Lipocalins and insulin resistance: etiological role of retinol-binding protein 4 and lipocalin-2?

The prevalence of type 2 diabetes is increasing dramatically worldwide. Excess adiposity is an important contributor to the development of type 2 diabetes and cardiovascular diseases (1 ). Insulin resistance, inflammation, hypertension, and dyslipidemia, components of the metabolic syndrome, have been implicated in the effects of adiposity on type 2 diabetes and cardiovascular diseases, but the mechanisms responsible for these detrimental effects of adiposity have not been fully elucidated. Two paradigms are currently areas of intense study: one focused on ectopic fat and the other on the endocrine function of adipose tissue (2 ). Ectopic fat is present in nonadipose tissues such as the liver, muscle, and probably pancreatic -cells. Both lipodystrophy (failure to develop adipose tissue) and obesity with full adipose cells are characterized by a lack of fat-storage capacity, resulting in overflow to other tissues of triglycerides and free fatty acids in the form of ectopic fat. Normal physiological processes can be disrupted by this ectopic fat, leading to insulin resistance and impaired insulin secretion. The endocrine function of adipose tissue is an important regulatory process throughout the body that is carried out by signaling proteins secreted by adipose tissue. These signaling proteins are called adipocytokines or adipokines, and they include leptin, adiponectin, resistin, tumor necrosis factor, and interleukin-6. Members of the lipocalin family of proteins have large sequence differences but share a common tertiary structure formed by segments termed lipocalin folds (3 ). Lipocalin folds consist of 8 antiparallel -sheets that surround a hydrophobic pocket and allow lipocalins to function as transport or carrier proteins. Several human lipocalins have been identified, including retinol-binding protein 4 (RBP4), which has recently been added to the list of adipokines that may link obesity and insulin resistance (4, 5). In this issue of Clinical Chemistry, Dr. Wang and colleagues report on the association of lipocalin-2 with obesity, insulin resistance, and inflammation in mice and humans (6 ). Lipocalin-2 (also known as neutrophil gelatinase-associated lipocalin, siderocalin, 24p3, and uterocalin) has previously been shown to play a role in the innate immune response to infection by binding to iron-laden bacterial siderophores and thereby limiting bacterial growth (7 ). Wang et al. now report that circulating concentrations of lipocalin-2, as well as expression of lipocalin-2 in adipose and liver tissue, were increased in db/db obese diabetic mice compared with normal mice (6 ). Furthermore, plasma lipocalin-2 concentrations were higher in obese than in lean humans and were correlated with body mass index (BMI) and various components of the metabolic syndrome. After adjustment for BMI, statistically significant associations remained with fasting glucose, the Homeostasis Model Assessment (HOMA) index for insulin resistance, and C-reactive protein concentration. Treatment with the peroxisome proliferator–activated receptoragonist rosiglitazone markedly decreased lipocalin-2 expression in mice and circulating concentrations in both mice and humans. These changes in lipocalin-2 were correlated with changes in C-reactive protein and HOMA insulin resistance (HOMA-IR) index. The authors should be commended for conducting this careful study using various techniques in both mice and humans. Their intriguing results raise the question whether lipocalin-2 can be useful as a predictor of cardiovascular diseases and whether lipocalin-2 has causal effects on insulin resistance, hyperglycemia, or inflammation and could thus be a treatment target. In the human study reported by Wang et al., the association between lipocalin-2 concentrations and indicators of glucose metabolism was attenuated after adjustment for BMI: No significant association remained with type 2 diabetes, 2-h plasma glucose concentrations, or fasting insulin concentrations, and only modest correlations remained with fasting glucose and the HOMA-IR index (6 ). Given that BMI may not have fully captured the effect of body fat or specific body fat depots, the strength of the reported associations independent of body fatness may have been overestimated. However, the parallel decrease in lipocalin-2 and HOMA-IR after rosiglitazone treatment provides some reassurance that cross-sectional associations with insulin resistance are not solely attributable to incomplete control for body fatness. Possibly, the independent associations with HOMA-IR and fasting glucose, but not 2-h glucose concentrations, reflect a specific association between lipocalin-2 and hepatic insulin sensitivity or glucose output rather than peripheral insulin sensitivity (8 ). Further human studies with more precise measures of body fatness and glucose homeostasis are needed to confirm the associations reported by Wang et al. (6 ). Decreased adipocyte expression of glucose transporter 4 (GLUT4) is associated with systemic insulin resistance, and based on studies in mice, it has been postulated that RBP4 acts as the mechanistic link by which decreased adipocyte GLUT4 expression contributes to insulin resistance (4 ). Circulating RBP4 concentrations were increased in insulin-resistant mice. Moreover, transgenic overexpression of human RBP4 and injection of recombinant RBP4 decreased insulin sensitivity in normal mice, whereas genetic deletion of the RBP4 (retinol binding protein 4, plasma) gene or normalization of RBP4 concentrations in obese mice improved insulin sensitivity. RBP4 concentrations were also increased in obese humans, and higher concentrations were correlated with lower insulin *Address correspondence to this author at: Department of Nutrition, Harvard School of Public Health, 665 Huntington Ave., Bldg II, Boston, MA 02115. Fax 617-432-2435; e-mail [email protected]. Previously published online at DOI: 10.1373/clinchem.2006.080432 Editorial