Journal of Invertebrate Pathology

Transcription

1 Journal of Invertebrate Pathology 109 (2012) Contents lists available at SciVerse ScienceDirect Journal of Invertebrate Pathology journal homepage: www. elsevier. com/ locate/ jip Diversity of Beauveria spp. isolates from pollen beetles Meligethes aeneus in Switzerland Nicolai V. Meyling a,, Christina Pilz b, Siegfried Keller b, Franco Widmer b, Jürg Enkerli b a Department of Agriculture and Ecology, Faculty of Life Sciences, University of Copenhagen, Denmark b Molecular Ecology, Agroscope Reckenholz-Tänikon Research Station ART, Reckenholzstrasse 191, Zurich, Switzerland a r t i c l e i n f o a b s t r a c t Article history: Received 13 May 2011 Accepted 3 October 2011 Available online 10 October 2011 Keywords: Entomopathogenic fungi Phylogeny Microsatellite markers Ecological host range Genetic diversity Pollen beetles Meligethes aeneus were collected in oilseed rape fields at different sites in Switzerland in spring and 32 isolates of the fungal genus Beauveria occurring as latent infections in the beetles were obtained and molecularly characterized. Three major clades, Beauveria bassiana sensu stricto (Clade A: n = 13), Beauveria brongniartii (Clade B: n = 1) and Beauveria Clade C (n = 18) were identified among the isolates based on sequences of the ITS region and the 5 0 end of EF1-a. B. bassiana s.s. was further separated in the two clades, Eu_1 (n = 10) and Eu_4 (n = 3). The intergenic region Bloc provided best resolution of the individual clades B. bassiana s.s. Eu_1, Eu_4 and B. brongniartii. No specific clade of Beauveria appeared to be associated with adult M. aeneus populations. However, data suggested high relative abundance of Beauveria Clade C among the fungal entomopathogens infecting M. aeneus. Characterization of the isolates by simple sequence repeats (SSR) revealed further genotypic diversity within the clades except B. bassiana s.s. Eu_4 which appeared to be clonal. However, the individual SSR markers were differentially amplifiable from isolates of the different clades. It is therefore important to identify the underlying phylogenetic affinity of Beauveria isolates to interpret results based on SSR markers. The data suggest that not all available SSR markers are suitable for reliable characterization of diversity within Beauveria Clade C. Ó 2011 Elsevier Inc. All rights reserved. 1. Introduction The pollen beetle Meligethes aeneus F. (Coleoptera: Nitidulidae) is a major pest in oilseed rape cultures and other cruciferous crops in Central and Northern Europe (Nilsson, 1987; Alford et al., 2003; Warner et al., 2008). Adult M. aeneus beetles hibernate in hedgerows and forested areas and migrate to oilseed rape fields in the following spring (e.g., Alford et al., 2003). There they feed on flower buds, flowers as well as pollen and oviposit into flower buds causing a severe reduction in seed production (e.g. Alford et al., 2003). M. aeneus is attacked by different natural enemies, including predators (e.g., Büchs and Alford, 2003; Warner et al., 2008), parasitoids (e.g. Nilsson, 2003) and microsporidian parasites (Hokkanen et al., 1988; Lipa and Hokkanen, 12). Fungal entomopathogens are known to infect a wide range of insect species from many different orders and are readily isolated from soils in various ecosystems (e.g., Roy et al., 2010). However, little is known about fungal entomopathogens infecting M. aeneus under natural conditions and only few field surveys have been Corresponding author. Address: Department of Agriculture and Ecology, Faculty of Life Sciences, University of Copenhagen, Thorvaldsensvej 40, DK-1871 Frederiksberg C, Denmark. Fax: address: nvm@life.ku.dk (N.V. Meyling). performed to elucidate this interaction. In Finland a study has revealed no natural infections by fungal entomopathogens (Hokkanen et al., 1988, 2003), while in Switzerland 1.8% of 2139 M. aeneus adults collected in oilseed rape fields in different regions were infected by fungi (Pilz, 2005; Pilz and Keller, 2006) identified as Beauveria spp. (Ascomycota: Hypocreales). Treatments of oilseed rape field soil with Beauveria bassiana as a biological control agent against M. aeneus have resulted in a 50% decrease in winter survival of adult beetles compared to non-treated controls (Hokkanen, 13). However, the mechanisms leading to the observed effect of the fungal application have not become clear as no epizootics have been observed in M. aeneus in the treated fields (Hokkanen, 13). Traditional identification of species of the genus Beauveria is principally based on conidial morphology. However, molecular phylogenies have revealed that the genus includes cryptic species (Rehner and Buckley, 2005). The globally distributed Beauveria bassiana sensu lato (s.l.) (Balsamo) Vuillemin morphospecies is not monophyletic and consists of members placed in separate clades which have been proposed to be considered separate species (Rehner and Buckley, 2005; Rehner et al., 2006; Ghikas et al., 2010; S.A. Rehner pers. comm.). One clade, Clade A, corresponds to B. bassiana sensu stricto (s.s.) and is the sister clade of Clade B which comprises Beauveria brongniartii (Saccardo) Petch (Rehner and Buckley, 2005). A third and distantly related clade, Clade C, /$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi: /j.jip

