Evaluation ofin vivo insulin action and glucose metabolism in milk-fed rats

Milk diet has long been recommended in the management of gastrointestinal pathologies. Since milk feeding represents a high fat-low carbohydrate diet and it is acknowledged that insulin resistance is one of the consequences of high fat feeding, it is important to know whether or not chronic milk feeding leads to an impairment of the insulin-mediated glucose metabolism. To examine this question, adult female rats were given raw cow's milk (50% of total calories as lipids) for 18 days. They were compared to rats raised in parallel and fed the standard laboratory diet (15% of total calories as lipids). At the end of the 18 day period, body weight, daily caloric intake, basal plasma glucose and insulin levels in the milk-fed rats were similar to those in the control rats.In vivo insulin action was assessed with the euglycemichyperinsulinemic clamp technique in anesthetized animals. These studies were coupled with the 2-deoxyglucose technique allowing a measurement of glucose utilization by individual tissues. In the milk fed rats: 1) the basal rate of endogenous glucose production was significantly (p<0.01) reduced (by 20%); 2) their hepatic glucose production was however normally suppressed by hyperinsulinemia; 3) their basal glucose utilization rate was significantly (p<0.01) reduced (by 20%); 4) their glucose utilization rate by the whole-body mass or by individual tissues was normally increased by hyperinsulinemia. These results indicate that insulin action in adult rats is not grossly altered after chronic milk-feeding, at least under the present experimental conditions.     

Role of palmitate-induced sphingoid base-1-phosphate biosynthesis in INS-1 β-cell survival

Sphingoid base-1-phosphates represent a very low portion of the sphingolipid pool but are potent bioactive lipids in mammals. This study was undertaken to determine whether these lipids are produced in palmitate-treated pancreatic β cells and what role they play in palmitate-induced β cell apoptosis. Our lipidomic analysis revealed that palmitate at low and high glucose supplementation increased (dihydro)sphingosine-1-phosphate levels in INS-1 β cells. This increase was associated with an increase in sphingosine kinase 1 (SphK1) mRNA and protein levels. Over-expression of SphK1 in INS-1 cells potentiated palmitate-induced accumulation of dihydrosphingosine-1-phosphate. N,N-dimethyl-sphingosine, a potent inhibitor of SphK, potentiated β-cell apoptosis induced by palmitate whereas over-expression of SphK1 significantly reduced apoptosis induced by palmitate with high glucose. Endoplasmic reticulum (ER)-targeted SphK1 also partially inhibited apoptosis induced by palmitate. Inhibition of INS-1 apoptosis by over-expressed SphK1 was independent of sphingosine-1-phosphate receptors but was associated with a decreased formation of pro-apoptotic ceramides induced by gluco-lipotoxicity. Moreover, over-expression of SphK1 counteracted the defect in the ER-to-Golgi transport of proteins that contribute to the ceramide-dependent ER stress observed during gluco-lipotoxicity. In conclusion, our results suggest that activation of palmitate-induced SphK1-mediated sphingoid base-1-phosphate formation in the ER of β cells plays a protective role against palmitate-induced ceramide-dependent apoptotic β cell death.     

cAMP-secretion coupling is impaired in diabetic GK/Par rat β-cells: a defect counteracted by GLP-1

cAMP-raising agents with glucagon-like peptide-1 (GLP-1) as the first in class, exhibit multiple actions that are beneficial for the treatment of type 2 diabetic (T2D) patients, including improvement of glucose-induced insulin secretion (GIIS). To gain additional insight into the role of cAMP in the disturbed stimulus-secretion coupling within the diabetic β-cell, we examined more thoroughly the relationship between changes in islet cAMP concentration and insulin release in the GK/Par rat model of T2D. Basal cAMP content in GK/Par islets was significantly higher, whereas their basal insulin release was not significantly different from that of Wistar (W) islets. Even in the presence of IBMX or GLP-1, their insulin release did not significantly change despite further enhanced cAMP accumulation in both cases. The high basal cAMP level most likely reflects an increased cAMP generation in GK/Par compared with W islets since 1) forskolin dose-dependently induced an exaggerated cAMP accumulation; 2) adenylyl cyclase (AC)2, AC3, and G(s)α proteins were overexpressed; 3) IBMX-activated cAMP accumulation was less efficient and PDE-3B and PDE-1C mRNA were decreased. Moreover, the GK/Par insulin release apparatus appears less sensitive to cAMP, since GK/Par islets released less insulin at submaximal cAMP levels and required five times more cAMP to reach a maximal secretion rate no longer different from W. GLP-1 was able to reactivate GK/Par insulin secretion so that GIIS became indistinguishable from that of W. The exaggerated cAMP production is instrumental, since GLP-1-induced GIIS reactivation was lost in the presence the AC blocker 2',5'-dideoxyadenosine. This GLP-1 effect takes place in the absence of any improvement of the [Ca(2+)](i) response and correlates with activation of the cAMP-dependent PKA-dependent pathway.     

