A class of proteins known for its involvement in muscle development, brain connectivity, and cancer has now been found in the liver, where it spurs sugar production when we need it most. The proteins, called histone deacetylases (HDACs), activate genes by hopping onto a section of DNA and helping switch on or off its transcription into RNA. They are the desired target of many potential drugs currently in development. The new study, published May 13, 2011, in the journal Cell, is the first to show that some of these HDACs play an important role in diabetes.

“Because of their potential use in different cancers, the whole area [of HDAC drug development] has just exploded.”
Reuben J. Shaw

The findings suggest that drugs that block some forms of HDACs could curb the sky-high levels of blood sugar in people with diabetes.

Several HDAC inhibitors are currently being evaluated in clinical trials as potential treatments for cancer. "Because of their potential use in different cancers, the whole area [of HDAC drug development] has just exploded," says lead investigator Reuben Shaw, an HHMI early career scientist and assistant professor of molecular and cell biology at Salk Institute for Biological Studies in La Jolla, California. "We now have reason to believe that some of the same types of drugs tested in the cancer setting might have potential to treat diabetes.

In the liver, a dizzying array of interconnected chemical pathways keeps our blood sugar levels steady. When we're asleep and must go many hours without food, these processes spur the liver to crank out glucose. Conversely, after we eat a big meal, the system dials down glucose production.

In people with diabetes, however, the liver does not respond properly to these fasting and feeding cues, leading to excessive levels of glucose in the blood. "The liver gets out of control, making glucose as if you're starving all the time," Shaw says.

Eight years ago, Shaw stumbled onto one of the key players in this complex signaling system. He had been studying cancer, and trying to figure out what proteins interact with a tumor suppressor called LKB1. Much to his surprise, he discovered that LKB1 activates a well-known metabolism protein: AMP kinase (AMPK). The results, published in Science in 2005, were among the first to molecularly link cancer and metabolism.

In that work, Shaw also showed that metformin—the world's most widely used diabetes drug—depends on the LKB1-AMPK pathway to curb glucose production.

AMPK triggers changes that lower blood sugar, sensitize cells to insulin, enable cells to burn more fat, suppress inflammation, and otherwise influence metabolic pathways. It does so by linking chemical phosphate groups to other proteins, which switches them on. In the new study, Shaw aimed to find out another piece of the pathway: what does AMPK regulate to cause the glucose shut-off?

Using liver cell cultures from mice and humans, Shaw's team compared normal cells and those in which AMPK was intentionally shut off. They also knew a specific code sequence AMPK needs in its targets. After treating the cells with metformin, the researchers looked for proteins containing the code sequence that were phosphorylated in the normal cells but not phosphorylated in the cells with impaired AMPK. These were assumed to be AMPK's targets.

They then tested each to see what functions it might have in liver cells. At first, they were surprised to find this subgroup of HDACs in the liver, as they weren’t known to be expressed there. In subsequent experiments, they showed that these specific HDACs are crucial for glucose production during times of food deprivation.

Maria Mihaylova, a graduate student in Shaw’s lab, had noticed that HDAC phosphorylation levels changed depending on how much time had passed since the mice ate: fasting reduced the phosphorylation, whereas feeding increased it. "This was the key insight to decoding the whole thing," Shaw says.

Mihaylova figured out that glucagon, a hormone the pancreas makes during fasting, was somehow involved in blocking HDAC phosphorylation.

Putting all the pieces together, the researchers propose that during fasting, members of this group of HDACs are stripped of their phosphate baggage, allowing them to rapidly invade the nucleus. Once there, they hop on to certain transcription factors that trigger the production of glucose. In contrast, after a meal, there is no glucagon, meaning that AMPK can phosphorylate HDACs and prevent them from entering the cell nucleus.

The findings could have immediate clinical implications. Many diabetics benefit from metformin, but it tends to lose effectiveness over years of treatment. "People are always trying to come up with a better metformin," Shaw says.

Over the past few years, researchers have developed an array of drugs that block HDACs for use in different forms of cancer. In the new study, Shaw simulated the drugs by reducing the activity of specific HDAC genes in several mouse models of diabetes. This led to a dramatic drop in the animals' blood glucose levels.

"We think that some of those drugs—in fact, some that may have been worthless for cancer—may now find usefulness in the treatment of diabetes," Shaw says. "It's certainly something that should be looked into."