2 N.V. Meyling et al. / Journal of Invertebrate Pathology 109 (2012) includes members being morphologically indistinguishable from those of Clade A (Rehner and Buckley, 2005). Identification of Beauveria isolates of Clade C is therefore only possible with the use of molecular markers. It has been recently shown that Clade C occurred at relatively low frequencies in a Beavueria community within a hedgerow habitat in Denmark (Meyling et al., 2009). Clade A furthermore consists of an assemblage of cryptic species for which an ad hoc identification system has been established designating phylogenetic species by continent and in the order of their discovery (e.g. Eu_1 and Eu_2 for the first two European phylogenetic species discovered) using terminology from Rehner et al. (2006) and Meyling et al. (2009). Phylogenetic assignment of Beauveria isolates to specific clades can be done by sequencing the genomic DNA regions of the internal transcribed spacer (ITS) region of the ribosomal RNA (rrna) gene cluster (White et al., 10), the 5 0 end of Elongation Factor 1-alpha (EF1-a) as described by Bischoff et al. (2006, 2009) and the intergenic Bloc region (Rehner et al., 2006). Using a global collection of Beauveria spp. Rehner and Buckley (2005) have shown that sequencing the entire EF1-a region provided more phylogenetic resolution than the ITS region. Sequencing both EF1-a and the Bloc region of Beauveria isolates from a single community in an agroecosystem in Denmark has resolved isolates among Clade A, B and C as well as identified five phylogenetic species within Clade A isolates (Meyling et al., 2009). However, the phylogenetic assignment of Beauveria isolates from a single host species has not yet been done using these markers. Further molecular characterization of isolates has become possible by using simple sequence repeat (SSR) markers developed for Clade A (Rehner and Buckley, 2003; Meyling et al., 2009) and Clade B (Enkerli et al., 2001). SSR markers have been used to assess genetic diversity of Beauveria isolates from single hosts (McGuire et al., 2005; Castrillo et al., 2008, 2010) but the phylogenetic affiliations of the isolates have not been determined in these studies. For example, it is not known whether the SSR markers may provide useful markers for Beauveria Clade C isolates. Few Beauveria isolates collected from M. aeneus have morphologically been typed as B. brongniartii (Pilz, 2005; Pilz and Keller, 2006). In Europe the ecological host range of B. brongniartii is assumed to be narrow because it has mainly been isolated from the field cockchafer Melolontha melolontha (L.) and forest cockchafer Melolontha hippocastani Fabricius (Coleoptera: Melolonthidae) and only occasionally from other host insects, mostly within Coleoptera (Vestergaard et al., 2003; Zimmermann, 2007). In contrast, Beauveria Clades A and C have not been found to be associated with particular host taxa (Rehner and Buckley, 2005; Ghikas et al., 2010). The aim of this study was to use molecular tools to characterize and identify selected Beauveria spp. isolates collected from M. aeneus in the previous study conducted in Switzerland (Pilz, 2005; Pilz and Keller, 2006) and assess the genetic diversity of Beauveria infections found in M. aeneus. For this purpose, phylogenetic assignment of the isolates to Clades A, B or C were done by sequencing the ITS and the 5 0 end of EF1-a. Furthermore, the intergenic Bloc region was used to sort identified Clade A isolates to phylogenetic species and as an attempt to provide further resolution to Clade B isolates. Finally SSR markers developed for either Clade A or B were applied to evaluate their affinities to the respective Beauveria clades identified and to test whether they provided additional resolution of the diversity of Beauveria spp. collected from M. aeneus. 2. Materials and methods 2.1. Fungal isolates and DNA extraction The Beauveria spp. collection consisted of 32 isolates obtained from M. aeneus beetles collected in different regions in Switzerland in spring of 2004 and 2005 (Pilz, 2005; Pilz and Keller, 2006) and four isolates obtained from other host insects collected in Switzerland, i.e., three isolates from M. melolontha morphologically identified as B. brongniartii and one isolate from the bark beetle Ips typographus (L.) (Coleoptera: Curculionidae) identified as B. bassiana (see Supplement Table S1). Fungal isolates were grown and maintained on complete medium (Riba and Ravelojoana, 1984) agar plates at 22 C in the dark. Mycelium for DNA extraction was produced by inoculation of 80 ml liquid CM with conidia collected from solid media plates and grown for 2 days at 20 C at 120 rpm in the dark. Mycelium was harvested by filtration as described by Enkerli et al. (2001) and lyophilized. Genomic DNA was extracted using the DNeasy Plant Mini kit (Qiagen, Hilden, Germany). Extracted DNA was quantified after gel electrophoresis using a GelDoc XRS (Bio-Rad Laboratories, Hercules, CA) gel imaging system with the Quantity One analysis software (Bio-Rad Laboratories) and the High Mass DNA ladder (Promega, Madison, WI) as standard Sequence analysis For PCR amplification of the ITS1 5.8S ITS2 region of the