 |
 |
|||
DNA-analysen som sannsynliggjorde at de omtalte dyrene var hybrider av Canis lupus lupus og Canis lupus familiarisPRESENTATION OF DNA ANALYSISMethodologyAnimal samples Direktoratet for naturforvaltning arranged for us to be provided with three samples from Norway for genetic analyses. Sample A. Blood from a juvenile individual run over and killed in Våler (Østfold) on 1 October 1999. Sample B. Snow with urine and blood collected 1 March 1999. Assumed to correspond to the alpha female in the wolf pack in Moss (in heat). Preserved frozen. Sample C. Drop of blood collected from the snow on 9 January 1999. From the injured back foot of the animal. Assumed to correspond to the alpha female in the wolf pack in Moss. Preserved in 95% ethanol. Information concerning the respective samples given above is as we have been informed through contacts with Terje Bø and Morten Kjørstad at Direktoratet for naturforvaltning and with Rune Bergstrøm and Asmund Fjellbakk at fylkesmannen in Østfold. We will generally refer to the samples by the letters A, B and C. We have analysed samples A, B and C in relation to our own stock of wolf and dog samples. The latter wolves are from Sweden and Norway (25), Finland (25), Russia (24) and Estonia (23), and have been collected mainly during the last few years, with the exception of some samples from Sweden which are from 1983 and onwards. Our 44 dog samples correspond to pure bred huskies (19), German shepherd dog (7), American eskimo (7), Alaskan malamute (6), elkhound (3), keeshound (1) and kuvasz (1), all from the USA. We assume that members of the same breed sampled on different continents will be sufficiently similar to each other to be useful as reference population in this analysis. Importantly, the critical issue is that they are more similar to each other than to different populations of wolves, which seems to be a reasonable assumption. Dog breeds were chosen based on somewhat morphological similarity with wolves. Laboratory and data analysis: mitochondrial DNA A general overview of the characteristics of the genetic markers used in this study is provided in Appendix 1. DNA was isolated using slight variations on phenol-chloroform extraction methods (Sambrook et al. 1989). For sample B, a part of the volume of snow containing the urine and blood had to be thawed and centrifugated for an extended period of time to concentrate the DNA before isolation. For sample C, a small amount of coagulated blood was taken from the ethanol for the extraction. Amplification of a fragment of about 400 base pairs (bp) from the mitochondrial DNA control region was performed via the polymerase chain reaction (PCR) using universal primers modified from Kocher et al. (1989). Precise primer sequences, PCR conditions and profile are available upon request. PCR products were sequenced using Big Dye Terminator cycle sequencing chemistry on an ABI 377 instrument (Perkin Elmer), following protocols provided by the manufacturer. Sequences were aligned using the program CLUSTAL W (Higgins et al. 1992) and checked by eye. All sequences were compared to each other and to sequences available in GenBank and databases previously developed (based on Ellegren et al. 1996, Okumura et al. 1996, Taberlet et al. 1996, Tsuda et al. 1997, Vilà et al 1997, Pilgrim et al. 1998, Vilà et al. 1999) using the program PAUP* (Swofford 1998). Laboratory and data analyses: Y chromosome microsatellite We made use of a canine Y chromosome microsatellite marker developed by Olivier et al. (1999), with some optimization to allow more specific amplification. The forward primer was labeled with a fluorescent dye and the PCR products run on an ABI 377 instrument (Perkin Elmer), following protocols provided by the manufacturer. PCR primers, conditions and profile, as well as technical details on the protocols are available upon request. Using the program GENESCAN (Perkin Elmer), the electrophoresis running times were correlated to the size of the DNA fragments, and the alleles observed for each microsatellite were subsequently scored using the software GENOTYPER (Perkin Elmer). Laboratory and data analysis: autosomal microsatellites Eightteen microsatellites developed for dogs were selected for this study: c2001, c2010, c2017, c2054, c2079, c2088 and c2096 (Francisco et al. 1996), vWF (Shibuya et al. 1994), u213, u250 and u253 (Ostrander et al. 1993), and PEZ01, PEZ03, PEZ05, PEZ06, PEZ08, PEZ12 and PEZ20 (Perkin Elmer, Zoogen). As for the electrophoresis of PCR products, allele identification and scores, we followed the same protocol used for the Y chromosome microsatellites. To study the likelihood of finding one of the observed genotypes in each one of the populations we used an assignment test (Paetkau et al. 1995, Waser and Strobeck 1998). This analysis calculates the likelihood of finding a certain genotype in each population and assigns the individual to the population for which it has the highest likelihood. If the frequency of allele A in