Search

TYLCV-IS76 is a recombinant begomovirus (family Geminiviridae) between Tomato yellow leaf curl virus (TYLCV) and Tomato yellow leaf curl Sardinia virus. It was detected in 2010 in Southern Morocco, and subsequent large field surveys showed that it had replaced its parental viruses. The fact that TYLCV-IS76 was positively selected in tomato plants carrying the begomovirus resistance gene Ty-1, and the co-occurrence of its emergence with the replacement of susceptible tomato cultivars with Ty-1 resistant cultivars, both support a Ty-1 resistancedriven emergence of TYLCV-IS76. The virus-plant interactions that determine positive selection of begomovirus recombinants are thought to involve gene silencing, because the Ty-1 gene encodes a gamma-clade RNAdependent RNA polymerase (RDRγ) implicated in biogenesis of viral small interfering RNAs (siRNAs) and methylation of viral DNA. To further investigate the mechanisms of RDRγ-mediated antiviral defence and defence evasion by TYLCV-IS76, we performed a comparative small RNA-ome and transcriptome profiling of Ty- 1 resistant and control susceptible plants infected with TYLCV or TYLCV-IS76. We found that RDRγ strongly enhances production of 22 and 24 nt viral siRNAs from both TYLCV and IS76 recombinant. The TYLCV-IS76 recombination event enhances transcription of viral silencing suppressor and coat protein genes and triggers more potent antiviral silencing responses. Compared to TYLCV, TYLCV-IS76 better evades transcriptional silencing of viral genes promoted by RDRγ, which may explain why the recombinant virus outcompetes its parents in Ty-1 plants.

National audience; Leptosphaeria maculans is responsible for the stem canker, a major disease of oilseed rape (Brassica napus). Specific resistance genes are used in commercial varieties to control the disease. The effectiveness of a given resistance gene is a function of the frequency of the corresponding avirulent allele in field populations of the pathogen. After the release of the oilseed rape resistance gene Rlm1 in the 1990’s in France, a very rapid increase in the frequency of isolates virulent towards this gene was observed [1]. More recently, a new resistance gene, Rlm7, was introduced into commercial hybrid varieties, at a time when most (>99.5%) of the L. maculans isolates possessed the corresponding avirulent allele AvrLm7 [2]. Since 2000, the frequency of isolates virulent against Rlm7 has been monitored in populations of L. maculans in either experimental fields with increased selection pressure, or at a national scale in more standard agronomic situations. While a rapid increase in frequency of virulent isolates was observed in an experimental field at Grignon that had minimum tillage and continuous oilseed rape cropping (36% of isolates virulent after 3 years; [3]), the breakdown of the Rlm7 gene appeared much less rapid than that observed previously for Rlm1 at the national level in France (less than 20% of isolates virulent after 10 years of widespread use of Rlm7. Among the possible reasons for this unexpected durability of Rlm7, the role of the negative interaction between AvrLm7 and AvrLm3 [4] was explored here. A detailed knowledge of molecular mechanisms responsible for virulence against Rlm3 and Rlm7 was obtained and suggested that pyramiding of the two resistance genes could have a negative impact on Rlm7 durability while the use of both resistance genes in different varieties could have slowed down the breakdown. These population and molecular information are currently used in an epidemiological model to test our hypotheses. References [1] Rouxel T, et al (2003). Eur J Plant Pathol 109:871-881. [2] Balesdent MH, et al (2006). Eur J Plant Pathol 114:53-65. [3] Daverdin G, et al (2012).. PLoS Pathog 8: e1003020. [4] Plissonneau C, et al (2016) New Phytol 209:1613-1624

International audience; Leptosphaeria maculans, the causal agent of blackleg disease, interacts with Brassica napus (oilseed rape, canola) and other Brassica hosts in a gene-for-gene manner. The avirulence gene AvrLmJ1 has been cloned previously and shown to interact with an unidentified Brassica juncea resistance gene. In this study, we show that the AvrLmJ1 gene maps to the same position as the AvrLm5 locus. Furthermore, isolates complemented with the AvrLmJ1 locus confer avirulence towards B. juncea genotypes harbouring Rlm5. These findings demonstrate that AvrLmJ1 is AvrLm5 and highlight the need for shared resources to characterize accurately avirulence and/or resistance genes.

