The molecular mechanisms underlying the link between obesity and diabetes have been elusive. A new protein, christened 'resistin', can now be added to the panoply of factors that may be involved.
Type II diabetes is the most common form of diabetes in the Western world, and is strongly linked to obesity — over 80% of sufferers are obese. The molecular basis for this link has remained a mystery. On page 307 of this issue1, however, Steppan and colleagues describe how they have identified a new hormone, which they have named 'resistin', that is produced by fat cells. Their results indicate that resistin may form at least part of the missing link between obesity and diabetes.
In patients with type II diabetes, insulin is less able to promote the uptake of glucose into muscle and fat, and to inhibit the production of glucose by the liver. Several molecular mechanisms are known by which this state — insulin resistance — comes about, but the full picture is not yet available2.
Obesity is characterized by the increased storage of triglycerides (fat molecules) in adipose tissue and causes insulin resistance. But how does this increased energy storage in fat cells (adipocytes) promote insulin resistance in muscles, the liver and elsewhere in the body? For many years, it looked as if free fatty acids would provide the link. These products of triglyceride metabolism are the main form in which energy is transferred from stores in adipose tissue to other sites in the body for metabolic use. After all, levels of free fatty acids in the bloodstream are higher in obese than in non-obese people, and free fatty acids can induce insulin resistance in tissues other than adipose tissue3. However, as well as having a role in energy storage, adipocytes also secrete numerous peptides that might lead to insulin resistance or other complications of obesity (Fig. 1).
Two such peptides are the cytokine tumour-necrosis factor-α (TNF-α) and the hormone leptin. Although better known for its roles in inflammation and immunity, TNF-α is expressed in normal adipocytes, is overexpressed in adipocytes from obese people, and can cause insulin resistance through effects on insulin-mediated cellular signalling pathways4. TNF-α is almost certainly involved in insulin resistance in vivo. But the balance of evidence suggests that other factors must be required, too.
Is leptin such a factor? It is best known as the adipocyte hormone whose absence causes profound obesity in rodents and humans, but it also has potent effects on insulin's action5,6. In rodents, for example, an inherited deficiency in leptin causes both severe insulin resistance and obesity. Returning leptin to these rodents reverses insulin resistance, but by mechanisms that are independent of effects on food intake and body weight7. This reversal in insulin resistance may result from leptin working within the brain as well as directly on target tissues. But in contrast to rodents, in which leptin deficiency causes obesity, levels of leptin are higher than normal in most obese people8. So it is not yet clear how alterations in leptin might affect insulin resistance.
Steppan et al.1 have now taken a creative approach to identifying new adipocyte-derived mediators of insulin resistance. The thiazoladinedione class of antidiabetes drugs reduces insulin resistance by acting through a nuclear receptor protein that is abundant in fat cells9. This protein is the peroxisome proliferator-activated receptor-γ (PPAR-γ)10, which has been found to guide the differentiation of adipocytes11. The results of genetic and pharmacological experiments revealed that this nuclear receptor also affects insulin sensitivity, by unknown mechanisms that presumably involve altered gene expression in adipocytes. Might the thiazoladinediones work through PPAR-γ to switch on or off an adipocyte-specific gene that is involved in insulin-mediated signalling pathways? Steppan et al. assumed that it does.
By treating a differentiated adipocyte cell line with thiazoladinediones, the authors have identified a new messenger RNA that is expressed only in adipose tissue and is suppressed by thiazoladinediones. The authors find that the protein encoded by the new adipocyte-specific mRNA is overexpressed in fat in a variety of rodent models of obesity. Moreover, thiazoladinediones reduce the amount of this protein that is secreted from adipocytes in vitro and released into the bloodstream in mice.
The evidence that this new protein is involved in insulin resistance is provocative, but, so far, circumstantial. So, to test this hypothesis directly, the authors administered recombinant protein, and antibodies that recognize the endogenous protein, to fat cells in vitro and to mice. Some of the effects were rather small, but they all point in the same direction — that the protein, named resistin, is an important link between obesity and diabetes, and may mediate the beneficial effects of a major class of antidiabetes drugs.
Nonetheless, questions remain. Steppan et al. have shown that resistin suppresses insulin's ability to stimulate glucose uptake into adipose cells. But it is not known whether resistin also acts on important physiological targets, such as muscles, the liver or even the brain. Resistin is presumably detected in the bloodstream by a receptor protein on the surface of target cells. But what is the identity of the resistin receptor? How exactly does resistin-mediated signalling antagonize insulin-mediated signalling and insulin's effects? And what is the role of resistin in normal physiology? Answers to these questions will probably require gene-knockout experiments, careful examination of physiological responses to administration of resistin, and identification of the factors that naturally regulate the amount of resistin in the blood. It will also be important to determine whether genetic differences in people's ability to produce or respond to resistin contribute to variations in insulin sensitivity and susceptibility to diabetes.
More generally, however, this work1 illustrates an important principle of modern molecular science. Drugs that are discovered empirically, such as thiazoladinediones, can often be used for much more than treating diseases. In the hands of good scientists, they can also be essential tools for discovering previously hidden molecular pathways.
About this article
Medical Molecular Morphology (2006)