Visfatin: A New Adipokine

See allHide authors and affiliations

Science  21 Jan 2005:
Vol. 307, Issue 5708, pp. 366-367
DOI: 10.1126/science.1106933

The developed world has been engulfed by a rapid rise in the incidence of obesity and diabetes. One hypothesis for the marked prevalence of obesity posits that selective evolutionary pressure favored those of our ancestors who were able to store excess food in the form of triglycerides, the primary component of adipose (fat) tissue. Unfortunately, what may have been an advantage eons ago is now manifest as a rapidly expanding waistline. The increased incidence of obesity and diabetes carry many sequelae, collectively known as the metabolic syndrome or syndrome X. Such sequelae include an increased risk of cardiovascular and kidney disorders as well as breast cancer, hepatocellular carcinoma, and Alzheimer's disease. How these and other conditions are related to the multiple defects in energy metabolism associated with diabetes and obesity is a major focus of current research efforts. An emerging theme is that adipose tissue, found in several locations throughout the body and long thought to be primarily a repository for triglycerides, is also important for regulating homeostasis and metabolism. Fat is an endocrine tissue and, indeed, may constitute the largest endocrine organ in the body (1, 2). Several hormones, called adipokines or adipocytokines, are synthesized by adipose tissue. These include adiponectin, tumor necrosis factor-α (TNF-α), leptin, resistin, and adipsin (see the figure). To be added to the list, as Fukuhara et al. report on page 426 of this week's issue (3), is a presumed new member of the adipokine family, visfatin.

Certain attributes of visceral fat, the adipose tissue surrounding abdominal organs, make its accumulation more worrisome than the accumulation of subcutaneous fat, which resides below the skin. Such concerns include an association of increased visceral fat with the metabolic syndrome, enhanced secretion of TNF-α and other proinflammatory adipokines, and reduced secretion of antidiabetic and anti-inflammatory adipokines relative to those secreted by other fat depots (4). Using differential display of expressed genes in visceral fat and subcutaneous fat obtained from two female volunteers, Fukuhara et al. (3) identified a molecule that is expressed at much higher levels in visceral fat than in subcutaneous fat. This molecule, denoted visfatin to indicate its abundance in visceral fat, turns out to have been already identified as a growth factor for early B cells called pre-B cell colony-enhancing factor (PBEF) (5). Adipose tissue and cells of the immune system are known to express many of the same genes; indeed, some researchers have postulated that adipose tissue may serve not only to coordinate metabolic pathways in the body but also to coordinate immunological responses (6).

The adipokine family.

The differentiation of preadipocytes into mature fat cells is accompanied by the expression of genes encoding hormones that regulate the metabolism of fats and sugars. Three of these hormones—TNF-α, resistin, and IL-6—induce resistance to insulin, the principal hormone that regulates blood glucose levels. TNF-α is a proinflammatory cytokine that suppresses expression of adipocyte-specific genes; resistin maintains blood glucose levels during fasting; and IL-6 production increases in those with obesity and diabetes. Adiponectin and visfatin are adipokines that work synergistically with insulin to enhance glucose uptake and metabolism in muscle and to block glucose formation (gluconeogenesis) in liver. Adiponectin activates AMP-activated protein kinase (AMPK), modulates signaling pathways controlled by the master transcription factor NF-κB, increases β-oxidation of fatty acids by muscle, protects endothelial cells, and is reduced in diabetic or obese individuals. Visfatin, the newly characterized hormone secreted by visceral fat, binds to the insulin receptor at a site separate from insulin and acts as a natural insulin mimetic. Leptin activates AMPK, acts centrally and peripherally to regulate metabolism and to reduce food intake, and is reduced in individuals with rare genetic obesity disorders.


Fukuhara and colleagues analyzed the relationship between serum visfatin levels and the amount of visceral and subcutaneous fat in humans and in several mouse models of obesity and diabetes. They found that visfatin levels in serum increased in parallel with visceral but not subcutaneous fat in both mice and humans. To address the importance of visfatin beyond it's known effect on B cells, Fukuhara et al. used recombinant visfatin produced by transfected cultured cells and purified from culture medium. The primary amino acid sequence of visfatin does not include a signal sequence, although Fukuhara et al. did detect low serum concentrations of visfatin indicating that this adipokine may be secreted; this observation was also made by the investigators who originally described PBEF (5). Other researchers working on PBEF, however, have found this cytokine primarily in the cell nucleus and cytoplasm, which suggests that visfatin's putative role as a secreted protein involved in metabolism should be interpreted with caution (7). Rongvaux and colleagues have characterized murine PBEF as an intracellular nicotinamide phosphoribosyl-transferase and have demonstrated that PBEF can rescue a mutant bacterium deficient in this enzymic activity (8). Precedent exists for secreted proteins lacking a signal sequence, such as the cytokine interleukin-(IL)-1β, although these examples represent the minority of secreted proteins. Inhibition of the secretory pathway in adipocytes did not seem to affect secretion of visfatin, raising the possibility that this hormone is released during lysis of fat cells rather than being secreted (8).

To address the ways in which visfatin could be involved in metabolism, Fukuhara et al. injected recombinant visfatin into KKAy obese mice (a model of type II diabetes where the mice become obese at 6 to 12 weeks of age) and mice rendered diabetic by injection of streptozocin. They observed a decrease in the concentration of plasma glucose in both mouse models similar to that induced by injection of insulin, suggesting that visfatin may be an insulin-mimetic. Chronic supplementation of visfatin in these mice (achieved by infecting animals with an adenoviral vector carrying the visfatin gene) resulted in a slightly reduced concentration of plasma glucose and a decrease in insulin levels. These effects were greater in obese KKAy mice than in wild-type mice. Next the investigators showed that mice heterozygous for a targeted mutation in the visfatin gene had elevated plasma glucose levels, both while fasting and after feeding, and exhibited a small defect in a glucose tolerance test. (Mice lacking both copies of the visfatin gene die in utero.)

Like insulin, visfatin stimulates glucose uptake by cultured adipocytes and muscle cells and suppresses glucose release by cultured hepatocyes (liver cells). Visfatin, like insulin, also induces phosphorylation of signal transduction proteins that operate downstream of the insulin receptor. Most intriguing of all, Fukuhara et al. show that visfatin binds to the insulin receptor but does not compete with insulin, suggesting that the two proteins bind to different sites. That this is the case is confirmed by the observation that visfatin binds tightly to a mutant insulin receptor with reduced affinity for insulin. The 3 nM affinity of visfatin for the wild-type insulin receptor is similar to that of insulin, suggesting that this interaction may have physiological relevance. However, the lower serum levels of visfatin, compared to those of insulin, and the fact that visfatin levels do not change after feeding imply that the hypoglycemic effects of visfatin may not be of physiological importance. There are other noninsulin molecules that activate the insulin receptor, notably membrane-permeable small molecules that bind to intracellular domains of the receptor and stimulate activity (9). These findings have spurred development of an orally bioavailable small molecule with insulin-mimicking effects (10), which may evolve into an antidiabetic drug in the future. Given that visfatin mimics insulin signaling by binding to the insulin receptor, this adipokine may serve as a valuable model for studying other insulin-mimetic molecules. Further study of the cell biology and physiology of this natural insulin-mimetic should help to determine its role in glucose homeostasis and will boost diabetes research and therapy.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
View Abstract

Stay Connected to Science

Navigate This Article