Sequence variation in Zmr1 promoter contributes to differences in melanin accumulation. Means and standard errors of gray values (0 = black, 255 = white) of 3D1, 3D7, 3D7Δzmr1, 3D7Δzmr1 + Zmr13D1, and 3D7Δzmr1 + Zmr13D7, 7 days post inoculation (dpi) based on at least 20 colonies. Asterisks (*) and pluses (+) indicate significant differences in mean gray values of each strain with respect to the mean gray value of 3D7 and 3D7Δzmr1, respectively (Kruskal-Wallis, p values ≤ 0.05). The experiment was performed twice with similar results. NA = not applicable. (PDF 335 kb)

Meiosis is one of the most conserved molecular processes in eukaryotes. The fidelity of pairing and segregation of homologous chromosomes has a major impact on the proper transmission of genetic information. Aberrant chromosomal transmission can have major phenotypic consequences, yet the mechanisms are poorly understood. Fungi are excellent models to investigate processes of chromosomal transmission, because many species have highly polymorphic genomes that include accessory chromosomes. Inheritance of accessory chromosomes is often unstable and chromosomal losses have little impact on fitness. We analyzed chromosomal inheritance in 477 progeny coming from two crosses of the fungal wheat pathogen Zymoseptoria tritici. For this, we developed a high-throughput screening method based on restriction site-associated DNA sequencing that generated dense coverage of genetic markers along each chromosome. We identified rare instances of chromosomal duplications (disomy) in core chromosomes. Accessory chromosomes showed high overall frequencies of disomy. Chromosomal rearrangements were found exclusively on accessory chromosomes and were more frequent than disomy. Accessory chromosomes present in only one of the parents in an analyzed cross were inherited at significantly higher rates than the expected 1:1 segregation ratio. Both the chromosome and the parental background had significant impacts on the rates of disomy, losses, rearrangements, and distorted inheritance. We found that chromosomes with higher sequence similarity and lower repeat content were inherited more faithfully. The large number of rearranged progeny chromosomes identified in this species will enable detailed analyses of the mechanisms underlying chromosomal rearrangement., Genome Biology and Evolution, 10 (6), ISSN:1759-6653

3D7 accumulates more melanin than 3D1. Means and standard errors of gray values (0 = black, 255 = white) of at least 60 colonies of Z. tritici strains 3D1 and 3D7 at 7–12 days post inoculation (dpi). The experiment was performed three times with similar results. Asterisks indicate significant differences between 3D1 and 3D7 at each time point according to Kruskal-Wallis test (p values ≤ 0.05). n = number of colonies analyzed. (PDF 110 kb)

Non-melanized mutants grow faster than the wild-types. Radial growth rates (slope of the curves) of wild-type strains and melanin-deficient mutants (3D1Δzmr1 and 3D7Δzmr1) were obtained by plotting radial size (mm) of the colonies over time. The fit of the radial growth curve estimated using a linear model was evaluated using Pearson’s correlation coefficient (r2 value). Asterisks (*) indicate significant differences in growth rate (slope) between the wild-type and the mutant according to ANCOVA analysis (p values ≤ 0.05). The experiment was performed three times with similar results. (PDF 332 kb)

Zmr1 expression levels are lower in 3D1 than in 3D7 at 7 days post inoculation. Mean and standard errors (se) of the relative quantification (RQ, fold change in expression level of Zmr1 with respect to the Zmr1 expression levels of the strain 3D1 at 7 days post inoculation, dpi) of the expression of Zmr1 in 3D1 and 3D7 at 7 and 9 dpi. The experiment was performed three times with similar results. (PDF 31 kb)

Reduced expression of genes in the Pks1 cluster in Δzmr1 mutants. Mean and standard error (se) of CPM (counts per million mapped reads) values of genes significantly downregulated (false discovery rates, FDR ≤ 0.05) in both Δzmr1 mutants compared to the wild-type strains 3D1 and 3D7. Means and standard errors of the mean of three independent replicates are indicated. Genes previously shown to be involved in melanin biosynthesis are shown in bold. (PDF 423 kb)

The transposable element insertion upstream of Zmr1 in 3D1 downregulates Zmr1 expression. Mean gray values (0 = black, 255 = white) based on at least 35 colonies of the wild-types 3D1 and 3D7, three independent TE deletion mutants in the 3D1 background (3D1 ΔTE #93, #131, #239) and the two ectopic controls (3D1+ Hyg #3 and #6), 7 days post inoculation. Asterisks (*) indicate that the strains are significantly darker than the wild-type 3D1 (Kruskal-Wallis, p value ≤ 0.05). The experiment was performed three times with 3D1ΔTE #93 and twice with #131 and #239. (PDF 330 kb)

