This article appeared in a journal published by Elsevier. The

Mar 18, 2008 - P. halstedii is a homothallic oomycete, whose cycle is made up of a single sexual generation permitting overwintering and one or perhaps two ...
540KB taille 9 téléchargements 306 vues
This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright

Author's personal copy

Infection, Genetics and Evolution 8 (2008) 534–540

Contents lists available at ScienceDirect

Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid

Single nucleotide polymorphisms reveal multiple introductions into France of Plasmopara halstedii, the plant pathogen causing sunflower downy mildew Franc¸ois Delmotte a,*, Xavier Giresse a, Sylvie Richard-Cervera a, Jessica M’Baya a, Felicity Vear b, Jeanne Tourvieille b, Pascal Walser b, Denis Tourvieille de Labrouhe b a b

INRA, Institut des Sciences de la Vigne et du Vin, UMR 1065 Sante´ Ve´ge´tale (INRA-ENITAB), Centre de Recherche Bordeaux-Aquitaine, BP 81, 33883 Villenave d’Ornon cedex, France INRA, UMR 1095 GEDEC (INRA-Universite Blaise Pascal), Centre de Recherches de Clermont-Ferrand, Domaine de Crouelle, 63100 Clermont-Ferrand, France

A R T I C L E I N F O

A B S T R A C T

Article history: Received 11 October 2007 Received in revised form 28 February 2008 Accepted 29 February 2008 Available online 18 March 2008

Plasmopara halstedii, the causal agent of sunflower downy mildew, displays a gene-for-gene interaction with its host plant, Helianthus annuus and other species of the genus. Monitoring of the evolution of virulent races in France over a 19-year period led to the identification of 14 different races (or pathotypes). Twelve expressed sequence tag (EST)-derived markers displaying SNPs and insertion– deletion variations have recently been identified in P. halstedii. We used these markers to study the genetic structure and the evolution of sunflower downy mildew races. Bayesian assignment analysis identified three genetically differentiated groups of isolates organized around the first three races described in France. Strong genetic substructuring according to geographic origin of races was observed, confirming that these three groups corresponded to three separate introductions into France of isolates with different genetic and phenotypic backgrounds. Our results suggest that multiple introductions of P. halstedii isolates may have provided the raw material for more complex processes in the evolution of races, such as recombination between races or clonal evolution through mitotic instability. ß 2008 Elsevier B.V. All rights reserved.

Keywords: Helianthus annuus Invasive parasite Oomycetes Pathotype Race SNP Virulence profile

1. Introduction Sunflower downy mildew is caused by Plasmopara halstedii (Berlese & de Toni), an invasive species native to North America, where wild sunflower (Helianthus annuus) is endemic. It was first described in Indiana, Iowa and Minnesota in the early 1920s (Yong and Morris, 1927). This pathogen spreads into Europe, probably in the seeds of its host plant (Ioos et al., 2007), and was first reported in Europe, in Russia, at the beginning of 1960s (Leppik, 1962; Novotel’nova, 1966). Since 1992, P. halstedii has been subject to quarantine regulation in the European Union (directive 92/103/ CEE). P. halstedii is a homothallic oomycete, whose cycle is made up of a single sexual generation permitting overwintering and one or perhaps two asexual generations which occur during the growing season (Spring et al., 2006). Gene-for-gene interactions have been demonstrated between this pathogen and its host plant (Tourvieille de Labrouhe et al., 2000), and many physiological races (or pathotypes) have been characterized worldwide (Gulya, 2007; Gulya et al., 1991; Lafon et al., 1996; Moinard et al., 2006;

* Corresponding author. E-mail address: [email protected] (F. Delmotte). 1567-1348/$ – see front matter ß 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.meegid.2008.02.012

Molinero-Ruiz et al., 2002; Rashid, 1993; Spring et al., 1994; Zimmer, 1974). The introduction of genetic resistance into sunflower varieties has been, and remains the most efficient method for controlling the disease. This resistance concerns a number of downy mildew races and is conferred by major genes, denoted Pl (Vear et al., 1997, 2003). In France, the monitoring of sunflower downy mildew races has shown that, after a long period of stability, the effectiveness of the major resistance genes (Pl1, Pl2, Pl5 and Pl6) has recently been overcome. Fourteen different races have been described in France, nine of which emerged during the last 7 years. When sunflower breeding programs were first initiated in France, only one race of P. halstedii was known (race 100). This first race of P. halstedii collected in Europe has a specific virulence profile different from that of American races. With changes in nomenclature, it has been called the ‘‘European race’’ (Sackston, 1981), ‘‘race 1’’, and now ‘‘race 100’’ (Tourvieille de Labrouhe, 1999). Until 1980s, race 100 was widely distributed throughout France and was the only race present in this country. Selection pressure on the pathogen was strong in 1980s, because all the hybrids on sale in France carried either Pl1 or Pl2, both of which confer resistance to race 100 (Sackston, 1981). At the end of this decade, two new virulent races (710 and 703) appeared in France: race 710 found in 1988 in the Indre (Tourvieille de Labrouhe, 1988)

Author's personal copy

F. Delmotte et al. / Infection, Genetics and Evolution 8 (2008) 534–540 Table 1 Name of race (pathotype), date of first isolation in France and virulence profiles for the 14 French ‘‘reference isolates’’ of P. halstedii used in this study Race namea

