root transcriptome in response to P. cinnamomi 77 Portugal, France and Spain to obtain hybrids tolerant to ink disease, while maintaining fruit production and quality traits to satisfy commercial demands (Vannini and Vettraino 2001; Martins et al. 2009). However, fruit quality produced by these hybrids is below current market standards, so there is demand from both researchers and producers, for developing genomic tools to understand resistance mechanisms against P. cinnamomi. Barakat et al (Barakat et al. 2009; Barakat et al. 2012) described the generation of more than 1,5 million cDNA sequences for the American and Chinese chestnuts that have been used to analyse chestnut resistance to C. parasitica. The data are available through the Fagaceae Genomics Web (http://www.fagaceae.org/) and represent the first public resource on chestnut transcriptomes. The data we present here contribute to this resource by identifying Japanese and European chestnut genes involved in the reaction to ink disease, another critical threat to Castanea. To compare the response of the resistant Japanese chestnut with the response of the susceptible European chestnut to P. cinnamomi infection, four cDNA libraries of C. sativa (Cs) and C. crenata (Cc) root tissues, inoculated (i) and non-inoculated (n) with the pathogen were prepared for 454 pyrosequencing. Contig annotation and analysis of transcript abundance supported the quantification of transcript expression on inoculated and noninoculated roots in each species, as well as, identifying differentially expressed genes upon pathogen inoculation. This allowed a comparison of each species’ response to the pathogen and the selection of candidate genes for resistance to ink disease. Materials and Methods Plant material and pathogen inoculation The TRAGSA nursery (Grupo TRAGSA-SEPI, Maceda, Spain) provided 36 micropropagated plants at five years of age, 18 of C. sativa (Cs, susceptible) and 18 of C. crenata (Cc, resistant). Four treatments were set, corresponding Chapter III 78 to C. sativa and C. crenata inoculated and non-inoculated with P. cinnamomi (Supplementary material 1). Plants were distributed in 15 L pots with peat. A hypervirulent isolate of P. cinnamomi (IMI 340340) provided by Trás-osMontes and Alto Douro University was grown at 22°C on Potato Dextrose Agar. For soil infestation P. cinnamomi inoculum was prepared by growing mycelia on sterilized millet seeds (Ponicium mileaceum), which were thoroughly moistened with vegetable juice (V8®) broth [20% (v/v) with 3 g/L of CaCO3]. The mixture was incubated for three weeks in darkness at 24°C. At the time of inoculation (0 h), P. cinnamomi was carefully added to each container substrate at a concentration of 5% (v/v), in order to minimize root disturbance and wounding. No pathogen was added to non-inoculated plants. After inoculation all containers were flooded for 3h to stimulate zoospore release and to promote disease development. At 2, 4 and 7 days after inoculation, 6 plants per treatment (3 of C. sativa and 3 of C. crenata, Supplementary material 1) were removed from containers and root samples were collected. After rinsing, roots were frozen in liquid nitrogen and stored at -80°C. RNA isolation Total RNA from root tissue was isolated based on Le Provost et al (Le Provost et al. 2007). RNA integrity and purity was determined with a 2100 Bioanalyser with the RNA 6000 Pico kit (Agilent Technologies, Palo Alto, CA, USA). In order to compare gene expression between the two chestnut species after pathogen inoculation, four RNA pools were prepared, based on the experimental design described by Barakat et al (Barakat et al. 2009; Barakat et al. 2012): Cci, Ccn, Csi and Csn (i: inoculated; n: non-inoculated). Each pool included the RNA from nine plants, 3 biological replicates collected at 3 time points after inoculation (2, 4 and 7 days, Supplementary material 1). Castanea root transcriptome in response to P. cinnamomi 79 Poly(A) RNA enrichment, cDNA library construction and pyrosequencing The procedures described in this section were provided by the Next Gen Sequencing Unit at Biocant (Cantanhede, Portugal). The integrity of all RNA pools was verified on a 2100 Bioanalyser as above and the quantity assessed by fluorometry with the Quant-iTRiboGreen RNA kit (Invitrogen, CA, USA). Poly(A)RNA was enriched from total RNA using two rounds of the MicroPoly(A) Purist Kit (Applied Biosystems, Ambion, CA, USA), according to the manufacturer’s instructions. The RNA quality was again assessed on a 2100 Bioanalyser and the quantity determined by fluorometry as described above. A fraction of 200 ng of Poly(A)+ RNA of each isolate was used as starting material for cDNA library construction using Multiplex Identifiers (MIDs) according to the cDNA Rapid Library Preparation Method Manual, ‘GS FLX Titanium Series, October 2009’ (Roche-454 Life Sciences, Brandford, CT, USA). The four dscDNA libraries were quantified by fluorescence, pooled in equimolar amounts and pyrosequenced in a single plate with GS FLX Titanium chemistry (Roche-454 Life Sciences, Brandford, CT, USA), according the standard manufacturers’ procedures. Transcript assembly and functional annotation After 454 sequencing, the raw reads were processed to remove sequences with less than 100 nucleotides and low quality regions. Ribosomal, mitochondrial and chloroplast reads were identified through BLASTx against the non-redundant NBCI database and any hits with an E value of 0.0 were