We evaluated the role of ATP-sensitive K+ (KATP) channels, somatostatin, and Zn2+ in the control of glucagon secretion from mouse islets. Switching from 1 to 7 mmol/L glucose inhibited glucagon release. Diazoxide did not reverse the glucagonostatic effect of glucose. Tolbutamide decreased glucagon secretion at 1 mmol/L glucose (G1) but stimulated it at 7 mmol/L glucose (G7). The reduced glucagon secretion produced by high concentrations of tolbutamide or diazoxide, or disruption of KATP channels (Sur1−/− mice) at G1 could be inhibited further by G7. Removal of the somatostatin paracrine influence (Sst−/− mice or pretreatement with pertussis toxin) strongly increased glucagon release, did not prevent the glucagonostatic effect of G7, and unmasked a marked glucagonotropic effect of tolbutamide. Glucose inhibited glucagon release in the absence of functional KATP channels and somatostatin signaling. Knockout of the Zn2+ transporter ZnT8 (ZnT8−/− mice) did not prevent the glucagonostatic effect of glucose. In conclusion, glucose can inhibit glucagon release independently of Zn2+, KATP channels, and somatostatin. Closure of KATP channels controls glucagon secretion by two mechanisms, a direct stimulation of α-cells and an indirect inhibition via somatostatin released from δ-cells. The net effect on glucagon release results from a balance between both effects.

ObjectiveNon-coding RNAs constitute a major fraction of the β-cell transcriptome. While the involvement of microRNAs is well established, the contribution of long non-coding RNAs (lncRNAs) in the regulation of β-cell functions and in diabetes development remains poorly understood. The aim of this study was to identify novel islet lncRNAs differently expressed in type 2 diabetes models and to investigate their role in β-cell failure and in the development of the disease.MethodsNovel transcripts dysregulated in the islets of diet-induced obese mice were identified by high throughput RNA-sequencing coupled with de novo annotation. Changes in the level of the lncRNAs were assessed by real-time PCR. The functional role of the selected lncRNAs was determined by modifying their expression in MIN6 cells and primary islet cells.ResultsWe identified about 1500 novel lncRNAs, a number of which were differentially expressed in obese mice. The expression of two lncRNAs highly enriched in β-cells, βlinc2, and βlinc3, correlated to body weight gain and glycemia levels in obese mice and was also modified in diabetic db/db mice. The expression of both lncRNAs was also modulated in vitro in isolated islet cells by glucolipotoxic conditions. Moreover, the expression of the human orthologue of βlinc3 was altered in the islets of type 2 diabetic patients and was associated to the BMI of the donors. Modulation of the level of βlinc2 and βlinc3 by overexpression or downregulation in MIN6 and mouse islet cells did not affect insulin secretion but increased β-cell apoptosis.ConclusionsTaken together, the data show that lncRNAs are modulated in a model of obesity-associated type 2 diabetes and that variations in the expression of some of them may contribute to β-cell failure during the development of the disease.

The mechanisms of control of glucagon secretion are largely debated. In particular, the paracrine role of somatostatin (SST) is unclear. We studied its role in the control of glucagon secretion by glucose and K channel blockers, using perifused islets and the in situ perfused pancreas. The involvement of SST was evaluated by comparing glucagon release of control tissue or tissue without paracrine influence of SST (pertussis toxin-treated islets, or islets or pancreas from mice). We show that removal of the paracrine influence of SST suppresses the ability of K channel blockers or K channel ablation to inhibit glucagon release, suggesting that in control islets, the glucagonostatic effect of K channel blockers/ablation is fully mediated by SST. By contrast, the glucagonostatic effect of glucose in control islets is mainly independent of SST for low glucose concentrations (0-7 mmol/L) but starts to involve SST for high concentrations of the sugar (15-30 mmol/L). This demonstrates that the glucagonostatic effect of glucose only partially depends on SST. Real-time quantitative PCR and pharmacological experiments indicate that the glucagonostatic effect of SST is mediated by two types of SST receptors, SSTR2 and SSTR3. These results suggest that alterations of the paracrine influence of SST will affect glucagon release.

The specific phenotype of mature differentiated beta cells not only depends on the specific presence of genes that allow beta cell function but also on the selective absence of housekeeping genes (“disallowed genes”) that would interfere with this function. Recent studies have shown that both histone modifications and DNA methylation via the de novo methyltransferase DNMT3A are involved in repression of disallowed genes in neonatal beta cells when these cells acquire their mature phenotype. It is unknown, however, if the environmental influence of advanced age, pregnancy and the metabolic stress of high fat diet or diabetes could alter the repression of disallowed genes in beta cells. In the present study, we show that islet disallowed genes—which are also deeply repressed in FACS-purified beta cells—remain deeply repressed in animals of advanced age and in pregnant females. Moreover, the stability of this repression was correlated with strong and stable histone repression marks that persisted in islets isolated from 2 year old mice and with overall high expression of Dnmt3a in islets. Furthermore, repression of disallowed genes was unaffected by the metabolic stress of high fat diet. However, repression of about half of the disallowed genes was weakened in 16 week-old diabetic db/db mice. In conclusion, we show that the disallowed status of islet genes is stable under physiological challenging conditions (advanced age, pregnancy, high fat diet) but partially lost in islets from diabetic animals.

The effect of a high-fat diet on the expression of the three main isoforms of caveolins in adipocytes isolated from rat retroperitoneal and subcutaneous white adipose tissue was investigated. Two distinct phases can be distinguished on a time-dependent response in adipocytes from both locations. The early stage affects only to retroperitoneal adipocytes and implies caveolin-1 activation and caveolin-2 inactivation, together with increased expression of insulin signaling intermediaries. This initial response would be aimed to counterbalance the energy overload. Continued exposure to the high-fat diet produces an increase in circulating glucose and insulin levels, inducing a late stage in which adipocytes from both locations are affected. This late stage is characterized by general increased caveolin-1 and caveolin-2 expression; while on the other hand, the insulin signaling intermediaries are downregulated, with the noticeable exception of GLUT-4, whose expression remains high. Therefore, it seems that at this stage caveolins and GLUT-4 are regulated independently of the insulin pathway, through a mechanism that could be mediated by inflammation and oxidative stress associated with obesity. Although this GLUT-4 upregulation suggests a response against the raise in circulating glucose, this might not be the case, since the developing insulin resistance at this stage indicates a prediabetic state. We have also found that the high-fat diet is able to induce the expression of muscle-specific caveolin-3 in retroperitoneal adipocytes since the initial phase. This observation is similar to what we reported previously in skeletal muscle (Gómez-Ruiz et al., 2009, FEBS Lett 583:3259-3264), suggesting a similar regulatory mechanism for this isoform.