Neandertal Nuclear Genome: Multiregional Evolution is the new Out of Africa

Green and colleagues announced the Neandertal nuclear genome in tomorrow’s issue of the journal Science. Hitherto only complete mitochondrial DNA (mtDNA) genomes had been recovered. These are only inherited maternally, and the genetic differences between the Neandertal mtDNA and that of modern humans seemed to suggest that Neandertals and humans didn’t mix, that is that they were replaced by “anatomically modern humans” (whatever that phrase means). mtDNA is special as far as genetic stuff goes – only inherited maternally, so only tells about one strain out of a slew of ancestors; doesn’t recombine; as a result, selection acting on a part results on selection of the entire mitochondrial genome; oh and it’s certainly not selectively neutral.
So should we have been wary when it was suggested by mtDNA that neandertals and humans were separate species (recall the issue was even crazier with the Denisova mtDNA specimen…)?

This is a big deal, because for the past several decades researchers have debated the nature of modern human origins. On morphological and shaky mtDNA evidence, several researchers have argued that modern humans emerged from a small African population, which then spread throughout the world between 100-200 thousand years ago and replaced all other ‘archaic’ human populations. Intuitively this doesn’t make sense, and today’s neandertal announcement renders the Out of Africa with Replacement model for human origins absolutely untenable.
So, were Neandertals and (even then-modern) humans the same species? Yes! If Neandertals were a different species, we would expect all humans to be equally genetically divergent from neandertals. But this is not what Green and colleagues found. Rather, the genomes of a French person, a Chinese person, and a Papua New Guinean were actually more similar to the Neandertal genomes than the two African human representatives were to the Neandertals. Such disparate divergences mean we’re dealing with genetic variation within a species, rather than between species.
In fact, the authors estimate that about 1-4% of modern, non-African genomes are derived from Neandertals. Plagnol and Wall (2006) estimated around 5% of human genes come from ‘archaic’ humans, so it is good to see corroborating evidence from two sources. It is interesting, however, that earlier candidates for introgression from archaics, such as the microcephalin haplogroup D, do not appear to have come from Neandertals (maybe another archaic population, then?).
The authors were also able to use these neandertal and modern human genomes to estimate regions of the human genome that have been under recent and accelerated evolution, including:
  • SPAG17 is associated with sperm motility – is this evidence for sperm competition and recent sexual selection?
  • Regions in which, among modern humans, mutations are associated with social-cognitive diseases like schizophrenia and autism
  • RUNX2, again where misexpression in humans is associated with dysgenesis of frontal bone (forehead), shoulder and rib-cage shape morphology
I think the only things I would have loved to have seen in this study are simple logistical issues, things that are probably simply not practical at the moment because of technological constraints. First, I’d love to see a much larger set of modern human reference genomes. The study included only 2 human nuclear genomes from sub-Saharan Africa, 1 from Europe, 1 from China and 1 from Papua New Guinea. Yes, this samples variation from all over the world, but it’s 5 out of nearly 7 billion genomes out there today. At the moment, however, it’s just not that easy to acquire and handle genomic data for many individuals.
Second, I’d like to see nuclear genome comparisons using Upper Paleolithic modern humans – ‘modern human’ contemporaries of Neandertals. The Denisova mtDNA was surprising because, at some 40 ka, its genome was about twice as different from modern humans as the neandertal mtDNA sequences were. Just what kind of genetic diversity are we looking at in ancient (anatomically both ‘archaic’ and ‘modern’) humans?
Green and colleagues should be lauded because of how meticulously they went about this project. They took major pains to circumvent issues of contamination, they maximized the DNA they could obtain in spite of preservation issues, they came up with some clever tests. And their results are really interesting.
Green RE et al. 2010. A draft sequence of the Neandertal genome. Science 328: 710 – 722.
Plagnol V and Wall JD. 2006. Possible Ancestral Structure in Human Populations. PLoS Genetics 2(7): e105

