Conserv Genet DOI 10.1007/s10592-010-0068-x

RESEARCH ARTICLE

Genetic outcomes of wolf recovery in the western Great Lakes states Steven R. Fain • Dyan J. Straughan Bruce F. Taylor

•

Received: 26 November 2008 / Accepted: 19 February 2010 Ó US Government 2010

Abstract Conflicting interpretations of the influence of coyote hybridization on wolf recovery in the western Great Lakes (WGL) states have stemmed from disagreement over the systematics of North American wolves. Questions regarding their recovery status have resulted. We addressed these issues with phylogenetic and admixture analysis of DNA profiles of western wolves, WGL states wolves and Wisconsin coyotes developed from autosome and Y-chromosome microsatellites and mitochondrial DNA control region sequence. Hybridization was assessed by comparing the haplotypes exhibited by sympatric wolves and coyotes. Genetic variability and connectivity were also examined. These analyses support the recognition of Canis lycaon as a unique species of North American wolf present in the WGL states and found evidence of hybridization between C. lupus and C. lycaon but no evidence of recent hybridization with sympatric coyotes. The recolonized WGL states wolves are genetically similar to historical wolves from the region and should be considered restored. Keywords Endangered species recovery
mtDNA
Y-chromosome
Autosomal microsatellites
Genetic diversity
Admixture analysis
Hybridization
Canis latrans
Canis lupus
Canis lycaon

The findings and conclusions in this article are those of the author(s) and do not necessarily represent the views of the U.S. Fish and Wildlife Service. S. R. Fain (&)
D. J. Straughan
B. F. Taylor National Fish and Wildlife Forensic Laboratory, 1490 East Main Street, Ashland, OR, USA e-mail: [email protected]

Introduction Although wolves in Wisconsin and Michigan had been exterminated by the late 1950s (Wydevan et al. 1995), cooperative state and federal recovery efforts over the past 40 years (ESA, Federal Register Vol. 39, January 4, 1974) have fostered a remnant of 350–700 wolves in northeast Minnesota to over 3000 individuals and recolonized Wisconsin and Michigan with some 1,000 wolves. In 2003, the FWS removed the recovered WGL population from the protection of the ESA. However, lawsuits followed that overturned this action as well as three subsequent attempts to delist WGL wolves. In April 2009, the FWS delisted WGL wolves as a Distinct Population Segment but this action was vacated in July 2009 in response to another lawsuit. As a result, WGL wolves remain protected under the ESA. Recovery decisions concerning wolves in the WGL states continue to be dogged by controversy regarding the scale and complexity of wolf–coyote hybridization (Mech 2008, Leonard and Wayne 2008a; Wheeldon and White 2008). Moreover, this issue is further complicated by the lack of general agreement surrounding the systematics of North American wolves and the manner in which wolf–coyote hybrids have been identified (Lehman et al. 1991; Wilson et al. 2000; Kyle et al. 2006; Murray Berger and Gese 2007; Kyle et al. 2008; Nowak 2002, 2003, 2009). Numerous studies have established that the wolves in eastern North America are phenotypically distinct from western gray wolves (C. lupus) and coyotes (C. latrans) (Young and Goldman 1944; Kolenosky and Stanfield 1975; Nowak 1995); but initial investigations of the mitochondrial DNA (mtDNA) of Minnesota wolves found gray wolf mtDNA haplotypes as well as ‘‘coyote-like’’ haplotypes that were attributed to ongoing wolf–coyote hybridization

123

Conserv Genet

(Lehman et al. 1991; Roy et al. 1994). However, subsequent DNA sequence analysis of mtDNA from eastern Canadian wolves showed that some ‘‘coyote-like’’ haplotypes were actually ancient, 150,000–300,000 years divergent from coyotes (Wilson et al. 2000), while others were found in historic eastern wolf samples collected before coyotes had invaded the region (Wilson et al. 2003). Wilson et al. (2000, 2003) reasoned from this that eastern wolves were a unique canid species, C. lycaon, of North American origin and closer affinity to coyote and red wolf (C. rufus) than to gray wolf. Alternatively, Leonard and Wayne (2008b) and Koblmu¨ller et al. (2009) recognized a unique Great Lakes ecotype of C. lupus lycaon with significant introgression from coyote. They also disputed the successful recovery of the WGL states population by arguing that pre-extirpation wolves showed no evidence of hybridization with coyotes, whereas a mtDNA haplotype cladistically associated with coyotes was common in the recovered population. Regardless of disagreements over wolf systematics, recent hybridization can be unambiguously detected by comparing the exact haplotypes (i.e., not cladistically associated haplotypes) exhibited by sympatric wolves and coyotes. Both mtDNA control region sequences and Y-chromosome microsatellites have sufficiently rapid mutation rates that individuals sharing the same types can be confidently assumed to share a relatively recent common ancestor. We employed this approach using three different kinds of genetic markers: mitochondrial DNA control region sequence, seven Y-chromosome microsatellite loci and eight autosomal microsatellite loci and assessed genetic variability, connectivity and hybridization among recovered wolves as follows. (1) Did colonizing wolves in Wisconsin and UpperPeninsula Michigan originate from a single founder source in Minnesota? (2) Has the recovery goal of maintaining Minnesota and Wisconsin/Upper-Peninsula Michigan as separate populations been achieved? (3) How does genetic variation among naturally recovered WGL states wolves compare to long-standing wolf populations and to naturally recovered and managed recovery populations in the northern Rocky Mountains? (4) Are WGL states wolves Canis lupus or Canis lycaon and do they exhibit evidence of ongoing hybridization with coyotes or domestic dogs (Canis familiaris)? An alarming number of wolf–dog hybrids have been documented roaming within Western Great Lakes wolf range (Wisconsin Department of Natural Resources (WIDNR) 1999) and there is concern that they may have hybridized with dispersing wolves from Minnesota, especially during the early years of the recovery program (Gottelli et al. 1994; Randi and Lucchini 2002; Mun˜oz-Fuentes et al. 2009b). (5) Finally, was the taxonomic provenance

123

of WGL wolves prior to their extirpation the same as the population of wolves residing there today?

Methods Samples Wolf biologists and veterinarians collected blood or tissue from 124 wolves and 132 coyotes from the study area in northeast Minnesota (MN; 42 wolves), northern Wisconsin (WI; 65 wolves and 132 coyotes) and Upper-Peninsula Michigan (UPMI; 17 wolves) Sampling began in 1990 in MN and UPMI and continued annually through 1999 in MN and 2000 in UPMI. WI wolves were sampled annually from 1991 to 2003, WI coyotes from 2001 to 2003. In order to provide additional scale to genetic admixture assessments, the analysis included gray wolves from Alaska (39 wolves), British Columbia (41 wolves) and Alberta (26 wolves) as well as 25 domestic dogs and 14 wolf–dog hybrids. The dog samples were from California, Florida, Illinois, Kentucky, Minnesota, Oregon and Texas and included pure-bred Akita, Bullmastiff, Collie, Corgi, German Shepherd, Golden Retriever, Great Pyrenees, Malamute, Rottweiler and Weimaraner, as well as mixedbreed individuals. Wolves sampled from the WGL states, British Columbia and Alberta as well as most wolf–dog hybrid samples were identified morphologically after the criteria in Mech (1974) by Paula Holahan (mammalogist, University of Wisconsin), Nancy J. Thomas (veterinarian pathologist, Wisconsin Dept of Natural Resources), Thomas Cooley (veterinarian pathologist, Michigan Dept of Natural Resources) and Mark R. Johnson (wildlife veterinarian, Global Wildlife Resources, Inc.). Some wolf–dog hybrid samples were from animals of known pedigree. The capture locations of the wolf and coyote samples from the WGL study area are presented in Fig. 1. The coyote samples used in this study were all from Wisconsin and were identified morphologically after the criteria in Bekoff (1977) by Dorothy Ginnett (wildlife ecologist, University of Wisconsin) and Eizabeth S. Williams (veterinary pathologist, University of Wyoming). Capture locations were only known to county of origin (Fig. 1). Total cellular DNA was prepared after manufacturer recommendations (Qiagen; DNeasy 96 Tissue; Cat #69582). Mitochondrial DNA sequencing Using the PCR and DNA sequencing reaction conditions and cycling profiles listed in Fain et al. (2000), a sequence containing 224 bp of the mtDNA-control region (CR) was obtained from each of 206 wolves and 132 coyotes with the primers L15188 (50 -ACATGAATTGGAGGACAACCA

Conserv Genet Fig. 1 Map of sampling locations of wolves (black dots) and Wisconsin coyotes (counties of origin are shaded) in the western Great Lakes study area. Sampling localities are abbreviated as northeast Minnesota (MN), north-central Wisconsin (WI) and UpperPeninsula Michigan (UPMI)

GT-30 ) and H15662 (50 AAGCCCTTATTGGACTAAGTG30 ) and internal sequencing primer L15368 (50 -GGTCTT GTAAACCAAAAATGG-30 ). Fifty-eight individuals representing the unique wolf haplotypes observed in this study were sequenced over 494 bp of hypervariable region I (HVI) to determine if sub-types could be identified. The HVI segment was amplified with primers L15188 and H16106 (50 -AAACTATATGTCCTGAAACC-30 ) and 494 bp were sequenced in both directions with primers L15368 and H16106 (amplification extension time increased to 1.5 min). All sequencing products were purified by gel filtration and characterized on an automated sequencer (ABI 3130xl). Negative controls for DNA extraction and PCR amplification reagents were included. The wolf and coyote mtDNA-CR sequences obtained in this study have been archived in GenBank (wolf accession numbers: FJ213912–FJ213916, GU647049– GU647053; coyote sequence accession numbers: FJ213917– FJ213930). Microsatellite typing Seven Y-chromosome-linked microsatellite loci (Table 1; Sundqvist et al. 2001; Bannasch et al. 2005) and eight autosomal microsatellite loci (Appendix 1; Ostrander et al. 1993), originally developed from domestic dog, were used in this study. All microsatellite amplifications were performed as in Ostrander et al. (1993), but with an annealing temperature of 58°C and a final extension at 60°C for 30 min. Allele sizes were determined with the internal lane standard CXR (Promega, Cat #DG6221) on an automated sequencer (ABI 3130xl) with GeneMapperÒ software.