removed from the data set. All remaining reads were then assembled into contigs using 454 Newbler 2.6 (Roche, Branford, CT, USA) with the default parameters (40 bp overlap and 90% identity). A three step analysis was carried out to identify genes. First, the translation frame of each contig was assessed through BLASTx searches against Swissprot (E value<1E-6) and the corresponding amino acid sequence was translated using an in-house Chapter III 80 script. Then any contigs without translation were submitted to FrameDP (Gouzy et al. 2009) software with default parameters. Finally, all remaining contigs were analysed with ESTScan (Lottaz et al. 2003) with default parameters. Transcripts identified by FrameDP or ESTScan were searched using BLASTp against the non-redundant NBCI database (E value<1E-2) to translate putative proteins. The functional annotation of all translated amino acid sequences was predicted through assignment into protein families and identification of protein domains using InterProScan version 4.6 (Hunter et al. 2009). Gene Ontology (GO) terms identified by InterProScan results for each translated amino acid sequence were additionally retrieved and added to classify each transcript product. The procedures above described were provided by the Next Gen Sequencing Unit at Biocant. All contigs were taxonomy annotated in order to separate the sequences belonging to the Streptophyta phylum for further analysis. To obtain the taxonomical assignments we uploaded the contigs to MG-RAST (Meyer et al. 2008) (with default parameters), an automated analysis platform for metagenomes based on sequence similarity to both protein and nucleotide databases. Identification of differentially expressed genes related to P. cinnamomi resistance Differentially expressed genes were identified as genes showing significant higher/lower expression levels in inoculated root tissue versus noninoculated root tissue. The number of reads mapping to each transcript (contig) in the two treatments (inoculated and non-inoculated) was counted and used as an approximate estimation of gene expression level in the corresponding tissues. First, the contigs from the different samples were clustered at 90% similarity and 95% identity by CD-Hit 454 (Niu et al. 2010) to eliminate redundant sequences and generate reference contigs. The reads from each sample were then mapped to those references with Newbler Castanea root transcriptome in response to P. cinnamomi 81 mapping 2.6 (Roche, Branford, CT, USA) using the default parameters, and the number of reads contributed by each sample counted. Reads with multiple hits were discarded. The number of reads per reference contig per sample was used to build a contingency table, which was analysed with the Myrna statistical analysis package (Langmead et al. 2010), with the normalization factor set to 95th percentile. Statistical significance of the differential expression was evaluated using a linear regression model based on a Gaussian distribution, and using only contigs with a minimum of eight mapped reads. All results were compiled into a SQL database developed as an information management system. The procedures described above were provided by the Next Gen Sequencing Unit at Biocant. For the selection of Differentially Expressed Genes (DEGs), contigs with a P value <1E-03 were considered. P value describes the probability that differences in counts between the two sets in comparison are due to chance (Langmead et al. 2010). Fold expression changes were calculated for the inoculated vs. non-inoculated comparisons, Csi-Csn and Cci-Ccn. Contigs with a fold change greater than 1 were classified as up-regulated genes and contigs with a fold change less than to 1 were classified as down-regulated genes. For further analysis of DEGs (P. cinnamomi resistance-related genes), the log2 of fold change>ǀ1ǀ criteria was applied. DEGs were also analysed for the two inoculated species in the comparison Csi-Cci (P value<1E-03) to reveal the genes that were significantly induced in both species after pathogen challenge. On the other hand, the comparison of the two non-inoculated species Csn-Ccn (P value<1E-03) revealed the constitutive genes in both species whithout inoculation. The application Blast2GO (Conesa and Götz 2008), namely the Enrichment Analysis, was used to statistically analyse GO annotation in the comparisons Csi-Csn, Cci-Ccn, Csi-Cci and Csn-Ccn for DEGs. It employs a Fisher's exact test with multiple testing correction of FDR (Benjamini and Hochberg). Upon selection of a single test and P value <5E-3, all GO terms were tested Chapter III 82 if they are enriched in the DEGs group when compared to a reference group (all contigs in the comparison). 