Reconstructing the Evolutionary History of China: A Caveat About Inferences Drawn from Ancient DNA
* Kunming Institute of Zoology, Chinese Academy of Sciences, Kunming, Yunnan, China/, http://www.100md.com
Graduate School of the Chinese Academy of Sciences, Beijing, China/, http://www.100md.com
Taishan Medical College, Taian, Shandong, China/, http://www.100md.com
Fachbereich Mathematik, Universität Hamburg, Hamburg, Germany/, http://www.100md.com
|| Yunnan University, Kunming, China/, http://www.100md.com
Abstract/, http://www.100md.com
The decipherment of the meager information provided by short fragments of ancient mitochondrial DNA (mtDNA) is notoriously difficult but is regarded as a most promising way toward reconstructing the past from the genetic perspective. By haplogroup-specific hypervariable segment (HVS) motif search and matching or near-matching with available modern data sets, most of the ancient mtDNAs can be tentatively assigned to haplogroups, which are often subcontinent specific. Further typing for mtDNA haplogroup-diagnostic coding region polymorphisms, however, is indispensable for establishing the geographic/genetic affinities of ancient samples with less ambiguity. In the present study, we sequenced a fragment (~ 982 bp) of the mtDNA control region in 76 Han individuals from Taian, Shandong, China, and we combined these data with previously reported samples from Zibo and Qingdao, Shandong. The reanalysis of two previously published ancient mtDNA population data sets from Linzi (same province) then indicates that the ancient populations had features in common with the modern populations from south China rather than any specific affinity to the European mtDNA pool. Our results highlight that ancient mtDNA data obtained under different sampling schemes and subject to potential contamination can easily create the impression of drastic spatiotemporal changes in the genetic structure of a regional population during the past few thousand years if inappropriate methods of data analysis are employed.
Key Words: mtDNA • ancient DNA • Han Chinese6j[[(s, 百拇医药
Introduction6j[[(s, 百拇医药
The prehistory and recent migration history of the people of China are very complex ). Information provided by ancient mitochondrial DNA (mtDNA), especially from the post-Neolithic period, has been regarded as one of the most powerful methods of understanding and reconstructing the past from the genetic perspective . In particular, investigated "temporal changes in genetic structure of human populations during the past 2,500 years in China using mtDNA sequences." In their study of a 2,500-year-old human population from Linzi, Shandong Province, China, together with a 2,000-year-old population and a modern population from the same location, they employed a network method to display the phylogenetic relationship among mtDNA haplotypes and used a distance method to construct a population tree. Among the six "radiation groups," designated I to VI , in their haplotype network, the frequencies of each radiation group varied substantially across the 2,500-year-old, the 2,000-year-old, and the modern Linzi populations. Their population tree clustered the 2,500-year-old population with the present-day Europeans, whereas the 2,000-year-old population occupied an intermediate position between modern East Asians (which contained the present-day Linzi population) and Europeans. Based on these results, they suggested that drastic spatiotemporal changes had occurred in the genetic structure of Chinese people during the past 2,500 years. Leaving potential problems of authenticity of the ancient DNAs aside, their data deserve further analyses for two reasons. First, four of their six radiation groups (I to IV) are paraphyletic . Second, the reported mtDNA control region hypervariable segment I (HVS-I) sequences were truncated to 185-bp segments for phylogenetic analysis of mtDNA haplotypes and to 172-bp segments for estimating genetic distances between populations, which would no longer permit the clear distinction between major European and East Asian mtDNA haplogroups in several cases.
A dissection of mtDNAs into major (and minor) haplogroups is essential for identifying spatial frequency patterns . Thus, there are two prerequisites necessary for an interpretation of relatively short fragments of ancient mtDNA from East Asia: (1) a reliable basal phylogeny of the major East Asian mtDNA lineages based on control region and coding region information ; and (2) a large database of modern East Asian mtDNA HVS-I (and HVS-II) sequences linked with partial coding region sequences, or at least single haplogroup-diagnostic sites screened by restriction fragment length polymorphism (RFLP) typing . With this information at hand, one can often infer the potential haplogroup status from a short segment of the control region (which is typical for ancient DNAs) through a haplogroup-specific HVS-I motif search and matching or near-matching with the mtDNA haplotypes in the database .{z'$*, http://www.100md.com
Until now, only two modern mtDNA samples (Zibo, with 185-bp fragments of HVS-I , and Qingdao, with HVS-I and HVS-II sequences and additional coding region information ) from Han populations of Shandong have been analyzed for mtDNA variation. In the present study, we sequenced the mtDNA control region in 76 Han individuals from Taian, Shandong, and combined these data with the reported Zibo and Qingdao samples in order to elucidate the (matrilineal) genetic structures of contemporary populations from Shandong. Then, by contrasting the frequency distributions of the mtDNA haplogroups tentatively identified in the ancient DNAs with the modern data from Shandong as well as from other provinces of China ( and references therein), we reassessed the conclusion of .