Comparison between D-[3-3H]- and D-[5-3H]glucose and fructose utilization in pancreatic islets from control and hereditarily diabetic rats

The metabolism of D-glucose and/or D-fructose was investigated in pancreatic islets from control rats and hereditarily diabetic GK rats. In the case of both D-glucose and D-fructose metabolism, a preferential alteration of oxidative events was observed in islets from GK rats. The generation of 3HOH from D-[5-3H]glucose (or D-[5-3H]fructose) exceeded that from D-[3-3H]glucose (or D-[3-3H]fructose) in both control and GK rats. This difference, which is possibly attributable to a partial escape from glycolysis of tritiated dihydroxyacetone phosphate, was accentuated whenever the rate of glycolysis was decreased, e.g., in the absence of extracellular Ca(2+) or presence of exogenous D-glyceraldehyde. D-Mannoheptulose, which inhibited D-glucose metabolism, exerted only limited effects upon D-fructose metabolism. In the presence of both hexoses, the paired ratio between D-[U-14C]fructose oxidation and D-[3-3H]fructose or D-[5-3H]fructose utilization was considerably increased, this being probably attributable, in part at least, to a preferential stimulation by the aldohexose of mitochondrial oxidative events. Moreover, this coincided with the fact that D-mannoheptulose now severely inhibited the catabolism of D-[5-3H]fructose and D-[U-14C]fructose. The latter situation is consistent with both the knowledge that D-glucose augments D-fructose phosphorylation by glucokinase and the findings that D-mannoheptulose, which fails to affect D-fructose phosphorylation by fructokinase, inhibits the phosphorylation of D-fructose by glucokinase.     

Development of beta-cell mass in fetuses of rats deprived of protein and/or energy in last trimester of pregnancy

Fetal malnutrition is now proposed as a risk factor of later obesity and type II diabetes. We previously analyzed the long-term impact of reduced protein and/or energy intake strictly limited to the last week of pregnancy in Wistar rats. Three protocols of gestational malnutrition were used: 1) low-protein isocaloric diet (5 instead of 15%) with pair feeding to the mothers receiving the control diet, 2) restricted diet (50% of control diet), and 3) low protein-restricted diet (50% of low-protein diet). Only isolated protein restriction induced a long-term beta-cell mass decrease. In the present study, we used the same protocols of food restriction to analyze their short-term impact (on day 21.5 of pregnancy) on beta-cell mass development. A 50% beta-cell mass decrease was present in the three restricted groups, but low-protein diet, either associated or not to energy restriction, increased fetal beta-cell insulin content. Among all the parameters analyzed to further explain our results, we found that the fetal plasma level of taurine was lowered by low-protein diet and was the main predictor of the fetal plasma insulin level (r = 0.63, P < 0.01). In conclusion, rat fetuses exposed to protein and/or energy restriction during the third part of pregnancy have a similar dramatic decrease in beta-cell mass, and their ability to recover beta-cell mass development retardation depends on the type of malnutrition used. Moreover, our results support the hypothesis that taurine might play an important role in fetal beta-cell mass function.     

Undernutrition does not alter the activation of beta-cell neogenesis and replication in adult rats after partial pancreatectomy

In previous work, we demonstrated that a 65% protein calorie food restriction started during the third trimester of gestation in rats caused a reduced beta-cell mass at 4 days of life that persisted until adult age. In this study with adult undernourished (U) rats, we investigated 1) whether undernutrition affects the beta-cell growth potential and both beta-cell proliferation and differentiation and 2) the implication of the IGFs, highly responsive to nutritional status, in these processes. To this end, we used the 90% pancreatectomy (Px) procedure in U and control (C) adult rats. The results show that, on day 2 after Px, beta-cell replication was significantly higher in C rats, whereas the beta-cell neogenesis was markedly increased in U/Px rats. Both the serum levels of IGF-I and the liver IGF-I mRNA expression were reduced in adult U rats before and after Px compared with C rats. Pancreatic IGF-I mRNA expression was reduced in U animals on day 0. However, on day 2 after Px, the increase of pancreatic IGF-I mRNA expression was significantly higher in U rats than in C rats. These data suggest that beta-cells still have the capacity to regenerate in the adult U rats, with a higher efficiency than C rats on day 2, and that both beta-cell neogenesis and beta-cell replication are stimulated. The increased pancreatic IGF-I mRNA may be instrumental in these processes.     

