After the 2010 publication of the Neanderthal draft genome sequence, evolutionary biologist Joshua Akey, then at the University of Washington in Seattle, and his graduate student Benjamin Vernot began looking into its most provocative implication: that the ancient hominins had bred with the ancestors of modern humans. Neanderthals had been living in Eurasia for more than 300 millennia when some human ancestors left Africa some 60,000–70,000 years ago, and according to the 2010 publication, in which researchers compared the Neanderthal draft genome with modern human sequences, about 2 percent of the DNA in the genomes of modern-day people with non-African ancestry is Neanderthal in origin.1
To investigate the archaic ancestry of the living human population, Akey and Vernot set to work searching for Neanderthal DNA in modern genomes. They developed a statistical approach to identify genetic signatures suggestive of Neanderthal ancestry in the genomes of 379 European and 286 East Asian individuals. The endeavor was further powered by the first high-quality Neanderthal genome sequence, which gave the duo confidence that the sequences they’d identified were indeed of archaic origin. Still, in the back of Akey’s mind, he had doubts about the research. “I remember telling Ben [when] we were working on this, ‘I wake up every day in a cold sweat that this is all just incomplete lineage sorting’”—a methodological artifact that would undermine their conclusions about Neanderthal ancestry, meaning the sequences were the result of the common ancestry the two groups shared.
Then, as Vernot and Akey were getting ready to submit their work for publication, their department got a visit from Svante Pääbo, a geneticist at the Max Planck Institute for Evolutionary Anthropology who had pioneered techniques for extracting and analyzing DNA from ancient specimens and had led the early Neanderthal genome efforts. They spoke with him about their ongoing project, and Pääbo noted that his collaborator, David Reich at Harvard Medical School, was pursuing a very similar line of research. So Akey gave Reich a call.
“The end result [of the conversation] was we agreed to coordinate publication,” Akey recalls. “We also agreed not to even look at each other’s papers because we didn’t want to influence the results in any way.”
See “Simultaneous Release”
Was it just this curious feature of human history that didn’t have an impact, or did it alter the trajectory of human evolution?—Joshua Akey, Princeton University
Vernot and Akey submitted to Science;2 Reich and his colleagues submitted to Nature.3 The two journals synchronized publication of the papers at the end of January 2014. The day they went live, Akey anxiously began to read the paper from the Reich group. “I remember sitting in my office, reading it, and really sort of just going through the checklist” of the key results, he says. Quickly, the relief set in. “We essentially said the exact same thing,” Akey recalls. “Usually when you publish something, it’s years before you see validation. . . . This was sort of instant gratification.”
The two groups had used different statistical approaches to identify Neanderthal DNA in modern human genomes, putting to bed any skepticism about the history of hominin group interbreeding. “[It was] the final nail on the coffin that it couldn’t be anything else,” says Janet Kelso, a computational biologist at the Max Planck Institute for Evolutionary Anthropology and a collaborator on Reich’s publication.
With the issue of Neanderthal/modern human mating settled, scientists could focus on a new goal, says Akey, now at Princeton University. Namely, what was the consequence of this interbreeding? “Was it just this curious feature of human history that didn’t have an impact, or did it alter the trajectory of human evolution?”
In the past five years, a flurry of research has sought to answer that question. Genomic analyses have associated Neanderthal variants with differences in the expression levels of diverse genes and of phenotypes ranging from skin and hair color to immune function and neuropsychiatric disease. But researchers cannot yet say how these archaic sequences affect people today, much less the humans who acquired them some 50,000–55,000 years ago.
“So far I have not seen any convincing functional studies where you take the Neanderthal variant and the human variant and do controlled experiments” to identify the physiological consequence, says Grayson Camp, a genomicist at the Institute of Molecular and Clinical Ophthalmology Basel (IOB) in Switzerland. “No one has actually shown yet in culture that a human and Neanderthal allele have a different physiological function. That will be exciting when someone does.”
A Mixed History
Some 350,000 or more years ago, the group of hominins that would evolve to become Neanderthals and Denisovans left Africa for Eurasia.