Materials and Methods/a7x[ry, 百拇医药
Samples and Data Sets/a7x[ry, 百拇医药
The ancient DNA population data reanalyzed here comprise 34 mtDNAs and 24 mtDNAs from Linzi, Shandong Province . Seventy-six modern Han samples from Taian, Shandong (with informed consent) were newly recruited in this study and, together with a series of modern mtDNAs from China (and references therein), taken for comparison./a7x[ry, 百拇医药
fig.ommitted/a7x[ry, 百拇医药
Table 1 mtDNA sequence variation in the 2,500-year-old, 2,000-year-old, and modern mtDNAs from Shandong Province, China./a7x[ry, 百拇医药
Sequence Analysis/a7x[ry, 百拇医药
Genomic DNA was extracted from whole blood by standard phenol/chloroform methods. The mtDNA control region sequences (relative to region 16000–407 in Cambridge reference sequence [CRS] ) were amplified with primer L15996 (5'-CTCCACCATTAGCACCCAAAGC-3') and H408 (5'-CTGTTAAAAGTGCATACCGCCA-3'). The products were purified with gel extraction kits (Watson BioMedicals Inc.) and were sequenced by using these two primers and three internal primers (L16209 [5'-CCCCATGCTTACAAGCAAGT-3'], L29 [5'- GGTCTATCACCCTATTAACCAC-3'], and H16498 [5'- CCTGAAGTAGGAACCAGATG-3']) on an ABI 377XL DNA Sequencer (Applied Biosystems). The length variation of the A and C stretches in 16180 to 16188 (triggered by the 16189 T to C substitution) were disregarded in the analyses. The polymorphisms in region 10171 to 10659 and at sites 4833 and 5178 relative to CRS in several Taian Han samples were typed by using the same primers and conditions as described in our previous study . The control region sequences have been submitted to GenBank under accession nos. AF519813 to AF519888.
Haplogroup Assignment3su^;-, http://www.100md.com
We adopted the mtDNA haplogroup classification tree and nomenclature system described in recent studies for mtDNA assignment. The haplogroup status of the 2,000-year-old and 2,500-year-old mtDNAs was tentatively inferred according to the strategy described in the Introduction and in recent studies . Some ancient mtDNAs that could not be classified unambiguously were left unassigned and were marked as "?" for haplogroup status. The frequency vectors of mtDNA basal haplogroups (M7, M8, M9, M10, M*/N*, G2, D, A, N9, B, R9, and R* ) in 16 regional Han samples across China ( and references therein; this study) and the two ancient samples were subjected to principal component (PC) analysis.3su^;-, http://www.100md.com
Results and Discussion3su^;-, http://www.100md.com
lists the nucleotide variation in the 2,500-year-old and 2,000-year-old samples along with the potential haplogroup status of each haplotype. A relatively high frequency of haplogroup F (9/24) is seen in the 2,000-year-old sample, whereas the 2,500-year-old sample harbors a very large amount of potential B types (15/34). It is conspicuous that in either data set quite a number of individuals were found to share the same haplotype in several instances, which could be due to close matrilineal relationships (thus biasing comparisons with modern mtDNA samples). In particular, six members of the 2,500-year-old sample are indistinguishable from the CRS within the short HVS-I fragment sequenced and may be close to the ancestral HVS-I type of haplogroup B. For this fragment, they match two mtDNAs (QD8141 and SD10313) of our Shandong samples. Interestingly, a central Honsh mtDNA sample (50 individuals) sequenced for 16040 to 16375 in HVS-I and 20 to 430 in HVS-II harbors 15 identical HVS-I sequences, differing from the CRS only at 16189. Truncated to the segment 16194 to 16375, both this Japanese sample and the 2,500-year-old Linzi sample would show the same sort of genetic affinity to European samples incurred by high frequency of the CRS (a typical Western European feature).
