|
Mitochondrial DNA Diversity in EuropeAntti Sajantila
We have studied mitochondrial DNA sequence variation from 637 individuals sampled in 14 European populations. While most populations show only little differentiation with respect to each other, the Saami distinguish themselves by a comparatively large amount of sequence differences when compared to the other populations, by a different distribution of sequence diversity within the population, and by the occurrence of particular sequence motifs. These findings apply also to four Saami subgroups studied from Saami living in Finland, Sweden and Norway. Thus, the Saami seem to have a long common history distinct from other European populations. No significant substructure according to geographical regions or linguistic affiliations could be demonstrated in European populations analyzed here. Mitochondrial DNA (mtDNA) has several advantages that make it useful for studies of biological history of human populations. First, it is almost exclusively maternally inherited (1). Therefore, the analysis of molecular lineages is not perturbed by recombination as is the case for most of the chromosomal polymorphisms (e.g. microsatellites). Second, the high rate of evolution of the mtDNA (2,3) permits the accumulation of substantial numbers of substitutions over time-periods relevant for human population history. In addition, each cell contains several hundreds or thousands of copies of mtDNA, compared to only two copies of autosomal chromosomes. For this reason mtDNA has proven to be retrievable from ancient human tissue samples, which did not allow the analysis of nuclear markers using DNA amplification methods (4). Thus, the mtDNA genome is a promising molecule also for forensic purposes in particular instances. In order to study genetic substructure in a particular geographical area, DNA sequence data from different regions of this area must be collected and compared with each other. To date, studies concerning mtDNA diversity in European populations have been relatively scarce. The majority of published mtDNA sequence data from European populations are from Indo-European linguistic groups. However, approximately 20-30 million Europeans speak Finno-Ugric languages (5). The principal idea behind our study was to investigate how linguistic barriers are reflected in mtDNA sequence comparisons among the European populations. We have determined a 360 bp hypervariable part of the mitochondrial control region (6) from five Finno-Ugric linguistic groups: the Saami, the Finns, the Karelians, the Estonians, and the Volga-Finnic (7). Theses sequences were compared to those of four Indo-European linguistic groups: the Icelanders, the Sardinians (8), the British (9), the Swiss (10); as well as the Basques (11), who form a linguistic isolate. DNA was extracted according to standard methods from blood samples of 115 Saami, 50 Finns, 83 Karelians, 28 Estonians, 34 Volga-Finnic, and 39 Icelanders. As far as possible, it was assured that individuals donating blood were unrelated. MtDNA sequence information from other populations were obtained from published studies. DNA Amplification and Sequencing PCR was performed in a programmable heating block (PE 4800, Perkin Elmer) using 20 nmol of each dNTP, 50 pmol of each primer (L15926: 5'-TCA AAC TTA CAC CAG TCT TGT AAA CC-3' and H 16498: 5'-CCT GAA GTA GGA ACC AGA TG-3') and 2 units of Taq DNA polymerase (Perkin-Elmer) in 50 ml of 50 mM Tris-HCl, pH 8.8, 15mM (NH4)2SO4, 1.5 mM MgCl2, 0.1 % triton X-100, 0.01% gelatin. In the PCR one of the primers used were labeled with biotin at its 5' end. The strands in the amplicon were separated using avidin coupled to a solid-phase (12) and subsequently sequenced according to Auto Read™ Sequencing kit protocol (Pharmacia, LKB Biotechnology, Uppsala, Sweden). The primers used for sequencing (L15997: 5'-CAC CAT TAG CAC CCA AAG CT-3' or H16401: 5'-TGA TTT CAC GGA GGA TGG TG-3') were labeled with fluorescein at their 5'end. The PCR and sequencing reactions were performed so in order that from each individual sequences could be determined for both directions. Alternatively, the Sequenase 2.0 (United States Biochemicals Corp., Cleveland, Ohio, USA) protocol was applied for radioactive sequencing using the solid-phase technique (12). Reaction products were electrophoresed at 75W for 2.5 hours through 6% polyacrylamide gel containing 7M urea. After electrophoresis the gels were rinsed with ddH20, dried under vacuum and exposed to Kodak XAR film for 24-72 hours. Sequences were