rrna gene cluster (White et al., 10), the 5 0 end of Elongation Factor 1-alpha (Bischoff et al., 2006, 2009) and the intergenic Bloc region (Rehner et al., 2006) PCR primer pairs ITS5 (5 0 -GGAAGTA AAAGTCGTAACAAGG-3 0 )/ITS4 (5 0 -TCCTCCGCTTATTGATATGC-3 0 ), EF2F (5 0 -GGAGGACAAGACTCACATCAACG-3 0 )/1567R (5 0 -ACHGTRCC RATACCACCSATCTT-3 0 ), and B22U (5 0 -GTCGCAGCCAGAGCAAC T-3 0 )/B822L (5 0 -AGATTCGCAACGTCAACTT-3 0 ), were used, respectively. The three regions were amplified from all isolates of the collection except that the Bloc region could not be amplified from isolates belonging to Beauveria Clade C using these primers. PCR targeting the ITS region were performed in reaction volumes of 25ll containing 1 Dynazyme buffer (Finnzymes, Espoo, Finland), 200 lm dntp, 0.5 lm of each primer, 2U Dynazyme II polymerase and 10 ng genomic DNA. Cycling conditions consisted of 3 min initial denaturation at 94 C, 35 cycles of 1 min at 94 C, 1 min at 55 C, and 1.5 min at 72 C followed by final extension of 7 min at 72 C. PCR targeting the 5 0 end of EF1-a and Bloc regions were performed in reaction volumes of 50ll consisting of 1 Phusion HF buffer (Finnzymes, Espoo, Finland), 200 lm dntp, 1lM of each primer, 2U proofreading Phusion polymerase (Finnzymes) and 10 ng genomic DNA. Cycling conditions for amplification of 5 0 end of EF1-a and Bloc regions consisted of a touch-down protocol with 30 s initial denaturation at 98 C followed by 10 cycles of 10 s at 98 C, 30 s at C (reducing annealing temperature by 1 C per cycle), and 30 s at 72 C. Subsequently, 35 cycles were performed with the same conditions, however, with a fixed annealing temperature of 60 C and a final extension of 10 min at 72 C. PCR products were purified using the GFX PCR DNA and Gel band purification kit (GE Healthcare, UK) according to the manufacturer s instructions. PCR products of the ITS and Bloc region were sequenced using the primers mentioned above, whereas PCR products of the 5 0 end of EF1-a were sequenced using primers EF2F (described above) and EFjR (5 0 -TGYTCNCGRGTYTGNCCRTCY TT-3 0 ). Sequencing was performed by MWG (Ebersberg, Germany). Sequence alignments were generated for all three regions including the generated sequences as well as published reference sequences retrieved from GenBank (see Supplement Table S2) using ClustalW and Sequence Alignment Editor in BioEdit ver (Hall, 19). Alignments were analyzed separately for each region using maximum parsimony in MEGA 4.1 (Tamura et al., 2007) by excluding gaps and using the Close-Neighbor-Interchange algorithm with search level 1 in which the initial trees were obtained with the random addition of sequences (10 replicates) and

3 78 N.V. Meyling et al. / Journal of Invertebrate Pathology 109 (2012) resampled with 1000 bootstrap resamplings. Sequence difference matrices were calculated for each region with BioEdit. Accession numbers of sequences deposited in GenBank are listed in Supplement Table S SSR marker analysis SSR marker analysis was based on 15 markers, i.e., Ba06, Ba08, Ba12, Ba13, Ba15, Ba18, Ba21, Ba26, and Ba27 isolated from B. bassiana s.s. (Rehner and Buckley, 2003; Meyling et al., 2009) and Bb1F4, Bb2A3, Bb2F8, Bb4H9, Bb5F4, and Bb8D6 isolated from B. brongniartii (Enkerli et al., 2001; Supplement Table S3). PCR amplifications were performed in 20ll reaction volumes containing 10 ng genomic DNA, 1 PCR Buffer, 1.5 mm MgCl 2, 0.6 mg/ml bovine serum albumin, 200 lm dntp, 0.2 lm forward (FAM, HEX, or NED labelled) and reverse primer, and 0.5 U HotStar Taq DNA polymerase (Qiagen). Cycling conditions consisted of 15 min of initial denaturation and enzyme activation at 95 C followed by 35 cycles of 30 s at 95 C, 30 s at 56 C (B. bassiana primers) or 58 C (B. brongniartii primers), and 40 s at 72 C, followed by a final extension of 7 min at 72 C. PCR product sizes (SSR allele sizes) were determined on an ABI Prism 3130xl genetic analyzer equipped with 36-cm capillaries and POP-7 polymer (Applied Biosystems, Foster City, CA). Fragment sizes were analyzed using Gen- Marker v1.5 (SoftGenetics LLC, State College, PA) software and GeneScan ROX400 (Applied Biosystems) as internal size standard JE ARSEF 1628 KVL KVL KVL ARSEF Gl Gl Gä Ne Nh ARSEF Wi Gä 2575 ARSEF Nh B. bass siana s.s. E u_1 B. bas ssiana a s.s. Eu_ Nw Lq 2609 Lq 2611 Lq 2613 Lq 2614 Lq 2616 Lz ARSEF 1969 B. amorpha B. bro ongniar artii Clade C Beau uveria 5 changes Fig. 1. One representative most parsimonious tree (total trees = 568; length = 110; consistency index = ; retention index = ; composite index = ) of the 5 0 end of EF1-a of Beauveria spp. isolates analyzed by maximum parsimony and Close-Neighbor-Interchange algorithm. Branch lengths indicate number of changes over the whole sequence. Numbers on branches indicate the percentages of replicate trees in which the associated taxa clustered together. Isolate identities are indicated for all isolates while host origin (black = Meligethes aeneus; gray = Ips typographus; white = Melolontha melolontha) and location of collection are shown for Swiss isolates. Isolate numbers and location abbreviations as listed in Supplement Table S1 are shown. Major Beauveria clades are indicated. Reference isolates (ARSEF or KVL) are listed in Supplement Table S2.