The mtDNA haplogroups that could be positively identified (albeit at different levels of certainty) in the two ancient populations are "Asian-specific" (see). Moreover, some of the haplotypes in the ancient DNAs match exactly those from modern samples. For example, types M7b2 (16223-16297-16298), M* (16223-16311), M* (16223), A (16223-16290-16319-16362), D5a (16223-16266-16362-150-263), and C (16223-16298-16327) in the ancient mtDNAs could be matched to haplotypes observed in the three modern samples from Shandong. The F1b type (16232A-16249-16304-16311-16344) in the 2,500-year-old population matches a haplotype observed in Han people from Yunnan Province, and the motif 16232A-16249-16304 is widespread, found in individuals from Xinjiang, Yunnan, Liaoning , Shaanxi , and Shandong (this study). The specific motif (16291-16304) of F2 types in the 2,000-year-old populations are also found in contemporary Han individuals from those provinces as well as individuals from some ethnic groups in Yunnan . It is therefore far from conclusive that "the 2,500-year-old population showed greater genetic similarity to present-day European populations than to present-day East Asian populations, and the 2,000-year-old population had features that were intermediate between the present-day European populations and the present-day East Asian populations" ( p. 1396).
gives a PC analysis view of the 16 regional Han samples across China and the two ancient samples based on the frequency profiles of the basal haplogroups , whereby the unassigned types marked as "?" for haplogroup status were disregarded (a tentative scoring as R* would only emphasize potential outlier status [data not shown]). The 2,500-year-old and especially the 2,000-year-old samples show closer affinity to the modern samples from the south than from the north, thus reflecting the trend in the haplogroup profiles discussed above: haplogroups F and B have higher frequencies in populations from the south (or of southern origin) than those from the north (or of northern origin) .3efd, http://www.100md.com
fig.ommitted3efd, http://www.100md.com
FIG. 1. Principal component (PC) map of 16 regional Han samples and the two ancient samples, with respect to the mtDNA basal haplogroup frequency profiles as described in . Ancient mtDNAs with haplogroup status marked as "?" in were excluded from the analyses. The abbreviations and data sources of the modern Han samples are as follows: YN = Yunnan; WH = Wuhan; QD = Qingdao; LN = Liaoning; XJ = Xinjiang; GD-ZJ = Zhanjiang, Guandong; HK = Hong Kong; TW-1 = Taiwan-1; TW-2 = Taiwan-2; SH = Shanghai; QH = Qinghai ( and references therein); GD-GZ = Guanzhou, Guangdong ; ZB = Zibo, Shandong ); XA = Xian, Shaanxi; CS = Changsha, Hunan ; and TA = Taian (this study)
Given the difficulties in retrieving authentic ancient DNAs from specimens and the fact that many haplogroups are defined by whole arrays of mutations in the control region as well as the relatively less mutable coding region, it would be economical and effective to sequence one or two short HVS-I (or HVS-II) stretches generally and then, depending on the potential haplogroup assignment(s) inferred from the resulting HVS-I fragments, to screen different small fragments in the coding region individually. Even with the meager information contained in very short HVS-I fragments, one would need only one or two extra sequencing rounds to confirm haplogroup status in most cases. For example, the potential M10 haplotype (16223-16311-73-263-315+C) in the recent Korean data comprising mtDNA control region and cytochrome b gene polymorphisms bears the two mutations 15071 and 15218. We therefore sequenced the two types, SD10364 and SD10334, of our Taian sample (having the HVS-I motif 16223-16311 combined with +5176 AluI) and confirmed that only the M10 type (SD10334, according to the mutation 10646) harbored these two cytochrome b mutations. Further typing of the five mtDNAs assigned to haplogroup M10 in showed that they all carry these mutations as well. Hence, the presence of mutations 15071 and 15218 constitutes additional markers for haplogroup M10. If this information was available at the time, then the ancient mtDNA, Liach09, showing the motif 16223-16311 could have been tested for the latter two mutations. In case these mutations turned out to be absent, a confirmation of the HVS-II motif 215-318-326 could point to another (yet unnamed) sub-haplogroup of M (cf. SD10324 and SD10364).
In short, both the 2,500-year-old and the 2,000-year-old Linzi populations had features in common with the modern populations from south China rather than any specific affinity to the European mtDNA pool. The following scenario could explain the changes in the matrilineal genetic structure of the Shandong populations during the past 3,000 years: before or during the Spring-Autumn and Warring States era (770 B.C. to 221 B.C.), migrations from south and central China could have heavily contributed to the gene pool of Linzi inhabitants, as Linzi was the capital of the feudal state of Qi, which was famous for its economy and development at that time . Later, the conquest by the Han coming from central and north China in historical time and the occasional invasion of northern nomadic tribes to this region might have brought major shifts in the composition of the resident population , which eventually led to the genetic pattern we observe today. To reconstruct the evolutionary history of China, it is thus important to take (pre)historic knowledge and archaeological assemblages into full consideration. Our results highlight that the interpretation of ancient DNA should begin with sufficient data and not draw conclusions on too limited a data set. Molecular anthropology would profit from embracing the research done in medical genetics and forensics: both are employing or are beginning to use coding region information ( and references therein). For unambiguous conclusions, it is indispensable to include coding region polymorphisms that are diagnostic for mtDNA haplogroup membership in addition to sequencing portions of the HVS-I when trying to establish the geographic/genetic affinities of ancient samples.