Year of isolation

Virulence profiles on differential hostsb D1

D2

D3

D4

D5

D6

D7

D8

D9

100 710 703 300 700 304 314 307 704 714 334 707 717 730

1966 1988 1989 1995 1995 2000 2001 2002 2002 2002 2004 2004 2004 2005

S S S S S S S S S S S S S S

R S S S S S S S S S S S S S

R S S R S R R R S S R S S S

R S R R R R S R R S S R S S

R R R R R R R R R R S R R S

R R R R R R R R R R R R R R

R R S R R R R S R R R S S R

R R S R R R R S R R R S S R

R R R R R S S S S S S S S R

a b

According to Gulya et al. (1998). S: susceptible (compatible interaction); R: resistant (incompatible interaction).

and mostly restricted to the northern part of France, and race 703 which predominates in southern France, where it was first described in 1989 (de Gue´nin, 1990). From 1989 to 1995, races 100, 703 and 710 were the only races found in French downy mildew populations (Meliala, 2001; Roeckel-Drevet et al., 1997). However, pathogen evolution has accelerated since 1995, due to the massive deployment of new resistance genes, such as Pl6 and Pl7, and as many as 11 new races of P. halstedii have been characterized (see Table 1 for a comprehensive chronology): 300, 700 (1995), 304 (2000), 314 (2001), 307, 704, 714 (2002), 334, 707, 717 (2004) and 730 (2005) (Moinard, 2002–2005; Tourvieille de Labrouhe et al., 2005). Today, the total ‘population’ of French races can overcome the resistance genes in seven of the nine downy mildew differential lines of H. annuus, and the Pl2 and Pl6 genes have each been overcome by eight races (Moinard et al., 2006; Tourvieille de Labrouhe et al., 2000). Our understanding of the recurrent breakdown of sunflower major resistance genes can be improved by new findings concerning the key processes governing the evolution of P. halstedii populations. Knowledge of the evolutionary potential of plant pathogen species has improved the management of resistance genes and maximized their durability in space and time (McDonald and Linde, 2002). A population genetics approach can be used to evaluate the major forces driving evolution—i.e. selection, mutation, recombination, genetic drift and gene flow. New virulence profiles in P. halstedii populations may evolve via several different mechanisms: (i) mutations in virulence genes in clonal lineages; (ii) recombination following crosses between different races; (iii) the introduction of races able to overcome the resistance of sunflower varieties grown in one area. Sound knowledge of the population genetic structure of P. halstedii is required to distinguish between these possibilities. Previous genetic studies of P. halstedii in Europe have reported low levels of genetic and genotypic diversity in this species, with no clear genetic structure revealed with RAPD markers (Komjati et al., 2004; Roeckel-Drevet et al., 1997, 2003), ISSR (Intelmann and Spring, 2002) or ITS sequences (Spring et al., 2006; Thines et al., 2005). This precluded any reliable conclusions on the mode of reproduction, genetic structuring or the extent to which races are related in this species. It is possible that no genetic structure in P. halstedii populations was identified in previous studies because the molecular markers used were non-specific and insufficiently polymorphic within this species. AFLP, ISSR and RAPD markers have been used in many studies of plant pathogens because they

535

allow the rapid screening of many loci within a genome (Fry, 2001). However, these markers are dominant, anonymous and not specific to the target species. This lack of specificity is a major drawback in studies of obligate endoparasites, such as downy mildews. In contrast, single nucleotide polymorphisms (SNPs) are promising molecular markers for population genetics as they are widespread throughout the genome, codominant, specific and have a high resolving power. However, SNPs have been little used for analyses of the genetic structure of populations of fungi, and of plant pathogens in particular (Morin et al., 2004). One reason for this is that it has been difficult to characterize SNPs in non-model organisms, principally due to a lack of genomic data (Schlotterer, 2004). The development of new methods for screening for DNA polymorphism has rendered possible the extensive development of such makers for plant pathogen species (Brumfield et al., 2003; Schlotterer, 2004). We recently developed a set of 12 EST-derived markers displaying SNPs and insertion–deletion variations for genetic studies of sunflower downy mildew populations (Giresse et al., 2007). The aim of this study was to investigate the evolutionary processes responsible for changes in the pathogenic features of P. halstedii. We used 12 EST-derived markers to perform a genetic analysis of races of P. halstedii characterized over a 19-year period of monitoring in France (1988–2006). We addressed four main questions, using a mixture of phenotypic characterization and molecular genotyping: how powerful are SNP and EST-derived markers for studies of the population structure of P. halstedii? What is the relative importance of outcrossing and clonal reproduction in the French P. halstedii population? How are the different races of P. halstedii related? Which evolutionary mechanisms are involved in the emergence of new races in this plant pathogenic species? 2. Materials and methods 2.1. Sample collection and DNA extraction We analyzed 24 isolates of P. halstedii collected in France between 1966 and 2006. Fourteen of these isolates are ‘‘references Table 2 Race (pathotype), collection site and date of isolation in France for the 24 isolates of P. halstedii of this study Race

Collection site (‘‘de´partement’’)

Year of collection

100* 100 100 710* 710 710 710 710 703* 703 703 703 703 300* 700* 304* 314* 307* 704* 714* 334* 707* 717* 730*

Unknown Charente-Maritime Cher Indre Cher Unknown Maine-et-Loire Deux-Se`vres Lot-et-Garonne Tarn Lot-et-Garonne Haute-Garonne Gers Aude Haute-Garonne Aude Manche Haute-Garonne Deux-Se`vres Gers Charente Lot-et-Garonne Gers Tarn

1966 1993 1993 1988 1993 2000 2004 2006 1989 1993 2001 2004 2006 1995 1995 2000 2001 2002 2002 2002 2004 2004 2004 2005

The asterisk (*) indicates isolates corresponding to the ‘‘reference race’’ (see Table 1).

