| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Centre de Biologie et de Gestion des Populations, Institut de Recherche pour le Développement (IRD) and Institut National de la Recherche Agronomique (INRA), Campus International de Baillarguet, CS30016, 34988 Montferrier sur lez, France
Cécile Dalleau-Clouet
Jacques Fargues
Centre de Biologie et de Gestion des Populations, Institut National de la Recherche Agronomique (INRA), Campus International de Baillarguet, CS30016, 34988 Montferrier sur Lez, France
Marie-Claude Bon
USDA-ARS-European Biological Control Laboratory, Campus International de Baillarguet, CS90013, 34988 Montferrier sur Lez, France
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONLITERATURE CITED |
|---|
The hyphomycete Paecilomyces fumosoroseus (Pfr) is a geographically widespread fungus capable of infecting various insect hosts. The fungus has been used for the biological control of several important insect pests of agriculture. However knowledge of the fungus’ genetic diversity and population structure is required for its sustainable use as a biological control agent. We investigated length and sequence polymorphisms of nine microsatellite loci for 33 Pfr accessions sampled from various host species and geographical locations, and our results reveal complex mutational processes for these molecular markers. Only Pfr isolates from Bemisia tabaci were amplified successfully, indicating the existence of Pfr genotypes specifically associated with B. tabaci. Genetic relationships among the 25 Pfr isolates from B. tabaci were inferred from allelic variability data at eight microsatellite loci that were polymorphic and subsequently from sequence data from the flanking regions of three selected loci. Maximum parsimony and neighbor joining analyses partitioned Pfr genetic diversity in two major lineages. One lineage included genotypes from the B-biotype of B. tabaci distributed across the Americas and was strongly supported in both analyses. Another lineage was distributed across Asia and consisted of four distinct clusters. Allele size homoplasy was found at the three microsatellite loci. We obtained better discrimination and resolution of the relationships among isolates with sequence data, although not all isolates could be typed. Thus sequencing of microsatellite flanking regions and repeats is a promising approach for the identification of Pfr isolates that specifically infect certain B. tabaci biotypes and phylogeographic studies.
Key words: Bemisia tabaci host influence, genetic relationships, length and sequence polymorphisms, microsatellites, mutational patterns
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONLITERATURE CITED |
|---|
, Bolckmans et al 1995). In addition it is naturally associated with the whitefly Bemisia tabaci (Homoptera: Aleyrodidae) in various geoclimatic regions (Lacey et al 1996) where this pest causes severe damage to a large variety of valuable crops (Gerling and Mayer 1996). Because of its pathogenicity to a broad spectrum of agricultural pests, Pfr was thought to be useful as a microbial agent against B. tabaci and the mass production of several Pfr isolates has been developed for inundative release for control of B. argentifolii, the B-biotype in B. tabaci. Such applications could be effective in greenhouses (Vidal et al 1998) as well as in the field (Wraight et al 2000). Thus accurate phenotypic and genotypic characterization of these microbial agents before their release is of paramount importance.
Only a few of the numerous studies of genetic diversity in entomopathogenic fungi reported in the literature concern Paecilomyces sp. (Tigano-Milani et al 1995a, b, Chew et al 1997, Cantone and Vandenberg 1998, Fargues et al 2002). Tigano-Milani et al (1995a) and Cantone and Vandenberg (1998) demonstrated that Pfr is a highly variable species exhibiting significant divergence in RAPD fingerprinting patterns but with little clear evidence of host specificity or geographic structuring. Sexual reproduction has not been observed in Pfr, but its parasexual cycle has been reproduced under laboratory conditions, although the significance of this cycle in nature remains unclear. A survey of the number of vegetative compatibility groups (VCG) in a worldwide sampling revealed high diversity, suggesting that the prospect of mitotic recombination at least between VCG seems low (Cantone and Vandenberg 1998). The same authors also demonstrated that Pfr isolates within the same VCG were not necessarily genetically related, indicating that these isolates were not clonal in origin. Although RAPD markers have proven to be valuable tools for assessing the genetic variability between large groups, inference of relationships at lower taxonomic ranks such as isolate level have limitations. Likewise the recent use of ITS-rDNA enabled Pfr isolates of B. tabaci sampled from various geoclimatic regions to be assigned to various ITS-rDNA lineages, but this genetic marker did not allow a precise typing of isolates (Fargues et al 2002).