a population is pA, the frequency of homozygote AA for a certain locus would be pA2 and the frequency of a heterozygote AB would be 2pApB, according to simple Hardy-Weinberg principles. For a multilocus genotype (assuming that loci are not linked, which seems reasonable in this case), the likelihood of finding the genotype in a population will be the product of the frequencies for all alleles at all loci. If the likelihood of finding a certain genotype is significantly higher for one population than for the others, this could suggest that the individual sample belongs to or originated from that population (for example, see Paetkau et al. 1998). Since the likelihood values are typically extremely low, the log likelihood is indicated. However, a higher likelihood may indicate that a certain genotype is more similar to the genotypes expected in one population than to the genotypes expected in others, but this does not indicate that it originated in the first population. Since we have only been able to genotype a limited number of individuals from each population, we can not expect our sample to represent most of the genotypic variability in the population. To characterize how well an individual genotype fits into the distribution of genotypes that should be expected from each population, we generated 1000 synthetic genotypes taking random alleles for each locus according to their frequency. These synthetic genotypes represent multilocus genotypes that could occur in the observed allelic distributions but may or may not have been sampled. Similarly, we generated populations of 1000 synthetic genotypes of hybrids between dogs and Swedish wolves. In this case, each synthetic genotype contained per locus one allele derived from each one of the parent populations. If the likelihood of the assignment of the target sample is outside the range observed for the 1000 synthetic animals corresponding to one population, we have a statistical basis for rejecting that population as the origin of the target sample. Since the number of microsatellites successfully scored was different for each target sample, a new set of histograms was constructed for each one. To standardize the histograms of Figure 2, the log likelihood of assignment of the target sample to the wolf population was subtracted from the log likelihoods of all synthetic genotypes. If the assignment for the target sample lies outside the distribution of assignment likelihoods for the synthetic population (or inside the 2.5% margins at each side of the distribution), -that is to say, if the value 0 is included in the distributions shown in the histograms- the hypothesis that the target sample belongs to that population should be rejected. In order to determine if the female of sample B could be the mother of the young canid (sample A), we compared the alleles found at each microsatellite. Assuming that she is the mother, we can determine what should be the allelic composition of the paternal contribution at some microsatellites. We constructed a synthetic genotype homozygous for those alleles and calculated its assignment likelihood to different populations.
Results and discussionMitochondrial DNA (mtDNA) sequences Sample A and sample B showed the same mtDNA D-loop haplotype. This is the same haplotype that was found in all Swedish and Norwegian wolves sampled from 1983 and onwards by Ellegren et al. (1996). Sample C showed a distinctly different haplotype, which excludes it representing the same individual as sample B as well as excluding it being the mother of sample A. Moreover, the haplotype shown by sample C (D10, according to Vila et al. 1997) has previously only been seen in dogs (e.g. dachshund and flat-coated retriever; Vilà et al. 1997). Consequently, samples A and B are compatible with being from animals that have wolf ancestry in the maternal line, while sample C is compatible with being an animal that is either a dog or has dog ancestry in the maternal line. Y chromosome microsatellite The Y chromosome microsatellite was successfully amplified in samples A and C, but not in sample B. A reasonable interpretation is thus that sample B comes from a female. Sample A and sample C had different alleles, suggesting that the sample C (now known to be a male) is not the father of the young of sample A. Importantly, neither the allele of sample A or of sample C has previously been seen among Scandinavian wolves (our unpublished results). However, both alleles have been seen in other wolf populations, as well as in dogs. Autosomal microsatellites An assignment test comparing, on the one hand, the Scandinavian wolf population and, on the other hand, our population of dogs shows that the observed allelic distributions allow for distinguishing between these two groups of animals (Figure 1). In this graphical illustration of the test, all dogs are located in the upper left corner (higher likelihood of being dogs than wolves), whereas the Scandinavian wolves are in the lower right corner.