During plant infection, pathogens secrete an arsenal of effectors, key elements of pathogenesis which modulate innate immunity of the plant and facilitate infection. Fungal effector genes typically encode small proteins, predicted to be secreted, with no homology in databases, and absence of known motif. As such their function or role in pathogenesis is mostly unknown. The phytopathogenic ascomycete Leptosphaeria maculans is the causal agent of stem canker of oilseed rape. More than 650 putative effector-encoding genes have been identified in its genome, 7 of them corresponding to avirulence proteins. We develop a project aiming at elucidating the involvement of L. maculans effectors in pathogenicity through the structural and functional characterization of a few major effector proteins and the determination of their interactants. Our strategy is illustrated here with AvrLm4-7, a 143 amino-acid long secreted protein important for fungal fitness and recognized by two oilseed rape resistance proteins, Rlm4 and Rlm7. One single amino-acid change is sufficient to lose recognition by Rlm4 while maintaining recognition by Rlm7. 3D-structure of an isoform of AvrLm4-7 only recognized by Rlm7 was previously determined, allowing us to define regions implicated in recognition by Rlm7 and translocation into plant cell. We recently determined the 3D-structure of another isoform of AvrLm4-7 recognized both by Rlm4 and Rlm7, showing that the amino-acid change allowing to escape Rlm4-recognition was located on an external loop and did not change the overall structure of the protein. AvrLm4-7 was also recently shown to suppress recognition of another L. maculans avirulence gene, AvrLm3, by its cognate resistance gene Rlm3, leading us to hypothesize a suppression of Effector-Triggered Immunity (ETI) by AvrLm4-7. In order to test that hypothesis, we transiently expressed AvrLm4-7 and several cell-death inducers in Nicotiana benthamiana epidermal cells: AvrLm4-7 was able to suppress cell death induced by BAX and AvrPto. We also generated transgenic lines of Arabidopsis thaliana constitutively expressing AvrLm4-7 and are currently characterizing the lines for their susceptibility to pathogens with contrasted lifestyles and for their ability to suppress recognition of Pseudomonas syringae avirulence proteins. A better understanding of the role of an effector implicated in the masking of another effector will allow us to develop alternative strategies to genetically control stem canker disease.

The stem canker disease, caused by Leptosphaeria maculans, is one of the most devastating diseases of oilseed rape (canola). It colonizes the plant in two stages: a short and early colonisation stage corresponding to cotyledon or leaf colonisation, and a late colonisation stage during which the fungus colonises systemically and symptomlessly the plant during several months before stem canker appears. To date, determinants of the late colonisation stage remain poorly understood. By a transcriptomic approach, we previously identified two waves of effector candidate expression during the early and late colonisation stages (Gervais et al, 2016). The late effector candidates are located in gene-rich genomic regions, whereas the early effector genes are located in gene-poor regions of the genome. Among the late effector candidates identified, we selected 6 genes for further characterization. We created mutants silenced for these effector candidates. For one of these genes, its expression level correlated negatively with the size of the necrosis observed in the stem. The identification of new effector genes would contribute to the identification of new resistance genes specific to these effectors. To easily identify matching resistance genes in oilseed rape, we created transgenic isolates expressing these 6 late effectors at the early steps of infection to provide medium-throughput strategies to screen more efficiently different cultivars. Preliminary results indicate that some cultivars with adult resistance were more resistant to these transgenic isolates in cotyledon assays. With this approach, we also identified a cultivar carrying a specific resistance to one these 6 effector candidates. Reference Gervais, J., Plissonneau, C., Linglin, J., Meyer, M., Labadie, K., Cruaud, C., Fudal, I., Rouxel, T. and Balesdent, M.H. (2016) Different waves of effector genes with contrasted genomic location are expressed by Leptosphaeria maculans during cotyledon and stem colonization of oilseed rape. Mol. Plant Pathol.

National audience; Plant pathogens secrete effector proteins into host tissues to promote infection through the manipulation of host processes. Sequencing and analyses of the genomes of fungal phytopathogens have shown that they contains tens to hundreds of genes predicted to encode putative effectors. Moreover, global analyses of gene expression revealed that several waves of concerted expression of effector genes take place during host invasion. In sharp contrast with the situation described inplanta, the expression of the effectors is difficult to detect and quantify in axenic cultures because their genes are expressed at a very low level. In the present study, we investigate biotic and abiotic factors that may relieve suppression of expression of effectors during axenic growth. Biotic factors (such as carbon source, nitrate source, antibiotics) as well as abiotic factors (pH, temperature) can influence their expression. Of major interest, incubation of the fungal mycelium with 1ug/ml of an antibiotic of the aminoglycoside family allowed an increase of effector gene expression 20-fold to 60- fold compared to regular axenic growth. An RNAseq analysis aiming at identifying the set of effectors up- and down-regulated in a culture medium supplemented or not with the antibiotic has been performed and results obtained will be presented. This simple system could be a good starting point to characterize the plant signals that trigger fungal effector gene expression.

