The gene variant was detected in a large genome-wide association study (GWAS) of more than 8,000 Mexicans and other Latin Americans.
People who carry the higher risk version of the gene are 25 per cent more likely to have diabetes than those who do not, and people who inherited copies from both parents are 50 per cent more likely to have diabetes, researchers found.
The higher risk form of the gene - named SLC16A11 - has been found in up to half of people with recent Native American ancestry, including Latin Americans, 'BBC News' reported.
The variant is found in about 20 per cent of East Asians and is rare in populations from Europe and Africa.
The elevated frequency of this risk gene in Latin Americans could account for as much as 20 per cent of the populations' increased prevalence of type 2 diabetes - the origins of which are not well understood.
"To date, genetic studies have largely used samples from people of European or Asian ancestry, which makes it possible to miss culprit genes that are altered at different frequencies in other populations," said co-corresponding author Jose Florez, a Broad associate member, an associate professor of medicine at Harvard Medical School and an Assistant Physician in the Diabetes Unit and the Center for Human Genetic Research at the Massachusetts General Hospital.
"By expanding our search to include samples from Mexico and Latin America, we've found one of the strongest genetic risk factors discovered to date, which could illuminate new pathways to target with drugs and a deeper understanding of the disease," Florez said.
In the study published in the journal Nature, researchers conducted genomic analyses, in collaboration with Svante Paabo of the Max Planck Institute for Evolutionary Anthropology, and discovered that the SLC16A11 sequence associated with risk of type 2 diabetes is found in a newly sequenced Neanderthal genome.
Analyses indicated that the higher risk version of SLC16A11 was introduced into modern humans through mixing with Neanderthal.
The researchers went on to show that altering the levels of the SLC16A11 protein can change the amount of a type of fat that has previously been implicated in the risk of diabetes.
These findings have led the team to hypothesise that SLC16A11 may be involved in the transport of an unknown metabolite that affects fat levels in cells and thereby increases risk of type 2 diabetes.