Did Mexicans Inherit Diabetes Risk from Neanderthals?
These days Neanderthals seem to pop up where you least expect them. When Medscape asked me a few days ago to write up a paper being published in the December 25 online Nature, the title sounded run of the mill: “Sequence variants in SLC16A11 are a common risk factor for type 2 diabetes in Mexico.”
Yawn. Another genome-wide association study (GWAS), showing stretches of single sites in the genome (SNPs) that track with type 2 diabetes in Mexicans and other Latin Americans, who have about twice the prevalence as European whites. I might have been more alert had I read the ending first, like I sometimes do with novels – a “Neanderthal analysis team” that included Svante Paabo. The Neanderthal connection is also not in the headline for the news release accompanying publication of the paper, but is buried a few paragraphs down. So I wondered, on Christmas day, would the media notice the missing link? If they don’t, here’s DNA Science blog’s take.
The study is terrific, with nice numbers — 9.2 million SNPs analyzed for 3,848 Mexicans and other Latin Americans who have type 2 diabetes and 4,366 who don’t. The researchers are a stellar team from Mexico, Boston, LA and others part of the The Slim Initiative in Genomic Medicine for the Americas (SIGMA) Type 2 Diabetes Consortium.
The findings zeroed in on five linked SNPs — a haplotype — in a gene called SLC16A11 on the short arm of chromosome 17. Four of the five mutations change an amino acid in the encoded protein, and the fifth is silent. The protein normally ferries certain lipids into liver cells, a complex function that makes sense in terms of past studies of insulin resistance. The association between the haplotype and disease risk is strong, and holds up in other populations. Perhaps it could be developed into a tool to predict elevated diabetes risk, or present a new drug target in lipid metabolism. So far, so good.
But it was the DNA sequence that grabbed my full attention, and the clues from geographic prevalence.
The five-site haplotype is in 50% of Native Americans, in about 10% of East Asians, much rarer in Europeans, and absent among Africans. And it’s ancient. Researchers determine the degree to which a mutant gene differs from the most common sequence (wild type), then impose a time scale in the form of known mutation rates. The SLC16A11 five-site haplotype is so divergent that it goes back to nearly 800,000 years ago — before our ancestors expanded out of Africa.
The most plausible explanation, unexpected I suspect, seemed to be that the haplotype came from an archaic human – a Neanderthal or Denisovan or their as-yet unnamed contemporaries. And the haplotype indeed shows up in the skeleton of a Neanderthal found in the Denisovan cave in Siberia — that’s the now-famous place where the genomes that led to us sorted themselves out. It is a peek into a sometimes promiscuous past.
Having a bit of one’s genome from a Neanderthal that predisposes to diabetes might be unexpected, but the presence of this DNA source shouldn’t be stigmatizing – from 1 to 4 percent of many of our genomes are from Neanderthals (sub-Saharan Africans have none). But for a disease population so scrutinized — note the Pima Indian investigations — it’s surprising to find that a risk gene for diabetes traces back to this side branch from the evolutionary road leading to humanity.
I was going to take a blogging break until this paper appeared — next week I’ll return to the genetic stories of the Amish.