Microsatellite markers have proven to be powerful tools for genome mapping, parentage and kinship studies and assessing the genetic diversity and structure of populations (Queller et al 1993, Goldstein and Schlötterer 1999). Their value is based on their wide distribution throughout the genome, their high level of polymorphism, their discriminating power and their codominant transmission (Li 1997). Their use continues to increase in popularity among most biological subdisciplines, including mycology (Fisher et al 2000, Enkerli et al 2001, Zhou et al 2001, Fournier et al 2002, Dettman and Taylor 2004, Dalleau-Clouet et al 2005). Although extensively used over a large range of fungal taxa, the potential of microsatellites for the exploration of species and lineage boundaries in the species complexes in fungi still requires investigation. This is particularly crucial for species lacking taxonomically informative morphological characters, such as Pfr.
Nine microsatellite loci have been developed for Pfr (Dalleau-Clouet et al 2005). However their usefulness in population genetic studies that depends on their discriminatory power has not been evaluated fully. The purpose of the present study was to assess the genetic diversity of 33 Pfr isolates sampled from various host species and geographic regions using these nine microsatellite loci. We first determined both the size and frequency distribution of alleles at all loci. We next sequenced the repeat and flanking regions of three specific microsatellite loci to reveal nucleotide substitutions that genotyping cannot detect. This more detailed analysis allows microsatellite repeat variability to be placed in the evolutionary context of less rapidly evolving flanking regions (Fisher et al 2000, Zhu et al 2000, Blankenship et al 2002).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONLITERATURE CITED |
|---|
were obtained from both INRA (Montpellier, France) and USDA-ARS-ARSEF (Ithaca, USA) entomopathogenic fungus collections (TABLE I). These accessions were collected from geographical regions, including Asia, North and South America and Europe, and from insect orders, including Homoptera, Lepidoptera, Coleoptera and Hemiptera (TABLE I).
|
. We employed nine dinucleotide microsatellite loci originally developed for Pfr isolates from B. tabaci (Dalleau-Clouet et al 2005). Their isolation procedure and GenBank accession numbers (BV211293
[GenBank]
–BV211301
[GenBank]
) were reported by Dalleau-Clouet et al (2005).
Microsatellite polymorphisms.— –
For each locus and isolate allele size polymorphisms were resolved after gel electrophoresis and radiolabeling visualization as described in Dalleau-Clouet et al (2005). Allele size was assigned by comparison with the reference cloned allele, for which the exact size was determined previously by direct sequencing. Polymorphisms among the Pfr isolates then were assessed based on the allelic diversity observed at each of the nine loci from unambiguous (e.g. one clear band) polymerase chain reaction (PCR) amplifications.
Haplotype polymorphisms were determined by sequencing all PCR products obtained at each polymorphic locus following the same PCR amplification conditions as specified for the allele-size polymorphism investigation (Dalleau-Clouet et al 2005). Each PCR reaction was performed in a total volume of 25 µL and consisted of 5 µL of genomic Pfr isolate DNA (15–18 ng), 2.5 µL of 10x QIAGEN buffer (with 1.5 mM MgCl2), 5 µL Q buffer (QIAGEN Inc.), 1.5 µL of each primer (10 µM), 1.8 µL of 2.5 mM dNTPs, 1.25 units of Taq polymerase (QIAGEN Inc.). Amplifications were run on a PTC-200 thermocycler (MJ Research). PCR product sizes were checked on a 2% agarose gel. PCR products were purified with the QIAquick PCR Purification Kit (QIAGEN Inc.). Both strands were sequenced directly with amplification primers using the Big-dye Terminator Sequencing Kit (Applied Biosystems Inc.) and an Applied Biosystems 3730 XLTM DNA analyser (GenomeExpress).
Data analyses.— –
Genetix 4.0.1 software (Belkhir et al 2000) was used to calculate descriptive statistics including the mean number of alleles per locus, the allele frequency and total genetic diversity (
). Haploid data were treated as diploid data, which introduces a small bias in values that was corrected by a factor of (2n-1)/(2n-2) where n is the number of isolates included in the analysis.
Sequence alignment for the eight polymorphic loci was performed manually. Sequences of all isolates were compared, and only those that revealed different haplotypes were included in the analysis. Nucleotide variability of the microsatellite loci with or without their repeats was described with these parameters: N (number of haplotypes), 
(nucleotidic diversity) using Proseq version 2.91 (Filatov 2002), K (number of variable sites) and P (parsimony informative characters) using PAUP 4b10 (Swofford 2001).
The incongruence length test (Farris et al 1994) also known as the partition homogeneity test in PAUP (Swofford 2001) was used to determine the significance of incongruence among the flanking-region sequences before combining data at all loci in a single dataset. Full heuristic searches were made based on 100 replicates (using the 0.05 significance level).
Relationships among isolates based on allele size data have been depicted by consensus trees constructed from Cavalli-Sforza’s and Edwards’ (1967) genetic distance data by importing allele frequency data into the PHYLIP 3.6c package (Felsenstein 1995). These consensus trees (program CONSENSE) were constructed with the neighbor joining (NJ) method with the NEIGHBOR program after 1000 bootstrap resamplings of the input gene frequency data (program SEQBOOT).
