This Article Abstract Full Text (PDF) Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fournier, E. Articles by Brygoo, Y. Search for Related Content PubMed Articles by Fournier, E. Articles by Brygoo, Y. Agricola Articles by Fournier, E. Articles by Brygoo, Y.

Partition of the Botrytis cinerea complex in France using multiple gene genealogies

     PMDV, INRA Centre de Versailles, Route de Saint-Cyr, F-78026 Versailles cedex, France

Tatiana Giraud

     ESE, Bât. 360, UMR 8079 Université Paris Sud-CNRS, F-91405 Orsay cedex, France

Catherine Albertini

     Phytopharmacie et Médiateurs Chimiques, INRA Centre de Versailles, Route de Saint-Cyr, F-78026 Versailles cedex, France

Yves Brygoo

     PMDV, INRA Centre de Versailles, Route de Saint-Cyr, F-78026 Versailles cedex, France

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION LITERATURE CITED

 

In micro-organisms biodiversity is often underestimated because relevant criteria for recognition of distinct evolutionary units are lacking. Phylogenetic approaches have been proved the most useful in fungi to address this issue. Botrytis cinerea, a generalist fungus causing gray mold, illustrates this problem. It long has been thought to be a single variable species. Recent population genetics studies have shown that B. cinerea is a species complex. However conflicting partitions were proposed. To identify the most relevant partitions within the B. cinerea complex we used a multiple-gene genealogies approach. We sequenced portions of four nuclear genes, of which genealogies congruently clustered into two well supported groups corresponding to Groups I and II previously described, indicating that they represent phylogenetic species. Estimates of migration rates and genetic differentiation showed that these groups had been isolated for a long time, without detectable gene flow. This was confirmed by the high number of polymorphic sites fixed within each group. The genetic diversity was lower within Group I, as revealed by DNA polymorphism and vegetative incompatibility tests. Groups I and II exhibited phenotypic differences in their phenology, host range, size of asexual spores and vegetative compatibility. All these morphological and molecular aspects suggest that B. cinerea Groups I and II may be different cryptic species, isolated for a long time. Phylogenies and molecular analyzes of variance revealed no genetic structure according to the other suggested partitions for the B. cinerea complex (i.e., among host plants, between strains with and without transposable elements, nor between strains responsible for noble rot and gray mold. This suggests that recombination regularly occurs, or occurred until recently, within B. cinerea Group II. This also was supported by recombination rates at each locus. Multiple-gene genealogies showed their utility by providing a relevant partition criterion for the B. cinerea complex.

Key words: cryptic species, genealogical concordance of the phylogenetic species recognition (GCPSR), maximum parsimony, speciation

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
In many micro-organisms, biodiversity often is underestimated because of a lack of relevant criteria for recognition of distinct evolutionary units. During the past decade many "species" of micro-organisms actually have been recognized as species complexes (i.e. composed of genetically distinct groups that are not readily distinguishable by morphological criteria). Such cases are not rare (Kunz 2002, Bush 1994, Via 2001) and seem to be common in fungi (Taylor et al 1999, Dettman et al 2003a). Detecting the cryptic structure of such complexes is essential in the case of pathogens because the different cryptic species might have evolved very different life history traits such as reproductive mode (Kumar et al 1999), phenology, host range (Poulin 1997) or drug resistance (Albertini et al 2002, Fournier et al 2003). The main problem in describing species complexes resides in the method chosen to circumscribe species limits. The most appropriate species criterion for a particular case indeed depends on the speciation mechanism and on the time since speciation (De Queiroz 1998, Le Gac and Giraud 2004, Giraud et al 2006). The morphological species concept is obviously not adapted to the case of sibling species. The biological species concept (Mayr 1940) has proven to be useful in some cases (e.g. for the basidiomycete fungus Armillaria mellea, Guillaumin et al 1996). However Dettman et al (2003b) showed that the use of tester isolates for mating compatibility can underestimate the number of species units. The extended version of the phylogenetic species concept, based on congruence between multiple genealogies (GCPSR for genealogical concordance of the phylogenetic species recognition), recently was described by Taylor et al (2000) and appears to be the most stringent and appropriate tool for the recognition of sibling species in fungi (e.g. Grube and Kroken 2000, Dettman et al 2003b). GCPSR avoids possible confusion of single genealogies based on genes subject to balancing selection (Ward et al 2002), persistence of ancient polymorphism in daughter species (Wu 1991), horizontal gene transfer (Yoshiyama et al 2001) or paralogy (Brocchieri 2001). It allows the detection of barriers to gene flow that in this approach are characterized as the transition from conflict to concordance among the different genealogies as one progresses from the branch tips to the base (Avise and Wollenberg 1997, Taylor et al 2000). The multiple-gene genealogies approach has proven to be useful in detecting genetic isolation within morphological species in agriculturally or medically important fungi (for a review, see O’Donnell 2004) but also within biological species (Dettman, Jacobson and Taylor 2003a, Dettman et al 2003b). Furthermore this approach also allows detection of life-history particularities such as mode of reproduction (Carbone et al 1999) or geographical structuring of populations within a given species (O’Donnell et al 2000, Fisher et al 2001).

The filamentous fungus Botrytis cinerea (Pers.) is a good illustration of the problem of finding the appropriate criterion to detect distinct evolutionary units within a described species of agronomic pathogens. This phytopathogenic ascomycete is responsible for gray mold on more than 200 host plants, causing severe damages to numerous crops (i.e. grapevines, kiwifruits, strawberries, lettuces). It is also the agent of noble rot on grapevine, used for the elaboration of sweet wines. B. cinerea long has been thought to be a single but morphologically variable and polyphagous species. Several recent studies however have shown that B. cinerea was likely to form a species complex, with restricted gene flow among different cryptic genetic groups (Giraud et al 1997, Albertini et al 2002, Fournier et al 2003). First Giraud et al (1997) identified two groups based on the presence or absence of two transposable elements (TE) in the genome: B. cinerea var vacuma (isolates without the two transposable elements) and B. cinerea var transposa (isolates with both active transposable elements). More recently Albertini et al (2002) and Fournier et al (2003) studied the DNA polymorphism of two different nuclear genes. Their results were congruent in showing that B. cinerea isolates clustered in two different genetically isolated subgroups, Group I and Group II. Group I strains were only of vacuma TE type, whereas Group II strains included vacuma and transposa TE types. All these works therefore suggest that B. cinerea is a complex of cryptic species, but their exact limits are still unclear and not concordant across all studies. Furthermore the different cryptic groups seem to exhibit differences in their host range (Muñoz et al 2002, Giraud et al 1999), their genetic diversities (Fournier et al 2003), their temporal succession (Giraud et al 1997; Martinez et al 2005) and some of their phenotypic characteristics (Martinez et al 2003), although they frequently are found in sympatry (Giraud et al 1999, Fournier et al 2003). Another partition of the B. cinerea complex that would be worth testing is between the strains responsible for noble rot and those responsible for gray mold on grapes. It indeed has never been investigated whether these dramatically different symptoms could be due to distinct and differentiated populations.