read and aligned using ESEE software (13). The distributions of pairwise differences within and between populations, lineage sharing between populations and permutation tests for detecting population subdivision (14) were calculated using an unpublished program by Dr. A. von Haeseler (Ludwig Maximilians University, Munich, Germany). Genetic distances (D) between populations were calculated as follows: D=DBET- (DW1+DW2)/2, where DBET= mean pairwise difference between populations 1 and 2, DW1= mean pairwise difference within population 1, and DW2= mean pairwise difference within population 2. Negative distances were converted to zero. Pairwise Comparisons of mtDNA Sequences in Europe Table 1 shows the mean pairwise sequence differences within as well as among the populations studied. The pairwise differences within populations vary between 3.24 and 5.03, whereas those among pairs of populations vary between 3.49 and 5.29. The major finding of pairwise comparisons in European populations was that the overall amount of sequence difference within populations is not different from that between populations. This is in contrast to what is observed e.g. in African populations, where differences within three populations vary between 3.68 and 8.74, whereas those between populations range from 9.07 to 9.95 (10, 15). The only exception in European populations studied here is the Saami. They differ in the amount of sequence variation that they display when compared to other European populations. This becomes particularly apparent when the distances between populations are corrected for the within-population differences. The distances between the Saami and other groups is between 44 and 78, whereas distances between the others vary from 0 to 9.5 (Table 1). We also computed a permutation based test for detecting genetic differentiation between populations (14). In this test procedure, eight out of nine comparisons involving the Saami show that they are different from the other European populations (P of Kst =0.01), excepting only the Basques (P of Kst =0.059). Distributions of Pairwise Sequence Differences Within European Populations The distributions of pairwise sequence differences within populations are shown in Figure 1. With the exception of the Saami, all European populations have distributions that tend to have a single mode at 3 to 5 substitutions and have identical sequence pairs at frequencies of 5% or less. On the contrary, the Saami exhibit a "ragged" distribution of pairwise sequence differences with approximately 18% identical sequence pairs within them. Distributions of pairwise differences reflect demographic history of populations in ways that are not fully understood. One scenario that can give rise to a unimodal distribution of pairwise sequence differences is an expansion of population size, whereas populations that have been of constant size over long time can exhibit ragged distributions with no specific mode (16, 17). Therefore, our observations of the distribution of pairwise sequence differences support the idea that the Saami may have not experienced an expansion in size whereas other European populations have done so. However, since other demographic scenarios might also give rise to unimodal and multimodal distributions (17), the interpretation of population sizes in prehistoric times based on distribution of pairwise sequence differences remains only speculative. Saami Share Unique Sequence Motifs That Are Absent in Other Populations A distinct sequence motif consisting of a combination of three substitutions at positions 16144, 16189 and 16270 (6) was found to be almost exclusively characteristic for Saami sequences. 43 of 115 (37%) Saami mtDNA sequences share this motif and it was found in Saami living in Finland, Sweden as well as Norway, at frequencies ranging from 32% to 52% (Table 2 and Table 3). When other populations from around the world were studied for the presence of this motif (Table 3), it was found in 1 of 50 (2%) Finns and 5 of 83 (5%) Karelians, but in none of 62 other Finno-Ugric speakers. It is also absent among 282 Indo-European speakers, 45 Basques, 101 indigenous Siberians, 47 Japanese, 165 Native Americans and 362 Africans. Overall, this motif is shared by all four Saami groups studied, but is almost exclusively restricted to Saami. Since the Finns and the Karelians live in the same or close geographical region with the Saami, the presence of the motif at low frequency in the Finns and the Karelians is most likely due to a mixture between