454 sequencing validation by real-time PCR The relative expression of a subset of genes was achieved by quantitative real-time PCR (qRT-PCR) to validate RNA-sequencing (RNA-seq). The four RNA pools used for sequencing (Cci, Ccn, Csi and Csn) were prepared for qRT-PCR as follows: RNA was treated with DNase (Turbo DNase-free kit Ambion, Inc., USA), according to manufacturer's instructions. cDNA was synthesized using RevertAid H Minus Reverse Transcriptase (Thermo Fisher Scientific, Waltham, USA) according to manufacturer's instructions. Gene specific primers were designed for six target genes (Supplementary material 2) using Primer Express (version 1.0, Applied Biosystems, Sourceforge, USA). Actin-7 was selected as a reference gene after verifying a similar number of reads for all cDNA libraries and used for normalization of expression. A final concentration of 0,2 μM of each primer was used in 25 μL reactions, together with cDNA as template and Maxima SYBR Green/ROX qPCR Master Mix (Fermentas, Ontario, Canada), on a StepOne™ Real-Time PCR system (Applied Biosystems, Foster City, CA, USA). Thermal cycling for all genes started with a denaturation step at 95°C for 10 min followed by 40 cycles of denaturation at 95°C for 15 s and annealing temperatures for 30 s. Three technical replicates were used per reaction set, including template and no template controls. Non-specific PCR products were analysed by dissociation curves. The relative expression value and mean absolute deviation values were calculated for the pool comparisons Cci-Ccn and CsiCsn according to the ΔΔCT method (Livak and Schmittgen 2001). Results 454 sequencing and assembly summary Castanea root transcriptome in response to P. cinnamomi 83 Two Japanese chestnut cDNA libraries were constructed, one from a RNA pool of inoculated root tissue (Cci) and the other from a RNA pool of noninoculated root tissue (Ccn). A half plate of sequencing was used, resulting in 220 412 reads for Cci and 182 314 reads for Ccn, with an average read length of 350 nt (Table 1). Approximately 77 and 64 megabases of cDNA were generated for Cci and Ccn respectively. After assembly, 8 528 contigs were generated for Cci and 8 300 contigs were generated for Ccn, with an average length of 885 nt. 2 712 Cci contigs and 2 214 Ccn contigs had more than 1000 nt, corresponding to 32% and 27% of all respective contigs. Table 1. Summary of 454 sequencing for Castanea crenata and Castanea sativa root transcriptomes. cDNA library Cci Ccn Csi Csn Roots sampled C. crenata inoculated C. crenata non-inoculated C. sativa inoculated C. sativa noninoculated No. of plates ¼ ¼ ¼ ¼ No. of reads 220 412 182 314 181 384 186 920 Average read length (nt) 350 350 357 367 No. of bp 77 175 000 63 823 300 64 884 000 68 672 896 No. of contigs 8 528 8 300 7 208 8 475 Average contig length (nt) 915 854 856 823 No. of large contigsa 2 712 2 214 1 943 2 065 No. of putative proteins 8 149 7 969 6 852 8 073 AA sequences assigned to InterPro terms 6 373 6 279 5 350 6 213 AA sequences assigned to GO terms 4 885 4 790 4 090 4 691 a Greater than 1000 nt Chapter III 90 Figure 4. Distribution of Castanea crenata and Castanea sativa differentially expressed genes (DEGs) into functional sub-categories of Gene Ontology. Enriched Analysis was applied separately to upregulated and downregulated DEGs and compared with the reference sets of all contigs in the comparison, to obtain significant GO terms. The selected P value on Fisher’s Exact Test was lower than 5E−3. P. cinnamomi resistance-related genes We consider that the most significant candidate genes of resistance to P. cinnamomi are C. crenata (resistant species) genes up-regulated at least 2 times after inoculation that are not present among C. sativa DEGs and are putatively related to stress response. A selection of C. crenata candidate genes was inferred from the Cci-Ccn DEGs list in Supplementary material 3 and assigned to the functional categories in Table 2. Castanea root transcriptome in response to P. cinnamomi 91 Table 2 Castanea crenata (Japanese, resistant) candidate genes to Phytophthora cinnamomi resistance. Genes were associated into the functional categories in the left column (Bold). Chapter III 92 Castanea root transcriptome in response to P. cinnamomi 93 Chapter III 94 Especially noteworthy are the kinase receptor genes that may be involved in pathogen recognition (as the Probable LRR receptor-like serine/threonineprotein kinase) (Diévart and Clark 2003) and genes corresponding to TF involved in the regulation of host response after pathogen perception (as WRKY TF) (Yang et al. 2009). The putative involvement of JA and salicylic acid (SA) signaling pathways was inferred from the categories ‘Regulation of host response after pathogen perception’ and ‘Lipid signaling’. Two genes in the ‘Regulation of plant immune response’ category are responsive to the plant hormone ethylene (e.g. Ocs element-binding factor 1) (Zhang and Singh 1994). Certain C. crenata up-regulated genes may prevent pathogen progress, such as the precursor of Cationic peroxidase 1 (Reimers et al. 1992) [category ‘Hypersensitive response’ (HR)], and Pectinesterase 2 (Wen et al. 2013) (category ‘Cell wall strengthening’). Three genes, e.g. Probable glutathione S-transferase, were associated in the category ‘HR recovery’ (Ryu et al. 2009). Genes involved in ‘Anti-fungal metabolite synthesis’ (such as UDPglycosyltransferase 85A2) (Woo et al. 2007) and ‘Anti-fungal enzymes’ (such as Probable carboxylesterase 120) (Marshall et al. 2003) may take part in the host response to enhance the defense to P. cinnamomi. Genes in the categories ‘Regulation of drought stress’, ‘Response to drought stress’ and ‘Stress recovery’ are also represented in the C. crenata candidate list, such as NAC domain-containing protein 72 (Singh et al. 2013), Phosphoprotein ECPP44 (Tan and Kamada 2000) and Lon protease homolog 2 peroxissomal (Lingard and Bartel 2009), respectively. C. sativa DEGs after P. cinnamomi inoculation are presented in Supplementary material 4. In order to identify the susceptible species’ response to the pathogen we selected up-regulated genes in inoculated C. sativa (at least two times) that are not present among C. crenata DEGs and