The goal of the present study was to identify the most relevant partition of the B. cinerea complex using multiple-genealogies approach. We examined these two questions: (i) Are there actual cryptic species within the morphospecies B. cinerea, and if so, what are their limits among the previously proposed partitions of the complex? (ii) Do the different evolutionary entities exhibit different genetic or life-history features ? To answer these questions, we applied the multiple gene genealogies approach by sequencing four nuclear loci in a set of 43 B. cinerea isolates collected from several host plant species mainly from France, including vacuma and transposa strains, isolates causing either gray mold or noble rot symptoms and strains from three additional Botrytis morphological species. When possible we added sequences from two outgroup species, Sclerotinia sclerotiorum and Sclerotinia minor. To further characterize the biological variability within the B. cinerea species complex, we analyzed the conidium size and the ability of the strains to fuse vegetatively.

	MATERIAL AND METHODS

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
Cultures.— – Forty-three B. cinerea monospore isolates were sequenced (TABLE I), including vacuma and transposa isolates, belonging to Group I and Group II, and collected from various host plants, mainly in France. One isolate that we identified as belonging to B. fabae based on morphological criteria was added in the analysis, but this origin remains uncertain. We also included one isolate of Sclerotinia sclerotiorum and one of S. minor that had been isolated on lettuce and identified morphologically (P. Davet, personal communication). All isolates were preserved in 20% glycerol at –80 C.

Sclerotinia is the closest genus to Botrytis (Carbone and Kohn 1993). We succeeded in the amplification of the ß-tubulin and Bc-hch loci for one isolate of S. sclerotiorum and one isolate of S. minor. Hence we used these sequences as outgroups for the phylogenies of these two loci.

DNA amplification and sequencing.— – Fragments in the four loci were PCR-amplified and sequenced (FIG. 1), using several pairs of primers. Primers 155 (5'-CAACCTTCAAAATGCGTGAG-3') and 1174 (5'-AGATGGGTTGCTGAGCTTCA-3'), primers Beg (5'-TGCGATGGGGATTCTTGAAC-3' and End (5'-TTATCGTCGCTCCCAAGCTAC-3'), and primers 258 (5'-TGATGATCGGTGAAGACTCC-3') and 866 (5'-CTTTACCAACAGGGCTCCAT-3'), were used for amplification and sequencing at both extremities of the resulting PCR products for loci ß-tubulin, CYP51 and 63R respectively. For Bc-hch, primers 262 (5'-AAGCCCTTCGATGTCTTGGA-3') and 520 L (5'-ACGGATTCCGAACTAAGTAA-3') were used for the amplification, and primers 262 and 101 L (5'-CCGTGTTTATTCTATCTTTG -3') for the sequencing. All PCR amplifications were done in a total volume of 50 µL containing 0.2 mg genomic DNA, 5 µL reaction buffer, 1 mM MgCl2, 2.5 µL of each 10 µM concentrated primer and 0.2 U of EUROGENTEC Taq polymerase, with 30 cycles of 30 s at 94 C, 1 min 30 s at 55 C (for Bc-hch, ß-tubulin and ITS) or 60 C (for CYP51 and 63R), and 1 min at 72 C. Sequences were determined directly from purified PCR products (GFX Purification Kit, Amersham; DNA Sequencing Dye Terminator Kit, Applied Biosystems)with an ABI Prism 310, then analyzed with Chromas and manually aligned with BioEdit.

View larger version (9K): [in this window] [in a new window]   FIG. 1. Localization of the sequenced fragments within the open reading frames of the three coding genes analyzed (Bc-hch, CYP51 and ß-tubulin,). Black boxes indicate introns, dashed boxes indicate sequenced fragments. Locations of the primers used for sequencing are underlined.

Once diversifying selection was rejected, the four individual gene genealogies were reconstructed by unweighted maximum parsimony using PAUP 4.0 (Swofford 1998), with heuristic search (tree bisection-reconnection [TBR] branch swapping; MAXTREES= 5000) and 500 parsimony bootstraps replications (simple stepwise addition). When sequences from Sclerotinia sclerotiorum and S. minor were available (i.e., for Bc-hch and ß-tubulin loci), they were used as outgroups to root the phylogenies.

Several isolates could not be amplified in some loci (TABLE I) despite several attempts of PCR optimisation and DNA re-extractions. This might be due to point mutations in these strains in one or both regions corresponding to primer sequences. Therefore individual gene genealogies were recalculated as above after having excluded the 13 missing sequences. The four "minimal" resulting datasets each encompassed 33 sequences. To determine the congruence among trees we first performed pairwise partition homogeneity tests (PHT, Farris et al 1995) as implemented in PAUP (HOMPART option). In this test pairwise sequences were pooled and resampled N times without replacement to give N artificial datasets. Maximum parsimonious (MP) trees then were reconstructed for each of the N artificial datasets, and their lengths were compared to the length of the observed summed MP trees. P, the probability of obtaining an artificial MP tree similar or shorter to the observed summed tree length, then was calculated. The significant threshold was corrected with the Bonferroni method (six different PHT, significant threshold for each test: P = 0.05/6 = 0.008); this means that for each pairwise comparison, the null hypothesis of incongruence was rejected when P < 0.008. This value is also in accordance with Darlu and Lecointre (2002), who argued that PHT was too conservative with a significant threshold of 0.05. All PHT were performed with N = 500 replicates. The four datasets also were compared globally using the PHT. Because several recent works have underlined that the ILD test is actually a poor indicator of dataset combinability (e.g. Barker and Lutzoni 2002), and because the GPSCR method requires that the different nodes are examined individually, congruence between gene phylogenies also was estimated by visual inspection of topologies and statistical supports (Mason-Gamer and Kellogg 1996). Nodes were considered as incongruent when strong statistical values of at least two of the four phylogenetic reconstruction methods supported conflicting nodes. Because the global PHT test was significant, no tree was constructed using a combined dataset.

