By David N. Leff

Obesity and diabetes go together like ham and eggs (triple helpings?).

How is it, then, that over 80 percent of individuals with Type II diabetes mellitus are obese, but only 10 percent of obese people are diabetic? This paradox has exercised two biochemists - Samuel Nadler and Alan Attie - at the University of Wisconsin, Madison.

Nadler is lead author, and Attie senior author, of a paper in the Oct. 10, 2000, issue of the Proceedings of the National Academy of Sciences (PNAS). Its title - reporting the paradox - reads: "The expression of adipogenic genes is decreased in obesity and diabetes mellitus."

Adipogenic genes, as their name implies, are expressed in adipocytes - the body's fat cells. These comprise the bulk of the weight gain in pathologically obese individuals. Such adipocytes tend to be hypertrophic - larger rather than more numerous.

"What we did," Attie told BioWorld Today, "was look at the pattern of gene expression that changes when you go from a lean mouse to an obese one, and from an obese non-diabetic animal to an obese diabetic one. Much to our surprise," he went on, "the expression of genes that are normally associated with differentiation of adipocytes from their precursor cells goes down in obesity - and for the most part stays down in diabetes.

"That told us," he went on, "that the burden of lipogenesis - the lipid content of fat cells - is shifted away from adipocytes to the wrong places. And one of those wrong places that we know occurs in the liver. So we think of that perhaps as a molecular definition of a prediabetic state."

To gauge the behavior of adipocyte genes in mammals of varying body weight and diabetic propensity, the co-authors set up a small menagerie featuring five leptin-lacking, overweight strains of mouse, with mild to severe hyperglycemia.

DNA Chip Scoped 'Over 11,000 Genes At Once'

"We isolated RNA from adipose tissue of lean, obese and obese-diabetic animals," Nadler recounted. "Then we applied those samples to DNA microarrays - gene chips. And the amounts of gene message that hybridized to the chips led to a level of intensity that gave us the amount of gene expression in the sample.

"We then compared the samples," Nadler continued, "to see if gene expression went up or went down. These chips allowed us to take a picture of gene expression in over 11,000 genes at once. What we saw was a pattern of gene expression in these animals that was the exact opposite of the genes and gene expression when adipocytes go from a preadipocyte state to a mature adipocyte state. That was very interesting to us because these animals are obese. They've got bigger fat pads. They're looking more and more like immature fat cells - the opposite of what we expected."

He and Attie "are looking at this as the dedifferentiation of adipose cells. They're becoming more like their precursors. Therefore, the role that they ordinarily would play in normal physiology is taken over by other organs, such as the liver, which are not as well suited to these functions."

Attie recalled that the evolutionary mission of adipocytes was "helping us to protect against starvation, back when caloric excess leading to obesity was no problem. Now we have - with obesity and diabetes - problems that are consequences of the way in which we defend against starvation. We have wonderful mechanisms for hanging on to calories to defend against starvation, and the better we have evolved those, the worse off we are when we have caloric excess."

He opined that "as the adipocytes stop playing their traditional role in metabolism, they become more and more insulin resistant in responding to circulating levels of that hormone. Their responsibilities are taken over by other tissues. So, when you dysregulate fat and glucose, then you start to have the onset of diabetes."

Any therapeutic potential in their research thus far, Attie pointed out, "has to be factored in with work from other people. There have been a few interesting surprises lately, and one of them has come from transgenic mouse experiments in which people successfully developed animals that did not have any fat at all - no adipose tissue. And the surprising result was that those fat-free mice were diabetic.

"Our findings fit very well with that result," he observed, "because we saw a pattern of gene expression that was consistent with dedifferentiation of adipose tissue. The idea is that you do need some kind of fat - and it should be well-differentiated fat - to protect you against diabetes.

"It also fits well," he added, with the new generation of drugs for treating Type II diabetes - the so-called insulin-sensitizers - which are thiazolidinediones. These are drugs that lower blood glucose by enhancing target-cell response to insulin. They are agonists for a transcription factor called PPAR-g - peroxisome proliferator activator receptor gamma. They promote adipocyte differentiation.

As for treating obesity per se - as a medical disorder in itself - Nadler observed, "I don't see a clear picture of the molecular events involved in obesity. In terms of treating it, I think that diet and exercise will be the way to go."

Four New Genes Aimed At Long-Range Prediction

In research now ongoing in his lab, Attie recounted, "We have mapped two genes that interact with obesity to cause diabetes, as well as two obesity genes. They reside, respectively, on murine chromosomes 16 and 19; 3 and 10 in humans. And we've made congenic mouse strains in which we explore the gene that's causing the diabetes. This gives us a uniform population of animal models so we can now look at the time course of disease progression. That's clinically important," he pointed out, "because most people who are diabetic are obese, but most people who are obese are not diabetic. We have no clue, no markers, to tell who's to become diabetic among all the obese people.

"By using these gene profiling methods," Attie concluded, "we hope we can ascertain what changes in gene expression exist that can predict which animals are going to develop diabetes - long before the onset of disease."

The university has filed for a provisional patent on the Attie-Nadler invention, covering leads for new drug targets and diagnostics.