A Single Nonsynonymous Substitution in the RNA-Dependent RNA Polymerase of Potato virus Y Allows the Simultaneous Breakdown of Two Different Forms of Antiviral Resistance in Capsicum annuum.

The dominant Pvr4 gene in pepper (Capsicum annuum) confers resistance to members of six potyvirus species, all of which belong to the Potato virus Y (PVY) phylogenetic group. The corresponding avirulence factor in the PVY genome is the NIb cistron (i.e., RNA-dependent RNA polymerase). Here, we describe a new source of potyvirus resistance in the Guatemalan accession C. annuum cv. PM949. PM949 is resistant to members of at least three potyvirus species, a subset of those controlled by Pvr4. The F1 progeny between PM949 and the susceptible cultivar Yolo Wonder was susceptible to PVY, indicating that the resistance is recessive. The segregation ratio between resistant and susceptible plants observed in the F2 progeny matched preferably with resistance being determined by two unlinked recessive genes independently conferring resistance to PVY. Inoculations by grafting resulted in the selection of PVY mutants breaking PM949 resistance and, less efficiently, Pvr4-mediated resistance. The codon substitution E472K in the NIb cistron of PVY, which was shown previously to be sufficient to break Pvr4 resistance, was also sufficient to break PM949 resistance, a rare example of cross-pathogenicity effect. In contrast, the other selected NIb mutants showed specific infectivity in PM949 or Pvr4 plants. Comparison of Pvr4 and PM949 resistance, which share the same target in PVY, provides interesting insights into the determinants of resistance durability.

Concurrent evolution of resistance and tolerance to potato virus Y in Capsicum annuum revealed by genome-wide association.

We performed a genome-wide association study of pepper (Capsicum annuum) tolerance to potato virus Y (PVY). For 254 pepper accessions, we estimated the tolerance to PVY as the coefficient of regression of the fresh weight (or height) of PVY-infected and mock-inoculated plants against within-plant virus load. Small (strongly negative) coefficients of regression indicate low tolerance because plant biomass or growth decreases sharply as virus load increases. The tolerance level varied largely, with some pepper accessions showing no symptoms or fairly mild mosaics, whereas about half (48%) of the accessions showed necrotic symptoms. We found two adjacent single-nucleotide polymorphisms (SNPs) at one extremity of chromosome 9 that were significantly associated with tolerance to PVY. Similarly, in three biparental pepper progenies, we showed that the induction of necrosis on PVY systemic infection segregated as a monogenic trait determined by a locus on chromosome 9. Our results also demonstrate the existence of a negative correlation between resistance and tolerance among the cultivated pepper accessions at both the phenotypic and genetic levels. By comparing the distributions of the tolerance-associated SNP alleles and previously identified PVY resistance-associated SNP alleles, we showed that cultivated pepper accessions possess favourable alleles for both resistance and tolerance less frequently than expected under random associations, while the minority of wild pepper accessions frequently combined resistance and tolerance alleles. This divergent evolution of PVY resistance and tolerance could be related to pepper domestication or farmer's selection.

Genome-wide association mapping of QTLs implied in potato virus Y population sizes in pepper: evidence for widespread resistance QTL pyramiding.

In this study, we looked for genetic factors in the pepper (Capsicum annuum) germplasm that control the number of potato virus Y (PVY) particles entering the plant (i.e. effective population size at inoculation) and the PVY accumulation at the systemic level (i.e. census population size). Using genotyping-by-sequencing (GBS) in a core collection of 256 pepper accessions, we obtained 10 307 single nucleotide polymorphisms (SNPs) covering the whole genome. Genome-wide association studies (GWAS) detected seven SNPs significantly associated with the virus population size at inoculation and/or systemic level on chromosomes 4, 6, 9 and 12. Two SNPs on chromosome 4 associated with both PVY population sizes map closely to the major resistance gene pvr2 encoding the eukaryotic initiation factor 4E. No obvious candidates for resistance were identified in the confidence intervals for the other chromosomes. SNPs detected on chromosomes 6 and 12 colocalized with resistance quantitative trait loci (QTLs) previously identified with a biparental population. These results show the efficiency of GBS and GWAS in C. annuum, indicate highly consistent results between GWAS and classical QTL mapping, and suggest that resistance QTLs identified with a biparental population are representative of a much larger collection of pepper accessions. Moreover, the resistance alleles at these different loci were more frequently combined than expected by chance in the core collection, indicating widespread pyramiding of resistance QTLs and widespread combination of resistance QTLs and major effect genes. Such pyramiding may increase resistance efficiency and/or durability.