4 N.V. Meyling et al. / Journal of Invertebrate Pathology 109 (2012) For selected clades identified by sequence analysis, allele sharing distances (Bowcock et al., 14) were calculated in Excel using the Microsatellite Toolkit (Park, 2001). 3. Results 3.1. Phylogenetic position of Beauveria isolates Sequences of the 5 0 end of EF1-a obtained from the 36 Swiss Beauveria spp. isolates were aligned with those of 10 Beauveria reference strains. Maximum parsimony analysis based on 1000 bootstrapped data sets (764 positions of which 66 were parsimony informative) generated 568 most parsimonious trees for the 5 0 end of EF1-a, one of which is presented in Fig. 1. This phylogenetic analysis consistently resolved four distinct clades, which were identified based on the reference strains as B. bassiana s.s. Eu_1 and Eu_4 (both in Clade A), B. brongniartii (Clade B) and Beauveria Clade C. Within Clade A, the clade B. bassiana s.s Eu_1 included 10 isolates from M. aeneus and the clade B. bassiana s.s Eu_4 included three isolates from M. aeneus. Isolate 2428 from M. aeneus and morphologically typed as B. brongniartii clustered within the B. brongniartii clade while 18 isolates clustered with Beauveria Clade C. Isolates obtained from M. melolontha and I. typographus clustered together with reference strain B. brongniartii JE276 or B. bassiana s.s. Eu_4 ARSEF 1848, respectively. The three isolates from M. aeneus belonging to B. bassiana s.s. Eu_4 were collected at the same location, tzenrüti (), while the 10 isolates from M. aeneus in B. bassiana s.s. Eu_1 were collected at five different locations in Switzerland. Five isolates from M. aeneus collected in Landquart (Lq) clustered all in Beauveria Clade C, while the remaining 13 isolates from M. aeneus in this clade were collected at eight other locations (Fig. 1). Maximum parsimony analysis of the ITS sequences from the 36 Swiss isolates and the 10 reference isolates partitioned the isolates among the same four distinct major clades as the analysis of the 5 0 end of EF1-a sequences although with lower bootstrap support of nodes, i.e. ranging from 81% to 94% (data not shown). Corresponding analysis of sequences from the intergenic Bloc region confirmed the topology of the three clades it could be PCR amplified from, i.e., B. bassiana s.s. Eu_1 and Eu_4, and B. brongniartii with bootstrap support of nodes ranging from of 91% to 100% (data not shown) Sequence based genotypic variability within and among Beauveria clades Sequences of the 5 0 end of EF1-a and ITS each identified a single genotype in B. bassiana s.s. Eu_1 and one genotype in Eu_4 (Fig. 1 and Table 1). The two groups differed by 24 bp in the sequences of the 5 0 end of EF1-a and by 5 bp in the ITS sequences. The Bloc sequences revealed two genotypes within B. bassiana s.s Eu_1 and one genotype within B. bassiana s.s. Eu_4 (Table 1). The two genotypes within B. bassiana s.s. Eu_1 differed by a single bp (isolates 2581, 2586 and 2587 versus the remaining seven isolates) and differences of bp were detected between the two genotypes of B. bassiana s.s. Eu_1 and B. bassiana s.s. Eu_4, respectively. The four B. brongniartii isolates were separated into two genotype groups by sequences of all three regions with one genotype consisting of isolate 2428 from M. aeneus and the other consisting of the three isolates from M. melolontha (Table 1). The two genotypes differed by a single bp in the ITS region, 4 bp in the 5 0 end of EF1-a, and 7 bp in the Bloc region. All 18 isolates of Beauveria Clade C had identical ITS sequences (Table 1). However, sequences of the 5 0 end of EF1-a revealed five different genotype groups among these isolates. The distinct Table 1 Genotypes identified within the three major clades, B. bassiana s.s. (Clade A), B. brongniartii (Clade B) and Beauveria Clade C based on sequence difference matrices for region specific alignments in BioEdit (Hall, 19). Genetic distances were based on Jukes Cantor distance determination in MEGA 4.1 (Tamura et al., 2007). Region ITS 5 0 EF1-a Bloc B. bassiana s.s. Eu_1 and Eu_4 (n = 14) Positions used (bp) Number of genotypes Mean distances (range) ( ) ( ) genotype represented by isolate 2574 differed by bp from all other genotypes while the remaining genotype groups only differed by 1 4 bp from each other (Table 2) SSR-marker based genotypic variability within and among Beauveria clades ( ) B. brongniartii (n = 4) Positions used (bp) Number of genotypes Mean distances (range) ( ) ( ) Beauveria Clade C (n = 18) Positions used (bp) Number of genotypes 1 5 Mean distances (range) ( ) ( ) The nine primer sets originally designed for B. bassiana s.s. yielded successful amplification from the ten B. bassiana s.s. Eu_1 isolates (Table 3; see Supplement Table S4 for all alleles). Polymorphism was