Acknowledgementsi, http://www.100md.com
We thank Zhu C.-L. and Gou S.-K. for technical assistance. This study was supported by grants from Natural Sciences Foundation of Yunnan Province and Natural Sciences Foundation of China (NSFC).i, http://www.100md.com
Literature Citedi, http://www.100md.com
Andrews, R. M., I. Kubacka, P. F. Chinnery, R. N. Lightowlers, D. M. Turnbull, and N. Howell. 1999. Reanalysis and revision of the Cambridge reference sequence for human mitochondrial DNA. Nat. Genet. 23:147.i, http://www.100md.com
Ge, J. X., S. D. Wu, and S. J. Chao. 1997. Zhongguo yimin shi (The migration history of China). Fujian People Press, Fuzhou [in Chinese].i, http://www.100md.com
Herrnstadt, C., J. L. Elson, E. Fahy, et al. (11 co-authors). 2002. Reduced-median-network analysis of complete mitochondrial DNA coding-region sequences for the major African, Asian, and European haplogroups. Am. J. Hum. Genet. 70:1152-1171.i, http://www.100md.com
Kivisild, T., H.-V. Tolk, J. Parik, Y. Wang, S. S. Papiha, H.-J. Bandelt, and R. Villems. 2002. The emerging limbs and twigs of the East Asian mtDNA tree. Mol. Biol. Evol. 19:1737-1751.
Koyama, H., M. Iwasa, Y. Maeno, et al. (11 co-authors). 2002. Mitochondrial sequence haplotype in the Japanese population. Forensic Sci. Int. 125:93-96.(*f, 百拇医药
Lee, S.-D., Y.-S. Lee, and J.-B. Lee. 2002. Polymorphism in the mitochondrial cytochrome B gene in Koreans: an additional marker for individual identification. Int. J. Legal. Med. 116:74-78.(*f, 百拇医药
Oota, H., T. Kitano, F. Jin, I. Yuasa, L. Wang, S. Ueda, N. Saitou, and M. Stoneking. 2002. Extreme mtDNA homogeneity of continental Asian populations. Am. J. Phys. Anthropol. 118:146-153.(*f, 百拇医药
Oota, H., N. Saitou, T. Matsushita, and S. Ueda. 1999. Molecular genetic analysis of remains of a 2,000-year-old human population in China—and its relevance for the origin of the modern Japanese population. Am. J. Hum. Genet. 64:250-258.(*f, 百拇医药
Torroni, A., H.-J. Bandelt, V. Macaulay, et al. (33 co-authors). 2001. A signal, from human mtDNA, of postglacial recolonization in Europe. Am. J. Hum. Genet. 69:844-852.(*f, 百拇医药
Torroni, A., M. Richards, V. Macaulay, P. Forster, R. Villems, S. Nørby, M. L. Savontaus, K. Huoponen, R. Scozzari, and H.-J. Bandelt. 2000. mtDNA haplogroups and frequency patterns in Europe. Am. J. Hum. Genet. 66:1173-1177.
