Peroxisome proliferator-activated receptor alpha (PPARalpha) potentiates, whereas PPARgamma attenuates, glucose-stimulated insulin secretion in pancreatic beta-cells

Fatty acids (FAs) are known to be important regulators of insulin secretion from pancreatic beta-cells. FA-coenzyme A esters have been shown to directly stimulate the secretion process, whereas long-term exposure of beta-cells to FAs compromises glucose-stimulated insulin secretion (GSIS) by mechanisms unknown to date. It has been speculated that some of these long-term effects are mediated by members of the peroxisome proliferator-activated receptor (PPAR) family via an induction of uncoupling protein-2 (UCP2). In this study we show that adenoviral coexpression of PPARalpha and retinoid X receptor alpha (RXRalpha) in INS-1E beta-cells synergistically and in a dose- and ligand-dependent manner increases the expression of known PPARalpha target genes and enhances FA uptake and beta-oxidation. In contrast, ectopic expression of PPARgamma/RXRalpha increases FA uptake and deposition as triacylglycerides. Although the expression of PPARalpha/RXRalpha leads to the induction of UCP2 mRNA and protein, this is not accompanied by reduced hyperpolarization of the mitochondrial membrane, indicating that under these conditions, increased UCP2 expression is insufficient for dissipation of the mitochondrial proton gradient. Importantly, whereas expression of PPARgamma/RXRalpha attenuates GSIS, the expression of PPARalpha/RXRalpha potentiates GSIS in rat islets and INS-1E cells without affecting the mitochondrial membrane potential. These results show a strong subtype specificity of the two PPAR subtypes alpha and gamma on lipid partitioning and insulin secretion when systematically compared in a beta-cell context.     

Peroxisome proliferator-activated receptor-alpha activation during pregnancy attenuates glucose-stimulated insulin hypersecretion in vivo by increasing insulin sensitivity,without impairing pregnancy-induced increases in beta-cell glucose sensing and responsiveness

We investigated the effects of acute (24-h) peroxisome proliferator-activated receptor (PPAR)alpha activation by WY14,643 (pirinixic acid) treatment on glucose-stimulated insulin secretion (GSIS) during pregnancy, in the rat, in relation to insulin sensitivity. GSIS after iv glucose challenge (500 mg/kg) was increased at d 15 of pregnancy but was attenuated by WY14,643 treatment in vivo, with decreases in acute insulin response (51%; P < 0.001) and total suprabasal 30-min area under the insulin curve (deltaI) (55%; P < 0.001). GSIS was unaffected by WY14,643 treatment in unmated rats. Islet perifusions were employed to identify persistent effects of PPARalpha activation. GSIS was enhanced, and the glucose threshold was reduced in perifused islets from pregnant rats, but WY14,643 treatment failed to reverse these effects. WY14,643 treatment of 15-d-pregnant rats significantly lowered (by 63%; P < 0.01) the insulin resistance index [total suprabasal 30-min area under insulin curve x suprabasal 30-min area under glucose curve (deltaI x deltaG)]. A strong positive linear relationship (r = 0.92) between acute insulin response and deltaI x deltaG was evident between groups. Our studies show that acute PPARalpha activation reverses the augmented GSIS evoked by pregnancy in vivo, whereas the isolated islets retain pregnancy-induced enhancement of beta-cell glucose sensing and responsiveness. Normalization of maternal GSIS to that found in the nonpregnant state is observed in association with alleviation of maternal insulin resistance.     

Regulation of hormone-sensitive lipase in islets

An unique isoform of hormone-sensitive lipase (HSL) is expressed in beta-cells. Recent findings suggest that HSL could be involved in the regulation of glucose stimulated insulin secretion (GSIS), however, these findings are controversial. To test the hypothesis that HSL is involved in control of normal GSIS via changes in its expression and/or activity in response to stimuli, we examined the effects of free fatty acid (FFA) loading and glucagon like peptide-1 (GLP-1) stimulation on the regulation of HSL expression and activity. With prolonged FFA loading, there was increased expression of beta-cell HSL and increased HSL hydrolytic activity in clonal beta-cells. Short-term treatment with GLP-1 increased HSL activity without changing the expression of the beta-cell isoform of HSL. Basal insulin secretion was increased, whereas GLP-1 potentiation of GSIS was decreased in islets isolated from HSL-/- mice, as compared to islets from wild type mice. Furthermore, using PancChip 2.2 cDNA microarrays (NIDDK consortium), the gene expression profile in the islets of HSL-/- mice was compared with wild type mice. Results showed changes in several metabolic pathways due to changes in lipid homeostasis caused by inactivation of HSL. Quantitative PCR for selected genes also revealed changes in genes that are related to insulin secretion, such as UCP-2. Therefore, these results suggest that the beta-cell isoform of HSL is involved in maintaining lipid homeostasis in islets and contributes to the proper control of GSIS.     