are putatively related to stress response. Those genes were distributed in a series of functional categories in Table 3. In the ‘Regulation of plant immune Castanea root transcriptome in response to P. cinnamomi 95 response’ category we emphasize the gene REF/SRPP-like protein At1g67360 (Taki et al. 2005), induced by a precursor of JA. C. sativa also invests in genes related to ‘Stress recovery’ (e.g. Aminophospholipid flippase 9) (López-Marqués et al. 2012) and ‘HR recovery’ (e.g. 4hydroxyphenylpyruvate dioxygenase) (Peal et al. 2011). Genes involved in ‘Anti-fungal metabolite synthesis’ (e.g. Flavonoid 3-hydroxylase) (Sharma et al. 2012) and ‘Cell wall strengthening’ (e.g. UPF0497 membrane protein At3g06390) (Roppolo et al. 2011) may prevent P. cinnamomi proliferation in the host. Finally, C. sativa up-regulated genes linked to drought stress regulation, such as the TF Homeobox-leucine zipper protein HAT5 (Henriksson et al. 2005) may play a role in host recovery from pathogenicity. Validation of RNA-seq The RNA-seq approach allowed for the quantification of gene expression levels by sequence read depth. DEGs were identified by estimating the ratio between reads in inoculated libraries and non-inoculated controls (Cci-Ccn, Csi-Csn). To validate the differential expression levels observed by RNAseq, qRT-PCR was used to obtain the expression level of DEGs in inoculated libraries (Cci, Csi) relative to non-inoculated libraries (Ccn, Csn). The selected DEGs (Figure 5) are putatively related to Castanea response to the pathogen and to host recovery, and include: Ethylene-responsive TF 4, Disease resistance protein At4g27190, Ethylene-responsive TF ABR1, Precursor of glucan 1,3-beta-glucosidase (family 5), Pectinesterase 2 and C2 domain-containing protein At1g53590. The differential gene expression for the comparisons Cci-Ccn and Csi-Csn acquired with the 454 sequencing was compared with the relative expression levels obtained with qRT-PCR for the selected DEGs. Chapter III 96 Table 3 Castanea sativa (European, susceptible) up-regulated genes upon Phytophthora cinnamomi inoculation related to stress response. Genes were associated into the functional categories in the left column (Bold). Castanea root transcriptome in response to P. cinnamomi 97 Chapter III 98 Castanea root transcriptome in response to P. cinnamomi 99 The results presented in Figure 5 reveal differences in the expression levels of C. crenata transcripts upon inoculation when compared to C. sativa transcripts upon inoculation. Those differences are in accordance with read data obtained by RNA-seq and may reflect Castanea root transcriptome in response to P. cinnamomi. Discussion Functional annotation GO annotation comparison of expressed genes after inoculation between Japanese (C. crenata, resistant to pathogen) and European chestnut (C. sativa, susceptible to pathogen) revealed a correlation of gene ontology, suggesting a convergent response after pathogen inoculation. However, among DEGs, GO annotation revealed differences that suggest distinct host susceptibility to the pathogen as well as variations in gene expression and timing. C. crenata inoculated with the pathogen up-regulated genes with the functional GO annotation ‘Oxidation reduction process’ (BP), disclosing genes involved in the synthesis of anti-fungal secondary metabolites (6 in 27) and in stress recovery (10 in 27). Examples are Squalene monooxygenase (Belchí-Navarro et al. 2013) and Prolyl 4-hydroxylase subunit alpha-2 (Vlad et al. 2007), respectively. On other hand, ‘Sequencespecific DNA binding transcription factor activity’ (MF) point to genes coding for TF related to pathogen recognition and biotic stress regulation (10 in 14, examples in Table 2). Contrasting with C. crenata, GO annotation for inoculated C. sativa revealed that the down-regulated genes in ‘Catalytic activity’ (MF) were involved in the synthesis of secondary metabolites, protein kinases and receptor-like protein kinases (24 in 84). In the same term are also included genes related to stress recovery (9 in 84). Chapter III 106 decarboxylase 1 codes for root-specific calcium/calmodulin-regulated GAD1, which plays a major role in GABA synthesis in plants responding to stress, thereby helping maintain plant homeostasis (Bouché and Fromm 2004). Thus the repression of GAD1 may affect host recovery from pathogen attack. Glutaredoxins are candidates for mediating redox regulation of transcriptional regulators that target genes associated with detoxification and pathogen defense (Ndamukong et al. 2007). The Glutaredoxin-C9 gene was highly up-regulated in Japanese chestnut; its down-regulation in European chestnut may also affect host recovery. Comparison between Japanese and European chestnut response DEGs annotation analyses revealed that Japanese and European chestnuts show many common features in their responses to P. cinnamomi. However, this evaluation must be carefully regarded as it is not supported by transcript profiling or functional analysis. Upon P. cinnamomi inoculation, both species up-regulated genes involved in HR/HR recovery, genes related to the regulation of JA pathway and genes induced by JA related to anti-fungal metabolite synthesis and anti-fungal enzymes. The presence of HR and JA signaling upon pathogen inoculation indicates that both host species were able to recognize the pathogen attack. According to Thomma (Thomma 1998), the JA-dependent defense response pathway is required for resistance