Leptosphaeria maculans is the fungus responsible for the stem canker disease of oilseed rape (Brassica napus). AvrLm3 and AvrLm7, two avirulence effector genes of L. maculans, are involved in an unusual relationship: the presence of AvrLm7 suppresses the Rlm3-mediated resistance of AvrLm3. Following the large-scale cropping of Rlm7 cultivars in Europe, the on-going breakdown of the Rlm7 resistance and the concomitant inactivation of AvrLm7 was accompanied by the resurgence of isolates expressing the AvrLm3 phenotype. Here, we evaluated adaptation mechanisms to escape Rlm3-mediated ETI by assessing AvrLm3 polymorphism in a collection of 236 L. maculans isolates. In isolates virulent towards both Rlm3 and Rlm7 (a3a7), the loss of the Rlm3-mediated resistance response was due to two distinct mechanisms. First, when AvrLm7 was inactivated (deletion or inactivating mutations), amino acid substitutions in AvrLm3 generated virulent isoforms of the protein. Secondly, point mutations in AvrLm7 could maintain the masking of the AvrLm3 phenotype while also allowing escape from the Rlm7 recognition. This virulence mechanism was found in 56.4% of the a3a7 isolates, and these contained an avirulent allele of AvrLm3. Signatures of positive selection were observed in AvrLm3 and, in spite of the telomeric location of AvrLm3, no a3a7 isolates exhibited deletion events or inactivating mutations in AvrLm3 as commonly observed for other avirulence genes of L. maculans. The complex evolutionary mechanisms enabling L. maculans to escape Rlm3-mediated resistance while preserving AvrLm3 sequence integrity, along with observed reduced aggressiveness of isolates silenced for AvrLm3, serves to emphasize importance of this effector in pathogenicity towards B. napus.

Leptosphaeria maculans, the causal agent of stem canker disease, colonizes oilseed rape (Brassica napus) in two stages: a short and early colonization stage corresponding to cotyledon or leaf colonization, and a late colonization stage during which the fungus colonizes systemically and symptomlessly the plant during several months before stem canker appears. To date, the determinants of the late colonization stage are poorly understood; L. maculans may either successfully escape plant defences, leading to stem canker development, or the plant may develop an ‘adult‐stage’ resistance reducing canker incidence. To obtain an insight into these determinants, we performed an RNA‐sequencing (RNA‐seq) pilot project comparing fungal gene expression in infected cotyledons and in symptomless or necrotic stems. Despite the low fraction of fungal material in infected stems, sufficient fungal transcripts were detected and a large number of fungal genes were expressed, thus validating the feasibility of the approach. Our analysis showed that all avirulence genes previously identified are under‐expressed during stem colonization compared with cotyledon colonization. A validation RNA‐seq experiment was then performed to investigate the expression of candidate effector genes during systemic colonization. Three hundred and seven ‘late’ effector candidates, under‐expressed in the early colonization stage and over‐expressed in the infected stems, were identified. Finally, our analysis revealed a link between the regulation of expression of effectors and their genomic location: the ‘late’ effector candidates, putatively involved in systemic colonization, are located in gene‐rich genomic regions, whereas the ‘early’ effector genes, over‐expressed in the early colonization stage, are located in gene‐poor regions of the genome.

Epidemics caused by fungal plant pathogens pose a major threat to agro-ecosystems and impact global food security. High-throughput sequencing enabled major advances in understanding how pathogens cause disease on crops. Hundreds of fungal genomes are now available and analyzing these genomes highlighted the key role of effector genes in disease. Effectors are small secreted proteins that enhance infection by manipulating host metabolism. Fungal genomes carry 100s of putative effector genes, but the lack of homology among effector genes, even for closely related species, challenges evolutionary and functional analyses. Furthermore, effector genes are often found in rapidly evolving chromosome compartments which are difficult to assemble. We review how population and comparative genomics toolsets can be combined to address these challenges. We highlight studies that associated genome-scale polymorphisms with pathogen lifestyles and adaptation to different environments. We show how genome-wide association studies can be used to identify effectors and other pathogenicity-related genes underlying rapid adaptation. We also discuss how the compartmentalization of fungal genomes into core and accessory regions shapes the evolution of effector genes. We argue that an understanding of genome evolution provides important insight into the trajectory of host-pathogen co-evolution. ISSN:1664-462X