Inter- and intragroup recombination at each locus.— – Recombination parameters at each locus were assessed on the total minimum dataset of each locus, using the SITES software (Hey and Wakeley 1997). This program implements several methods, including the calculation of the minimum set of recombination intervals (Hudson and Kaplan 1985) and several estimators of the population recombination rate. Here we used Hey’s and Wakeley’s (1997) estimator, with = 4Nc for diploid organisms, N being the effective population size and c the crossing-over rate per generation. The same analysis also was performed considering only isolates from Group I or only isolates from Group II to assess whether recombination was still a contemporary active force within these groups.

Divergence history and dating of the divergence event.— – We first tested the null hypothesis of lack of genetic differentiation between Groups I and II by using four test statistics implemented in DNAsp 3.5: the traditional ² test proposed by Nei (1987), based on allele frequencies, and the three nonparametric sequence statistics proposed by Hudson et al (1992a). Among these latter H_S represents the weighted average of H_I and H_II, which are the estimated haplotype diversities in Groups I and II, respectively; K_S* represents the weighted average between K_I and K_II, which are the numbers of nucleotide differences among sequences of Groups I and II respectively, this weighted average being calculated with a correction that does not give as much weighting to large numbers of nucleotide differences; Z* is a weighted sum of Z*_I and Z*_II, where these are rank statistics representing the average of the logarithm of 1 plus the rank of the number of nucleotide differences for all pairs of sequences among Groups I and II, respectively. According to Hudson et al (1992a), these three statistics, together with the Nei’s ² test, are the most powerful tests of the null hypothesis of no genetic differentiation between subpopulations under various conditions (various migration or mutations rates). For each observed value of these four statistics, the associated P value, representing the probability of obtaining either the observed value of the statistic or a more extreme one, was estimated by using a permutation-based method with 1000 replicates. This method first pools Group I and Group II samples, then randomly partitions this pooled sample 1000 times and recalculates the statistics each time. If the P value of the statistic is small, in effect lower than 0.05, then the null hypothesis is rejected.

We also tested whether there was gene flow between Group I and Group II by estimating the product Nm (where N is the effective population size and m the fraction of migrants per generation) using DNAsp 3.5. The chosen estimator of Nm was based on F_ST estimates because this has been reported as the most appropriate estimator in case of haploid genomes (Hudson et al 1992b).

The date of the speciation event between B. cinerea Group I and Group II was estimated using the pairwise divergence between these two cryptic species, which is the average number of nucleotide substitutions per site between the two groups, d_I–II (Nei 1987).

Genetic diversity analyzes.— – Nucleotidic and haplotypic diversities were estimated for each locus in each clade using DNAsp 3.5 (Rozas and Rozas 1999). To investigate possible genetic differentiation within Group II according to host plant, vacuma or transposa type, and symptom on grapevine, we performed molecular analyzes of variance (AMOVA, Excoffier, Smouse and Quattro 1992) using the package Arlequin 2.000 (Schneider 2000). We used the global sequence dataset after having removed all isolates belonging to Group I, Sclerotinia, and other Botrytis sp. The first sample to be analyzed consisted of the 20 isolates out of the 26 B. cinerea Group II isolates for which all four loci could be amplified, which included five vacuma nongrape isolates, five transposa nongrape isolates, six vacuma grape isolates and four transposa grape isolates. To partition the total variance in this sample we used two different models: (i) a model opposing the "grape" population (10 individuals) to the "nongrape" population (10 individuals) to test the effect of host plant on the observed genetic variance; (ii) a model opposing the "vacuma" population (11 isolates) to the "transposa" (nine isolates) population. The second sample to be analyzed consisted of the 16 isolates from grape for which the four loci could be amplified, including six vacuma grape gray mold isolates, four transposa grape gray mold isolates, one vacuma grape noble rot isolate and five transposa grape noble rot isolates. AMOVA opposed the "gray mold" population (10 individuals) to the "noble rot" population (six individuals) to test the effect of symptom on the observed genetic variance.

Vegetative compatibility analyzes within clades.— – Vegetative compatibility analyzes were used to characterize the biological diversity within each group. Vegetative compatibility is the ability, unique to filamentous fungi, to form a viable heterokaryon by hyphal fusion. Vegetative compatibility groups (VCGs) have proven useful for characterizing population diversity in several fungal species (e.g. Katan et al 1991, Elena 1999), including B. cinerea (Beever and Parkes 2003). Vegetative compatibility tests were performed by placing mycelium from cultures of the 44 B. cinerea isolates that were sequenced on the surface of PDA medium. Three different isolates were placed at the edge of each dish grown at 23 C in the dark and observed each day until isolates came into contact, usually 4–6 d later. A strongly colored "barrage" at the confrontation line was considered an incompatible reaction. Each confrontation was repeated at least twice.

Size of asexual spores.— – Giraud et al (1997) showed that asexual spores were significantly smaller in transposa isolates than in vacuma, Groups I and II being pooled. However vacuma and transposa size ranges overlapped (tranposa: 11–12.5 microns, with a mode at 11.5 microns; vacuma: 11–14 microns, with a mode at 12.5 microns). In this study we focused on the difference of spore size between vacuma Group I and vacuma Group II. Five Group I strains (637, 1258, plus three other isolates not used for the phylogenies) and six vacuma Group II strains (32, 619, 646, 715, 1259, plus one other isolate not used for the gene genealogies) were grown on PDA medium until sufficient asexual sporulation (ca. 2 wk). Spores were suspended in sterile water solution, observed under a confocal microscope coupled with a Sony CDD camera. The longest diameter was calculated with Optimas Bioscan (Imasys) for 100 spores of each isolate. The mean length of spores was compared between vacuma Group I and Group II using a nested ANOVA, the "group" effect being nested into the "isolate" effect (SAS package 1989).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
Adequacy of individual datasets to GCPSR purpose.— – We amplified 642 bp in the ß-tubulin locus, 587 bp in the Bc-hch locus, 713 bp in the CYP51 locus, and 442 bp in the 63R locus. One locus or another could not be properly amplified or sequenced in some fungal strains so the datasets of the different loci are not exactly overlapping (TABLE I).