Genetic analysis Haplotypes were constructed from the allele scores at the seven Y-chromosome microsatellite loci and the mtDNACR sequences. To facilitate the direct comparison of the results of this study to the different mtDNA-CR sequence lengths presented in the wolf literature (i.e., 224–425 bp, Table 2), the 224 bp sequence was used for all calculations except the phylogenetic analysis. Phylogenetic relationships between wolf and coyote mtDNA-CR haplotypes were estimated from the 494 bp sequences with the neighborjoining (NEIGHBOR) subroutine of the Phylogeny Inference Package (PHYLIP; version 3.57; Felsenstein 1989, 1995). Phylogeny reconstructions were evaluated with 1000 bootstrap resamplings (DNABOOT; PHYLIP) and a single consensus tree (CONSENSE; PHYLIP). Tree files were viewed using TREEVIEW (version 2.0 Page 1996). Phylogenetic relationships among wolf and coyote Y-chromosome haplotypes were examined with the Median-Joining-network method (Bandelt et al. 1999) as implemented in Network 4.5 (http://www.fluxus-engineering.com). To simplify the coyote complement of the resulting networks, only haplotypes observed more than once were used in the analysis. Unique WGL states wolf Y-chromosome types were given frequencies of two to prevent them from being excluded. In accordance with the phylogenetic results obtained in this study, as well as previous taxonomic assignments, mtDNA (Wilson et al. 2000; Wilson et al. 2003) and Y-chromosome haplotypes (Hailer and Leonard 2008), were coded as Canis lupus or C. lycaon for frequency calculations. We performed genetic admixture analysis on western wolves, WGL states wolves, WI coyotes and dogs and

123

Conserv Genet Table 1 Haplotype frequencies for seven Y-chromosome microsatellite loci Frequency of Y-STR haplotypes MN

WI

UPMI

FWSCluA

0.1

0.061 –

FWSCluB

–

–

–

FWSCluC FWSCluD

– 0.034

– – 0.025 –

FWSClyE

0.207

0.367 0.500

Y-STR haplotype by allele size

WGL AK

AB

BC

West

Coy

MS MS MS 34.5 650 650 MS 79.2Ab 79.2Bb 34Aa 34Ba 41Aa 41Ba

0.068

–

–

–

–

–

172

178

208

216

122

116

128

–

–

0.133 0.048 –

172

184

208

220

122

116

128

0.023

– –

– –

0.067 0.024 – – 0.071 –

172 172

178 180

208 208

212 222

122 122

116 116

128 128

0.330

–

–

–

170

182

212

224

130

120

130

–

–

FWSCluF

0.034

0.102 –

0.068

0.167

0.067 –

0.071 –

172

178

208

224

122

116

128

FWSCluFF

–

–

–

–

–

–

0.067 0.024 –

172

178

208

218

122

116

130

FWSCluG

–

–

–

–

0.333

–

–

0.095 –

172

178

208

218

122

116

128

FWSCluI

–

–

–

–

0.083

0.800 0.333 0.429 –

172

176

208

214

122

116

128

FWSCluJ

–

0.102 –

0.068

–

0.133 0.200 0.119 –

172

180

208

220

122

116

128

FWSCluL

–

–

–

–

0.083

–

176

178

208

212

124

116

130

FWSCluM

–

–

–

–

0.250

–

–

0.071 –

172

180

208

212

122

116

132

FWSClyO

–

0.041 –

0.023

–

–

–

–

–

170

182

212

224

128

120

130

FWSClyR

0.103

0.122 0.100

0.102

–

–

–

–

172

180

212

212

130

120

126

FWSCluU

0.138

0.122 –

0.114

0.083

–

0.200 0.095 –

172

178

208

216

122

116

130

FWSCluW

–

0.025 –

0.011

–

–

–

–

–

172

178

208

222

122

116

128

FWSClyX

0.034

–

0.011

–

–

–

–

–

170

182

212

224

122

116

130

FWSClyY FWSCluZ

0.0.69 0.276

– – 0.041 0.400

0.023 0.160

– –

– –

– –

– –

– –

170 172

182 178

212 208

224 214

122 122

116 116

128 128

–

–

0.024 –

FWSClaCCC

–

–

–

–

–

–

–

–

0.025 172

178

212

222

132

120

126

FWSClaCDD

–

–

–

–

–

–

–

–

0.025 172

178

212

222

128

120

126

FWSClaCEE

–

–

–

–

–

–

–

–

0.025 170

180

210

222

130

120

130

FWSClaCFF

–

–

–

–

–

–

–

–

0.025 176

180

210

222

130

120

126

FWSClaCG

–

–

–

–

–

–

–

–

0.075 172

178

210

214

130

120

126

FWSClaCGG

–

–

–

–

–

–

–

–

0.050 170

182

210

222

130

120

130

FWSClaCH

–

–

–

–

–

–

–

–

0.025 172

178

212

214

130

120

128

FWSClaCHH

–

–

–

–

–

–

–

–

0.075 170

182

210

222

128

120

126

FWSClaCII

–

–

–

–

–

–

–

–

0.025 170

182

210

222

130

120

128

FWSClaCJ

–

–

–

–

–

–

–

–

0.025 172

178

210

214

132

120

126

FWSClaCKK

–

–

–

–

–

–

–

–

0.025 180

182

210

222

128

120

126

FWSClaCL

–

–

–

–

–

–

–

–

0.025 172

178

212

214

124

116

130

FWSClaCLL

–

–

–

–

–

–

–

–

0.025 174

178

208

224

128

120

126

FWSClaCM

–

–

–

–

–

–

–

–

0.025 176

178

210

214

122

116

130

FWSClaCMM FWSClaCN

– –

– –

– –

– –

– –

– –

– –

– –

0.025 174 0.050 172

178 178

208 212

224 212

130 130

120 120

130 126

FWSClaCO

–

–

–

–

–

–

–

–

0.050 172

180

210

214

126

120

126

FWSClaCQ

–

–

–

–

–

–

–

–

0.025 172

178

212

216

130

120

132

FWSClaCQQ

–

–

–

–

–

–

–

–

0.050 172

178

212

214

130

120

130

FWSClaCR

–

–

–

–

–

–

–

–

0.025 174

178

212

216

130

120

132

FWSClaCS

–

–

–

–

–

–

–

–

0.025 172

180

210

216

128

120

126

FWSClaCSS

–

–

–

–

–

–

–

–

0.025 172

180

210

214

132

120

126

FWSClaCU

–

–

–

–

–

–

–

–

0.050 172

176

210

218

130

120

126

FWSClaCV

–

–

–

–

–

–

–

–

0.025 172

176

210

218

128

120

126

FWSClaCW

–

–

–

–

–

–

–

–

0.025 172

178

210

218

126

120

126

FWSClaCWW –

–

–

–

–

–

–

–

0.025 172

178

208

224

128

120

126

FWSClaCX

–

–

–

–

–

–

–

0.025 172

178

208

224

122

116

130

123

–

Conserv Genet Table 1 continued Frequency of Y-STR haplotypes

Y-STR haplotype by allele size MS MS MS 34.5 650 650 MS 79.2Ab 79.2Bb 34Aa 34Ba 41Aa 41Ba

MN

WI

UPMI

WGL AK

AB

BC

West

Coy

FWSClaCXX

–

–

–

–

–

–

–

–

0.025 172

182

210

212

FWSClaCY

–

–

–

–

–

–

–

–

0.050 172

178

212

218

130

120

126

FWSClaCZ

–

–

–

–

–

–

–

–

0.025 172

180

210

218

128

120

126

Allele numbers correspond to base-pair size and are calibrated to Sundqvist et al. (2001). Sampling localities are abbreviated as northeast Minnesota (MN), north-central Wisconsin (WI), Upper-Peninsula Michigan (UPMI), Gates of the Arctic NP, Alaska (AK), Hinton, Alberta (AB), Fort St Johns, British Columbia (BC), Wisconsin Coyotes (Coy), combined frequency of wolves from MN, WI and MI (WGL) and Alaska, Alberta and British Columbia (West) a For primer sequence information and locus description please see Sundqvist et al. (2001) b

Locus 650 79.2 amplifies a duplicated region on wolf, dog and coyote Y chromosomes and generate two polymorphic PCR products labeled in this study as 650 79.2A and 650 79.2B. For primer sequence information and locus description please see Bannasch et a.l (2005)

wolf–dog hybrids with the Bayesian clustering program STRUCTURE (version 2, Pritchard et al. 2000). The population admixture model of STRUCTURE with independent allele frequencies was used to infer population division within a sample as sets of genetic clusters of greatest similarity (K value). STRUCTURE also calculated the proportion of ancestry (qi) of the individual genotypes in the sample with respect to each of the identified population clusters. The number of populations represented by the data was determined by running STRUCTURE for K = 1–12 with five repetitions of 100,000 iterations following a burn-in period of 10,000 iterations. The respective probabilities calculated for each K were averaged over the five runs at each K. We used the maximal value of LnP(D) (Pritchard et al. 2000), DK (Evanno et al. 2005) and overall individual ancestry assignments as criteria to infer the number of different populations represented by the data. Taking this into account, the data was run through STRUCTURE ten times at the selected K value with burnin set at 100,000 and 1,000,000 iterations. Individual admixture proportions (i.e., qi values) were taken from the run with the highest probability and lowest variance. The STRUCTURE inferred reference canid and WGL states wolf populations were tested for departures from genotypic linkage equilibrium and Hardy–Weinberg Equilibrium (HWE) in GENEPOP (vers. 3.4, Raymond and Rousset 1995). The linkage and HWE results were adjusted for multiple tests using the Bonferroni correction (Rice 1989; Sacks et al. 2004). Pairwise estimates of FST (Weir and Cockerham 1984) were calculated in ARLEQUIN for the populations inferred from the data set by STRUCTURE. Private allele estimates of gene flow among the WGL states wolf and coyote sampling localities were calculated with GENEPOP as the effective number of migrants per generation (Nem, Slatkin 1985). Allelic richness (i.e., number of alleles independent of sample size, El

