Ilya.S
06-20-2010, 01:20 PM
I haven't read it yet(will read it a bit later) but it think it must be interesting.
__________________________________________________ _______________
Article Excerpt
The origin of Eastern Slavs (Russians, Ukrainians, and Belorussians) is a complex issue that has been debated by linguists, anthropologists, archeologists, and historians for more than 300 years. The Slavonic colonization of eastern Europe [which was most intensive in the 6th-7th centuries A.D. according to archeological data (Sedov 1995)] has entailed changes in the anthropological composition of Slavs because of their interpopulation contacts with Baltic tribes in the northwest, Finno-Ugric tribes in the north and east, and Iranian (or Iranic-speaking) tribes in the south of eastern Europe (Alekseeva and Alekseev 1989). Notwithstanding the vast factual material concerning ancient and modern Eastern Slavonic populations, accumulated by archeologists, anthropologists, and historians, many problems still remain unsolved, including the character of interactions between Slavs and autochthonous populations of eastern Europe.
Previous genetic studies of classical genetic markers (Rychkov et al. 1999) and anthropological data (Alekseeva 1973; Rychkov and Balanovska 1988) have shown geographic differences between Russian populations and have allowed researchers to recognize at least two groups of Russians. It has been suggested that the most western Russian populations appear to be descendants of the Slavs, whereas northern, eastern, and southern Russian populations are the result of admixture between Slavonic tribes and pre-Slavonic populations of eastern Europe (Rychkov and Balanovska 1988). In addition, principal components analyses performed on the basis of anthropological data revealed that modern Russians are represented by two regional groups of populations, southeastern and northwestern (Alekseeva 1999). Using computer cartography of anthropological and genetic data, Balanovsky and Balanovska (2002) recently identified a latitudinal (north-south) trend as the main geographic pattern of the Russian gene pool variation.
Although there are several examples of research on Russian populations using nuclear DNA (Limborska 1999; Spitsyn et al. 2001), Y-chromosome (Rosser et al. 2000; Wells et al. 2001), and mitochondrial DNA (mtDNA) variability (Orekhov et al. 1999; Malyarchuk et al. 2002), no detailed mtDNA and Y-chromosome studies have been performed to analyze the differentiation of regional populations of Russians. Previously, population samples of Russians were analyzed in different ways: Some studies covered only mtDNA hypervariable segment I (HVSI) sequences (Malyarchuk et al. 1995; Orekhov et al. 1999); other studies also included coding region RFLPs (Malyarchuk and Derenko 2001; Malyarchuk et al. 2002). As for the Y chromosome, a limited number of Russian population samples have been investigated but only in the framework of global Y-chromosome marker surveys in Europeans. These studies focused on placing Y-chromosome variation in Russians in a broader phylogeographic perspective (Karafet et al. 1999; Rosser et al. 2000; Wells et al. 2001). Given the large number of the Russian people and the complexity of their population history, it is necessary to perform a more detailed study of Russian populations inhabiting the European part of Russia (the East European, or Russian, plain), where the first stages of Russian ethnic history occurred. To obtain a better estimate of the level of genetic differentiation in Russian populations, we present here mtDNA and Y-chromosome data in Russian individuals (325 mtDNA and 338 Y-chromosome samples) originating from the indigenous geographic area of the Russian population.
Materials and Methods
Samples. For the mtDNA variation study, a sample of 325 Russian individuals (125 females and 200 males) from the following 5 populations of the European part of Russia was analyzed (sample sizes are given in parentheses): Tula (73), Kaluga (71), Vladimir (72), Yaroslavl (41), and Pskov (68) regions (Figure 1). All individuals were maternally unrelated, and their mothers had been born in the area considered for this study. For the Y-chromosome variation study, 338 Russian males from the same populations [Tula (44), Kaluga (42), Vladimir (51), Yaroslavl (23), and Pskov (40) regions] and three additional populations from the Belgorod (44), Orel (36), and Nizhniy Novgorod (58) regions were analyzed (Figure 1). All of these individuals were paternally unrelated and originated from the area considered for this study. Appropriate informed consent was obtained from all participants in this study. All samples were randomly taken mainly from towns of the corresponding administrative areas of Russia.
HVSI Sequencing and RFLP Analysis of mtDNA. Genomic DNA was extracted from whole blood by a standard phenol/chloroform method. Hypervariable segment I (HVSI) of the mtDNA noncoding control region was amplified and sequenced as described elsewhere (Malyarchuk et al. 2002). The nucleotide sequences of HVSI from position 15991 to 16400 were determined and compared with the Cambridge Reference Sequence (CRS) (Anderson et al. 1981; Andrews et al. 1999). Nucleotide positions showing point indels and transversions located between positions 16180 and 16193 were excluded from further analysis.
RFLP typing was performed using restriction endonuclease analysis of PCR-amplified mtDNA fragments; the same primer pairs and amplification conditions as described elsewhere by Torroni et al. (1996) and Finnila et al. (2000) were used. The samples were typed for a restricted set of RFLPs (Table 1) that were diagnostic of all major Eurasian clusters, on the basis of the hierarchical mtDNA RFLP scheme (Macaulay et al. 1999; Richards et al. 2000; Yao et al. 2002). HVSI sequences, belonging to haplogroups N1a, N1b, U1, U3, U5, U7, M1, L1b, and L3b were identified on the basis of the HVSI motif classification (Richards et al. 2000; Salas et al. 2002; Yao et al. 2002).
Analysis of Y-Chromosome Binary Polymorphism. Twelve binary markers [RPS4Y, SRY-8299, M89, 12f2, M9, M20, 92R7, SRY-10831, DYS199. Tat, LLY22g, and the Alu insertion polymorphism (YAP) at DYS287] defining the structure of the North Eurasian Y-chromosome variation (Underhill et al. 2001; Y Chromosome Consortium 2002) were analyzed in this study (Figure 2). The binary markers were typed according to methods described elsewhere: M9 (Underhill et al. 1997) was typed according to the method of Hurles et al. (1998); YAP (Hammer 1994) was typed according to the method of Hammer and Horai (1995); SRY-10831 and SRY-8299 (Whitfield et al. 1995) were typed according to the methods of Kwok et al. (1996) and Santos et al. (1999), respectively; 92R7 (Mathias et al. 1994) was typed according to the method of Hurles et al. (1999); 12f2 (Casanova et al. 1985) was typed according to the method of Rosser et al. (2000); Tat was typed according to the method of Zerjal et al. (1997); LLY22g and M20 were typed according to the method of Underhill et al. (2000); RPS4Y-711 (Bergen et al. 1999) was typed according to the method of Kayser et al. (2000); DYS199 (Underhill et al. 1997) was typed according to the method of Santos et al. (1999); and M89 (Underhill et al. 2000) was typed according to the method of Ke et al. (2001).
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
SNP variants at haplogroup-specific sites were determined using PCR RFLP assay, followed by an electrophoretic analysis of DNA fragments in 8% polyacrylamide gels and visualization through ethidium bromide staining. The Y-chromosome SNP haplogroup nomenclature used is that recommended by the Y Chromosome Consortium (2002).