Relationships among isolates based on only the flanking-region sequences were inferred from consensus trees generated under a parsimony optimality criterion. The analysis was performed with unweighted characters with gaps stated as missing data or a new character. A heuristic search for the most parsimonious tree was performed with 100 replicates of random stepwise addition of data and branch swapping via tree-bisection-reconnection (TBR). Bootstrap values were computed by 1000 data resamplings with default PAUP settings.
Consensus trees were described with these parameters: L (length), CI (consistency index), RI (retention index) and P (number of parsimony informative characters used for the tree construction). In addition analyses were conducted using the distance-matrix method with the NJ algorithm (Saitou and Nei 1987) on Kimura 2-parameter distances with PAUP. The NJ consensus tree was constructed after a bootstrap procedure of 1000 iterations. No outgroup could be used for both MP and NJ because all Pfr isolates for which we generated data were members of the same species.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONLITERATURE CITED |
|---|
) failed to provide amplification products, regardless of the PCR conditions tested. These results indicate mutational alteration of primer binding sites among insect hosts.
Length polymorphism in microsatellite loci.— –
Nine microsatellite loci displayed different levels of allele length polymorphism. The number of alleles at these loci were 1–5 and genetic diversity () was 0–0.79, with the higher values always observed for the locus PfrBtB02 (TABLE II). Higher allelic richness and genetic diversity were not necessarily linked to loci with larger allele sizes. Some alleles were rare (f < 0.05) and detected only once, such as allele 176 bp for PfrBtB02, allele 275 bp for PfrBtD01 and allele 91 bp for PfrBtD11b (TABLE II). All three of these unique alleles were detected in the Pfr 84 isolate from India (TABLE III). In contrast allele 172 bp (PfrBtB02) was specific to all isolates from Nepal and hence could be defined as a diagnostic allele for Pfr isolates from this country. Most other alleles however were shared among several Pfr isolates with different geographic origins.
|
|
). All Pfr isolates collected from the B. tabaci B-biotype (B. argentifolii) showed the same pattern, [P6]. In several instances Pfr isolates originating from the same region exhibited different patterns (i.e. India Pfr with [P2], [P3] and [P4] and Pakistan Pfr with [P1], [P2]), while Pfr from Nepal displayed a unique pattern [P5] (TABLE III).
Sequence polymorphism in microsatellite loci.— –
Sequence polymorphism in microsatellite loci was assessed at three levels, the microsatellite repeat region, the flanking region and the whole micro-satellite locus; both the microsatellite repeat and flanking regions also are called haplotypes. Length and sequence polymorphisms were studied at three loci, PfrBtA08, PfrBtB02 and PfrBtD05 (TABLE IV).
|
, FIG. 1). Allele size did not appear to influence the total polymorphism expressed at these microsatellite loci. When considering separately PfrBtB02, PfrBtA08 and PfrBtD05 for the 25 Pfr isolates from B. tabaci, more haplotypes than allele size variants (21 vs. 13) were observed. For all loci a total of 14 haplotypes was revealed, which was greater than the total number of allele size variants (FIG. 1, TABLE III). Some haplotypes were unique, including Pfr 69, 72, 73, 74, 79, 80, 84, while others were shared by several isolates (Pfr 85, 86, 87, 88, 89). Sequence polymorphisms were due to diverse mutational processes such as point mutations and more important insertion/deletion events, not only in the flanking regions but also in the micro-satellite repeats (FIG. 1).
|
). The mutations seemed to be distributed randomly across the nonrepetitive flanking regions and in the poly-N regions, in the form of either a single-base substitution (with 67% of these being transitions) or a single-base insertion/ deletion (e.g. at sites 154 and 502). Another important indel event also was detected. At the locus PfrBtD05, 11 isolates belonging to the [P2] and [P4] allele patterns displayed a GG(A)10 repeat between poly-A regions and CA repeats (FIG. 1). The number of unique flanking sequence variants do not seem to increase with the length of the flanking-region sequence because the most polymorphic locus (PfrBtB02) was also the shortest (TABLE IV).
Microsatellite-repeat region..
The highest level of variation in microsatellite repeat regions seems to correspond to the longest microsatellite repeats and the level of variation in the corresponding flanking region (TABLE IV), as observed for PfrBtB02. In addition the PfrBtB02 microsatellite-repeat region also had the highest number of variants either because of its higher number of repeats or mutation events. The interruption of microsatellite repeats due to transversion of C 
G, as for PfrBtB02, was observed in two out of the eight loci analyzed.