Dexamethasone counteracts the effect of prolactin on islet function: implications for islet regulation in late pregnancy

Islets undergo a number of up-regulatory changes to meet the increased demand for insulin during pregnancy, including increased insulin secretion and beta-cell proliferation. It has been shown that elevated lactogenic hormone is directly responsible for these changes, which occur in a phasic pattern, peaking on day 15 of pregnancy and returning to control levels by day 20 (term). As placental lactogen levels remain elevated through late gestation, it was of interest to determine whether glucocorticoids (which increase during late gestation) could counteract the effects of lactogens on insulin secretion, beta-cell proliferation, and apoptosis. We found that insulin secretion measured over 24 h in culture and acute secretion measured over 1 h in response to high glucose were increased at least 2-fold by PRL treatment after 6 days in culture. Dexamethasone (DEX) treatment had a significant inhibitory effect on secretion in a dose-dependent manner at concentrations greater than 1 nM. At 100 nM, a concentration equivalent to the plasma corticosteroid level during late pregnancy, DEX inhibited secretion to below control levels. The addition of DEX (>1 nM) inhibited secretion from PRL-treated islets to levels similar to those produced by DEX treatment alone. Bromodeoxyuridine (10 microM) staining for the final 24 h of a 6-day culture showed that PRL treatment increased cell proliferation 6-fold over the control level. DEX treatment alone (1-1000 nM) did not reduce cell division below the control level, but significantly inhibited the rate of division in PRL-treated islets. YoYo-1, an ultrasensitive fluorescent nucleic acid stain, was added (1 microM; 8 h) to the medium after 1-3 days of culture to examine cell death. Islets examined under confocal microscopy showed that DEX treatment (100 nM) increased the number of cells with apoptotic nuclear morphologies. This was quantified by counting the number of YoYo-labeled nuclei per islet under conventional epifluorescence microscopy. The numbers of YoYo-1-positive nuclei per islet in control and PRL-treated islets were not different after 3 days of culture. However, DEX treatment increased YoYo-1 labeling 7-fold over that in controls. DEX also increased YoYo-1 labeling in PRL-treated islets 3-fold over the control level. These data show that the increased plasma glucocorticoid levels found during the late stages of pregnancy could effectively reverse PRL-induced up-regulation of islet function by inhibiting insulin secretion and cell proliferation while increasing apoptosis.     

Metallothionein induction in islets of Langerhans and insulinoma cells

Isolated pancreatic islets from rat and mouse and the insulinoma cell lines, betaHC9 and RINm5F, were investigated to determine the regulation of metallothionein (MT). Dexamethasone (DEX) increased rat and mouse islet and insulinoma cell MT levels in a time- and concentration-dependent manner. Rat islet MT expression was increased with interleukin-1beta (IL-1beta), but not tumor necrosis factor-alpha (TNF). However, MT induction by IL-1beta and TNF was synergistic with DEX in rat islets and insulinoma cells. Mouse islet MT failed to respond to IL-1beta alone, although IL-1beta and TNF were synergistic. IL-1beta and TNF did not synergize with DEX for mouse islet MT induction. Zinc sulfate induced MT in rat islets but not mouse islets. MT messenger RNA levels were significantly increased in rat islets in response to DEX and IL-1beta plus DEX. The inducible nitric oxide synthase inhibitors N(G)-monomethyl-L-arginine and aminoguanidine failed to inhibit IL-1beta induced MT levels in insulinoma cells, and the nitric oxide generating agent sodium nitroprusside failed to significantly affect MT levels. Phorbol dibutyrate increased MT levels in rat islets and betaHC9 cells, but phorbol dibutyrate and IL-1beta effects were not additive. Transgenic MT-null and wild-type mouse islets had similar insulin contents, but basal and glucose-stimulated insulin release from MT-null islets were significantly lower than in wild-type islets. Blood glucose levels in MT-null mice were, however, slightly lower than those in wild-type mice. Thus, MT induction in pancreatic islets and beta-cells is regulated by cytokines and DEX, and protein kinase C activation may play a role. However, regulation of MT induction in mouse and rat islets differs. MT also appears to modulate insulin release from pancreatic islets.     

