Writing in the journal Nature, the researchers - led by a team from the Salk Institute for Biological Studies, La Jolla, CA - say they believe their findings will lead to a new generation of safer and more effective drugs for diabetes.
Type 2 diabetes is managed with varying levels of success through drugs and lifestyle changes, including diet, weight loss and regular exercise.
There are two types of diabetes. Type 1 is where the immune system attacks the insulin-producing cells of the pancreas, effectively destroying the body's ability to make the hormone that keeps blood sugar or glucose in check.
Type 2 diabetes, which usually results from carrying too much weight and being physically inactive, is when the body develops resistance to insulin - so even though the pancreas continues to make the hormone, the cells that need it cannot use it properly, resulting in high levels of circulating glucose.
Rates of type 2 diabetes have risen in the last few decades, to the point where estimates suggest some 30 million Americans have the disease.
As a chronic disease, type 2 diabetes can lead to serious health problems. There is no cure as such, instead the disease is managed with varying levels of success through drugs and lifestyle changes, including diet, weight loss and regular exercise.
Treating diabetic mice with FGF1 reversed insulin resistance
What surprised the researchers in this new study was that treating the mice with FGF1 - fibroblast growth factor 1 protein - did not only keep glucose in check, but also reversed insulin resistance. Plus, there were none of the side-effects that normally accompany most current diabetes treatments.
Corresponding author Ronald M. Evans, professor and director of Salk's Gene Expression Laboratory, says:
"Controlling glucose is a dominant problem in our society. And FGF1 offers a new method to control glucose in a powerful and unexpected way."
Current drugs for diabetes attempt to reduce blood glucose by changing gene expression to boost insulin levels and reverse insulin resistance. One example is Byetta, which increases insulin production. But sometimes this can result in glucose dropping too far, which leads to life-threatening hypoglycemia, and other undesirable side effects.
In 2012, Prof. Evans and colleagues reported making an unexpected discovery: mice lacking FGF1 quickly develop diabetes when fed on a high-fat diet. They suggested this meant the protein was important for managing glucose levels.
That result led the team to wonder whether giving the growth factor to diabetic mice might affect the symptoms of the disease.
After a single dose, blood glucose dropped to normal levels, with no side effects
The researchers proceeded by injecting doses of FGF1 into obese mice with diet-induced diabetes - a mouse model that is often used as the equivalent of type 2 diabetes in humans. They were stunned by the profound impact that the protein had on the mice's metabolism: after just a single dose, blood glucose in all the treated mice quickly dropped to normal levels and stayed there for over 2 days.
After just a single dose of FGF1, blood glucose in all the treated mice quickly dropped to normal levels and stayed there for over 2 days.
As well the risk of glucose dropping to dangerous levels, among the drawbacks of current diabetic drugs, is that they have unwanted side effects such as weight gain and heart and liver problems. This is the problem with Actos, for example.
But the team found that even at high doses, FGF1 did not produce these side effects in the mice. By triggering the body's natural ability to regulate insulin, the protein kept blood glucose in a safe range - effectively reversing the number one symptom of diabetes.
The researchers believe among the reasons why FGF1 has a more "normal" response are that it targets specific cell types and metabolizes quickly.
However, they acknowledge they do not fully understand how FGF1 works - just as there still many unanswered questions around insulin resistance itself.
Team is planning human trials, but still long way to go
But the researchers say they did find the protein's ability to stimulate growth is totally separate from its effect on glucose, something that is important to know when considering it as a drug candidate.
Prof. Evans says he and his colleagues now want to find out which signaling pathways are involved when FGF1 acts on glucose to affect diabetes and metabolism.
They are already planning human trials, but say it will take time to fine-tune the protein into a drug for clinical use.
"We want to move this to people by developing a new generation of FGF1 variants that solely affect glucose and not cell growth," says Prof. Evans. "If we can find the perfect variation, I think we will have on our hands a very new, very effective tool for glucose control."