to necrotrophic pathogens. Eshragui et al (Eshraghi et al. 2014) suggest that a P. cinnamomi challenge activates JA-related plant defense responses in leaves of A. thaliana Col-0. HR is suggested to be associated with all forms of resistance to Phytophthora (Kamoun et al. 1999) and is believed to constitute one of the primary mechanisms of resistance to plant pathogens. Induction of HR is often associated with synthesis of antimicrobial compounds and cell wall thickening (Hammond-Kosack and Jones 1996). We identified C. crenata and C. sativa DEGs related to cell wall strengthening and anti-fungal metabolite synthesis. HR also induces several genes Castanea root transcriptome in response to P. cinnamomi 107 involved in cellular protection (Jabs et al. 1996). We also identified DEGs related to HR recovery in both species. The suggested occurrence of HR in inoculated European chestnut, the susceptible species, potentially points to partial resistance within this genotype. Partial resistance to Phytophthora infestans is common in wild Solanum species, which may reveal HR-like necrotic reactions and, occasionally, late or trailing HR. This suggests a weak R gene-Avr gene interaction or a gene-dosage effect resulting in ineffective HR and partly resistant phenotypes (Kamoun et al. 1999). Japanese chestnut resistance to ink disease may in part result from a set of up-regulated genes during P. cinnamomi attack involved in pathogen recognition, regulation of host response after pathogen perception, and signaling through lipids. When compared to Japanese chestnut, European chestnut up-regulated much less genes in those functional categories. Current knowledge describes the plant immune response as starting with the recognition of pathogen elicitors by plant receptors, followed by induction of resistance genes (R genes) that initiate signal transduction cascades leading to: a) HR and rapid cell death and b) the activation of phytohormone signaling pathways [reviewed in (Bari and Jones 2009)]. In our study inoculated Japanese chestnut induced DEGs involved in the SA pathway regulation (e.g. Calcium-dependent protein kinase isoform 3) (Chung et al. 2004) as well as DEGs induced by SA related to the regulation of plant immune response (e.g. Sulfate transporter 3,1) (Marsolais et al. 2007) and HR (e.g. Arginine decarboxylase) (Nakane et al. 2003). García-Pineda et al (2009) observed that SA inhibited avocado root colonization in the interaction between Persea americana and P. cinnamomi. In the complex web of defense responses JA, SA, Ethylene and Abscisic Acid are essential players (Bari and Jones 2009). SA is activated during and following HR (Jabs et al. 1996) and is generally involved in the activation of defense responses against biotrophic and hemi-biotrophic pathogens, as well as in the establishment of systemic acquired resistance (Bari and Jones 2009). Chapter III 108 Vleeshouwers et al (Vleeshouwers et al. 2000) studied the P. infestansSolanum interaction using wild species and reported that in fully resistant genotypes, the HR was faster and resulted in smaller lesions than in partially resistant clones. The authors suggest that the difference between compatibility (non-resistant host response) and incompatibility (resistant host response) is quantitative rather that qualitative. In our study, Japanese chestnut regulated a higher number of genes involved in biotic stress upon P. cinnamomi inoculation when compared to the European chestnut. The identified DEGs are not only related to HR but also with cell wall strengthening, anti-fungal metabolite synthesis and anti-fungal enzyme synthesis, and may account for the Japanese chestnut’s adequate resistance to ink disease. Castanea response to P. cinnamomi and C. parasitica: brief comparison The reports of Barakat et al. (2009, 2012) provided the first insights into chestnut resistance to C. parasitica using high-throughput RNA-seq. The response of chestnut to C. parasitica and P. cinnamomi may be comparable, as fungi and Oomycetes share similar infection mechanisms (Latijnhouwers et al. 2003). When comparing Chinese and American chestnut responses to C. parasitica with the Japanese and European chestnut responses to P. cinnamomi, we found similar DEGs that fall in the following functional categories: a) Regulation of biotic stress response (ATPase transporter, Pyridine nucleotide-disulphide oxidoreductase), b) HR and cell wall lignification (Peroxidase), c) HR recovery (Arginine decarboxylase, Manganese superoxide dismutase), d) Anti-fungal enzymes (Thaumatin-like protein, β-1,3-glucanase, Chitinase), e) Anti-fungal metabolite synthesis (family 1 Cytochrome P450 glycosyltransferase, Abscisic acid 8’- hydroxylase, Squalene monooxygenase, UDP-glucosyltransferase), f) Cell wall synthesis (β-expansin), and g) Stress recovery (ABC transporter family, Glyceraldehyde 3-phosphate dehydrogenase). Other shared Castanea Castanea root transcriptome in response to P. cinnamomi 109 responses to both pathogens include DEGs related to kinase genes involved in pathogen recognition and JA pathway activation, gene regulation by Myb TF and Ethylene-responsive TF, and genes of the 26S proteasome regulatory unit. The response of all four species to both pathogens further includes genes from the flavonoid pathway that promote phytoalexin synthesis. In summary, the DEG analysis of C. sativa and C. crenata root transcriptomes after P. cinnamomi inoculation revealed similarities among the