AMP-activated protein kinase agonist dose dependently improves function and reduces apoptosis in glucotoxic beta-cells without changing triglyceride levels

Prolonged hyperglycaemia leads to impaired glucose-stimulated insulin secretion (GSIS) and apoptosis in insulin-producing beta-cells. The detrimental effects have been connected with glucose-induced lipid accumulation in the beta-cell. AMP-activated protein kinase (AMPK) agonist, 5-aminoimidazole-4-carboxamide ribonucleoside (AICAR), promotes utilization of nutrient stores for energy production. It was tested how impaired GSIS and elevated apoptosis observed in insulinoma (INS)-1E cells after prolonged culture at 27 mM glucose were affected by the inclusion of 0.3 or 1 mM AICAR during culture. Glucose-induced impairment of insulin release was reverted by the inclusion of 0.3 but not 1 mM AICAR, which did not affect insulin content. The glucose-induced rise in triglyceride (TG) content observed in the cells cultured at 27 mM glucose was not altered by the inclusion of either 0.3 or 1 mM AICAR. Inclusion of 1 but not 0.3 mM AICAR during culture induced phosphorylation of AMPK and its downstream target acyl-CoA carboxylase. Phosphorylation was paralleled by reduced number of apoptotic cells and lowered expression of pro-apoptotic C/EBP homologous protein (CHOP). In conclusion, AICAR dose dependently improves beta-cell function and reduces apoptosis in beta-cells exposed to prolonged hyperglycaemia without changing TG levels.     

Beta-cell lipotoxicity in the pathogenesis of non-insulin-dependent diabetes mellitus of obese rats: impairment in adipocyte-beta-cell relationships.

Hyperinsulinemia, loss of glucose-stimulated insulin secretion (GSIS), and peripheral insulin resistance coexist in non-insulin-dependent diabetes mellitus (NIDDM). Because free fatty acids (FFA) can induce these same abnormalities, we studied their role in the pathogenesis of the NIDDM of obese Zucker diabetic fatty (ZDF-drt) rats from 5 weeks of age (before the onset of hyperglycemia) until 14 weeks. Two weeks prior to hyperglycemia, plasma FFA began to rise progressively, averaging 1.9 +/- 0.06 mM at the onset of hyperglycemia (P < 0.001 vs. controls). At this time GSIS was absent and beta-cell GLUT-2 glucose transporter was decreased. The triacylglycerol content of prediabetic islets rose to 10 times that of controls and was correlated with plasma FFA (r = 0.825; P < 0.001), which, in turn, was correlated with the plasma glucose concentration (r = 0.873; P < 0.001). Reduction of hyperlipacidemia to 1.3 +/- 0.07 mM by pair feeding with lean littermates reduced all beta-cell abnormalities and prevented hyperglycemia. Normal rat islets that had been cultured for 7 days in medium containing 2 mM FFA exhibited increased basal insulin secretion at 3 mM glucose, and first-phase GSIS was reduced by 68%; in prediabetic islets, first-phase GSIS was reduced by 69% by FFA. The results suggest a role for hyperlipacidemia in the pathogenesis of NIDDM; resistance to insulin-mediated antilipolysis is invoked to explain the high FFA despite hyperinsulinemia, and sensitivity of beta cells to hyperlipacedemia is invoked to explain the FFA-induced loss of GSIS.     

Non-esterified fatty acids are deleterious for human pancreatic islet function at physiological glucose concentration

Aims/hypothesis Whether excess glucose (glucotoxicity) and excess non-esterified fatty acids (lipotoxicity) act synergistically or separately to alter beta-cell function on Type 2 diabetes remains controversial. We examined the influence of non-esterified fatty acids, with or without concomitant increased glucose concentrations, on human islet function and on the expression of genes involved in lipid metabolism.Methods Human islets isolated from non-diabetic and non-obese donors were cultured with 5.5, 16 or 30 mmol/l glucose, and when appropriate with 1 or 2 mmol/l non-esterified fatty acids. After 48 h, glucose-stimulated insulin secretion, insulin content, triglyceride content and expression of different genes were evaluated.Results Non-esterified fatty acids decreased glucose-stimulated insulin secretion, insulin content and increased triglyceride content of human isolated islets, independently from the deleterious effect of glucose. Increased glucose concentrations also decreased glucose-stimulated insulin secretion and insulin content, but had no influence on triglyceride content. Glucose-stimulated insulin secretion of islets appeared to be significantly correlated with their triglyceride content. Glucose and non-esterified fatty acids modified the gene expression of carnitine palmitoyltransferase-I, acetyl-CoA carboxylase, acyl-CoA oxidase and uncoupling protein 2.Conclusion/interpretation In our model of isolated human islets, increased glucose and non-esterified fatty acids separately reproduced the two major beta-cell alterations observed in vivo, i.e. loss of glucose-stimulated insulin secretion and reduction in islet insulin content. Our results also suggest that this deleterious effect was, at least in part, mediated by modifications in lipid metabolism gene expression.Abbreviations ACC Acetyl-CoA carboxylase - ACO acyl-CoA oxidase - CPT-I carnitine palmitoyltransferase-I - GSIS glucose-stimulated insulin secretion - PPAR peroxisome proliferator-activated receptor - PDX-1 pancreatic/duodenal homeobox-1 - PPRE peroxisomal proliferator response element - TG triglyceride - UCP-2 uncoupling protein 2     