A few hundred millennia later, about 60,000 to 70,000 years ago, the ancestors of modern non-Africans followed a similar path out of Africa and began interbreeding with these other hominin groups. Researchers estimate that much of the Neanderthal DNA in modern human genomes came from interbreeding events that took place around 50,000 to 55,000 years ago in the Middle East. Thousands of years later, humans moving into East Asia interbred with Denisovans.
the scientist staff
Neanderthal in our skin
Most Neanderthal variants exist in only around 2 percent of modern people of non-African descent. But some archaic DNA is much more common, an indication that it was beneficial to ancient humans as they moved from Africa into Eurasia, which Neanderthals had called home for more than 300,000 years. In their 2014 study, Vernot and Akey found several sequences of Neanderthal origin that were present in more than half of the genomes from living humans they studied. The regions that contained high frequencies of Neanderthal sequences included genes that could yield clues to their functional effect. Base-pair differences between Neanderthal and human variants rarely fall in protein-coding sequences, but rather in regulatory ones, suggesting the archaic sequences affect gene expression. (See “Denisovans in the Mix” below.)
A number of segments harbor genes that relate to skin biology, such as a transcription factor that regulates the development of epidermal cells called keratinocytes. These variants may underlie traits that were adaptive in the different climatic conditions and lower levels of ultraviolet light exposure at more northern latitudes. Reich’s group similarly found genes involved in skin biology enriched in Neanderthal ancestry—that is, more than just a few percent of people carried Neanderthal DNA in these parts of the genome.
No one has actually shown yet in culture that a human and Neanderthal allele have a different physiological function. That will be exciting when someone does.—Grayson Camp,
Institute of Molecular and Clinical Ophthalmology Basel
It was unclear, however, what specific effect the Neanderthal variants had on phenotype. For that, researchers needed phenotypic data on many different kinds of traits, paired with genetic information, for thousands of people. Vanderbilt University evolutionary geneticist Tony Capra has access to such a resource: the Electronic Medical Records and Genomics (eMERGE) Network. Right around the time the scientific community was beginning to map Neanderthal DNA in the genomes of living people, eMERGE organizers were compiling electronic health records and associated genetic data for tens of thousands of patients from nine health-care centers across the US. “We felt like we had a chance to evaluate some of those hypotheses [about functionality] on a larger scale in a real human population where we had rich phenotype data,” says Capra.
In collaboration with Akey and Vernot, who helped identify Neanderthal variants in the genetic data included in the database, Capra’s group looked for links between the archaic DNA and more than 1,000 phenotypes across some 28,000 people of European ancestry. They reported in 2016 that Neanderthal DNA at various sites in the genome influences a range of immune and autoimmune traits, and there was some association with obesity and malnutrition, pointing to potential metabolic effects. The researchers also saw an association between Neanderthal ancestry and two types of noncancerous skin growths associated with dysfunctional keratinocyte biology—supporting the idea that the Neanderthal DNA was at one point selected for its effects on skin.4
“This was crazy to me,” says Capra. “What these other groups had predicted based on just the pattern of occurrence—the presence and absence of Neanderthal ancestry around certain types of genes—we were actually seeing in a real human population, that having Neanderthal ancestry influenced traits related to those types of skin cells.” What remains unclear, however, is what the benefits of the Neanderthal sequences were for those early humans.
At the same time, Kelso and her postdoc Michael Dannemann were taking a similar approach with a relatively new database called the UK Biobank (UKB), which includes data from around half a million British volunteers who filled out questionnaires about themselves, underwent medical exams, and gave blood samples for genotyping. Formally launched in 2006, the UKB published its 500,000-person-strong resource in 2015, and Kelso and Dannemann decided to see what information they could extract. Conveniently, the genotyping data specifically includes SNPs that can identify variants of Neanderthal origin, thanks to Reich’s group, which provided UKB architects with a list of 6,000 Neanderthal variants.
Among the many links Kelso and Dannemann identified as they dug into data from more than 112,000 individuals in the UKB was, once again, an association between certain Neander-thal variants and aspects of skin biology.5 Specifically, the archaic sequences spanning the BNC2 gene—a stretch of the genome that Vernot and Akey had identified as having Neanderthal origin in some 70 percent of non-Africans—were very clearly associated with skin color. People who carried Neanderthal DNA there tended to have pale skin that burned instead of tanned, Kelso says. And the stretch that included BNC2 was just one of many, she adds: around 50 percent of Neanderthal variants linked with phenotype in her study have something to do with skin or hair color.