No GC bias was observed in any of the four datasets (average G + C frequency = 0.470, 0.483, 0.449 and 0.465 for ß-tubulin, Bc-hch, CYP51 and 63R, respectively), suggesting an equal probability of mutation toward all nucleotide states in the four loci.

For the three coding loci, the Fisher’s exact test was not significant (P = 1, 0.051 and 0.197 for ß-tubulin, Bc-hch and CYP51, respectively), indicating that the ratio of nonsynonymous to synonymous fixed substitutions between groups did not differ significantly from the ratio of nonsynonymous to synonymous polymorphisms within groups. This result matches the pattern expected under nondiversifying evolution and ensures that the four chosen loci are good candidates for the establishment of neutral genealogies letting one infer speciation events.

ß-tubulin.— – Because nonoverlapping fragments were amplified in the ß-tubulin (FIG. 1), the statistical congruence between the two sequenced parts was checked by a partition homogeneity test. The nonsignificant resulting probability (P = 0.68, with N = 100 replicates) indicated that MP topologies obtained from the two parts of ß-tubulin loci were congruent. Visual inspection of the nodes and their supports also supported this conclusion (not shown). Both sequenced parts therefore were merged and treated as a single 642 bp-long fragment.

This locus could be amplified in all 46 strains (TABLE I). The 10 MP trees obtained after parsimony analysis (FIG. 2a) were 120 steps long and had a consistency index of 0.942. As expected the two Sclerotinia isolates appeared as a strongly supported outgroup (bootstrap value of 100). The nine isolates belonging to Group I clustered into a monophyletic clade supported by a bootstrap value of 100. The 35 remaining isolates all clustered into a second clade, including the B. fabae isolate in a basal position, confirming the uncertainty about its correct identification. The ß-tubulin locus thus clearly separates Group I from Group II. This clustering, supported by a bootstrap score of 100, corresponds to eight fixed nucleotide differences between Groups I and II isolates (without considering the B. fabae isolate).

View larger version (34K): [in this window] [in a new window]   FIG. 2. One of the most parsimonious trees obtained for ß-tubulin (a), Bc-hch (b), CYP51 (c) and 63R (d) loci, with the total datasets for each locus. Bootstrap values more than 50% (N = 500 replicates) are indicated on the corresponding nodes. CI: consistency index. I: Botrytis cinerea Group I. II: Botrytis cinerea Group II. V: vacuma isolates, T: transposa isolates. Gray stars indicate gray mold isolates coming from nongrape hosts, gray squares indicate gray mold isolates coming from grape and white squares indicate noble rot isolates coming from grape. Non-French isolates are marked respectively by e, g, i and j for England, Germany, Italy and Japan.

CYP51.— – As in the ß-tubulin analysis, two non-overlapping fragments were amplified (FIG. 1c). The partition homogeneity test (P = 0.602, with N = 100 replicates) indicated statistical congruence between the two sequenced parts, in agreement with visual inspection of the nodes and supports (not shown). The two parts consequently were merged and treated as a single 713 bp-long fragment.

This locus could neither be amplified in the two Sclerotinia isolates, in the B. fabae isolate, nor in the B. cinerea isolates No. 395 and 686. The resulting topology (48 steps, CI = 0.833; FIG. 2c), recovered in six MP trees, again clearly separated two strongly supported clades, corresponding to Groups I and II. This segregation corresponds to 14 fixed mutations separating these two clades. There was a rather high level of DNA polymorphism within Group II and subclades were present, some being well supported. These subclades did not bring together strains of similar TE type, host species nor grape symptom (gray vs noble rot).

63R.— – This 442 bp noncoding sequence contains a microsatellite locus between base 31 and base 78 (included), whose core sequence is four bp long, with variable numbers of repetitions among isolates. The exact mutational model associated to this microsatellite has not been investigated. We assumed a stepwise mutation model, so characters 31 to 78 received a weight of 0.25 to give a relative weight of 1 to indel events involving the four bp-repeat (in flanking regions any mutational event had a weight of 1). A 20 bp insertion also was observed in the B. cinerea Group I isolates between positions 312 and 331 (included). We considered that this insertion event implied a single mutational event and we thus weighed each of these 20 characters to 1/20. The same principle was applied for another nine bp sequence between characters 400 and 408 (included) deleted in B. cinerea Group I isolates (given weight for each bp; 1/9). Note that the same tree topology was recovered when the analysis was done on microsatellite flanking regions only (data not shown).

This locus could not be amplified in strains No. 750 and 774 and in the two Sclerotinia isolates. In the each of the nine resulting MP trees (40.1 steps, CI = 0.890, FIG. 2d), the 40 remaining isolates clustered in two strongly supported clades, again corresponding to Groups I and II. At this locus the segregation between Groups I and II corresponds to 10 fixed polymorphisms (microsatellite region excluded). As found with the CYP51 locus, Group II appeared more polymorphic than Group I and some subclustering was present, although weakly supported. Here again the substructure did not correspond to expected patterns (TE type, host or symptom).

GCPSR application: comparison of independent minimal topologies.— – To compare the four datasets, isolates in which any of the four genes could not be amplified were removed from the analysis. The resulting datasets encompassed 33 isolates lacking the Sclerotinia and the B. fabae isolates (FIG. 3). Partition homogeneity tests were performed pairwisely and globally. For these tests only microsatellite flanking regions of locus 63R were taken into account and not the microsatellite locus in itself. With a Bonferroni corrected significance threshold of 0.008 (see Material and Methods), pairwise P values were never significant. The global PHT was significant (P = 0.012), probably due to the nodes within Group II that were not consistent across the four phylogenies. The significance of the global PHT also might be due partly to the weak relative number of informative sites in the global dataset, which places the global PHT in a limit case as described by Darlu and Lecointre (2002).

View larger version (24K): [in this window] [in a new window]   FIG. 3. One of the most parsimonious trees obtained for ß-tubulin (a), Bc-hch (b), CYP51 (c) and 63R (d) loci, with the minimum datasets (33 isolates) for each locus. Bootstrap values more than 50% (N = 500 replicates) are indicated on the corresponding nodes. CI: consistency index. I: Botrytis cinerea Group I. II: Botrytis cinerea Group II. Symbols for isolates origin and phenotype have the same meaning as in FIG. 2.

