December 22, 2010

Post-glacial expansions of mtDNA haplogroups C and D in northern Asia

Table S1 has the C frequencies and Table S2 the D ones.

PLoS ONE 5(12): e15214. doi:10.1371/journal.pone.0015214

Origin and Post-Glacial Dispersal of Mitochondrial DNA Haplogroups C and D in Northern Asia

Miroslava Derenko et al.


More than a half of the northern Asian pool of human mitochondrial DNA (mtDNA) is fragmented into a number of subclades of haplogroups C and D, two of the most frequent haplogroups throughout northern, eastern, central Asia and America. While there has been considerable recent progress in studying mitochondrial variation in eastern Asia and America at the complete genome resolution, little comparable data is available for regions such as southern Siberia – the area where most of northern Asian haplogroups, including C and D, likely diversified. This gap in our knowledge causes a serious barrier for progress in understanding the demographic pre-history of northern Eurasia in general. Here we describe the phylogeography of haplogroups C and D in the populations of northern and eastern Asia. We have analyzed 770 samples from haplogroups C and D (174 and 596, respectively) at high resolution, including 182 novel complete mtDNA sequences representing haplogroups C and D (83 and 99, respectively). The present-day variation of haplogroups C and D suggests that these mtDNA clades expanded before the Last Glacial Maximum (LGM), with their oldest lineages being present in the eastern Asia. Unlike in eastern Asia, most of the northern Asian variants of haplogroups C and D began the expansion after the LGM, thus pointing to post-glacial re-colonization of northern Asia. Our results show that both haplogroups were involved in migrations, from eastern Asia and southern Siberia to eastern and northeastern Europe, likely during the middle Holocene.



German Dziebel said...

The authors stress how widely C and D lineages are dispersed and that they show up as far west as West Asia and Europe. But they fail to mention C1 in Iceland.

This can be explained by their focus on "Asian" Cs and Ds, while C1 is mostly Amerindian, but this would be a lame excuse, especially since they list C1 frequencies in their Suppl Table 1.

German Dziebel said...

"Importantly, we have not found in northern Asia any genetic signatures of sufficient antiquity to indicate traces of pre-LGM expansions, that originated from the Upper Paleolithic industries that were present both in the southern Siberia and Siberian Arctic, and that date back to ,30 kya, well before the LGM [1,34,36]. Apparently, the Upper Paleolithic population of northern Asia did not leaving a genetic mark on the female lineages of modern Siberians. It is probable that the initial population expansion in the southern Siberia region involved maternal lineages other than haplogroups C and D. Nevertheless none of the remaining northern Asian haplogroups became as frequent in Siberia as haplogroups C and D."

This is a conclusion important for theories of the peopling of the Americas. Amerindians don't have the same C and D lineages as Siberians. They also have mtDNA lineages that are entirely missing in Siberia, such as B and X. This means Amerindians couldn't have derived from Siberians as a result of the recolonization of Northeast Asia after LGM. It means Amerindians are much older than 12,000 years and they colonized America at least during the initial expansion of C and D lineages in East Asia, i.e. around 40K years ago. This is consistent with another study that said: "So far, there is no clearly identifiable progenitor of the Native American D1 in contemporary populations of Siberia and Beringia, and it is also absent from Asia."

Andrew Oh-Willeke said...

The absence of common mtDNA C and D haplogroup lineages that are private to Japan is notable, as the paper suggests that basal mtDNA C and D haplogroups should have been emerging somewhere in the general vicinity around the time of Jomon settlement of Japan ca. 30,000 years ago. Since the archaeological record in Japan is so clear, and the history of migration prior to the historic era is so thin, this example is one of the most helpful in putting minimum dates on lineages that are not dependent upon mutation rates.

"The matrilineal side (mtDNA) is more varied. There 15 main lineages and many more subclades or minor haplogroups. Those of typical Sino-Korean origin are haplogroups A (mostly A4 and A5), B (mostly B4 and B5), C, D (mostly D4 and D5), F, M8a, M9, M10, M11 and Z. Together they are present in 67% of the Japanese population.

M7, although present in most of East Asia, is usually present as M7a in Japan, and has been associated with the Jomon and Ainu people. Haplogroup N9b is unique to Japan[.] N9a is found in Southeast Asia, parts of China, and throughout Japan, but is absent from Korea or Eastern China, and is consequently also surely of Jomon origin.

Haplogroups G and Y are normally found in Western and Eastern ends of Siberia. These lineages are more common among the Ainu, both inside and outside Japan. The Ainu would have acquired these haplogroups through population exchange (intermarriages, probably) with their Siberian neighbours. . . .

[A] few samples of haplogroup D1 were found in skeletons from the Jomon period (although N9b was the dominant lineage). This is interesting because D1 is normally not found among East Asians but among Native Americans. . . . D1 is now extremely rare in the modern Japanese population, if it still exist at all.. The most likely explanation is that D1 first appeared in Siberia then migrated to the Americas, and that a few women carrying this lineage married into other Siberian tribes that eventually came into contact with the Ainu then Jomon people, after many generations of geographic drift."

The linked source doesn't make clear the age of the Jomon ancient mtDNA D1 specimens, which could be anywhere in the time range from 30,000 years ago to 2,000 years ago. An ancient DNA D1 specimen late in that range (particularly if it is after Austronesian, Paleoeskimo and/or Inuit migrations established the existence of viable long distance sea travel in the region) together with its low frequency in Japan would suggest that D1 was not part of the founding Jomon mix and arrived through minor admixture events after Jomon populations had stabilized near their peaks. On the other hand, if the specimen is very old, one might favor a theory that mtDNA D1 was present at a very low frequency in the founding population and simply wasn't very successful, perhaps due to random chance (or at any rate for reasons other than selective genetic advantage).

GFC said...

Following up on the Andrew Oh-Willeke post, regardless of whether or not D1 originated in the proto-Jomon range or in Siberia, it is clearly not a Beringian lineage, as was reported by Tamm et al.(2007).

The erroneous identification of D1 as a Beringian lineage has led to the erroneous conclusion of Chatters et al.(2014) that "Naia," the 12,000 - 13,000 year old skeleton from the Yucatan cave, was a descendant of the Siberian-Beringian migration.

As can be seen from the references in Chatters et al. (2014) there is no consideration of the Adachi et al.(2009/10) or Derenko et al. (2010) papers, which postdate the Tamm et al.(2007) paper. That is a very serious omission. The Tamm et al.(2007) paper is based on a very shaky foundation after those two papers.

Unknown said...
This comment has been removed by the author.