Wang, L., H. Oota, N. Saitou, F. Jin, T. Matsushita, and S. Ueda. 2000. Genetic structure of a 2,500-year-old human population in China and its spatiotemporal changes. Mol. Biol. Evol. 17:1396-1400.2w2g8uq, 百拇医药
Yao, Y.-G., Q.-P. Kong, H.-J. Bandelt, T. Kivisild, and Y.-P. Zhang. 2002a. Phylogeographic differentiation of mitochondrial DNA in Han Chinese. Am. J. Hum. Genet. 70:635-651.2w2g8uq, 百拇医药
Yao, Y.-G., L. Nie, H. Harpending, Y.-X. Fu, Z.-G. Yuan, and Y.-P. Zhang. 2002b. Genetic relationship of Chinese ethnic populations revealed by mtDNA sequence diversity. Am. J. Phys. Anthropol. 118:63-76.2w2g8uq, 百拇医药
Yao, Y.-G., and Y.-P. Zhang. 2002. Phylogeographic analysis of mtDNA variation in four ethnic populations from Yunnan Province: new data and a reappraisal. J. Hum. Genet. 47:311-318.2w2g8uq, 百拇医药
Accepted for publication October 9, 2002.(Yong-Gang Yao Qing-Peng Kong Xiao-Yong Man, Hans-Jürgen Bandelt and Ya-Ping Zhang)
Graduate School of the Chinese Academy of Sciences, Beijing, China/, http://www.100md.com
Taishan Medical College, Taian, Shandong, China/, http://www.100md.com
Fachbereich Mathematik, Universität Hamburg, Hamburg, Germany/, http://www.100md.com
|| Yunnan University, Kunming, China/, http://www.100md.com
Abstract/, http://www.100md.com
The decipherment of the meager information provided by short fragments of ancient mitochondrial DNA (mtDNA) is notoriously difficult but is regarded as a most promising way toward reconstructing the past from the genetic perspective. By haplogroup-specific hypervariable segment (HVS) motif search and matching or near-matching with available modern data sets, most of the ancient mtDNAs can be tentatively assigned to haplogroups, which are often subcontinent specific. Further typing for mtDNA haplogroup-diagnostic coding region polymorphisms, however, is indispensable for establishing the geographic/genetic affinities of ancient samples with less ambiguity. In the present study, we sequenced a fragment (~ 982 bp) of the mtDNA control region in 76 Han individuals from Taian, Shandong, China, and we combined these data with previously reported samples from Zibo and Qingdao, Shandong. The reanalysis of two previously published ancient mtDNA population data sets from Linzi (same province) then indicates that the ancient populations had features in common with the modern populations from south China rather than any specific affinity to the European mtDNA pool. Our results highlight that ancient mtDNA data obtained under different sampling schemes and subject to potential contamination can easily create the impression of drastic spatiotemporal changes in the genetic structure of a regional population during the past few thousand years if inappropriate methods of data analysis are employed.
Key Words: mtDNA • ancient DNA • Han Chinese6j[[(s, 百拇医药
Introduction6j[[(s, 百拇医药
The prehistory and recent migration history of the people of China are very complex ). Information provided by ancient mitochondrial DNA (mtDNA), especially from the post-Neolithic period, has been regarded as one of the most powerful methods of understanding and reconstructing the past from the genetic perspective . In particular, investigated "temporal changes in genetic structure of human populations during the past 2,500 years in China using mtDNA sequences." In their study of a 2,500-year-old human population from Linzi, Shandong Province, China, together with a 2,000-year-old population and a modern population from the same location, they employed a network method to display the phylogenetic relationship among mtDNA haplotypes and used a distance method to construct a population tree. Among the six "radiation groups," designated I to VI , in their haplotype network, the frequencies of each radiation group varied substantially across the 2,500-year-old, the 2,000-year-old, and the modern Linzi populations. Their population tree clustered the 2,500-year-old population with the present-day Europeans, whereas the 2,000-year-old population occupied an intermediate position between modern East Asians (which contained the present-day Linzi population) and Europeans. Based on these results, they suggested that drastic spatiotemporal changes had occurred in the genetic structure of Chinese people during the past 2,500 years. Leaving potential problems of authenticity of the ancient DNAs aside, their data deserve further analyses for two reasons. First, four of their six radiation groups (I to IV) are paraphyletic . Second, the reported mtDNA control region hypervariable segment I (HVS-I) sequences were truncated to 185-bp segments for phylogenetic analysis of mtDNA haplotypes and to 172-bp segments for estimating genetic distances between populations, which would no longer permit the clear distinction between major European and East Asian mtDNA haplogroups in several cases.
A dissection of mtDNAs into major (and minor) haplogroups is essential for identifying spatial frequency patterns . Thus, there are two prerequisites necessary for an interpretation of relatively short fragments of ancient mtDNA from East Asia: (1) a reliable basal phylogeny of the major East Asian mtDNA lineages based on control region and coding region information ; and (2) a large database of modern East Asian mtDNA HVS-I (and HVS-II) sequences linked with partial coding region sequences, or at least single haplogroup-diagnostic sites screened by restriction fragment length polymorphism (RFLP) typing . With this information at hand, one can often infer the potential haplogroup status from a short segment of the control region (which is typical for ancient DNAs) through a haplogroup-specific HVS-I motif search and matching or near-matching with the mtDNA haplotypes in the database .{z'$*, http://www.100md.com
Until now, only two modern mtDNA samples (Zibo, with 185-bp fragments of HVS-I , and Qingdao, with HVS-I and HVS-II sequences and additional coding region information ) from Han populations of Shandong have been analyzed for mtDNA variation. In the present study, we sequenced the mtDNA control region in 76 Han individuals from Taian, Shandong, and combined these data with the reported Zibo and Qingdao samples in order to elucidate the (matrilineal) genetic structures of contemporary populations from Shandong. Then, by contrasting the frequency distributions of the mtDNA haplogroups tentatively identified in the ancient DNAs with the modern data from Shandong as well as from other provinces of China ( and references therein), we reassessed the conclusion of .