detected at eight loci while loci Ba06 and Ba12 were monomorph. Seven of the nine B. bassiana s.s. SSR loci were successfully amplified from the four B. bassiana s.s. Eu_4 isolates and each of these loci were represented by single alleles. Four of the nine loci were successfully amplified from all four B. brongniartii. Among the B. brongniartii isolates, polymorphism was detected at loci Ba08 and Ba26. Isolates of Beauveria Clade C revealed positive amplification from loci Ba12 and Ba13 yielding one and two alleles, respectively. Three of the six SSR loci that were originally isolated from B. brongniartii yielded amplification products from all B. bassiana s.s. Eu_1 and Eu_4 isolates (Table 3 and Supplement Table S4). No polymorphisms were detected, but alleles at loci Bb1F4 and Bb8D6 were different for the two clades. All six loci were successfully amplified from the four B. brongniartii isolates and they all were polymorphic. Four of the six B. brongniartii loci yielded amplification products from the Beauveria Clade C isolates. Polymorphism was detected at locus Bb8D6 only. Neighbor joining analysis based on allele sharing distances of the 10 successfully amplified SSR loci revealed a clear separation of the two clades Eu_1 and Eu_4 of B. bassiana s.s. (Fig. 2). The SSR analysis separated the 10 isolates of B. bassiana s.s. Eu_1 into seven unique genotypes with allele sharing distances ranging from In contrast isolates of B. bassiana s.s. Eu_4 were not discriminated and displayed a single genotype. Another set of ten SSR markers, i.e., all six loci designed for B. brongniartii and four loci designed for B. bassiana (Table 3), was successfully amplified from all four B. brongniartii isolates and allowed for discrimination of three genotypes. Isolates 384 and 2753 were identical and showed a genetic distance of 0.6 to isolate 376. The genetic distance among isolate 2428 and the three M. melolontha isolates 384, 2753 and 376 was 0.78.

5 80 N.V. Meyling et al. / Journal of Invertebrate Pathology 109 (2012) Table 2 Base pair difference between pairs of 18 Beauveria Clade C isolates in sequences of the 5 0 end of Elongation Factor 1-a (total positions 789 bp) Table 3 Number of alleles [and allele sizes or range of sizes] at 15 SSR loci for each of the four clades identified with DNA sequence based analyses. Unsuccessful amplifications are indicated with. Loci originally isolated from B. bassiana s.s. are designated Ba while loci originally isolated from B. brongniartii are designated Bb. Locus B. bassiana s.s. Eu_1 (n = 10) B. bassiana s.s. Eu_4 (n = 4) B. brongniartii (n = 4) Beauveria Clade C (n = 18) Ba06 1 [101] 1 [104] 1 [88] Ba08 2 [198; 201] 1 [195] 2 [193; 196] Ba12 1 [203] 1 [213] 1 [177] 1 [186] Ba13 2 [149; 152] 1 [158] 1 [178] a 2 [148; 151] Ba15 2 [153; 156] Ba18 4 [ ] Ba21 2 [189; 190] 1 [188] Ba26 5 [ ] 1 [126] 3 [ 159] Ba27 6 [ ] 1 [143] Bb1F4 1 [195] 1 [192] 3 [ ] 1 [193] Bb2F8 1 [148] 1 [148] 3 [ ] 1 [148] b Bb8D6 1 [180] 1 [176] 2 [162; 168] 3 [ ] Bb4H9 2 [173; 179] c 1 [203] d Bb2A3 3 [ ] Bb5F4 3 [ ] a b c d Only a single isolate, 2428, yielded fragment for this locus. Two isolates, 2579 and 2616, failed to yield a fragment for this locus. Isolate 2428 failed to yield a fragment for this locus. Isolate 2574 failed to yield a fragment for this locus. For Beauveria Clade C, six SSR markers discriminated four genotypes separated by allele sharing distances ranging from to 0.5. Isolate 2574 had one unique genotype while isolates 2575, 2582, 2583 and 2589 shared another. The three isolates 2578, 2580 and 2608 shared an allele of 186 bp at locus Bd8D6. The remaining ten isolates had mostly the same allele sizes at all loci, except isolates 2579 and 2616 which shared a null allele at one locus (Bb2F8). 4. Discussion The two genome regions sequenced for all Beauveria spp. isolates, i.e., ITS and 5 0 end of EF1-a, resolved the major clades, i.e. B. bassiana s.s. Eu_1 and Eu_4 within Clade A, B. brongniartii Clade B, and Beauveria Clade C. The 5 0 end of EF1-a provided the best over all resolution, and the intergenic Bloc region proved to be most successful in resolving B. bassiana s.s. and B. brongniartii. Phylogenetic analyses based on the three genome regions corresponded to previous analyses, which confirmed the reliability of this sequencing approach to identify and classify isolates belonging to B. bassiana s.l. (Rehner and Buckley, 2005; Rehner et al., 2006; Meyling et al., 2009). Bischoff et al. (2009) highlighted the 5 0 end of EF1-a as a valuable marker for sorting isolates to clades within the Metarhizium anisopliae lineage. Similarly, sequencing the 5 0 end of EF1-a is well suited for assigning Beauveria spp. isolates to specific clades. For detection of genotype variation within clades, SSR-markers provided