Statistical Analysis. Genetic variation was analyzed using methods implemented in the Arlequin 1.1 software (Schneider et al. 1997). The statistical significance of [F.sub.ST] values was estimated by permutation analysis using 10,000 permutations. Multidimensional scaling analysis of pairwise interpopulation [F.sub.ST] values was performed by means of the software package Statistica (StatSoft Inc., Tulsa, Oklahoma).
mtDNA HVSI data from the following Russian populations were used for comparison (sample sizes are given in parentheses): the Stavropol (62), Orel (76), Saratov (63) regions (Malyarchuk et al. 2002); the Kursk (34) and Kostroma (55) regions (Orekhov et al. 1999); and the large Cossack village of Razdorskaya in the Rostov region (25) (Richards et al. 2000). Additional samples from 9 populations across central and eastern Europe were also included in the comparative analysis. These populations were Adygei (50) from North Caucasus (Macaulay et al. 1999); Tatars (45) and Mari (60) from the Middle Volga region (Orekhov 2002); Austrians (101) (Parson et al. 1998); Finns (50), Estonians (47), and Karelians (83) (Sajantila et al. 1995); Poles (436) (Malyarchuk et al. 2002); and Lithuanians (120) (Kasperaviciute and Kucinskas 2002).
In a comparative analysis of Y-chromosome variation data, the Russian populations of northern Russia [Northern Russians (49) and Pomors (28) (Wells et al. 2001)] were also used. In addition, samples from nine central and eastern European populations (Rosser et al. 2000) were also analyzed: Mari (48) and Chuvash (17) from the Middle Volga region; Finns (57), Estonians (207), Lithuanians (38), Latvians (34), Ukrainians (27), Belorussians (41), and Poles (112).
Results
mtDNA and Y-Chromosome Haplogroups in Russians. The analysis of HVSI variation in combination with RFLP typing of the coding region haplogroup-diagnostic sites allowed detection of 188 different mtDNA haplotypes in a total sample of 325 Russian individuals (Table 2). It was found that these haplotypes are clustered, according to human mtDNA classification (Macaulay et al. 1999; Richards et al. 2000; Salas et al. 2002; Yao et al. 2002), into haplogroups H, pre-V, HV*, R1, U1, U2, U3, U4, U5, U7, U8, K, J, T*, T1, I, N1a, N1b, X, W, C, D, G, M1, L1b, and L3b.
Table 3 summarizes the frequencies of the mtDNA haplogroups and subgroups found in the Russian populations studied. Russians present relatively high means of pairwise nucleotide differences between HVSI sequences. This parameter ranges from 4.74 in the Yaroslavl sample to 5.46 in the Vladimir sample and is in accordance with diversity estimates found in European and Middle Eastern populations [from 2.95 in Basques to 7.08 in populations of the Middle East according to Comas et al. (1997)].
The majority (97.8%) of mtDNA haplotypes found in Russians fall into major West Eurasian haplogroups. The frequency of East Asian--specific haplotypes in Russians is very low (1.5% on average). All of them belong to macrohaplogroup M (C, D, G). A single haplotype found in Russians belongs to specific haplogroup M1, which is distributed mainly in East Africa (Somalia and Ethiopia) and the Red Sea region (Richards et al. 2003). However, the low frequencies of haplogroup M1 were observed in southern Europe and the Caucasus region (Richards et al. 2000), from where this haplogroup may have been introduced into the Russian population. An interesting finding of the present study is the occurrence of African-specific mtDNA lineages (L1b and L3b) in the Russian gene pool. Their frequency is very low (0.6%), but this estimate should be lowered to 0.3% to take into account the previous data on mtDNA variability in Russian populations (Orekhov et al. 1999; Malyarchuk et al. 2002). These results indicate that the appearance of African mtDNAs among Russians was probably casual.
Table 4 reports the frequencies of the Y-chromosome haplogroups found in Russian samples. The pattern of Y-chromosome haplogroup distributions in Russians is typical for northeastern European populations (Rosser et al. 2000). The principal diagnostic haplogroups for Russians are haplogroups R1a and N3, which occur in the studied populations at a total frequency of 50-72%. The remaining Y-chromosome haplogroups represent mostly paraphyletic groups distributed with similar frequencies in European populations.
It should be noted that the frequency of paragroup P* [or haplogroup 1 according to Rosser et al.'s (2000) nomenclature] in Russians (2.3-19%) reflects the known pattern of distribution, with a high incidence of P* in western European populations (especially those of Celtic origin) and a lower frequency in the east (Zerjal et al. 2001). Haplogroups E* and J* were found only in the most southwestern Belgorod population at frequencies of 6.8% and 13.6%, respectively. These haplogroups probably are associated with the contributions of Neolithic farmers to the European gene pool, because both haplogroups E* and J* show the highest frequencies in Anatolia and southern Europe (Rosser et al. 2000; Semino et al. 2000; Underhill et al. 2001). Thus the lack of haplogroups E* and J* in most Russian populations studied suggests a minor genetic effect from southern European populations. A single haplotype revealed in Russians from the Belgorod population belongs to the East Asian--specific haplogroup C*, which is widespread in Central Asia (Zerjal et al. 2002). Therefore Y-chromosome variation data allow us to speculate that East Eurasian male gene flow in Russian populations was insignificant (0.3%).
In general, the data indicate that the Russian male gene pool preserves the main peculiarity of Y-chromosome diversity in central-eastern European populations. In the first instance, it is the high frequency of haplogroup R1a that is rare in western Europe but widely present in eastern and central Europe, being found at a frequency of 50-60% in Poles, Hungarians, Ukrainians, and Belorussians (Rosser et al. 2000; Semino et al. 2000; Wells et al. 2001; Behar et al. 2003). Analysis of the haplogroup R1a distribution in Russian populations reveals a relatively homogeneous pattern, with a population frequency of 50-60%, except for the most northwestern Pskov population and the most eastern Nizhniy Novgorod population from the Volga region, where this haplogroup is observed with a lower frequency of about 35%.
It is noteworthy that among the Russian populations studied, the Pskov population shows the highest frequency (35%) of haplogroup N3, which is characteristic of all Uralic- and Baltic-speaking populations north and east of the Baltic Sea (Rosser et al. 2000; Zerjal et al. 2001). This haplogroup is considered a signature of Uralic Finno-Ugric-speaking tribes migrating to northern Europe about 5,000 years ago (Passarino et al. 2002). The presence of haplogroup N3 in Russians suggests that some admixture between Slavonic and Finno-Ugric and/ or Baltic tribes occurred during the colonization of the East European plain by Slavs in the early Middle Ages and later. In addition, the distribution pattern of haplogroup N3 in other Slavonic-speaking populations suggests that proto-Slavs probably did not carry this lineage at a substantial frequency (Barac et al. 2003).