Impact of genetic drift, selection and accumulation level on virus adaptation to its host plants.

The efficiency of plant major resistance genes is limited by the emergence and spread of resistance-breaking mutants. Modulation of the evolutionary forces acting on pathogen populations constitutes a promising way to increase the durability of these genes. We studied the effect of four plant traits affecting these evolutionary forces on the rate of resistance breakdown (RB) by a virus. Two of these traits correspond to virus effective population sizes (Ne ) at either plant inoculation or during infection. The third trait corresponds to differential selection exerted by the plant on the virus population. Finally, the fourth trait corresponds to within-plant virus accumulation (VA). These traits were measured experimentally on Potato virus Y (PVY) inoculated to a set of 84 pepper doubled-haploid lines, all carrying the same pvr23 resistance gene, but having contrasting genetic backgrounds. The lines showed extensive variation for the rate of pvr23 RB by PVY and for the four other traits of interest. A generalized linear model showed that three of these four traits, with the exception of Ne at inoculation, and several pairwise interactions between them had significant effects on RB. RB increased with increasing values of Ne during plant infection or VA. The effect of differential selection was more complex because of a strong interaction with VA. When VA was high, RB increased as the differential selection increased. An opposite relationship between RB and differential selection was observed when VA was low. This study provides a framework to select plants with appropriate virus evolution-related traits to avoid or delay RB.

Estimating virus effective population size and selection without neutral markers.

By combining high-throughput sequencing (HTS) with experimental evolution, we can observe the within-host dynamics of pathogen variants of biomedical or ecological interest. We studied the evolutionary dynamics of five variants of Potato virus Y (PVY) in 15 doubled-haploid lines of pepper. All plants were inoculated with the same mixture of virus variants and variant frequencies were determined by HTS in eight plants of each pepper line at each of six sampling dates. We developed a method for estimating the intensities of selection and genetic drift in a multi-allelic Wright-Fisher model, applicable whether these forces are strong or weak, and in the absence of neutral markers. This method requires variant frequency determination at several time points, in independent hosts. The parameters are the selection coefficients for each PVY variant and four effective population sizes Ne at different time-points of the experiment. Numerical simulations of asexual haploid Wright-Fisher populations subjected to contrasting genetic drift (Ne ∈ [10, 2000]) and selection (|s| ∈ [0, 0.15]) regimes were used to validate the method proposed. The experiment in closely related pepper host genotypes revealed that viruses experienced a considerable diversity of selection and genetic drift regimes. The resulting variant dynamics were accurately described by Wright-Fisher models. The fitness ranks of the variants were almost identical between host genotypes. By contrast, the dynamics of Ne were highly variable, although a bottleneck was often identified during the systemic movement of the virus. We demonstrated that, for a fixed initial PVY population, virus effective population size is a heritable trait in plants. These findings pave the way for the breeding of plant varieties exposing viruses to stronger genetic drift, thereby slowing virus adaptation.

Quantitative trait loci in pepper control the effective population size of two RNA viruses at inoculation.

Infection of plants by viruses is a complex process involving several steps: inoculation into plant cells, replication in inoculated cells and plant colonization. The success of the different steps depends, in part, on the viral effective population size (Ne), defined as the number of individuals passing their genes to the next generation. During infection, the virus population will undergo bottlenecks, leading to drastic reductions in Ne and, potentially, to the loss of the fittest variants. Therefore, it is crucial to better understand how plants affect Ne. We aimed to (i) identify the plant genetic factors controlling Ne during inoculation, (ii) understand the mechanisms used by the plant to control Ne and (iii) compare these genetic factors with the genes controlling plant resistance to viruses. Ne was measured in a doubled-haploid population of Capsicum annuum. Plants were inoculated with either a Potato virus Y (PVY) construct expressing the green fluorescent protein or a necrotic variant of Cucumber mosaic virus (CMV). Newas assessed by counting the number of primary infection foci on cotyledons for PVY or the number of necrotic local lesions on leaves for CMV. The number of foci and lesions was correlated (r=0.57) and showed a high heritability (h2=0.93 for PVY and h2=0.98 for CMV). The Ne of the two viruses was controlled by both common quantitative trait loci (QTLs) and virus-specific QTLs, indicating the contribution of general and specific mechanisms. The PVY-specific QTL colocalizes with a QTL that reduces PVY accumulation and the capacity to break down a major-effect resistance gene.