good genotype resolution in the present study, but with different resolution for the different clades. Overall, SSR markers worked most consistently and detected highest levels of genetic variability for isolates belonging to the clade for which the markers originally were designed. It is therefore advisable to assign the isolates within Beauveria populations to clades prior to assessment of their genetic structures using SSR markers. Markers that were amplifiable across clades were mostly not polymorphic except for Ba08 and Ba26 in B. brongniartii and Ba13 and Bb8D6 in Beauveria Clade C. SSR markers may therefore underestimate genetic diversity within clades, e.g. Beauveria Clade C, for which they were not designed. These data correspond to those of previous studies reporting limited cross taxa amplification of SSR markers in fungi as compared to animals and plants (Barbara et al., 2007; Vogelgsang et al., 2009). Similar to this study, loci that could be amplified across taxa showed no or only low levels of

6 N.V. Meyling et al. / Journal of Invertebrate Pathology 109 (2012) B. bassiana s.s. Eu_ 1 B. bassiana s.s. Eu_ Fig. 2. Neighbor joining tree based on allele sharing distances of SSR data for B. bassiana s.s. Eu_1 and Eu_4 isolates from M. aeneus. Ten SSR loci which revealed PCR amplification products for all isolates were included in the analysis. Circles indicate the clades B. bassiana s.s. Eu_1 and Eu_4, respectively. polymorphism suggesting that transfer of SSR markers among fungal species may generally be limited. It will therefore be important to isolate SSR markers specifically for Beauveria Clade C for which currently no markers are available to allow accurate assessment of genetic variability. Four of the SSR markers which were detectable across all the Beauveria clades i.e., Ba12, Ba13, Bb1F4, and Bb8D8, revealed clade-specific allele sizes. These markers represent a potential tool for efficient assignment of isolates to specific Beauveria clades. However, more isolates must be analyzed to validate cladespecificity of SSR markers. Some recent studies focused on characterizing B. bassiana s.l. isolates from North America by 7-8 SSR markers as the only molecular tool (McGuire et al., 2005; Castrillo et al., 2008, 2010). In these studies all PCR amplifications were successful indicating that isolates belonged to B. bassiana s.s. (Clade A). Of the markers applied in the present study, loci Ba06, Ba08, Ba12 and Ba13 were also used by Castrillo et al. (2008, 2010) while McGuire et al. (2005) used Ba06, Ba08 and Ba13. Although these loci were not the most polymorphic ones in the present study, Castrillo et al. (2010) recorded 20 different alleles for Ba06, 29 for Ba08, 17 for Ba12 and 12 for Ba13 among 82 isolates, suggesting a genetically highly diverse isolate collection of Beauveria spp. However, comparisons to the present study are not possible without assignment of isolates to clades. Within Clade A isolates were separated into two subclades, which have previously been recognized as phylogenetic species, i.e. B. bassiana s.s. Eu_1 and Eu_4 (Rehner et al., 2006; Meyling et al., 2009). Neither characterization with ten SSR markers nor DNA sequencing of the three regions revealed any differences among the four Swiss isolates of B. bassiana s.s. Eu_4. This suggests a clonal origin of these isolates of which three were collected from M. aeneus in May 2004 at the same site and one isolate was collected from the bark beetle I. typographus 2 years later at another location. This further indicates that the detected genotype may have a wide host range and good persistence in the environment. In contrast to the B. bassiana s.s. Eu_4 isolates, a high allelic diversity allowing for discrimination of seven genotypes was detected among the 10 B. bassiana s.s. Eu_1 collected from M. aeneus at five different locations. This indicates that several genotypes of B. bassiana s.s. Eu_1 are widespread in Switzerland and that many of these include M. aeneus in their ecological host range. Comparable patterns of genetic diversity among isolates of B. bassiana s.s. Eu_4 and Eu_1 have been reported in a previous study performed at an agroecosystem in Denmark (Meyling et al., 2009). In that study only minor allelic differences have been detected among 10 B. bassiana s.s. Eu_4 isolates by applying 18 SSR markers, whereas 33 B. bassiana s.s. Eu_1 isolates collected at a single field site were highly diverse (Meyling et al., 2009). However, detailed studies based on thorough sampling of different habitats and hosts in different regions will have to be performed in order to substantiate the observed differences between these two European clades. Genetic analyses confirmed the morphological identification of an isolate obtained from M. aeneus (2428) as B. brongniartii. This isolate was genotypically different from the Swiss B. brongniartii isolates obtained from M. melolontha, but similar to an isolate (KVL 03-91) collected from a carabid beetle in Denmark. Results reported here and in a previous study (Enkerli et al., 2001) indicate that different host taxa among Coleoptera may be infected by distinct genotypes of B. brongniartii. However, preliminary data suggested that the B. brongniartii isolates studied here cause mycosis at low frequencies in M. aeneus indicating that B. brongniartii may be a weak and unspecific pathogen of M. aeneus, and natural infections of M. aeneus by B. brongniartii may be relatively rare (Pilz and Keller, unpublished). About half of the Beauveria isolates collected from infected M. aeneus at nine locations belonged to Beauveria Clade C (18 of 32). This indicated that Beauveria Clade C may be widespread in the investigated Swiss farmland. Furthermore, the observed frequency was relatively high compared to the frequencies of this clade reported in other studies. For instance, 5% of the Beauveria spp. isolates collected from soil and from infected insects in an agroecosystem in Denmark have been demonstrated to belong to Beauveria Clade C (Meyling et al., 2009). Studies based on Beauveria isolates obtained from culture collections and representing worldwide distributions have revealed frequencies of Beauveria Clade C of 17% (Rehner and Buckley, 2005) and 22% (Ghikas et al., 2010), respectively. Furthermore, Beauveria Clade C isolates have been shown to originate from five separate insect host orders (Rehner and Buckley, 2005) indicating a wide host range of Beauveria Clade C. The genetic diversity detected among isolates of Beauveria Clade C in the present study suggests that this clade of Beauveria may contain variation in ecologically significant traits. In Denmark Beauveria Clade C isolates were only found in hedgerow habitats while B. bassiana s.s. Eu_1 was the only clade identified among isolates from an agricultural field (Meyling et al., 2009) suggesting that factors other than host insect is determining the distribution of Beauveria Clade C in the agricultural landscape. If these distribution patterns are transferable to Switzerland the M. aeneus hosts may have acquired the infection of Beauveria Clade C at their overwintering sites. The observed low prevalence of Beauveria spp. infections in Swiss M. aeneus populations (1.8%) during spring migration to oilseed rape fields (Pilz, 2005; Pilz and Keller, 2006) suggests that mortality by fungal entomopathogens in M. aeneus may not significantly contribute to the population regulation of adult beetles. However, since Beauveria spp. genotypes naturally infect adult pollen beetles the dispersal of fungal genotypes due to migration of infected individuals of M. aeneus may contribute to fungal distribution in the investigated farmlands. It remains unknown where the M. aeneus individuals in Switzerland had acquired their infections with Beauveria spp., but it is possible that the infections were initiated in the overwintering sites of the adult beetles, i.e. hedgerows and forest edges (e.g., Alford et al., 2003). Future studies specifically focusing on the ecology of Beauveria spp., particularly Beauveria Clade C, may provide further important information on this interesting group of fungal entomopathogens.

7 82 N.V. Meyling et al. / Journal of Invertebrate Pathology 109 (2012) Acknowledgments We would like to thank Christina Wolsted and Diego Bünter for valuable technical assistance. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi: /j.jip References Alford, D.V., Nilsson, C., Ulber, B., Insect pests of oilseed rape crops. In: Alford, D.V. (Ed.), Biocontrol of Oilseed Rape Pests. Blackwell Science Ltd., pp Barbara, T., Palma-Silva, C., Paggi, G.M., Bered, F., Fay, M.F., Lexer, C., Crossspecies transfer of nuclear microsatellite markers: potential and limitations. Mol. Ecol. 16, Bischoff, J.F., Rehner, S.A., Humber, R.A., Metarhizium frigidum sp. nov.: a cryptic species of M. anisopliae and a member of the M. flavoviride complex. Mycologia 98, Bischoff, J.F., Rehner, S.A., Humber, R.A., A multilocus phylogeny of the Metarhizium anisopliae lineage. Mycologia 101, Bowcock, A.M., Ruiz-Linares, A., Tomfohrde, J., Minch, E., Kidd, J.R., Cavalli-Sforza, L.L., 14. High resolution of human evolutionary trees with polymorphic microsatellites. Nature 368, Büchs, W., Alford, D.V., Predators of oilseed rape pests. In: Alford, D.V. (Ed.), Biocontrol of Oilseed Rape Pests. Blackwell Science Ltd., pp Castrillo, L.A., Ugine, T.A., Filotas, M.J., Sanderson, J.P., Vandenberg, J.D., Wraight, S.P., Molecular characterization and comparative virulence of Beauveria bassiana isolates (Ascomycota: Hypocreales) associated