Author's personal copy

F. Delmotte et al. / Infection, Genetics and Evolution 8 (2008) 534–540

536

isolates’’, corresponding to the first description of the race concerned in France (Table 1). The other 10 isolates (Table 2) were obtained from the French Plant Protection Service monitoring program (Moinard et al., 2006). Each sample resulted from the bulking of several infected sunflower leaves collected on one or two plants in a field. For samples corresponding to the first identification of a race in France (Tables 1 and 2), these bulks were ‘‘purified’’, i.e. a monozoosporangium isolation was achieve before multiplication of the isolate on a susceptible host–plant genotype (Sakr et al., 2007). 2.2. Race characterization Downy mildew races are defined on the basis of virulence profiles on a set of differential hosts carrying different Pl resistance genes (Table 1). Resistance tests were performed as described by Cohen and Sackston (1974), with the modifications proposed by Mouzeyar et al. (1993): 15 days after inoculation, plants were incubated for 48 h in a saturated atmosphere. Plants were scored as susceptible if sporulation was observed in cotyledons and leaves, and as resistant if no sporulation or only light sporulation was seen on cotyledons. Races were named according to the international nomenclature of P. halstedii races proposed by Gulya et al. (1998). 2.3. DNA extraction and molecular typing For each isolate, DNA was isolated from infected plant tissue, as previously described for P. viticola (Delmotte et al., 2006; Chen et al., 2007). We then used the 12 recently identified polymorphic EST-derived markers (Giresse et al., 2007) to genotype P. halstedii isolates. Pha6, Pha54, Pha56, Pha82, Pha99, Pha79 and Pha120 are seven SNPs transformed into cleaved amplified polymorphism sequence (CAPS) markers or directly screened by PCR-SSCP. Pha39, Pha42, Pha43, Pha106 and Pha47 correspond to five insertion– deletion (indel) polymorphisms automated on a Beckman Coulter Ceq8000 capillary sequencer or visualized directly by agarose gel electrophoresis (Giresse et al., 2007). 2.4. Statistical analysis The resolving power of EST-derived markers was evaluated with Multilocus 1.2 (Agapow and Burt, 2001), which calculates the percentage of discriminated multilocus genotypes commensurate to the number of combined loci after 1000 resamplings. Identical multilocus genotypes sharing the same alleles at all loci may result from distinct sexual reproduction events or clonal reproduction. We assessed the likelihood of copies of multilocus genotypes

resulting from sexual reproduction, by calculating the probability Pgen(f), taking into account departure from Hardy–Weinberg equilibrium, as estimated for the sample set with GenClone 1.0 (Arnaud-Haond and Belkhir, 2007). Genepop version 3.2a was used to calculate expected and observed heterozygosities, unbiased estimates of FIS and FST (Weir and Cockerham, 1984). We investigated the genetic structure of P. halstedii races, using a Bayesian approach to genetic mixture analysis based on STRUCTURE v2.2 (Pritchard et al., 2000; Falush et al., 2003). This method can be used to estimate parameters separately from the posterior probability distribution of allele frequencies. Parameter estimation assumes panmixia, and that each locus is at Hardy– Weinberg equilibrium (HWE) and independent of the others. Nonetheless, this Bayesian approach is quite robust even when some deviations from these assumptions occur. Using the admixture model, we estimated the number of genetic clusters, K, to which the pathogens should be assigned. Using the full data set, 3–10 parallel Markov chains were run for all models of K, with a burn-in of 10,000 iterations and a run length of 105 iterations following the burn-in. For each run, the ln likelihood of each model was calculated. The genetic relationships between French races was assessed by correspondence analysis (Benze´cri, 1982) performed using the AFC procedure available in genetix 4.05 (Belkhir et al., 2001) on the matrix of multilocus genotypes. Allele shared distance (DAS, Jin and Chakraborty, 1993) were computed using Populations 1.2.28 software (Langella, 1999). A Mantel test was performed to assess the correlation between pairwise geographic and genetic distance (DAS) matrices for each isolate, using Genepop software. 3. Results 3.1. Genotypic structure Based on combinations of the 12 genomic markers, we identified 11 different multilocus genotypes among the 24 isolates analyzed. Three multilocus genotypes (MLG1, MLG2, MLG9) were found to have multiple isolates: five isolates for MLG1, six for MLG2, five for MLG9 (Table 3). A combination of eleven EST markers was sufficient to discriminate all the multilocus genotypes in the dataset, demonstrating the high resolving power of the markers (Fig. 1). Three races were represented by more than one isolate (race 100 represented by three isolates, races 703 and 710 each represented by five isolates; Table 2). Isolates of the same race

Table 3 Multilocus genotypes (MLG) characterized using 12 EST-derived genomic markers on the 24 French isolates of P. halstedii MLG