Differentiation of mtDNA and Y-Chromosome in Russian Populations. In order to study differentiation of Russian populations, we performed an analysis of molecular variance (AMOVA) separately on the level of mtDNA haplogroups and HVSI sequences. Analysis of between-population differentiation based on the frequencies of the mtDNA haplogroups revealed that 1.4% of the variation was due to differences among populations ([F.sub.ST] = 1.4%, p = 0.007). Significant pairwise [F.sub.ST] differences (p < 0.05) were found only between the Tula population and samples from Kaluga, Vladimir, and Pskov. In addition, the Vladimir and Pskov populations differed at a significant level. One should note, however, that the level of between-population differentiation was considerably lower when additional data on the mtDNA haplogroup frequencies in the Stavropol, Orel, and Saratov populations were taken into account ([F.sub.ST] = 0.6%, p = 0.048).
The AMOVA results of the mtDNA HVSI sequence data show that when the five Russian populations studied here were treated as a single group, 99.28% of the total variance was within populations ([F.sub.ST] = 0.72%, p = 0.009). Similar results were obtained ([F.sub.ST] = 0.42%, p = 0.005) when six additional Russian samples from the Stavropol, Orel, Saratov, Kostroma, Kursk, and Rostov regions were taken into account. Therefore the AMOVA results indicate the low but statistically significant level of mtDNA differentiation in Russian populations.
To further investigate genetic relationships between Russian populations, we used additional published data on the mtDNA HVSI variability in eastern and central European populations. These were the North Caucasus population of Adygei, Finno-Ugric-speaking Mari and Turkic-speaking Tatars from the Middle Volga region, North European Finno-Ugric populations of the Finns, Estonians, and Karelians, and the Baltic-speaking population of Lithuanians. Austrians and Poles were chosen because of their geographic location relative to the putative Central European homeland of the Slavs in the early Middle Ages. The AMOVA results showed that there is a low but statistically significant level of differentiation between the European populations analyzed ([F.sub.ST] = 0.41%, p = 0).
The multidimensional scaling analysis performed on the basis of pairwise [F.sub.ST] values (Table 5) revealed that Russian populations do not cluster together (Figure 3). It is noteworthy that all southern and western Russian populations (Stavropol, Saratov, Rostov, Kursk, Orel, and Kaluga) as well as Poles, Austrians, Lithuanians, and Estonians are separated from other, mostly northern and eastern Russian populations (such as Tula, Vladimir, Yaroslavl, Kostroma, and Pskov), which are clustered together with Finns, Karelians, Mari, Tatars, and Adygei. In the latter group of populations, Kostroma and Pskov Russians cluster more with Finno-Ugric-speaking Karelians and Mari than they do with the other Russian populations. The Tula, Vladimir, and Yaroslavl populations are close together but separated from the remaining Russian populations. In general, the results of the multidimensional scaling analysis of the mtDNA data indicate that, despite the low values of between-population differentiation, Russian populations appear to be heterogeneous.
The AMOVA results of Y-chromosome binary data in eight Russian populations (i.e., based on the haplogroup frequencies) indicate that the highest fraction of variability was due to within-population differences (97.53%), but a substantial fraction was due to differences among populations (2.47%, p = 0.0057). However, significant pairwise [F.sub.ST] differences (p < 0.05) were found only between the Pskov population and the rest of the Russian populations, with the exception of the Nizhniy Novgorod and Orel populations. In addition, the Nizniy Novgorod sample differs significantly from the Vladimir, Belgorod, and Yaroslavl populations. Similar results ([F.sub.ST] = 3.09%, p = 0) were obtained when two additional Russian samples from the north of European Russia (Northern Russians and Pomors) were included in analysis. The Russian Pomors sample was found to be distinct from all the Russian populations except for Pskov, Northern Russians, and Orel. Meanwhile, the Northern Russian sample differs significantly only from the Nizhniy Novgorod and Yaroslavl samples.
[FIGURE 3 OMITTED]
To provide the broader context for the study of the subdivision of Russian populations based on the Y-chromosome SNP [F.sub.ST] distances, we considered additional European populations in the analysis: Slavonic-speaking Poles, Ukrainians, and Belorussians; Baltic-speaking Latvians and Lithuanians; Finno-Ugric-speaking Finns, Estonians, and Mari; and Turkic-speaking Chuvash [all the data are from Rosser et al.'s (2000) study]. The AMOVA results show a significant level of differentiation between the European populations analyzed ([F.sub.ST] = 7.04%, p = 0). The multidimensional scaling analysis based on Y-chromosome haplogroup [F.sub.ST] values (Table 6) revealed that the strongest division was between Pskov and Pomor Russians together with all Finno-Ugric and Baltic-speaking populations, on the one hand, and the remaining Russian populations clustered together with other Slavonic-speaking groups, on the other hand (Figure 4). As for Baltic-speaking populations, it is well known that the Latvians and Lithuanians show genetic similarity with Finno-Ugric populations based on the increased frequency of haplogroup N3 (Rosser et al. 2000; Zerjal et al. 2001). Similarly, clustering of northern Pskov and Pomor Russian populations with Finno-Ugric and Baltic ones may be explained by increased frequency of haplogroup N3 found in these populations.
[FIGURE 4 OMITTED]
In general, Y-chromosome and mtDNA variation data indicate that the differentiation of the geographically distinct Russian populations appears to be more prominent in the male than in the female lineages ([F.sub.ST] values are 3.09% and 0.42%, respectively). Such a pattern has been found before in many human populations and is usually explained by patrilocal residence (Seielstad et al. 1998; Oota et al. 2001; Kayser et al. 2003). Thus the discordance between mtDNA and Y-chromosome variation in Russian populations may be a consequence of cultural features, such as patrilocality and the patrilineal social system that Russians practiced in the past, especially before the peasant reforms of the first half of the 20th century (Alekseev and Pershits 1999). In modern times, however, these features are preserved as a community tradition only in some rural localities of Russians in the north of eastern Europe (Alekseev and Pershits 1999). However, we cannot exclude other reasons (e.g., sex-specific migration or genetic drift in isolated populations) to explain why both genetic loci, Y chromosomes and mtDNA, can provide different pictures. Therefore the problem of contrasting patterns for Y-chromosome and mtDNA differentiation in Russian populations needs to be studied in more detail.