Mousadik and Petit 1996), inbreeding coefficient FIS (Weir and Cockerham 1984) and average gene diversity (Nei 1987) were calculated with the program FSTAT (Goudet 1995, 2001). Mitochondrial DNA and Y-chromosome haplotype diversities (Nei 1973) were calculated for wolf and coyote populations with the program ARLEQUIN (version 2.0, Schneider et al. 2000). The significance of statistical differences between populations was evaluated by comparing 95% confidence intervals. The frequency distributions of the autosomal microsatellite alleles observed in WGL states wolves were compared to western wolves, coyote, dogs and wolf–dog hybrids (Appendix 1). Comparisons included the autosomal microsatellite data from Roy et al. (1994) for 20 Minnesota coyotes and from Garcia-Moreno et al. (1996) for 42 domestic dogs. These studies used the same loci employed herein and we calibrated our data by characterizing samples common to Roy et al. (1994); Forbes and Boyd (1996, 1997) and Garcia-Moreno et al. (1996). Recolonized WGL states wolves were assessed for evidence of genetic bottleneck effects with the program BOTTLENECK (Piry et al. 1999). The test characterizes a population with respect to differences in observed and expected heterozygosities from the number of alleles observed at microsatellite loci. We used BOTTLENECK with both the stepwise mutation model (SMM, Valdes et al. 1993) and the two-phase model of mutation (TPM, Di Rienzo et al. 1994; Primmer et al. 1998) with multistep mutations accounting for 30% of all mutations in the TPM application and 10,000 replications. The significance of calculated heterozygosity excess or deficiency was evaluated with a one-tailed Wilcoxon sign-rank test (Luikart and Cornuet 1998). Hybrid WGL states wolves were defined as any individual that exhibited mixed ancestry (i.e., qWGL \0.7 and qClupus/Coy/Dog [0.1; Randi and Lucchini 2002). We deduced the taxonomic source and degree of introgression

123

123

FJ213915

FJ213916

FJ213924

FJ213912

FJ213913 GU647049

GU647050

GU647051

GU647052

GU647053

FWSCly21

FWSCla31

FWSClu06

FWSClu07 FWSClu01

FWSClu02

FWSClu03

FWSClu04

FWSClu05

–

–

–

W23

W20 W24

W22

–

–

–

–

–

–

AF005312

AF005308 AF005312

AF005309

–

–

–

–

–

–

–

258

257 258

257

–

–

–

–

Size

lu38

lu37

lu61

–

lu28 –

lu32

–

–

–

–

Type

– 406 401 425

– FM201760a FN298189b

257 –

257

–

–

–

–

Size

FM201767a

AF005308 –

AF005309

–

–

–

–

Acc#

Vila and Wayne (1999)

–

–

–

–

C23 –

C22

–

C3

C13

C1

Type

–

–

–

–

FJ687609 –

FJ687608

–

AY267720

AY267730

AY267718

Acc#

–

–

–

–

228 –

228

–

224

224

224

Size

Wilson et al. (2000, 2003)

–

–

–

–

– –

–

la76

–

–

–

Type

–

–

–

–

– –

–

FM209421

–

–

–

Acc#

–

–

–

–

– –

–

396

–

–

–

Size

Hailer and Leonard (2008)

–

–

–

–

– –

–

–

GL2

GL10

GL1

Type

328 328

N/Ac N/Ac

–

–

–

–

– –

–

–

–

–

–

– –

–

–

328

N/Ac

–

Size

Acc#

Leonard and Wayne (2008b)

c

b

Personal communication

Mun˜oz-Fuentes et al. (2009a, b) originally described as AF812731 by Vila and Wayne (1999)

The various nomenclatures used in earlier studies to describe these haplotypes are also presented. The descriptors are haplotype name (Type), GenBank accession number (Acc#) and sequence length in base pairs (Size) a Mun˜oz-Fuentes et al. (2009a, b) originally described as AF812741 and AF812730 by Vila and Wayne (1999)

494

494

494

494

494 494

494

494

494

494

–

FWSCly12

494

FJ213914

FWSCly04

Acc#

Type

Acc#

Type

Size

Vila et al. (1997)

This study

Table 2 Control region haplotypes observed in this study in western and WGL wolves, as well as the coyote haplotype FWSCla31 discussed in the text

Conserv Genet

Conserv Genet

in the genetic ancestries of WGL states wolves by combining the admixture results from the autosomal microsatellite genotypes with the population/species associations determined for mtDNA and Y-chromosome lineage markers and the inferred origins of private alleles.

Results Mitochondrial DNA sequence haplotypes Ten different mtDNA-CR sequence haplotypes were observed among the 206 wolves characterized in this study whereas the 132 WI coyotes exhibited 22. The 224 bp segment of the wolf mtDNA-CR encompassed 71% of the variation observed in the 494 bp HVI sequences. None of the wolf haplotypes derived from the 224 bp segment were further sub-typed by considering the larger 494 bp sequences. WGL states wolves exhibited five haplotypes, western gray wolves seven. Haplotypes FWSClu06, FWSCly12 and FWSCly21 occurred across the WGL study area (Fig. 2). Haplotype diversity was lowest among WGL states wolves and highest among WI coyotes (Table 3). Only 34% of wolves in the WGL sample exhibited gray wolf mtDNA-CR haplotypes FWSClu06 and FWSClu07, 50% exhibited the ‘‘coyote-like’’ haplotype FWSCly12 and 16% exhibited C. lycaon haplotypes FWSCly04 and FWSCly21 (identical to published C. lycaon sequences; Wilson et al. 2000, 2003). Neither of the FWSCly04, FWSCly12 or FWSCly21 haplotypes was observed among western gray wolves or WI coyotes; all three have been reported in wolves from Ontario (Wilson et al. 2003; Grewal et al. 2004). The mtDNA haplotypes of WGL states wolves, western gray wolves and WI coyotes descend from three different phylogenetic clades (Fig. 3). The FWSClu06 and FWSClu07 sequences, observed among both western gray wolves and WGL states wolves, contributed to a strongly supported gray wolf clade (93% bootstrap value) while the FWSCly12 sequence nested well within the coyote clade. The FWSCly04 and FWSCly21 C. lycaon haplotypes were included in the third clade (100% bootstrap support) which was intermediate between gray wolves and coyotes. Haplotypes in the same clade differed from each other by 0.04–2.4%; gray wolf and C. lycaon haplotypes differed by 6.1–6.5% and the C. lycaon and coyote clades differed by 1.7–3.5%. Y-chromosome microsatellite haplotypes Twelve different Y-chromosome microsatellite haplotypes were observed among male WGL states wolves, 11 in gray

wolves and 30 in WI coyotes (Table 1). Gray wolves and WGL states wolves shared four haplotypes, all wolf and coyote haplotypes were exclusive. Colonizing male wolves in WI exhibited three Y-chromosome haplotypes (i.e., FWSCluJ, FWSClyO and FWSCluW) not observed in MN or UPMI and haplotypes FWSClyE, FWSClyR and FWSCluZ occurred across the study area. Haplotype diversity was significantly higher among WGL states wolves than western gray wolves but was highest among WI coyotes (Table 4). Seven of the 19 Y-chromosome haplotypes observed in wolves in this study have been observed previously among gray wolves (i.e., FWSCluD, FWSCluF, FWSCluI, FWSCluJ, FWSCluL, FWSCluU, FWSCluZ; Sundqvist et al. 2006; Musiani et al. 2007; Hailer and Leonard 2008). Five haplotypes (i.e., FWSClyE, FWSClyO, FWSClyR, FWSClyY, FWSClyX) were unique to WGL states wolves and seven haplotypes (i.e., FWSCluB, FWSCluC, FWSCluFF, FWSCluG, FWSCluI, FWSCluL, FWSCluM) were unique to western gray wolves. Similar to the phylogenetic analysis of mtDNA, the Median-Joining-network (Fig. 4) divided the Y-chromosome haplotypes of gray wolves, WGL states wolves and WI coyotes into three distinct groups. All of the gray wolf haplotypes FWSCluA, FWSCluB, FWSCluC, FWSCluD, FWSCluF, FWSCluFF, FWSCluG, FWSCluI, FWSCluJ, FWSCluL, FWSCluM, FWSCluU, FWSCluW and FWSCluZ grouped together, while the unique WGL states wolf haplotypes FWSClyE, FWSClyO, FWSClyY and FWSClyX sorted into a second group intermediate to coyotes and gray wolves and consistent with a C. lycaon origin (Shami 2002). Lastly, the haplotype FWSClyR was observed exclusively among WGL states wolves, but nested in the third group among the coyote haplotypes. Half of the male wolves in the WGL states sample exhibited gray wolf Y-chromosome haplotypes and 40% exhibited the unique C. lycaon haplotypes FWSClyE, FWSClyO, FWSClyY and FWSClyX. The unique ‘‘coyote-like’’ haplotype FWSClyR, was observed in 10% of male WGL states wolves (Table 1). Autosomal microsatellite genotypes Analysis with STRUCTURE found that the selected canid reference groups and WGL states wolves described five different genetic populations. The large DK value observed at K = 2 (Fig. 5) corresponded to the segregation of western wolves from WGL wolves and coyotes (Fig. 6; the same result was obtained without the inclusion of dogs). The smaller DK peak at K = 5 combined with the high LnP(D) (Fig. 5) and high ancestry assignment values observed for individual animals at K = 5 identified it as the highest level of subdivision supported by the data. Likelihood at K = 6 and K = 7 was higher than at K = 5, but these divisions were not supported by DK (Fig. 5b).

123

Conserv Genet Fig. 2 Map of the WGL wolf sampling locations color-coded according to the species type of the identified mtDNA CR haplotype. Counties of origin of Wisconsin coyotes are shaded. C. lupus types (FWSClu06 and FWSClu07 red dots), C. lycaon types (FWSCly04 and FWSCly21, green dots) and C. lycaon type (FWSCly12, blue dots)

Table 3 Mitochondrial DNA control region haplotype diversity in western Great Lakes wolves, western wolves and WI coyotes Sample

WGL

116

5

0.624 (0.005)

WW

90

7

0.794 (0.004)

132

22

0.919 (0.002)

WI coyote

No. types

CR Dh (95% CI)

Population

Dh haplotype diversity, CI confidence interval

Additionally, although wolves are continuously distributed over the *600 km separating the Alberta and British Columbia samples (Forbes and Boyd 1997), they were separated into different clusters at K = 6. Significant gene flow has been demonstrated between these localities (Forbes and Boyd 1997) and justifying population structure under such conditions is questionable as STRUCTURE is known to have difficulty making assignments with data of this sort (Pritchard and Wen 2003; Schwartz and McKelvey 2009). Moreover, at K = 7, the C. lupus cluster of WGL wolves was subdivided into two different clusters even though no individuals were strongly assigned to either subgroup, all individuals were highly admixed, and the proportions of the sample assigned to each subgroup were symmetrical. Similar results were obtained when WGL wolves were analyzed alone. Taken together, these results demonstrate the lack of real biologically sensible population structure in the data beyond K = 5 (Pritchard and Wen 2003). At K = 3, WGL states wolves separated from WI coyotes, at K = 4, WI coyotes separated from dogs. The

