AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 93:189-199 (1994)

Mitochondrial DNA Analysis in Tibet: Implications for the Origin of the Tibetan Population and Its Adaptation to High Altitude

ANTONIO TORRONI, JULIE A. MILLER. LORNA G. MOORE. STACY ZAMUDIO, JIANGUO ZHUANG, TAFSHI DROMA, AND DOUGLAS C. WALLACE Department of Genetics and Molecular Medicine, Emory University School of Medicine, (A.T., J.A.M., D.C. W.), and Department of Anthropology, Emory Uniuersity, (D.C. W.), Atlanta, Georgia 30322; Cardiovascular Pulmonary Research Lab, University of Colorado Health Sciences Center, and Departm,ent of Anthropology, University of Colorado at Denuer, Denver, Colorado 80262 (L.G.M., S.Z.); Tibet Institute of Medical Science, Lhasa, Tibet Autonomous Region, China, 850000 (J.Z., T.D.)

KEY WORDS Mitochondrial DNA variation, Tibetans, Asian mtDNA lineages, Haplotypes, High altitude adaptation

ABSTRACT Mitochondrial DNAs (mtDNAs) of 54 Tibetans residing at altitudes ranging from 3,0004,500 m were amplified by polymerase chain reaction (PCR), examined by high-resolution restriction endonuclease analy- sis, and compared with those previously described in 10 other Asian and Siberian populations. This comparison revealed that more than 50% of Asian mtDNAs belong to a unique mtDNA lineage which is found only among Mon- goloids, suggesting that this lineage most likely originated in Asia a t an early stage of the human colonization of that continent. Within the Tibetan mtDNAs, sets of additional linked polymorphic sites defined seven minor lineages of related mtDNA haplotypes (haplogroups). The frequency and dis- tribution of these haplogroups in modern Asian populations are supportive of previous genetic evidence that Tibetans, although located in southern Asia, share common ancestral origins with northern Mongoloid populations. This analysis of Tibetan mtDNAs also suggests that mtDNA mutations are un- likely to play a major role in the adaptation of Tibetans to high altitudes. 0 1994 Wiley-Liss, Inc

The nature and the extent of the genetic variation in the Tibetan population are par- ticularly interesting for at least two reasons. First, there is a marked paucity of genetic data concerning Tibetans because of both their geographic isolation and political events occurring in the late 1950s which fur- ther isolated them. Consequently, a limited amount of information is available to recon- struct their genetic relationships with other Mongoloid populations. Second, Tibetans are likely the largest and oldest high-alti- tude population (Zhimin, 1982), and their adaptation to high altitudes could have a genetic component (Lahiri et al., 1976).

Some of the genetic data about this popu- lation has been obtained from Tibetans who moved to India and Nepal after 1959 (Bhalla, 1972; Patel, 1971, 1973; Santachi- ara et al., 1976; Bhattacharjee et al., 1977; Roychoudhury, 1981; Sharma, 1983; Papiha et al., 1989), and, more recently, several studies of blood cell antigens and immuno- globulins have been carried out on the Tibet-

Received May 17,1993; accepted October 4,1993 Address reprint requests to Douglas C. Wallace, Department

of Genetics and Molecular Medicine, Emory University School of Medicine, Atlanta, Georgia 30322.

0 1994 WILEY-LISS, INC.

A. TORRONI ET AL. 190

ans who still live in Tibet (Ai et al., 1987; Zhao et al., 1987; Matsumoto, 1987; Zhao and Lee, 1989).

Extensive studies of the Gm and Km allo- types, HLA antigens, and dental morphol- ogy of Mongoloid populations have sug- gested that Mongoloids are divided into two groups: a northern group encompassing northern Asia and Siberian populations and a southern group encompassing southeast Asian populations (Matsumoto, 1987; Turner, 1987; Lee et al., 1988; Zhao and Lee, 1989). Interestingly, the same studies also showed that, although Tibetans inhabit a relatively southern region of Asia, they ge- netically clustered within the northern Mongoloid group. This result may be ex- plained by the peculiar orography of Tibet which makes this region more accessible from the north than the southeast.

Analysis of mtDNA is relevant to the Ti- betan population for two reasons. MtDNA studies have provided a new approach to de- termine the genetic relationships and pat- terns of migrations of human populations (Stoneking et al., 1990; Schurr et al., 1990; Vigilant et al., 1991; Ballinger et al., 1992; Torroni et al., 1992). In addition, mtDNA encodes for 37 essential genes of oxidative phosphorylation (Wallace, 1992) and their variation could represent one of the genetic factors involved in the human adaptation to hypoxia.

To clarify the origin of modern Tibetans as well as the possible role of mtDNA muta- tions in their adaptation to high altitudes, mtDNA variation of a sample of Tibetans from Tibet was determined and compared to that reported in east and southeast Asians (Ballinger et al., 1992), and Siberians (Tor- roni et al., 1993b).

This analysis showed that Tibetan mtDNA haplotypes are distributed through- out the entire range of the Asian mtDNA variation, suggesting that modern Tibetans could have experienced gene flow from pop- ulations of the surrounding regions. How- ever, Tibetans maintain particularly high frequencies of certain mtDNA haplotypes which are found only in northern Asian and Siberian populations. Consequently, this finding supports the notion that Tibetans and northern Asian populations share a

common ancestral origin. In addition, the comparison of the Tibetan mtDNA variation with that reported for other Asian popula- tions suggests that mtDNA mutations are unlikely to play a major role in the adapta- tion of Tibetans to high altitude.