Neandertal mtDNA genome sequenced

A neandertal mtDNA genome has been sequenced <!–[if supportFields]> ADDIN EN.CITE Green200840440417Green, Richard E.Malaspinas, Anna-SapfoKrause, JohannesBriggs, Adrian W.Johnson, Philip L. F.Uhler, CarolineMeyer, MatthiasGood, Jeffrey M.Maricic, TomislavStenzel, UdoPrüfer, KaySiebauer, MichaelBurbano, Hernán A.Ronan, MichaelRothberg, Jonathan M.Egholm, MichaelRudan, PavaoBrajkovic, DejanaKucan, ZeljkoGusic, IvanWikström, MårtenLaakkonen, LiisaKelso, JanetSlatkin, MontgomeryPääbo, SvanteA Complete Neandertal Mitochondrial Genome Sequence Determined by High-Throughput SequencingCellCell416-4261343CHEMBIODNAECO_EVOL2008 <![endif]–>(Green et al. 2008)<!–[if supportFields]><![endif]–>, the specimen coming from the Croatian site of Vindija, some 38 kya. The paper’s verdict: “Neandertal mtDNA falls outside the range of variation of modern humans.” So though they don’t explicitly say it, it sounds like the conclusion is that, based on mtDNA, neandertals were a separate species from the humans that inhabit the globe today. Is this the end of the story? Hardly.

First, a technical note. The team used the “high-throughput 454 sequencing technique”, and since I am not a geneticist and could barely understand what the technique involves when I looked into it, all I can gather is that the method creates more sequence copies than traditional PCR (polymerase chain reaction). Perhaps a colleague can enlighten me and other readers on this technique? Also, the team took strong precautions that pretty much ensured that the sample wasn’t (significantly) contaminated with modern human mtDNA. Cool beans, the future today.

Anyway, what’s important is that this complete sequence (from a single individual) allows researchers to do whole mtDNA comparisons of neandertals with modern humans, to try to answer a riddle that is hotly debated in Paleoanthropology—whither Neandertals? Was there admixture between the archaic humans endemic to Eurasia (Neandertals) and the immigrating modern humans coming from Africa?

Now, in general I don’t care that much about neandertals. In my mind, they’re just a form of H. sapiens, albeit probably a homely form. But what I do care about (lately) are patterns of speciation in primates and human origins, so the question of modern-human-neandertal admixture is an interesting one to me. Green and colleagues inferred from the Vindija mtDNA that humans and neandertals were distinct (i.e. probably separate species)—a level of separation I don’t know I can agree with. When the team compared sequence differences between the neandertal and 53 modern humans from around the globe, they found that there are more differences between the neandertal and each human than there are between any pair of humans. This is in contrast to previous studies that looked only at the HRVI and HRVII regions of mtDNA, which found more overlap (less difference) between humans and neandertals. So this underscores the importance of using whole genomes for analysis, rather than a few genes.

Next the team estimated the human-neandertal divergence, assuming a molecular clock with a HomoPan divergence of 6-8 mya. This yielded a divergence date of 660,000 years, with the 95% confidence interval of 800,000 to 520,00 years. I suppose this is not too unreasonable. Some 600 kya is roughly the time when H. heidelbergensis is running around Europe and Africa. Their HomoPan divergence estimate is not so much to my liking, however. They based this estimate on the fossil record, 8 mya being based on the ~7 my-old Sahelanthropus tchadensis cranium and 6 mya based on the ~6 my-old Orrorin tugenensis material. I might have just stuck with the 6 mya divergence, because Sahelanthropus is not convincingly a hominin or “pre-hominin,” really it’s not convincingly anything but an ape. And 5-6 mya is when we start seeing fossils that really look like hominins, be it the Orrorin femora or the dental and mandibular fossils from E. Africa.

Now, as I asked before, is this the end of the story? No. For starters, this paper only looks at mtDNA, which is only maternally inherited. So we could deduce from this paper that perhaps no neandertal females interbred with modern humans. What will be more informative is a look at nuclear DNA—which the team hopes to have sequenced by the end of this year. Moreover, this single neandertal falls outside the range of variation of modern humans. There are several human mtDNA haplotypes—different lineages of mtDNA (again, I’m not a geneticist, so I don’t know how many or how different—a little help, anyone?). From this single individual we cannot get a good picture of neandertal mtDNA variation (haplotypes). Plausibly if we had more samples of mtDNA from archaic humans (are there any from any Upper Paleolithic modern humans?) we may well see the gap between humans and this neandertal bridged. Of course, on the other hand, we might not. So this paper demonstrates considerable difference between human and neandertal mtDNA, but the case is anything but closed.