with the greenhouse shore fly, Scatella tenuicosta (Diptera: Ephydridae). Biol. Control 45, Castrillo, L.A., Bauer, L.S., Liu, H., Griggs, M.H., Vandenberg, J.D., Characterization of Beauveria bassiana (Ascomycota: Hypocreales) isolates associated with Agrilus planipennis (Coleoptera: Buprestidae) populations in Michigan. Biol. Control 54, Enkerli, J., Widmer, F., Gessler, C., Keller, S., Strain-specific microsatellite markers in the entomopathogenic fungus Beauveria brongniartii. Mycol. Res. 105, Ghikas, D.V., Kouvelis, V.N., Typas, M.A., Phylogenetic and biogeographical implications inferred by mitochondrial intergenic region analyses and ITS1-5.8S-ITS2 of the entomopathogenic fungi Beauveria bassiana and B. brongniartii. BMC Microbiol. 10, Art. No doi: / Hall, T.A., 19. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41, Hokkanen, H.M.T., 13. Overwintering survival and spring emergence in Meligethes aeneus: effects of body weight, crowding, and soil treatment with Beauveria bassiana. Entomol. Exp. Appl. 67, Hokkanen, H.M.T., Husberg, G.-B., Söderblom, M., Natural enemy conservation for the integrated control of the rape blossom beetle Meligethes aeneus F. Ann. Agric. Fenn. 27, Hokkanen, H.M.T., Menzler-Hokkanen, I., Butt, T.M., Pathogens of oilseed rape pests. In: Alford, D.V. (Ed.), Biocontrol of Oilseed Rape Pests. Blackwell Science Ltd., pp Lipa, J.J., Hokkanen, H.M.T., 12. Nosema meligethi I. and R. (Microsporidia) in populations of Meligethes spp. in Europe. Biocontrol Sci. Technol. 2, McGuire, M.R., Ulloa, M., Park, Y.H., Hudson, N., Biological and molecular characteristics of Beauveria bassiana isolates from California Lygus hesperus (Hemiptera: Miridae) populations. Biol. Control 33, Meyling, N.V., Lübeck, L., Buckley, E.P., Eilenberg, J., Rehner, S.A., Community composition, host-range and genetic structure of the fungal entomopathogen Beauveria in adjoining agricultural and semi-natural habitats. Mol. Ecol. 18, Nilsson, C., Yield losses in summer rape caused by pollen beetles (Meligethes spp.). Swed. J. Agric. Res. 17, Nilsson, C., Parasitoids of pollen beetles. In: Alford, D.V. (Ed.), Biocontrol of Oilseed Rape Pests. Blackwell Science Ltd., pp Park, S.D.E., Trypanotolerance in West African Cattle and the Population Genetic Effects of Selection. Ph.D. thesis, University of Dublin, eland. Pilz, C., Natürliches ftreten insektenpathogener Pilze beim Rapsglanzkäfer (Meligethes aeneus) und Versuche der Eignung von Metarhizium anisopliae zur mikrobiellen Bekämpfung. M.Sc. thesis, Universität für Bodenkultur (BOKU), Vienna, stria. Pilz, C., Keller, S., Pilzkrankheiten bei adulten Rapsglanzkäfern [Fungal diseases in adult pollen beetles]. AGRARForschung 13, Rehner, S.A., Buckley, E.P., Isolation and characterization of microsatellite loci from the entomopathogenic fungus Beauveria bassiana (Ascomycota: Hypocreales). Mol. Ecol. Notes 3, Rehner, S.A., Buckley, E.P., A Beauveria phylogeny inferred from nuclear ITS and EF1-alpha sequences: evidence for cryptic diversification and links to Cordyceps teleomorphs. Mycologia 97, Rehner, S.A., Posada, F., Buckley, E.P., Infante, F., Castillo, A., Vega, F.E., Phylogenetic origins of African and Neotropical Beauveria bassiana s.l. pathogens of the coffee berry borer, Hypothenemus hampei. J. Invert. Pathol. 93, Riba, G., Ravelojoana, A.M., The parasexual cycle in the entomopathogenic fungus Paecilomyces fumoso-roseus (Wize) Brown and Smith. Can. J. Microbiol. 30, Roy, H.E., Vega, F.E., Chandler, D., Goettel, M.S., Pell, J.K., Wajnberg, E., The Ecology of Fungal Entomopathogens. Springer Verlag. Tamura, K., Dudley, J., Nei, M., Kumar, S., MEGA 4: molecular evolutionary genetics analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24, Vestergaard, S., Cherry, A., Keller, S., Goettel, M., Safety of hyphomycete fungi as microbial control agents. In: Hokkanen, H.M.T., Hajek, A.E. (Eds.), Environmental Impacts of Microbial Insecticides. Kluwer Academic Publishers, Dordrecht, NL, pp Vogelgsang, S., Widmer, F., Jenny, E., Enkerli, J., Characterisation of novel Fusarium graminearum microsatellite markers in different Fusarium species from various countries. Eur. J. Plant Pathol. 123, Warner, D.J., Allen-Williams, L., Warrington, S., Ferguson, A.W., Williams, I.H., Implications for conservation biocontrol of spatio-temporal relationships between carabid beetles and coleopterous pests in winter oilseed rape. Agric. Forest Entomol. 10, White, T.J., Bruns, T.D., Lee, S., Taylor, J., 10. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis, M.A., Gelfand, D.H., Sninsky, J.J., White, T.J. (Eds.), PCR Protocols. Academic Press, San Diego, USA, pp Zimmermann, G., Review on safety of the entomopathogenic fungi Beauveria bassiana and Beauveria brongniartii. Biocontrol Sci. Technol. 17,