Wow. It’s like discovering that the orchestra’s second violinist, who does a good job but just isn’t as glamorous as the first violinist, also has a fantastic operatic singing voice. In this case, the surprise comes from metformin, the solid performer that since the 1950s has been the first non-insulin drug that doctors prescribe to newly diagnosed type 2s.
For years the understanding has been that metformin makes the liver more sensitive to insulin. How does it work? Between meals, the liver produces glucose-“fuel”-to supply energy and maintain metabolism. When people eat, their pancreases release insulin, which tells the body to absorb the glucose and makes the liver reduce or temporarily shut down glucose production.
In people without diabetes, the insulin successfully carries out those tasks. But people with type 2 diabetes have developed “insulin resistance”-their livers do not sense the presence of insulin and therefore do not respond to it. Their blood glucose levels spike, creating inflammation that over the long term can lead to cardiovascular, kidney, eye, and limb problems.
But according to a new study from the Johns Hopkins Children’s Center in Baltimore, metformin isn’t just a messenger that says, “Hey, liver, knock it off!” Instead, the drug is the message itself. In other words, metformin doesn’t tell the liver what to do; it makes the liver tamp down glucose production by mimicking and goading CBP, a signaling protein that communicates between the liver and pancreas.
The researchers tested how metformin works by feeding rats a high-fat diet over several months to induce insulin resistance. Once the mice became insulin-resistant and their blood glucose levels did not drop after eating, they began receiving metformin. As expected, the drug succeeded in returning their glucose levels to normal first by mimicking, and therefore activating, their CBP.
However, the success of the drug depended on whether the mice had normal CBP. Researchers focused on the one segment of CBP that metformin appeared to affect and removed that segment from some of the insulin-resistant mice. Their blood glucose levels were not affected by metformin-a sure indication that the drug works only when the recipient has intact CBP.
Aside from the disease treatment possibilities gained from a deeper knowledge of CBP, which is involved in the growth, development, and metabolic processes of other organs, the Johns Hopkins team came up with another finding that could have a dramatic effect on diabetes therapy: a biomarker that can predict how well a person will respond to metformin.
The team found that metformin changes CBP in white blood cells, creating an easily detected molecular marker that can be measured in a standard blood test. By tracking how much of a change it produces in an individual’s CBP, doctors will now be able to tailor doses to each individual.
The study was published in the May 15, 2009, issue of Cell.