Also, as paper commentator A. Clark noted, there are many genes in modern human nuclear DNA that appear to be over 1 my old <!–[if supportFields]> ADDIN EN.CITE Clark200840340317Clark, Andrew G.Genome Sequences from Extinct RelativesCellCell388-38913432008 <![endif]–>(Clark 2008)<!–[if supportFields]><![endif]–>, and this may suggest that modern humans and archaic populations (including neandertals) may have interbred at least sporadically. He notes, “The long period of coexistence of modern humans and Neanderthals, as well as the great depth of common ancestry of modern human nuclear genes, make it quite plausible that there was opportunity for interbreeding . . . If there had been admixture, say 100,000 years ago, giving modern humans small segregating pieces of our genome with Neanderthal ancestry, it would be nearly impossible to identify them as such, even with full genome sequences.” When two populations intermingle, their offsprings’ genomes will not necessarily simply be a mix of ½ one parent, ½ the other. Rather, often only adaptive genes are able to ‘sneak’ into the other population’s gene pool—a phenomenon known as introgression. It looks like the human FOXP2 gene may well be an example of introgression, and in fact may have introgressed from an archaic population into modern humans <!–[if supportFields]> ADDIN EN.CITE Coop200826526517Coop, GrahamBullaughey, KevinLuca, FrancescaPrzeworski, MollyThe Timing of Selection at the Human FOXP2 GeneMolecular Biology and EvolutionMol Biol EvolMolecular Biology and EvolutionMol Biol Evol1257-12592572008<![endif]–>(Coop et al. 2008)<!–[if supportFields]><![endif]–>. On an interesting aside, geneticist Chung-I Wu has formulated the “genic species concept,” in which species are formed when they can still interbreed and exchange genetic material, but adaptive regions are not exchanged; obviously this intriguing concept is also controversial <!–[if supportFields]> ADDIN EN.CITE Noor200240540517Noor, Mohamed A. F.Is the biological species concept showing its age?Trends in Ecology & EvolutionTrends in Ecology & Evolution153-154174species conceptsspeciationreproductive isolationadaptationhybridizationbiological species concept2002 <![endif]–>(Noor 2002)<!–[if supportFields]><![endif]–>.

A final point to consider that didn’t come up in Green et al.’s paper is the growing body of evidence that human evolution is accelerating, and has been for the past 40 ky, but especially in the past 10-20 ky <!–[if supportFields]> ADDIN EN.CITE Hawks20071117Hawks, JohnWang, Eric T.Cochran, Gregory M.Harpending, Henry C.Moyzis, Robert K.Recent acceleration of human adaptive evolutionProceedings of the National Academy of SciencesProc Nat Acad SciProceedings of the National Academy of SciencesProc Nat Acad Sci20753-2075810452


Adaptive evolutionhuman evolutionlinkage disequilibriumdemography2007December 26, 2007<![endif]–>(Hawks et al. 2007)<!–[if supportFields]><![endif]–>. This is interesting as the neandertal specimen is 38 ky-old, and other neandertal DNA has come from even older specimens (Krause et al. 2007). I’m not sure at the moment how to interpret this in the context of mtDNA and recent sequencing of neandertal mtDNA. But it should be very important when the team (or someone else) analyzes ancient nuclear DNA, especially given that neandertals (arguably) ‘disappeared’ before human adaptive evolution really began to sprint.

This is an exciting time for anthropological genetics. Techniques are being developed for the extraction and analysis of ancient DNA, which will help shed light on the nature of the emergence of modern humans, and their interactions with archaic populations. At the same time, I am always wary of papers in genetics because of the numbers of assumptions/parameters required by their models.

<!–[if supportFields]> ADDIN EN.REFLIST <![endif]–>Clark AG (2008) Genome Sequences from Extinct Relatives. Cell 134(3):388-389

Coop G, Bullaughey K, Luca F, Przeworski M (2008) The Timing of Selection at the Human FOXP2 Gene. Mol Biol Evol 25(7):1257-1259

Green RE, Malaspinas A-S, Krause J, Briggs AW, Johnson PLF, Uhler C, Meyer M, Good JM, Maricic T, Stenzel U, Prüfer K, Siebauer M, Burbano HA, Ronan M, Rothberg JM, Egholm M, Rudan P, Brajkovic D, Kucan Z, Gusic I, Wikström M, Laakkonen L, Kelso J, Slatkin M, Pääbo S (2008) A Complete Neandertal Mitochondrial Genome Sequence Determined by High-Throughput Sequencing. Cell 134(3):416-426

Hawks J, Wang ET, Cochran GM, Harpending HC, Moyzis RK (2007) Recent acceleration of human adaptive evolution. Proceedings of the National Academy of Sciences 104(52):20753-20758