four Castanea species response to both pathogens, namely genes related to systemic acquired resistance, HR that may prevent pathogen spread and the putative involvement of JA pathway. Some of these DEGs may also promote cell wall strengthening through lignification and synthesis of flavonoids as anti-fungal metabolites. Final considerations RNA-seq using 454 platform was adequate for comparing the root transcriptomes of two Fagaceae species, Castanea sativa and Castanea crenata when either inoculated or non-inoculated with the pathogen Phytophthora cinnamomi. The four sequenced transcript libraries allowed a draft comparison of both species’ responses to the pathogen in terms of gene regulation and pathways, together with the selection of candidate genes for host resistance to P. cinnamomi. Although further research is required on gene expression at specific time points after inoculation, in silico analysis has shown that Japanese and European chestnut, despite the association of expressed genes in similar functional categories, differ in the distribution of DEGs after pathogen inoculation. The most noteworthy result from DEG analysis was the overall down-regulation of genes in susceptible C. sativa, which may facilitate the pathogenicity of P. cinnamomi. On the other hand, in the resistant C. crenata there was the regulation of a higher number of genes related with biotic stress when compared to C. sativa, mostly up- Chapter III 110 regulated. Analysis of homology and functional annotation revealed associations between many of those up-regulated genes with pathogen response in other plant species, and suggests involvement in pathogen recognition, regulation of the host immune response, signaling, hypersensitive response, cell wall strengthening and encoding of enzymes and synthesis of metabolites against Oomycetes and fungal pathogens. The regulation of DEGs in C. crenata and many of the specific transcripts we identified may account for the adequate resistance level of this species to P. cinnamomi. SSR markers were also developed from the sequences of these candidate genes in order to improve the mapping approach for identification of QTLs related to pathogen resistance in Japanese and European chestnut (Costa et al. 2011 and Chapter V). In Chapter V we will analyse if the candidate genes map to disease resistance QTLs, which will provide further support for a major role in chestnut resistance to the pathogen. Acknowledgments We acknowledge Dr. Beatriz Cuenca (TRAGSA-SEPI) for providing the plant material used in this study. The authors are also grateful to Dr. Andreia Figueiredo (BioFIG) and Dr. Filipa Monteiro (BioFIG) for help on qRT-PCR, Dr. Conceição Egas (Next Gen Sequencing Unit, Biocant) for submitting raw data to NCBI, Dr. Dana Nelson (USDA Forest Service, MS, Unit of Forest genetics and Ecosystems Biology) for arrangements on submitting data to Fagaceae.org, and Prof. William Powell and Andrew Newhouse (SUNY College of Environmental Science and Forestry, NY, Department of Environmental and Forest Biology) for critical review of the manuscript. Data Archiving Statement Raw data files can be accessed in the Short Read Archive at NCBI (http://www.ncbi.nlm.nih.gov) with the reference PRJNA215368. Nucleotide Castanea root transcriptome in response to P. cinnamomi 111 and aminoacid sequences are publicly available in the Fagaceae Genomics Web (http://www.fagaceae.org/). References Anagnostakis SL (2001) The effect of multiple importations of pests and pathogens on a native tree. Biol Invasions 3:245–254. doi: 10.1023/A:1015205005751 Barakat A, DiLoreto DS, Zhang Y, et al (2009) Comparison of the transcriptomes of American chestnut (Castanea dentata) and Chinese chestnut (Castanea mollissima) in response to the chestnut blight infection. BMC Plant Biol 9:51. doi: 10.1186/1471-2229-9-51 Barakat A, Staton M, Cheng C-H, et al (2012) Chestnut resistance to the blight disease: insights from transcriptome analysis. BMC Plant Biol 12:38. doi: 10.1186/1471-222912-38 Bari R, Jones JDG (2009) Role of plant hormones in plant defence responses. Plant Mol Biol 69:473–88. doi: 10.1007/s11103-008-9435-0 Belchí-Navarro S, Almagro L, Sabater-Jara AB, et al (2013) Induction of trans-resveratrol and extracellular pathogenesis-related proteins in elicited suspension cultured cells of Vitis vinifera cv Monastrell. J Plant Physiol 170:258–64. doi: 10.1016/j.jplph.2012.10.003 Benhamou N, Mazau D, Grenier J, Esquerr-Tugay M-T (1991) Time-course study of the accumulation of hydroxyproline-rich glycoproteins in root cells of susceptible and resistant tomato plants infected by Fusarium oxysporum f. sp.radicis-lycopersici. Planta 184:196–208. doi: 10.1007/BF01102419 Bocca SN, Kissen R, Rojas-Beltrán JA, et al (1999) Molecular cloning and characterization of the enzyme UDP-glucose: protein transglucosylase from potato. Plant Physiol Biochem 37:809–819. doi: 10.1016/S0981-9428(99)00117-5 Bouché N, Fromm H (2004) GABA in plants: just a metabolite? Trends Plant Sci 9:110–5. doi: 10.1016/j.tplants.2004.01.006 Brasier CM, Jung T (2006) Recent developments in Phytophthora diseases of trees and natural ecosystems in Europe. Prog. Res. Phytophthora Dis. For. Trees. Proceedings, 3rd Int. IUFRO Work. Party. pp 5–16 Broeckling CD, Huhman D V, Farag MA, et al (2005) Metabolic profiling of Medicago truncatula cell cultures reveals the effects of biotic and abiotic elicitors on metabolism. J Exp Bot 56:323–36. doi: 10.1093/jxb/eri058 Cantu