Discussion
Y-chromosome variation data obtained in the present study demonstrate that the Russian gene pool appears to be close to central-eastern European populations. The data show a high frequency of haplogroups R1a and N3. The high incidence of R1a haplotypes in Eastern Slavs and in Baltic and some Finno-Ugric populations of eastern Europe is a well-established fact (Rosser et al. 2000; Wells et al. 2001; Zerjal et al. 2001), but little is known about the spread of haplogroup N3 in different Russian populations. Our data indicate that this haplogroup is unevenly distributed in Russian populations, with frequencies ranging from 4.8% to 35%. It is noteworthy that in most of the populations studied (with the exception of the Pskov sample) the frequency of haplogroup N3 does not exceed 20%. An important feature distinguishing Russians from their eastern European neighbors is almost complete absence of paragroup N*, which is ancestral to haplogroup N3. In Europe the highest frequencies of paragroup N* (up to 42.5%) were detected in Volga Finno-Ugric populations: Udmurts and Mari (Rosser et al. 2000; Khusnutdinova et al. 2002). Therefore Y-chromosome data suggest that the Russian male gene pool is characterized by its own structural features, although genetic influences on Russians from Uralic- or Baltic-speaking populations are also highlighted. The latter is clearly defined in the most northern Russian populations, such as Pskov Russians and Pomors. A possible explanation for this phenomenon includes replacement of an earlier Uralic or Baltic language in populations of the north of eastern Europe by the present Russian language, occurring with a low input of Slavonic mtDNA and Y-chromosome haplotypes. The high frequency of haplogroup N3 in Pskov and Pomor Russians (and probably in other northern Russian populations) is striking and should be investigated further using Y-chromosome STR analysis and a comparison with published N3-associated Y-chromosome STR haplotype data in Finno-Ugric and Baltic populations. These data should be studied with respect to the hypothesis that the genetic history of N3 Y chromosomes in Baltic-speaking populations is distinct from that of the Uralic speakers and "that there were two distinct early migrations of haplogroup 16 into Europe" (Zerjal et al. 2001). To test this and other hypotheses concerning the recent historical interactions between eastern European populations, more populations need to be studied for more genetic markers.
The present study demonstrates that Russian populations contain relatively high levels of mtDNA sequence diversity (in the means of pairwise nucleotide differences), but the level of [F.sub.ST]-based between-population differentiation is low. However, the results of the multidimensional scaling analysis performed on the basis of pairwise [F.sub.ST] values for mtDNA HVSI sequences suggest that the Russian populations can be differentiated into subregions. The multidimensional scaling analysis revealed that Russian populations analyzed in combination with other central and eastern European populations do not cluster together and that populations from the southern and western parts of Russia (such as Stavropol, Rostov, Kursk, Orel, Kaluga, and Saratov) are separated from eastern and northern populations (Vladimir, Tula, Yaroslavl, Kostroma, and Pskov). Interestingly, southwestern Russian populations demonstrate genetic similarities with a set of central and northern European populations of different linguistic affiliation [Slavonic, Baltic, Germanic, and even Finno-Ugric (Estonians)], whereas northeastern Russian populations cluster together with eastern and northern European populations speaking not only Finno-Ugric languages but also Turkic (Tatars) and North Caucasus (Adygei) ones.
The multidimensional scaling analysis performed on the basis of pairwise [F.sub.ST] values for Y-chromosome haplogroup data shows a somewhat different picture (for mtDNA sequences) of population differentiation in Russia. The most important subdivision was found only between northern populations of Pskov and Pomor Russians and the rest of the Russian populations studied. Historically, this observed discrepancy in the depth of penetration of mtDNA and Y-chromosome lineages characteristic for the most southwestern Russian populations into the east and north of eastern Europe may indicate [in accordance with anthropological data (Rychkov and Balanovska 1988)] that only the most western Russian populations appear to be descendants of the Slavs, whereas northern and eastern Russian populations seem to be the result of an admixture between Slavonic tribes and pre-Slavonic populations of eastern Europe. In addition, the data allow one to assume the possibility that Slavonic male lineages penetrated the original eastern European populations further than mtDNA lineages. One should note, however, that this scenario should be tested with additional mtDNA and Y-chromosome data in multiple Russian-, Finno-Ugric-, and Turkic-speaking populations of eastern Europe, which are poorly characterized genetically.
In addition, to estimate the degree of population replacement in eastern Europe associated with the Slavonic colonization starting in the early Middle Ages [6th-7th centuries A.D. according to archeological data (Sedov 1995)], it would be important to perform analyses of multiple samples from small urban areas, because such an approach seems to be informative in genetic history studies (Capelli et al. 2003).
Table 1. RFLPs Analyzed to Identify Major Eurasian mtDNA Haplogroups Haplogroup Characteristic Restriction Site(s) West Eurasian HV - 14766 MseI H - 14766 MseI, - 7025 AluI pre*V1 - 14766 MseI, - 15904 MseI, + 4577 NlaIII pre*V2 - 14766 MseI, + 15904 MseI, + 4577 NlaIII V - 14766 MseI, + 15904 MseI, - 4577 NlaIII U + 12308 HinfI K + 10394 DdeI, + 12308 HinfI, - 9052 HaeII U2 + 12308 HinfI, + 15907 RsaI U4 + 12308 HinfI, + 4643 RsaI U8 + 12308 HinfI, - 7055 AluI J + 10394 DdeI, - 13704 BstOI T + 13366 BamHI, + 15606 AluI TI + 13366 BamHI, + 15606 AluI, - 12629 AvaII NI - 12498 NlaIII I - 4529 HaeII, + 8249 AvaII, + 10032 AluI, + 10394 DdeI, - 12498 NlaIII W + 8249 AvaII, - 8994 HaeIII X - 1715 DdeI, + 14465 AccI East Eurasian M + 10394 DdeI, + 10397 AluI C + 10394 DdeI, + 10397 AluI, - 13259 HincII/ + 13262 AluI D + 10394 DdeI, + 10397 AluI, - 5176 AluI E + 10394 DdeI, + 10397 AluI, - 7598 HhaI G + 10394 DdeI, + 10397 AluI, + 4830 HaeII/ + 4831 HhaI A + 663 HaeIII B 9-bp deletion F1 - 12406 HpaI/HincII RFLP markers are given relative to the revised version of the Cambridge Reference Sequence (Andrews et al. 1999) belonging to haplogroup H. Site gains and losses are indicated by a plus sign (+) or a minus sign (-). Table 2. mtDNA Haplotypes in Russians Obtained by Combining the Data on the HVSI Sequence and Haplogroup-Specific RFLPs HVSI Variation (-16000 Relative to CRS Sequence Haplo- Tula Kaluga Vladimir Numbering) group (n = 73) (n = 71) (n = 72) Cambridge Reference Sequence (CRS) H 11 6 7 70 93 311 H 1 93 311 H 2 93 H 126 H 129 H 1 147 H 1 162 H 51 162 H 1 51 162 213 256 H 1 162 311 H 2 168 H 170 H 1 176 H 1 169 176 H 180 H 1 183delA H 189 H 209 H 2 221 H 1 1 222 H 234 H 1 86 239 H 256 352 H 1 261 H 2 261 311 H 2 261 335 H 1 274 H 274 317 H 1 241 274 287 H 278 293 311 H 93 278 293 311 H 1 134 278 293 311 H 1 172 278 293 311 H 1 224 278 293 311 H 1 92 140 265 293 311 H 4 3 286 H 1 235 291 H 249 291 H 1 293 H 1 296 H 300 H 1 304 H 3 3 192 304 H 1 209 304 H 240 304 H 1 258AT 304 H 1 304 352 H 1 304 390 H 1 304 400 H 1 1 294CA 304 H 294 304 H 1 271 294 304 H 311 H 1 311 319 H 320 H 1 354 H 1 5 193 354 H 1 189 356 H 1 188 189 356 H 75 111 183AC 189 209 356 H 1 183AC 189 193.1C 294 356 H 1 80 189 356 H 1 80 129 189 193.1C 356...