Materials and Methods/a7x[ry, 百拇医药
Samples and Data Sets/a7x[ry, 百拇医药
The ancient DNA population data reanalyzed here comprise 34 mtDNAs and 24 mtDNAs from Linzi, Shandong Province . Seventy-six modern Han samples from Taian, Shandong (with informed consent) were newly recruited in this study and, together with a series of modern mtDNAs from China (and references therein), taken for comparison./a7x[ry, 百拇医药
fig.ommitted/a7x[ry, 百拇医药
Table 1 mtDNA sequence variation in the 2,500-year-old, 2,000-year-old, and modern mtDNAs from Shandong Province, China./a7x[ry, 百拇医药
Sequence Analysis/a7x[ry, 百拇医药
Genomic DNA was extracted from whole blood by standard phenol/chloroform methods. The mtDNA control region sequences (relative to region 16000–407 in Cambridge reference sequence [CRS] ) were amplified with primer L15996 (5'-CTCCACCATTAGCACCCAAAGC-3') and H408 (5'-CTGTTAAAAGTGCATACCGCCA-3'). The products were purified with gel extraction kits (Watson BioMedicals Inc.) and were sequenced by using these two primers and three internal primers (L16209 [5'-CCCCATGCTTACAAGCAAGT-3'], L29 [5'- GGTCTATCACCCTATTAACCAC-3'], and H16498 [5'- CCTGAAGTAGGAACCAGATG-3']) on an ABI 377XL DNA Sequencer (Applied Biosystems). The length variation of the A and C stretches in 16180 to 16188 (triggered by the 16189 T to C substitution) were disregarded in the analyses. The polymorphisms in region 10171 to 10659 and at sites 4833 and 5178 relative to CRS in several Taian Han samples were typed by using the same primers and conditions as described in our previous study . The control region sequences have been submitted to GenBank under accession nos. AF519813 to AF519888.
Haplogroup Assignment3su^;-, http://www.100md.com
We adopted the mtDNA haplogroup classification tree and nomenclature system described in recent studies for mtDNA assignment. The haplogroup status of the 2,000-year-old and 2,500-year-old mtDNAs was tentatively inferred according to the strategy described in the Introduction and in recent studies . Some ancient mtDNAs that could not be classified unambiguously were left unassigned and were marked as "?" for haplogroup status. The frequency vectors of mtDNA basal haplogroups (M7, M8, M9, M10, M*/N*, G2, D, A, N9, B, R9, and R* ) in 16 regional Han samples across China ( and references therein; this study) and the two ancient samples were subjected to principal component (PC) analysis.3su^;-, http://www.100md.com
Results and Discussion3su^;-, http://www.100md.com
lists the nucleotide variation in the 2,500-year-old and 2,000-year-old samples along with the potential haplogroup status of each haplotype. A relatively high frequency of haplogroup F (9/24) is seen in the 2,000-year-old sample, whereas the 2,500-year-old sample harbors a very large amount of potential B types (15/34). It is conspicuous that in either data set quite a number of individuals were found to share the same haplotype in several instances, which could be due to close matrilineal relationships (thus biasing comparisons with modern mtDNA samples). In particular, six members of the 2,500-year-old sample are indistinguishable from the CRS within the short HVS-I fragment sequenced and may be close to the ancestral HVS-I type of haplogroup B. For this fragment, they match two mtDNAs (QD8141 and SD10313) of our Shandong samples. Interestingly, a central Honsh mtDNA sample (50 individuals) sequenced for 16040 to 16375 in HVS-I and 20 to 430 in HVS-II harbors 15 identical HVS-I sequences, differing from the CRS only at 16189. Truncated to the segment 16194 to 16375, both this Japanese sample and the 2,500-year-old Linzi sample would show the same sort of genetic affinity to European samples incurred by high frequency of the CRS (a typical Western European feature).