D, Vicente AR, Labavitch JM, et al (2008) Strangers in the matrix: plant cell walls and pathogen susceptibility. Trends Plant Sci 13:610–617. Chen K, Fan B, Du L, Chen Z (2004) Activation of hypersensitive cell death by pathogen- Chapter III 112 induced receptor-like protein kinases from Arabidopsis. Plant Mol Biol 56:271–83. doi: 10.1007/s11103-004-3381-2 Chung E, Park JM, Oh S-K, et al (2004) Molecular and biochemical characterization of the Capsicum annuum calcium-dependent protein kinase 3 (CaCDPK3) gene induced by abiotic and biotic stresses. Planta 220:286–95. doi: 10.1007/s00425-004-1372-9 Coelho AC, Horta Jung M, Ebadzad G, Cravador A (2011) Quercus suber – Phytophthora cinnamomi interaction: a hypothetical molecular mechanism model. New Zeal J For Sci 41S:S143–S157. Conesa A, Götz S (2008) Blast2GO: A Comprehensive Suite for Functional Analysis in Plant Genomics. Int J Plant Genomics. doi: doi:10.1155/2008/619832 Costa R, Santos C, Tavares F, et al (2011) Mapping and transcriptomic approches implemented for understanding disease resistance to Phytophthora cinnamomi in Castanea sp. BMC Proc 5:O18. doi: 10.1186/1753-6561-5-S7-O18 Dhondt S, Geoffroy P, Stelmach BA, et al (2000) Soluble phospholipase A2 activity is induced before oxylipin accumulation in tobacco mosaic virus-infected tobacco leaves and is contributed by patatin-like enzymes. Plant J 23:431–440. doi: 10.1046/j.1365313x.2000.00802.x Diévart A, Clark SE (2003) Using mutant alleles to determine the structure and function of leucine-rich repeat receptor-like kinases. Curr Opin Plant Biol 6:507–516. Dixon RA, Paiva NL (1995) Stress-Induced Phenylpropanoid Metabolism. Plant Cell 7:1085– 1097. doi: 10.1105/tpc.7.7.1085 Eshraghi L, Anderson JP, Aryamanesh N, et al (2013) Defence signalling pathways involved in plant resistance and phosphite-mediated control of Phytophthora cinnamomi. Plant Mol Biol Report 32:342–356. doi: 10.1007/s11105-013-0645-5 Feng B, Li P (2012) Genome-wide identification of laccase gene family in three Phytophthora species. Genetica 140:477–84. doi: 10.1007/s10709-012-9696-z Fernandes CT (1955) A luta contra a doença da tinta nos soutos do norte de Portugal e ensaios diversos para a sua maior eficiência e economia., Direcção-G. 61 p. Foster J, Kim HU, Nakata PA, Browse J (2012) A previously unknown oxalyl-CoA synthetase is important for oxalate catabolism in Arabidopsis. Plant Cell 24:1217–29. doi: 10.1105/tpc.112.096032 García-Pineda E, Benezer-Benezer M, Gutiérrez-Segundo A, et al (2009) Regulation of defence responses in avocado roots infected with Phytophthora cinnamomi (Rands). Plant Soil 331:45–56. doi: 10.1007/s11104-009-0225-5 Gouzy J, Carrere S, Schiex T (2009) FrameDP: sensitive peptide detection on noisy matured sequences. Bioinformatics 25:670–1. doi: 10.1093/bioinformatics/btp024 Castanea root transcriptome in response to P. cinnamomi 113 Hammond-Kosack KE, Jones JD (1996) Resistance gene-dependent plant defense responses. Plant Cell 8:1773–91. Hammond-Kosack KE, Jones JDG (1997) Plant disease resistance genes. Annu Rev Plant Physiol Plant Mol Biol 48:575–607. doi: 10.1146/annurev.arplant.48.1.575 Hartmann U, Sagasser M, Mehrtens F, et al (2005) Differential combinatorial interactions of cis-acting elements recognized by R2R3-MYB, BZIP, and BHLH factors control lightresponsive and tissue-specific activation of phenylpropanoid biosynthesis genes. Plant Mol Biol 57:155–71. doi: 10.1007/s11103-004-6910-0 He X, Miyasaka SC, Fitch MMM, et al (2013) Taro (Colocasia esculenta) Transformed with a Wheat Oxalate Oxidase Gene for Improved Resistance to Taro Pathogen Phytophthora colocasiae. HortScience 48:22–27. Henriksson E, Olsson ASB, Johannesson H, et al (2005) Homeodomain leucine zipper class I genes in Arabidopsis. Expression patterns and phylogenetic relationships. Plant Physiol 139:509–18. doi: 10.1104/pp.105.063461 Hondo D, Hase S, Kanayama Y, et al (2007) The LeATL6-associated ubiquitin/proteasome system may contribute to fungal elicitor-activated defense response via the jasmonic acid-dependent signaling pathway in tomato. Mol Plant Microbe Interact 20:72–81. doi: 10.1094/MPMI-20-0072 Hu X, Neill S, Cai W, Tang Z (2003) Hydrogen peroxide and jasmonic acid mediate oligogalacturonic acid-induced saponin accumulation in suspension-cultured cells of Panax ginseng. Physiol Plant 118:414–421. doi: 10.1034/j.1399-3054.2003.00124.x Hunter S, Apweiler R, Attwood TK, et al (2009) InterPro: the integrative protein signature database. Nucleic Acids Res 37:D211–5. doi: 10.1093/nar/gkn785 Irshad M, Canut H, Borderies G, et al (2008) A new picture of cell wall protein dynamics in elongating cells of Arabidopsis thaliana: confirmed actors and newcomers. BMC Plant Biol 8:94. doi: 10.1186/1471-2229-8-94 Jabs T, Dietrich RA, Dangl JL (1996) Initiation of Runaway Cell Death in an Arabidopsis Mutant by Extracellular Superoxide. Science (80-) 273:1853–1856. doi: 10.1126/science.273.5283.1853 Jackson D, Culianez-Macia F, Prescott AG, et al (1991) Expression patterns of myb genes from Antirrhinum flowers. Plant Cell 3:115–25. doi: 10.1105/tpc.3.2.115 Jacobs DF, Dalgleish HJ, Nelson CD (2013) A conceptual framework for restoration of threatened plants: the effective model of American chestnut (Castanea dentata) reintroduction. New Phytol 197:378–93. doi: 10.1111/nph.12020 Jiang N, Xiao D, Zhang D, et al (2009) Negative roles of a novel nitrogen metabolite repression-related gene, TAR1, in laccase production and nitrate utilization by the