SUBJECTS AND METHODS Samples

The 54 Tibetans analyzed were chosen from a sample of 171 subjects living in three different geographic regions. All samples were collected from subjects living at least 20 miles from the closest town. Twenty sam- ples were from nomads residing in the re- gion surrounding the town of Nachu (4,500 m), which is approximately 250 miles north- west of Lhasa. Nineteen samples were from farming villages located in the region of the town of Tsedang (3,600 m) which is 150 miles southeast of Lhasa. This town is lo- cated in a valley which the Tibetans con- sider to be the cradle of Tibetan civilization and the birthplace of their people. The re- maining 15 samples were from farming vil- lages in the region surrounding the town of Linchi (3,000 m), which is about 300 miles southeast of Lhasa.

Information was obtained by interviews of all individuals concerning their birthplace, current or former places of residency, and birthplace and residence of their parents and grandparents. Only subjects of Tibetan ancestry from families who were unrelated through their maternal line and who had lived in the same area for several genera- tions were analyzed for mtDNA variation.

Molecular genetic analysis

DNAs were extracted from buffy coats us- ing the method described in Torroni et al. (1992). The entire mtDNAs were amplified through polymerase chain reaction (PCR) (Saiki et al., 1985) in nine overlapping seg- ments using the primer pairs and amplifica- tion conditions described in Torroni et al. (1993a). Each PCR segment was digested with 14 restriction endonucleases (AluI, AvaII, BamHI, DdeI, HaeII, HaeIII, HhaI, HincII, HinfI, HpaI, MspI, MboI, RsaI, TaqI). The resulting fragments were re- solved through electrophoresis in NuSieve plus SeaKem agarose (FMC BioProducts)

MtDNA VARIATION IN TIBETANS 191

TABLE I. MtDNA haplotypes observed in Tibetans

+ 10394d Haplotype N2 +10397a Haplogroup Altitude3

AS56* AS62* AS65‘

AS123 AS124 AS125 AS126 AS127 AS128 AS129 AS130 AS131 AS132 AS133 AS134 AS135 AS136 AS137 AS138 AS139 AS140 AS141 AS142 AS143 AS144 As145 AS146 AS147 AS148 AS149 As150 AS151 AS152 AS153 AS154 AS155 AS156 AS157 AS158 AS159 AS160

~ ~ 1 1 8 4

1 2 2 2 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 2 1 1 3 1 1 1 1 1 1 1 1 2 5 1 1 1 1 1 1 1

Total 54

A Other

C Other

A A A A A B B B D D D D D D D D E E E F F F F F F G G G

Other Other Other Other Other Other Other Other Other Other

1 1

2 (l), 3 (1) 3 3 1 1 1 3 3 3 1 1 3

3 2 3 1 3 2 1 3 2

1(1) , 2 (31, 3 (1) 1 1 3 3 2 1 1

2 (2), 3 (1)

‘The nomenclature ofthe haplogroups corresponds to that reported in Fig. 1. The polymorphic restriction sites which define the 42 haplo- types observed in the Tibetans (AS56, AS62, AS65, AS118, AS123- AS1601 are listed the Appendix The haplotype of the African outgroup is described in Torroni et al. (1993a). ‘N = number of subjects. ‘Altitudes of 3,000 m, 3,700 m and 4,500 m indicated by 1, 2 and 3, respectively. “Haplotypes already described in Asian populations and numbered according to Ballinger et al. (1992).

gels and visualized by W-induced fluores- cence. This restriction analysis probes about 15-20% of the mtDNA nucleotides and per- mits definition of mtDNA haplotypes (Ta- ble 1).

Phylogenetic analysis

The haplotypes observed in the Tibetans were compared to those previously reported

Koreans

TIBETANS -%& Malaysian Chinese --B_-_--1 Taiwanese Han

Vietnamese Malay Aborigines

Fig. 1. Geographic location of the Asian and Siberian populations whose mtDNA haplotypes were compared to those observed in the Tibetans. Data from east Asian and Siberian populations are from Ballinger e t al. (1992) and Torroni e t al. (1993b1, respectively.

in the other Asian and Siberian populations analyzed with the same set of restriction en- zymes (Fig. 1) (Ballinger et al., 1992; Tor- roni et al., 1993b), and their evolutionary relationships were inferred using parsi- mony analysis (PAUP 3.0s [Swofford 19921). All dendrograms were rooted using a Sene- galese mtDNA haplotype (“African out- group” [Torroni et al., 1993a,b]). This haplo- type was obtained from a mtDNA characterized by the presence of an HpaI site at np 3592. This site defines haplotypes which are African specific and observed in 70-100% of the sub-Saharan Africans but are absent in Asians and Caucasians (HpaI morph-3) (Denaro et al., 1981; Cann et al., 1987; Scozzari et al., 1988, in press). Maxi- mum parsimony (MP) trees were generated through random addition of sequences using the Tree Bisection and Reconnection (TBR) algorithm. Because of the large number of terminal taxa, thousands of MP trees could be obtained. We terminated our searches at 3,000 trees after 1,148 replications and saved no more than 10 MP trees for each replication. Consequently, shorter trees

192 A. TORRONI ET AL

could exist, although none were observed in our analyses.