MLG1 MLG2 MLG3 MLG4 MLG5 MLG6 MLG7 MLG8 MLG9 MLG10 MLG11

N

5 6 1 1 1 1 1 1 5 1 1

Races

100, 300, 304 703, 307 700 707 730 314 334 704 710 714 717

Clustera

1 2 2 2 2 3 3 3 3 3 3

EST-derived markers Pha6

Pha39

Pha42

Pha43

Pha54

Pha56

Pha74

Pha79

Pha82

Pha99

Pha106

Pha120

2/2 1/1 1/1 1/1 1/1 2/2 1/1 2/2 1/1 2/2 1/1

2/2 1/1 1/1 1/1 1/1 1/1 1/1 1/1 1/1 1/1 1/1

1/1 1/1 1/1 1/1 1/1 2/2 2/2 1/1 2/2 2/2 1/2

1/1 1/1 1/2 1/1 2/2 2/2 1/1 2/2 2/2 2/2 1/2

1/1 1/1 1/1 1/1 1/1 1/1 2/2 1/1 1/1 1/1 1/1

1/1 1/1 1/1 1/1 1/1 2/2 1/1 2/2 2/2 2/2 2/2

1/1 1/1 1/1 1/1 1/1 2/2 2/2 2/2 2/2 1/1 2/2

3/3 2/2 1/2 2/2 1/1 1/1 1/1 1/1 1/1 1/1 1/1

2/2 1/1 1/1 2/2 1/1 1/1 1/1 2/2 1/1 1/1 1/1

2/2 1/1 1/1 1/1 1/1 1/1 1/1 2/2 1/1 1/1 2/2

1/1 2/2 2/2 1/1 2/2 2/1 2/2 2/2 2/2 2/2 2/2

2/2 1/1 1/1 1/1 1/1 2/2 1/1 1/1 1/1 2/2 1/1

For each MLG, we indicate the number of isolates (N), race and cluster, and for each locus, the two alleles found. a Cluster number was obtained from the Bayesian assignation analysis performed with STRUCTURE software.

Author's personal copy

F. Delmotte et al. / Infection, Genetics and Evolution 8 (2008) 534–540

537

Fig. 1. Proportion of P. halstedii genotypes discriminated (number of multilocus genotypes divided by the total number of isolates) as a function of the number of loci combined in the analysis.

presented an identical multilocus genotype, indicating that the three races may correspond to three clonal lineages (Table 3). Conversely, two multilocus genotypes included more than one race: the first multilocus genotype (MLG1) comprised races 100, 300 and 304 and the second (MLG2) comprised races 307 and 703 (Table 3). We checked that isolates with the same multilocus genotype belonged to the same clonal lineage, by calculating the probability of identical multilocus genotypes resulting from sexual reproduction events (taking the selfing rate into account), using GenClone v1. We were able to reject this hypothesis at the P < 10 5 level, confirming that identical multilocus genotypes probably resulted from asexual reproduction. As clonal reproduction leads to strong deviation from panmixia, further population genetic analysis were performed on a dataset comprising only one copy of each multilocus genotype (Sunnucks et al., 1997). 3.2. Population structure and genetic relations of races The expected heterozygosity level was consistent with high levels of genetic diversity across loci (0.349 < HE < 0.677). Observed heterozygosity levels were much lower, with a mean HO of 0.026. Only three of the 11 distinct multilocus genotypes were heterozygous: two at two loci (MLG3, MLG11) and one at one loci (MLG6) (Table 3). For the bayesian analysis, a single splitting solution was found for K = 3 for the full data set (Evanno et al., 2005). Probabilities of assignment of each race to each cluster are shown in Fig. 2. Assuming an arbitrary threshold of q = 0.9 for assignment to clusters, 12 over the 14 races (85.7%) were assigned to one of the three clusters, indicating high level of genetic differentiation between clusters (FST = 0.637 with a probability P < 10 5). Fig. 3 shows the scores of the races on the first two components of the factorial analysis. For each race, the probability of assignment obtained from the Bayesian analysis was reported on the correspondence analysis. Axis 1 (representing 34.2% of total genetic variability) discriminates cluster 1 from clusters 2 and 3. Axis 2 (representing 25.2% of the total genetic variability) separated races belonging to cluster 2 from races belonging to cluster 3 (Fig. 3). The first cluster corresponded to a single multilocus genotype, including three races: 100 (N = 3), 300 (N = 1) and 304 (N = 1). The second cluster included four multilocus genotypes and five races: 703 (N = 5), 307 (N = 1), 700 (N = 1), 730 (N = 1), 707 (N = 1). The third cluster included six races with different multilocus genotypes: 710 (N = 5), 334 (N = 1), 314 (N = 1), 714 (N = 1), 717 (N = 1) and 704 (N = 1). Two races (707, 334) could be considered as hybrid genotypes since they were assigned to two different group with a

Fig. 2. Probabilities of assignment of each race to each of the clusters inferred with structure software (white, gray, black). The date of the first description in France of each race is indicated in brackets under the race name.

threshold >20% (Fig. 2). For each of the three differentiated genetic clusters, we found one of the three races described before 1990 in France: cluster 1 included race 100 (1966), cluster 2 race 703 (1989), and cluster 3 race 710 (1988). Moreover, the genetic groups identified by the Bayesian analysis clustered races according to their geographical origin: cluster 1 corresponds to races 100 plus two races found in south-eastern France (Aude), cluster 2

Fig. 3. Factorial correspondence analysis (FCA) performed on the multilocus genotypes of P. halstedii races based on 12 EST-derived markers. Probabilities of assignment of each race to each cluster are reported on the FCA analysis using a colored pie-chart: white = cluster 1, gray = cluster 2, black = cluster 3.