Diabetogenic impact of long-chain omega-3 fatty acids on pancreatic beta-cell function and the regulation of endogenous glucose production

In healthy individuals, peripheral insulin resistance evoked by dietary saturated lipid can be accompanied by increased insulin secretion such that glucose tolerance is maintained. Substitution of long-chain omega-3 fatty acids for a small percentage of dietary saturated fat prevents insulin resistance in response to high-saturated fat feeding. We substituted a small amount (7%) of dietary lipid with long-chain omega-3 fatty acids during 4 wk of high-saturated fat feeding to investigate the relationship between amelioration of insulin resistance and glucose-stimulated insulin secretion (GSIS). We demonstrate that, despite dietary delivery of saturated fat throughout, this manipulation prevents high-saturated fat feeding-induced insulin resistance with respect to peripheral glucose disposal and reverses insulin hypersecretion in response to glucose in vivo. Effects of long-chain omega-3 fatty acid enrichment to lower GSIS were also observed in perifused islets suggesting a direct effect on islet function. However, long-chain omega-3 fatty acid enrichment led to hepatic insulin resistance with respect to suppression of glucose output and impaired glucose tolerance in vivo. Our data demonstrate that the insulin response to glucose is suppressed to a greater extent than whole-body insulin sensitivity is enhanced by enrichment of a high-saturated fat diet with long-chain omega-3 fatty acids. Additionally, reduced GSIS despite glucose intolerance suggests that either long-chain omega-3 fatty acids directly impair the beta-cell response to saturated fat such that insulin secretion cannot be augmented to normalize glucose tolerance or beta-cell compensatory hypersecretion represents a response to insulin resistance at the level of peripheral glucose disposal but not endogenous glucose production.     

Exercise and dexamethasone oppositely modulate beta-cell function and survival via independent pathways in 90% pancreatectomized rats

Long-term dexamethasone (DEX) treatment is well known for its ability to increase insulin resistance in liver and adipose tissues leading to hyperinsulinemia. On the other hand, exercise enhances peripheral insulin sensitivity. However, it is not clear whether DEX and/or exercise affect beta-cell mass and function in diabetic rats, and whether their effects can be associated with the modulation of the insulin/IGF-I signaling cascade in pancreatic beta-cells. After an 8-week study, whole body glucose disposal rates in 90% pancreatectomized (Px) and sham-operated male rats decreased with a high dose treatment of DEX (0.1mg DEX/kg body weight/day)(HDEX) treatment, while disposal rates increased with exercise. First-phase insulin secretion was decreased and delayed by DEX via the impairment of the glucose-sensing mechanism in beta-cells, while exercise reversed the impairment of first-phase insulin secretion caused by DEX, suggesting ameliorated beta-cell functions. However, exercise and DEX did not alter second-phase insulin secretion except for the fact that HDEX decreased insulin secretion at 120 min during hyperglycemic clamp in Px rats. Unlike beta-cell functions, DEX and exercise exhibited increased pancreatic beta-cell mass in two different pathways. Only exercise, through increased proliferation and decreased apoptosis, increased beta-cell mass via hyperplasia, which resulted from an enhanced insulin/IGF-I signaling cascade by insulin receptor substrate 2 induction. By contrast, DEX expanded beta-cell mass via hypertrophy and neogenesis from precursor cells, rather than increasing proliferation and decreasing apoptosis. In conclusion, the improvement of beta-cell function and survival via the activation of an insulin/IGF-I signaling cascade due to exercise has a crucial role in preventing the development and progression of type 2 diabetes.