Inflammatory ideas: new thoughts about causes of diabetes.

Eat right, exercise regularly, maintain a healthy weight, and you can reduce your risk of diabetes as you get older. Since following that advice is significantly harder than giving it, more and more people are developing type II diabetes, and researchers continue to explore its underlying causes in the hope of developing effective therapies. The task is more urgent than ever because the average age of people being diagnosed with the disease is falling as the U.S. population gets fatter. Overall, according to the American Diabetes Association, about 10 million people in the United States know they have type II diabetes, while more than 5 million with the disease remain undiagnosed.

Recently, scientists searching for causes of type II diabetes have begun to focus on the immune process called inflammation. In inflammation, the body responds to injury by sending specialized blood cells to damaged areas where they destroy invaders and clean up dead and dying cells. In recent years, scientists have discovered that this process plays a role in a surprising number of diseases (SN: 6/14/97, p. 374). Now, they're gathering evidence that inflammation precedes and predicts diabetes. Previous research had already linked inflammation to heart disease and obesity, which are both common in people with diabetes.

"Looking for a link between inflammation and diabetes was a matter of logic," notes Joshua Barzilay of Kaiser Permanente of Georgia in Tucker.

BLOOD WORK Both type II diabetes and type I, or juvenile, diabetes are characterized by high blood-sugar concentrations that result from defects in the body's use or production of insulin. Normally, insulin ushers sugar, the body's basic fuel, from the bloodstream into cells. Uncontrolled high concentrations of blood sugar can lead to heart disease, blindness, kidney disease, and circulation problems that may require amputations.

Scientists on the frontier of diabetes research are testing people's blood samples not only for insulin and glucose but also for a variety of compounds associated with inflammation. Some of these, such as interleukin-6 (IL-6) and tumor necrosis factor alpha (TNF-alpha), are cytokines--chemical signals that the immune system uses to marshal inflammatory activity.

Others, such as so-called acute-phase proteins made by the liver, rise in response to increased cytokine concentrations. These proteins don't contribute directly to inflammation but, because they remain detectable in blood longer than cytokines do, they're a convenient measure of it.

The concentrations of cytokines and acute-phase proteins, such as C-reactive protein, or CRP, rise at least a hundredfold when a person contracts an infection. By contrast, Barzilay says, in most studies linking inflammation to diabetes and to heart disease, these inflammation markers reach only perhaps twice-normal amounts. It's not clear whether at such low concentrations the cytokines trigger swelling or other inflammatory responses.

Although analyses of different indicators of inflammation sometimes produce conflicting findings, evidence is piling up that what the researchers call low-grade inflammation does indicate an increased risk for type II diabetes. It also seems to boost the likelihood of a condition called insulin resistance, which often precedes the disease.

Most of the evidence comes from analyses of blood samples and data collected in studies that have followed the health of large numbers of people over several years. All these investigations have used statistical techniques to take into account various factors, such as obesity, that might confound their results.

Recent support for the CRP-diabetes link comes from a study presented in the October 2001Diabetes. Barzilay and his colleagues tracked 5,888 U.S. residents without diabetes, 65 or older, who got their health care from the Kaiser Permanente health maintenance organization. Among the quarter of people with the highest CRP blood concentrations at the beginning of the study, twice as many had been diagnosed with diabetes after 3 to 4 years, compared with the quarter of people with the lowest CRP concentrations, Barzilay says.

Similar findings came from an offshoot of a study by Scottish researchers focusing on the heart benefits of the cholesterol-lowering drug pravastatin. It indicated a link between diabetes and CRP. Over 5 years, they found that 139 of the 5,974 people in the heart study developed diabetes. The researchers reported in the January 23, 2001 Circulation that the more elevated people's CRP concentrations were at the study's start, the higher their risk of developing diabetes.