Author's personal copy

538

F. Delmotte et al. / Infection, Genetics and Evolution 8 (2008) 534–540

Fig. 4. Geographic origin of 21 isolates of P. halstedii collected in France from 1966 to 2005. French reference races are shown in bold. Using a threshold value of 60%, each race was assignment to a cluster inferred from the Bayesian analysis (white = cluster 1, gray = cluster 2, black = cluster 3). The matrix of pairwise genetic distance (DAS) estimates between isolates (including repeated genotypes) was significantly correlated with the matrix of geographical distance (Mantel test, P = 0.0293).

French population of P. halstedii. Previous genetic studies of P. halstedii suffered from a lack of variability (Komjati et al., 2004; Roeckel-Drevet et al., 1997, 2003; Spring et al., 2006; Thines et al., 2005; Intelmann and Spring, 2002), whereas the EST-derived markers used here displayed high levels of resolving power and intraspecific polymorphism. Unlike the fingerprints obtained with dominant markers (Intelmann and Spring, 2002; Roeckel-Drevet et al., 1997, 2003), these codominant markers accurately estimated intra-individual heterozygosity. Moreover, whereas previous studies on P. halstedii has not indicated any relationship between genetic profile and race, EST-derived markers made it possible to demonstrate the presence of relationships between races, and to assess those relationships using assignment methods. Finally, these markers are specific and could therefore be used for the highthroughput genotyping of isolates directly from sporulating lesions collected from host leaves, avoiding the need for laborintensive isolate subculture (Giresse et al., 2007). Because Oomycetes are diploid organisms, the use of codominant markers is much more powerful in population genetic studies. Microsatellite markers may have provided an interesting alternative to SNPs for population genetic studies of this species, as reported for Plasmopara viticola, a closely related species causing grapevine downy mildew (Delmotte et al., 2006; Chen et al., 2007). They are also reliable, codominant and specific, but may in some cases be very difficult to isolate (Dutech et al., 2007). We, therefore, believe that SNPs may be the most useful type of marker in diploid plant pathogenic species for which genomic nucleotide sequences are available. 4.2. Genetic structure and evolution of races

comprises races from the south-western region of France and cluster 3 brings together races found in western and northern France (with the exception of one 710 isolated in the Gers) (Table 2, Fig. 4). Confirming this result, the matrix of pairwise genetic distance (DAS) estimates between isolates was significantly correlated with the matrix of geographical distance (Mantel test, P = 0.0293), indicating that the spatial distribution of races in France is not random (Fig. 4). Calculations of the genetic distance within each cluster indicated that cluster 3 was more heterogeneous than cluster 2 (Table 4). The clusters were well differentiated, with a particularly marked distinction between cluster 1 and clusters 2 and 3 (Table 4). This result was confirmed by the identification of two private alleles for cluster 1, at loci Pha39 and Pha79 (Table 3). 4. Discussion 4.1. The value of EST-derived markers for genetic studies of plant pathogens The use of EST-derived markers for this study resulted in the first comprehensive description of the genetic structure of the Table 4 Allele shared distance within and between the three clusters, as defined by the Bayesian analysis conducted with STRUCTURE software Clusters

DAS

Intracluster C1 C2 C3

0 0.167 0.336

Intercluster C1–C2 C1–C3 C2–C3

0.536 0.691 0.297

The finding that P. halstedii races cluster into three genetically differentiated groups, each including one of the first races described in France (i.e. 100, 703, 710), sheds new light on sunflower downy mildew evolution. Races 100, 710 and 703 correspond to three clonal lineages strongly differentiated from each other genetically and with different virulence profiles (Table 1). These results reflect multiple introductions of this pathogen in France: the first corresponding to race 100 (before 1966), which is now widely distributed throughout France, the second to race 710 in northern France (before 1988) and the third to race 703 in south-western France (before 1989). These three introductions of P. halstedii may have provided the raw genetic materials for more complex evolutionary processes, such as recombination between races or the accumulation of mutations in clonal lineages (clonal evolution), in the emergence of new races. Oomycetes have already been reported to display a large range of processes generating phenotypic variability, including interspecific hybridization (Brasier et al., 1999), recombination resulting from sexual reproduction, and genome instability caused by mitotic recombination, gene conversion, transposable elements or dispensable chromosomes (Kamoun, 2003). Progress in genetic studies of Phytophthora is very rapid, but very little is known about the population genetics of downy mildews which are obligate parasite. Our study provides several important insights into the genetic basis of phenotypic evolution in P. halstedii. First, our data indicate that occasional recombination events between genetically differentiated races may have led to the emergence of new races: race 334 is intermediate between races 710 and 703 (Figs. 2 and 3), indicating that it may have resulted from crosses between isolates from clusters 2 and 3. In this case, the recombination of genetically differentiated races has generated a new phenotypic profile, characterized by the ability of races 334 to overcome the resistance gene of the differential host D5 (Table 1). Similarly, both genetic (Figs. 2 and 3) and phenotypic

