Type 2 diabetes is a common disease characterized by high blood glucose levels and unresponsiveness to insulin. Individuals with type 2 diabetes may produce insulin at normal levels but do not respond to it sufficiently, either because the insulin receptors in their cells have become less sensitive or because the insulin produced in the pancreas isn’t being delivered to other cells. One reason that insulin might not make it where it’s supposed to go is that insulin secretion from pancreatic β cells is impaired.
In an article published in Science last week, Anders Rosengren, Erik Renstrom and their team from the Lund University Diabetes Centre in Sweden used newly created rat strains — called N1I5 and N1I11, both derived from a well-studied type 2 diabetes strain — to identify the Adra2a gene as one of the genetic culprits behind defective insulin secretion. They then tied a variation in this gene to type 2 diabetes in humans.
When Rosengren and his colleagues stimulated β cells from each strain with glucose, they found that N1I5 cells secreted less insulin than both N1I11 and normal cells. Other experiments verified that whatever was causing impaired insulin secretion was unique to N1I5.
Of the handful of possible gene candidates, only Adra2a showed different expression between the two strains, causing N1I5 cells to produce almost twice as much of the encoded α(2A)AR protein as N1I11 cells. Additional experiments confirmed that α(2A)AR was indeed responsible for impaired insulin secretion. α(2A)AR is known to be involved in suppressing insulin secretion but variations in the gene had not previously been associated with the development of type 2 diabetes.
Based on the strong evidence from their rat experiments, Rosengren’s group hypothesized that variations in the human version of Adra2a, helpfully called ADRA2A, would contribute to type 2 diabetes risk in humans. Using two independent populations totaling about 5500 people, they first found that the less common A version of was associated with impaired insulin secretion.
To see whether the results extended to actual diabetes risk, Rosengren and his team then compared a group of 2830 type 2 diabetics to 3740 non-diabetics. They found that was also significantly associated with type 2 diabetes, with each copy of the A version increasing the odds of having the disease by about 1.4 times. The final piece of evidence came from laboratory experiments. Pancreatic cells from people carrying the riskier A version of behaved just like the cells from the insulin-secretion impaired N1I5 rats.