PerspectiveCell Biology

Selective Insulin Sensitizers

See allHide authors and affiliations

Science  25 Mar 2011:
Vol. 331, Issue 6024, pp. 1529-1531
DOI: 10.1126/science.1204504

Type 2 diabetes mellitus and its complications are, with cardiovascular diseases, leading threats to public health in the 21st century. In the United States, type 2 diabetes care accounts for a third of federal health insurance (Medicare) expenditures—nearly half of it to treat associated macrovascular problems (1). The cornerstone of type 2 diabetes is insulin resistance—a decreased sensitivity of tissues and organs, such as the liver, to the metabolic effects of the hormone insulin (the other major cause is failure of the pancreas to produce insulin). Yet, except for thiazolidinediones—whose checkered safety history, troublesome side effects, and regulatory setbacks stifled widespread adoption by clinicians—treatment options for insulin resistance have generally remained unchanged since the 1940s. Although insulin signaling pathways in cells have been largely deciphered (2), the key mediators of insulin signaling are poor drug targets because they either lack a suitable ligand-binding domain, or are shared with other cellular pathways that regulate cell growth and proliferation. This realization spawned research into “alternative pathways” that control insulin resistance. On page 1621 of this issue, Kir et al. (3) find that human fibroblast growth factor 19 (FGF19) can boost certain effects of insulin on the mammalian liver, raising interest and questions about possible therapies involving this molecule.

After food intake, blood glucose concentration increases, triggering the secretion of insulin. Glucose is taken up by liver cells (hepatocytes), where insulin then stimulates its conversion into glycogen (as an energy reserve). Insulin also stimulates protein and lipid production in the liver (see the figure). Made in the small intestine in response to food intake, FGF19 acts on the liver to inhibit the synthesis of bile acids, thereby regulating cholesterol and triglyceride levels in the plasma. It also promotes energy expenditure, by acting on the central nervous system and by stimulating fat oxidation (4). Kir et al. reveal that like insulin, FGF19 also regulates hepatic glycogen and protein buildup in the mouse liver, through seemingly independent and possibly synergistic signaling pathways with insulin. Therapeutically, this could be beneficial to the insulin-insensitive diabetic liver. However, there are two key aspects of hepatic insulin resistance that must be considered.

Integrated pathways in the liver.

In response to food intake and an increase in blood glucose concentration, the hormones FGF19 and insulin control hepatic protein synthesis and glycogen production. Insulin also controls liver lipogenesis. In the diabetic liver, the effects of insulin are blocked. Treatment with FGF19 may selectively rescue glycogen and protein production without affecting lipid synthesis.


The diabetic liver overproduces glucose and atherogenic lipoproteins [very-low-density lipoproteins (VLDLs) and small dense LDLs, resulting in lower concentrations of high-density lipoproteins (HDLs)], which cause microvascular and macrovascular complications, respectively. Understanding the linkage between these two conditions would likely provide insight into potential treatments for type 2 diabetes. A leading view is that increased hepatic glucose production stems from impaired insulin action, whereas increased lipoprotein secretion stems from the preserved ability of insulin to promote lipogenesis (5). If so, redressing this balance to boost some actions of insulin, while curtailing others, would require identification of selective agonists of insulin signaling (similar to selective agonists developed for G protein–coupled receptors or nuclear receptors in other diseases). FGF19 could be viewed as one such agonist, in view of the demonstration by Kir et al. that it stimulates hepatic glycogen and protein synthesis without promoting lipogenesis.

But this conclusion remains conjectural to the extent that it attributes the increase in VLDL secretion in diabetes to increased de novo lipogenesis, which is usually a minor contributor to the process of lipoprotein assembly and secretion (6). In fact, there remain yawning gaps in our understanding of the effects of insulin on VLDL secretion. Although in fasting conditions insulin appears to drive VLDL secretion from the liver, especially in obese individuals with insulin resistance, it also inhibits this process in the postprandial state, independent of its ability to shunt free fatty acids away from liver into adipocytes (6). And in isolated liver cells, insulin can suppress processing, maturation, and degradation of the main VLDL protein, apolipoprotein B (6). Hence, the net effect of an insulin sensitizer on VLDL secretion by the liver cannot be predicted from its ability to promote lipogenesis.

Although Kir et al. did not analyze hepatic glucose production, FGF19 is likely to be effective in treating its abnormal elevation in the diabetic liver (7). FGF19 may help restore the liver's ability to prevent excessive glucose release by modulating the relative contributions of glycogen breakdown and de novo glucose production to this process.

The clinical future of FGFs is uncertain. Peptide-based diabetes therapeutics, once considered impractical and viewed as an obstacle to initiating insulin therapy (because of the complexity of measuring the correct dose as well as patient aversion to needles), have gained renewed attention from the experience with injectable agonists that mimic the effect of an incretin peptide hormone called glucagon-like peptide–1 (Glp-1), which stimulates insulin release. But FGF19 production is normal in diabetics, raising doubts about the benefits of boosting its actions. A similar concern, however, could have been raised over incretin-based treatments, but hasn't prevented their rapid adoption. Moreover, FGF19 production falls in response to administration of bile acid absorption inhibitors (8), a moderately effective and seemingly safe, if somewhat impractical, antidiabetic and lipid-lowering treatment. These data suggest that the therapeutic benefits of FGF19 may be secondary to its ability to restrict the expanded bile acid pool in diabetics.

Notably, Kir et al. draw attention to the crucible of insulin signaling and nutrient-activated nuclear receptors (those that control the transcription of genes involved in glucose and fat metabolism) that likely holds the key to unraveling the paradox of the diabetic liver, in which resistance to the actions of insulin on glucose production seemingly coexists with preserved sensitivity to the lipogenic effects of insulin. The looming threat of young type 2 diabetics coming of age with their cardiovascular comorbidities lends urgency to the search for remedies.


View Abstract

Stay Connected to Science

Navigate This Article