123

subdivision of WGL wolves from dogs and Wisconsin coyotes at K = 3 was unexpected given the shared phylogenetic history of wolves and dogs. The program STRUCTURE uses the assumption of HWE to assign individuals to likely source populations and it is not understood how significant departures from HWE, such as results from the inbreeding of domestic dogs (e.g., four of eight loci were out of HWE in this study), effects the reliability of population assignment (Pritchard and Wen 2003; Verardi et al. 2006). Pairwise FST estimates of the divergence between the inferred populations (Table 5) clearly distinguished western gray wolves, WGL states wolves, coyotes and dogs (FST = 0.123–0.206), but WGL states wolves were somewhat less differentiated from WI coyotes (FST = 0.159) than were western gray wolves (FST = 0.206). The same approach revealed high gene flow and little differentiation among WGL state locality samples (FST = 0.006–0.016). The inferred populations departed from HWE at 1–4 loci and from linkage equilibrium expectations at 0–3 locus pairs. The WGL states population was out of HWE at one locus and one locus pair was out of linkage equilibrium. Such modest disequilibria could result from localized inbreeding, over-sampling of related individuals or introgression from a population with different allele frequencies. Consistent with the latter, all but two of the 44 alleles observed in MN wolves (95%) were also observed among WI/UPMI colonizing animals, whereas, colonizing wolves exhibited 13 private alleles. We detected significant heterozygosity deficiency among WGL states wolves at seven

Conserv Genet Fig. 3 Neighbor-joining tree of mitochondrial control region sequences from the western gray wolves (C. lupus), western Great Lakes wolves (C. lupus and C. lycaon), and Wisconsin coyotes (C. latrans) characterized in this study. A red wolf (C. rufus) sequence (AY280913; Adams et al. 2003) and a golden jackal (C. aureus) sequence (AY289996; Aggarwal et al. 2007) were included as references. Selected coyote haplotypes were removed from this analysis for clarity. All of the clades identified in the complete analysis are represented

Table 4 Y-chromosome STR haplotype diversity in western Great Lakes wolves, western wolves and WI coyotes Population

Sample

No. types

Y Dh (95% CI)

WGL

61

12

0.851 (0.006)

WW WI coyote

37 39

11 30

0.742 (0.022) 0.984 (0.003)

Dh, haplotype diversity, CI confidence interval

of eight loci tested under the SMM (P = 0.006). Neither heterozygosity excess nor deficiency was detected under the TPM. Estimates of heterozygote deficiency FIS, the inbreeding coefficient, ranged from 0.04 in WI coyotes to 0.08 in western wolves and was significantly positive among colonizing wolves from WGL––consistent with increased inbreeding (Table 6). Average gene diversity was similar across all populations, and allelic richness ranged from 5.1 in western wolves to 7.2 in WGL states wolves (Table 6).

Hybridization among WGL states wolves Significant admixture was detected (qWGL \0.7 and qClupus/ Coy/Dog [0.1) in 38% of WGL states wolves. Hybrids occurred at roughly similar frequency (35–39%) in the three states within the study area. Twenty-five percent were C. lupus–C. lycaon hybrids while wolf–dog hybrids and C. lupus–C. lycaon–C. latrans hybrids each made up about 6% of the population. In addition to significant dog ancestry (qDog = 0.1–0.7), four wolf–dog hybrids were admixed WGL wolves (qWGL = 0.4–0.6). The genotypes of the three remaining wolf–dog hybrids were predominantly western gray wolf (qClupus = 0.7) or dog (qDog = 0.7). Together, these individuals presented five private alleles, three of which were common among dogs. Domestic dog lineage markers were absent. Even though the STRUCTURE assignments of WGL states wolves to the WI coyote cluster were low (96% with qCoy B0.09), seven individuals were intermediate to both coyote

123

Conserv Genet

Fig. 4 Median-Joining network based on Y-chromosome microsatellite repeat variation for western wolves, WGL states wolves and WI coyotes. Cross-hatches represent mutational events. Haplotypes are lettered next to circles and correspond to designations in Table 1. Haplotypes are color-coded by species association (Green = C. lycaon; Yellow = C. latrans; Orange = C. lupus). Overall sizes of circles are proportional to the number of individuals with the haplotype. Red circles are median vectors

(qCoy = 0.1–0.4) and wolf clusters (qWGL/Clupus = 0.1–0.7). These results were consistent with direct hybridization with coyote, but other data were contradictory. All WGL states wolf and WI coyote mtDNA and Y-chromosome lineage markers were mutually exclusive; nuclear gene flow between sympatric WGL states wolves and coyotes was low (FST = 0.159; Nem = 0.3–0.7); and STRUCTURE analysis did not detect introgression from WGL states wolves into WI coyotes (100% with qWGL B0.1). Nonetheless, seven WGL states wolves exhibited detectable coyote ancestry. Six were C. lycaon–C. latrans hybrids, one a C. lupus–C. lycaon–C. latrans hybrid. Five of the private alleles observed among WGL colonizers were consistent with coyote origin. Only two WI coyotes exhibited ‘‘wolf-like’’ Y-chromosome haplotypes, FWSClaCL and FWSClaCM were most similar to the widely occurring gray wolf haplotype FWSCluU (equals H33, Hailer and Leonard 2008; Fig. 4). These Y-chromosome results were inconsistent with ongoing or even recent introgression into sympatric coyotes from male WGL states wolves.

123

Fig. 5 Bayesian clustering analysis of western wolves, WGL states wolves, WI coyotes, dogs and wolf–dog hybrids for determination of K values. a Mean Ln probability for five runs at each of 1–12 clusters (K). b DK rate of change of Ln probability between successive K values from the data in Fig. 5a. The modal value of the distribution is the uppermost level of structure

Discussion Ancestry of recovered WGL states wolves Concordant results from autosomal microsatellite, mtDNA control region sequence and Y-chromosome microsatellite characterizations found the recovered WGL states population to be comprised of C. lupus, C. lycaon and their hybrids. Phylogenetic analysis of mtDNA-CR sequence and Y-chromosome microsatellite haplotypes support the recognition (Wilson et al. 2000; 2003; Kyle et al. 2006) of C. lycaon as a unique species of North American wolf. Contrary to the findings of Leonard and Wayne (2008b) and Koblmu¨ller et al. (2009), we did not find WGL states wolves to exhibit either coyote mtDNA or Y-chromosome haplotypes. Two WI coyotes exhibited ‘‘wolf-like’’ Y-chromosome haplotypes, but neither was observed among WGL states wolves or western gray wolves. These observations are consistent with the absence of coyote mtDNA haplotypes among WGL states wolves as the proliferation of ‘‘coyote-like’’ mtDNA among eastern wolves has long been explained (Lehman et al. 1991; Roy et al. 1994; Leonard and Wayne 2008b) by the backcrossing into wolves of hybrid females from directional matings of male gray wolves with female coyotes. Analysis of autosomal microsatellite variation also demonstrated

Conserv Genet Fig. 6 Bar plots of individual proportional assignments to K = 2–5 clusters inferred by STRUCTURE. In the K = 5 cluster plot, green bars represent likelihood of assignment to C. lycaon, pink bars to C. lupus. Sample names are abbreviated as Gates of the Arctic NP, Alaska (AK); Fort St Johns, British Columbia (BC); Hinton, Alberta (AB); northeast Minnesota (MN); north-central Wisconsin (WI); UpperPeninsula Michigan (UPMI); domestic dog (DOG); wolf–dog hybrids (WOLF–DOG); Wisconsin Coyotes (COY)

Table 5 Pairwise estimates of FST between groups assigned to populations inferred (Q [0.75) by STRUCTURE Population

WI

UPMI

WGL

WW

WI coyote

Dogs

MN

0.006

0.009

na

0.133

0.149

0.117

0.016

na

0.118

0.167

na

0.139 0.125

WI UPMI WGL WW WI Coyote

Table 6 Allelic richness (A), gene diversity (H) and inbreeding coefficient (Fis) in western Great Lakes wolves, western wolves, WI coyotes and dogs and wolf–dog hybrids Population

Sample

A

H

FIS

0.124

WGL

112

7.2 (0.13)

0.67 (0.07)

0.06 (0.01)

0.154

0.142

WW

103

5.1 (0.11)

0.68 (0.07)

0.08 (0.01)

0.159

0.123

WI coyote

36

6.5 (0.30)

0.69 (0.12)

0.04 (0.02)

0.146

Dogs

39

5.9 (0.19)

0.67 (0.11)

0.19 (0.02)

0.157

95% confidence intervals are in parentheses

0.206

The WGL population sampling localities (MN, WI, UPMI) are included separately to demonstrate gene flow across the study region

that WGL states wolves and sympatric coyotes were highly differentiated. WGL states wolves are reputed to be subject to ongoing gray wolf–coyote hybridization, largely due to the common occurrence in the recovered population of the ‘‘coyotelike’’ mtDNA-CR haplotype FWSCly12 (Fig. 2; Leonard and Wayne 2008b; Koblmu¨ller et al. 2009; but see Wheeldon and White 2009). We did not find FWSCly12 among 132 sympatric Wisconsin coyotes nor in any other coyote (comparison included 44 additional coyotes from

six western US states as well as comprehensive searches of the coyote sequence archive in GenBank). In addition, haplotype FWSCly12 has been found in museum specimens of eastern wolves collected in New York and Minnesota before these locations had been invaded by western coyotes (i.e., circa 1900, Wilson et al. 2003; Wheeldon and White 2009). These observations and the wide distribution of FWSCly12 among modern eastern wolves from the WGL states, Ontario and Quebec (Grewal et al. 2004; Koblmu¨ller et al. 2009) along with its absence in non-hybridizing coyotes (Pilgrim et al. 1998; Wilson et al. 2003; Hailer and Leonard 2008) argue for an ancient