RESULTS Haplotype analysis

Forty-two haplotypes (AS56, AS62, AS65, AS118, AS123-160) were observed among the 54 Tibetans whose mtDNA variation was determined (Table 1). Four of these hap- lotypes, AS56, AS62, AS65 and AS118, were previously described in other Asian popula- tions (Ballinger et al., 1992) whereas the re- maining thirty-eight (AS123-160) were ob- served for the first time in this analysis. A total of 71 polymorphic restriction sites and the 9-base pair (9-bp) COII-tRNALys inter- genic deletion (Cann and Wilson, 1983; Ho- rai and Matsunaga, 1986; Wrischnik et al., 1987; Hertzberg et al., 1989) defined the 42 haplotypes (Appendix). Virtually all of the Tibetan haplotypes were characterized by sets of linked polymorphic sites already re- ported in Mongoloid populations (Stoneking et al., 1990; Ballinger et al., 1992; Torroni et al., 1992, 1993a,b; Passarino et al., 1993). These linked polymorphic sites define groups of haplotypes which share a common ancestral origin.

Tibetan mtDNAs, as well as mtDNAs from all Mongoloid populations, can be di- vided into two major lineages (Ballinger et al., 1992; Torroni et al., 1993a,b; Passarino et al., 1993). One is characterized by the con- comitant presence of a DdeI site at np 10394 (A to G at np 10398) and an Nu1 site at np 10397 (C to T at np 10400). The DdeI site is generated by an A to G transition at np 10398. The AluI site is generated by the same A to G transition at np 10398 plus an additional C to T transition at np 10400. Of the 42 Tibetan haplotypes, 24 belong to the lineage that possesses these two sites (Table 1). Conversely, the second haplotype lineage lacks these two sites. The remaining 18 Ti- betan haplotypes are members of this sec- ond major lineage.

Within these two major lineages, several additional restriction sites define smaller groups of related haplotypes or haplogroups. Seven of these haplogroups (named A, B, C,

D, E, F, and G1) are common in Tibetans and all together represent approximately 60% of their mtDNAs. Haplogroups C, D, E, and G belong to the + 10394 DdeI\+ 10397 AluI lin- eage, whereas haplogroups A, B, and F are members of the lineage lacking these two sites (Table 1).

Tibetan haplotypes AS56 and AS123- AS127 belong to haplogroup A which is de- fined by a HaeIII site at np 663 (Torroni et al., 1992). Haplotype AS56 was previously reported in the Han from Taiwan (Ballinger et al., 1992) and in Native Americans and is one of the four Asian mtDNAs which colo- nized the Americas (Torroni et al., 1993a). In contrast, haplotypes AS123-AS127 have not been observed in any other human popu- lation.

Haplotypes AS128-AS130 belong to hap- logroup B which is defined by the 9-bp COII- tRNALy" intergenic deletion. Haplotypes of this haplogroup have been observed in nu- merous Asian, Melanesian, Polynesian, and Amerind populations (Horai and Mat- sunaga, 1986; Hertzberg et al., 1989; Schurr et al., 1990; Stoneking et al., 1990; Ballinger et al., 1992; Harihara et al., 1992; Torroni et al., 1992, 1993a; Passarino et al., 1993). However, haplogroup B haplotypes are ab- sent in Siberian and NaDene populations (Shields et al., 1992; Torroni et al., 1993a,b).

Haplotype AS65 is the only group C haplo- type observed in the Tibetans. Haplogroup C is defined by the HincII np 13259 site loss and the AluI np 13262 site gain (Torroni et al., 1992). This haplotype has been reported at low frequencies in Taiwanese Han (Bal- linger et al., 1992) and at high frequencies in Aboriginal Siberian and Amerind popula- tions (Torroni et al., 1993a,b). Similar to haplotype AS56, haplotype AS65 is sug- gested as being one of the four mtDNA hap- lotypes which migrated to the Americas dur- ing the first human migration into that continent (Torroni et al., 1993a,b).

Haplotypes AS131-138 belong to haplo- group D which is defined by a AluI site loss

'Nomenclature of haplogroups A-D is according to Torroni et al. (1992,1993a,b). Haplogroups A, B, C, D, and F correspond to haplotype groupings H, D,+C, R, L, and A (Ballinger et al., 19921, respectively. Haplogroups E and G are new haplotype groupings.

APPENDIX. Polymorphic restriction sites observed in 42 distinct Tibetan mtDNA halotypes (AS56, AS62. AS65. AS1 18, AS1234S160)

Sites Haplotypes

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 5 6 6 1 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 5 5 6 6 2 5 8 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0

663e 10040 1667d1670a 1941f 2734a 2856c 3397k 4732k 4769a* 4830d4831f 5176a 5259b/5261e 5370Y5732j 5566k 6022a 6260e 6331b 6618e 7025a 73351 7598f 7607e 7641a 7828f 8148e 8198a 8309c 8858f 9052d9053f 9644a 9683g 10097e 10394c 10397a 10737k 11092e 11403g 11447k 12008g 12026hio 12406hio 12560a 12629b 132590/13262a 13633eA36340 13702e* 141990* 14268g* 14368g* 14394c 147493' 14773c 15434d15437a 15520e 15925i 16049k 16226a 16246a 16254a 16270c 16303k 16310k 16388198e 164941 16517e

