Aim/hypothesis The distinct metabolic properties of visceral and subcutaneous adipocytes may be due to inherent characteristics of the cells that are resident in each fat depot. To test this hypothesis, human adipocytes were differentiated in vitro from precursor stromal cells obtained from visceral and subcutaneous fat depots and analysed for genetic, biochemical and metabolic endpoints. Methods Stromal cells were isolated from adipose tissue depots of nondiabetic individuals. mRNA levels of adipocytespecific proteins were determined by real-time RT-PCR.Insulin signalling was evaluated by immunoblotting with specific antibodies. Glucose transport was measured by a 2-deoxy-glucose uptake assay. Adiponectin secretion in the adipocyte-conditioned medium was determined by a specific RIA. Results With cell differentiation, mRNA levels of PPARG, C/EBPα (also known as CEBPA), AP2 (also known as GTF3A), GLUT4 (also known as SLC2A4) were markedly upregulated, whereas GLUT1 (also known as SLC2A1) mRNA did not change. However, expression of C/EBPα, AP2 and adiponectin was higher in subcutaneous than in visceral adipocytes. By contrast, adiponectin was secreted at threefold higher rates by visceral than by subcutaneous adipocytes while visceral adipocytes also showed two-to threefold higher insulin-stimulated glucose uptake. Insulininduced phosphorylation of the insulin receptor, IRS proteins, Akt and extracellular signal-regulated kinase-1/2 was more rapid and tended to decrease at earlier time-points in visceral than in subcutaneous adipocytes. Conclusions/interpretation Subcutaneous and visceral adipocytes, also when differentiated in vitro from precursor stromal cells, retain differences in gene expression, adiponectin secretion, and insulin action and signalling. Thus, the precursor cells that reside in the visceral and subcutaneous fat depots may already possess inherent and specific metabolic characteristics that will be expressed upon completion of the differentiation programme.

Glucagon-like peptide-1 and its analogs may preserve pancreatic beta-cell mass by promoting resistance to cytokine-mediated apoptosis. The mechanisms of TNFalpha-induced apoptosis and of its inhibition by exendin-4 were investigated in insulin-secreting cells. INS-1 and MIN6 insulinoma cells were exposed to 20 ng/ml TNFalpha, with or without pretreatment with 10 nm exendin-4. Treatment with TNFalpha increased c-Jun N-terminal protein kinase (JNK) phosphorylation 2-fold, reduced inhibitor-kappaBalpha (IkappaBalpha) protein content by 50%, induced opposite changes in caspase-3 and Bcl-2 protein content, and increased cellular apoptosis. Moreover, exposure to TNFalpha resulted in increased serine phosphorylation of both insulin receptor substrate (IRS)-1 and IRS-2 and reduced basal and insulin-induced Akt phosphorylation. However, in the presence of a JNK inhibitor, TNFalpha-induced apoptosis was diminished and serine phosphorylation of IRS proteins was prevented. When cells were pretreated with exendin-4, TNFalpha-induced JNK and IRS-1/2 serine phosphorylation was markedly reduced, Akt phosphorylation was increased, caspase-3 and Bcl-2 protein levels were restored to normal, and TNFalpha-induced apoptosis was inhibited by 50%. This was associated with a 2-fold increase in IRS-2 expression levels. A similar ability of exendin-4 to prevent TNFalpha-induced JNK phosphorylation was found in isolated pancreatic human islets. The inhibitory effect of exendin-4 on TNFalpha-induced JNK phosphorylation was abrogated in the presence of the protein kinase A inhibitor H89. In conclusion, JNK activation mediates TNFalpha-induced apoptosis and impairment of the IRS/Akt signaling pathway in insulin-secreting cells. By inhibiting JNK phosphorylation in a PKA-dependent manner, exendin-4 counteracts TNFalpha-mediated apoptosis and reverses the inhibitory events in the IRS/Akt pathway, resulting in promotion of cell survival.