Recombination analyzes.— – Recombination analyzes were conducted on the minimum datasets for each locus (33 isolates, see FIG. 3), considering either all isolates, isolates from Group I only or isolates from Group II only. No recombination event was detected within Group I at any of the four loci. Low polymorphism within Group I yielded actually low power to detect recombination. No recombination event was detected at the locus Bc-hch, neither for the entire dataset nor within Group II. For the ß-tubulin locus, two recombination intervals were detected within Group II (positions 378–506, positions 506–635). On the entire dataset, a single interval of recombination was detected between positions 348 and 635, corresponding to one of the recombination events detected within Group II. At this locus, the Hey and Wakeley (1997) estimation of recombination rate per base pair within Group II was = 10^–5. At the locus CYP51, we found three recombination intervals when both groups were considered together (positions 137–228, positions 228–248, positions 542–707). Within Group II, three recombination intervals were detected, all three being equal to or including the intervals detected on the entire dataset (positions 137–228, positions 228–248, positions 248–707). The recombination estimate per base pair within Group II was = 0.0043. A single recombination interval was detected at locus 63R when both groups were taken into account (positions 18–360), identical to the interval recovered when Group II was considered alone. At this locus the recombination estimate per base pair within Group II was = 0.004.

Divergence between groups I and II.— – Estimates of genetic differentiation were calculated on both total datasets and minimum datasets for each locus. For all loci, Nei’s ² test as well as H_S, K*_ST and Z* Hudson’s statistics were all highly significant, indicating that the null hypothesis of no genetic differentiation between Group I and Group II could unambiguously be rejected (TABLE II). The estimators of F_ST and Nm confirmed that the differentiation between the two groups was strong, with F_ST values varying from 0.749 (CYP51) to 0.922 (ß-tubulin) for the total datasets and from 0.874 (CYP51) to 0.931 (63R) for the minimal datasets. Corresponding Nm estimates varied from 0.04 (Bc-hch, ß-tubulin) to 0.17 (CYP51) for the total datasets and from 0.04 (63R, Bc-hch) to 0.07 (CYP51, ß-tubulin) for minimal datasets. These Nm estimates indicate that the level of migration between the two groups is close to zero.

View this table: [in this window] [in a new window]   TABLE II. Tests of the null hypothesis of no genetic differentiation between Groups I and II following Hudson et al. (1992a) and estimates of gene flow

Recombination analyzes showed that recombination events recently occurred within Group II and examination of its subclades showed that none was strongly supported and consistently recovered across the four individual gene genealogies. In particular individual gene genealogies did not show any pattern of segregation within Group II according to host plant, symptom or ET status (FIG. 3). This might be due to a lack of resolution of this type of analysis at such a low phylogenetic level. To confirm the apparent absence of genetic structure within Group II, we performed AMOVA analyzes, using the global sequence dataset with strains from Group II only. Two AMOVAs were performed on a first sample (20 isolates) to test for the effects of host plant and transposable elements separately. Both factors explained a low percentage of the observed genetic variance (1.11% for the host plant, 0.93% for the transposable elements) and their effects were nonsignificant (F_ST = 0.011, P = 0.47 for host plant, F_ST = 0.0093, P = 0.29 for transposable elements). A third AMOVA was performed on a sample composed of 16 grape gray mold and noble rot isolates to test for the effect of the symptom on grape. Results showed that this factor was not significant either (3.07% of the observed genetic variance was explained by this factor, F_ST = 0.031, P = 0.29). Altogether these results confirmed the absence of strong genetic structuring within Group II.

Comparison of VCG diversity within the two clades.— – In vegetative compatibility tests all Group I isolates formed a unique vegetative compatibility group (VCG), whereas several VCGs were detected within Group II (FIG. 4). No VCG overlapped between the two groups.

View larger version (24K): [in this window] [in a new window]   FIG. 4. Results of the pairwise confrontations of the vegetative compatibility experiments. Gray boxes indicate the occurrence of a melanized barrage (i.e. an incompatible reaction between mycelia), white boxes indicate a compatible reaction between mycelia and dashed boxes indicate uncertain results.

View larger version (21K): [in this window] [in a new window]   FIG. 5. Mean size of 100 conidiospores for 5 vacuma Group I and 6 vacuma Group II isolates.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION LITERATURE CITED

The nonparametric tests that we performed following Hudson et al (1992a) consistently let us reject unambiguously the null hypothesis of no genetic differentiation between Group I and Group II. We also obtained high values of F_ST estimates for all loci, varying from 0.749 to 0.931, leading to low values of Nm estimates. These low values of migration rates, together with the presence of recombination at three of the four loci, place our datasets in the range of maximal power of the tests for detecting subdivision (Hudson et al 1992a). Therefore gene flow and migration can be considered as null between Group I and Group II.

To measure the time since divergence between the two groups we estimated the net pairwise between groups. These estimates dated the divergent event around 10⁷ generations ago and would be consistent with those obtained in the recombination analyzes if we assume an effective size of N = 2*10⁵.

The main result of this study thus is to firmly establish that B. cinerea as a whole is not a panmictic species and that the main barrier to gene flow is between Group I and Group II. These groups appeared to have been strongly isolated, exchanging no migrants, for a long time. This indicates that Group I and Group II are true phylogenetic species. In contrast the other entities previously proposed as cryptic species (Giraud et al 1997) were not supported as such by our multiple-gene phylogenies. A phylogenetic revision of the whole Botrytis genus recently was published (Staats et al 2005), based on multiple gene genealogies. Only strains from Group II however were included in this study. The next step then is to precisely locate the phylogenetic position of Group I within the whole genus, to determine whether Group I and Group II form are monophyletic or whether other Botrytis spp. are phylogenetically closer to Group II than is Group I.

This study again illustrates the usefulness of GCPSR, especially in the study of fungal biodiversity. In our case the four chosen loci appeared to be appropriate for GCPSR application because diversifying evolution was not detected for them. Even the CYP51 locus was not under diversifying evolution, although the protein coded by CYP51 is the potential target of fenhexamid, an anti-Botrytis fungicide (Albertini et al 2002, Leroux et al 2001). This can be explained by the fact that all the isolates analyzed in the present study were collected before the use of fenhexamid in the field, which started in 2000 in France. This locus however might not be a good candidate for the analysis of neutral genetic variability in current populations.