Regardless the type of substitutions observed, they often are shared by various isolates. At PfrBtB02 the same imperfect repeat, (GT)n (CTGT)2 CTC (GT)n, was shared by all isolates belonging to the [P4], [P6] and [P7] patterns (FIG. 1). At locus PfrBtD01 the (CA)3 (GA) (CA)n repeat was shared by all isolates belonging to the [P1], [P3] and [P5] allele pattern (data not shown). The duplication of repeats with substitutions (imperfect repeats [(GTAT)2–3 (GT)11–15]) within the same microsatellite allele (PfrBtA08) was observed in all Pfr isolates (FIG. 1). The same nonrepetitive flanking region was observed to be associated with multiple repeat numbers. At PfrBtD05 the same nonrepetitive flanking region was found in alleles 199 bp [P2] and 197 bp [P4] that displayed different repeat numbers (FIG. 1). At PfrBtD01 (data not shown), 11 and 12 dinucleotide repeats of CA were flanked by the same conserved region. Some repeats also were observed to share different flanking-sequence variants. At PfrBtA08 (FIG. 1) the repeat (GTAT)3 (GT)15 was present in alleles 256 bp [P6, P7] and 259 bp [P2, P5]. Although identical in state it is unlikely that all alleles at the same locus were identical by descent, especially when shared among isolates from different geographic regions. Nevertheless in Pfr the allele size polymorphisms essentially were linked with variability in repeat unit number (single or more units), instead of mutations in the flanking regions (TABLE IV). Indeed it was only through the complete sequencing of alleles that we detected higher levels of polymorphism than would have been possible by genotyping three loci (14 haplotypes vs. seven allele size patterns).
Genetic relationships..
NJ consensus tree based on Cavalli-Sforza’s and Edwards’ genetic distances after 1000 bootstrap resamplings of the allele data showed that the clustering of isolates in specific allele patterns [P] was strongly supported (bootstrap values >93) (FIG. 2a). [P7] consisting of the Cuban isolate clustered only with [P6], which corresponds to the American Pfr isolates sampled from the B-biotype of B. tabaci. Most clusters (or allele patterns) are made up of accessions from the same geographic region (e.g. [P4] and [P5]), although accessions from India and Pakistan are found within several clusters. Bootstrap values for all clusters are generally high (81–100%), however the relationship among clusters is not strongly supported (
55%).
|
= 0.99 > 0.05) allowed the joint analysis of the PfrBtB02, A08 and D05 sequence data in the MP and NJ approaches. This analysis detected 14 haplotypes of 529 bp that displayed high sequence similarities. The percentage of shared nucleotides was above 99% within each allele pattern and 92.2–99.8% between patterns, with the highest percentage of shared nucleotides observed between isolates from America and Cuba and between the isolates from Pakistan in [P1] and the Indian isolate Pfr84. The 529-bp microsatellite flanking-region sequences amplified from the 25 Pfr isolates showed a low level of polymorphism (4.7%). Sixteen of the 25 variable sites were parsimony informative (64%).
Both MP and NJ analyses revealed similar relationships with lower support for the main branches of the MP analysis (FIG. 2b). All unrooted analyses revealed a separation of the Pfr isolates into two distinct lineages. Within each lineage the isolates clustered into subgroups equivalent to the allele patterns previously described (TABLE IV). The first strongly supported lineages included the American and Cuban isolates that were assigned to the B-biotype of B. tabaci. Only the B-biotype has been reported from Cuba. This lineage was represented by four different haplotypes (FIG. 1). Pfr isolates from America and Cuba shared numerous synapomorphies (e.g. sites 68, 102, 104, 105, 161, 329 and 413) and a high percentage of nucleotide similarity, which explain their high bootstrap value (99%). The second lineage included all isolates from Asia. Relationships among the Asian isolates are poorly resolved. The Indian isolates although collected from the same geographic region occur in three distinct locations in the MP and NJ trees. One of these locations (corresponding to allele pattern [P4]) was always the closest to that of the American-Cuban isolates. Isolates from Pakistan always occur in two distinct locations in these trees. In the NJ tree isolates from Nepal form a distinct cluster [P5], while in the MP tree the relationship among these isolates is less well resolved.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONLITERATURE CITED |
|---|
) and is used as a biological control agent of agricultural pests. Isolate genotyping and robust inference into population genetic structure of this fungus are required before it can be used against pests such as B. tabaci. Therefore we investigated allelic and sequence variability of nine microsatellite loci in 33 isolates of Pfr, which are representative of the global diversity of the species.
