Gene killer may prevent diabetes in even those at most risk
The New York Times
A new study based on genetic testing of 150,000 people has found a rare mutation that protects even overweight people from getting Type 2 diabetes. The effect is so pronounced — the mutation reduces risk by two-thirds — that it provides a promising new target for developing a drug to mimic the mutation’s effect.
The mutation destroys a gene used by pancreas cells where insulin is made. Those with the mutation seem to make slightly more insulin and have slightly lower blood glucose levels for their entire lives.
Already Pfizer, which helped finance the study, and Amgen, which owns a company whose data played a key role in the research, are starting programs aimed at developing drugs that act like the mutation, the companies said.
But Timothy Rolph, a Pfizer vice president, cautioned it can take 10 to 20 years to get a drug to market after discovering something new about human genetics and disease.
The study, published Sunday in Nature Genetics, involved a mutation so rare that finding it was only recently possible, with massive data from large numbers of people, researchers said.
“The study is a tour de force and the authors are the top people in the field,” said Dr. Samuel Klein, director of the center for human nutrition at Washington University School of Medicine, who was not involved in the study.
This is the first time in diabetes research that a mutation that destroys a gene has proved beneficial, noted Louis Philipson, director of the Kovler Diabetes Center at the University of Chicago. For drug development, he said, “that is very powerful.”
For scientists, the result was a surprise because the same mutation that protects people from diabetes, by destroying one copy of the gene, known as ZnT8, has the opposite effect in some strains of mice. Destroying that gene actually causes diabetes in the animals.
The work began four years ago when a group of geneticists from academic institutions and Pfizer decided to search for gene mutations that protect against diabetes. Usually researchers look for mutations that increase — rather than decrease — the risk of diseases, with the aim of determining who gets a disease, and why.
The group started with populations in Finland and Sweden, where 28,000 people had been studied for years. The data included their ages, weights and diseases, including diabetes.
They compared people at either end of the spectrum of diabetes risk. One group of 352 people had Type 2 diabetes even though their risk seemed low. Their average age was about 50, they were lean and they did not smoke. The other group of 406 people was just the opposite. Their average age was about 80, and, Rolph said, “they had all the bad habits — they were overweight, they drank, they smoked.” And yet these people did not have diabetes.
Two of the overweight older people who were free of diabetes turned out to have a mutation that destroyed one copy of the ZnT8 gene. It was intriguing but hard to know if the association was meaningful with only two people.
So the researchers expanded their work, studying the genes of 18,000 people in Sweden, fat and thin, old and young, with diabetes and without. They found another 31 people who seemed protected from diabetes and had mutations that destroyed the ZnT8 gene.
Then Dr. David Altshuler, deputy director of the Broad Institute of Harvard and MIT and the study’s lead author, met with Dr. Kari Stefanson, chief executive of deCODE Genetics, a company with data on genes and diseases for the entire population of Iceland. The American drug company, Amgen, bought deCODE and its valuable genetic database.
Stefanson searched deCODE’s database and quickly found 39 people out of 5,440 who had a mutation that destroyed the gene and who did not have diabetes. In contrast, just nine out of 3,727 diabetes patients had the mutation.
“It took us five minutes,” Stefanson said. “It was a lovely little afternoon in our conference room.”
At that point, Altshuler said, the group wrote a paper and submitted it to a medical journal. It was rejected, he said, after one of the reviewers said it must be wrong because it contradicted what was known from studies with mice.
The group went back for more data. They mapped the genes of 13,000 more people and once again found mutations destroying the same gene and associated with a markedly reduced risk of Type 2 diabetes.
This time their paper was accepted for publication by Nature Genetics, Altshuler said.
Now the researchers are asking whether the mutation has any bad health effects. So far, Stefanson said, none has been found. With his data he has established that people with the mutation are no more likely to get 750 diseases he searched for.