__________________________________________________ _______________
Article Excerpt
The origin of Eastern Slavs (Russians, Ukrainians, and Belorussians) is a complex issue that has been debated by linguists, anthropologists, archeologists, and historians for more than 300 years. The Slavonic colonization of eastern Europe [which was most intensive in the 6th-7th centuries A.D. according to archeological data (Sedov 1995)] has entailed changes in the anthropological composition of Slavs because of their interpopulation contacts with Baltic tribes in the northwest, Finno-Ugric tribes in the north and east, and Iranian (or Iranic-speaking) tribes in the south of eastern Europe (Alekseeva and Alekseev 1989). Notwithstanding the vast factual material concerning ancient and modern Eastern Slavonic populations, accumulated by archeologists, anthropologists, and historians, many problems still remain unsolved, including the character of interactions between Slavs and autochthonous populations of eastern Europe.
Previous genetic studies of classical genetic markers (Rychkov et al. 1999) and anthropological data (Alekseeva 1973; Rychkov and Balanovska 1988) have shown geographic differences between Russian populations and have allowed researchers to recognize at least two groups of Russians. It has been suggested that the most western Russian populations appear to be descendants of the Slavs, whereas northern, eastern, and southern Russian populations are the result of admixture between Slavonic tribes and pre-Slavonic populations of eastern Europe (Rychkov and Balanovska 1988). In addition, principal components analyses performed on the basis of anthropological data revealed that modern Russians are represented by two regional groups of populations, southeastern and northwestern (Alekseeva 1999). Using computer cartography of anthropological and genetic data, Balanovsky and Balanovska (2002) recently identified a latitudinal (north-south) trend as the main geographic pattern of the Russian gene pool variation.
Although there are several examples of research on Russian populations using nuclear DNA (Limborska 1999; Spitsyn et al. 2001), Y-chromosome (Rosser et al. 2000; Wells et al. 2001), and mitochondrial DNA (mtDNA) variability (Orekhov et al. 1999; Malyarchuk et al. 2002), no detailed mtDNA and Y-chromosome studies have been performed to analyze the differentiation of regional populations of Russians. Previously, population samples of Russians were analyzed in different ways: Some studies covered only mtDNA hypervariable segment I (HVSI) sequences (Malyarchuk et al. 1995; Orekhov et al. 1999); other studies also included coding region RFLPs (Malyarchuk and Derenko 2001; Malyarchuk et al. 2002). As for the Y chromosome, a limited number of Russian population samples have been investigated but only in the framework of global Y-chromosome marker surveys in Europeans. These studies focused on placing Y-chromosome variation in Russians in a broader phylogeographic perspective (Karafet et al. 1999; Rosser et al. 2000; Wells et al. 2001). Given the large number of the Russian people and the complexity of their population history, it is necessary to perform a more detailed study of Russian populations inhabiting the European part of Russia (the East European, or Russian, plain), where the first stages of Russian ethnic history occurred. To obtain a better estimate of the level of genetic differentiation in Russian populations, we present here mtDNA and Y-chromosome data in Russian individuals (325 mtDNA and 338 Y-chromosome samples) originating from the indigenous geographic area of the Russian population.
Materials and Methods
Samples. For the mtDNA variation study, a sample of 325 Russian individuals (125 females and 200 males) from the following 5 populations of the European part of Russia was analyzed (sample sizes are given in parentheses): Tula (73), Kaluga (71), Vladimir (72), Yaroslavl (41), and Pskov (68) regions (Figure 1). All individuals were maternally unrelated, and their mothers had been born in the area considered for this study. For the Y-chromosome variation study, 338 Russian males from the same populations [Tula (44), Kaluga (42), Vladimir (51), Yaroslavl (23), and Pskov (40) regions] and three additional populations from the Belgorod (44), Orel (36), and Nizhniy Novgorod (58) regions were analyzed (Figure 1). All of these individuals were paternally unrelated and originated from the area considered for this study. Appropriate informed consent was obtained from all participants in this study. All samples were randomly taken mainly from towns of the corresponding administrative areas of Russia.
HVSI Sequencing and RFLP Analysis of mtDNA. Genomic DNA was extracted from whole blood by a standard phenol/chloroform method. Hypervariable segment I (HVSI) of the mtDNA noncoding control region was amplified and sequenced as described elsewhere (Malyarchuk et al. 2002). The nucleotide sequences of HVSI from position 15991 to 16400 were determined and compared with the Cambridge Reference Sequence (CRS) (Anderson et al. 1981; Andrews et al. 1999). Nucleotide positions showing point indels and transversions located between positions 16180 and 16193 were excluded from further analysis.
RFLP typing was performed using restriction endonuclease analysis of PCR-amplified mtDNA fragments; the same primer pairs and amplification conditions as described elsewhere by Torroni et al. (1996) and Finnila et al. (2000) were used. The samples were typed for a restricted set of RFLPs (Table 1) that were diagnostic of all major Eurasian clusters, on the basis of the hierarchical mtDNA RFLP scheme (Macaulay et al. 1999; Richards et al. 2000; Yao et al. 2002). HVSI sequences, belonging to haplogroups N1a, N1b, U1, U3, U5, U7, M1, L1b, and L3b were identified on the basis of the HVSI motif classification (Richards et al. 2000; Salas et al. 2002; Yao et al. 2002).
Analysis of Y-Chromosome Binary Polymorphism. Twelve binary markers [RPS4Y, SRY-8299, M89, 12f2, M9, M20, 92R7, SRY-10831, DYS199. Tat, LLY22g, and the Alu insertion polymorphism (YAP) at DYS287] defining the structure of the North Eurasian Y-chromosome variation (Underhill et al. 2001; Y Chromosome Consortium 2002) were analyzed in this study (Figure 2). The binary markers were typed according to methods described elsewhere: M9 (Underhill et al. 1997) was typed according to the method of Hurles et al. (1998); YAP (Hammer 1994) was typed according to the method of Hammer and Horai (1995); SRY-10831 and SRY-8299 (Whitfield et al. 1995) were typed according to the methods of Kwok et al. (1996) and Santos et al. (1999), respectively; 92R7 (Mathias et al. 1994) was typed according to the method of Hurles et al. (1999); 12f2 (Casanova et al. 1985) was typed according to the method of Rosser et al. (2000); Tat was typed according to the method of Zerjal et al. (1997); LLY22g and M20 were typed according to the method of Underhill et al. (2000); RPS4Y-711 (Bergen et al. 1999) was typed according to the method of Kayser et al. (2000); DYS199 (Underhill et al. 1997) was typed according to the method of Santos et al. (1999); and M89 (Underhill et al. 2000) was typed according to the method of Ke et al. (2001).
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
SNP variants at haplogroup-specific sites were determined using PCR RFLP assay, followed by an electrophoretic analysis of DNA fragments in 8% polyacrylamide gels and visualization through ethidium bromide staining. The Y-chromosome SNP haplogroup nomenclature used is that recommended by the Y Chromosome Consortium (2002).