Chapter III 114 basidiomycete Cryptococcus neoformans. Appl Environ Microbiol 75:6777–82. doi: 10.1128/AEM.00708-09 Kamoun S, Huitema E, Vleeshouwers V (1999) Resistance to oomycetes: a general role for the hypersensitive response? Trends Plant Sci 4:196–200. Keinänen SI, Hassinen VH, Kärenlampi SO, Tervahauta AI (2007) Isolation of genes upregulated by copper in a copper-tolerant birch (Betula pendula) clone. Tree Physiol 27:1243–52. Langmead B, Hansen KD, Leek JT (2010) Cloud-scale RNA-sequencing differential expression analysis with Myrna. Genome Biol 11:R83. doi: 10.1186/gb-2010-11-8-r83 Latijnhouwers M, de Wit PJGM, Govers F (2003) Oomycetes and fungi: similar weaponry to attack plants. Trends Microbiol 11:462–469. doi: 10.1016/j.tim.2003.08.002 Le Provost G, Herrera R, Paiva JA, et al (2007) A micromethod for high throughput RNA extraction in forest trees. Biol Res 40:291–7. doi: /S0716-97602007000400003 Leivar P, Antolín-Llovera M, Ferrero S, et al (2011) Multilevel control of Arabidopsis 3-hydroxy3-methylglutaryl coenzyme A reductase by protein phosphatase 2A. Plant Cell 23:1494– 511. doi: 10.1105/tpc.110.074278 Li G, Liu K, Baldwin SA, Wang D (2003) Equilibrative nucleoside transporters of Arabidopsis thaliana. cDNA cloning, expression pattern, and analysis of transport activities. J Biol Chem 278:35732–42. doi: 10.1074/jbc.M304768200 Lin Q, Buckler ES, Muse S V, Walker JC (1999) Molecular evolution of type 1 serine/threonine protein phosphatases. Mol Phylogenet Evol 12:57–66. doi: 10.1006/mpev.1998.0560 Lingard MJ, Bartel B (2009) Arabidopsis LON2 is necessary for peroxisomal function and sustained matrix protein import. Plant Physiol 151:1354–65. doi: 10.1104/pp.109.142505 Liu X, Bush DR (2006) Expression and transcriptional regulation of amino acid transporters in plants. Amino Acids 30:113–20. doi: 10.1007/s00726-005-0248-z Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2−ΔΔCT Method. Methods 25:402–8. doi: 10.1006/meth.2001.1262 López-Marqués RL, Poulsen LR, Palmgren MG (2012) A putative plant aminophospholipid flippase, the Arabidopsis P4 ATPase ALA1, localizes to the plasma membrane following association with a β-subunit. PLoS One 7:e33042. doi: 10.1371/journal.pone.0033042 Lottaz C, Iseli C, Jongeneel C V., Bucher P (2003) Modeling sequencing errors by combining Hidden Markov models. Bioinformatics 19:ii103–ii112. doi: 10.1093/bioinformatics/btg1067 Marshall SDG, Putterill JJ, Plummer KM, Newcomb RD (2003) The carboxylesterase gene Castanea root transcriptome in response to P. cinnamomi 115 family from Arabidopsis thaliana. J Mol Evol 57:487–500. doi: 10.1007/s00239-0032492-8 Marsolais F, Boyd J, Paredes Y, et al (2007) Molecular and biochemical characterization of two brassinosteroid sulfotransferases from Arabidopsis, AtST4a (At2g14920) and AtST1 (At2g03760). Planta 225:1233–44. doi: 10.1007/s00425-006-0413-y Martins L, Anjos R, Costa R, Gomes-Laranjo J (2009) COLUTAD: um clone de castanheiro resistente à doença da tinta. In: Gomes-Laranjo J, Peixoto F, Ferreira-Cardoso J (eds) Castanheiros, Técnicas e Práticas, UTAD - Vil. Vila Real, pp 135–142 Meyer F, Paarmann D, D’Souza M, et al (2008) The metagenomics RAST server - a public resource for the automatic phylogenetic and functional analysis of metagenomes. BMC Bioinformatics 9:386. doi: 10.1186/1471-2105-9-386 Mutuku JM, Nose A (2012) Changes in the contents of metabolites and enzyme activities in rice plants responding to Rhizoctonia solani Kuhn infection: activation of glycolysis and connection to phenylpropanoid pathway. Plant Cell Physiol 53:1017–32. doi: 10.1093/pcp/pcs047 Nakane E, Kawakita K, Doke N, Yoshioka H (2003) Elicitation of primary and secondary metabolism during defense in the potato. J Gen Plant Pathol 69:378–384. doi: 10.1007/s10327-003-0075-6 Ndamukong I, Abdallat A Al, Thurow C, et al (2007) SA-inducible Arabidopsis glutaredoxin interacts with TGA factors and suppresses JA-responsive PDF1.2 transcription. Plant J 50:128–39. doi: 10.1111/j.1365-313X.2007.03039.x Niu B, Fu L, Sun S, Li W (2010) Artificial and natural duplicates in pyrosequencing reads of metagenomic data. BMC Bioinformatics 11:187. doi: 10.1186/1471-2105-11-187 Ohta M, Matsui K, Hiratsu K, et al (2001) Repression domains of class II ERF transcriptional repressors share an essential motif for active repression. Plant Cell 13:1959–68. Oßwald W, Fleischmann F, Rigling D, et al (2014) Strategies of attack and defence in woody plantPhytophthora interactions. For Pathol 44:n/a–n/a. doi: 10.1111/efp.12096 Peal L, Jambunathan N, Mahalingam R (2011) Phylogenetic and expression analysis of RNAbinding proteins with triple RNA recognition motifs in plants. Mol Cells 31:55–64. doi: 10.1007/s10059-011-0001-2 Porta H, Rocha-Sosa M (2002) Plant lipoxygenases. Physiological and molecular features. Plant Physiol 130:15–21. doi: 10.1104/pp.010787 Reeksting BJ, Coetzer N, Mahomed W, et al (2014) De novo sequencing, assembly, and analysis of the root transcriptome of Persea americana (Mill.) in response to Phytophthora cinnamomi and flooding. PLoS One 9:e86399. doi: 10.1371/journal.pone.0086399 Chapter III 122 Castanea root transcriptome in response to P. cinnamomi 123 Chapter III 124 Castanea root transcriptome in response to P. cinnamomi 125 Chapter III 126 Castanea root transcriptome in response to P. cinnamomi 127 Chapter III 128 Castanea root transcriptome in response to P. cinnamomi 129 Chapter III 130 Castanea root transcriptome in response to P. cinnamomi 131