Understanding Evolution

The deep roots of diabetes
February 2014

bust of Neanderthal
A hyper-realistic bust of Homo neanderthalensis by artist John Gurche

The modern diabetes epidemic is caused, not by a virulent pathogen, but by the spread of an even stealthier invader: the Western lifestyle. As people around the world have begun to eat less healthily, lead more sedentary lives, and live to older ages, adult onset diabetes (type 2 diabetes) has become common in places where the disease was previously unknown. Between 1985 and 2002, the number of people with diabetes grew from 30 million to 217 million, and this figure is expected to exceed 366 million by 2030. But the epidemic has not been even-handed. Even accounting for differences in lifestyle, some populations have been hit particularly hard. Mexicans and Latin Americans, for example, have nearly twice the chance of developing diabetes that non-Hispanic white Americans do. New research addresses these disparities. Last month, scientists announced that they'd discovered a gene that helps explain the difference in diabetes risk among many populations. In a strange twist, the gene version in question traces its ancestry back to Neanderthals! What exactly is going on here?

Where's the evolution?

To understand the evolutionary back story, you first need to know a little about the gene itself. The gene in question encodes a protein that helps move certain lipids into liver cells. The diabetes-contributing version of this gene differs from the standard gene version by five mutations—and these seem to alter the function of the protein enough to increase diabetes risk. Carriers of the mutated version of the gene are more likely to get diabetes at a younger age and with a lower degree of obesity than non-carriers.

Anyone can carry this gene—but the new research found that it is more common in some populations than others. Among people with many Native American ancestors, the likelihood of carrying at least one copy of the mutated gene is greater than 50%. Among East Asians, the frequency is about 10%. Among people with mainly European ancestors, the gene version is extremely rare, and it seems to be not present at all in Africans. Because people from Mexico and Latin America are much more likely to have Native American ancestry, they are also much more likely to carry this gene version, and hence, have higher odds of developing diabetes.

So diabetes risk in modern populations is tied to that population's evolutionary history. That should come as no surprise. Recent research has uncovered hundreds of gene versions contributing to diseases that range from asthma to Alzheimer's disease. The surprise in this research lies in the provenance of the disease-contributing gene.

Something about the new gene version struck the researchers as surprising: it had evolved too much. In big, slow-to-reproduce organisms like humans, mutations take a while to accumulate and evolution proceeds fairly slowly. Based on human DNA's usual rates of evolution, this diabetes gene version must have started diverging from the standard version almost 800,000 years ago. That's before our modern human anatomy had evolved and long before we had left Africa. Now, there's nothing surprising about a really old gene—but if this gene version first evolved in Africa in the ancestral lineage of all humans, then why don't all human populations, in particular Africans, carry it?

The researchers hypothesized that perhaps the diabetes-contributing linked gene version didn't actually evolve in our direct ancestral lineage, but in Neanderthals, as shown in the diagram below. In this scenario, the gene version would have acquired many of its mutations in the Neanderthal lineage some time after the human and Neanderthal lineages split from one another. When modern humans eventually left Africa between 60,000 and 80,000 years ago and arrived in Europe and the Middle East, Neanderthals were already living there. Those humans and Neanderthals interbred, introducing some Neanderthal DNA (including the diabetes-linked gene version) into the human lineage—but not into all humans. Human lineages that had remained in sub-Saharan Africa never encountered Neanderthals and so did not wind up carrying any Neanderthal DNA. This scenario would help explain how the diabetes-contributing gene could be so old and not be found in Africans.

Neanderthal/Human Diabetes Tree

Through recent advances in recovering DNA from ancient bones, the genomes of several Neanderthal individuals have been reconstructed. The researchers searched through the DNA sequences of these samples and found what they were looking for. One of the Neanderthals (a newer fossil discovery from Denisova Cave) carried the diabetes-linked sequence! It seems that this gene version, now common among people of Native American ancestry, is a relic from the period of our history when humans walked the earth alongside other hominids.

This discovery does not mean that people of Native American descent (or for that matter anyone who carries the diabetes gene version) are particularly closely related to Neanderthals. Human populations from all over the world seem to have similar degrees of Neanderthal ancestry (between 1 and 4%); we all just carry different subsets of Neanderthal-derived genes—that is, unless your ancestors are from sub-Saharan Africa, where many people have no Neanderthal ancestry at all.

Neither does this discovery mean that Neanderthals had diabetes. Type 2 diabetes is a disease of the modern world, borne of a mismatch between modern, unhealthy lifestyles and a metabolism that, for the vast majority of our evolutionary history, existed in an environment where food was relatively scarce and lots of physical activity was necessary to survive. In that harsh environment, even individuals carrying genes that contribute to diabetes when food is plentiful and sedentary lifestyles are common are unlikely to develop diabetes. This helps explain how such genes can be common today. At no point in our evolutionary history have they been exposed to the rigors of natural selection. Only recently have such genes become detrimental to human health!

This discovery highlights the importance of evolutionary history in understanding and improving human health. Even the deepest roots of our past, which lead back to Africa and to our common ancestors with other, now-extinct hominid species, may become relevant at your next doctor's appointment. And we are just beginning to understand these ramifications. Advances in DNA technology have only recently allowed us to study the intersections between ancient DNA, large-scale genomic data, and modern epidemiology. So stay tuned to learn more about the results of these exciting investigations!

Primary literature

  • Green, R. E., Krause, J., Briggs, A. W., Maricic, T., Stenzel, U., Kircher, M., ... Pääbo, S. (2010). A draft sequence of the Neandertal genome. Science. 328: 710-722.
    read it
  • The SIGMA Type 2 Diabetes Consortium. (2013). Sequence variants in SLC16A11 are a common risk factor for type 2 diabetes in Mexico. Nature. doi:10.1038/nature12828.
    read it

Discussion and extension questions

  1. In your own words, explain why this diabetes-linked gene version is not found among Africans.
  2. If the diabetes-contributing gene version had arisen in our direct ancestral lineage 800,000 years ago, what would you expect to observe in terms of the distribution of the gene version across different populations? Explain your reasoning.
  3. In your own words, explain why the diabetes-contributing gene version was not weeded out of human populations by natural selection long ago.
  4. Do some research, and find an example of a human disease with a genetic component that has different frequencies in different populations. Describe the disease and the gene version that contributes to it. Describe the population in which the gene version is common.
  5. Advanced: Do some research on type 2 diabetes, and review the concept of evolutionary fitness. Do you think that this disease decreases a person's evolutionary fitness? Explain your reasoning.
  6. Advanced: Based on your answer to the item above, do you think that natural selection is acting against the diabetes-contributing gene version in modern populations? Explain why or why not.


  • Green, R. E., Krause, J., Briggs, A. W., Maricic, T., Stenzel, U., Kircher, M., ... Pääbo, S. (2010). A draft sequence of the Neandertal genome. Science. 328: 710-722.
  • Smyth, S., and Heron, A. (2006). Diabetes and obesity: the twin epidemics. Nature Medicine. 12: 75-80.
  • The SIGMA Type 2 Diabetes Consortium. (2013). Sequence variants in SLC16A11 are a common risk factor for type 2 diabetes in Mexico. Nature. doi:10.1038/nature12828


View this article online at:

Image of Neanderthal bust courtesy of Smithsonian Institute: http://smithsonianscience.org/2010/03/hall-of-human-origins/

Understanding Evolution © 2015 by The University of California Museum of Paleontology, Berkeley, and the Regents of the University of California