International audience; Leptosphaeria maculans, causal agent of blackleg disease, colonises oilseed rape (Brassica napus) in two stages: a short and early colonisation stage corresponding to cotyledon or leaf colonisation leading to leaf spot development, and a late colonisation stage during which the fungus colonises systemically and symptomlessly the plant during several months before stem canker appears. To date, determinants of the late colonisation stage are poorly understood; L. maculans may either successfully escape plant defences leading to the stem canker development, or the plant can develop an “adult-stage” resistance reducing canker incidence. Here, we hypothesized that L. maculans deploys effectors different from the one deployed in the early colonisation to enable the long symptomless stem colonisation. To get insight into these determinants, we performed a RNA-seq pilot project comparing fungal gene expression in infected cotyledons and in infected stems. Despite the low fraction of fungal material in infected stems with or without symptoms, enough fungal transcripts were detected to conduct a RNA-seq analysis focused on effector genes. With this approach 175 late effector candidates, under-expressed in the early colonisation stage and over-expressed in the infected stems, were identified. These new effector genes, putatively involved in the systemic colonisation, are located in gene-rich genomic regions, whereas the early effector genes involved in cotyledon colonisation are located in gene-poor regions of the genome. Our analysis reveals a link between the expression pattern of effectors and their genomic location.

National audience; Leptosphaeria maculans is responsible for the stem canker (phoma) disease of oilseed rape (Brassica napus). AvrLm3 and AvrLm4-7, two avirulence effector genes of L. maculans, are involved in an unusual relationship: the presence of AvrLm4-7 induces the Rlm7-mediated resistance but suppresses the Rlm3-mediated recognition (Plissonneau et al., 2015; doi: 10.1111/nph.13736). To investigate this relationship, we assessed the sequence diversity of AvrLm3 in L. maculans populations. The analysis of more than 200 isolates revealed a high level of allelic polymorphism for this gene, but no deletion event nor inactivating mutations were found. This observation contrasts with the presence/absence or RIP-(Repeat-induced-point-mutation) inactivation polymorphisms usually found for other L. maculans avirulence genes and questioned the role of AvrLm3 in fungal fitness. Two distinct mechanisms responsible for the «double virulent» phenotype (i.e. virulent toward both Rlm3 and Rlm7, «a3a7») were identified. In isolates displaying an inactivation of AvrLm4-7, amino acid changes in AvrLm3 were responsible for the virulent phenotype towards Rlm3. However, 56% of the a3a7 isolates displayed an avirulent allele of AvrLm3, combined with point mutations in AvrLm4-7. Such situations allow the fungus to escape Rlm7-mediated resistance while maintaining the suppression of the AvrLm3 phenotype effective. They also allow the fungus to escape to the recognition by two resistance genes while keeping the secretion of the two corresponding effector proteins. The complex evolutionary mechanisms displayed by L. maculans to escape Rlm3-mediated resistance while preserving AvrLm3 sequence integrity, along with the reduced virulence in isolates silenced for AvrLm3 confirmed the importance of this effector in pathogenicity towards B. napus. This work was funded by the INRA SMaCH metaprogram «K-masstec» and the CTPS project «ICOSCOP»

Fungal plant pathogens rapidly evolve virulence on resistant hosts through mutations in genes encoding proteins that modulate the host immune responses. The mutational spectrum likely includes chromosomal rearrangements responsible for gains or losses of entire genes. However, the mechanisms creating adaptive structural variation in fungal pathogen populations are poorly understood. We used complete genome assemblies to quantify structural variants segregating in the highly polymorphic fungal wheat pathogen Zymoseptoria tritici. The genetic basis of virulence in Z. tritici is complex, and populations harbor significant genetic variation for virulence; hence, we aimed to identify whether structural variation led to functional differences. We combined single-molecule real-time sequencing, genetic maps, and transcriptomics data to generate a fully assembled and annotated genome of the highly virulent field isolate 3D7. Comparative genomics analyses against the complete reference genome IPO323 identified large chromosomal inversions and the complete gain or loss of transposable-element clusters, explaining the extensive chromosomal-length polymorphisms found in this species. Both the 3D7 and IPO323 genomes harbored long tracts of sequences exclusive to one of the two genomes. These orphan regions contained 296 genes unique to the 3D7 genome and not previously known for this species. These orphan genes tended to be organized in clusters and showed evidence of mutational decay. Moreover, the orphan genes were enriched in genes encoding putative effectors and included a gene that is one of the most upregulated putative effector genes during wheat infection. Our study showed that this pathogen species harbored extensive chromosomal structure polymorphism that may drive the evolution of virulence., IMPORTANCE Pathogen outbreak populations often harbor previously unknown genes conferring virulence. Hence, a key puzzle of rapid pathogen evolution is the origin of such evolutionary novelty in genomes. Chromosomal rearrangements and structural variation in pathogen populations likely play a key role. However, identifying such polymorphism is challenging, as most genome-sequencing approaches only yield information about point mutations. We combined long-read technology and genetic maps to assemble the complete genome of a strain of a highly polymorphic fungal pathogen of wheat. Comparisons against the reference genome of the species showed substantial variation in the chromosome structure and revealed large regions unique to each assembled genome. These regions were enriched in genes encoding likely effector proteins, which are important components of pathogenicity. Our study showed that pathogen populations harbor extensive polymorphism at the chromosome level and that this polymorphism can be a source of adaptive genetic variation in pathogen evolution.