Other researchers have looked at both CRP and cytokines. As part of a long-running national study, Paul Ridker of Brigham and Women's Hospital in Boston and his colleagues compared the medical histories of 188 middle-aged women who had diabetes with records on 362 women of similar age and weight who didn't. The quarter of women who had the highest CRP concentrations early in the study were four times as likely to develop the disease as were the 25 percent of women with the lowest CRP concentrations.

Also, women with the highest IL-6 concentrations were more than twice as likely to develop diabetes as were the women with the lowest IL-6 concentrations. Finally, those with the highest concentrations of both IL-6 and CRP were six times as likely to develop the disease over the course of the study as were women with low concentrations of the two compounds, Ridker and his colleagues reported in the July 18, 2001 Journal of the American Medical Association (SN: 8/11/2001, p. 89).

An international study of originally nondiabetic men and women has also suggested that "something is going on" between inflammation and type II diabetes, says Bruce Duncan of the Federal University of Rio Grande do Sul in Porto Alegre Brazil. During long-term tracking of 10,397 people, 651 developed type II diabetes. Comparing that group with 643 age- and weight-matched people in the study who remained free of the disease, Duncan and his colleagues found that high concentrations of CRP didn't predict diabetes risk. But people with the highest concentrations of sialic acid, a measure of CRP and several other acute-phase proteins, were 70 percent more likely to develop diabetes than were those with the lowest concentrations, Duncan reported at the American Diabetes Association (ADA) meeting in San Francisco last June.

"It's a little disappointing we didn't find stronger associations Duncan says, "but these findings add to a growing list of studies suggesting that markers of inflammation precede and predict diabetes. The question is less now if this association exists and more why it exists."

"Different cytokines seem to interact to increase risk of type II diabetes," says Heiner Boeing of the German Institute of Human Nutrition in Bergholz-Rehbruke. He and his colleagues reported at the San Francisco meeting that in a study of 188 people with diabetes and 377 others of similar age and weight--all selected from a much larger sample--the cytokine IL-1 alone had no association with diabetes risk. On the other hand, Boeing's group found that high blood concentrations of IL-6 tripled a person's chance of developing type II diabetes. Taken together, these cytokines had an even greater predictive value: The people with the highest concentrations of both cytokines were more than five times as likely to develop the disease as people with the lowest concentrations were.

"It is fair to say there is an association between diabetes and inflammation, "concludes Barzilay, "but there is not enough information to say whether the chicken or the egg comes first."

CAUSE AND EFFECT? None of these epidemiological studies can prove that inflammation causes diabetes. It's possible that some unknown factor predisposes people to both. One candidate is obesity. Fat cells are known to produce cytokines, and CRP is typically elevated in people who are overweight. Since obesity is a known risk factor for diabetes, says Vivian Fonseca of Tulane University Health Sciences Center in New Orleans, analyzing inflammatory markers and acute-phase proteins "may just be an expensive way of measuring obesity."

Epidemiological studies try to avoid this situation by adjusting their findings to account for people's weight. Though most of these studies continue to support the idea that inflammation is linked to diabetes, not all do.

However, laboratory studies are bolstering the view that inflammation causes insulin resistance. Animals with infections and those with cancer naturally have high concentrations of cytokines, and scientists have also detected increased insulin resistance in these animals.

Now, researchers are examining whether animals without underlying disease but with altered amounts of inflammatory cytokines are vulnerable to diabetes. For instance, mice lacking the gene for the cytokine TNF-alpha are less likely to develop obesity-linked insulin resistance than are mice with that gene. Recently scientists have shown that TNF-alpha blocks insulin from getting into cells.

This is "a clear-cut scenario where one of the inflammatory agents blocks insulin action," says Paresh Dandona of the University of Buffalo School of Medicine.

