MEC1291.fm Page 1423 Monday, May 7, 2001 4:58 PM Molecular Ecology (2001) 10, 1423–1432 HierarchicalBlackwell Science, Ltd genetic structure of the introduced wasp Vespula germanica in Australia MICHAEL A. D. GOODISMAN,*‡ ROBERT W. MATTHEWS† and ROSS H. CROZIER* *Department of Genetics, La Trobe University, Bundoora, VIC, 3083, Australia and Department of Zoology and Tropical Ecology, James Cook University, Townsville, QLD, 4811, Australia, †Department of Entomology, University of Georgia, Athens, GA, 30602, USA Abstract The wasp Vespula germanica is a highly successful invasive pest. This study examined the population genetic structure of V. germanica in its introduced range in Australia. We sampled 1320 workers and 376 males from 141 nests obtained from three widely separated geographical areas on the Australian mainland and one on the island of Tasmania. The genotypes of all wasps were assayed at three polymorphic DNA microsatellite markers. Our analyses uncovered significant allelic differentiation among all four V. germanica popu- lations. Pairwise estimates of genetic divergence between populations agreed with the results of a model-based clustering algorithm which indicated that the Tasmanian popu- lation was particularly distinct from the other populations. Within-population analyses revealed that genetic similarity declined with spatial distance, indicating that wasps from nests separated by more than ~25 km belonged to separate mating pools. We suggest that the observed genetic patterns resulted from frequent bottlenecks experienced by the V. germanica populations during their colonization of Australia. Keywords: DNA microsatellite, introduced species, migration, social insects, yellowjacket wasp Received 7 October 2000; revision received 8 January 2001; accepted 8 January 2001 siderable harm to indigenous wildlife because they are Introduction voracious polyphagous predators on native arthropods The organization of natural ecosystems may be disrupted and outcompete other animals for food (e.g. Moller & by the invasion of foreign species (Mooney & Drake 1986; Tilley 1989; Beggs & Wilson 1991; Fordham 1991; Harris Drake et al. 1989; Williamson 1996). Social insects are a 1991; Beggs 2000). particularly notorious group of invaders. Various intro- V. germanica was first discovered in Australia on the duced termites, ants and wasps have caused substantial island of Tasmania in 1959 (V. germanica colonization of environmental and economic harm in many of their intro- Australia reviewed by Crosland 1991; Spradbery & duced habitats (Vinson 1986; Williams 1994; Moller 1996). Maywald 1992). However, it was not recorded on the The European wasp or German yellowjacket, Vespula Australian mainland until 15 years later. By then it was germanica, is an especially successful and destructive social found in Adelaide, Melbourne, Perth and Sydney. Sub- insect in its introduced range. Originating from Europe, sequently V. germanica established itself as a major pest Asia and North Africa, V. germanica now enjoys a nearly throughout the south-east of the country. Estimates of the circumglobal distribution, having established populations natural dispersal rate of V. germanica indicate that the wasp in North America, South America, South Africa, New can spread no more than ~1000 m per year (Thomas 1960; Zealand and Australia (Spradbery 1973; Akre & MacDonald Edwards 1980; Crosland 1991). This relatively short dis- 1986; Clapperton et al. 1989). Vespula wasps cause con- persal distance cannot account for the rapid appearance of V. germanica in Australian cities separated by hundreds of kilometres. Rather, the movement of V. germanica over such long distances probably results from the accidental trans- Correspondence: Michael A. D. Goodisman. ‡Present address: c/o Dr Wells Lab-Department of Biochemistry, Biological Sciences West, port of hibernating queens (Thomas 1960; Crosland 1991; PO Box 210088, University of Arizona, Tucson, AZ 85721-0088, USA. Spradbery & Maywald 1992). Such human-aided dispersal Fax: 520-621-9288; E-mail: [email protected] normally involves the movement of relatively few queens © 2001 Blackwell Science Ltd MEC1291.fm Page 1424 Monday, May 7, 2001 4:58 PM 1424 M. A. D. GOODISMAN, R. W. MATTHEWS and R. H. CROZIER Fig. 1 Current distribution of Vespula germanica in Australia by interim biogeographic region (modified and reprinted with permission from Clarke et al. 2001) and locations of four populations sampled in this study. (Crosland 1991), and thus causes bottlenecks during the Materials and methods establishment of wasp populations. The purpose of this study is to investigate the population Sampling genetic structure of the V. germanica in Australia using DNA microsatellite markers. The manner in which V. germanica Vespula germanica wasps were collected from annual colon- has colonized Australia leads us to predict that introduced ies in the austral summer and autumn of 1998/99. Samples populations will display a particular pattern of hierarchical were obtained from populations around the cities of genetic