Harvard Public Health Review Home

Harvard Public Health Review/Summer 2002

random samples

Getting to the Heart (and stomach and gene) of Diabetes

The poet Ralph Waldo Emerson once wrote, "Every sweet hath its sour; every evil its good." He could have been talking about diabetes, an insidious disease caused by the harmful build-up of sugars in the body, but also a contemporary health "evil" that has rallied a unique group of Harvard School of Public Health researchers to advance its prevention and treatment. Diverse in background but united in purpose, these scientists have found innovative ways--from large populations to minute genes--for exploring how diabetes inflicts its damage. Their findings may mean a sweeter future for those at risk for diabetes and related conditions.

Diabetes has all the makings of a modern epidemic. Almost 17 million Americans now have the disease, with about one million additional cases diagnosed each year, at an annual health care cost of nearly $100 billion. According to the Centers for Disease Control and Prevention, the prevalence of diabetes has risen by almost 50 percent just over the past decade, with increases in both sexes and across all ages, ethnicities, and education levels. And we may only have ourselves to blame. "We've probably been susceptible to them for hundreds of thousands of years," says Associate Professor Eric Rimm of chronic diseases like diabetes and heart disease. "It's just that we didn't sit at desk chairs, eat red meat, smoke, and do all the things that we do today. It hasn't been until the 20th century that we actually had all the exposures that pushed us this way."

When it comes to diabetes, the direction we're headed is far from good. The condition can be both debilitating and deadly. Diabetes is caused by defects in the action or secretion of insulin, a hormone produced by the pancreas that helps usher sugar into the cells of the body where it can be converted into energy. In the absence or dysfunction of insulin, sugars accumulate in the blood and wear away at the body's vital systems. Heart disease, kidney failure, and blindness are just a few of the effects of uncontrolled diabetes. In the past primarily an over-age-40 problem, diabetes is now hitting people younger and younger. Indeed, the scientific community actually renamed adult-onset diabetes, which accounts for up to 95 percent of all cases of the disease, Type 2 diabetes to reflect the recent shift. (The risk factors for the less common Type 1 diabetes, which generally strikes in childhood, are not well defined.) "It's devastating to hear about 11-year-olds, 100 pounds overweight, being diagnosed with adult-onset diabetes," notes Rimm, "and that people are having heart attacks at age 40 and 50 because they've already had diabetes for 25 years."

It was, in fact, this connection to heart disease that lured Rimm, a cardiovascular epidemiologist and nutrition expert, to the perimeter of diabetes research. Type 2 diabetes increases the risk of heart disease up to sevenfold in women and up to fourfold in men. Eventually 75 percent of people with Type 2 diabetes will die from some kind of cardiovascular problem. Interested in better understanding the predictors of heart disease, it was a natural progression for Rimm and his colleagues to step back and examine related--perhaps even causal--conditions like diabetes and obesity. Using the storehouses of dietary data from the renowned Health Professionals Follow-Up Study, which Rimm directs, and its sister studies, the Nurses' Health Studies I and II, they have shed new light on the risk factors for diabetes and its cardiovascular consequences.

"We found that 90 percent of diabetes could be prevented," observes Frank Hu, Rimm's colleague in the Department of Nutrition, "which was actually very surprising, because even though we suspected that many diabetes cases were caused by diet and lifestyle, we didn't know the magnitude--it's just so big." Hu, associate professor of nutrition and cardiovascular disease, has become an expert on the subset of almost 10,000 volunteers with diabetes in the health professionals and nurses studies. He has concluded that if we ate well, drank moderately, exercised regularly, and stayed lean, we could potentially live diabetes-free lives.

But isn't this old news? Not exactly, according to Hu. While there's plenty of advice to eat right and stay active, few Americans are heeding its call. Obesity is on the rise and sedentary lifestyles are the norm. And it may be that all that good advice isn't really clear at all. "Of course, it's important to emphasize eating a balanced diet, a healthy diet," Hu says, "but if you don't know what a healthy diet is, what do you do? Most people would say that exercise is good for you but when you talk about exercise, what does that mean? If people always think about exercise as always going to the gym or running marathons, then few people are going to do it and it loses its public health importance. So this is why it's really important to pinpoint the exact message you want to convey and have scientific support for that message."

Hu's work has shown, for example, that physical activity doesn't have to be vigorous to be beneficial; even regular brisk walking at least 30 minutes a day can reduce the risk of Type 2 diabetes by 30 to 40 percent. His research has also illustrated the importance of distinguishing between different types of fats and carbohydrates. Overall fat intake is not associated with the risk of Type 2 diabetes, whereas trans fatty acids, often found in commercially baked or fried goods, increase it and polyunsaturated fat, in foods such as nuts and olive oil, actually reduce it. "The low-fat campaign was clearly misleading," notes Hu.

Alcohol is another surprise. Not only do healthy people who drink moderately have a 30 to 40 percent lower risk of developing diabetes, but even those with the disease have almost half the risk of having a heart attack if they drink moderately as opposed to teetotalers. The protective effects, regardless of the kind of alcohol, improve with more frequent consumption (4 to 5 times per week). The results fly in the face of the traditional view that alcohol is bad and, in particular, harmful for diabetics who were often advised to avoid it altogether. "It's alcohol so it's nothing you are necessarily going to prescribe to people," remarks Rimm, who has become something of an expert on the subject, "although there are cardiologists who tell their diabetic patients to drink if they don't. But it really would have to be on an individual basis." In the big picture, however, alcohol's a bit player. It will provide greatest benefit only to those who are already at high risk for conditions like diabetes and heart disease, and other factors clearly outweigh its advantages. "If you got rid of obesity in this country and got everyone to exercise and eat a prudent diet, you'd get rid of 80 to 90 percent of diabetes," asserts Rimm. "So a lot of the things that we study, like alcohol--great, big deal, it lowers your risk by 30 to 40 percent. But if you could lower your risk of diabetes by 90 percent, the further impact of alcohol would hardly matter."