123

Conserv Genet

hybridization event (Wheeldon and White 2009). This explanation is also supported by phylogenetic analysis. The neighbor-joining tree in Fig. 3 identified the WI coyote mtDNA-CR haplotype FWSCla31 as being most similar to FWSCly12. Haplotype FWSCla31 was the most commonly observed haplotype among WI coyotes (i.e., 17% of the sample) and is widely distributed from Nebraska (Hailer and Leonard 2008) to Maine (Lance et al. 2008, GenBank Acc.# EF508172, direct submission). Haplotypes FWSCly12 and FWSCla31 differ by two substitutions over 490 bp of control region sequence (i.e., one substitution in each line). Casting back to the ancestral coyote haplotype, a mutation rate of 3.7 9 10-8 substitutions/site/year (Leonard and Wayne 2008b) places the original hybridization event some 55,000 years ago when the Laurentide ice sheet would have driven eastern wolves to southern refugia and increased contact with coyotes (Nowak 1983). Although of coyote origin, haplotype FWSCly12 has long been transmitted as a C. lycaon lineage marker (Wilson et al. 2003; Wheeldon and White 2009). The ‘‘coyote-like’’ Y-chromosome haplotype FWSClyR parallels the phylogenetic and demographic features of FWSCly12, but on the paternally inherited Y-chromosome (Fig. 4). Less is understood about the mutation rate of canid Y-chromosome microsatellites (Jobling et al. 1999, Hellborg and Ellegren 2004, Natanaelsson et al. 2006), but we are convinced that Y-chromosome haplotype FWSClyR is also a C. lycaon lineage marker and not the result of recent ongoing hybridization with coyotes. Genetic variability Minnesota was the primary source of founder stock for the re-establishment of wolves in Wisconsin and UpperPeninsula Michigan. The observations of field and radiotracking studies (Thiel and Hammill 1988; Wydeven et al. 1995; Mech et al. 1995) that found consistent exchange from MN to WI and WI to UPMI are reflected in the genetic composition of recovered WGL states wolves today. Recovered WGL states wolves constitute a single interbreeding population as exemplified by high gene flow across the study area (Table 5), however, as late as 1994, established packs had not yet formed in the eastern third of northern Wisconsin (Mladenoff et al. 1995). Future connectivity may be threatened by major north-south highways dividing wolf range (Mech et al. 1988), especially as they will be primary access points for wolf hunters following federal delisting. WGL states wolves underwent a severe demographic bottleneck and a corresponding founder effect of significant inbreeding was detected (Table 6). Nonetheless, overall genetic variability remained high. Heterozygosity was similar to that of long-standing wolf populations in

123

Table 7 Genetic variation in selected natural and recovered (recolonized/reintroduced) wolf populations Population

N

A

He

Reference

Hinton, Alberta (natural)

25

4.5

0.66

This study

NW Montana (recolonized)

59

4.4

0.65

Forbes and Boyd (1997)

Yellowstone NP (reintroduced)

90

5.1

0.66

Fain, unpublished data

NW Ontario (natural)

30

6.9

0.71

Wilson et al. (2009)

NE Minnesota (recolonized)

25

5.4

0.68

This study

112

7.2

0.68

This study

WGL (recolonized)

All data are from the same eight microsatellite loci used in this study N Sample size, A mean number of alleles per locus, He expected heterozygosity

western Alberta and northwest Ontario, Canada as well as to both naturally recovered and managed recovery populations in western Montana and northwestern Wyoming, USA (Table 7). Y-chromosome and mtDNA haplotype diversities were also similar to a long-standing wolf population from Northwest Territories, Canada (Musiani et al. 2007). These results suggest that even though dispersing Minnesota founders were genetically variable, they may not have been of adequate number, or sufficiently reproductive to prevent the occurrence of inbreeding in the recovered WGL states population. Immigration is the most likely cause of the heterozygosity deficiency observed in WGL states wolves. Admixture analysis found one in six WGL states wolves to be from outside the WGL study area or significantly admixed with non-WGL states wolf genes (see below). The influx of novel alleles from genetically divergent sources (e.g., C. lycaon–C. latrans hybrids, domestic wolves, wolf– dog hybrids) inflated expected heterozygosity without substantially effecting observed heterozygosity (Cornuet and Luikart 1996). Hybridization Our results support the conclusions of Mech and Federoff (2002) and Kyle et al. (2006) that C. lupus and C. lycaon hybridize in the wild. Forty-six percent of male WGL states wolves displayed both C. lupus and C. lycaon lineage markers. Forty-one percent of males exhibited C. lycaon mtDNA and Y-chromosome markers but only 13% exhibited C. lupus mtDNA and Y-chromosome markers. Fully 2/3 of the combined WGL states sample exhibited C. lycaon mtDNA and only 1/3 C. lupus mtDNA. Similar proportions were observed among wolves in northwestern Ontario (Wilson et al. 2009) which contains significant numbers of C. lupus–C. lycaon hybrids (Grewal et al. 2004; Wilson et al. 2009) and, as an extension of this core

Conserv Genet

population, WGL states wolves reflect similar admixture. Wheeldon and White (2009) reported that C. lupus–C. lycaon hybrids were distributed throughout northwestern Ontario into Manitoba, but Wilson et al. (2003) found a predominance of C. lupus mtDNA haplotypes among Manitoba wolves. The distinction we observed between Alberta wolves and WGL states wolves (Table 5; Fig. 6) is likely due to allele frequencies varying continuously over the large distance (i.e., *1500 km) separating the samples and not strict genetic isolation (Pritchard et al. 2000; Schwartz and McKelvey 2009). Wolves are continuously distributed from Alberta to Minnesota and significant isolation-bydistance has been demonstrated between these localities (Forbes and Boyd 1997). Even without barriers to gene flow, local genetic relatedness due to neighbor mating can result in significant genetic difference between distant localities. Sampling more localities would reveal the underlying continuum. This was observed in skull morphometrics of wolves sampled through this region (i.e., Jasper National Park, Alberta; Prince Albert National Park, Saskatchewan; Riding Mountain National Park, Manitoba and southwest Ontario) where size characters overlapped continuously from the largest animals in Alberta to the smallest animals in southwest Ontario (Skeel and Carbyn 1977). Some Alberta gray wolves showed evidence of admixture with WGL states wolf Cluster IV (but not with Cluster V, Table 8). Given the absence of C. lycaon mtDNA and Y-chromosome haplotypes among Alberta wolves (Fig. 4; Table 1), and that both Alberta and WGL wolves exhibited C. lupus mtDNA-CR haplotype FWSCluW06 and Ychromosome haplotype FWSCluJ, eastern C. lupus nubilus (Nowak 1995) was the likely source of the admixture in these western wolves (see pink bars in the K = 5 cluster of Fig. 6). Similarly, Cluster IV also represents the C. lupus nubilus contribution to WGL wolves whereas Cluster V depicts C. lycaon (Table 8). The high proportion of WGL ancestry (qWGL = 0.6– 0.7) and low proportion of coyote ancestry (qCoy = 0.1– 0.2) of most of the individuals exhibiting coyote admixture was consistent with their being the progeny of WGL states wolves and migrant C. lycaon–C. latrans hybrids from western and central Ontario (Kyle et al. 2006). Such hybrids were common in central Ontario, where two-thirds of the wolves in Algonquin Park exhibited five different coyote mtDNA-CR haplotypes (Grewal et al. 2004). However, one individual was a likely F1 hybrid and may have been a direct migrant. This male, taken in Mackinac Co., in far east Upper-Peninsula MI, was the only individual found to exhibit the C. lycaon mtDNA-CR haplotype FWSCly04 (Wilson et al. 2000; Leonard and Wayne 2008b).

Numerous genetic studies have identified wolf–dog hybrids in the wild (Randi and Lucchini 2002; Vila et al. 2003; Lucchini et al. 2004; Verardi et al. 2006; Mun˜ozFuentes et al. 2009b) and these observations extend to Western Great Lakes wolves (Wisconsin Department of Natural Resources (WIDNR) 1999). Three wolf–dog hybrid individuals presented high western C. lupus ancestry and western C. lupus lineage markers common among domestic hybrids. But the remaining four wolf–dog hybrid individuals presented significant WGL wolf ancestry and reduced dog ancestry and three of the four exhibited at least one WGL lineage marker. These individuals were probably the result of local domestic breeding and recent release, but multiple generations of back-crossing into wild WGL states wolves would give similar results. Historical WGL wolves Fifteen historical (i.e., 1899–1912) wolf samples from Minnesota, Wisconsin and Upper-Peninsula Michigan (Leonard and Wayne 2008b; Wheeldon and White 2009) all exhibited what we have interpreted to be C. lycaon mtDNA haplotypes and confirm that the WGL states were included in the historic range of C. lycaon (Kyle et al. 2006). Overall, 66% of the recovered WGL states wolves exhibited C. lycaon mtDNA. Seventy-nine percent of the historic samples exhibited the same C. lycaon haplotypes, but three haplotypes identified in historical WI/UPMI wolves (Leonard and Wayne 2008b) were absent from the recovered population. The haplotype frequencies of the historical samples do not track precisely with those observed among recovered WGL states wolves, but do establish that haplotypes common among pre-recovery wolves (FWSCLY12:13%, FWSCLY21:33%, FWSCLY04:33%), were, for the most part, common among recovered wolves (FWSCLY12:48%, FWSCLY21:21%, FWSCLY04:1%). The most significant difference between the recovered and historical populations was that gray wolf mtDNA was found in a third of the recovered wolves but not in any historical sample (Leonard and Wayne 2008b; Wheeldon and White 2009). Leonard and Wayne (2008b) attributed this to the eastward expansion of C. lupus into Great Lakes wolf range in northwest Ontario and Minnesota after the extirpation of the historical population, although Grewal et al. (2004) and Wilson et al. (2009) found C. lycaon to have expanded westward into Manitoba, northward in Ontario and eastward in Quebec. To date, the historical WGL states sample is small and may not accurately represent the historical population. For example, the mtDNA haplotype FWSClyC12, common in the recovered population, was not found in the initial collection of historical samples

123

Conserv Genet Table 8 Proportion of membership of western wolf, WGL wolf, dog, wolf–dog hybrid and Wisconsin coyote samples to STRUCTURE inferred clusters at K = 5

Cluster I C. lupus

Cluster II C. latrans

Cluster III C. familiaris

Cluster IV C. lupus

Cluster V C. lycaon

AK

0.89

0.02

0.02

0.03

0.04

BC

0.95

0.01

0.01

0.02

0.01

AB

0.78

0.01

0.01

0.16

0.04

MN

0.04

0.02

0.02

0.44

0.48

WI

0.05

0.02

0.03

0.42

0.49

UP

0.05

0.07

0.02

0.47

0.39

Dog

0.01

0.02

0.94

0.01

0.02

Wolf–dog

0.05

0.08

0.69

0.09

0.09

WI coyote

0.01

0.94

0.02

0.02

0.01

Western wolf

WGL wolf

Highest proportion of membership in bold type

characterized by Leonard and Wayne (2008b), but was detected in additional samples studied by Wheeldon and White (2009) and Koblmu¨ller et al. (2009). Similarly, Wheeldon and White (2009) found evidence of C. lupus introgression in the nuclear DNA of two of the three historical WGL states samples they examined. Individual ancestries inferred from the autosomal genotypes of recovered WGL states wolves did not always correlate with respective mtDNA or Y-chromosome ancestries (and see Koblmu¨ller et al. 2009; Wilson et al. 2009), suggesting that cross-species display of lineage markers does not necessarily indicate recent hybridization. Rather, hybridization may be generations removed and/or have occurred in multiple generations. Moreover, mtDNACR haplotype frequencies are dynamic––we observed a [15% increase in the frequency of FWSClyC12 among colonizing Wisconsin wolves in a 10 year period. These observations raise questions as to how the loss of rare mtDNA haplotypes over 100 years of extirpation and recovery should be interpreted with regard to the comparability of the ancestral pedigrees of historical and recolonized WGL wolves.