Author's personal copy

F. Delmotte et al. / Infection, Genetics and Evolution 8 (2008) 534–540

(Table 1) data have provided strong evidence to suggest that race 707 resulted from the recombination of races 703 and 304. These examples clearly illustrate the contribution of outcrossing events to the emergence of a new race in the French population of P. halstedii. Second, our results provide evidence to suggest that a clonal lineage (one multilocus genotype) can include several races (Table 3). This suggests that mutations in a clonal lineage may lead to the emergence of new races in the same genetic background: this is the cases for races 100, 300 and 304 and for races 703 and 307 (Figs. 2 and 3; Table 3). Two non-exclusive mechanisms may be responsible for this phenotypic variability in avirulence determinants. The first of these mechanisms is recombination within the lineage via the homothallic fusion of gametangia. This hypothesis is certainly the most likely, given the high selfing rate of sunflower downy mildew inferred from our analysis. Alternatively, genetic recombination via parasexual events may have generated this variability as it has been shown recently that mitotic recombination between different isolates of P. halstedii may be involved in the generation of new phenotypes within a population (Spring and Zipper, 2006). In this case, mitotic recombination within a genetic lineage would be required, and no such phenomenon has ever been described. The importance of mutation events in race evolution has already been highlighted in several Oomycetes plant pathogens such as Bremia lactucae (Lebeda and Petrzelova, 2004; Lebeda and Zinkernagel, 2003) and Phytophtora infestans (Andrivon, 1994). Although surprising at first sight, the greater importance of mutation than of recombination in populations able to outcross may be related to the large number of clonal cycles of fungal multiplication during epidemics. The combination of large populations, due to asexual reproduction of the pathogen, with strong selective pressures induced by resistance genes in the plant is likely to favor the emergence of new virulent races. Finally, the origin of five of the reference races (314, 700, 704, 714, 717, 730) remains unexplained. First, they may have resulted from introduction via seed transportation, provided that the introduced isolates had new virulence profiles in a genetic background similar to that of race 710 or 703 (Fig. 2). If this was the case, it would raise questions about identification of the source populations of this plant pathogen recently introduced into France. Also, there was a long delay between observation of races 703 and 710 and appearance of the new races (11 years, in spite of a concerted effort to detect new races), which except for race 700, have never been reported in the USA. Second, the new races may have arisen via the accumulation of mutations in ‘‘established races’’ (as discussed above). This hypothesis is probably the least likely, as it implies very high mutation rates for molecular markers to account for the genetic distances observed between races of the same cluster (difference of 10 alleles between the most distant races of cluster 3). Third, a two-step process, involving crosses between two races generating new phenotypic profiles followed by several generations of selfing resulting in the random fixation of one parental allele at each locus may have occurred. This would generate races resembling either race 710 or race 703. It is difficult to determine with certainty which of these hypotheses is correct based solely on an analysis of the genetic structure of natural populations of the pathogen. Improvements in our understanding of the importance of these mechanisms would require investigations of race evolution under controlled experimental conditions. 5. Conclusion The EST-derived markers used in this study displayed high levels intraspecific polymorphism which made it possible to infer

539

the reproductive mode of P. halstedii and to assess the relationships among French races. The biology of P. halstedii is indicative of a pathogen species for which there is a high risk of the population adapting to plant resistance genes: the pathogen population is large and this pathogen produces oospores, which can survive in the soil in the absence of crops, and displays obligate sexual recombination, mainly by selfing, and asexual multiplication. The potential of P. halstedii to evolve new races has also been enhanced by the introduction of several genetically differentiated genotypes: repeated introductions of P. halstedii isolates, combined with the selective pressure exerted by host resistance genes, may have greatly accelerated the evolution of new races in France (DesprezLoustau et al., 2007; Novak, 2007).