Genetic variability of the microsatellite markers.— –
Eight out of the nine microsatellite loci tested in the 25 Pfr isolates from B. tabaci were polymorphic with an average of 3.5 alleles per locus. The average number of alleles per locus in Pfr is slightly lower than the value reported in most previous studies on entomopathogenic fungi (Rehner and Buckley 2003; Enkerli et al 2001, 2005). PfrBtB02, which was the most variable locus, also had the highest range of repeat numbers and flanking-region variants. Our findings are consistent with other studies that have reported high variability among microsatellite loci with complex repeat motifs (Enkerli et al 2001, Dettman and Taylor 2004, Jany et al 2006). Previous studies also have reported higher allelic diversity in loci with a larger number of repeats (Schlötterer and Tautz 1992, Goldstein and Schlötterer 1999, Shinde et al 2003, Dettman and Taylor 2004). Two factors contribute to such loci having greater mutation rates and genetic diversity: (i) During replication longer stretches of repeats are more prone to strand-slippage mispairing and provide more opportunities for misalignment during re-annealing of the nascent strand (Dettman and Taylor 2004); and (ii) the presence of single-step base substitutions in repeat regions results in an interruption in the synthesis of an exact repeat. Two loci in Pfr presented such interrupted micro-satellite repeats due to transversions, which are shared by several isolates. In the context of micro-satellite evolution, mutations that interrupt stretches of exact repeats are important because they should decrease the frequency of slippage and thus lead to reduced polymorphisms (Richards and Sutherland 1994, Jin et al 1996, Goldstein and Schlötterer 1999). Much of the variation at the Pfr loci we analyzed occurred in the flanking regions as a single-base substitution (mainly transitions) or single-base insertions/deletions. At PfrBtD05 we also detected indels such as repeats GG(A)10 followed by a stretch of As. Although increased variability of nonrepetitive DNA near microsatellite repeats has been reported (Brohede and Ellegren 1999, Dettman and Taylor 2004) base substitutions in Pfr seem to be distributed randomly. It is noteworthy that the same flanking-sequence variants were associated with different alleles, or repeat numbers, and that similar alleles, or repeat numbers, were shared by divergent flanking-sequence variants. This suggests size homoplasy in the Pfr loci examined in this study. Homoplasy has been reported to occur within populations of simple sequence repeat loci (Blankenship et al 2002, Estoup et al 2002) and could lead to an underestimation of divergence. Although apparently identical in state it was unlikely that all copies of the same repeat were identical by descent, especially when shared among well differentiated genealogical lineages (i.e. the repeat [GTAT]3 [GT]15/locus PfrBtA08 shared by distantly related isolates from the [P2, P5] and [P6, P7] allele patterns). Size homoplasy was found in one or two alleles per homoplasic locus. The average number of alleles per homoplasic locus was 2.5 in PfrBtA08 and 3 in PfrBtB02 and D05. These values are high compared with previous studies of infraspecific size homoplasy of microsatellite loci (Blankenship et al 2002, Estoup et al 2002). Many low frequency alleles however might have inflated the level of homoplasy in the current study.
Advantages of the sequencing approach.— –
Microsatellite polymorphisms in populations typically are assessed based on variation in allele size alone and do not include the detection of base substitutions within the repeat array or in the flanking regions through direct sequencing. However variation in microsatellite flanking regions is commonly found, even in a single species (Dettman and Taylor 2004). Substitutions in the flanking regions even can lead to amplification failure as exemplified in the eight Pfr isolates for which no amplification occurred. For Pfr 50% of the total allelic variation would have gone undetected if the sequencing approach had not been used (e.g. seven allele patterns were identified across the eight polymorphic loci, but 14 haplotypes were detected based on allele size and sequencing at only three loci [PfrBtB02-A08-D05]). This approach unfortunately did not resolve relationships among Pfr isolates (FIG. 2). Indeed based on the allele size data only the clustering based on allele patterns was supported strongly (FIG. 2a). With a sequence-based study of microsatellite markers, relationships among microsatellite alleles can be depicted independently of their repeat number through the phylogenetic analysis of molecular variation observed in flanking regions. Microsatellite repeats and their associated flanking regions might share the same evolutionary history (and lines of descent) because of physical linkage. While population genetic inference can benefit from the sequencing approach, doing so requires a larger amount of time and effort even for a moderate number of populations and markers. A more reasonable approach would involve a preliminary assessment, before initiating larger scale studies, of the levels of variation both in the flanking regions and microsatellite repeats. Thus microsatellite loci can be used to infer species-level relationships but only when using the combined approach.