IGF-I regulates bone acquisition and maintenance, even though the cellular targets and signaling pathways responsible for its action in human bone cells are poorly understood. Whether abnormalities in IGF-I action and signaling occur in human osteoblasts under conditions of net bone loss has not been determined. Herein we carried out a comparative analysis of IGF-I signaling in primary cultures of human osteoblasts from osteoporotic and control donors. In comparison with control cells, osteoporotic osteoblasts showed increased tyrosine phosphorylation of the IGF-I receptor in the basal state and blunted stimulation of receptor phosphorylation by IGF-I. Augmentation of basal IGF-I receptor phosphorylation was associated with coordinate increases in basal tyrosine phosphorylation of insulin receptor substrate (IRS)-2 and activation of Erk, which were also minimally responsive to IGF-I stimulation. By contrast, phosphorylation levels of IRS-1, Akt, and glycogen synthase kinase-3 were similar in the basal state in control and osteoporotic osteoblasts and showed marked increases after IGF-I stimulation in both cell populations, even though these responses were significantly lower in the osteoporotic osteoblasts. The IGF-I signaling abnormalities in osteoporotic osteoblasts were associated with reduced DNA synthesis both under basal conditions and after stimulation with IGF-I. Interestingly, treatment of the osteoporotic osteoblasts with the MAPK kinase inhibitor PD098059 reduced the elevated levels of Erk phosphorylation and increased basal DNA synthesis. Collectively, our data show that altered osteoblast proliferation in human osteoporosis may result from dysregulation of IGF-I receptor signaling, including constitutive activation of the IRS-2/Erk signaling pathway, which becomes unresponsive to IGF-I, and defective induction of the IRS-1/Akt signaling pathway.

The p66(Shc) protein isoform regulates MAP kinase activity and the actin cytoskeleton turnover, which are both required for normal glucose transport responses. To investigate the role of p66(Shc) in glucose transport regulation in skeletal muscle cells, L6 myoblasts with antisense-mediated reduction (L6/p66(Shc)as) or adenovirus-mediated overexpression (L6/p66(Shc)adv) of the p66(Shc) protein were examined. L6/(Shc)as myoblasts showed constitutive activation of ERK-1/2 and disruption of the actin network, associated with an 11-fold increase in basal glucose transport. GLUT1 and GLUT3 transporter proteins were sevenfold and fourfold more abundant, respectively, and were localized throughout the cytoplasm. Conversely, in L6 myoblasts overexpressing p66(Shc), basal glucose uptake rates were reduced by 30% in parallel with a approximately 50% reduction in total GLUT1 and GLUT3 transporter levels. Inhibition of the increased ERK-1/2 activity with PD98059 in L6/(Shc)as cells had a minimal effect on increased GLUT1 and GLUT3 protein levels, but restored the actin cytoskeleton, and reduced the abnormally high basal glucose uptake by 70%. In conclusion, p66(Shc) appears to regulate the glucose transport system in skeletal muscle myoblasts by controlling, via MAP kinase, the integrity of the actin cytoskeleton and by modulating cellular expression of GLUT1 and GLUT3 transporter proteins via ERK-independent pathways.

Increased apoptosis of cardiac progenitor cells (CPCs) has been proposed as a mechanism of myocardial damage and dysfunction. Glucagon-like peptide-1 (GLP-1) has been shown to improve heart recovery and function after ischemia and to promote cell survival. The protective effects of GLP-1 on oxidative stress-induced apoptosis were investigated in human CPCs isolated from human heart biopsies. Mesenchymal-type cells were isolated from human heart biopsies, exhibited the marker profile of CPCs, differentiated toward the myocardiocyte, adipocyte, chondrocyte, and osteocyte lineages under appropriate culture conditions, and expressed functional GLP-1 receptors. CPCs were incubated with GLP-1 with or without hydrogen peroxide (H(2)O(2)). Phospho- and total proteins were detected by immunoblotting and immunofluorescence analysis. Gene expression was evaluated by quantitative RT-PCR. The role of the canonical GLP-1 receptor was assessed by using the receptor antagonist exendin(9-39) and receptor-specific silencer small interfering RNAs. Cell apoptosis was quantified by an ELISA assay and by flow cytometry-detected Annexin V. Exposure of CPCs to H(2)O(2) induced a 2-fold increase in cell apoptosis, mediated by activation of the c-Jun N-terminal protein kinase (JNK) pathway. Preincubation of CPCs with GLP-1 avoided H(2)O(2)-triggered JNK phosphorylation and nuclear localization, and protected CPCs from apoptosis. The GLP-1 effects were markedly reduced by coincubation with the receptor antagonist exendin(9-39), small interfering RNA-mediated silencing of the GLP-1 receptor, and pretreatment with the protein kinase A inhibitor H89. In conclusion, activation of GLP-1 receptors prevents oxidative stress-mediated apoptosis in human CPCs by interfering with JNK activation and may represent an important mechanism for the cardioprotective effects of GLP-1.