Plant Genetic Background Increasing the Efficiency and Durability of Major Resistance Genes to Root-knot Nematodes Can Be Resolved into a Few Resistance QTLs.

With the banning of most chemical nematicides, the control of root-knot nematodes (RKNs) in vegetable crops is now based essentially on the deployment of single, major resistance genes (R-genes). However, these genes are rare and their efficacy is threatened by the capacity of RKNs to adapt. In pepper, several dominant R-genes are effective against RKNs, and their efficacy and durability have been shown to be greater in a partially resistant genetic background. However, the genetic determinants of this partial resistance were unknown. Here, a quantitative trait loci (QTL) analysis was performed on the F2:3 population from the cross between Yolo Wonder, an accession considered partially resistant or resistant, depending on the RKN species, and Doux Long des Landes, a susceptible cultivar. A genetic linkage map was constructed from 130 F2 individuals, and the 130 F3 families were tested for resistance to the three main RKN species, Meloidogyne incognita, M. arenaria, and M. javanica. For the first time in the pepper-RKN pathosystem, four major QTLs were identified and mapped to two clusters. The cluster on chromosome P1 includes three tightly linked QTLs with specific effects against individual RKN species. The fourth QTL, providing specific resistance to M. javanica, mapped to pepper chromosome P9, which is known to carry multiple NBS-LRR repeats, together with major R-genes for resistance to nematodes and other pathogens. The newly discovered cluster on chromosome P1 has a broad spectrum of action with major additive effects on resistance. These data highlight the role of host QTLs involved in plant-RKN interactions and provide innovative potential for the breeding of new pepper cultivars or rootstocks combining quantitative resistance and major R-genes, to increase both the efficacy and durability of RKN control by resistance genes.

Diversity of genetic backgrounds modulating the durability of a major resistance gene. Analysis of a core collection of pepper landraces resistant to Potato virus Y.

The evolution of resistance-breaking capacity in pathogen populations has been shown to depend on the plant genetic background surrounding the resistance genes. We evaluated a core collection of pepper (Capsicum annuum) landraces, representing the worldwide genetic diversity, for its ability to modulate the breakdown frequency by Potato virus Y of major resistance alleles at the pvr2 locus encoding the eukaryotic initiation factor 4E (eIF4E). Depending on the pepper landrace, the breakdown frequency of a given resistance allele varied from 0% to 52.5%, attesting to their diversity and the availability of genetic backgrounds favourable to resistance durability in the plant germplasm. The mutations in the virus genome involved in resistance breakdown also differed between plant genotypes, indicating differential selection effects exerted on the virus population by the different genetic backgrounds. The breakdown frequency was positively correlated with the level of virus accumulation, confirming the impact of quantitative resistance loci on resistance durability. Among these loci, pvr6, encoding an isoform of eIF4E, was associated with a major effect on virus accumulation and on the breakdown frequency of the pvr2-mediated resistance. This exploration of plant genetic diversity delivered new resources for the control of pathogen evolution and the increase in resistance durability.

Host genetic resistance to root-knot nematodes, Meloidogyne spp., in Solanaceae: from genes to the field.

Root-knot nematodes (RKNs) heavily damage most solanaceous crops worldwide. Fortunately, major resistance genes are available in a number of plant species, and their use provides a safe and economically relevant strategy for RKN control. From a structural point of view, these genes often harbour NBS-LRR motifs (i.e. a nucleotide binding site and a leucine rich repeat region near the carboxy terminus) and are organised in syntenic clusters in solanaceous genomes. Their introgression from wild to cultivated plants remains a challenge for breeders, although facilitated by marker-assisted selection. As shown with other pathosystems, the genetic background into which the resistance genes are introgressed is of prime importance to both the expression of the resistance and its durability, as exemplified by the recent discovery of quantitative trait loci conferring quantitative resistance to RKNs in pepper. The deployment of resistance genes at a large scale may result in the emergence and spread of virulent nematode populations able to overcome them, as already reported in tomato and pepper. Therefore, careful management of the resistance genes available in solanaceous crops is crucial to avoid significant reduction in the duration of RKN genetic control in the field. From that perspective, only rational management combining breeding and cultivation practices will allow the design and implementation of innovative, sustainable crop production systems that protect the resistance genes and maintain their durability.