The existence of an ancient species barrier between B. cinerea Groups I and II is further supported by several biological particularities. The significant difference in the size of asexual spores reported here was consistent with previous studies (Giraud et al 1997); we showed that vacuma Group I conidiospores were significantly longer than vacuma Group II conidiospores, the latter being not different from tranposa Group II conidiospores length reported by Giraud et al (1997). The bimodal curve obtained by these authors for vacuma isolates probably was due to the fact that, unaware of the existence of Groups I and II, they pooled vacuma isolates from the two groups. The difference in conidiospore length thus may be considered as a diagnostic morphological difference between Group I and Group II, although this result has to be confirmed on a larger number of isolates. Leroux et al (2001) also reported differences in mycelial growth speed on synthetic medium between the two groups. Groups I and II also presented different genetic features. In particular the genetic variability was much lower within Group I. The results of confrontation tests in Group II indicate that the genetic variability may be very high at the involved het locus or loci (which is the case in all filamentous fungi examined to date; Bégueret 1994, Glass 2000, Saupe 2000) and that Group II probably includes more than one VCG. On the contrary, B. cinerea Group I seems to form a unique VCG. This low genetic and biologic variability in Group I can be due either to demographical processes, such as a recent colonization event, a strong bottleneck or a selective sweep (Galtier et al 2000) or to an absence of sexual reproduction (Tibayrenc et al 1991) and loss of variability by drift.

The main question that remains to be answered is the biological specificities of Group I and Group II isolates. As a preliminary approach we re-analyzed part of the data from Giraud et al (1997), taking into account the assignation of vacuma strains to Group I or Group II (on the basis of their fenhexamid resistance phenotype). The temporal distribution of isolates sampled on grapevine in 1995 were compared between vacuma Group I, vacuma Group II and tranposa Group II. Vacuma Group I isolates were present mainly at the beginning of the season, whereas tranposa Group II dominated in August and October (FIG. 6a), in agreement with previous studies (Giraud et al 1997, Martinez et al 2005). We also reexamined the distribution of the different groups on several host plants among the vegetation surrounding the sampled vineyards. Group II had a larger host range than Group I and were not present on the same plants (FIG. 6b). Although these results have to be completed and actualized, they indicate that the two cryptic species might not be adapted to the same host plants and might not have the same biological abilities, tranposa Group II isolates being the most pathogenic on grapevine.

View larger version (23K): [in this window] [in a new window]   FIG. 6. Differences in the phenologies between vacuma Group I, vacuma Group II and tranposa Group II. (a) Seasonal occurrence on grapevine. (b) Occurrence on different host plants surrounding the vineyard.

Whether the cessation of gene flow between B. cinerea groups I and II has been caused, or has been accompanied, by the evolution of an active reproductive barrier between the two groups will be an interesting issue to settle, but is not necessarily relevant to the identification of evolutionary independent lineages. Intersterility is indeed not necessarily the appropriate criterion for detecting the most terminal evolutionary units in fungi (Dettman et al 2003a, 2003b, Giraud et al in press). A study of intersterility patterns, involving systematic tests of mating types and pairwise crosses, is now in project within the whole Botrytis genus to assess whether the phylogenetic species identified in this study correspond to single intersterility groups. More generally there might not be a universal criterion for recognizing evolutionary units. As assessed by De Queiroz (1998), the evolution of intersterility and of diagnostic characters depend on the mode of speciation and the time since divergence, and the same author (2005) recently argued that intersterility may not immediately follows the emergence of new species. The most appropriate approach therefore may be to combine several species criteria.

	ACKNOWLEDGMENTS

 
We thank J.-Y. Rasplus, J.-M. Cornuet, A. Estoup and J.A. Shykoff and anonymous reviewers for useful comments of previous drafts of the manuscript and C. Giraud-Delville and B. Faivre for help in the acquisition of the sequence data. This work was partially supported by Fondation Singer-Polignac and by an ACI Jeunes Chercheurs (CNRS).

	FOOTNOTES

 
Accepted for publication October 31, 2005.

¹ Corresponding author. E-mail: efournie{at}versailles.inra.fr

	LITERATURE CITED

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
Albertini C, Thébaud G, Fournier E, Leroux P. 2002. Eburicol 14–demethylase gene (cyp51) polymorphism and speciation in Botrytis cinerea. Mycol Res 106: 1171–1178.[CrossRef]

Avise JC, Wollenberg K. 1997. Phylogenetics and the origin of species. Proc Natl Acad Sci USA 94:7748–7755.[Abstract/Free Full Text]

Barker FK, Lutzoni FM. 2002. The utility of the incongruence length difference test. Syst Biol 51:625–637.[CrossRef][Medline]

Bégueret J, Turcq B, Clavé C. 1994. Vegetative incompatibility in filamentous fungi: het genes begin to talk. Trends Genet 10:441–445.[CrossRef][Medline]

Beever RE, Parkes SL. 2003. Use of nit mutants to determine vegetative compatibility in Botryotinia fuckeliana (Botrytis cinerea). Eur J Plant Pathol 109: 607–613.[CrossRef]

Brocchieri L. 2001. Phylogenetic inferences from molecular sequences: review and critique. Theoretical Pop Biol 59:27–40.[CrossRef]

Burt A, Carter DA, Koenig GL, White TJ, Taylor JW. 1996. Molecular markers reveal cryptic sex in the human pathogen Coccidioides immitis. Proc Natl Acad Sci USA 93:770–773.[Abstract/Free Full Text]

Bush GL. 1994. Sympatric speciation in animals: new wine in old bottles. Trends Ecol Evol 9:285–288.[CrossRef]

Carbone I, Anderson JB, Kohn LM. 1999. Patterns of descent in clonal lineages and their multilocus fingerprints are resolved with combined gene genealogies. Evolution 53:11–21.

———, Kohn LM. 1993. Ribosomal DNA sequence divergence within internal transcribed spacer 1 in the Sclerotiniaceae. Mycologia 85:415–427.[CrossRef]

Darlu P, Lecointre G. 2002. When does the incongruence length difference test fail? Mol Biol Evol 19:432–437.[Abstract/Free Full Text]

De Queiroz K. 1998. The general lineage concept of species, and the process of speciation. In: Howard DJ, Belocher SH, eds. Endless Forms: species and speciation. Oxford University Press. p 57–75.