Obesity. Diabetes. Heart disease. The connections between these seemingly disparate conditions keep cropping up again and again. But Rimm and Hu's large observational studies can't tell us much about what's going on behind the curtain biologically to produce these effects. That's where another Department of Nutrition faculty member comes in. Associate Professor Gokhan Hotamisligil is a molecular geneticist looking at a cluster of diseases from the gene on up. Known as metabolic syndrome, or syndrome X, this cluster includes, as he puts it, "all of the horrible things you can think of all together"--diabetes, obesity, insulin resistance, high cholesterol, hypertension, atherosclerosis. By studying their molecular and genetic foundation, he hopes to understand why these conditions occur together and why one tends to increase the risk for the others--so he can ultimately break them apart, using the "good old divide and conquer strategy." Using virtually every organism on earth, from fruit flies to humans, to tease out culpable genes, he and his coworkers identify critical pathways of disease, then "knock out" the genes from the DNA of mice, and observe the outcome with respect to the disease in question. If mice deprived of a certain gene get sicker or healthier, then the gene and the protein it codes for are suspect; if the mice remain as they were, it's evidence that the gene is likely a bystander.

Hotamisligil's research group has found a couple of knock-out knock-outs in this regard. The first is a gene that codes for tumor necrosis factor (TNF), a cytokine produced excessively in fat cells, which, after years of work, they suspected was involved in blocking the action of insulin. When they bred mice that were both obese and lacking TNF genes, they found that, despite the extra weight, the mice were less insulin resistant and less diabetic than their obese but genetically intact counterparts. In addition to establishing a genetic link between diabetes and obesity, these discoveries have also improved the genetic understanding of what Hotamisligil calls "the inflammatory connection" in metabolic syndrome. TNF, for example, is released in the inflammatory response of the body's immune system--a response that has also been strongly implicated in cardiovascular disease. "My personal view is that all of these diseases develop from a joint venture between the metabolic system and the inflammatory system, which are coupled together very tightly," says Hotamisligil. "We think the reason for this is evolutionary--that they all emerged from one single functional entity, act in concert on many occasions, and hence have maintained many common elements."

Hotamisligil's "new darlings" are the fatty acid binding proteins (FABPs), a unique class of molecules that bind to lipids. When one of the FABP genes, thought to be specific only to adipocytes, or fat cells, is destroyed, as in the TNF case, the resulting knock-out mice become resistant to diabetes. Recently, the FABPs have also done their part in substantiating Hotamisligil's inflammation theory. Hotamisligil has found the gene that produces FABPs exists not only in the fat cell, as originally presumed, but in the macrophage too; hypercholesterolemic mice lacking the FABP gene in these immunoregulatory cells become dramatically resistant to atherosclerosis.

With a great deal of time and effort, Hotamisligil has found that by blocking certain combinations of FABPs in adipocytes and macrophages, he can create mice that won't become obese, or diabetic, or atherosclerotic no matter how unlimited or unhealthy their diet--a pretty appealing scenario for those of us who like to have our cake without the consequences. But unfortunately it's not as simple as all that. First, it remains uncertain if such a scenario would be possible in humans. Second, completely knocking out genes may have unintended negative results or may be compensated by other genes designed to back up their function. "We want to go deeper into the pathway, we want to be more specific," says Hotamisligil, "because we can't possibly propose that we get rid of immune system genes and we'll be all set. It would be like proposing, let's cut off your head and you'll never have a headache."

Now Hotamisligil is working closely with population scientists like Rimm and Hu on projects to help put a genetic face on risk. Looking at the combination of genetic susceptibility and environmental variables will ideally provide a more accurate picture of both their additive and synergistic effects on our increased risk of diabetes and related diseases. "What we're trying to do is to make the bridge between the discovery of genes and their function in mouse models and the discovery of genetic variation of these same genes in human populations to identify patterns of susceptibility to diabetes," says Rimm. "Hypothetically, once you do that, there are several avenues of public health intervention. You could screen people to find someone who would benefit from a tailored diet or develop a drug that would mimic the function of a certain gene." While cooperation between epidemiologists and basic scientists is relatively new in public health, Rimm, Hu, and Hotamisligil all believe that it's something that comes naturally in the School's Department of Nutrition, where conversations over a cup of coffee or lunch spur new ideas and projects. Down the road they would like to see something more formalized to seal their joint efforts--collective grant proposals, perhaps even a center--to improve fundraising and better communicate their results to the public. But until that time, be it from the heart, the stomach, or the gene, these three hope that their more informal teamwork will pay off in advances that improve the prevention and treatment of diabetes for all. How sweet it is!

Alexandra Molloy

Harvard Public Health Review Home

next story

previous story

HSPH home

top of page

This page is maintained by Development Communications in the Office of Resource Development.
To contact us with suggestions, comments, and questions, please e-mail: editor@hsph.harvard.edu

Copyright, 2005, President and Fellows of Harvard College