The cytokine IL-6 has also been implicated in blocking insulin action. Robert A. Mooney and his colleagues at the University of Rochester School of Medicine in Rochester, N.Y., have shown that IL-6 indirectly blocks the biological effects of the insulin receptor in liver cells. In unpublished work, Mooney has found that injections of IL-6 act through another compound to blunt insulin response to sugar injections, leaving the test animals with higher concentrations of blood sugar. "This begins to show that there truly is a cause-and-effect relationship between IL-6 and insulin resistance and later diabetes," he says.

Mooney says that inflammation is the process through which obesity can cause diabetes. Without fat cells, these cytokines wouldn't be at higher-than-normal concentrations.

Further evidence that inflammation may cause diabetes comes from studies suggesting that aspirin and other anti-inflammatory agents can reverse insulin resistance in mice, rats, and people, says Steven E. Shoelson of the Joslin Diabetes Center at Harvard Medical School in Boston. In the Aug. 31, 2001 Science, he reported that high doses of aspirinlike compounds lower concentrations of sugar, insulin, and fat in the blood of obese rodents. They do this by interfering with an enzyme that triggers a cascade of reactions that drives the production of cytokines and inflammation.

"These findings implicate an inflammatory process in the pathogenesis of insulin resistance in obesity and type II diabetes," Shoelson says.

In the May Journal of Clinical Investigation, he and his colleagues report that daily doses of 7 grams of aspirin for 2 weeks boosted insulin's effectiveness in people with diabetes. Because such extremely high doses of aspirin tend to cause ulcers, the team is now investigating the effects of an aspirinlike drug called disalcid, which is used to treat rheumatoid arthritis. So far, Shoelson says, it too seems to increase people's response to insulin. Disalcid or a drug like it may eventually be useful in preventing and treating type II diabetes, he predicts.

SWEET SOLUTIONS? Researchers are debating whether excess fat cells are the primary source of the inflammatory cytokines that seem to be triggering type II diabetes. Dandona argues that eating frequent heavy meals may boost inflammation, regardless of whether a person is overweight. He and his colleagues gave people with diabetes doses of sugar or a high-calorie fast-food breakfast. The result in both cases was increased production of free radicals, cell-damaging molecules that can trigger inflammation, the team reported at the San Francisco ADA meeting.

"An ordinary meal induces a transient inflammatory effect for 3 to 4 hours," Dandona says. Taking vitamins E and C, which block free radicals, seems to avoid the inflammatory effect of the high-calorie meals, he reports.

There are other approaches that might benefit people prone to diabetes. Dandona has shown that either insulin (SN: 9/1/01, p. 143) or the diabetes drugs called thiazolidinediones can reduce several markers of inflammation. That could explain why thiazolidinediones make people less insulin resistant, he says. Further, he adds, he has experimental evidence suggesting that reduced inflammation lowers a diabetic person's risk of developing heart disease.

Not only do some diabetes drugs appear to reduce risk of heart disease but some heart drugs may stave off diabetes. Last year, the Scottish study reported in Circulation suggested that the drug called pravastatin lowers the incidence of diabetes by a third. Pravastatin and the other cholesterol-lowering drugs known collectively as statins seem to reduce inflammation.

Intrigued by reports suggesting that statins might improve blood sugar control and prevent the onset of diabetes, Jeffrey A. Johnson of the University of Alberta in Edmonton and his colleagues examined a database of about 11,000 Saskatchewan people with diabetes. Looking back at medical records for 5 to 9 years, the researchers noted what medications the patients had taken. Johnson and his colleagues calculated that the use of statins correlated with a 10-month delay in a person's need to start taking insulin in response to severe type II disease.

"The anti-inflammatory effects of statins might improve insulin sensitivity," Johnson speculates.

Most researchers consider it too early in the study of the inflammation-diabetes connection for people with or at high risk for diabetes to receive treatment for low-grade inflammation. Still, the field is an active area of study for scientists pursuing new diabetes treatments.

"I like to think of science as a big puzzle," says Shoelson. The image of inflammation as it fits into type II diabetes is provocative, "but we're not quite there yet with the whole picture."