Interaction patterns between potato virus Y and eIF4E-mediated recessive resistance in the Solanaceae.

UNLABELLED: The structural pattern of infectivity matrices, which contains infection data resulting from inoculations of a set of hosts by a set of parasites, is a key parameter for our understanding of biological interactions and their evolution. This pattern determines the evolution of parasite pathogenicity and host resistance, the spatiotemporal distribution of host and parasite genotypes, and the efficiency of disease control strategies. Two major patterns have been proposed for plant-virus genotype infectivity matrices. In the gene-for-gene model, infectivity matrices show a nested pattern, where the host ranges of specialist virus genotypes are subsets of the host ranges of less specialized viruses. In contrast, in the matching-allele (MA) model, each virus genotype is specialized to infect one (or a small set of) host genotype(s). The corresponding infectivity matrix shows a modular pattern where infection is frequent for plants and viruses belonging to the same module but rare for those belonging to different modules. We analyzed the structure of infectivity matrices between Potato virus Y (PVY) and plant genotypes in the family Solanaceae carrying different eukaryotic initiation factor 4E (eIF4E)-coding alleles conferring recessive resistance. Whereas this system corresponds mechanistically to an MA model, the expected modular pattern was rejected based on our experimental data. This was mostly because PVY mutations involved in adaptation to a particular plant genotype displayed frequent pleiotropic effects, conferring simultaneously an adaptation to additional plant genotypes with different eIF4E alleles. Such effects should be taken into account for the design of strategies of sustainable control of PVY through plant varietal mixtures or rotations. IMPORTANCE: The interaction pattern between host and virus genotypes has important consequences on their respective evolution and on issues regarding the application of disease control strategies. We found that the structure of the interaction between Potato virus Y (PVY) variants and host plants in the family Solanaceae departs significantly from the current model of interaction considered for these organisms because of frequent pleiotropic effects of virus mutations. These mutational effects allow the virus to expand rapidly its range of host plant genotypes, make it very difficult to predict the effects of mutations in PVY infectivity factors, and raise concerns about strategies of sustainable management of plant genetic resistance to viruses.

Pyramiding, alternating or mixing: comparative performances of deployment strategies of nematode resistance genes to promote plant resistance efficiency and durability.

BACKGROUND: Resistant cultivars are key elements for pathogen control and pesticide reduction, but their repeated use may lead to the emergence of virulent pathogen populations, able to overcome the resistance. Increased research efforts, mainly based on theoretical studies, explore spatio-temporal deployment strategies of resistance genes in order to maximize their durability. We evaluated experimentally three of these strategies to control root-knot nematodes: cultivar mixtures, alternating and pyramiding resistance genes, under controlled and field conditions over a 3-years period, assessing the efficiency and the durability of resistance in a protected crop rotation system with pepper as summer crop and lettuce as winter crop. RESULTS: The choice of the resistance gene and the genetic background in which it is introgressed, affected the frequency of resistance breakdown. The pyramiding of two different resistance genes in one genotype suppressed the emergence of virulent isolates. Alternating different resistance genes in rotation was also efficient to decrease virulent populations in fields due to the specificity of the virulence and the trapping effect of resistant plants. Mixing resistant cultivars together appeared as a less efficient strategy to control nematodes. CONCLUSIONS: This work provides experimental evidence that, in a cropping system with seasonal sequences of vegetable species, pyramiding or alternating resistance genes benefit yields in the long-term by increasing the durability of resistant cultivars and improving the long-term control of a soil-borne pest. To our knowledge, this result is the first one obtained for a plant-nematode interaction, which helps demonstrate the general applicability of such strategies for breeding and sustainable management of resistant cultivars against pathogens.

Farther, slower, stronger: how the plant genetic background protects a major resistance gene from breakdown.

Genetic resistance provides efficient control of crop diseases, but is limited by pathogen evolution capacities which often result in resistance breakdown. It has been demonstrated recently, in three different pathosystems, that polygenic resistances combining a major-effect gene and quantitative resistance controlled by the genetic background are more durable than monogenic resistances (with the same major gene in a susceptible genetic background), but the underlying mechanisms are unknown. Using the pepper-Potato virus Y system, we examined three mechanisms that could account for the greater durability of the polygenic resistances: (i) the additional quantitative resistance conferred by the genetic background; (ii) the increase in the number of mutations required for resistance breakdown; and (iii) the slower selection of adapted resistance-breaking mutants within the viral population. The three mechanisms were experimentally validated. The first explained a large part of the variation in resistance breakdown frequency and is therefore expected to be a major determinant of resistance durability. Quantitative resistance factors also had an influence on the second mechanism by modifying the virus mutational pathways towards resistance breakdown and could also have an influence on the third mechanism by increasing genetic drift effects on the viral population. The relevance of these results for other plant-pathogen systems and their importance in plant breeding are discussed.