The effect that Neanderthal DNA might have on skin appearance and function is “fascinating,” says Akey. “Something that we’re still really interested in and starting to do some experimental work on is: Can we understand what these genes do and then maybe what the selective pressure was that favored the Neanderthal version?”
Denisovans in the mix
BENCE VOILA, MAX PLANCK INSTITUTE FOR EVOLUTIONARY ANTHROPOLOGY
Neanderthals thrived in Eurasia as a dominant hominin group for hundreds of thousands of years and have long been a focus of scientific inquiry. But less than a decade ago, researchers discovered that there was another group of archaic hominins that coexisted with Neanderthals and the ancestors of modern humans. DNA collected from a single finger bone and two teeth appeared to be neither Neanderthal nor human, and scientists named a new group, the Denisovans, after the Siberian cave in which the remains were found in 2008.
Once researchers reconstructed the entire high-quality Denisovan genome in 2012 (Science, 338:222–26, 2012), it became clear that, like Neanderthals, Denisovans had interbred with modern humans during the time that they coinhabited Eurasia, with analyses suggesting that the introgressed DNA likely came from multiple Denisovan populations within the last 50,000 years, sometime after mixing occurred between Neanderthals and human ancestors (Cell, 173:P53–61.E9, 2018; Cell, 177:P1010–21.E32, 2019). Denisovan DNA makes up 4–6 percent of the genomes of people native to the islands of Melanesia, a subregion of Oceania, and to a lesser extent they left their genetic mark in other Pacific island populations and some modern East Asians, while it is largely absent from the genetic code of most other people. As with Neanderthal introgression, the question that remains to be answered is: What effect did these variants have on our own lineage—and are we still experiencing Denisovans’ genetic influence?
As with Neanderthal DNA, experts have identified regions of modern human genomes that are significantly depleted of Denisovan DNA, and they saw that these “deserts” were the same ones that lacked Neanderthal sequences—indications of selection against deleterious variants (Science, 352:235–39, 2016). “That’s as close as you can get to sort of a replication in this type of work,” says Princeton University evolutionary biologist Joshua Akey. In terms of introgressed bits of Denisovan DNA that might have been beneficial to modern humans, researchers have found links to toll-like receptors and other contributors to immune function, similar to links found with Neanderthal variants.
Denisovan DNA may have also offered some unique benefits to ancient humans. One scientific team identified Denisovan variants in the genomes of Greenland Inuits that include genes involved in the development and distribution of adipose tissue, perhaps pointing to advantages in cold tolerance and metabolism (Mol Biol Evol, 34:509–24, 2017). And maybe the strongest suggestion of beneficial Denisovan introgression comes from a 2014 study in which researchers linked the archaic sequences with high altitude adaptation among populations that live in the Tibetan highlands (Nature, 512:194–97, 2014). The particular variant they focused on was so highly selected, notes Kelso, that “almost everyone living on the plateau carries this piece of Denisovan DNA.”
Another area of human biology tightly linked to Neanderthal variants in the genome is the immune system. Given that human ancestors were exposed to a menagerie of different pathogens—some of which came directly from the Neanderthals—as they migrated through Eurasia, the Neanderthal sequences introgressed into the human genome may have helped defend against these threats, to which Neanderthals had long been exposed.
“Viral challenges, bacterial challenges are among the strongest selective forces out there,” says Kelso. Unlike changes in other environmental conditions such as daylight patterns and UV exposure, “pathogens can kill you in one generation.”
Hints of archaic DNA’s role in immune function surfaced as early as 2011, as soon as the Neanderthal genome was available for cross-referencing with sequences from modern humans. A team led by researchers at Stanford University found that certain human leukocyte antigen (HLA) alleles, key players in pathogen recognition, held signs of archaic ancestry—from Neanderthals, but also from another hominin cousin, the Denisovans.6 “It’s a cool paper and one that contributed to a lot of people thinking about the effects of introgression,” says Capra.