Conclusion The recovery goal of establishing a large, genetically diverse population of wolves in the WGL states has been attained (U.S. Fish and Wildlife Service 1992), but significant inbreeding may have occurred in the course of re-establishment. Cross-species display of mtDNA and Y-chromosome haplotypes in male wolves, as well as introgression detected in multi-locus autosomal genotypes, indicated that C. lycaon–C. lupus hybrids are common. Hybridization between WGL states wolves and sympatric coyotes has not occurred recently, although some

123

introgression has been mediated indirectly through immigrant C. lycaon–C. latrans hybrids. The history of hybridization in the WGL states predates significant human intervention in the region, as similar results have been obtained in genetic characterizations of historical WGL states wolves (Leonard and Wayne 2008b; Wheeldon and White 2009). Recovered WGL states wolves represent an ancient component of the northeast ecosystem and have been established throughout the region for thousands of years. WGL states wolves differ significantly from wolves to the east of Lake Superior and Lake Huron in that they have not hybridized with coyotes. Eastern wolves range in size from smaller animals in southeast Ontario to larger animals in the northwest (Kolenosky and Stanfield 1975). Small size has been associated with C. latrans–C. lycaon hybridization, large size with C. lupus–C. lycaon hybridization (Kyle et al. 2006) and this introgression has been documented with corresponding genetic evidence (Wilson et al. 2009). The larger wolves of northwest Ontario are thought to be better adapted to hunting large prey, but they may also be better able to exclude coyote competitors from wolf range (Pacquet 1992, Berger and Gese 2007). Furthermore, disparity in body size has been proposed as a reason for the lack of hybridization between western C. lupus and coyotes (Lehman et al. 1991; Roy et al. 1994; Pilgrim et al. 1998). Perhaps size incompatibility and coyote-directed aggression have contributed to the conspicuous absence of coyote hybridization among recovered WGL states wolves. However, hybridization with domestic dogs threatens the continued stability of the recovered WGL states population (Hope 1994; Saetre et al. 2004; Ito et al. 2004). The introgression of dog genes in wolf populations may reduce viability with the erosion of native adaptation and has been associated with uncharacteristically aggressive behaviors

Conserv Genet

(see van den Berg et al. 2005; Wydeven et al. 1999). Bad wolf–dog hybrid behavior may be confused with wolf behavior and deteriorating public relations can lead to reduced initiative to protect this recently recovered endangered species (Treves 2008). Finally, the historic range of C. lycaon within the continuous US includes the WGL states and the northeast US (e.g., Maine, Massachusetts, New Hampshire, New York, Pennsylvania and Vermont; Kyle et al. 2006). Wolves have been delisted in the WGL states but C. lupus remains the endangered wolf species of record in the northeast US and should be replaced by C. lycaon. The northeast US includes unoccupied wolf habitat as well as that occupied by coyotes. Since extirpation, wolves have occasionally passed

through this region but have not established resident packs. If the northeast US were to be considered for wolf reintroduction, C. lycaon should be the object of the recovery program. Acknowledgements We are grateful for the help of the many colleagues who have provided the wolf samples that this work is based upon, particularly Adrian Wydevan (WIDNR), Peter Gogan (NPS), Thomas Cooley (MIDNR), Dave Duncan (USFWS) and Ed Spoon (USFWS). We also thank Thomas Cooley (MIDNR), Paula Holahan (Univ of WI) and Nancy Thomas (WIDNR) for morphological characterization of some of the wolves in this study and Jennifer Leonard (Uppsala Univ) for providing sequence data that was invaluable for the confirmation of some of the haplotypes we identified. We also would like to thank Brian Hamlin and Doina Voin (USFWS) for laboratory and analytical assistance.

Appendix 1 Allele frequencies for eight autosomal microsatellite loci in WGL wolves, Western wolves, coyotes and reference canids. Allele sizes are in base pairs Wolves a