Acknowledgments This work was supported by the CETIOM (Centre Technique Interprofessionnel des Ole´agineux Me´tropolitains). We thank Cyril Dutech and Marie-Laure Desprez-Loustau for useful comments on an earlier version of the manuscript, and Julie Sappa for correcting the English. References Andrivon, D., 1994. Race structure and dynamics in populations of Phytophthora infestans. Canadian Journal of Botany 72, 1681–1687. Arnaud-Haond, S., Belkhir, K., 2007. Genclone: a computer program to analyse genotypic data, test for clonality and describe spatial clonal organization. Molecular Ecology Notes 7, 15–17. Brasier, C.M., Cooke, D.E.L., Duncan, J.M., 1999. Origin of a new Phytophthora pathogen through interspecific hybridization. Proceedings of the National Academy of Sciences of the United States of America 96, 5878–5883. Belkhir, K., Borsa, P., Chikhi, L., Raufaste, N., Bonhomme, F., 2001. GENETIX 4.02 logiciel sous Windows TM. pour la ge´ne´tique des populations. Laboratoire Ge´nome et Populations, Interactions, Adaptations, CNRS UMR500, Universite´ Montpellier II, Montpellier, France. Benze´cri, J.P., 1982. L’analyse des donne´es. Dunod, Paris. Brumfield, R.T., Beerli, P., Nickerson, D.A., Edwards, S.V., 2003. The utility of single nucleotide polymorphisms in inferences of population history. Trends in Ecology & Evolution 18, 249–256. Chen, W.J., Delmotte, F., Richard-Cervera, S., Douence, L., Greif, C., Corio-Costet, M.F., 2007. Multiple origins of fungicide resistance in grapevine downy mildew populations. Applied and Environmental Microbiology 73, 5162–5172. Cohen, Y., Sackston, W.E., 1974. Seed infection and latent infection of sunflower by Plasmopara halstedii. Canadian Journal of Botany 52, 231–238. de Gue´nin, M.C., 1990. Maladies du tournesol: un nouveau mildiou et toujours le Sclerotinia. Cultivar 269, 92–93. Delmotte, F., Chen, W.J., Richard-Cervera, S., Greif, C., Papura, D., Giresse, X., Mondor-Genson, G., Corio-Costet, M.F., 2006. Microsatellite DNA markers for Plasmopara viticola, the causal agent of downy mildew of grapes. Molecular Ecology Notes 6, 379–381. Desprez-Loustau, M.L., Robin, C., Buee, M., Courtecuisse, R., Garbaye, J., Suffert, F., Sache, I., Rizzo, D.M., 2007. The fungal dimension of biological invasions. Trends in Ecology and Evolution 22, 472–480. Dutech, C., Enjalbert, J., Fournier, C., Delmotte, F., Barre`s, B., Carlier, J., Tharreau, D., Giraud, T., 2007. Challenges of microsatellite isolation in fungi. Fungal Genetics and Biology 44, 933–949. Evanno, G., Regnaut, S., Goudet, J., 2005. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology 14, 2611–2620. Falush, D., Stephens, M., Pritchard, J.K., 2003. Inference of population structure using multilocus genotype data: linked loci and correlated allele frequencies. Genetics 164, 1567–1587. Fry, W.E., 2001. Molecular tools in epidemiology of plant diseases with special reference to potato-Phytophthora infestans pathosystem. Plant Disease Research 16, 1–9. Giresse, X., Tourvieille de Labrouhe, D., Richard-Cervera, S., Delmotte, F., 2007. Twelve polymorphic EST-derived markers for Plasmopara halstedii, the causal agent of sunflower downy mildew. Molecular Ecology Notes 7, 1363–1365. Gulya, T., 2007. Distribution of Plasmopara halstedii races from sunflower around the world. In: The Downy Mildews 2nd International Symposium, Olomouc, Czech Republic. Gulya, T., Tourvieille de Labrouhe, D., Masirevic, S., Penaud, A., Rashid, K., Viranyi, F., 1998. Proposal for standardized nomenclature and identification of races of Plasmopara halstedii. In: Symposium III: Sunflower Downy Mildew. International Sunflower Association, Fargo, pp. 130–136.

Author's personal copy

540

F. Delmotte et al. / Infection, Genetics and Evolution 8 (2008) 534–540

Gulya, T.J., Sackston, W.E., Viranyi, F., Masirevic, S., Rashid, K.Y., 1991. New races of the sunflower downy mildew pathogen (Plasmopara halstedii) in Europe and North and South America. Journal of Phytopathology 132, 303–311. Intelmann, F., Spring, O., 2002. Analysis of total DNA by minisatellite and simplesequence repeat primers for the use of population studies in Plasmopara halstedii. Canadian Journal of Microbiology 48, 555–559. Ioos, R., Laugustin, L., Rose, S., Tourvieille, J., Tourvieille de Labrouhe, D., 2007. Development of a PCR test to detect the downy mildew causal agent Plasmopara halstedii in sunflower seeds. Plant Pathology 56, 209–218. Jin, L., Chakraborty, R., 1993. Estimation of genetic distance and coefficient of gene diversity from single-probe multilocus DNA fingerprinting data. Molecular Biology and Evolution 11, 120–127. Kamoun, S., 2003. Molecular genetics of pathogenic oomycetes. Eukaryotic Cell 2, 191–199. Komjati, H., Fekete, C., Viranyi, F., 2004. Genetic and molecular characterization of Plasmopara halstedii isolates from Hungary. In: J Spencer-Phillips, P., Jeger, M. (Eds.), Avances in Downy Mildew Research. Kluwer Academic Publisher, London, UK, pp. 193–201. Lafon, S., Penaud, A., Walser, P., Guenin, M.C.D., Molinero, V., Mestres, R., Tourvieille, D., 1996. Sunflower downy mildew: ongoing monitoring. Phytoma 484, 35–36. Langella, O., 1999. Populations 1.2.28 software (http://www.pge.cnrs-gif.fr/bioinfo/ populations). Lebeda, A., Petrzelova, I., 2004. Variation and distribution of virulence phenotypes of Bremia lactucae in natural populations of Lactuca serriola. Plant Pathology 53, 316–324. Lebeda, A., Zinkernagel, V., 2003. Evolution and distribution of virulence in the German population of Bremia lactucae. Plant Pathology 52, 41–51. Leppik, E.E., 1962. Distribution ge´ographique du mildiou du tournesol et de quelques autres agents pathoge`nes transmis par les semences. F.A.O. Plant Protection Bulletin 10, 126–129. McDonald, B.A., Linde, C., 2002. Pathogen population genetics, evolutionary potential, and durable resistance. Annual Review of Phytopathology 40, 349–379. Meliala, C., 2001. Biologie du Plasmopara halstedii (Farl.) Berl. et de Toni, agent du mildiou du tournesol (Helianthus annuus L.). Universite´ Blaise Pascal, ClermontFerrand, 150 pp. Moinard, J., 2002–2005. Mildiou du tournesol – Prospection et surveillance du territoire – rapport annuel, SDQV/SRPV Midi-Pyre´ne´e (France). Moinard, J., Mestries, E., Penaud, A., Pinochet, X., Tourvieille de Labrouhe, D., Vear, F., Tardin, M.-C., Pauchet, I., Eychenne, N., 2006. Le mildiou du tournesol: panorama du pathoge`ne. Phytoma 589, 34–38. Molinero-Ruiz, M.L., Dominguez, J., Melero-Vara, J.M., 2002. Races of isolates of Plasmopara halstedii from Spain and studies on their virulence. Plant Disease 86, 736–740. Morin, P.A., Luikart, G., Wayne, R.K., 2004. SNPs in ecology, evolution and conservation. Trends in Ecology and Evolution 19, 208–216. Mouzeyar, S., Tourvieille de Labrouhe, D., Vear, F., 1993. Histopathological studies of resistance of sunflower (Helianthus annuus L.) to downy mildew (Plasmopara halstedii). Journal of Phytopathology 139, 289–297. Novak, S.J., 2007. The role of evolution in the invasion process. Proceedings of the National Academy of Sciences of the United States of America 104, 3671–3672. Novotel’nova, N.S., 1966. Sunflower Downy Mildew. Nauka AS, Moscow, USSR. Pritchard, J.K., Stephens, M., Donnelly, P., 2000. Inference of population structure using multilocus genotype data. Genetics 155, 945–959.