Statistical Analysis. Genetic variation was analyzed using methods implemented in the Arlequin 1.1 software (Schneider et al. 1997). The statistical significance of [F.sub.ST] values was estimated by permutation analysis using 10,000 permutations. Multidimensional scaling analysis of pairwise interpopulation [F.sub.ST] values was performed by means of the software package Statistica (StatSoft Inc., Tulsa, Oklahoma).
mtDNA HVSI data from the following Russian populations were used for comparison (sample sizes are given in parentheses): the Stavropol (62), Orel (76), Saratov (63) regions (Malyarchuk et al. 2002); the Kursk (34) and Kostroma (55) regions (Orekhov et al. 1999); and the large Cossack village of Razdorskaya in the Rostov region (25) (Richards et al. 2000). Additional samples from 9 populations across central and eastern Europe were also included in the comparative analysis. These populations were Adygei (50) from North Caucasus (Macaulay et al. 1999); Tatars (45) and Mari (60) from the Middle Volga region (Orekhov 2002); Austrians (101) (Parson et al. 1998); Finns (50), Estonians (47), and Karelians (83) (Sajantila et al. 1995); Poles (436) (Malyarchuk et al. 2002); and Lithuanians (120) (Kasperaviciute and Kucinskas 2002).
In a comparative analysis of Y-chromosome variation data, the Russian populations of northern Russia [Northern Russians (49) and Pomors (28) (Wells et al. 2001)] were also used. In addition, samples from nine central and eastern European populations (Rosser et al. 2000) were also analyzed: Mari (48) and Chuvash (17) from the Middle Volga region; Finns (57), Estonians (207), Lithuanians (38), Latvians (34), Ukrainians (27), Belorussians (41), and Poles (112).
Results
mtDNA and Y-Chromosome Haplogroups in Russians. The analysis of HVSI variation in combination with RFLP typing of the coding region haplogroup-diagnostic sites allowed detection of 188 different mtDNA haplotypes in a total sample of 325 Russian individuals (Table 2). It was found that these haplotypes are clustered, according to human mtDNA classification (Macaulay et al. 1999; Richards et al. 2000; Salas et al. 2002; Yao et al. 2002), into haplogroups H, pre-V, HV*, R1, U1, U2, U3, U4, U5, U7, U8, K, J, T*, T1, I, N1a, N1b, X, W, C, D, G, M1, L1b, and L3b.
Table 3 summarizes the frequencies of the mtDNA haplogroups and subgroups found in the Russian populations studied. Russians present relatively high means of pairwise nucleotide differences between HVSI sequences. This parameter ranges from 4.74 in the Yaroslavl sample to 5.46 in the Vladimir sample and is in accordance with diversity estimates found in European and Middle Eastern populations [from 2.95 in Basques to 7.08 in populations of the Middle East according to Comas et al. (1997)].
The majority (97.8%) of mtDNA haplotypes found in Russians fall into major West Eurasian haplogroups. The frequency of East Asian--specific haplotypes in Russians is very low (1.5% on average). All of them belong to macrohaplogroup M (C, D, G). A single haplotype found in Russians belongs to specific haplogroup M1, which is distributed mainly in East Africa (Somalia and Ethiopia) and the Red Sea region (Richards et al. 2003). However, the low frequencies of haplogroup M1 were observed in southern Europe and the Caucasus region (Richards et al. 2000), from where this haplogroup may have been introduced into the Russian population. An interesting finding of the present study is the occurrence of African-specific mtDNA lineages (L1b and L3b) in the Russian gene pool. Their frequency is very low (0.6%), but this estimate should be lowered to 0.3% to take into account the previous data on mtDNA variability in Russian populations (Orekhov et al. 1999; Malyarchuk et al. 2002). These results indicate that the appearance of African mtDNAs among Russians was probably casual.
Table 4 reports the frequencies of the Y-chromosome haplogroups found in Russian samples. The pattern of Y-chromosome haplogroup distributions in Russians is typical for northeastern European populations (Rosser et al. 2000). The principal diagnostic haplogroups for Russians are haplogroups R1a and N3, which occur in the studied populations at a total frequency of 50-72%. The remaining Y-chromosome haplogroups represent mostly paraphyletic groups distributed with similar frequencies in European populations.
It should be noted that the frequency of paragroup P* [or haplogroup 1 according to Rosser et al.'s (2000) nomenclature] in Russians (2.3-19%) reflects the known pattern of distribution, with a high incidence of P* in western European populations (especially those of Celtic origin) and a lower frequency in the east (Zerjal et al. 2001). Haplogroups E* and J* were found only in the most southwestern Belgorod population at frequencies of 6.8% and 13.6%, respectively. These haplogroups probably are associated with the contributions of Neolithic farmers to the European gene pool, because both haplogroups E* and J* show the highest frequencies in Anatolia and southern Europe (Rosser et al. 2000; Semino et al. 2000; Underhill et al. 2001). Thus the lack of haplogroups E* and J* in most Russian populations studied suggests a minor genetic effect from southern European populations. A single haplotype revealed in Russians from the Belgorod population belongs to the East Asian--specific haplogroup C*, which is widespread in Central Asia (Zerjal et al. 2002). Therefore Y-chromosome variation data allow us to speculate that East Eurasian male gene flow in Russian populations was insignificant (0.3%).
In general, the data indicate that the Russian male gene pool preserves the main peculiarity of Y-chromosome diversity in central-eastern European populations. In the first instance, it is the high frequency of haplogroup R1a that is rare in western Europe but widely present in eastern and central Europe, being found at a frequency of 50-60% in Poles, Hungarians, Ukrainians, and Belorussians (Rosser et al. 2000; Semino et al. 2000; Wells et al. 2001; Behar et al. 2003). Analysis of the haplogroup R1a distribution in Russian populations reveals a relatively homogeneous pattern, with a population frequency of 50-60%, except for the most northwestern Pskov population and the most eastern Nizhniy Novgorod population from the Volga region, where this haplogroup is observed with a lower frequency of about 35%.
It is noteworthy that among the Russian populations studied, the Pskov population shows the highest frequency (35%) of haplogroup N3, which is characteristic of all Uralic- and Baltic-speaking populations north and east of the Baltic Sea (Rosser et al. 2000; Zerjal et al. 2001). This haplogroup is considered a signature of Uralic Finno-Ugric-speaking tribes migrating to northern Europe about 5,000 years ago (Passarino et al. 2002). The presence of haplogroup N3 in Russians suggests that some admixture between Slavonic and Finno-Ugric and/ or Baltic tribes occurred during the colonization of the East European plain by Slavs in the early Middle Ages and later. In addition, the distribution pattern of haplogroup N3 in other Slavonic-speaking populations suggests that proto-Slavs probably did not carry this lineage at a substantial frequency (Barac et al. 2003).