SPICY: towards automated phenotyping of large pepper plants in the greenhouse.

Most high-throughput systems for automated plant phenotyping involve a fixed recording cabinet to which plants are transported. However, important greenhouse plants like pepper are too tall to be transported. In this research we developed a system to automatically measure plant characteristics of tall pepper plants in the greenhouse. With a device equipped with multiple cameras, images of plants are recorded at a 5cm interval over a height of 3m. Two types of features are extracted: (1) features from a 3D reconstruction of the plant canopy; and (2) statistical features derived directly from RGB images. The experiment comprised 151 genotypes of a recombinant inbred population of pepper, to examine the heritability and quantitative trait loci (QTL) of the features. Features extracted from the 3D reconstruction of the canopy were leaf size and leaf angle, with heritabilities of 0.70 and 0.56 respectively. Three QTL were found for leaf size, and one for leaf angle. From the statistical features, plant height showed a good correlation (0.93) with manual measurements, and QTL were in accordance with QTL of manual measurements. For total leaf area, the heritability was 0.55, and two of the three QTL found by manual measurement were found by image analysis.

A point mutation in the polymerase of Potato virus Y confers virulence toward the Pvr4 resistance of pepper and a high competitiveness cost in susceptible cultivar.

To understand why the Pvr4 resistance of pepper against Potyvirus spp. remained durable in field conditions while virulent Potato virus Y (PVY) variants could be selected in the laboratory, we studied the molecular mechanisms which generated these variants and the consequences on viral fitness. Using a reverse genetics approach with an infectious cDNA clone of PVY, we found that the region coding for the NIb protein (RNA-dependent RNA polymerase) of PVY was the avirulence factor corresponding to Pvr4 and that a single nonsynonymous nucleotide substitution in that region, an adenosine to guanosine substitution at position 8,424 of the PVY genome (A(8424)G), was sufficient for virulence. This substitution imposed a high competitiveness cost to the virus against an avirulent PVY variant in plants devoid of Pvr4. In addition, during serial passages in susceptible pepper plants, the only observed possibility of the virulent mutant to increase its fitness was through the G(8424)A reversion, strengthening the high durability potential of the Pvr4 resistance. This is in accordance with the fact that the NIb protein is one of the most constrained proteins expressed by the PVY genome and, more generally, by Potyvirus spp., and with a previously developed model predicting the durability of virus resistances as a function of the evolutionary constraint applied on corresponding avirulence factors.

Constraints on evolution of virus avirulence factors predict the durability of corresponding plant resistances.

SUMMARY Understanding the factors driving pathogen emergence and re-emergence is a major challenge, particularly in agriculture, where the use of resistant plant cultivars imposes strong selective pressures on plant pathogen populations and leads frequently to 'resistance breakdown'. Presently, durable resistances are only identified after a long period of large-scale cultivation of resistant cultivars. We propose a new predictor of the durability of plant resistance. Because resistance breakdown involves modifications in the avirulence factors of pathogens, we tested for correlations between the evolutionary constraints acting on avirulence factors or their diversity and the durability of the corresponding resistance genes in the case of plant-virus interactions. An analysis performed on 20 virus species-resistance gene combinations revealed that the selective constraints applied on amino acid substitutions in virus avirulence factors correlate with the observed durability of the corresponding resistance genes. On the basis of this result, a model predicting the potential durability of resistance genes as a function of the selective constraints applied on the corresponding avirulence factors is proposed to help breeders to select the most durable resistance genes.

Allele mining in the pepper gene pool provided new complementation effects between pvr2-eIF4E and pvr6-eIF(iso)4E alleles for resistance to pepper veinal mottle virus.