1 0 0 0 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 ~~ ~ ~ ~ ~ ~ ~ ~ . ~ ~ _ _ _

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 ~~ ~~ ~~~- 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 1 1 1 0 1 1 1 1 1 1 0 0 0 0 0 0 1 1 1 1 1 1 0 1 1 1 1 1 1 0 1 1 0 0 0 0 0 0 0 0 0 0 1 1 1 0 1 1 1 1 1 1 0 0 0 0 0 0 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 0 0 0 0 0 0 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 111111111111111111111111111111111111111111 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 1 0 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 1 0 0 0 1 1 1 0 0 1 0 1 1 1 1 0 0 0 0 1 1 1 1 0 1 0 1 1 0 0 1 1 1 0 1 0 0

16528c 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 9-bp del 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

'A 1 indicates the presence of a site, and 0 indicates the absence of a site, except for 9-hp del, where 1 indicates the presence o f a 9-hp deletion between the COII and tRNALys genes and a 0 indicates the absence of the deletion. Sites are numbered from the first nucleotide of the recognition sequence according to the published sequence (Anderson et al., 1981). The restriction enzymes used in the analysis are designated by the following single-letter code: a, Alu I: h, Aua 11: c, Dde I; e, Hae 111; f, Hha I; g, HinfI; h, Hpo I; i, Hpa 11; J , Mbo I; k, Rsa I; I , Taq I; m, Barn HI: n , Hae 11; 0, Hinc 11. Sites separated hy a diagonal line indicate either simultaneous site gains or site lasses for two different enzymes or a site gain for one enzyme and a site loss for another because of a single inferred nucleotide substitution; these sites are considered to be only one restriction site polymorphism in the parsimony analysis. Sitcn marked with an asterisk werc found to be present or absent in all samples contrary to the published sequence.

194 A. TORRONI ET AL.

at np 5176. Different haplotypes belonging to this haplogroup have been observed in numerous Asian (Ballinger et al., 1992), Si- berian (Torroni et al., 1993b), and Amerind (Torroni et al., 1993a) populations.

Haplotypes AS139-141 are defined by a HhaI site loss at np 7598 and belong to hap- logroup E. A different haplotype from the same haplogroup has been previously re- ported in the Koreans (Ballinger e t al., 1992).

Haplotypes AS142-147 are members of haplogroup F and are characterized by the lack of the HincII and HpaI sites at np 12406. Haplotypes from this grouping have a particularly high incidence in southeast Asian populations (Ballinger et al., 1992), but are absent in Native American popula- tions (Torroni et al., 1993a).

Haplotypes AS148-150 are characterized by the presence of a HaeIII site at np 4830 and a HhaI site at np 4831 which define haplogroup G. Members of this haplogroup have been previously reported in the Kore- ans (Ballinger et al., 1992) and Nivkhs (Tor- roni et al., 199313).

Most of the remaining Tibetan haplotypes fall into the +lo394 DdeI\+10397 AluI lin- eage (Table 1). However, because they do not share additional mutations with other hap- lotypes, they do not group in specific haplo- groups. Only two of these haplotypes, AS62 and AS118, have been previously reported. Haplotype AS62 has been observed in the Han from Taiwan, the Malays, and the Ma- lay Aborigines, whereas haplotype AS118 has been observed in the Sabah Aborigines (Ballinger et al., 1992).

Phylogenetic analysis

The genetic relationships among the Ti- betan mtDNA haplotypes and between the Tibetan haplotypes and those described in other Mongoloid populations were further defined through parsimony analysis. One phylogeny encompassing the 42 haplotypes observed in the Tibetans (AS56, AS2, AS65, AS118, AS123-160),106 haplotypes (AS17- AS122) described in other east Asian popu- lations (Ballinger et al., 1992), and 34 haplo- types (Sl-S34) described in three Siberian populations (Torroni et al., 1993b) is shown in Figure 2. This dendrogram is one of the

thousands of MP trees generated by the TBR branch swapping algorithm. It is 334 steps in length, with consistency and retention in- dices of 0.446 and 0.801, respectively. All MP trees are defined by the two major hap- lotype lineages, one characterized by the +lo394 DdeI +lo397 AluI sites, and an- other which lacks these sites. In Figure 2, the Tibetan haplotypes belonging to haplo- groups A-G cluster together with the other Asian haplotypes which harbor the same sets of linked polymorphic sites.

DISCUSSION The analysis of 54 Tibetans has revealed

that more than 60% of their mtDNAs belong to a major haplotype lineage that is defined by the DdeI np 10394 and AluI np 10397 sites. Not only is this lineage prevalent in Tibetans, but 193 out of the 360 Asian and Siberian mtDNAs included in the phylogeny of Figure 2 fall into this lineage. The + 10394 DdeI/+ 10397 AluI lineage also represents a large proportion of the Native American, Australian, and Melanesian mtDNAs (Cann et al., 1987; Stoneking et al., 1990; Torroni et al., l992,1993a), indicating the antiquity of the two mutations which create the two overlapping sites. However, while the DdeI site has been described in mtDNAs from every racial group (Cann et al., 1987; Brown et al., 1992; Shoffner e t al., 1993) and in the most divergent African haplotypes (Cann et al., 1987; Wallace, un-

Fig. 2. A phylogenetic tree which includes 42 Tibetan (AS56, AS62, AS65, AS118, AS123-1601, 106 Asian (AS17-AS1221, and 34 Siberian (Sl-S34) haplotypes. The Asian and the Siberian mtDNA data are from Ball- inger et al. 11992) and Torroni et al. (1993b), respec- tively. The presence or the absence of the 10394 DdeI and 10397 AluI sites define the two major lineages ofthe phylogeny. The capital letters A-G in shaded boxes indi- cate the seven major haplogroups observed in Tibetans. Haplotypes observed in the Tibetans are indicated by bold lines. The numbers at the end of each branch indi- cate different mtDNA haplotypes. The horizontal branch lengths are proportional to the number of muta- tional events that separate the haplotypes. This tree is one of the 3,000 MP trees generated with the TBR branch swapping algorithm.