Rashid, K.Y., 1993. Incidence and virulence of Plasmopara halstedii on sunflower in western Canada during 1988–1991. Canadian Journal of Plant Pathology 15, 206–210. Roeckel-Drevet, P., Coelho, V., Tourvieille, J., Nicolas, P., Tourvieille de Labrouhe, D., 1997. Lack of genetic variability in French identified races of Plasmopara halstedii, the cause of downy mildew in sunflower Helianthus annuus. Canadian Journal of Microbiology 43, 260–263. Roeckel-Drevet, P., Tourvieille, J., Gulya, T.J., Charmet, G., Nicolas, P., Tourvieille de Labrouhe, D., 2003. Molecular variability of sunflower downy mildew, Plasmopara halstedii, from different continents. Canadian Journal of Microbiology 49, 492–502. Sackston, W.E., 1981. Downy mildew of sunflower. In: Spencer, D.M. (Ed.), The Downy Mildew. Academic Press, London, UK, pp. 545–575. Sakr, N., Ducher, M., Tourvieille, J., Walser, P., Tourvieille de Labrouhe, D., 2007. A new method to obtain monozoosporangial isolates of sunflower downy mildew. Cryptogamie Mycologie 8, 123–131. Schlotterer, C., 2004. The evolution of molecular markers – just a matter of fashion? Nature Reviews Genetics 5, 63–69. Spring, O., Bachofer, M., Thines, M., Riethmuller, A., Goker, M., Oberwinkler, F., 2006. Intraspecific relationship of Plasmopara halstedii isolates differing in pathogenicity and geographic origin based on ITS sequence data. European Journal of Plant Pathology 114, 309–315. Spring, O., Miltner, F., Gulya, T.J., 1994. New races of sunflower downy mildew (Plasmopara halstedii) in Germany. Journal of Phytopathology 142, 241–244. Spring, O., Zipper, R., 2006. Evidence for asexual genetic recombination in sunflower downy mildew, Plasmopara halstedii. Mycological Research 110, 657–663. Sunnucks, P., DeBarro, P.J., Lushai, G., Maclean, N., Hales, D., 1997. Genetic structure of an aphid studied using microsatellites: cyclic parthenogenesis, differentiated lineages and host specialization. Molecular Ecology 6, 1059–1073. Thines, M., Komjati, H., Spring, O., 2005. Exceptional length of ITS in Plasmopara halstedii is due to multiple repetitions in the ITS-2 region. European Journal of Plant Pathology 112, 395–398. Tourvieille de Labrouhe, D., 1988. Une nouvelle race de mildiou en France, identification, test et controˆle. Information Technique CETIOM 104, 3–9. Tourvieille de Labrouhe, D., 1999. La nouvelle nomenclature des races de Plasmopara halstedii, agent du mildiou du tournesol, applique´e aux races franc¸aises. Ole´agineux Corps Gras Lipides 6, 219–222. Tourvieille de Labrouhe, D., Mestries, E., Walser, P., 2005. Quelles perspectives pour la lutte ge´ne´tique vis-a`-vis du mildiou du tournesol? Ole´agineux Corps gras Lipides 12, 85–93. Tourvieille de Labrouhe, D., Pilorge´, E., Nicola, P., Vear, F., 2000. Le mildiou du tournesol. INRA and CETIOM, Paris, France. Vear, F., Gentzbittel, L., Philippon, J., Mouzeyar, S., Mestries, E., Roeckel-Drevet, P., Tourvieille de Labrouhe, D., Nicolas, P., 1997. The genetics of resistance to five races of downy mildew (Plasmopara halstedii) in sunflower (Helianthus annuus L.). Theoretical and Applied Genetics 95, 584–589. Vear, F., Tourvieille de Labrouhe, D., Miller, J.F., 2003. Inheritance of the wide-range downy mildew resistance in the sunflower line RHA 419. Helia 26, 19–24. Weir, B.S., Cockerham, C.C., 1984. Estimating F-statistics for the analysis of population structure. Evolution 38, 1358–1370. Yong, P.A., Morris, H.E., 1927. Plasmopara downy mildew of cultivated sunflower. American Journal of Botany 14, 551–552. Zimmer, D.E., 1974. Physiological specialization between races of Plasmopara halstedii in America and Europe. Phytopathology 64, 1465–1467.