Rosenthal and Spielman
Total Page:16
File Type:pdf, Size:1020Kb
December, 2004 Journal of Vector Ecology 227 Reduced variation among northern deer tick populations at an autosomal microsatellite locus Benjamin M. Rosenthal and Andrew Spielman Animal Parasitic Diseases Laboratory, USDA-Agricultural Research Service, Beltsville, MD 20705, U.S.A. Department of Immunology and Infectious Diseases, Harvard School of Public Health, 665 Huntington Avenue, Boston, MA 02115, U.S.A. Received 28 October 2003; Accepted 5 March 2004 ABSTRACT: To determine whether genes flow freely between populations of the Ixodes ricinus-like ticks of eastern North America, and to determine whether the abundant northerly populations of these vectors of Lyme disease and other zoonotic infections may have arisen recently from a small cohort of ancestral founders, we characterized the nuclear IR27 microsatellite alleles in ticks sampled from the geographic extremes of their ranges. These microsatellite alleles differentiated populations located in the southeastern, northcentral and northeastern United States, respectively. Although evident heterozygous genotypes are about as frequent as would be expected in randomly mating populations, particular microsatellite alleles and diploid genotypes occur more frequently in certain populations than in others. Ticks from the Northeast and upper Midwest are markedly more related to each other than to ticks from the Southeast. Patterns of diversity present in this nuclear microsatellite marker correspond to those evident at a mitochondrial locus and indicate that the deer ticks of the Northeast and upper Midwest are genetically isolated from those in the Southeast. The Ixodes ricinus-like ticks that impose a public health burden in the northeastern and northcentral United States originated recently in a common founder population. Journal of Vector Ecology 29 (2): 227-235. 2004. Keyword Index: Lyme disease, Ixodes, scapularis, dammini, microsatellite, ticks. INTRODUCTION recorded over a seven-year period in Westchester County, NY (Falco et al. 1995). The frequency Surprisingly limited genetic diversity may distribution of mitochondrial 16S haplotypes of Ixodes characterize even very large populations if these deer ticks in New England, and the genetic similarity of organisms derived recently from small ancestral cohorts. these haplotypes, substantiate survey data indicating a Because small populations tend to lose genetic diversity remarkable, recent explosion in this tick’s regional more rapidly than do larger populations, the effective abundance (Qiu et al. 2002). size of a population may be estimated as its harmonic The geographic isolation and distinct nymphal mean, an expression dominated by its smallest terms morphology of deer ticks from coastal New England (Hartl and Clark 1989, Nei 1975). The diminution of were initially interpreted as evidence for their status as genetic variance through population bottlenecks poses a a distinct species, I. dammini Spielman, Clifford, critical concern for conservation programs, but may also Piesman, and Corwin (Spielman et al. 1979). Subsequent occur when highly successful invasive species proliferate studies of phenotypic and genotypic variation, however, from small originating populations. In this respect, the concluded that I. dammini should be relegated as a junior genetic structure of epidemic disease agents strikingly synonym of I. scapularis Say (McLain et al. 1995, Oliver reflects that of rapidly and recently proliferating et al., 1993b, Wesson et al. 1993). Although northerly- populations (Grassly et al. 1999). Until recent range distributed deer ticks descend from solely one of two, expansion, deer ticks in the eastern United States were significantly differentiated matrilineages (Rich et al. recognized only in small geographic foci (Dennis et al. 1995, Norris et al. 1996), analyses of nuclear genetic 1998). Successive surveys, as in Wisconsin (Riehle and markers have to date indicated no such geographic Paskewitz 1996) and Indiana (Pinger et al. 1996), attest differentiation; distinct geographic distributions of two to their expanding range and increasing local population forms of the D3 expansion sequence of nuclear rDNA density. A thirty-fold increase in larval abundance was were interpreted as evidence that I. scapularis was 228 Journal of Vector Ecology December, 2004 comprised of hybridized, previously divergant lineages (Invitrogen Corp.). The IR27F primer was modified at (McLain et al. 2001). A study of nuclear diversity in the 5’ end by the addition of the 6-FAM fluorescent dye. these ticks analyzed the internal transcribed spacer region Products were electrophoresed through uniform, pre- (ITS-2) of nuclear rDNA and found no evidence of mixed acrylamide sequencing gels (Long