Differentiation of mtDNA and Y-Chromosome in Russian Populations. In order to study differentiation of Russian populations, we performed an analysis of molecular variance (AMOVA) separately on the level of mtDNA haplogroups and HVSI sequences. Analysis of between-population differentiation based on the frequencies of the mtDNA haplogroups revealed that 1.4% of the variation was due to differences among populations ([F.sub.ST] = 1.4%, p = 0.007). Significant pairwise [F.sub.ST] differences (p < 0.05) were found only between the Tula population and samples from Kaluga, Vladimir, and Pskov. In addition, the Vladimir and Pskov populations differed at a significant level. One should note, however, that the level of between-population differentiation was considerably lower when additional data on the mtDNA haplogroup frequencies in the Stavropol, Orel, and Saratov populations were taken into account ([F.sub.ST] = 0.6%, p = 0.048).
The AMOVA results of the mtDNA HVSI sequence data show that when the five Russian populations studied here were treated as a single group, 99.28% of the total variance was within populations ([F.sub.ST] = 0.72%, p = 0.009). Similar results were obtained ([F.sub.ST] = 0.42%, p = 0.005) when six additional Russian samples from the Stavropol, Orel, Saratov, Kostroma, Kursk, and Rostov regions were taken into account. Therefore the AMOVA results indicate the low but statistically significant level of mtDNA differentiation in Russian populations.
To further investigate genetic relationships between Russian populations, we used additional published data on the mtDNA HVSI variability in eastern and central European populations. These were the North Caucasus population of Adygei, Finno-Ugric-speaking Mari and Turkic-speaking Tatars from the Middle Volga region, North European Finno-Ugric populations of the Finns, Estonians, and Karelians, and the Baltic-speaking population of Lithuanians. Austrians and Poles were chosen because of their geographic location relative to the putative Central European homeland of the Slavs in the early Middle Ages. The AMOVA results showed that there is a low but statistically significant level of differentiation between the European populations analyzed ([F.sub.ST] = 0.41%, p = 0).
The multidimensional scaling analysis performed on the basis of pairwise [F.sub.ST] values (Table 5) revealed that Russian populations do not cluster together (Figure 3). It is noteworthy that all southern and western Russian populations (Stavropol, Saratov, Rostov, Kursk, Orel, and Kaluga) as well as Poles, Austrians, Lithuanians, and Estonians are separated from other, mostly northern and eastern Russian populations (such as Tula, Vladimir, Yaroslavl, Kostroma, and Pskov), which are clustered together with Finns, Karelians, Mari, Tatars, and Adygei. In the latter group of populations, Kostroma and Pskov Russians cluster more with Finno-Ugric-speaking Karelians and Mari than they do with the other Russian populations. The Tula, Vladimir, and Yaroslavl populations are close together but separated from the remaining Russian populations. In general, the results of the multidimensional scaling analysis of the mtDNA data indicate that, despite the low values of between-population differentiation, Russian populations appear to be heterogeneous.
The AMOVA results of Y-chromosome binary data in eight Russian populations (i.e., based on the haplogroup frequencies) indicate that the highest fraction of variability was due to within-population differences (97.53%), but a substantial fraction was due to differences among populations (2.47%, p = 0.0057). However, significant pairwise [F.sub.ST] differences (p < 0.05) were found only between the Pskov population and the rest of the Russian populations, with the exception of the Nizhniy Novgorod and Orel populations. In addition, the Nizniy Novgorod sample differs significantly from the Vladimir, Belgorod, and Yaroslavl populations. Similar results ([F.sub.ST] = 3.09%, p = 0) were obtained when two additional Russian samples from the north of European Russia (Northern Russians and Pomors) were included in analysis. The Russian Pomors sample was found to be distinct from all the Russian populations except for Pskov, Northern Russians, and Orel. Meanwhile, the Northern Russian sample differs significantly only from the Nizhniy Novgorod and Yaroslavl samples.
[FIGURE 3 OMITTED]
To provide the broader context for the study of the subdivision of Russian populations based on the Y-chromosome SNP [F.sub.ST] distances, we considered additional European populations in the analysis: Slavonic-speaking Poles, Ukrainians, and Belorussians; Baltic-speaking Latvians and Lithuanians; Finno-Ugric-speaking Finns, Estonians, and Mari; and Turkic-speaking Chuvash [all the data are from Rosser et al.'s (2000) study]. The AMOVA results show a significant level of differentiation between the European populations analyzed ([F.sub.ST] = 7.04%, p = 0). The multidimensional scaling analysis based on Y-chromosome haplogroup [F.sub.ST] values (Table 6) revealed that the strongest division was between Pskov and Pomor Russians together with all Finno-Ugric and Baltic-speaking populations, on the one hand, and the remaining Russian populations clustered together with other Slavonic-speaking groups, on the other hand (Figure 4). As for Baltic-speaking populations, it is well known that the Latvians and Lithuanians show genetic similarity with Finno-Ugric populations based on the increased frequency of haplogroup N3 (Rosser et al. 2000; Zerjal et al. 2001). Similarly, clustering of northern Pskov and Pomor Russian populations with Finno-Ugric and Baltic ones may be explained by increased frequency of haplogroup N3 found in these populations.
[FIGURE 4 OMITTED]
In general, Y-chromosome and mtDNA variation data indicate that the differentiation of the geographically distinct Russian populations appears to be more prominent in the male than in the female lineages ([F.sub.ST] values are 3.09% and 0.42%, respectively). Such a pattern has been found before in many human populations and is usually explained by patrilocal residence (Seielstad et al. 1998; Oota et al. 2001; Kayser et al. 2003). Thus the discordance between mtDNA and Y-chromosome variation in Russian populations may be a consequence of cultural features, such as patrilocality and the patrilineal social system that Russians practiced in the past, especially before the peasant reforms of the first half of the 20th century (Alekseev and Pershits 1999). In modern times, however, these features are preserved as a community tradition only in some rural localities of Russians in the north of eastern Europe (Alekseev and Pershits 1999). However, we cannot exclude other reasons (e.g., sex-specific migration or genetic drift in isolated populations) to explain why both genetic loci, Y chromosomes and mtDNA, can provide different pictures. Therefore the problem of contrasting patterns for Y-chromosome and mtDNA differentiation in Russian populations needs to be studied in more detail.