Krause J, Lalueza-Fox C, Orlando L, Enard W, Green RE, Burbano HA, Hublin J-J, Hänni C, Fortea J, de la Rasilla M, Bertranpetit J, Rosas A, Pääbo S (2007) The derived FOXP2 variant of modern humans was shared with neandertals. Current Biology 17

Noor MAF (2002) Is the biological species concept showing its age? Trends in Ecology & Evolution 17(4):153-154<!–[if supportFields]><![endif]–>

A Tale of Two Lineages

[Hey, I’ll bet I’m the first person to make that allusion…]
In a paper published in JHE today, M. Schillaci posits that human facial anatomy suggests the existence of two human lineages in the late Pleistocene. Schillaci’s analysis reveals that faces of early Australasian crania (40-8 ka) are very similar to those of Levantine specimens Skhul 5, Qafzeh 6 and Qafzeh (100-90 ka). The overall results of the study suggest that the Australasian and Levantine populations share an earlier common ancestor than modern humans, including Upper Paleolithic Europeans. Schillaci interprets this to mean that modern humans first dispersed from Africa around 100 ka, long before the supposed “revolution” of Paleolithic Europe, and made it as far as Australia; second dispersal then occurred some 50 ka later.

The article brings up the issue of Out-of-Africa (Replacement) vs. Multiregional models, but does not clearly come out directly in favor of either one. But by setting up a scenario in which two human lineages are present throughout the Old World in the last 100 ka, the possibility is opened up for these lineages to accrue genetic differences simply by drift or even by selection, then to come into contact again and admix, and for the “archaic” genes to be incorporated into the newer (and modern) genome (introgressive hybridization; cf Evans et al. 2006, Garrigan and Kingan 2007, Hawks and Cochran 2006).

Schillaci does note that the Levantine sample

exhibits a slightly closer genetic relationship to Neandertals (d=0.7318) than to Upper Paleolithic Europeans (d=0.7483). . . . This observed relationship is probably not the result of phenotypic convergence, and likely reflects a slightly more recent common ancestry and/or perhaps hybridization between early modern humans and Neandertals (Trinkaus, 2007). (p. 6)

However, he later notes,

In the present study, the relationship between early modern humans from the Levant and early Australasians (d=0.348) is more than 2.1 times closer than between early modern humans and Neandertals (d=0.7318), and Neandertals do not show a close relationship with early Australasians (d=1.2615). If the observed relationship between Neandertals and early modern humans is the product of hybridization, there is no craniometric evidence indicating that there was substantial introgression of Neandertal alleles into the dispersing modern human population… (p. 7)

Can Schillaci make these claims about “genetic relationship[s]” based on his data? Let’s look at the opening line of the abstract: “This study examines the genetic affinities of various modern human groupings using a multivariate analysis of morphometric data.” A rewording might be: This study examines the craniofacial affinities of various modern human groupings, and to thereby infer genetic relationships. Basically, Schillaci assumes that genetic relationships between populations are accurately reflected in facial anatomy, a bold statement to make. Indeed, he acknowledges the problems with this assumption, but also cites studies (which I haven’t yet read) suggest that craniometric variation in humans throughout the world fits a neutral model of genetic variation. So, Schillaci talks about genetic relationships throughout the paper, but these aren’t based on actual genetic data, but rather inferences from craniometric data–quite confusing. His aforementioned lack of evidence for introgression between neandertals and early modern humans does not preclude real genetic evidence for introgression (cf. what I cf-ed above.)

What I care about: the study allows for, and possibly corroborates, a Multiregional model of human evolution (of course, what I really care about is the possibility of such a model prior to, and in the early stages of, the genus Homo). Hey, I guess old-school craniometrics hasn’t outlived its usefulness in physical anthropology.

Evans PD et al. 2006. Evidence that the adaptive allele of the brain size gene microcephalin introgressed into Homo sapiens from an archaic Homo lineage. Proc Nat Acad Sci 103(48): 18178-18183.

Garrigan D and SB Kingan. 2007. Archaic Human Admixture. Curr Anthropol 48(6): 895-902.

Hawks J and GM Cochran. 2006. Dynamics of adaptive introgression from archaic to modern humans. Paleoanthropol 4: 101-115.

Schillaci M, in press. Human cranial diversity and evidence for an ancient lineage of modern humans. J Hum Evol xx: 1-13.