195 MtDNA VARIATION IN TIBETANS

___

- -1 0394 Ddei -10397Alul 1 -___I_ .

AS105lSI :$

~ AS23

524 AS50

a519

a5142

a536

a5102

1 534 911

AS114 tgii3 AS41 AS1 51

r I a538

AFRICAN OUTGROUP

A TORRONI ET AL

TABLE 2 Percent frequencies of MtDNA haplogroups in mongoloid populations ‘ 196

Haplogroups

Other -~ Populations N A B C D E F G ~.

Na-Dene’ Northern Southern

Amerinds’ Northern Central Southern

Siberians3 Eskimos Chukchi Koryaks Yukagirs Evens Nivkhs Udegeys Evenks Nganasans Sel’kups

Koreans Han (Taiwan) Han (Malaysia) Vietnamese Malays Sabah Aborigines Malay Aborigines Tibetans

Asians4

- - 1.8 57 98.2 - 73 60.3 30.1 4.1 2.7

115 39.1 19.1 20.9 8.7 137 65.7 28.5 2.9 2.2 146 14.4 18.5 28.8 38.4 -

-

- -

50 80.0 24 37.5 46 23.9 27 43 57 46 51 3.9 49 2.0 20

- - - -

-

13 7.7 15.4 20 10.0 35.0 14 7.1 28 ~ 14.3 14 ~ 14.3 32 - 18.8 32 - 3.1 54 11.1 5.6

-

- 20.0 -

16.7 16.7 ND 21.7 8.7 ND 59.3 33.3 ND 58.1 7.0 ND - 28.1 -

19.6 84.3 9.8 38.8 36.7 ND 35.0

- - -

- -

- 23.1 5.0 5.0 - 14.3 - - - - - - - -

3.7 16.7

- -

- - -

-

ND ND - - - -

2.0 ND ND

15.4 5.0

32.2 21.4

3.1 6.3

14.8

-

Melanesia’ 0.8 - New Guineans 119 - 19.3 - 0.8

- -

ND 29.2 ND 45.6 ND 7.4 ND 34.9 5.9 66.0 - 80.4

ND 22.4 - 65.0

15.4 15.3 - 40.0 - 78.6 - 53.5 - 64.3 - 78.1 - 90.6 5.6 35.1

- 79.1

- -

‘ N = number of subjects analyzed; “Other” indicates haplotypes not belonging to haplogroups A-G. Most of the Siberian population samples were only screened for the presence or absence of the mutations that define haplogroups A-D. Therefore, the frequencies of haplogroups E, F, and Gin Siberians is not determined (ND). ‘From Torroni et al. (1992,1993a). ’From Torroni et al. (1993b). From Ballinger et al. (1992). From Stoneking et al. (1990).

6The “Other” haplotypes observed in Native Americans are Caucasian mtDNAs acquired through admixture with Europeans.

published data), the AluI site at np 103972 was not found in approximately 200 Cauca- sian and 150 African mtDNAs (Cann et al., 1987; Brown et al., 1992; Shoffner et al., 1993; Torroni and Wallace, unpublished data). Hence, the AluI np 10397 mutation appears to be a marker of Mongoloid popula- tions. It most likely arose in Asia at a very early stage of human evolution of that conti- nent. Haplotypes belonging to this lineage were carried from southeast Asia to Austra- lia and from Siberia to the Americas during the human colonization of these continents.

In addition to the presence or absence of the DdeI np 10394 and AluI np 10397 sites,

‘The 10397 AluI site was initially misplaced at np 1403 in Cann et al. (1987) and Stoneking et al. (1990) and correctly as- signed at np 10397 in Ballinger et al. (1992).

the majority of Tibetan mtDNAs showed ad- ditional linked polymorphic sites that define specific mtDNA haplogroups (Fig. 2). The frequencies of the seven haplogroups A-G in Tibetans together with those reported in other Mongoloid populations are given in Table 2. Although haplogroups A, B, and F lack the 10397 Mu1 site, they are character- ized by sets of mutations which are found only in Mongoloid populations (Cann et al., 1987; Brown et al., 1992; Shoffner et al., 1993). Therefore, these mutations must also have originated in Asia.

However, these haplogroups have a much more limited distribution than the AluI np 10397 lineage and, hence, a more recent ori- gin. Haplogroup A is found in the Tibetans, the Han, and the Koreans and, a t consider- ably higher frequencies, in Siberian and Na- tive American populations, whereas it was

MtDNA VARIATION IN TIBETANS 197

never observed in southeast Asians and Melanesians (Table 2). Haplogroup B is present in almost all Asian, Melanesian, Polynesian, and Amerind populations. How- ever, haplotypes belonging to this haplo- group were not found in Siberian and north- ern Na-Dene populations. Haplogroup F is common in central and southeast Asian pop- ulations but is rare or absent in Siberians, Native Americans, and Melanesians.