Several other studies since then have strengthened the link between archaic DNA and immune function, branching out from the HLA system to include numerous other pathways.7 For example, multiple labs have tied Neanderthal variants to altered expression levels of genes encoding toll-like receptors (TLRs), key players in innate immune responses. In 2016, Kelso, Dannemann, and a colleague found that pathogen response and susceptibility to develop allergies were associated with Neanderthal sequences that affect TLR production.8
Viruses, in particular, appear to be potent drivers of adaptation. Last year, University of Arizona population geneticist David Enard and colleagues found that one-third of Neanderthal variants under positive selection were linked to genes encoding proteins that interact with viruses.9
Viral challenges, bacterial challenges are among the strongest selective forces out there. Pathogens can kill you in one generation.—Janet Kelso, Max Planck Institute for Evolutionary Anthropology
Researchers have also identified several less-easily explainable phenotypic associations with Neanderthal introgression. In their 2017 analysis, for example, Kelso and Dannemann found that Neanderthal variants were associated with chronotype—whether people identify as early birds or night owls—as well as links with susceptibility to feelings of loneliness or isolation and low enthusiasm or interest. The associations with mood-related phenotypes jibe with what Capra’s group had found the year before in its dataset of medical information, which linked Neanderthal variants to risks for depression and addiction. “These were associations that were quite strong,” says Capra. “And when we looked at genes where these regions of Neanderthal ancestry fell, in many cases they made sense given what we already know about those genes.”
Why these associations exist is still a mystery. Kelso suspects that light might be a unifying factor, with both changes in day-length patterns and UV exposure reductions as they moved to more-northern latitudes. But that’s just a hunch, she emphasizes.
“It’s fun speculating about how [Neanderthal introgression] could have been advantageous, or how variants that make us depressed in the modern environment could have been beneficial,” says Capra. “I don’t really even know what depression meant 40,000 years ago. That’s both the challenge and the fun, provocative part about all this.”
A question of functionality
Even with more straightforward associations, such as with skin traits or immune responses, conclusions thus far are drawn from correlations between genotypes and phenotypes. While such genetic and statistical approaches can conceptually link Neander-thal introgression with phenotypes and hint at why such sequences may have been selected for in humans’ early history, researchers have not yet published in vitro validation studies.
“Studying Neanderthal DNA more closely on a molecular level in the lab is pretty tricky,” says Dannemann. Neanderthal variants tend to come in packages, and the linkage between the variants makes it difficult to identify the function of each one, he explains.
That challenge hasn’t stopped researchers from trying. As a postdoc in Pääbo’s lab in Germany, Camp, along with Vernot, Kelso, and Dannemann, established a handful of brain organoids from induced pluripotent stem cell lines of modern Europeans who vary in their Neanderthal-derived genetics, and tracked single-cell transcriptomes as the cultured cells matured. The early data suggest that the Neanderthal variants affect gene expression in the same way as documented by previous work, validating the model.
But such research is still in the proof-of-principle stage, says Camp, who is continuing this work in his own lab in Switzerland. “Now you just need to increase throughput. You need to do this for 100 or 200 individuals.” Even then, he adds, the conclusions researchers will be able to draw will be limited. “I am a bit cautious and maybe pessimistic [about whether] you can really identify . . . impacts [of Neanderthal variants] on some physiological outcomes.”
There are other fundamental questions that are proving difficult to answer about Neanderthal introgression, says Akey, from the number of hybridization events to the timescale over which those events took place, and whether there was sex bias in patterns of gene flow. “There are all these important things that are really hard to estimate,” he says. “I think the field is kind of stuck right now.” But he’s hopeful that as more genomes from various archaic hominin groups and from modern humans come online, researchers’ power to model how all of these groups interbred will strengthen. A second high-quality Neanderthal genome was published in 2017 (Science, 358:655–58), and researchers now have the genome of a 40,000-year-old human who had a Neanderthal ancestor just a few generations back. Last year, researchers published the sequence of a first-generation hybrid of Denisovans and Neanderthals.
Those data will likely yield some surprises. Capra has found evidence, for example, that some of the Neanderthal segments that correlated with modern phenotypes may not affect those pheno-types directly. His work has uncovered cases in which the correlation was driven by sequences close enough in the genome to Neanderthal variants that the two always appear together. These sequences were carried by the common ancestor of Neanderthals and modern humans but were missing from the group of humans who founded the modern Eurasian population. These variants, which had been retained by Neanderthals, were then reintroduced to the ancestors of modern non-Africans during periods of interbreeding.10 “These genetic variants existed in modern [Eurasians only] in the Neanderthal context, but these were not of Neanderthal ancestry,” Capra says.
