Studies of the early peopling of east Asia have long been a subject of interest in the field of human evolutionary history. Current hypotheses can be classified into two major models of the early migration routes into east Asia. The first model postulates that a southeast Asian origin is most likely, followed by a northward migration (Turner 1990). Recent genetic surveys with autosomal microsatellite markers (Chu et al. 1998) and Y-chromosomal binary markers (Su et al. 1999) support this model. In contrast, the second model suggests a bi- and/or multidirectional route: one migration through central Asia and one through southeast Asia (Nei and Roychoudhury 1993; Cavalli-Sforza et al. 1994; Karafet et al. 2001). Ding et al. (2000) have also noted that southeast Asia is not the homeland for northeast Asian populations, but the potential importance of more recent gene flow from central Asia needs to be stressed for the peopling of northeast Asia. Since Korea and Japan lie between the southeast and northeast Asian gene pools, their population genetic data can give us valuable information about the prehistoric migration route(s) and population expansions in east Asia.

The Koreans are generally considered a northeast Asian group, since the Korean peninsula is bounded to the north by China, to the northeast by Russia, and to the south by the Korean Strait and Japan. The Korean Peninsula and Japanese Archipelago were contiguous land before the rise in sea levels between 10,000 and 7,000 years ago (Glover 1980). Based on the results of archeological data, the earliest modern human lithic cultures are found from 25,000 to 45,000 years ago in the Altai Mountains and southeastern Siberia and the Korean Peninsula (Vasil'ev 1993; Choi 1993). Anthropological and/or archeological evidence suggests that the early Korean population was related to Mongolian ethnic groups who inhabited the general area of the Altai Mountains in central Asia (Kim 1970). Therefore, the ancient Koreans (proto-Koreans) may have shared a common origin with the northeast Asian groups who inhabited the general area of the Altai Mountains and Lake Baikal regions of southeastern Siberia.

There is also evidence for recent migration and range expansions via north China to Korea (Chard 1974; Hammer and Horai 1995; Yun 1998; Choi and Rhee 2001). According to Korea's founding myths, the Ancient Chosun (the first state-level society) was established around 2,333 BC in the region of southern Manchuria but later moved into the Pyongyang area of northwest Korea. Recent northeastward migration from China, beginning in the 3rd century BC, led to the decline of the Ancient Chosun, with various Ancient States in the regions of southern Manchuria and the Korean Peninsula emerging. In addition, archeological evidence implies that rice cultivation had spread to all parts of the Korean Peninsula by around 1,000 BC, introduced from the Yangtze River basin in southern China (Choi and Rhee 2001).

Studies of classical genetic markers (protein and nuclear DNA) show that Koreans tend to have a close genetic affinity with Mongolians among northeast Asians (Goedde et al. 1987; Saha and Tay 1992; Hong et al. 1993). In contrast, genetic surveys of mitochondrial DNA (mtDNA) variation indicate that the Koreans are more closely related to the Chinese and Japanese among east Asian populations (Harihara et al. 1988; Horai et al. 1996; Jin et al. 1999). Recent studies of Y-chromosomal DNA markers show that the Koreans possess lineages from both northeast and southeast Asia (Kim et al. 2000; Kwak and Kim 2001; Karafet et al. 2001). These results have led us to consider that the peopling of Korea is likely to have involved multiple events, and that different aspects could be revealed by different molecular genetic markers and DNA samples. Thus, further markers and DNAs from diverse regions of east Asia are required to explore the origin and genetic history of modern Koreans.

The present study, therefore, was designed to trace the male lineage history of the Koreans and their relationships to other east Asian ethnic groups. We analyzed eight Y-chromosomal binary markers (YAP, RPS4Y₇₁₁, M9, M175, LINE1, SRY₊₄₆₅, 47z, and M95) and three Y-STR markers (DYS390, DYS391, and DYS393) in samples from a total of 738 males from several regions of southeast and northeast Asia. We report that the peopling of Korea can be seen as a complex process with genetic contributions involving the northern Asian settlement and range expansions mostly from southern-to-northern China. A recent migration event leading to a Y-chromosome contribution from Korea to Japan is also discussed.

We observed nine different Y haplogroups by using the eight binary markers (Fig. 1). The structure of this tree is consistent with previous results from east Asia (Su at al. 1999; Karafet et al. 2001; Tajima et al. 2002; Kayser et al. 2003). The frequency distributions of the eight binary markers and corresponding haplogroups are listed in Tables 1 and 2 and displayed in Fig. 2. The east Asian populations were characterized by a high frequency (96.6%) of three major Y-chromosomal lineages (haplogroups DE-YAP, C-RPS4Y₇₁₁, and K-M9). These three distinct lineages coalesce to the M168-T lineage, which probably originated in Africa (Underhill et al. 2000; Ke et al. 2001). Our results, therefore, provide additional data confirming that the east Asian Y-chromosomal gene pool is almost completely contained within these three Y lineages (Tajima et al. 2002).