these populations and the Saami. Of those Saami who do not carry sequences characterized by the three substitutions, 74% share a single substitution at position 16298 (C:T transition). This substitution is found in Saami populations with frequencies varying from 9 to 60% (Table 2 and Table 3); it exists in other populations as well, although generally at a lower frequencies (0-12.7%). In total, 83.5% of the Saami carry either the sequence motif or the substitution at position 16298. This indicates that the Saami share a common history distinct from that of other groups included in this study. Comparisons of Mitochondrial DNA Sequences Among Saami Groups When mtDNA sequences were compared within the Saami groups, the distributions of pairwise differences in all groups were similar in that they did not show a distinctive mode and had a high proportion of identical sequences (data not shown). However, Saami from Inari Lake (Finland) showed a mean of 6.52 substitutions within the population, whereas the mean numbers of substitutions within Skolt Saami (Finland), Saami from Norrboten (Sweden), and from Karasjok (Norway) were 3.12, 3.25, and 2.76, respectively (Table 4). In comparisons between the groups, Inari Lake Saami stand out with mean distances of 5.13 to 5.71, whereas comparisons between other Saami groups vary between 3.08 and 3.19. Thus, individual Saami groups can differ substantially from other groups with respect to mtDNA sequence diversity. When tested for genetic differentiation (14), the Inari Lake Saami were found to differ from the Skolt Saami and the Saami from Norrbotten (P of Kst=0.01), whereas the other Saami groups did not reveal any significant differences. The differentiation that exists among Saami groups represents a further distinction between those and other European populations, which show little if any differentiation among them, and very similar amounts of intrapopulation differences (10). Patterns of Shared Mitochondrial Lineages We also investigated the pattern of shared mtDNA sequences between European populations. Figure 2 shows results from these analyses. The sequences shared between the Saami and other populations tend to be more frequent in one of the groups than in the other. In addition, the total number of shared lineages is small, varying from 4 to 7 (Fig. 2). This is the case in comparisons both with groups that are similar to the Saami in that they speak Finno-Ugric languages and with other groups. In contrast, the number of shared sequences is substantially higher in comparisons among European populations that do not involve Saami, ranging from 7 to 22 (Fig. 2). Furthermore, the frequency of each lineage in each respective population is more equal than for lineages shared with the Saami. Whereas the pattern observed in comparison between Saami and the other European populations may reflect sharing of lineages due to a limited amount of gene flow between populations and/or convergent evolution of the mtDNA, the pattern observed in the other comparisons is consistent with the claim that the populations are derived from a gene pool that shares of common history for the mtDNA sequences. Independent of their linguistic or geographical origin, most of the European populations show no major characteristics in their mitochondrial gene pool that are not also found in the other populations. In contrast, the Saami mtDNA sequences differ substantially from those of the other European populations studied. The data thus strongly suggest that the Saami have a genetic history distinct from other European populations. The Saami speak several mutually non-intelligible languages and have different current subsistence patterns (5). They are linguistically and culturally diverse and it may be asked to what extent the mitochondrial gene pool is homogeneous among Saami. As shown above, two mtDNA lineages, common among Saami but rare or absent elsewhere in Europe, exist in all Saami groups studied. Furthermore, the mean pairwise difference among mtDNA sequences within the Saami groups (with the exception of the Inari Lake Saami) varies between 3.08 and 3.19, whereas between other European groups it varies between 3.70 and 5.50. Thus, the Saami not only have a genetic history distinct from that of other Europeans but also share this history among themselves. The unique position of the Saami in the genetic landscape of Europe could mean that they are an old population in Europe, which diverged from other European populations prior to subsequent linguistic and cultural