———. 2005. Ernst Mayr and the modern concept of species. Proc Natl Acad Sci USA 102:6600–6607.[Abstract/Free Full Text]

Dettman JR, Jacobson DJ, Taylor JW. 2003a. A multilocus genealogical approach to phylogenetic species recognition in the model eukaryote Neurospora. Evolution 57:2703–2720.[CrossRef][Medline]

———, ———, Turner E, Pringle A, Taylor JW. 2003b. Reproductive isolation and phylogenetic divergence in Neurospora: comparing methods of species recognition in the model eukaryote Neurospora. Evolution 57: 2721–2741.[CrossRef][Medline]

Elena K. 1999. Genetic relationships among Verticillum dahliae isolates from cotton in Greece based on vegetative compatibility. Eur J Plant Pathol 105: 609–616.[CrossRef]

Excoffier L, Smouse P, Quattro J. 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human and mitochondrial DNA restriction data. Genetics 131: 479–491.[Abstract]

Farris JS, Kallersjo M, Klug AG, Bult C. 1995. Testing significance of incongruence. Cladistics 10:315–319.[CrossRef]

Fisher MC, Koenig GL, White TJ, San-Blas G, Negroni R, Alvarez IG, Wanke B, Taylor JW. 2001. Biogeographic range expansion into South America by Coccidioides immitis mirrors New World patterns of human migration. Proc Natl Acad Sci USA 98:4558–4562.[Abstract/Free Full Text]

———, ———, ———, Taylor JW. 2000. A test for concordance between the multilocus genealogies of genes and microsatellites in the pathogenic fungus Coccidioides immitis. Mol Biol Evol 17:1164–1174.[Abstract/Free Full Text]

Fournier E, Giraud T, Loiseau A, Vautrin D, Estoup A, Solignac M, Cornuet JM, Brygoo Y. 2002. Characterization of nine polymorphic microsatellite loci in the fungus Botrytis cinerea (Ascomycota). Mol Ecol Notes 2:253–255.[CrossRef]

———, Lévis C, Fortini D, Giraud T, Leroux P, Brygoo Y. 2003. Characterization of Bc-hch, the Botrytis cinerea homolog of the Neurospora crassa het-c vegetative incompatibility locus, and its use as population marker. Mycologia 95:251–261.[Abstract/Free Full Text]

Galtier N, Depaulis F, Barton NH. 2000. Detecting bottlenecks and selective sweeps from DNA sequence polymorphism. Genetics 155:981–987.[Abstract/Free Full Text]

Geiser DM, Pitt JI, Taylor JW. 1998. Cryptic speciation and recombination in the aflatoxin-producing fungus Aspergillus flavus. Proc Natl Acad Sci USA 95:338–393.

Giraud T, Fortini D, Levis C, Lamarque C, Leroux P, LoBuglio K, Brygoo Y. 1999. Two sibling species of the Botrytis cinerea complex, transposa and vacuma, are found in sympatry on numerous host plants. Phytopathology 89:967–973.[Medline]

———, ———, ———, Leroux P, Brygoo Y. 1997. RFLP markers show genetic recombination in Botryotinita fuckeliana (Botrytis cinerea) and transposable elements reveal two sympatric species. Mol Biol Evol 14: 1177–1185.[Abstract]

———, Villaréal LMMA, Austerlitz F, Le Gac M, Lavigne C. 2006. Importance of the life cycle in sympatric host race formation and speciation of pathogens. Phytopathology 96:280–287.[Medline]

Glass NL, Jacobson DJ, Shiu PKT. 2000. The genetics of hyphal fusion and vegetative incompatibility in filamentous Ascomycete fungi. Ann Rev Genet 34: 165–186.[CrossRef][Medline]

Grube M, Kroken S. 2000. Molecular approaches and the concept of species and species complexes in lichenized fungi. Mycol Res 104:1284–1294.[CrossRef]

Guillaumin JJ, Anderson JB, Legrand P, Gharari S, Berthelay S. 1996. A comparison of different methods for the identification of genets of Armillaria spp. New Phytologist 133:333–343.[CrossRef]

Hennig W. 1950. Grundzüge einer Theorie des phylogen-etischen Systematik. Deutscher Zentral Verlag, Berlin.

Hey J, Wakeley J. 1997. A coalescent estimator of the population recombination rate. Genetics 145:833–846.[Abstract]

Holst-Jensen A, Kohn LM, Jakobsen KS, Schmacher T. 1997. Molecular phylogeny and evolution of Monilia (Sclerotiniaceae) based on coding and noncoding rDNA sequences. Am J Bot 84:686–701.[Abstract]

Hudson RR, Boos DD, Kaplan NL. 1992a. A statistical test for detecting geographic subdivision. Mol Biol Evol 9:198–151.

———, Kaplan NL. 1985. Statistical properties of the number of recombination events in the history of a sample of DNA sequences. Genetics 111:147–164.[Abstract/Free Full Text]

———, Slatkin M, Maddison WP. 1992b. Estimation of levels of gene flow from DNA sequence data. Genetics 132:583–589.[Abstract]

Katan T, Zamir D, Sarfatti M, Katan J. 1991. Vegetative compatibility groups and subgroups in Fusarium oxysporum f. sp. radicis-lycopersici. Phytopathology 81:255–262.[CrossRef]

Kumar J, Nelson RJ, Zeigler RS. 1999. Population structure and dynamics of Magnaporthe grisea in the Indian Himalayas. Genetics 152:971–984.[Abstract/Free Full Text]

Kunz W. 2002. When is a parasite species a species? Trends Parasitol 18:121–124.[CrossRef][Medline]

Le Gac M, Giraud T. 2004. What is sympatric speciation in parasites? Trends Parasitol 20:207–208.[CrossRef][Medline]

Leroux P, Debieu D, Albertini C, Arnold A, Bach J, Chapeland F, Fournier E, Fritz R, Gredt M, Hugon M, Lanen C, Malosse C, Thebaud G. 2001. The hydroxyanilide botryticide fenhexamid: mode of action and mechanisms of resistance. In: Dehne HW, et al, eds. 2001. Modern fungicides and antifungal compounds III, 13th International Reinhardsbrumm symposium AgroConcept: Bonn. p 29–40.