Fig. 1  Y-chromosome parsimonious tree showing the evolutionary relationship of nine haplogroups constructed from the eight binary markers. Nomenclature is according to the Y Chromosome Consortium (2002)

---

Table 2  Distribution of Y-chromosomal haplogroup frequencies in east Asian populations

Population (n)	No. (%) of
	Y*(xC,DE,K)	C-RPS4Y711	DE-YAP	K*-M9	O*-M175	O-LINE1	O-SRY+465	O-47z	O-M95
Northeast Asians:
N. Chinese (113)
Beijing-Han (69)	12 (17.4)	4 (5.8)	1 (1.4)	3 (4.3)	35 (50.7)	10 (14.5)	4 (5.8)	0	0
Manchurian (44)	0	10 (22.7)	1 (2.3)	2 (4.5)	22 (50.0)	2 (4.5)	2 (4.5)	5 (11.4)	0
Japanese (108)	0	10 (9.3)	37 (34.3)	2 (1.9)	28 (25.9)	2 (1.9)	7 (6.5)	21 (19.4)	1 (0.9)
Koreans (160)	1 (0.6)	24 (15.0)	4 (2.5)	12 (7.5)	68 (42.5)	20 (12.5)	22 (13.8)	9 (5.6)	0
Mongolians (100)
Buryats (51)	2 (3.9)	19 (37.3)	2 (3.9)	17 (33.3)	7 (13.7)	3 (5.9)	1 (2.0)	0	0
Khalkhs (49)	2 (4.1)	21 (42.9)	1 (2.0)	12 (24.5)	10 (20.4)	3 (6.1)	0	0	0
Southeast Asians:
S. Chinese (39)
Yunnan (39)	1(2.6)	3 (7.7)	1 (2.6)	9 (23.1)	19 (48.7)	5 (12.8)	0	0	1 (2.6)
Indonesians (36)	0	3 (8.3)	0	2 (5.6)	10 (27.8)	4 (11.1)	6 (16.7)	1 (2.8)	10 (27.8)
Philippines (77)	3 (3.9)	3 (3.9)	0	8 (10.4)	52 (67.5)	9 (11.7)	1 (1.3)	0	1 (1.3)
Thais (55)	4 (7.3)	0	1 (1.8)	3 (5.5)	15 (27.3)	3 (5.5)	1 (1.8)	2 (3.6)	26 (47.3)
Vietnamese (50)	0	11 (22.0)	0	2 (4.0)	23 (46.0)	2 (4.0)	5 (10.0)	2 (4.0)	5 (10.0)

Fig. 2  Distribution of Y haplogroups in east Asia. Circle area is proportional to sample size, and the nine haplogroups are represented by different colors

---

The distribution of Y-chromosomal variation surveyed here reveals significant genetic differences among east Asian populations. Haplogroup DE-YAP (the YAP+ allele) was present at high frequency only in the Japanese and was rare in other parts of east Asia (Table 2, Fig. 2). This result is consistent with previous findings of YAP+ chromosomes only in populations from Japan and Tibet in east Asia (Hammer and Horai 1995; Hammer et al. 1997; Kim et al. 2000; Tajima at al. 2002). However, haplogroup DE-YAP is also found at low frequencies in all the other northeast Asian populations sampled here (2.4% overall, excluding the Japanese; 9.6%, including the Japanese), but only in two of the southern populations (0.8% overall), suggesting that the Korean YAP+ chromosomes are unlikely to have been derived from a southeast Asian source. The prevalence of the YAP+ allele in central Asian populations suggests a genetic contribution to the east Asian populations from the northwest, probably from central Asia (Altheide and Hammer 1997; Jin and Su 2000; Karafet et al. 2001).