Leptosphaeria maculans is a Dothideomycete responsible for stem canker on oilseed rape (Brassica napus). Genetic control, encompassing mostly the use of major resistance genes, is the most effective method to control this pathogen. When the first cultivars harboring Rlm7 have been deployed in the early 2000's, all of the European isolates were avirulent towards this resistance gene and AvrLm3 was considered to be absent from the populations. In 2012, Daverdin et al. showed that L. maculans has the ability to rapidly overcome the Rlm7 resistance whenever a strong selection pressure is applied. The phenotyping of isolates sampled by Daverdin et al. also showed that more than 98 % of the isolates virulent towards Rlm7 had become avirulent towards Rlm3. This result led to the hypothesis that the presence of AvrLm4-7 can suppress AvrLm3 recognition by Rlm3 and this was validated by the complementation of an isolate avirulent towards Rlm3 with a functional allele of AvrLm4-7. AvrLm3 is genetically linked to AvrLm4-7, at a distance of 20 cM. However, no candidate gene was identified before the beginning of my PhD. The objective of my PhD project was to identify AvrLm3, in order to better understand the antagonistic relationship between AvrLm3 and AvrLm4-7 phenotypes. The combination of genetic and genomic approaches (genetic mapping, RNA-seq, de novo sequencing of an isolate avirulent towards Rlm3 and BAC clone sequencing) allowed me to identify AvrLm3, whose sequence was absent from the reference genome assembly. AvrLm3 has common characteristics with others genes encoding fungal effector: it codes for a small, cysteine-rich protein, is highly expressed at early infection stages and shows no homology with others fungal genes. Although the mechanism allowing the suppression of AvrLm3 recognition due to the presence of AvrLm4-7 has not been elucidated, several hypotheses were invalidated: the presence of AvrLm4-7 has no impact on AvrLm3 expression and the two avirulence proteins do not interact physically.The sampling of field isolates in 2012 and 2013 showed the currently ongoing breakdown of Rlm7, with ca. 10% of virulent isolates, and the resurgence of the AvrLm3 phenotype in these populations, only 0.5 % of isolates being virulent towards both Rlm3 and Rlm7. A large collection of isolates from worldwide origin was genotyped for AvrLm3. It revealed that all isolates possess AvrLm3, with a high level of allelic diversity. The resurgence of the avirulent phenotype towards Rlm3 following the breakdown the Rlm7, along with the important role of AvrLm3 and AvrLm4-7 in fungal fitness, suggest that the antagonistic relationship between these two avirulence phenotypes could be an opportunity to propose original strategies to increase the durability of Rlm3 and Rlm7, by alternating both resistance genes at the landscape level or using pyramiding strategies. However, we identified novel isoforms of AvrLm4-7 allowing the fungus to escape Rlm7 recognition, while maintaining the suppression of Rlm3 recognition effective. This work allowed to better characterize an unusual gene-for-gene relationship. Indeed, only one other example of antagonism between avirulence phenotypes has been identified to date, in the phytopathogenic fungus Fusarium oxysporum (Houterman et al., 2008). The diversity of mechanisms allowing L. maculans to overcome the Rlm3 resistance illustrates perfectly the complexity of the arms race between plants and pathogens.; Leptosphaeria maculans est l'agent de la nécrose du collet des crucifères, principale maladie fongique du colza (Brassica napus). Lorsque des variétés de colza possédant la résistance Rlm7 ont été commercialisées au début des années 2000, la totalité des souches européennes étaient avirulentes vis-à-vis de ce gène de résistance. A l'inverse, AvrLm3 était considéré comme absent de ces populations. Un précédent projet de thèse réalisé dans l'équipe a montré la rapide capacité d'adaptation de L. maculans à la pression de sélection exercée par Rlm7 (Daverdin et al. 2012) et le phénotypage des populations isolées lors de cette étude avait montré que 98% des souches virulentes vis-à-vis de Rlm7 étaient avirulentes vis-à-vis de Rlm3. L'hypothèse d'un masquage du phénotype avirulent dû à la présence d'AvrLm4-7 a alors été validée par la complémentation d'une souche avirulente vis-à-vis de Rlm3 par AvrLm4-7. Sur ces bases, l'objectif de mon projet de thèse était d'identifier AvrLm3 afin d'analyser l'antagonisme entre les phénotypes AvrLm3 et AvrLm4-7, ainsi que les mécanismes menant au contournement de Rlm3 et Rlm7.Le gène AvrLm3 avait été identifié comme génétiquement lié à AvrLm4-7. Toutefois il n'avait pas été possible d'identifier un gène candidat. Par la combinaison d'approches de génétique et de génomique (clonage positionnel, RNA-seq, séquençage de novo d'une souche avirulente et de clones BAC), j'ai pu identifier AvrLm3, dont la séquence était absente du génome de référence de L. maculans. AvrLm3 est très fortement exprimé lors des phases précoces de l'infection et code pour une petite protéine sécrétée et ne présentant pas d'homologies avec d'autres protéines fongiques. Ce gène a donc les caractéristiques classiques des gènes codant pour des effecteurs, mais présente la particularité d'être le premier gène d'avirulence de L. maculans localisé en région télomérique. Bien que le mécanisme par lequel la présence du gène AvrLm4-7 supprime la reconnaissance d'AvrLm3 dans l'interaction entre B. napus et L. maculans n'ait pas été élucidé au terme de ma thèse, plusieurs hypothèses ont pu être invalidées, notamment l'absence de régulation de l'expression d'AvrLm3 par AvrLm4-7. De plus, une approche double hybride a mis en évidence une absence d'interaction directe entre les deux protéines d'avirulence. L'isolement de souches de L. maculans issues de parcelles expérimentales en 2012 et 2013 a mis en évidence le contournement actuel de la résistance Rlm7, avec environ 10 % de souches virulentes. Le phénotypage de ces collections a confirmé la présence majoritaire d'un allèle avirulent d'AvrLm3 chez celles-ci, moins de 0,5 % des souches isolées étant virulentes à la fois vis-à-vis de Rlm3 et de Rlm7. L'analyse de 592 souches d'origine mondiale a montré qu'AvrLm3 est toujours présent chez L. maculans, sous diverses formes alléliques, suggérant un rôle majeur d'AvrLm3 dans la fitness fongique. La résurgence du phénotype avirulent vis-à-vis de Rlm3 suite au contournement de Rlm7 fait envisager l'opportunité d'exploiter l'interaction entre les deux gènes d'avirulence pour une gestion plus durable des gènes de résistance de B. napus, par l'alternance de cultivars possédant Rlm3 et Rlm7 ou le pyramidage de ces deux gènes dans des variétés de colza. De plus, le rôle démontré d'AvrLm3 et d'AvrLm4-7 dans l'agressivité lors de l'infection suppose un fort coût de fitness lié à la virulence. Toutefois, des mécanismes originaux de compensation permettant au champignon d'échapper à la reconnaissance par Rlm3 et Rlm7 tout en conservant a priori fonctionnelle la fonction effectrice d'AvrLm3 et AvrLm4-7 ont été identifiés. Les résultats obtenus lors de ce travail de thèse ont permis la meilleure caractérisation d'une interaction gène-pour-gène inhabituelle. La diversité des mécanismes moléculaires permettant à L. maculans de contourner la résistance Rlm3 illustre la complexité de la course aux armements entre les plantes et les agents pathogènes.