The frequency comparison of haplogroups A-F in Mongoloid populations permits a def- inition of genetic relationships between Ti- betans and other Asian populations. The distribution of haplogroups A, C , D, E, and G clearly separates northern Mongoloids from southern Mongoloids. In the Tibetans, hap- logroups A, C, and D have frequencies of 11.1%, 3.7%, and 16.7%, respectively. These haplogroups have been reported in Han, Ko- rean, and Siberian populations at similar or higher frequencies but were not found in southeast Asians (Table 2) . Similarly, the less common haplogroups E and G, observed in 7.4% and 5.6% of the Tibetans, were pre- viously found only in the Koreans and the Nivkhs from Sakhalin Island. In contrast, two of the haplogroups, B and F, were less informative than the others, because the fre- quencies of haplogroups B and F are similar in northeastern and southeastern Asian populations, and their frequencies in Tibet- ans are in the same range (Table 2) .

The total frequency of the haplogroups A, C, D, E, and G in Tibetans and is 44.5%. Since haplotypes belonging to these haplo- groups were observed only in northern Asians and Siberians, their detection at such a high frequency in Tibetans indicates that the Tibetan population is likely to be closely related to northern Asians. While Ti- bet as well as much of central-east Asia was subject to Mongolian political influence at various times from the thirteenth through the seventeenth centuries (Stein, 1972), it would appear unlikely that the affinities of Tibetan mtDNAs with northern Asian mtDNAs are due to admixture during this relatively recent period. Since both China and Tibet were subject to Mongol rule, it seems unlikely that extensive mtDNA ad- mixture would have occurred with Tibetans

but not with Han or other Chinese popula- tions living in the Yangtze River region in South China (Zhao and Lee, 1989). In addi- tion, a large number of Mongolian women would be required to contribute 40% of the mtDNAs. Since the number of males associ- ated with military occupations is usually much higher than that of females, it seems more likely that most if not all of the mtD- NAs are endogenous to the Tibetan popula- tion.

MtDNA variation and adaptation to high altitude

We employed two approaches to deter- mine whether the nature of the mtDNA variation observed in Tibetans could relate to their adaptation to high altitude. The first was to compare the mtDNA variation observed among Tibetans living at different altitudes (4,500 m, 3,600 m and 3,000 m). In this comparison, no significant differences were found between altitudes in either hap- lotypes or haplogroups (Table 1). Further- more, most of the haplotypes described in more than one subject (AS62, AS65, AS131, AS143, AS1531 were observed in Tibetan in- dividuals from different altitudes. However, this absence of an association between mtDNA haplotypes and different altitudes does not preclude the possibility that spe- cific mtDNA genotypes could facilitate ad- aptation to high altitude, since all samples studied were collected from relatively high altitudes (above 3,000 m).

Second, we compared the frequencies of the mtDNA haplogroups and haplotypes ob- served in high-altitude Tibetans with those reported in the other Asian populations. Our analysis indicates that the mtDNA haplo- groups in Tibetans are at similar frequen- cies to those observed in other southeast Asian or Siberian populations (Table 2). This finding suggests that no major selec- tive pressure occurred for any of the haplo- groups at high altitude.

As for specific haplotypes, most of the haplotypes observed in Tibetans (38 out of 42) are new, being described for the first time in this population and were found in one or few subjects. This result confirms previous observations indicating that most

198 A. TORRONI ET AL.

of the mtDNA variation in Asian non-tribal populations is represented by a large num- ber of new haplotypes (population-specific haplotypes), each of which is usually ob- served at very low frequency (Ballinger e t al., 1992; Torroni et al., 1993a,b). Because of their low frequencies, it is unlikely that these population-specific haplotypes carry mutations conferring a strong selective ad- vantage. Most likely, they represent neutral variation originating in and limited to a spe- cific population. In Tibetans, only one of the population-specific haplotypes, AS153, was observed in a relatively high frequency (9.3%). Haplotype AS153 belongs to +lo394 DdeI\ + 10397 AluI lineage and, in addition to these two mutations, is characterized by the presence of a DdeI site at np 14394. This site is generated by an A to G transition at np 14398 which results in a synonymous mutation in the ND6 gene, probably a neu- tral mutation. Additional mutations could be associated with this haplotype but not detected by the restriction enzymes used. Even if all AS153 haplotypes carry an addi- tional mutation which confers some selec- tive advantage, they would represent only a limited fraction of the Tibetan mtDNA vari- ation, not sufficient to explain the distinc- tive adaptations of the Tibetan population to high altitude (Groves et al., 1993; Zhuang et al., 1993; Zamudio et al., 1993). Therefore, this haplotype most likely represents an- other example of neutral population-specific haplotypes observed in human populations (Stoneking et al., 1990; Torroni e t al., 1993a).

In conclusion, our mtDNA analysis shows that mtDNAs from Asian populations are divided in two major lineages, one of which is widespread among all modern Mongoloid populations and probably originated at the very beginning of the human radiation in Asia. Within each of these two major lin- eages, additional sets of mtDNA linked poly- morphic sites define smaller groups of re- lated haplotypes. The comparison of these haplogroups in Tibetans with those reported for other Mongoloid populations supports previous genetic and physical anthropologi- cal evidence that Tibetans, although living in a southern region of Asia, belong to the northern Mongoloid genetic stock and sug- gests that mtDNA mutations are unlikely to

play a major role in the adaptation of Tibet- ans to high altitudes.

