Researchers have identified a protein that plays a central and hitherto-undescribed role in glucose trafficking in humans but isn't even expressed in mice, they report in this week's Science.

"We always knew that mice and humans were different from each other," said Yale Medical School cell biologist Jonathan Bogan, who was not involved in the study. "This gives us some insight into perhaps how, specifically, they're different."

The protein, called CHC22, is an isoform of the membrane protein clathrin, which forms a proteinaceous coat around vesicles containing the glucose transport molecule GLUT4, in muscle and fat cells. "We think [CHC]22 is involved in helping these vesicles form," said Frances Brodsky, a cell biologist from the University of California, San Francisco, who led the study, adding that the protein may also help to stabilize already-formed vesicles.

GLUT4 is found in muscle and fat cells, translocating to the plasma membrane in response to insulin signaling, which typically occurs after meals. Normally, when blood glucose is high, GLUT4 moves to the membrane to collect excess sugars. When blood glucose is low, GLUT4 sinks back into the cytosol in specialized vesicles that can deliver the protein back to the membrane if the insulin signal calls. In people with type 2 diabetes, there is some disruption in this process.

CHC22 was originally identified in the late 1990s, and Brodsky's lab has been studying the protein's functional role in human muscle cells for the past few years. The group conducted siRNA knockdown experiments of CHC22 in a human cell line, and to Brodsky's surprise, found the CHC22-deficient cells didn't form the glucose compartment -- implicating the protein in vesicle formation and/or trafficking. The group also found that administering functional CHC22 to mice disrupts how their muscle cells sequester excess glucose. Mice use a different clathrin isoform, CHC17, to perform the same vesicle-forming role that CHC22 appears to play in humans. "It was a very exciting 9-month period," she said.

Most researchers have used rodent models to investigate GLUT4 trafficking, but since CHC22 isn't present in mice, its link to diabetes would not have been found in those models. The study "doesn't negate previous results at all," Brodsky told The Scientist. "What it does say is that humans do something differently than mice."

And these differences are important to keep in mind for researchers studying mouse models of diabetes, Brodsky said. "We need to be aware of these differences as much as we are aware of the conserved parts of the pathway."

"In a sense, we always knew we could only go so far in a [diabetes] mouse model," said Bogan, who wrote an accompanying commentary on the study in this week's Science. "This now gives us a handle on why" that's the case.

Brodsky's results might offer further insight into the pathophysiology of type 2 diabetes because they suggest vesicle trafficking defects, as well as insulin signaling problems, may contribute to the disease -- as suggested by a pair of studies done about ten years ago in the University of Alabama at Birmingham lab of diabetes researcher Timothy Garvey. "Those older studies suggested that there might also be some vesicle trafficking defects," Bogan told The Scientist.

Brodsky also said that there is some preliminary evidence that human CHC22 exists in about 20 different variations. These variations, she said, might possibly be studied and correlated to different risk factors or phenotypic variations related to diabetes.

The differences in mouse vs. human glucose vesicle trafficking could play an important role in developing novel therapeutics for diabetes, said Bogan. "For a company that's developing drugs, for example, they might want to use cells from a species that's more related to humans." Brodsky added that there are some species, such as chickens and dogs, that do express CHC22.


Related stories:

Beta eye-lets
[August 2008]

A diabetes variant
[February 2008]

The Skinny Fat
[January 2008]