Since the information provided by Direktoratet for Naturforvaltning states that the female from which sample B originated was the mother of the young of sample A, we have made this assumption for determining the alleles that should come from the father (Table 2). The likelihood of a genotype homozygous for these alleles will be the square of the likelihood for the paternal haplotype. In the same way that the origin of each one of the samples was assessed by comparison to synthetic genotypes, the origin of the father of sample A is analyzed in Table 1. The likelihood of obtaining this haplotype from a Scandinavian, Finnish or Russian wolf is very low and outside the distribution expected for each of them. However, the likelihood for the paternal haplotype falls inside the distribution for pure dogs. Consequently, the microsatellite data suggest the young of sample A having one wolf and one dog as parents. Conclusions Given the available reference material (wolves from Scandinavia, Finland, Russia and Estonia; dogs from North America), we come to the following conclusions. • Sample A, stated to be from a young wolf-like animal run over by a vehicle in Østfold on 1 October 1999, has a DNA profile that is compatible with that arising from a hybridization between a wolf and a dog, but is not compatible with that of a pure Scandinavian wolf. It thus seems reasonable to assume that the animal is a wolf-dog hybrid. • Sample A has a mitochondrial DNA haplotype that suggests it to have a wolf as mother. • Sample A has a Y chromosome DNA profile that suggests it to not have a Scandinavian wolf as father, unless recent immigration has occured. • The blood samples B and C, stated to be collected from the snow on 1 March and 9 January 1999, respectively, correspond to two different animals. Most likely, sample B is a female wolf from the Scandinavian population whereas sample C is a male dog. • The presumed male dog of sample C is not the father of the young animal of sample A. References
Figure 1. Assignment log likelihood for dogs and Swedish+Norwegian wolves. The three target samples are also indicated. Figure 2. Distribution of the assignment likelihood to the Scandinavian wolf population of 1000 synthetic genotypes corresponding to dogs, Scandinavian wolves and F1 hybrids between dogs and wolves. The synthetic genotypes are generated taking into consideration the allelic distributions in the parental populations (see text). Since the number of microsatellites successfully amplified for each target sample varied, the distributions for the synthetic populations were repeated considering only the loci that had been successfully amplified for the sample studied. The likelihoods are standardized by subtracting the likelihood calculated for each target sample. Consequently, if the distribution of likelihoods in the synthetic populations does not encompass the value observed for the target sample (corresponds to the value "0" after standardization), or if this value is inside the 2.5% margins at each side of the distribution, we can conclude that the sample does not fit the expected allelic distribution for the synthetic population. That combination of alleles is uncommon.
Table 2. Microsatellite alleles identified for each target sample. Last column indicates alleles that could be identified as coming from the father of A assuming that B is the mother. Microsatellite Locus Sample A (991001) Sample B (990301) Sample C (990109) Paternal Alleles for sample A c2001 149/153 153/153 133/145 149 c2010 225/237 x/x 237/237 x c2017 258/266 x/x 262/270 x c2054 148/152 148/148 148/160 152 c2079 275/283 271/271 275/279 x c2088 131/135 127/135 x/x 131 c2096 95/103 95/95 99/103 103 PEZ01 112/120 120/120 120/120 112 PEZ03 132/138 138/138 132/144 132 PEZ05 96/104 96/96 104/108 104 PEZ06 174/174 174/174 186/190 174 PEZ08 238/238 x/x 234/238 238 PEZ12 272/272 x/x 258/272 272 PEZ20 177/177 177/177 173/177 177 u213 159/162 x/x x/x x u250 126/138 x/x x/x x u253 106/112 106/106 108/108 112 VWF 157/157 x/x 157/187 157
Appendix 1. Types of genetic markers used in this studyMitochondrial DNA sequences
Y chromosome microsatellites Y chromosomes are only present in males of mammals and are always directly inherited from the father. Consequently, any polymorphic genetic marker on this chromosome could allow to follow the paternal lineage in the same way as mtDNA can reveal the maternal lineage. Olivier et al. (1999) were able to isolate some DNA sequences that would amplify only in male dogs and thus would be expected to be in the Y chromosome. One of the sequences reported contained one microsatellite. After constructing primers specific for the amplification of that microsatellite, two PCR products were obtained, both of them variable in male dogs, suggesting that the microsatellite was duplicated inside the Y chromosome. The analysis of the variability in these two copies of the microsatellite can be used to infer paternal lineages. However, as for the mtDNA, the information obtained is only partial (i.e. paternal lineage whereas mtDNA is maternal lineage). Autosomal microsatellites
|
|
|
||