National audience

National audience; absent

National audience

National audience

Leptosphaeria maculans est un champignon ascomycète responsable de la nécrose du collet du colza, principale maladie sur cette culture. La méthode de lutte la plus utilisée contre cet agent pathogène est la lutte génétique. Celle-ci repose sur l'interaction entre un gène de résistance Rlm chez la plante et un gène d'avirulence AvrLm chez le champignon. L'un d'eux, AvrLm4-7, joue un rôle important dans la fitness fongique1 et a été cloné2. Récemment, nous avons mis en évidence une corrélation négative entre les phénotypes induits par deux gènes d'avirulence, AvrLm3 et AvrLm4-73. Lorsqu'une souche possède un allèle fonctionnel du gène AvrLm4-7, le phénotype induit par AvrLm3 est masqué. En cas de contournement de la résistance Rlm7, le phénotype AvrLm3 apparait. Ceci offre des perspectives originales pour la gestion durable des résistances variétales Rlm3 et Rlm7. Afin d'élucider les mécanismes moléculaires sous-jacents à cette interaction négative, deux approches ont été menées en parallèle. D'une part, nous analysons des événements de mutation du gène AvrLm4-7 au sein de populations sélectionnées sur des génotypes de colza Rlm3 et/ou Rlm7 et présentant le phénotype simple virulent (AvrLm3-avrLm7) ou double virulent (avrLm3-avrLm7). Cette analyse a permis de mettre en évidence une mutation ponctuelle dans la protéine AvrLm4-7 qui semble spécifique aux souches double virulentes. D'autre part, nous avons initié le clonage du gène AvrLm3 par une approche de clonage positionnel classique et par des approches de génomique visant à identifier des effecteurs candidats spécifiquement exprimés chez une souche au phénotype AvrLm3, lors de phases primaires de l'infections. 40 gènes candidats ont été identifiés et sont actuellement en cours d'analyse.