ACKNOWLEDGMENTS We are indebted to Dr. X.F. Cai for his

intellectual contribution and support in the sample collection. This work was supported by NIH grant GM 46915, US Army grant DAMD 17-87-C7202, National Heart, Lung, and Blood Institute grant HL-14985, and NSF grant BNS 8919645.

LITERATURE CITED Ai Q, Yuan Y, Zhao H, Li S, and Du R (1987) Distribu-

tion of red cell blood group systems in Yi, Tibetan and Manchu ethnic groups of China. Gene Geogr. 1:169- 176.

Anderson, SA, AT Bankier, BG Barrell, MHL de Bruijn, AR Coulson, J Drouin, IC Eperon, DP Nierlich, BA Roe, F Sanger, PH Schreier, AJH Smith, R Staden, and IG Young (1981) Sequence and organization of the human mitochondrial genome. Nature 290:457- 465.

Ballinger SW, Schurr TG, Torroni A, Gan YY, Hodge JA, Hassan K, Chen KH, and Wallace DC (1992) South- east Asian mitochondrial DNA analysis reveals ge- netic continuity of ancient mongoloid migrations. Ge- netics 130:139-152.

Bhalla V (1972) Variations in taste threshold for PTC in populations of Tibet and Ladakh. Hum. Hered. 22: 453-458.

Bhattacharjee PN, Ghosh S, and Bhattacharya S (1977) The Tamang: A case of hybridization. Hum. Hered. 27: 1-8.

Brown MD, Voljavec AS, Lott MT, Torroni A, Yang C-C, and Wallace DC (1992) Mitochondrial DNA complex I and 111 mutations associated with Leber’s hereditary optic neuropathy. Genetics 130r163-173.

Cann RL, and Wilson AC (1983) Length mutations in human mitochondrial DNA. Genetics 104t699-711.

Cann RL, Stoneking M, and Wilson AC (1987) Mitochon- drial DNA and human evolution. Nature 325:31-36.

Denaro M, Blanc H, Johnson MJ, Chen KH, Wilmsen E, Cavalli Sforza LL, and Wallace DC (1981) Ethnic vari- ation in HpaI endonuclease cleavage patterns of hu- man mitochondrial DNA. Proc. Natl. Acad. Sci. U.S.A.

Groves BM, Droma T, Sutton JR, McCullough RG, Mc- Cullough RE, Zhuang J, Rapmund G , Sun S, Janes C, and Moore LG (1993) Minimal hypoxic pulmonary hy- pertension in normal Tibetans at 3,658 m. J. Appl. Physiol. 74r312-318.

Haribara S, Hirai M, Suutou Y, Shimizu K, and Omoto K(1992) Frequency ofa 9-bp deletion in the mitochon- drial DNA among Asian populations. Hum. Biol. 64: 161-166.

Hertzberg M, Mickleson KNP, Serjeantson SW, Prior JF, and Trent R J (1989) An Asian-specific 9-base pair deletion of mitochondrial DNA is frequently found in Polynesians. Am. J . Hum. Genet. 44.504-510.

Horai S, and Matsunaga E (1986) Mitochondrial DNA polymorphism in Japanese. 11. Analysis with restric-

78t5768-5772.

MtDNA VARIATION IN TIBETANS 199

tion enzymes of four or five base pair recognition. Hum. Genet. 72105-117.

Lahiri S, Delaney RG, Brody JS, Simpser M, Velasquez T, Motoyama EK, and Polgar C (1976) Relative role of environmental and genetic factors in respiratory ad- aptation to high altitude. Nature 261:133-135.

Lee TD, Zhao TM, Mickey R, Sun YP, Lee G, Song CX, Cheng DZ, Zhou MS, Ding SQ, Cheng DX, Song FJ, Lee PY, An JB, and Mittal KK (1988) The polymor- phisms of HLA antigens in the Chinese. Tissue Anti- gens 32.188-208.

Matsumoto H (1987) Characteristics of the Mongoloid and neighboring populations on the basis of the ge- netic merkers of immunoglobulins. J. Anthrop. SOC. Nippon 95.291304.

Papiha SS, Mastana SS, and Stephenson A (1989) Sero- genetic investigations of Tibetans and Himachalis from Himachal Pradesh, India: Genetic relationship between Tibetans and certain selected Mongoloid pop- ulations. Jpn. J . Hum. Genet. 34:143-157.

Passarino G, Semino 0, Modiano G, and Santachiara- Benerecetti AS (1993) COIIItRNALY” intergenic 9-bp deletion and other mtDNA markers clearly reveal that the Tharu (Southern Nepal) have Oriental affin- ities. Am. J. Hum. Genet. 53:609-618.

Pate1 S (1971) ABO blood groups, PTC tasting and der- matoglyphics of Tibetans at Chandragiri, Orissa. Am. J . Phys. Anthropol. 35t149-150.

Patel S (1973) Study on some genetic traits of Tibetans at Chandragiri, district Ganjam, Orissa. Am. J. Phys. Anthropol. 38:755-756.

Roychoudhury AK (1981) The genetic composition of the people in Eastern India. J. Indian. Anthropol. SOC. 16: 153-1 70.