Martinez F, Blancard D, Lecomte P, Levis C, Dubos B, Fermaud M. 2003. Phenotypic differences between vacuma and transposa subpopulations of Botrytis cinerea. Eur J Plant Pathol 109:479–488.[CrossRef]

———, Dubos B, Fermaud M. 2005. The role of saprophy and virulence in the population dynamics of Botrytis cinerea in vineyards. Phytopathology 95:692–700.[Medline]

Mason-Gamer RJ, Kellogg EA. 1996. Testing for phylogenetic conflict among molecular datasets in the tribe Triticeae (Gramineae). Syst Biol 45:524–545.[CrossRef]

Mayr E. 1940. Speciation phenomena in birds. Am Nat 74:249–278.[CrossRef]

McDonald JH, Kreitman M. 1991. Adaptive protein evolution at the Adh locus in Drosophila. Nature 351: 652–654.[CrossRef][Medline]

Möller E, Bahnweg MG, Sandermann H, Geige HH. 1992. A simple and efficient protocol for isolation of high molecular weight DNA from filamentous fungi fruit bodies and infected plant tissues. Nucleic Acid Res 20:6115–6116.[Free Full Text]

Muñoz G, Hinrichsen P, Brygoo Y, Giraud T. 2002. Genetic characterisation of Botrytis cinerea populations in Chile. Mycol Res 106:595–602.

Nei M. 1987. Molecular Evolutionary Genetics. Columbia Univ Press, New York.

O’Donnell K, Cigelnik E, Casper HH. 1998. Molecular phylogenetic, morphological, and mycotoxin data support re-identification of the Quorn mycoprotein fungus as Fusarium venenatum. Fungal Genet Biol 23:57–67.[CrossRef][Medline]

———, Corby-Kistler H, Tacke BK, Casper HH. 2000. Gene genealogies reveal global phylogeographic structure and reproductive isolation among lineages of Fusarium graminearum, the fungus causing wheat scab. Proc Natl Acad Sci USA 97:7905–7910.[Abstract/Free Full Text]

———, Ward TJ, Geiser DM, Corby Kistler H, Aoki T. 2004. Genealogical concordance between the mating type locus and seven other nuclear genes supports formal recognition of nine phylogenetically distinct species within the Fusarium graminearum clade. Fung Genet Biol 41:600–623.

Poulin R. 1997. Parasite faunas of freshwater fish: the relationship between richness and the specificity of parasites. Int J Parasitol 27:1091–1098.[CrossRef][Medline]

Rozas J, Rozas R. 1999. DnaSP version 3: an integrated program for molecular population genetics and molecular evolution analysis. Bioinformatics 15:174–175.[Abstract/Free Full Text]

SAS 1989 SAS/STAT User’s guide, version 6, 4th ed. Cary, North Carolina: SAS Institute Inc.

Saupe S. 2000. Molecular genetics of heterokaryon incompatibility in filamentous ascomycetes. Microbiol. Mol Biol Rev 64:489–502.[Abstract/Free Full Text]

Schneider S, Roessli D, Excoffier L. 2000. Arlequin version 2.000: A software for population genetics data analysis. Genetics and Biometry Laboratory, University of Geneva, Switzerland.

Staats M, van Baalren P, van Kan J. 2005. Molecular phylogeny of the plant pathogenic genus Botrytis and the evolution of host specificity. Mol Biol Evol 22(2):333–46.[Abstract/Free Full Text]

Swofford DL. 1998. PAUP* Phylogenetic analysis using parsimony (*and other methods). Sinauer Associates, Sunderland, Massachusetts, Version 4.0.

Taylor JW, Geiser DM, Burt A, Kofoupanou V. 1999. The evolutionnary biology and population genetics underlying fungal strain typing. Clin Microbiol Rev 12:126–146.[Abstract/Free Full Text]

SAS Institute Inc., Jacobson DJ, Kroken S, Kasuga T, Geiser DM, Hibbett DS, Fisher MC. 2000. Phylogenetic species recognition and species concepts in fungi. Fungal Genet Biol 31:21–32.[CrossRef][Medline]

Tibayrenc M, Kjellberg F, Arnaud J, Oury B, Breniére SF, Dardé ML, Ayala F. 1991. Are eukaryotic microorganisms clonal or asexual? A population genetic vantage. Proc Natl Acad Sci USA 88:5129–5133.[Abstract/Free Full Text]

Via S. 2001. Sympatric speciation in animals: the ugly duckling grows up. Trends Ecol Evol 16:381–390.[CrossRef][Medline]

Ward TJ, Bielawski JP, Kistler HC, Sullivan E, O’Donnell K. 2002. Ancestral polymorphism and adaptative evolution in the trichothecene mycotoxin gene cluster of phytopathogenic Fusarium. Proc Natl Acad Sci USA 99:9278–9283.[Abstract/Free Full Text]

Wu CI. 1991. Inferences of species phylogeny in relation to segregation of ancient polymorphisms. Genetics 127:429–435.[Abstract]

Yoshiyama M, Tu Z, Kainoh Y, Honda H, Shono T, Kimura K. 2001. Possible horizontal transfer of a transposable element from host to parasitoid. Mol Biol Evol 18:1952–1958.[Abstract/Free Full Text]

This article has been cited by other articles:

				P. Inderbitzin, Y. R. Mehta, and M. L. Berbee Pleospora species with Stemphylium anamorphs: a four locus phylogeny resolves new lineages yet does not distinguish among species in the Pleospora herbarum clade Mycologia, May 1, 2009; 101(3): 329 - 339. [Abstract] [Full Text] [PDF]

				S. Fillinger, P. Leroux, C. Auclair, C. Barreau, C. Al Hajj, and D. Debieu Genetic Analysis of Fenhexamid-Resistant Field Isolates of the Phytopathogenic Fungus Botrytis cinerea Antimicrob. Agents Chemother., November 1, 2008; 52(11): 3933 - 3940. [Abstract] [Full Text] [PDF]

				J. Montarry, P. Cartolaro, F. Delmotte, J. Jolivet, and L. Willocquet Genetic Structure and Aggressiveness of Erysiphe necator Populations during Grapevine Powdery Mildew Epidemics Appl. Envir. Microbiol., October 15, 2008; 74(20): 6327 - 6332. [Abstract] [Full Text] [PDF]

				D. T. O'Gorman, P. L. Sholberg, S. C. Stokes, and J. Ginns DNA sequence analysis of herbarium specimens facilitates the revival of Botrytis mali, a postharvest pathogen of apple Mycologia, March 1, 2008; 100(2): 227 - 235. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fournier, E. Articles by Brygoo, Y. Search for Related Content PubMed Articles by Fournier, E. Articles by Brygoo, Y. Agricola Articles by Fournier, E. Articles by Brygoo, Y.