Epidemics caused by fungal plant pathogens pose a major threat to agro-ecosystems and impact global food security. High-throughput sequencing enabled major advances in understanding how pathogens cause disease on crops. Hundreds of fungal genomes are now available and analyzing these genomes highlighted the key role of effector genes in disease. Effectors are small secreted proteins that enhance infection by manipulating host metabolism. Fungal genomes carry 100s of putative effector genes, but the lack of homology among effector genes, even for closely related species, challenges evolutionary and functional analyses. Furthermore, effector genes are often found in rapidly evolving chromosome compartments which are difficult to assemble. We review how population and comparative genomics toolsets can be combined to address these challenges. We highlight studies that associated genome-scale polymorphisms with pathogen lifestyles and adaptation to different environments. We show how genome-wide association studies can be used to identify effectors and other pathogenicity-related genes underlying rapid adaptation. We also discuss how the compartmentalization of fungal genomes into core and accessory regions shapes the evolution of effector genes. We argue that an understanding of genome evolution provides important insight into the trajectory of host-pathogen co-evolution., Frontiers in Plant Science, 8, ISSN:1664-462X

Fungal plant pathogens rapidly evolve virulence on resistant hosts through mutations in genes encoding proteins that modulate the host immune responses. The mutational spectrum likely includes chromosomal rearrangements responsible for gains or losses of entire genes. However, the mechanisms creating adaptive structural variation in fungal pathogen populations are poorly understood. We used complete genome assemblies to quantify structural variants segregating in the highly polymorphic fungal wheat pathogen Zymoseptoria tritici. The genetic basis of virulence in Z. tritici is complex, and populations harbor significant genetic variation for virulence; hence, we aimed to identify whether structural variation led to functional differences. We combined single-molecule real-time sequencing, genetic maps, and transcriptomics data to generate a fully assembled and annotated genome of the highly virulent field isolate 3D7. Comparative genomics analyses against the complete reference genome IPO323 identified large chromosomal inversions and the complete gain or loss of transposable-element clusters, explaining the extensive chromosomal-length polymorphisms found in this species. Both the 3D7 and IPO323 genomes harbored long tracts of sequences exclusive to one of the two genomes. These orphan regions contained 296 genes unique to the 3D7 genome and not previously known for this species. These orphan genes tended to be organized in clusters and showed evidence of mutational decay. Moreover, the orphan genes were enriched in genes encoding putative effectors and included a gene that is one of the most upregulated putative effector genes during wheat infection. Our study showed that this pathogen species harbored extensive chromosomal structure polymorphism that may drive the evolution of virulence., mBio, 7 (5), ISSN:2150-7511, ISSN:2161-2129

Background Fungal plant pathogens pose major threats to crop yield and sustainable food production if they are highly adapted to their host and the local environment. Variation in gene expression contributes to phenotypic diversity within fungal species and affects adaptation. However, very few cases of adaptive regulatory changes have been reported in fungi and the underlying mechanisms remain largely unexplored. Fungal pathogen genomes are highly plastic and harbor numerous insertions of transposable elements, which can potentially contribute to gene expression regulation. In this work, we elucidated how transposable elements contribute to variation in melanin accumulation, a quantitative trait in fungi that affects survival under stressful conditions. Results We demonstrated that differential transcriptional regulation of the gene encoding the transcription factor Zmr1, which controls expression of the genes in the melanin biosynthetic gene cluster, is responsible for variation in melanin accumulation in the fungal plant pathogen Zymoseptoria tritici. We show that differences in melanin levels between two strains of Z. tritici are due to two levels of transcriptional regulation: (1) variation in the promoter sequence of Zmr1 and (2) an insertion of transposable elements upstream of the Zmr1 promoter. Remarkably, independent insertions of transposable elements upstream of Zmr1 occurred in 9% of Z. tritici strains from around the world and negatively regulated Zmr1 expression, contributing to variation in melanin accumulation. Conclusions Our studies identified two levels of transcriptional control that regulate the synthesis of melanin. We propose that these regulatory mechanisms evolved to balance the fitness costs associated with melanin production against its positive contribution to survival in stressful environments., BMC Biology, 16 (1), ISSN:1741-7007