Ranger Singel diagnostic variation between specimens from packs, BioWhittaker Molecular Applications, Rockland, geographically separated sites (Wesson et al. 1993). ME) detected on an Applied Biosystems 377 automated Another study, which analyzed the diversity of randomly fluorescent sequencer and analyzed using Genescan and amplified polymorphic DNA (RAPD) markers, Genotypr software (Applied Biosystems). suggested a similar wide-ranging panmixia (Norris et Each sample was run in a lane that also included al. 1996). If genetic material does not regularly flow molecular weight size standards (TAMRA-350, Applied between regional deer tick populations, as indicated by Biosystems) to minimize misclassification that might the structured geographical distribution of mitochondrial otherwise result from variation among gels or among marker alleles, the apparently homogeneous allelic the lanes of a given gel. Each allele was classified distribution of nuclear genes would present a paradox. according to the molecular weight at which peak Such a discrepancy could result if dispersal was greater fluorescence was detected by GeneScan. Typically, each among male than among female North American allele was also accompanied by a minor fluorescent peak members of the Ixodes ricinus complex as has recently that was shorter by two bases (which subsequent been suggested for their European counterparts (De sequencing confirmed to lack a CA repeat) and a peak Meeus et al. 2002). longer by one base (to which Taq polymerase had added An analysis of the geographical distribution of a terminal A). polymorphic nuclear microsatellite loci may help resolve Reproducibility of the assay was confirmed by the conflict between mitochondrial markers, which replicate amplification and sequencing of selected differentiate eastern North American deer tick individuals. To verify the absence of contaminating populations, and nuclear markers, which find no such DNA, PCR was also attempted from tubes processed by evidence for genetic structure. To explore this suggestion, identical extraction and amplification procedures but to we determined whether alleles of a single copy nuclear which no template was added. To verify that the basis of microsatellite locus are distributed homogeneously size variation among alleles at this locus were variable between southeastern, northcentral, and northeastern sites numbers of CA dinucleotide repeats, selected PCR in the United States. In addition, we determined whether products were cloned using the TOPO TA kit (Invitrogen heterozygous genotypes occur in these populations less Corp.) and sequenced on an Applied Biosystems 3100 frequently than would be expected if the Ixodes ricinus- automated DNA sequencer using Big Dye Terminator like ticks of eastern North America were panmictic. chemistry. Additionally, we characterized the extent to Finally, we compared the relatedness of specimens which in vitro amplification might artifactually contribute sampled from each of these sites in order to determine to allelic diversity estimates by sequencing products that whether the abundant deer tick populations responsible were amplified from a single, cloned IR27 allele. for the Lyme disease epidemic have proliferated recently To evaluate allelic and genotypic differentiation, from a small number of ancestors. contingency analyses were used to compare the observed geographic distribution of alleles and genotypes to MATERIALS AND METHODS distributions expected under the null hypothesis of geographic homogeneity, simulated using a Markov Adult deer ticks were sampled by flagging Chain Monte Carlo approach implemented in GenePop vegetation in sites in the northcentral (Spooner, WI and Ver. 3.3 (Delaye 1998, Rousset and Raymond 1995). Long Point, Ontario), southeastern (Gainesville, FL), and To test for assortative mating, actual heterozygote two northeastern (Naushon Island and Nantucket Islands, frequencies were compared to what would be expected MA) parts of North America. in populations of randomly mating individuals. In such Fifty-eight of 72 sampled ticks were successfully populations conforming to Hardy-Weinberg equilibrium, genotyped at the autosomal microsatellite locus IR27 genotypes occur with frequencies proportional to the locus. Microsatellite alleles of locus IR27 were amplified product of their constituent allele frequencies. To and scored according to a modification of previously determine whether heterozygote genotypes occurred less described methods (Delaye 1998, Delaye et al. 1998). frequently than would be expected if genes flowed freely DNA extracted from adult ticks using a CTAB buffer among or within geographically defined populations was used asa template in 20 microliter Polymerase Chain (Wahlund’s effect), expected heterozygote frequencies