Saiki RK, Scharf S, Faloona F, Mullis KB, Horn GT, Erlich HA, and Arnheim N (1985) Enzymatic amplifi- cation of a-globin genomic sequences and restriction site analyses for diagnosis of sickle cell anemia. Sci- ence 230:1350-1354.

Santachiara-Benerecetti AS, Baur EW, Beretta M, Ran- zani G, Morpurgo G, Carter ND, D’Udine B, Ranjit SK, and Modiano G (1976) A study of several biochem- ical markers in Sherpas with description of some vari- ant phenotypes. Hum. Hered. 26.351-359.

Schurr TG, Ballinger SW, Gan YY, Hodge JA, Merri- wether DA, Lawrence DN, Knowler WC, Weiss KM, and Wallace DC (1990) Amerindian mitochondrial DNAs have rare Asian mutations a t high frequencies, suggesting they derived from four primary maternal lineages. Am. J. Hum. Genet. 46513-623.

Scozzari R, Torroni A, Semino 0, Sirugo G, Brega A, and Santachiara-Benerecetti AS (1988) Genetic studies on the Senegal population. I. Mitochondrial DNA poly- morphisms. Am. J . Hum. Genet. 43:534-544.

Scozzari R, Torroni A, Semino 0, Cruciani F, Spedini G, and Santachiara-Benerecetti AS (in press) Mitochon- drial DNA polymorphisms in Bamileke. Hum. Biol.

Sharma JC (1983) Dental morphology and odontometry of the Tibetan immigrants. Am. J . Phys. Anthropol. 61 t495-505.

Shields GF, Hecker K, Voevoda MI, and Reed JK (1992) Absence of the Asian-specific region V mitochondrial marker in Native Beringians. Am. J. Hum. Genet. 50: 758-765.

Shoffner JM, Brown MD, Torroni A, Lott MT, Cabell M, Mirra SS, Beal MF et al. (1993) Mitochondrial DNA mutations associated with Alzheimer disease and Parkinson disease patients. Genomics 17r171-184.

Stein RA (1972) Tibetan Civilization. Stanford: Stanford University Press.

Stoneking M, Jorde LB, Bhatia K, and Wilson AC (1990) Geographic variation in human mitochondrial DNA from Papua New Guinea. Genetics 124.717-733,

Swofford D (1992) Phylogenetic Analysis Using Parsi- mony (PAUP), Version 3.0s. Illinois Natural History Survey, Champaign, Illinois.

Torroni A, Schurr TG, Yang C-C, Szathmary EJE, Williams RC, Schanfield MS, Troup GA, Knowler WC, Lawrence DN, Weiss KM, and Wallace DC (1992) Na- tive American mitochondrial DNA analysis indicates that the Amerind and the Nadene populations were founded by two independent migrations. Genetics 130.153-162.

Torroni A, Schurr TG, Cabell MF, Brown MD, Nee1 JV, Larsen M, Smith DG, Vullo CM, and Wallace DC (1993a) Asian affinities and continental radiation of the four founding Native American mitochondrial DNAs. Am. J. Hum. Genet. 53:563-590.

Torroni A, Sukernik RI, Schurr TG, Starikovskaya YB, Cabell MF, Crawford MH, Comuzzi AG, and Wallace DC (1993b) Mitochondrial DNA variation of Aborigi- nal Siberians reveals distinct genetic affinities with Native Americans. Am. J. Hum. Genet. 53591- 608.

Turner CG (1987) Late Pleistocene and Holocene popu- lation hystory of East Asia based on dental variation. Am. J. Phys. Anthropol. 73.305-321.

Vigilant L, Stoneking M, Harpending H, Hawkes K, and Wilson A (1991) African populations and the evolution of human mitochondrial DNA. Science 253.1503- 1507.

Wallace DC (1992) Diseases of the mitochondrial DNA. Annu. Rev. Biochem. 61:1175-1212.

Wrischnik LA, Higuchi RG, Stoneking M, Erlich HA, Arnheim N, and Wilson AC (1987) Length mutations in human mitochondrial DNA Direct sequencing of enzymatically amplified DNA. Nucleic Acids Res. 15: 529-542.

Zamudio S, Droma T, Yonzon K, Aharya G, Zamudio JA, Niermeyer SN, and Moore LG (1993) Protection from intrauterine growth retardation in Tibetans at high altitude. Am. J. Phys. Anthropol. 91:215-224.

Zhao TM, and Lee TD (1989) Gm and Km allotypes in 74 Chinese populations: A hypothesis of the origin of the Chinese nation. Hum. Genet. 83t101-110.

Zhao TM, Zhang GL, Zhu YM, Zhen SQ, Liu DY, Chen Q, and Zhang X (1987) The distribution of immunoglobu- lin Gm allotypes in forty Chinese populations. Acta Anthropologica Sinica 6:l-9.

Zhimin A (1982) Palaeoliths and microliths from Shenia and Shuanghu, Northern Tibet. Current Anthropol- ogy 23:493-499.

Zhuang J, Droma T, Sun S, Janes C, McCullough RE, McCullough RG, Cymerman A, Huang SY, Reeves JT, and Moore LG (1993) Hypoxic ventilatory responsive- ness in Tibetan compared with Han residents of 3,658 m. J. Appl. Physiol. 74t303-311.