The origin of the Africa-into-Neandertal mtDNA introgression hypothesis

5 minute read

I was reading through some papers for a post on neutral evolution versus selection in human diversification. That’s a topic I’ve written about several times, and one of my papers (written with Jim Ahern and Sang-Hee Lee) in particular focused on the relatively great morphological differences between human populations compared to some other primates.

The following passage, near the end of the paper, comes directly after a discussion of morphological distances (which is why genetics are “on the other hand” here).

On the other hand, the genetic similarities across the geographic range of hominoid species can indicate a history of gene flow (Gagneux et al., 2001). Genetic evidence for recent gene flow includes great similarities among some genes, amid a background of strong genetic differentiation among subspecies. From ancient DNA evidence, it would appear that the Neandertal-human difference is smaller than that among subspecies of either chimpanzees or gorillas (Krings et al., 1999; Hawks and Wolpoff, 2001), despite the relatively greater morphological difference. It is currently premature to claim that the known pattern of ancient mtDNA diversity is evidence of gene flow between the ancestors of Neandertals and the ancestors of living people, and indeed such a finding does not preclude the hypothesis that Neandertals speciated after this shared ancestor. But certainly the critical morphological evidence to address whether Neandertals were a distinct species is the pattern of change in both populations over time (Hawks and Wolpoff, 2001). The hypothesis that Neandertals and modern humans were conspecific predicts that they shared a common evolutionary trajectory, regardless of the degree of morphological difference between them. Future multivariate work may pursue the extent to which different characters may provide comple- mentary evidence of such shared trajectories among ancient human populations.

I’ve bolded the part that struck me as I was re-reading the paper today. Look at that, we predicted that the mtDNA of Neanderthals came from modern humans as an introgression!

What drove us to this provocative (and in our words, “premature”) hypothesis?

We were writing in this paper about how to interpret what seemed like large morphological differences between modern humans and Neanderthals. Modern humans and Neanderthals are morphologically more different from each other than different subspecies of chimpanzees, for example. We pointed out that it is puzzling that two “species” that look morphologically different still have mtDNA sequence divergence that (at the time) was estimated to be less than 500,000 years. That was a lot less than the estimated mtDNA divergence of chimpanzee subspecies.

At that time, what we knew about Neanderthal DNA came exclusively from Svante Pääbo’s research group. They had found that the mtDNA sequences of Neanderthals share sequence motifs that they never found in any mtDNA sequences of living humans. In particular, a paper by David Serre and coworkers from Pääbo’s lab had shown that 24 Neanderthal and 40 early modern humans in Europe had complete lack of haplotype sharing. This reinforced the finding that a Neanderthal mtDNA clade had once existed, an outgroup to all the mtDNA clades found in living people, and that no surviving descendants of the Neanderthal mtDNA clade had yet been found. Laurent Excoffier and Mathias Currat (2004) had also emphasized the lack of modern humans who share the Neanderthal mtDNA haplotype, especially unexpected in a growing population of modern humans.

That view carried a lot of weight before the initial sequencing of a part of the Vindija 33.16 genome in 2006, which for the first time raised the substantial likelihood of introgression from Neanderthals. And of course the last eight years have completely transformed matters. We now know that (1) modern people have Neanderthal ancestors, (2) the Neanderthal component of ancestry is higher in East Asia than elsewhere, (3), modern populations almost everywhere in the world experienced massive genetic turnovers during the last 30,000 years, meaning that the immediate post-Neanderthal populations of Europe have little to do with today’s Europeans.

At any rate, Krings and coworkers (1997) first recovered Neandertal mtDNA from the Feldhofer 1 skeleton, finding that it had a type never yet found in any living people. They estimated that this mtDNA haplotype had diverged from the mtDNA clade leading to modern humans around 465,000 years ago. That time of divergence appeared to put an upper limit on the time that modern humans and Neandertals parted ways.

That limit was surprisingly young, much shorter than the divergence times at which other primate sister species have become intersterile. In particular, as we pointed out in 2005, the time was very young compared to the mtDNA divergence times of today’s chimpanzee subspecies.

One possibility to consider for Neandertals and modern humans is that mtDNA had introgressed from one to the other, despite the fact that Neandertals and modern humans were more morphologically different than chimpanzee subspecies.

This has turned out to be almost certainly true. Yet the morphological difference between Neandertals and modern humans is still anomalously large compared to that between chimpanzee subspecies—or, more probably, the chimpanzee morphological differences are anomalously small compared to their great genetic diversification. The good evidence for recurrent introgression among Neandertals, modern humans, and Denisovans (and among chimpanzee subspecies) makes none of this much of a surprise today.

But the reason for the mtDNA introgression from ancestral African to Neandertal populations is unknown. Was there some adaptive value to the African mtDNA that caused it to succeed in later Neandertals? Did the original Neandertal mtDNA variation suffer from the high genetic load that other Neandertal genes later did? Was this all just chance?

At the moment, we have no easy way to tell.


Ahern, J. C., Hawks, J. D., & Lee, S. H. (2005). Neandertal taxonomy reconsidered… again: a response to Harvati et al. (2004). Journal of human evolution, 48(6), 647-652.

Currat, M., & Excoffier, L. (2004). Modern humans did not admix with Neanderthals during their range expansion into Europe. PLoS biology, 2(12), e421.

Serre, D., Langaney, A., Chech, M., Teschler-Nicola, M., Paunovic, M., Mennecier, P., ... & Pääbo, S. (2004). No evidence of Neandertal mtDNA contribution to early modern humans. PLoS biology, 2(3), e57.