New perspectives.— –
These microsatellite markers were shown to be highly discriminant and valuable in reconstructing an intraspecific phylogeny of Pfr. Using the rDNA-ITS RFLP approach Fargues et al (2002) found that all Pfr isolates obtained from B. tabaci formed a single lineage. From this single lineage seven clusters based on specific allele pattern and 14 haplotypes were resolved in our study. Although the typing of all isolates was not achieved with these loci some Pfr isolates were characterized by a unique haplotype and a diagnostic allele was detected in the isolates from Nepal. The development of microsatellite loci, such as those described here, will allow the fate of specific Pfr isolates to be monitored in the environment. This information can be used to assess the survival of isolates in specific habitats and hosts and would allow changes in isolates to be evaluated during co-infections.
Biodiversity and relationships in the Pfr species.— –
Bemisia-host affiliation The microsatellite loci used in this study led to successful amplifications in only the 25 Pfr isolates obtained from B. tabaci. This indicates that mutations in the microsatellite priming site of Pfr isolates originating from other insect hosts are extensive enough to preclude amplification. The degree to which priming sites are conserved appears to be related to taxonomic divergence among insect host-species. As in Beauveria brongniartii (Enkerli et al 2001) the utility of microsatellite markers among Pfr isolates was low, raising the possibility of cryptic species within the Pfr complex. Pfr isolates that specifically are associated with B. tabaci clearly appear to be genetically divergent from other isolates. Studies using DNA-based genetic markers have not revealed such insect-host influence on Pfr species (Tigano-Milani et al 1995, Cantone and Vandenberg 1998).
B. tabaci is a species complex that exhibits considerable biological and genetic diversity among natural populations (Brown et al 1995, Fröhlich et al 1999, Berry et al 2004, De Barro et al 2005). This has led to the subdivision of this species into a series of biotypes (more than 20 have been described so far) based on their ability to utilize different host plants and to transmit begomovirus, inducing physiological changes in some hosts (Perring 2001). A broad genetic survey of B. tabaci has suggested that its populations were genetically variable with respect to host plant (Brown et al 1995) and geographic location (Berry et al 2004, De Barro et al 2005). Taking into account the biodiversity of B. tabaci and especially the B-biotype the detection of genetically differentiated Pfr isolates was not surprising. In addition to the role of host affiliation in Pfr genetic diversity, geographic structuring of diversity also was observed. Intraspecific phylogeny of Pfr isolates, based on the sequencing of the flanking region, provides evidence for two well supported lineages: American-Cuban and Asian, which consists of four distinct populations groups. These lineages were not reported by Cantone and Vandenberg (1998) using RAPDs. The position of the [P4] allele pattern as sister of the American-Cuban isolates would suggest a putative origin of the American-Cuban lineage from certain isolates found in India. Isolates from India were found to be the most diverse with three allele patterns detected among seven isolates, and they occurred in three clusters (FIG. 2a). These results are consistent with the findings of Tigano-Milani et al (1995b), which indicate the putative origin of B. tabaci in India. The Indian subcontinent is postulated to be center of the origin of B. tabaci based on the species’ genetic variability and parasitoid abundance and diversity in this region (Mound et al 1978, Brown and Idris 2005). Comparable patterns of evolution among entomopthogenic fungi and host insects have been reported (B. bassiana, Maurer et al 1997, Couteaudier et al 1998; Metarhizium in tropical and subtropical regions, Bidochka and Small 2005). Our results provide a phylogeographic perspective on the worldwide population structure of Pfr capable of infecting B. tabaci. Given the moderate sampling size in this study and our lack of knowledge concerning the biology and reproductive behavior of these isolates, it would be premature to consider the population clusters we have identified as different species. Thus we think it is more accurate to consider these clusters as genetically divergent populations. However the range of genetic diversity reported here among Pfr isolates infecting whiteflies in Asia or in America might prove to be valuable in biological control efforts (e.g. combinations of genetically diverse Pfr isolates might be required to provide broad control during B. tabaci outbreaks). This recommendation however requires further testing.
|
CONCLUSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONLITERATURE CITED |
|---|
). On the other hand development of a complementary approach, such as SNP, might be the next step for typing Pfr isolates. Finally for biological control and to comply with biopesticide registration requirements, the potential for genetic exchange within and between VCG also should be investigated.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Corresponding author. E-mail: nathalie.gauthier{at}supagro.inra.fr
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONLITERATURE CITED |
|---|
Berry SD, Fondong VN, Rey C, Rogan D, Fauquet CM, Brown J. 2004. Molecular evidence for five distinct Bemisia tabaci (Homoptera: Aleyrodidae) geographic haplotypes associated with cassava plants in sub-Saharan Africa. Ann Entomol Soc Am 97:852–859.[CrossRef]
Bidochka MJ, Small CL. 2005. Phylogeography of Metarhizium, an insect pathogenic fungus. In: Vega F, Blackwell M, eds. Insect-Fungal Associations: ecology and evolution. New York: Oxford University Press. p 28–50.