Discussion
Y-chromosome variation data obtained in the present study demonstrate that the Russian gene pool appears to be close to central-eastern European populations. The data show a high frequency of haplogroups R1a and N3. The high incidence of R1a haplotypes in Eastern Slavs and in Baltic and some Finno-Ugric populations of eastern Europe is a well-established fact (Rosser et al. 2000; Wells et al. 2001; Zerjal et al. 2001), but little is known about the spread of haplogroup N3 in different Russian populations. Our data indicate that this haplogroup is unevenly distributed in Russian populations, with frequencies ranging from 4.8% to 35%. It is noteworthy that in most of the populations studied (with the exception of the Pskov sample) the frequency of haplogroup N3 does not exceed 20%. An important feature distinguishing Russians from their eastern European neighbors is almost complete absence of paragroup N*, which is ancestral to haplogroup N3. In Europe the highest frequencies of paragroup N* (up to 42.5%) were detected in Volga Finno-Ugric populations: Udmurts and Mari (Rosser et al. 2000; Khusnutdinova et al. 2002). Therefore Y-chromosome data suggest that the Russian male gene pool is characterized by its own structural features, although genetic influences on Russians from Uralic- or Baltic-speaking populations are also highlighted. The latter is clearly defined in the most northern Russian populations, such as Pskov Russians and Pomors. A possible explanation for this phenomenon includes replacement of an earlier Uralic or Baltic language in populations of the north of eastern Europe by the present Russian language, occurring with a low input of Slavonic mtDNA and Y-chromosome haplotypes. The high frequency of haplogroup N3 in Pskov and Pomor Russians (and probably in other northern Russian populations) is striking and should be investigated further using Y-chromosome STR analysis and a comparison with published N3-associated Y-chromosome STR haplotype data in Finno-Ugric and Baltic populations. These data should be studied with respect to the hypothesis that the genetic history of N3 Y chromosomes in Baltic-speaking populations is distinct from that of the Uralic speakers and "that there were two distinct early migrations of haplogroup 16 into Europe" (Zerjal et al. 2001). To test this and other hypotheses concerning the recent historical interactions between eastern European populations, more populations need to be studied for more genetic markers.
The present study demonstrates that Russian populations contain relatively high levels of mtDNA sequence diversity (in the means of pairwise nucleotide differences), but the level of [F.sub.ST]-based between-population differentiation is low. However, the results of the multidimensional scaling analysis performed on the basis of pairwise [F.sub.ST] values for mtDNA HVSI sequences suggest that the Russian populations can be differentiated into subregions. The multidimensional scaling analysis revealed that Russian populations analyzed in combination with other central and eastern European populations do not cluster together and that populations from the southern and western parts of Russia (such as Stavropol, Rostov, Kursk, Orel, Kaluga, and Saratov) are separated from eastern and northern populations (Vladimir, Tula, Yaroslavl, Kostroma, and Pskov). Interestingly, southwestern Russian populations demonstrate genetic similarities with a set of central and northern European populations of different linguistic affiliation [Slavonic, Baltic, Germanic, and even Finno-Ugric (Estonians)], whereas northeastern Russian populations cluster together with eastern and northern European populations speaking not only Finno-Ugric languages but also Turkic (Tatars) and North Caucasus (Adygei) ones.
The multidimensional scaling analysis performed on the basis of pairwise [F.sub.ST] values for Y-chromosome haplogroup data shows a somewhat different picture (for mtDNA sequences) of population differentiation in Russia. The most important subdivision was found only between northern populations of Pskov and Pomor Russians and the rest of the Russian populations studied. Historically, this observed discrepancy in the depth of penetration of mtDNA and Y-chromosome lineages characteristic for the most southwestern Russian populations into the east and north of eastern Europe may indicate [in accordance with anthropological data (Rychkov and Balanovska 1988)] that only the most western Russian populations appear to be descendants of the Slavs, whereas northern and eastern Russian populations seem to be the result of an admixture between Slavonic tribes and pre-Slavonic populations of eastern Europe. In addition, the data allow one to assume the possibility that Slavonic male lineages penetrated the original eastern European populations further than mtDNA lineages. One should note, however, that this scenario should be tested with additional mtDNA and Y-chromosome data in multiple Russian-, Finno-Ugric-, and Turkic-speaking populations of eastern Europe, which are poorly characterized genetically.
In addition, to estimate the degree of population replacement in eastern Europe associated with the Slavonic colonization starting in the early Middle Ages [6th-7th centuries A.D. according to archeological data (Sedov 1995)], it would be important to perform analyses of multiple samples from small urban areas, because such an approach seems to be informative in genetic history studies (Capelli et al. 2003).
Table 1. RFLPs Analyzed to Identify Major Eurasian mtDNA Haplogroups Haplogroup Characteristic Restriction Site(s) West Eurasian HV - 14766 MseI H - 14766 MseI, - 7025 AluI pre*V1 - 14766 MseI, - 15904 MseI, + 4577 NlaIII pre*V2 - 14766 MseI, + 15904 MseI, + 4577 NlaIII V - 14766 MseI, + 15904 MseI, - 4577 NlaIII U + 12308 HinfI K + 10394 DdeI, + 12308 HinfI, - 9052 HaeII U2 + 12308 HinfI, + 15907 RsaI U4 + 12308 HinfI, + 4643 RsaI U8 + 12308 HinfI, - 7055 AluI J + 10394 DdeI, - 13704 BstOI T + 13366 BamHI, + 15606 AluI TI + 13366 BamHI, + 15606 AluI, - 12629 AvaII NI - 12498 NlaIII I - 4529 HaeII, + 8249 AvaII, + 10032 AluI, + 10394 DdeI, - 12498 NlaIII W + 8249 AvaII, - 8994 HaeIII X - 1715 DdeI, + 14465 AccI East Eurasian M + 10394 DdeI, + 10397 AluI C + 10394 DdeI, + 10397 AluI, - 13259 HincII/ + 13262 AluI D + 10394 DdeI, + 10397 AluI, - 5176 AluI E + 10394 DdeI, + 10397 AluI, - 7598 HhaI G + 10394 DdeI, + 10397 AluI, + 4830 HaeII/ + 4831 HhaI A + 663 HaeIII B 9-bp deletion F1 - 12406 HpaI/HincII RFLP markers are given relative to the revised version of the Cambridge Reference Sequence (Andrews et al. 1999) belonging to haplogroup H. Site gains and losses are indicated by a plus sign (+) or a minus sign (-). Table 2. mtDNA Haplotypes in Russians Obtained by Combining the Data on the HVSI Sequence and Haplogroup-Specific RFLPs HVSI Variation (-16000 Relative to CRS Sequence Haplo- Tula Kaluga Vladimir Numbering) group (n = 73) (n = 71) (n = 72) Cambridge Reference Sequence (CRS) H 11 6 7 70 93 311 H 1 93 311 H 2 93 H 126 H 129 H 1 147 H 1 162 H 51 162 H 1 51 162 213 256 H 1 162 311 H 2 168 H 170 H 1 176 H 1 169 176 H 180 H 1 183delA H 189 H 209 H 2 221 H 1 1 222 H 234 H 1 86 239 H 256 352 H 1 261 H 2 261 311 H 2 261 335 H 1 274 H 274 317 H 1 241 274 287 H 278 293 311 H 93 278 293 311 H 1 134 278 293 311 H 1 172 278 293 311 H 1 224 278 293 311 H 1 92 140 265 293 311 H 4 3 286 H 1 235 291 H 249 291 H 1 293 H 1 296 H 300 H 1 304 H 3 3 192 304 H 1 209 304 H 240 304 H 1 258AT 304 H 1 304 352 H 1 304 390 H 1 304 400 H 1 1 294CA 304 H 294 304 H 1 271 294 304 H 311 H 1 311 319 H 320 H 1 354 H 1 5 193 354 H 1 189 356 H 1 188 189 356 H 75 111 183AC 189 209 356 H 1 183AC 189 193.1C 294 356 H 1 80 189 356 H 1 80 129 189 193.1C 356...