structure. Specifically, if V. germanica populations Adelaide, Canberra and Melbourne, which are found on underwent bottlenecks when they were founded, then the Australian mainland, and Hobart, which is located on they should also have been subjected to strong episodes of the Australian island of Tasmania (Fig. 1). Wasps were genetic drift (Chakraborty & Nei 1977). Genetic drift, in turn, netted individually, collected from excavated nests, or would have resulted in significant differentiation among captured in Malaise traps. All wasps captured away independently founded populations. We do not anticipate from their nests (e.g. those captured in Malaise traps) substantial structure within populations because we expect were assumed to originate from different nests and thus that the flight capabilities of queens would be sufficient to represent genetically independent samples. Once caught, homogenize allele frequencies on a local level. Overall, this specimens were immediately frozen or placed in 95% study should help elucidate the manner in which social ethanol for subsequent DNA analysis. The samples con- insects establish populations when introduced into new sisted mostly of workers; however, males were also obtained habitats. opportunistically. © 2001 Blackwell Science Ltd, Molecular Ecology, 10, 1423–1432 MEC1291.fm Page 1425 Monday, May 7, 2001 4:58 PM GENETIC STRUCTURE OF INTRODUCED V. GERMANICA 1425 AluI, AscI, BspDI, BsrGI, DdeI, DraI, EcoRV, HinfI, MseI, Laboratory methods SspI, Tsp509I and XbaI according to supplier recommenda- We assayed the genotypes of all individuals at three poly- tions. The fragments were electrophoresed on a 3.5% agarose morphic microsatellite markers, Rufa 5, Rufa 18 and Rufa gel containing ethidium bromide and visualized with UV 19, originally developed for V. rufa (P. Thorén, unpublished light. results). Genomic DNA for microsatellite analysis was extracted from single V. germanica legs using a variation Statistical analyses of the Chelex® protocol (Walsh et al. 1991) as described by Crozier et al. (1999). The loci were polymerase chain Allele frequencies at the three microsatellite markers within reaction (PCR) -amplified using either fluorescently or each of the four populations were estimated using the radioactively labelled PCR primers. program relatedness 4.2, with nests weighted equally To visualize microsatellites using fluorescent techniques, (Queller & Goodnight 1989). Nei’s (1987) unbiased estimate the forward primers of Rufa 5, 18 and 19 were first labelled of gene diversity, with TET, FAM and TET dyes, respectively. PCRs were µ µ − Σ 2 − conducted in a final volume of 15 L containing 3 L hjk = 2n (1 i pijk )/(2n 1) (1) genomic DNA and 0.75 U Taq DNA polymerase (Promega), and a final concentration of 200 µm dNTPs, 0.5 µm of each of where n is the number of diploid individuals sampled × the forward and reverse PCR primers and 1 Promega Buffer and pijk is the frequency of allele i at locus j (for j = 1 – 3 (with 1.5 mm MgCl2). PCR products were then electro- where Rufa 5, Rufa 18 and Rufa 19 are loci 1, 2 and 3) in phoresed and scored on an ABI Prism 377 DNA Sequencer. population k (for k = 1 – 4 where Adelaide, Canberra, To visualize microsatellites using radioactivity, the for- Hobart and Melbourne are populations 1, 2, 3 and 4), was ward PCR primer for amplifying each locus was first end- used as a measure of variability for each marker in each labelled with [γ33P]ATP. PCRs were then carried out in a population. final volume of 10 µL containing 2 µL genomic DNA, 5 µg We investigated if the levels of genetic variation differed bovine serum albumin (New England Biolabs) and 0.4 U across southeastern Australia by testing if the mean gene Taq DNA polymerase, and a final concentration of 167 µm diversity of all three microsatellite markers, dNTPs, 0.4 µm of unlabelled reverse primer, 0.1 µm of µ Σ unlabelled forward primer, 0.03 m of radioactively end- Hk = ( j hjk)/3 labelled forward primer and 1× Promega Buffer (with 1.5 mm MgCl2). PCR products were run on 5% denaturing varied among the four populations. We conducted these polyacrylamide gels at a constant power of 65 W for 3–4 h comparisons using two different population sample sizes and then scored after exposing the dried gel to film (Kodak to account for instances where multiple wasps were col- or Fuji) overnight. For both the fluorescent and radioactive lected per nest. First, we conservatively took the sample methods, the PCR cycling profiles for the three markers size within each population to equal the number of nests ° began with an initial denaturation at 94 C for 2 min, and from which wasps were obtained (Nk). Second, we considered then proceeded with 30 or 35 cycles of 93 °C for 30 s, 55 °C a less restrictive sample size, which incorporated informa- (Rufa 5 and 19) or 52 °C (Rufa 18) for 30 s, and 72 °C for tion on the total number of haploid genomes detected 30 s, followed by a final extension of 72 °C for 10 min. in a given population. This method relies on the fact The wasps were also screened for restriction fragment- that annual V.