Search

Insulin resistance during pregnancy provokes gestational diabetes mellitus (GDM); however, the cellular mechanisms for this type of insulin resistance are not well understood. We evaluated the mechanisms(s) for insulin resistance in skeletal muscle from an animal model of spontaneous GDM, the heterozygous C57BL/KsJ-db/+ mouse. Pregnancy triggered a novel functional redistribution of the insulin-signaling environment in skeletal muscle in vivo. This environment preferentially increases a pool of phosphatidylinositol (PI) 3-kinase activity associated with the insulin receptor, away from insulin receptor substrate (IRS)-1. In conjunction with the redistribution of PI 3-kinase to the insulin receptor, there is a selective increase in activation of downstream serine kinases Akt and p70S6. Furthermore, we show that redistribution of PI 3-kinase to the insulin receptor increases insulin-stimulated IRS-1 serine phosphorylation, impairs IRS-1 expression and its tyrosine phosphorylation, and decreases the ability of IRS-1 to bind and activate PI 3-kinase in response to insulin. Thus, the pool of IRS-1–associated PI 3-kinase activity is reduced, resulting in the inability of insulin to stimulate GLUT4 translocation to the plasma membrane. These defects are unique to pregnancy and suggest that redistribution of PI 3-kinase to the insulin receptor may be a primary defect underlying insulin resistance in skeletal muscle during gestational diabetes.

Body weight is tightly regulated by food intake and energy dissipation, and obesity is related to decreased energy expenditure (EE). Herein, we show that nucleotide pyrophosphatase/phosphodiesterase 2 (ENPP2, autotaxin) is an adipose-derived, secreted enzyme that controls adipose expansion, brown adipose tissue (BAT) function, and EE. In mice, Enpp2 was highly expressed in visceral white adipose tissue and BAT and is downregulated in hypertrophied adipocytes/adipose tissue. Enpp2+/− mice and adipocyte-specific Enpp2 knockout mice fed a high-fat diet showed smaller body weight gains and less insulin resistance than control mice fed the same diet. BAT was functionally more active and EE was increased in Enpp2-deficient mice. In humans, ENPP2 expression in subcutaneous fat and ENPP2 levels in serum were reduced in obese subjects. Taken together, our results establish ENPP2 as an adipose-derived, secreted enzyme that regulates adipose obesity and systemic metabolism. They also suggest ENPP2 could be a useful therapeutic target for the treatment of metabolic disease.

OBJECTIVE—The expansion of adipose tissue mass seen in obesity involves both hyperplasia and hypertrophy of adipocytes. However, little is known about how adipocytes, adipocyte precursors, blood vessels, and stromal cells interact with one another to achieve adipogenesis. RESEARCH DESIGN AND METHODS—We have developed a confocal microscopy-based method of three-dimensional visualization of intact living adipose tissue that enabled us to simultaneously evaluate angiogenesis and adipogenesis in db/db mice. RESULTS—We found that adipocyte differentiation takes place within cell clusters (which we designated adipogenic/angiogenic cell clusters) that contain multiple cell types, including endothelial cells and stromal cells that express CD34 and CD68 and bind lectin. There were close spatial and temporal interrelationships between blood vessel formation and adipogenesis, and the sprouting of new blood vessels from preexisting vasculature was coupled to adipocyte differentiation. CD34+ CD68+ lectin-binding cells could clearly be distinguished from CD34− CD68+ macrophages, which were scattered in the stroma and did not bind lectin. Adipogenic/angiogenic cell clusters can morphologically and immunohistochemically be distinguished from crown-like structures frequently seen in the late stages of adipose tissue obesity. Administration of anti–vascular endothelial growth factor (VEGF) antibodies inhibited not only angiogenesis but also the formation of adipogenic/angiogenic cell clusters, indicating that the coupling of adipogenesis and angiogenesis is essential for differentiation of adipocytes in obesity and that VEGF is a key mediator of that process. CONCLUSIONS—Living tissue imaging techniques provide novel evidence of the dynamic interactions between differentiating adipocytes, stromal cells, and angiogenesis in living obese adipose tissue.

The cellular mechanisms for the insulin resistance of pregnancy and gestational diabetes mellitus (GDM) are unknown. The membrane protein plasma cell membrane glycoprotein-1 (PC-1) has been identified as an inhibitor of insulin receptor tyrosine kinase (IRTK) activity. We investigated insulin receptor function and PC-1 levels in muscle from three groups of obese subjects: women with GDM, pregnant women with normal glucose tolerance, and nonpregnant control subjects. Subjects (n = 6 for each group) were similar in age and degree of obesity (body fat >30%). IRTK activity, insulin receptor tyrosine phosphorylation, and protein levels of membrane glycoprotein PC-1 were determined in rectus abdominus muscle biopsies obtained at the time of either elective cesarean section or gynecological surgery. No significant differences were evident in basal insulin receptor tyrosine phosphorylation or IRTK activity in the three groups. After maximal insulin (10(-7) mol/l) stimulation, IRTK activity measured with the artificial substrate poly(Glu,Tyr) increased in all subjects but was lower in women with GDM by 25% (P < 0.05) and 39% (P < 0.001) compared with pregnant and nonpregnant control subjects, respectively. Similarly, insulin receptor tyrosine phosphorylation was significantly decreased in subjects with GDM (P < 0.05) compared with pregnant and nonpregnant control subjects. Treatment of the insulin receptors with alkaline phosphatase to dephosphorylate serine/threonine residues increased insulin-stimulated IRTK activity significantly in pregnant control and GDM subjects (P < 0.05), but these rates were still lower compared with nonpregnant control subjects (P < 0.05). PC-1 content in muscle from GDM subjects was increased by 63% compared with pregnant control subjects (P < 0.05) and by 206% compared with nonpregnant control subjects (P < 0.001). PC-1 content was negatively correlated with insulin receptor phosphorylation (r = -0.55, P < 0.05) and IRTK activity (r = -0.66, P < 0.05). These results indicate that pregnant control and GDM subjects had increased PC-1 content and suggest excessive phosphorylation of serine/threonine residues in muscle insulin receptors and that both may contribute to decreased IRTK activity. These changes worsen in women with GDM when controlling for obesity. These postreceptor defects in insulin signaling may contribute to the pathogenesis of GDM and the increased risk for type 2 diabetes later in life.