Glucose-induced insulin mRNA accumulation is impaired in islets from neonatal streptozotocin-treated rats.

According to the "glucose toxicity" hypothesis, hyperglycemia contributes to defective beta-cell function in type 2, non-insulin-dependent diabetes mellitus. This concept is supported by substantial data in rodent models of diabetes. However, the ability of glucose to stimulate the accumulation of insulin mRNA, a critical feature of normal beta-cell physiology, has not been investigated in in vivo models with chronic hyperglycemia. The aim of this study was to determine whether glucose-induced insulin mRNA accumulation is impaired in the neonatal streptozotocin-treated rat (n0-STZ rat), a model of non-obese, non-insulin-dependent diabetes mellitus. Islets of Langerhans isolated from n0-STZ and control rats were cultured for 24 h in the presence of 2.8 or 16.7 mmol/l glucose, and insulin mRNA levels were measured by Northern analysis. Insulin mRNA levels were increased more than twofold by glucose in control islets. In contrast, no significant effect of glucose was found on insulin mRNA levels in n0-STZ islets. We conclude that insulin gene regulation by glucose is impaired in n0-STZ rat islets.     

Effect of early dietary restriction on insulin action and secretion in the GK rat, a spontaneous model of NIDDM

The availability of the Goto-Kakisaki (GK) rat model of non-insulin-dependent diabetes mellitus prompted us to test the effect of a limited period of undernutrition in previously diabetic young rats on their insulin secretion and insulin action during adult age. Four-week-old female GK rats were either food restricted (35% restriction, 15% protein diet) or protein and energy restricted (35% restriction, 5% protein diet) for 4 wk. Food restriction in the young GK rat lowered weight gain but did not aggravate basal hyperglycemia or glucose intolerance, despite a decrease in basal plasma insulin level. Furthermore, the insulin-mediated glucose uptake by peripheral tissues in the GK rat was clearly improved. We also found that food restriction, when it is coupled to overt protein deficiency in the young GK rat, altered weight gain more severely and slightly decreased basal hyperglycemia but conversely aggravated glucose tolerance. Improvement of basal hyperglycemia was related to repression of basal hepatic glucose hyperproduction, despite profound attenuation of basal plasma insulin level. Deterioration of tolerance to glucose was related to severe blunting of the residual glucose-induced insulin secretion. It is, however, likely that the important enhancement of the insulin-mediated glucose uptake helped to limit the deterioration of glucose tolerance.     

Energy restriction with protein restriction increases basal metabolism and meal-induced thermogenesis in rats

We previously observed an increased sympathetic nervous system (SNS) activity that was partly responsible for a defect in the insulin secretion response to glucose after postweaning protein-energy restriction (PER) in female rats. These results, together with other data on low-protein feeding, suggested that a low protein-to-energy ratio (P/E) in the diet could stimulate energy expenditure (EE), but direct measurements of EE have never been reported under conditions of PER. The goal of the present study was thus to quantify the changes induced by PER to body composition, the various parameters of EE, and plasma triiodothyronine levels. PER induced severe growth retardation, but the subcutaneous white and interscapular brown adipose tissue masses were preserved. Basal metabolism, meal-induced thermogenesis, and triiodothyronine levels were increased, but substrate utilization by the working muscles was unaffected. Meal-induced thermogenesis was increased by spontaneous activity in PER rats only. These results suggest that rats adapt to a low P/E in the diet by burning part of their excess nonprotein energy and storing the remaining excess in subcutaneous adipose tissue.     

Insulin binding and insulin receptor tyrosine kinase activity are not altered in the liver of rats with non-insulin-dependent diabetes

Using the insulin-glucose clamp technique, we have previously shown that an increased sensitivity to insulin in vivo is a characteristic of the liver in rats with non-insulin-dependent diabetes induced by neonatal streptozotocin administration. We have thus studied the properties of liver insulin receptor in that model. 125I-porcine insulin binding was found normal both in isolated plasma membranes and in solubilized, wheat germ agglutinin purified receptors prepared from livers of rats with non-insulin-dependent diabetes, when compared to controls. Basal and insulin-stimulated insulin receptor kinase activities were also found normal for both the autophosphorylation of the beta subunit of the insulin receptor and the phosphorylation of the artificial substrate poly (Glu-Tyr) 4:1. Thus, in that model of chronic insulin deficiency and mild hyperglycemia: 1) liver insulin receptors are not up-regulated; 2) tyrosine kinase activity remains unaffected. This last observation supports the hypothesis that the increased insulin effect in the liver of rats with non-insulin-dependent diabetes is probably distal to the insulin receptor kinase.