The mtDNA haplogroups that could be positively identified (albeit at different levels of certainty) in the two ancient populations are "Asian-specific" (see). Moreover, some of the haplotypes in the ancient DNAs match exactly those from modern samples. For example, types M7b2 (16223-16297-16298), M* (16223-16311), M* (16223), A (16223-16290-16319-16362), D5a (16223-16266-16362-150-263), and C (16223-16298-16327) in the ancient mtDNAs could be matched to haplotypes observed in the three modern samples from Shandong. The F1b type (16232A-16249-16304-16311-16344) in the 2,500-year-old population matches a haplotype observed in Han people from Yunnan Province, and the motif 16232A-16249-16304 is widespread, found in individuals from Xinjiang, Yunnan, Liaoning , Shaanxi , and Shandong (this study). The specific motif (16291-16304) of F2 types in the 2,000-year-old populations are also found in contemporary Han individuals from those provinces as well as individuals from some ethnic groups in Yunnan . It is therefore far from conclusive that "the 2,500-year-old population showed greater genetic similarity to present-day European populations than to present-day East Asian populations, and the 2,000-year-old population had features that were intermediate between the present-day European populations and the present-day East Asian populations" ( p. 1396).
gives a PC analysis view of the 16 regional Han samples across China and the two ancient samples based on the frequency profiles of the basal haplogroups , whereby the unassigned types marked as "?" for haplogroup status were disregarded (a tentative scoring as R* would only emphasize potential outlier status [data not shown]). The 2,500-year-old and especially the 2,000-year-old samples show closer affinity to the modern samples from the south than from the north, thus reflecting the trend in the haplogroup profiles discussed above: haplogroups F and B have higher frequencies in populations from the south (or of southern origin) than those from the north (or of northern origin) .3efd, http://www.100md.com
fig.ommitted3efd, http://www.100md.com
FIG. 1. Principal component (PC) map of 16 regional Han samples and the two ancient samples, with respect to the mtDNA basal haplogroup frequency profiles as described in . Ancient mtDNAs with haplogroup status marked as "?" in were excluded from the analyses. The abbreviations and data sources of the modern Han samples are as follows: YN = Yunnan; WH = Wuhan; QD = Qingdao; LN = Liaoning; XJ = Xinjiang; GD-ZJ = Zhanjiang, Guandong; HK = Hong Kong; TW-1 = Taiwan-1; TW-2 = Taiwan-2; SH = Shanghai; QH = Qinghai ( and references therein); GD-GZ = Guanzhou, Guangdong ; ZB = Zibo, Shandong ); XA = Xian, Shaanxi; CS = Changsha, Hunan ; and TA = Taian (this study)
Given the difficulties in retrieving authentic ancient DNAs from specimens and the fact that many haplogroups are defined by whole arrays of mutations in the control region as well as the relatively less mutable coding region, it would be economical and effective to sequence one or two short HVS-I (or HVS-II) stretches generally and then, depending on the potential haplogroup assignment(s) inferred from the resulting HVS-I fragments, to screen different small fragments in the coding region individually. Even with the meager information contained in very short HVS-I fragments, one would need only one or two extra sequencing rounds to confirm haplogroup status in most cases. For example, the potential M10 haplotype (16223-16311-73-263-315+C) in the recent Korean data comprising mtDNA control region and cytochrome b gene polymorphisms bears the two mutations 15071 and 15218. We therefore sequenced the two types, SD10364 and SD10334, of our Taian sample (having the HVS-I motif 16223-16311 combined with +5176 AluI) and confirmed that only the M10 type (SD10334, according to the mutation 10646) harbored these two cytochrome b mutations. Further typing of the five mtDNAs assigned to haplogroup M10 in showed that they all carry these mutations as well. Hence, the presence of mutations 15071 and 15218 constitutes additional markers for haplogroup M10. If this information was available at the time, then the ancient mtDNA, Liach09, showing the motif 16223-16311 could have been tested for the latter two mutations. In case these mutations turned out to be absent, a confirmation of the HVS-II motif 215-318-326 could point to another (yet unnamed) sub-haplogroup of M (cf. SD10324 and SD10364).
In short, both the 2,500-year-old and the 2,000-year-old Linzi populations had features in common with the modern populations from south China rather than any specific affinity to the European mtDNA pool. The following scenario could explain the changes in the matrilineal genetic structure of the Shandong populations during the past 3,000 years: before or during the Spring-Autumn and Warring States era (770 B.C. to 221 B.C.), migrations from south and central China could have heavily contributed to the gene pool of Linzi inhabitants, as Linzi was the capital of the feudal state of Qi, which was famous for its economy and development at that time . Later, the conquest by the Han coming from central and north China in historical time and the occasional invasion of northern nomadic tribes to this region might have brought major shifts in the composition of the resident population , which eventually led to the genetic pattern we observe today. To reconstruct the evolutionary history of China, it is thus important to take (pre)historic knowledge and archaeological assemblages into full consideration. Our results highlight that the interpretation of ancient DNA should begin with sufficient data and not draw conclusions on too limited a data set. Molecular anthropology would profit from embracing the research done in medical genetics and forensics: both are employing or are beginning to use coding region information ( and references therein). For unambiguous conclusions, it is indispensable to include coding region polymorphisms that are diagnostic for mtDNA haplogroup membership in addition to sequencing portions of the HVS-I when trying to establish the geographic/genetic affinities of ancient samples.