Blankenship SM, May B, Hedgecock D. 2002. Evolution of a perfect simple sequence repeat locus in the context of its flanking sequence. Mol Biol Evol 19:1943–1951.
Bolckmans K, Sterk G, Eyal J, Sels B, Stepman W. 1995. PreFeRal (Paecilomyces fumosoroseus (Wize) Brown & Smith, strain Apopka 97), a new microbial insecticide for the biological control of whiteflies in greenhouses. Mededelingen van de Faculteit Landbouwetenschappen Univ Gent 60:719–724.
Brohede J, Ellegren H. 1999. Microsatellite evolution: polarity of substitutions within repeats and neutrality of flanking sequences. Proc R Soc London B 266:825–833.[Medline]
Brown JK, Fröhlich DR, Rosell RC. 1995. The sweet potato whitefly or silverleaf whiteflies, biotypes of Bemisia tabaci or a new species complex. Ann Rev Entomol 40: 511–534.[CrossRef]
———, Idris AM. 2005. Genetic differentiation of whitefly Bemisia tabaci mitochondrial cytochrome oxydase I, and phylogeographical concordance with the coat protein of the plant virus genus Begomovirus. Ann Entomol Soc Am 98:827–837.[CrossRef]
Cantone FA, Vandenberg JD. 1998. Intraspecific diversity in Paecilomyces fumosoroseus. Mycol Res 102:209–215.[CrossRef]
Cavalli-Sforza LL, Edwards AWF. 1967. Phylogenetic analysis: models and estimation procedures. Am J Hum Genet 19:233–257.[Medline]
Chew JSK, Strongmann DB, McKay RM. 1997. RFLP analysis of rDNA intergenic spacer region of 23 isolates of the entomopathogen Paecilomyces farinosus. Can J Bot 75: 2038–2044.
Couteaudier Y, Viaud M, Neuvéglise C. 1998. Combination of different independent molecular markers to understand the genetic structure of Beauveria populations. In: Bridge PD, Couteaudier Y, Clarkson J, eds. Molecular Variability of Fungal Pathogens. Walingford, UK: CAB International. p 95–104.
Dalleau-Clouet C, Gauthier N, Risterucci A, Bon MC, Fargues J. 2005. Isolation and characterization of microsatellite loci in the entomopathogenic fungus Paecilomyces fumosoroseus. Mol Ecol Notes 5:496–498.[CrossRef]
De Barro PJ, Trueman JWH, Fröhlich DR. 2005. Bemisia argentifolii is a population of B. tabaci, the molecular differentiation of B. tabaci populations around the world. Bull Entomol Res 95:193–203.[CrossRef][Medline]
Dettman JR, Taylor JW. 2004. Mutation and evolution of microsatellite loci in Neurospora. Genetics 168:1231–1248.
Enkerli J, Widmer F, Gessler C, Keller S. 2001. Strain specific microsatellite markers in the entomopathogenic fungus Beauveria brongniartii. Mycol Res 102: 250–256.[CrossRef]
———, Kolliker R, Keller S, Widmer F. 2005. Isolation and characterization of microsatellite markers from the entomopathogenic fungus Metarhizium anisopliae. Mol Ecol Notes 5:384–386.[CrossRef]
Estoup A, Jarne P, Cornuet JM. 2002. Homoplasy and mutation model at microsatellite loci and consequences for population genetics analysis. Mol Ecol 11: 1591–1604.[CrossRef][Medline]
Fargues J, Bon M-C, Manguin S, Couteaudier Y. 2002. Genetic variability among Paecilomyces fumosoroseus isolates from various geographical and host insect origins based on the rDNA-ITS regions. Mycol Res 106:1066–1074.[CrossRef]
Farris JS, Källersjö M, Kluge AG, Bult C. 1994. Testing significance of congruence. Cladistics 10:315–319.[CrossRef]
Felsenstein J. 1995. Phylip (Phylogeny Inference Package). Version 3.6. Seattle: University of Washington.
Filatov DA. 2002. Proseq: a software for preparation and evolutionary analysis of DNA sequence data sets. Mol Ecol Notes 2:621–624.[CrossRef]
Fisher MC, Koenig G, White TJ, 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.
Fournier E, Giraud T, Loiseau A, Vautrin D, Estoup A, Solignac M, Cornuet JM, Brygoo Y. 2002. Characterization of nine polymorphic microsatellite loci in he fungus Botrytis cinerea (Ascomycota). Mol Ecol Notes 2: 253–255.[CrossRef]
Fröhlich DR, Torres-Jerez I, Bedford ID, Markham PG, Brown JK. 1999. A phylogenetic analysis of the Bemisia tabaci species complex based on mitochondrial DNA markers. Mol Ecol 8:1593–1602.[CrossRef][Medline]
Gerling D, Mayer RJ. 1996. Bemisia: taxonomy, damage, control and management. Andover, UK: Hants, Intercept Ltd.
