Overexpression of adiponectin targeted to adipose tissue in transgenic mice: impaired adipocyte differentiation.

Adiponectin (ApN) is an adipokine whose expression and plasma levels are inversely related to obesity and insulin-resistant states. Chronic repercussions of ApN treatment or overexpression on adiposity and body weight are still controversial. Here, we generated a transgenic (Tg) mouse model allowing persistent and moderate overexpression of native full-length ApN targeted to white adipose tissue. Adipose mass and adipocyte size of Tg mice were reduced despite preserved calorie intake. This reduction resulted from increased energy expenditure and up-regulation of uncoupling proteins, and from abrogation of the adipocyte differentiation program, as shown by the loss of a key lipogenic enzyme and of adipocyte markers. Adipose mass remodeling favors enhanced insulin sensitivity and improved lipid profile of Tg mice. Alteration of the adipocyte phenotype was likely to result from increased expression of the preadipocyte factor-1 and from down-regulation of the transcription factor, CCAAT/enhancer binding protein-alpha, which orchestrates adipocyte differentiation. We further found that recombinant ApN directly stimulated pre- adipocyte factor-1 mRNA and attenuated CCAAT/enhancer binding protein-alpha expression in cultured 3T3-F442A cells. Conversely, opposite changes in the expression of these genes were observed in white fat of ApN-deficient mice. Thus, besides enhanced energy expenditure, our work shows that impairment of adipocyte differentiation contributes to the anti-adiposity effect of ApN.

Leptin treatment markedly increased plasma adiponectin but barely decreased plasma resistin of ob/ob mice.

Adiponectin (ApN) and leptin are two adipocytokines that control fuel homeostasis, body weight, and insulin sensitivity. Their interplay is still poorly studied. These hormones are either undetectable or decreased in obese, diabetic ob/ob mice. We examined the effects of leptin treatment on ApN gene expression, protein production, secretion, and circulating levels of ob/ob mice. We also briefly tackled the influence of this treatment on resistin, another adipocytokine involved in obesity-related insulin resistance. Leptin-treated (T) obese mice (continuous sc infusion for 6 days) were compared with untreated lean (L), untreated obese (O), and untreated pair-fed obese (PF) mice. Blood was collected throughout the study. At day 3 or day 6, fat pads were either directly analyzed (mRNA, ApN content) or cultured for up to 24 h (ApN secretion). The direct effect of leptin was also studied in 3T3-F442A adipocytes. Compared with L mice, ApN content of visceral or subcutaneous fat and ApN secretion by adipose explants were blunted in obese mice. Accordingly, plasma ApN levels of O mice were decreased by 50%. Leptin treatment of ob/ob mice increased ApN mRNAs, ApN content, and secretion from the visceral depot by 50-80%. Leptin also directly stimulated ApN mRNAs and secretion from 3T3-F442A adipocytes. After 6 days of treatment, plasma ApN of ob/ob mice increased 2.5-fold, a rise that did not occur in PF mice. Plasma resistin of T mice was barely decreased. Leptin treatment, but not mere calorie restriction, corrects plasma ApN in obese mice by restoring adipose tissue ApN concentrations and secretion, at least in part, via a direct stimulation of ApN gene expression. Such a treatment only minimally affects circulating resistin. ApN restoration could, in concert with leptin, contribute to the metabolic effects classically observed during leptin administration.

Complex haplotypes of IRS2 gene are associated with severe obesity and reveal heterogeneity in the effect of Gly1057Asp mutation.

In order to understand the role of the insulin receptor substrate-2 (IRS2) gene (chromosome region: 13q34) in obesity, a complex disorder associated with insulin resistance and glucose intolerance, we determined single nucleotide polymorphims (SNPs) and complex haplotypes in women with morbid obesity and a body mass index (BMI) of 41+/-0.8 kg/m2 ( n=99) compared with controls having a BMI of 23.8+/-0.1 kg/m2 ( n=92). Sequencing of unphased DNA or digestion of polymerase chain reaction fragments revealed seven SNPs, including a new C/T(-769) replacement at the 5' untranslated region. Considering four or seven SNPs, we reconstructed with the PHASE program nine or 24 haplotypes, respectively, that were well correlated into the cladogram. Logistic regression analysis with nine haplotypes in the whole sample revealed that obesity was associated with haplotype H3, with P<0.025, an odds ratio (OR) of 1.9 and a 95% confidence interval (CI) of 1.1-3.4, or pairs 3/3 ( P<0.005, OR=8.7, CI=1.9-40.1) and 3/4 ( P<0.023, OR=2.5, CI=1.1-5.6), all containing the the Gly1057Asp allelic variant of IRS2, whereas controls were associated with H5 and H6 ( P<0.02, OR=0.2, CI=0.01-0.81). Although obese H5 carriers (also containing Gly1057Asp mutation) were the most insulin resistant, haplotypes of IRS2 were poorly correlated (analysis of variance) with insulin resistance. By contrast, haplotypes H3, H4 and pairs 3/3 were consistently associated with increased 2-h glucose levels during an oral glucose tolerance test in obese individuals ( P<0.0005, 0.025 and 0.027, respectively). These data indicate that IRS2 is an influential gene in severe obesity and glucose intolerance in this population, whereas gene-based haplotypes of IRS2 have revealed heterogeneity in the behaviour of the Gly1057Asp mutation in relation to insulin resistance.