Locus and allele

Coyotes

Domestic canids b

Dogsc

Dogs

Wolf–dogs

–

0.026

–

–

0.100

0.410

0.474

0.625

0.108

0.075

–

0.026

0.042

0.284

0.450

0.077

–

0.021

MN

WI

MI

WGL

AK

AB

BC

West

WI

MN

140

0.132

0.096

0.031

0.093

–

0.049

0.400

0.117

0.014

142

0.618

0.561

0.750

0.606

0.351

0.439

0.160

0.340

0.095

144

–

–

0.063

0.008

0.081

0.049

0.240

0.107

146

–

0.009

–

0.004

0.378

0.280

0.060

0.262

148

0.184

0.184

0.094

0.169

0.149

0.110

0.040

0.107

0.189

0.175

0.205

0.237

–

150

0.053

0.149

0.063

0.114

0.041

–

–

0.015

0.095

0.200

0.244

0.237

0.313

152

0.013

–

–

0.004

–

0.730

0.100

0.053

0.095

–

0.038

–

–

154

–

–

–

–

–

–

–

–

0.108

–

–

–

–

156

–

–

–

–

–

–

–

–

0.014

–

–

–

–

158 123

–

–

–

–

–

–

–

–

–

–

–

0.026

–

141

0.013

–

–

0.004

–

–

–

–

–

–

–

–

–

143

0.211

0.331

0.313

0.269

0.459

0.768

0.780

0.660

–

0.100

0.149

0.158

0.167

145

0.053

0.017

0.063

0.037

0.135

–

–

0.049

0.054

0.100

0.027

0.105

0.063

147

0.158

0.119

0.156

0.149

0.176

0.049

0.060

0.097

0.351

0.200

0.365

0.211

0.354

149

0.066

0.025

0.125

0.050

0.189

0.171

–

0.136

0.297

0.425

0.270

0.263

0.063

151

–

0.008

–

0.004

0.041

0.012

0.16

–

0.270

0.025

0.149

0.184

0.188

153

0.303

0.331

0.344

0.326

0.838

0.366

0.36

0.058

0.027

0.150

0.041

0.079

0.146

155

–

–

–

–

–

–

–

–

–

–

–

–

0.021

157

0.197

0.169

–

0.161

–

–

–

–

–

–

–

–

–

109

172 140

0.211

0.178

0.313

0.202

–

–

–

–

0.486

0.300

–

–

0.065

142

–

–

–

–

–

–

–

–

–

–

–

–

–

144

–

–

–

–

–

–

–

–

–

–

–

–

–

146 148

– –

– –

– –

– –

– –

– –

– –

– –

– –

– –

– –

– –

– –

150

0.079

0.203

0.063

0.161

–

–

–

–

0.014

0.067

–

–

–

152

0.145

0.085

0.156

0.124

0.270

0.463

0.480

0.398

0.027

0.567

0.012

–

0.087

154

0.513

0.500

0.469

0.479

0.730

0.537

0.520

0.602

0.338

0.067

0.952

1.000

0.761

156

0.053

0.034

–

0.033

–

–

–

–

0.135

–

0.036

–

0.087

123

Conserv Genet Appendix 1 continued Wolves Locusa and allele

Coyotes

Domestic canids

MN

WI

MI

WGL

AK

AB

BC

West

WI

MNb

Dogsc

Dogs

Wolf–dogs

114

–

–

–

–

–

–

–

–

–

–

0.024

–

–

116

–

–

–

–

–

–

–

–

–

–

0.061

–

–

118

–

0.009

–

0.004

–

–

–

–

0.081

0.447

0.061

–

–

120

0.447

0.578

0.531

0.525

0.459

0.232

0.500

0.379

0.095

0.342

0.561

0.184

0.091

122

–

–

–

0.204

–

–

–

–

0.014

–

0.195

0.079

0.023

124

0.289

0.172

0.094

0.025

–

–

–

0.005

0.095

–

0.061

–

0.250

126

–

0.026

0.063

0.025

0.014

–

–

0.175

0.311

–

0.037

0.237

0.023

128 130

0.013 0.158

0.026 0.095

0.031 0.188

0.129 0.083

0.135 0.149

0.293 0.220

0.040 0.260

0.204 0.053

0.041 0.162

0.105 0.026

– –

0.132 0.132

0.227 0.023

132

0.092

0.086

0.094

–

0.095

0.012

0.056

0.184

0.108

–

–

0.053

0.318

134

–

–

–

0.004

0.149

0.244

0.140

–

0.014

0.053

–

0.184

0.045

136

–

0.009

–

–

–

–

–

–

0.081

0.026

–

–

–

200

0.632

0.518

0.656

0.585

0.230

0.049

0.200

0.150

0.824

0.711

0.303

0.368

0.273

202

0.237

0.348

0.219

0.288

0.081

0.134

0.260

0.146

0.176

0.289

0.026

0.053

0.114

204

–

–

–

–

–

–

–

–

–

–

–

–

0.023

206

0.079

0.080

0.125

0.085

0.486

0.378

0.340

0.408

–

–

0.026

–

0.091

208

0.053

0.045

–

0.038

0.203

0.439

0.200

0.296

–

–

–

–

0.023

210

–

–

–

–

–

–

–

–

–

–

–

–

0.045

212

–

0.009

–

0.004

–

–

–

–

–

–

0.513

0.579

0.432

214

–

–

–

–

–

–

–

–

–

–

0.066

–

–

216

–

–

–

–

–

–

–

–

–

–

–

–

–

218 220

– –

– –

– –

– –

– –

– –

– –

– –

– –

– –

0.013 –

– –

– –

222

–

–

–

–

–

–

–

–

–

–

–

–

–

224

–

–

–

–

–

–

–

–

–

–

0.053

–

–

154

–

–

–

–

–

0.024

–

0.010

–

–

–

–

–

156

–

–

–

–

–

0.012

0.020

0.010

0.014

–

–

–

0.002

158

0.354

0.398

0.250

0.384

.622

0.341

0.460

0.471

0.027

–

0.439

0.421

0.307

160

0.307

0.356

0.406

0.331

.230

0.402

0.300

0.316

0.500

0.325

0.012

0.026

0.278

162

0.276

0.203

0.188

0.223

.135

0.183

–

0.121

0.162

0.050

0.488

0.526

0.282

164

0.064

0.042

0.125

0.058

.014

0.037

0.220

0.073

0.203

0.175

0.049

0.026

0.112

166

–

–

0.031

0.004

–

–

–

–

0.068

0.100

0.012

–

0.014

168

–

–

–

–

–

–

–

–

–

–

–

–

–

170

–

–

–

–

–

–

–

–

–

–

–

–

0.003

172

–

–

–

–

–

–

–

–

–

–

–

–

–

174 250

–

–

–

–

–

–

–

–

0.027

0.350

–

–

0.003

130

–

0.008

–

0.004

–

–

–

–

0.108

0.031

0.024

0.222

0.064

132

0.289

0.178

0.156

0.202

–

–

0.019

0.027

0.063

–

–

0.083

134

0.158

0.203

0.031

0.174

0.207

0.180

0.165

0.216

0.125

0.083

0.139

0.175

136

0.500

0.517

0.688

0.537

0.220

0.320

0.194

0.054

0.094

0.214

0.194

0.346

138

0.026

0.025

0.031

0.025

.541

0.244

0.380

0.383

0.176

0.063

0.131

0.111

0.105

140

–

0.025

–

0.017

.095

0.012

0.000

0.039

0.176

0.250

0.298

0.222

0.087

200

204

225

123

.054 1.08 .081

Conserv Genet Appendix 1 continued Wolves Locusa and allele

Coyotes AK

WI

MNb

Dogsc

MN

WI

MI

WGL

142

0.013

0.017

–

0.012

144

0.013

0.025

0.094

0.029

–

146

–

–

–

–

–

0.122

0.040

0.058

0.041

0.031

0.048

0.028

0.019

148

–

–

–

–

–

–

–

–

0.014

–

0.036

–

0.019

150

–

–

–

–

–

–

–

–

–

–

–

–

0.005

152

–

–

–

–

–

–

–

–

–

–

–

–

–

154

–

–

–

–

–

–

–

–

–

–

–

–

–

156

–

–

–

–

–

–

–

–

–

–

–

–

–

158 160

– –

– –

– –

– –

– –

– –

– –

– –

– –

– –

– –

– –

– –

136

–

–

0.031

0.004

–

–

–

–

–

–

–

–

140

–

–

–

–

–

–

–

–

–

–

–

142

–

–

–

–

–

–

–

–

–

–

–

144

–

–

0.031

0.004

0.108

0.088

0.080

0.093

–

–

0.025

0.026

0.217

146

0.132

0.069

0.031

0.088

0.081

0.088

0.160

0.103

–

–

0.538

0.737

0.239

148

–

–

–

–

–

–

–

–

–

–

–

0.065

150

0.105

0.138

0.188

0.146

–

–

–

0.079

0.053

–

–

–

152

–

–

0.031

0.004

–

–

–

–

0.053

–

–

0.022

154

0.053

0.034

0.094

0.050

0.041

0.050

0.060

0.049

0.132

0.026

0.025

–

0.022

156

0.026

0.026

0.094

0.033

0.014

0.038

0.020

0.025

0.079

–

–

–

–

158

–

–

–

–

0.014

–

–

0.005

0.132

0.132

–

0.026

0.109

160

0.197

0.284

0.219

0.238

0.230

0.163

0.440

0.255

0.079

0.395

–

–

0.022

162 164

0.066 0.342

0.121 0.276

0.031 0.250

0.083 0.300

0.027 0.392

0.138 0.438

0.080 0.160

0.083 0.353

0.158 0.184

0.026 0.132

– –

0.079 0.026

0.130 –

166

–

–

–

–

0.054

–

–

0.020

0.132

0.184

0.087

0.105

0.174

168

0.079

0.052

–

0.050

0.041

–

–

0.015

0.026

–

0.138

–

–

170

–

–

–

–

–

–

–

–

–

–

0.138

–

–

172

–

–

–

–

–

–

–

–

–

–

0.038

–

–

174

–

–

–

–

–

–

–

–

–

–

0.013

–

–

.122

West

Domestic canids

AB

BC

Dogs

Wolf–dogs

0.195

0.080

0.141

0.122

0.250

0.048

–

0.053

–

–

–

0.068

0.094

0.119

0.083

0.043

377

Data from other sources are noted. Sampling localities are abbreviated as northeast Minnesota (MN), north-central Wisconsin (WI), UpperPeninsula Michigan (UPMI), Alaska (AK), Alberta (AB), British Columbia (BC), combined frequency of wolves from MN, WI and MI (WGL) and Alaska, Alberta and British Columbia (West) a

For primer sequence information and locus description please see Ostrander et al. (1993)

b

Allele frequencies from Roy et al. (1994)

c

Allele frequencies from Garcia-Moreno et al. (1996)

References Adams JR, Kelly BT, Waits LP (2003) Using faecal DNA sampling and GIS to monitor hybridization between red wolves (Canis rufus) and coyotes (Canis latrans). Mol Ecol 12:2175–2186 Aggarwal RK, Kivisild T, Ramadevi J, Singh L (2007) Mitochondrial DNA coding region sequences support the phylogenetic distinction of two Indian wolf species. J Zoolog Syst Evol Res 45: 163–172 Bandelt H-J, Forster P, Ro¨hl A (1999) Median-joining networks for inferring intraspecific phylogenies. Mol Biol Evol 16:37–48

Bannasch DL, Bannasch MJ, Ryun JR, Famula TR, Pedersen NC (2005) Y chromosome haplotype analysis in purebred dogs. Mamm Genome 16:273–280 Berger KM, Gese EM (2007) Does interference competition with wolves limit the distribution and abundance of coyotes? J Anim Ecol 76:1075–1085 Cornuet JM, Luikart G (1996) Description and power analysis of two tests for detecting recent population bottlenecks from allele frequency data. Genetics 144:2001–2014 Di Rienzo A, Peterson AC, Garza JC, Valdes AM, Slatkin M, Freimer NB (1994) Mutational processes of simple-sequence repeat loci in human populations. Proc Natl Acad Sci USA 91:3166–3170

123

Conserv Genet El Mousadik A, Petit RJ (1996) High level of genetic differentiation for allelic richness among populations of the argan tree [Argania spinosa (L.) Skeels] endemic to Morroco. Theor Appl Genet 92:832–839 Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Mol Ecol 14:2611–2620 Fain SR, Hamlin BC, Straughan DJ (2000) Genetic variation in the river sturgeon scaphirhynchus (Acipenseridae) as inferred from partial mtDNA sequences of cytochrome b. Final report, pp 1–20. http://www.lab.fws.gov/pdfs/Fain_etal.2000.pdf Felsenstein J (1989) PHYLIP––phylogeny inference package (version 3.2). Cladistics 5:164–166 Felsenstein J (1995) PHYLIP––phylogeny inference package (version 3.57). http://evolution.genetics.washington.edu/phylip.html. Updated from Felsenstein (1989) Fish US, Service Wildlife (2007) Proposed rule. Fed Reg 72:6051– 6103 Forbes SH, Boyd DK (1996) Genetic variation of naturally colonizing wolves in the central Rocky Mountains. Conserv Biol 10:1082– 1090 Forbes SH, Boyd DK (1997) Genetic structure and migration in native and reintroduced Rocky Mountain wolf populations. Conserv Biol 11:1226–1234 Garcia-Moreno J, Roy MS, Geffen E, Wayne RK (1996) Relationships and genetic purity of the endangered Mexican wolf based on analysis of microsatellite loci. Conserv Biol 10:376–387 Gottelli D, Sillero-Zubirl C, Applebaum GD, Roy MS, Girman DJ, Garcia-Moreno J, Ostrander EA, Wayne RK (1994) Molecular genetics of the most endangered canid: the Ethiopian wolf Canis simensis. Mol Ecol 3:301–312 Goudet J (1995) FSTAT (vers. 1.2) a computer program to calculate F-statistics. J Hered 86:485–486 Goudet J (2001) FSTAT, a program to estimate and test gene diversities and fixation indices (version 2.9.3). http://www.unil.ch/izea/ softwares/fstat.html. Updated from Goudet (1995) Grewal SK, Wilson PJ, Kung TK, Sharmi K, Theberge MT, Theberge JB, White BN (2004) A genetic assessment of the eastern wolf (Canis lycaon) in Algonquin Provincial Park. J Mamm 85: 625–632 Hailer F, Leonard JA (2008) Hybridization among three native North American Canis species in a region of natural sympatry. PLoS One. doi:10.1371/journal.pone.0003333 Hellborg L, Ellegren H (2004) Low levels of nucleotide diversity in mammalian Y chromosomes. Mol Biol Evol 21:158–163 Hope J (1994) Wolves and wolf hybrids as pets are big business––but a bad idea. Smithsonian 25:34–45 Ito H, Nara H, Inoue-Muayama M, Shimada MK, Koshimura A, Ueda Y, Kitagawa H, Takeuchi Y, Mori Y, Murayama Y, Morita M, Iwasaki T, Ota K, Tanabe Y, Ito S (2004) Allele frequency distribution of the canine dopamine receptor D4 gene exon III and I in 23 breeds. J Vet Med Sci 66:815–820 Jobling MA, Heyer E, Dieltjes P, deKnijff P (1999) Y-chromosomespecific microsatellite mutation rates re-examined using a minisatellite, MSY1. Hum Mol Genet 8:2117–2120 Koblmu¨ller S, Nord M, Wayne RK, Leonard JA (2009) Origin and status of the Great Lakes wolf. Mol Ecol 18:2313–2326 Kolenosky GB, Stanfield R (1975) Morphological and ecological variation among gray wolves (Canis lupus) of Ontario. In: Fox MW (ed) The wild canids––their systematics, behavioral ecology and evolution. Van Nostrand, New York Kyle CJ, Johnson AR, Patterson BR, Wilson PJ, Shami K, Grewal SK, White BN (2006) Genetic nature of eastern wolves: past, present and future. Conserv Genet 7:273–287 Kyle CJ, Johnson AR, Patterson BR, Wilson PJ, White BN (2008) The conspecific nature of eastern and red wolves: conservation