Haplogroups C-RPS4Y₇₁₁ and K-M9 were widely but not evenly distributed in the east Asian populations. Haplogroup C-RPS4Y₇₁₁ appears to be the predominant northeast Asian haplogroup, with high frequencies in Mongolians (Buryats, 37.3%; Khalkhs, 42.9%) and Manchurians (22.7%; Table 2, Fig. 2). The moderate frequency of haplogroup C-RPS4Y₇₁₁ Y-chromosomes in Korea (15.0%) implies a genetic influence from northern populations of east Asia, starting possibly in east Siberia. Su and Jin (2001) suggest that the RPS4Y₇₁₁-T chromosome originated in east Asia, probably in the southeast, and then expanded to the north (Siberia), based on the genetic diversity of Y-STR markers. However, the observed low Y-STR diversity of haplogroup C-RPS4Y₇₁₁ chromosomes in their surveys of Siberian and central Asian populations compared with east Asian populations could also be explained by a more northern (Mongolian and/or Siberian) origin followed by genetic drift resulting from small effective population sizes (Pakendorf et al. 2002). Recently, Cavalli-Sforza and Feldman (2003) have suggested that haplogroup C-RPS4Y₇₁₁ expanded both through a southern route from Africa (e.g., India) to Oceania, and a northern one to Mongolia, Siberia, and eventually to northwest America. Further genetic surveys are required to test these hypotheses, with additional markers and more samples from diverse regions of Asia.

Although haplogroup O-LINE1 was found at moderate to low frequencies in all the east Asian populations examined (8.5% overall), Han Chinese have the highest frequency of this L1 retroposon insertion polymorphism (15%). We therefore suspect that this lineage might have emerged in China (probably southern China) and been carried south and east, eventually extending into the surrounding regions. This result is consistent with recent surveys showing that O-LINE1 Y-chromosomes are present at high frequencies in samples of the Han Chinese and two minority populations, the Tujia and Miao from Hunan located in the southern areas of the Yangtze River (Santos et al. 2000; Kwak and Kim 2001). The moderate frequency of haplogroup O-LINE1 Y-chromosomes in the Korean population (12.5%) may have resulted from its interaction with Chinese populations. Southeast Asian populations, except for the Philippines, are characterized by a high frequency of haplogroup O-M95 (16.7%; Table 2, Fig. 2). This result is concordant with recent surveys carried out by Su et al. (1999) and Kayser et al. (2003). Based on the result of the O-M95 Y-chromosome distribution, Koreans are not closely related to the most southern east Asians but tend to be more related to the population of southern-to-northern China.

The prevalence of haplogroups carrying the SRY₊₄₆₅-T allele in Korean and Japanese populations suggests a strong genetic affinity between these two populations (Tables 1 and 2, Fig. 2). Haplogroup O-47z Y-chromosomes are direct descendants (a sublineage) of haplogroup O-SRY₊₄₆₅ (Fig. 1). Shinka et al. (1999) reported that Y-chromosomes carrying the SRY₊₄₆₅-T allele were not present in most European and African males examined in their survey. Lin et al. (1994) suggested that the O-47z Y-chromosomes (Y2 allele) might have originated from an ancestral population in Henan or southern parts of Shanxi near the Yellow River in China. Many ancient Chinese moved from the estuary of the Yellow River to the middle and downstream regions of the Yangtze River ~4,000 years ago (Ruofu and Yip 1993; Su et al. 2000). With increasing political chaos in the Chinese mainland during its Warring Period (476–221 BC), many Chinese moved further southward and eastward, and eventually inhabited all of China (Eberhard 1980). There is also historical evidence that many Chinese fled and sought refuge in the Korean Ancient Chosun during the Warring Period (Yun 1998; Choi and Rhee 2001). In addition, archeological evidence indicates that rice cultivation had spread to all parts of the Korean peninsula around 1,000 BC, introduced from the Yangtze River basin in southern China (Choi and Rhee 2001). The recent range expansion and introduction of rice cultivation from southern China may have resulted in the appearance of Y-chromosomal lineages carrying haplogroup O-M175-derived markers in Korea.

In this study, the Koreans appear to be most closely related overall to the Manchurians among east Asian ethnic groups (Fig. 2), although a principal components analysis of haplogroup frequencies reveals that they also cluster with populations from Yunnan and Vietnam (Fig. 3). The genetic relationship with Manchuria is consistent with the historical evidence that the Ancient Chosun, the first state-level society, was established in the region of southern Manchuria and later moved into the Pyongyang area of the northwestern Korean Peninsula. Based on archeological and anthropological data, the early Korean population possibly had a common origin in the northern regions of the Altai Mountains and Lake Baikal of southeastern Siberia (Han 1995; Choi and Rhee 2001). Recent studies of mtDNA (Kivisild et al. 2002) and the Y-chromosome (Karafet et al. 2001) have also indicated that Koreans possess lineages from both the southern and the northern haplogroup complex. In conclusion, the peopling of Korea can be seen as a complex process with an initial northern Asian settlement followed by several migrations, mostly from southern-to-northern China.

Fig. 3  Principal components (PC) analysis of haplogroup frequencies in 11 east Asian populations (circle Koreans, open diamonds southeast populations, closed diamonds northeast populations)

---