Goldstein DB, Schlötterer C. 1999. Microsatellites: evolution and applications Oxford University Press.
Jany J-L, Bousquet J, Gagné A, Khasa DP. 2006. Simple sequence repeat (SSR) markers in the ectomycorrhizal fungus Laccaria bicolor for environmental monitoring of introduced strains and molecular ecology applications. Mycol Res 110:51–59.[CrossRef][Medline]
Jin L, Macaubas C, Hallmayer J, Kimura A, Mignot E. 1996. Mutation rate varies among alleles at a microsatellite locus: phylogenetic evidence. Proc Nat Acad Sc USA 93: 15285–15288.
Lacey LA, Fransen JJ, Carruthers RI. 1996. Global distribution of naturally occurring fungi of Bemisia, their biology and use as biological control agents. In: Gerling D, Mayer R, eds. Bemisia 1995: taxonomy, biology, damage, control and management. Andover, U.K: Hants, Intercept. p 401–433.
Li WH. 1997. Molecular evolution. Sunderland, Massachusetts: Sinauer Associates.
Maurer P, Couteaudier Y, Girard PA, Bridge PB, Riba G. 1997. Genetic diversity of Beauveria bassiana and relatedness to host insect range. Mycol Res 101:159–164.[CrossRef]
Mound LA, Halsey HA. 1978. Whitefly of the world: a systematic catalogue of the Aleyrodidae (Homoptera) with host plant and natural enemy data. British Museum (Natural History), London. Chichester, UK: John Wiley & Sons.
Osborne LS, Landa Z. 1992. Biological control of whiteflies with entomophatogenic fungi. Florida Entomol 75: 456–471.[CrossRef]
Perring TM. 2001. The Bemisia tabaci species complex. Crop Protect 20:725–737.[CrossRef]
Queller DC, Strassman JE, Hughes CR. 1993. Microsatellites and kinship. Trend Ecol Evol 8:285–288.[CrossRef]
Richards RI, Sutherland GR. 1994. Simple repeat DNA is not replicated simply. Nat Genet 6:114–116.[CrossRef][Medline]
Saitou N, Nei M. 1987. The neighbor joining method: a new method for reconstructing phylogenic trees. Mol Biol Evol 4:406–425.[Abstract]
Schlötterer C, Tautz D. 1992. Slippage synthesis of simple sequence DNA. Nucleic Acid Res 20:211–215.
Shinde D, Lai YL, Sun FZ, Arnheim N. 2003. Taq DNA polymerase slippage mutation rates measured by PCR and quasi-likelihood analysis: (CA/GT)n and (A/T)n microsatellites. Nucleic Acid Res 31:974–980.
Swofford DL. 2001. PAUP*: phylogenetic analysis using parsimony (*and other methods). Sunderland, Massachusetts: Sinauer Associates.
Tigano-Milani MS, Honeycutt RJ, Lacey LA, Assis R, McClelland M, Sobral BW. 1995a. Genetic variability of Paecilomyces fumosoroseus isolates revealed by molecular markers. J Invert Pathol 65:274–282.[CrossRef][Medline]
———, Carneiro RG, Faria MR, Frazao HS, McCoy CW. 1995b. Isozyme characterization and pathogenicity of Paecilomyces fumosoroseus and P. lilacinus to Diabrotica speciosa (Coleoptera: Chrysomelidae) and Meloidogyne javanica (Nematoda: Tylenchidae). Biol Control 5: 378–382.[CrossRef]
Vidal C, Osborne LS, Lacey LA, Fargues J. 1998. Effect of host plant on the potential of Paecilomyces fumosoroseus (Deuteromycotina: Hyphomycetes) for controlling the silverleaf whitefly, Bemisia argentifolii (Homoptera: Aleyrodidae), in greenhouses. Biol Control 12:191–199.[CrossRef]
Wraight SP, Carruthers RI, Jaronski ST, Bradley CA, Garza CJ, Galaini-Wraight S. 2000. Evaluation of the entomopathogenic fungi Beauveria bassiana and Paecilomyces fumosoroseus for microbial control of the silverleaf whitefly, Bemisia argentifolii. Biol Control 17:203–217.[CrossRef]
Zhou Z, Miwa M, Hogetsu T. 2001. Polymorphism of simple sequence repeats gene flow within and between ectomycorrhizalo Suillus grevillei populations. New Phytol 149:339–348.[CrossRef]
Zhu Y, Queller DC, Strassman JE. 2000. A phylogenic perspective on sequence evolution in microsatellite loci. J Mol Evol 50:324–338.[Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