123

and management implications. Conserv Genet. doi:10.1007/ s10592-007-9380-5 Lehman N, Eisenhawer A, Hansen K, Mech LD, Peterson RO, Gogan PJP, Wayne RK (1991) Introgression of coyote mitochondrial DNA into sympatric North American gray wolf populations. Evolution 45:104–119 Leonard JA, Wayne RK (2008a) Wishful thinking: imagining that the current Great Lakes wolf is the same entity that existed historically. Biol Lett. doi:10.1098/rsbl.2008.0533 Leonard JA, Wayne RK (2008b) Native Great Lakes wolves were not restored. Biol Lett 4:95–98 Lucchini V, Galov A, Randi E (2004) Evidence of genetic distinction and long-term population decline in wolves (Canis lupus) in the Italian Apennines. Mol Ecol 13:523–536 Luikart G, Cornuet J-M (1998) Empirical evaluation of a test for identifying recently bottlenecked populations from allele frequency data. Conserv Biol 12:228–237 Mech LD (1974) Canis lupus. Mammalian Species 37:1–6 Mech LD (2008) Crying wolf: concluding that wolves were not restored. Biol Lett. doi:10.1098/rsbl.2008.0440 Mech LD, Federoff NE (2002) Alpha1-antitrypsin polymorphism and systematics of eastern North American wolves. Can J Zool 80:961–963 Mech LD, Fritts SH, Radde G, Paul WJ (1988) Wolf distribution in Minnesota relative to road density.Wildlife Soc B 16:85–88 Mech LD, Fritts SH, Wagner D (1995) Minnesota wolf dispersal to Wisconsin and Michigan. Am Mid Nat 133:368–370 Mladenoff DJ, Sickley TA, Haight RG, Wydevan AP (1995) A regional landscape analysis and prediction of favorable gray wolf habitat in the northern Great Lakes Region. Conserv Biol 9:279–294 Mun˜oz-Fuentes V, Darimont CT, Wayne RK, Paquet PC, Leonard J (2009a) Ecological factors drive differentiation in wolves from British Columbia. J Biogeograp. doi:10.1111/j.1365-2699. 2008.02067.x Mun˜oz-Fuentes V, Darimont CT, Paquet PC, Leonard J (2009b) The genetic legacy of extirpation and re-colonization in Vancouver Island wolves. Conserv Genet. doi:10.1007/s10592-009-9974-1 Musiani M, Leonard J, Cluff HD, Gates CC, Mariani S, Paquet PC, Vilas C, Wayne RK (2007) Differentiation of tundra/taiga and boreal coniferous forest wolves: genetics, coat color and association with migratory caribou. Mol Ecol 16:4149– 4170 Natanaelsson C, CR Oskarsson M, Angleby H, Lundeberg J, Kirkness E, Savilainen P (2006) Dog Y chromosomal DNA sequence: identification, sequencing and SNP discovery. BMC Genet 7:45. doi:10.1186/1471-2156-7-45 Nei M (1973) Analysis of gene diversity in subdivided populations. Proc Natl Acad Sci USA 70:3321–3323 Nei M (1987) Molecular evolutionary genetics. Columbia University Press, New York Nowak RM (1983) A perspective on the taxonomy of wolves in North America. In: Carbyn LN (ed) Wolves in Canada and Alaska: their status, biology, and management. Report series, no. 45. Canadian Wildlife Service, Edmonton, Alberta, pp 10–19 Nowak RM (1995) Another look at wolf taxonomy. In: Carbyn LN, Fritts SH, Seip DR (eds) Ecology and conservation of wolves in a changing world. Canadian Circumpolar Institute, Edmonton, Alberta, pp 375–397 Nowak RM (2002) The original status of wolves in eastern North America. Southeast Nat 1:95–130 Nowak RM (2003) Wolf evolution and taxonomy. In: Mech LD, Boitani L (eds) Wolves, behavior, ecology, and conservation. Chicago, University of Chicago Press, pp 239–258 Nowak RM (2009) Chapter 15, taxonomy, morphology, and genetics of wolves in the Great Lakes region. In: Wydeven AP, Van

Conserv Genet Deelen TR, Heske E (eds) Recovery of wolves in the great lakes region. New York, Springer, pp 233–250 Ostrander EA, Sprague GF, Rine J (1993) Identification and characterization of dinucleotide repeat (CA)n markers for genetic mapping in dog. Genomics 16:207–213 Pacquet PC (1992) Prey use strategies of sympatric wolves and coyotes in Riding Mountain National Park, Manitoba. J Mamm 73:337–343 Page RDM (1996) TREEVIEW: an application to display phylogenetic trees on personal computers. Comput Appl Biosci. 12:357– 358. http://taxonomy.zoology.gla.ac.uk/rod/treeview.html Pilgrim KL, Boyd DK, Forbes SH (1998) Testing for wolf-coyote hybridization in the Rocky Mountains using mitochondrial DNA. J Wildl Manag 62:683–686 Piry S, Luikart G, Cornuet J-M (1999) Bottleneck: a computer program for detecting recent reductions in the effective population size using allele frequency data. J Hered 90:502–503 Primmer CR, Saino N, Moller AP, Ellegren H (1998) Unravelling the processes of microsatellite evolution through analysis of germ line mutations in Barn Swallows Hirundo rustica. Mol Biol Evol 15:1047–1054 Pritchard JK, Wen W (2003) Documentation for STRUCTURE software: version 2. http://pritch.bsd.uchicago.edu Pritchard JK, Stephens M, Donnelly P (2000) Inference of population structure using multilocus genotype data. Genetics 155:945– 959 Randi E, Lucchini V (2002) Detecting rare introgression of domestic dog genes into wild wolf (Canis lupus) populations by Bayesian admixture analysis of microsatellite variation. Conserv Genet 3:31–45 Raymond M, Rousset F (1995) GENEPOP (version 1.2): population genetics software for exact tests and ecumenism. J Hered 86:248– 249. http://kimura.univ-montp2.fr/*rousset/Genepop.htm Rice WR (1989) Analyzing tables of statistical tests. Evolution 43:223–225 Roy MS, Geffen E, Smith D, Ostrander EA (1994) Patterns of differentiation and hybridization in North American wolflike canids, revealed by analysis of microsatellite loci. Mol Biol Evol 11:553–570 Sacks BN, Brown SK, Ernest HB (2004) Population structure of California coyotes corresponds to habitat-specific breaks and illuminates species history. Mol Ecol 13:1265–1275 Saetre P, Lindberg J, Leonard JA, Olsson K, Pettersson U, Ellegren H, Bergstrom TF, Vila C, Jazin E (2004) From wild wolf to domestic dog: gene expression changes in the brain. Mol Brain Res 126:198–206 Schneider S, Kueffer JM, Roessli D, Excoffier L (2000) Arlequinn ver 2000: a software for population genetics data analysis. Genetics and Biometry Lab, University of Geneva, Geneva Schwartz MK, McKelvey KS (2009) Why sampling scheme matters: the effect of sampling scheme on landscape genetic results. Conserv Genet 10:441–452 Shami K (2002) Evaluating the change in distribution of the eastern timber wolf (Canis lycaon) using the Y-chromosome. M.Sc. Thesis, McMaster University, 73 pp Skeel MA, Carbyn LN (1977) The morphological relationship of gray wolves (Canis lupus) in national parks of central Canada. Can J Zool 55:737–747 Slatkin M (1985) Rare alleles as indicators of gene flow. Evolution 39:53–65 Sundqvist A-K, Ellegren H, Olivier M, Vila C (2001) Y chromosome haplotyping in Scandinavian wolves (Canis lupus) based on microsatellite markers. Mol Ecol 10:1959–1966

˚, Sundqvist A-K, Bjo¨rnerfeldt S, Leonard JA, Hailer F, Hedhammar A Ellegren H, Vila C (2006) Unequal contribution of sexes in the origin of dog breeds. Genetics 172:1121–1128 Thiel RP, Hammill JH (1988) Wolf specimens in Upper Michigan, 1960–1986. Jack-Pine Warbler 66:153–179 Treves A (2008) Beyond recovery: Wisconsin’s wolf policy 1980– 2008. Hum Dimens Wild 13:329–338. doi:10.1080/108720080 2277716 U.S. Fish and Wildlife Service (1992) Recovery plan for the eastern timber wolf. Twin Cities, Minnesota, 73 pp Valdes AM, Slatkin M, Freimer NB (1993) Allele frequencies at microsatellite loci: the stepwise mutation model revisited. Genetics 133:737–749 Van den Berg L, Kwant L, Hestand MS, van Oost BA, Leegwater PAJ (2005) Structure and variation of three canine genes involved in serotonin binding and transport: the serotonin receptor IA gene (htrIA), serotonin receptor 2A gene (htr2A), and serotonin transporter gene (slc6A4). J Hered 96:786–796 Verardi A, Lucchini V, Randi E (2006) Detecting introgressive hybridization between free-ranging domestic dogs and wild wolves (Canis lupus) by admixture linkage disequilibrium analysis. Mol Ecol 15:2845–2855 Vila C, Wayne RK (1999) Hybridization between wolves and dogs. Conserv Biol 13:195–198 Vila C, Walker C, Sundqvist A-K, Flagstad Ø, Andersone Z, Casulli A, Kojola I, Valdmann H, Halverson J, Ellegren H (2003) Combined use of maternal, paternal and bi-parental genetic markers for the identification of wolf–dog hybrids. Heredity 90:17–24 Weir BS, Cockerham CC (1984) Estimating F-statistics for the analysis of population structure. Evolution 38:1358–1370 Wheeldon T, White BN (2009) Genetic analysis of historic western Great Lakes region wolf samples reveals early Canis lupus/ lycaon hybridization. Biol Lett. doi:10.1098/rsbl.2008.0516 Wilson PJ, Grewal S, Lawford ID, Heal JNM, Granacki AG, Pennock D, Theberge JB, Theberge MT, Voigt DR, Waddell W, Chambers RC, Paquet PC, Goulet G, Cluff D, White BN (2000) DNA Profiles of the eastern Canadian wolf and the red wolf provide evidence for a common evolutionary history independent of the gray wolf. Can J Zool 78:2156–2166 Wilson PJ, Grewal S, McFadden T, Chambers RC, White BN (2003) Mitochondrial DNA extracted from eastern North American wolves killed in the 1800 s is not of gray wolf origin. Can J Zool 81:936–940 Wilson PJ, Grewal SK, Mallory FF, White BN (2009) Genetic characterization of hybrid wolves across Ontario. J Hered 100(Supplement 1):S80–S89 Wisconsin Department of Natural Resources (WIDNR) 1999 wolf management plan. http://dnr.wi.gov/org/land/er/publications/ wolfplan/toc.htm. Madison, WI Wydevan AP, Schultz RN, Thiel RP (1995) Monitoring of a recovering gray wolf population. In: Wisconsin, 1979–1991. In: Carbyn LN, Fritts SH, Seip DR (eds) Ecology and conservation of wolves in a changing world. Canadian Circumpolar Institute, Edmonton, pp 147–156 Wydeven AP, Weidenhoeft JE, Kohn BE, Thiel RP, Schultz RN, Boales SR (1999) Progress report of wolf population monitoring in Wisconsin for the period April–September 1999. Wisconsin Department of Natural Resources. http://www.timberwolfinformation. org/updates/oct201999/october1999wolfprogressreport.pdf Young SP, Goldman GE (1944) The wolves of North America. American Wildlife Institute, Washington, DC

123