diversification. Alternatively, they may have come to Europe from another, currently unknown region. However, the latter possibility does not, derive support from the survey of human populations that has so far failed to identify the mtDNA lineages typical of Saami at any appreciable frequencies in other world populations. (Table 3). Genetic and Language Evolution Our analyses of mtDNA lineages show no congruence between linguistic and genetic affiliations of populations studied here; e.g., the Finns and the Saami speak linguistically closely related languages and live in geographically overlapping areas, but are genetically distant. The Finns fall into the common mitochondrial gene pool with the populations that speak Indo-European languages, whereas the Saami are outliers in the European mtDNA pool. Such a disagreement with linguistic and genetic analyses has been observed not only in Europe (7, 18) but also in the Americas (19). Patterns of relationships of genetic lineages may actually reflect temporally deeper affinities of populations relatively unperturbed by founder effects and other changes in effective population size. Thus, the analysis of genetic lineages may be the instrument of choice to study population history at a time-depth that exceeds the linguistic history. The time-depth that a particular genetic lineage reflects will be determined by its evolutionary rate. For the mtDNA region studied here, estimates of the evolutionary rate vary between approximately one substitution in 13,000 years (3,20) to one substitution in 5,000 years (21). Thus, the temporal perspective in which the distinct history of the Saami and the shared history of other populations in Europe must be viewed in the order of tens of thousands of years. This study was supported by the Finnish Cultural Foundation. I would like to thank Oliva Handt, Matthias Krings, Hendrik Poinar and Abdel Halim Salem for constructive criticism concerning my presentation and manuscript. I am also grateful to Svante Pääbo for ideas and enthusiasm concerning this project. Finally, I would like to thank Promega Corporation for their kind invitation to the Sixth Symposium on Human Identification. A more detailed information of the data summarized in this manuscript can be found in references 7 and 18.
*Population abbreviations are as follows: saa=Saami; fin=Finns; kar=Karelians; est=Estonians; vol=Volga-Finnic; swi=Swiss; bri=British; sar=Sardinians; ice=Icelanders; and bas=Basques. Note: The positions refer to Andersson sequence (CAMB) and are indicated as e.g. 126 referring to position 16126, 129 to position 16129, etc. (6). The following sites are variable:
Table 3. Presence of Saami specific motif and substitution at position 16298 in various populations.
(a)Pult et al. 1994 (10); (b) Piercy et al. 1993 (9); (c) Di Rienzo and Wilson 1991 (8); (d) Bertranpetit et al. 1995 (11); (e) Vigilant et al. 1991 (15) and Watson et al. unpublished data; (f) Ward et al. 1991 (3) and unpublished data; (g) Horai et al. 1990 (22).
Populations abbreviations are as follows: Inari=Saami from Inari Lake region, Finland; Skolt=Skolt Saami from eastern Finland; Norrbotten=Saami from Norrbotten, Sweden; Karasjok=Saami from Karasjok, Norway.
Distribution of pairwise sequence differences within 10 European populations. The comparisons were performed for 115 Saami, 50 Finns, 83 Karelians, 28 Estonians, 34 Volga-Finnic speakers, 100 British, 74 Swiss, 39 Icelanders, 51 Sardinians and 45 Basques. The abscissa gives the number of substitutions in a pair and the ordinate of the relative frequency (%) of pairs. Sequences are from DiRienzo and Wilson 1991 (Sardinians), Piercy et al. 1993 (British), Pult et al. 1994 (Swiss) and Bertranpetit et al. 1995 (Basques). Previously unpublished Saami, Finnish, Karelian, Estonian, Volga-Finnic and Icelandic sequences have been deposited in the EMBL Data library under accession numbers X90001-X90115, X90116-X90165, X90166-X90248, X90249-X90276, X90277-X90310, and X90311-90349, respectively.
Pattern of mitochondrial lineage sharing among 115 Saamis, 183 Finno-Ugric speakers (50 Finns, 83 Karelians, 28 Estonians and 34 Volga-Finnic speakers, but excluding the Saamis), 282 Indo-European speakers and 45 Basques. The lineages shared between the groups analyzed are given on the abscissa starting with them most common lineage. The relative proportion of each of the lineages in the respective population is shown on the ordinate. Go to proceedings home page
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