Acknowledgementsi, http://www.100md.com
We thank Zhu C.-L. and Gou S.-K. for technical assistance. This study was supported by grants from Natural Sciences Foundation of Yunnan Province and Natural Sciences Foundation of China (NSFC).i, http://www.100md.com
Literature Citedi, http://www.100md.com
Andrews, R. M., I. Kubacka, P. F. Chinnery, R. N. Lightowlers, D. M. Turnbull, and N. Howell. 1999. Reanalysis and revision of the Cambridge reference sequence for human mitochondrial DNA. Nat. Genet. 23:147.i, http://www.100md.com
Ge, J. X., S. D. Wu, and S. J. Chao. 1997. Zhongguo yimin shi (The migration history of China). Fujian People Press, Fuzhou [in Chinese].i, http://www.100md.com
Herrnstadt, C., J. L. Elson, E. Fahy, et al. (11 co-authors). 2002. Reduced-median-network analysis of complete mitochondrial DNA coding-region sequences for the major African, Asian, and European haplogroups. Am. J. Hum. Genet. 70:1152-1171.i, http://www.100md.com
Kivisild, T., H.-V. Tolk, J. Parik, Y. Wang, S. S. Papiha, H.-J. Bandelt, and R. Villems. 2002. The emerging limbs and twigs of the East Asian mtDNA tree. Mol. Biol. Evol. 19:1737-1751.
Koyama, H., M. Iwasa, Y. Maeno, et al. (11 co-authors). 2002. Mitochondrial sequence haplotype in the Japanese population. Forensic Sci. Int. 125:93-96.(*f, 百拇医药
Lee, S.-D., Y.-S. Lee, and J.-B. Lee. 2002. Polymorphism in the mitochondrial cytochrome B gene in Koreans: an additional marker for individual identification. Int. J. Legal. Med. 116:74-78.(*f, 百拇医药
Oota, H., T. Kitano, F. Jin, I. Yuasa, L. Wang, S. Ueda, N. Saitou, and M. Stoneking. 2002. Extreme mtDNA homogeneity of continental Asian populations. Am. J. Phys. Anthropol. 118:146-153.(*f, 百拇医药
Oota, H., N. Saitou, T. Matsushita, and S. Ueda. 1999. Molecular genetic analysis of remains of a 2,000-year-old human population in China—and its relevance for the origin of the modern Japanese population. Am. J. Hum. Genet. 64:250-258.(*f, 百拇医药
Torroni, A., H.-J. Bandelt, V. Macaulay, et al. (33 co-authors). 2001. A signal, from human mtDNA, of postglacial recolonization in Europe. Am. J. Hum. Genet. 69:844-852.(*f, 百拇医药
Torroni, A., M. Richards, V. Macaulay, P. Forster, R. Villems, S. Nørby, M. L. Savontaus, K. Huoponen, R. Scozzari, and H.-J. Bandelt. 2000. mtDNA haplogroups and frequency patterns in Europe. Am. J. Hum. Genet. 66:1173-1177.
Wang, L., H. Oota, N. Saitou, F. Jin, T. Matsushita, and S. Ueda. 2000. Genetic structure of a 2,500-year-old human population in China and its spatiotemporal changes. Mol. Biol. Evol. 17:1396-1400.2w2g8uq, 百拇医药
Yao, Y.-G., Q.-P. Kong, H.-J. Bandelt, T. Kivisild, and Y.-P. Zhang. 2002a. Phylogeographic differentiation of mitochondrial DNA in Han Chinese. Am. J. Hum. Genet. 70:635-651.2w2g8uq, 百拇医药
Yao, Y.-G., L. Nie, H. Harpending, Y.-X. Fu, Z.-G. Yuan, and Y.-P. Zhang. 2002b. Genetic relationship of Chinese ethnic populations revealed by mtDNA sequence diversity. Am. J. Phys. Anthropol. 118:63-76.2w2g8uq, 百拇医药
Yao, Y.-G., and Y.-P. Zhang. 2002. Phylogeographic analysis of mtDNA variation in four ethnic populations from Yunnan Province: new data and a reappraisal. J. Hum. Genet. 47:311-318.2w2g8uq, 百拇医药
Accepted for publication October 9, 2002.(Yong-Gang Yao Qing-Peng Kong Xiao-Yong Man, Hans-Jürgen Bandelt and Ya-Ping Zhang)