Molecular cloning of recessive resistance genes to potyviruses in a large range of host species identified the eukaryotic translation initiation factor 4E (eIF4E) as an essential determinant in the outcome of potyvirus infection. Resistance results from a few amino acid changes in the eIF4E protein encoded by the recessive resistance allele that disrupt the direct interaction with the potyviral protein VPg. In plants, several loci encode two protein subfamilies, eIF4E and eIF(iso)4E. While most eIF4E-mediated resistance to potyviruses depends on mutations in a single eIF4E protein, simultaneous mutations in eIF4E (corresponding to the pvr2 locus) and eIF(iso)4E (corresponding to the pvr6 locus) are required to prevent pepper veinal mottle virus (PVMV) infection in pepper. We used this model to look for additional alleles at the pvr2-eIF4E locus that result in resistance when combined with the pvr6-eIF(iso)4E resistant allele. Among the 12 pvr2-eIF4E resistance alleles sequenced in the pepper gene pool, three were shown to have a complementary effect with pvr6-eIF(iso)4E for resistance. Two amino acid changes were exclusively shared by these three alleles and were systematically associated with a second amino acid change, suggesting that these substitutions are associated with resistance expression. The availability of new resistant allele combinations increases the possibility for the durable deployment of resistance against this pepper virus which is prevalent in Africa.

LTR-retrotransposons Tnt1 and T135 markers reveal genetic diversity and evolutionary relationships of domesticated peppers.

Plant genetic resources often constitute the foundation of successful breeding programs. Pepper (Capsicum annuum L.) is one of the most economically important and diversely utilized Solanaceous crop species worldwide, but less studied compared to tomato and potato. We developed and used molecular markers based on two copia-type retrotransposons, Tnt1 and T135, in a set of Capsicum species and wild relatives from diverse geographical origins. Results showed that Tnt1 and T135 insertion polymorphisms are very useful for studying genetic diversity and relationships within and among pepper species. Clusters of accessions correspond to cultivar types based on fruit shape, pungency, geographic origin and pedigree. Genetic diversity values, normally reflective of past transposition activity and population dynamics, showed positive correlation with the average number of insertions per accession. Similar evolutionary relationships are observed to that inferred by previous karyosystematics studies. These observations support the possibility that retrotransposons have contributed to genome inflation during Capsicum evolution.

QTL analysis of plant development and fruit traits in pepper and performance of selective phenotyping.

A QTL analysis was performed to determine the genetic basis of 13 horticultural traits conditioning yield in pepper (Capsicum annuum). The mapping population was a large population of 297 recombinant inbred lines (RIL) originating from a cross between the large-fruited bell pepper cultivar 'Yolo Wonder' and the small-fruited chilli pepper 'Criollo de Morelos 334'. A total of 76 QTLs were detected for 13 fruit and plant traits, grouped in 28 chromosome regions. These QTLs explained together between 7% (internode growth time) and 91% (fruit diameter) of the phenotypic variation. The QTL analysis was also performed on two subsets of 141 and 93 RILs sampled using the MapPop software. The smaller populations allowed for the detection of a reduced set of QTLs and reduced the overall percentage of trait variation explained by QTLs. The frequency of false positives as well as the individual effect of QTLs increased in reduced population sets as a result of reduced sampling. The results from the QTL analysis permitted an overall glance over the genetic architecture of traits considered by breeders for selection. Colinearities between clusters of QTLs controlling fruit traits and/or plant development in distinct pepper species and in related solanaceous crop species (tomato and eggplant) suggests that shared mechanisms control the shape and growth of different organs throughout these species.

Key determinants of resistance durability to plant viruses: insights from a model linking within- and between-host dynamics.

The emergence of new genotypes of parasites involves several evolutionary, epidemiological and ecological processes whose individual effects and interactions are difficult to disentangle using experimental approaches. Here, a model is proposed to investigate how these processes lead to the emergence of plant viral genotypes breaking down qualitative resistance genes. At the individual plant scale, selection, drift and mutation processes shape the evolution of viral populations from a set of differential equations. The spatial segregation of virus genotypes in their hosts is also considered. At the host population scale, the epidemiological dynamics is given by an individual-based algorithm. Global sensitivity analyses allowed ranking the ten demo-genetic and epidemiological parameters of the model according to their impact on the mean and variance of the risk of breakdown of a plant resistance. Demo-genetic parameters (number and nature of mutations involved in breakdown, fitness of mutant genotypes) had the largest impact on the mean breakdown risk, whereas epidemiological parameters had more influence on its standard deviation. It is discussed how these results can be used to choose the potentially most durable resistance genes among a pool of candidates. Finally, our analyses point out the parameters which should be estimated more precisely to improve durability predictions.