Akey has come upon another interesting twist: Africans do have Neanderthal ancestry. Unpublished work from his group points to the possibility that some of the ancient modern humans that bred with Neanderthals migrated back to Africa, where they mixed with the modern humans there, sharing bits of Neanderthal DNA. If true, that would mean that Africa is not devoid of Neanderthals’ genetic influence, Akey notes. “There’s Neanderthal basically all over the world.”
All About Regulation
Am J Hum Genet, doi:10.1016/j.ajhg.2019.04.016, 2019; the scientist staff
In their seminal 2014 studies, the groups of David Reich of Harvard Medical School and Joshua Akey, then at the University of Washington, noted that the Neanderthal variants that correlated with human phenotypes did not appear in coding regions. Two years later, a genome-wide analysis published by investigators in France found that Neanderthal ancestry was enriched in areas tied to gene regulation (Cell, 167:643–56.e17, 2016). The implication was that sequences that originated in Neanderthals tend to have “less impact through protein and more impact through gene expression,” says coauthor Maxime Rotival, a geneticist at the Pasteur Institute in Paris.
To ask this question more directly, Akey turned to the Genotype-Tissue Expression (GTEx) Project, which has cataloged gene expression data from roughly 50 tissues for each of 10,000 individuals. “It’s this really fine-scale record of gene expression,” says Akey. His then-postdoc Rajiv McCoy, now an assistant professor at Johns Hopkins University, developed a method to assess messenger RNA levels based on which allele was being expressed—the one from an individual’s father or mother—and the researchers applied this approach to people in the GTEx database who were heterozygous for a particular Neanderthal variant. Comparing expression levels based on which allele was being expressed, the researchers found that a quarter of the stretches of Neanderthal DNA in human genomes affect the regulation of the genes in or near those stretches (Cell, 168:P916–27.E12, 2017).
“We’ve known for a long time that gene expression variation is an important source of phenotypic variation within populations and phenotypic divergence between species,” says Akey. “We were interested in asking whether Neanderthal sequences make any contribution to gene expression variability.” The answer was a resounding yes.
Earlier this year, Rotival and two colleagues calculated ratios of Neanderthal to non-Neanderthal variants across the genome and compared those ratios for protein-coding
“This is not at all a surprise,” says Vanderbilt University’s Tony Capra, whose lab has generated similar findings in people of Eurasian descent, “but it’s really nice to see it quantified very comprehensively.”
- R.E. Green et al., “A draft sequence of the Neandertal genome,” Science, 328:710–22, 2010.
- B. Vernot, J. Akey, “Resurrecting surviving Neandertal lineages from modern human genomes,” Science, 343:1017–21, 2014.
- S. Sankararaman et al., “The genomic landscape of Neanderthal ancestry in present-day humans,” Nature, 507:354–57, 2014.
- C.N. Simonti et al., “The phenotypic legacy of admixture between modern humans and Neandertals,” Science, 351:737–41, 2016.
- M. Dannemann, J. Kelso, “The contribution of Neanderthals to phenotypic variation in modern humans,” Am J Hum Genet, 101:P578–89, 2017.
- L. Abi-Rached et al., “The shaping of modern human immune systems by multiregional admixture with archaic humans,” Science, 334:89–94, 2011.
- H. Quach et al., “Genetic adaptation and Neandertal admixture shaped the immune system of human populations,” Cell, 167:643–56.e17, 2016.
- M. Dannemann et al., “Introgression of Neandertal- and Denisovan-like haplotypes contributes to adaptive variation in human toll-like receptors,” Am J Hum Genet, 98:P22–33, 2016.
- D. Enard and D.A. Petrov, “Evidence that RNA viruses drove adaptive introgression between Neanderthals and modern humans,” Cell, 175:P360–71.E13, 2018.
- D.C. Rinker et al., “Neanderthal introgression reintroduced functional alleles lost in the human out of Africa bottleneck,” bioRxiv, doi:10.1101/533257, 2019.
Jef Akst is the managing editor of The Scientist. Email her at firstname.lastname@example.org.