Do cell membrane dynamics participate in insulin resistance?

Fat feeding potentiates the diabetogenic effect of dexamethasone in Wistar rats

Background

The role of cortisol and its increased action/availability is implicated in the pathogenesis of insulin resistance associated with obesity and metabolic syndrome but the mechanism of increased action/availability is not known. Availability of several other lipophilic hormones, drugs and pollutants are also reported to be increased in obesity. Increased lipids in the circulation are reported to alter the fluidity and permeability of membranes. Hyperlipidemia is also reported to alter the pharmacokinetics and pharmacodynamics of lipophilic molecules and also membrane fluidity and permeability. In this context we assumed that the hyperlipidemia associated with human obesity might play a role in the altered action/availability of cortisol and this in turn might have initiated the metabolic complications. To evaluate our assumption we have administered dexamethasone [low [50 μg/kg/day] or high [250 μg/kg/day] dose] to high-fat [coconut oil & vanaspati] fed rats and the results were compared with rats administered with either dexamethasone or high-fat.

Results and Discussion

Within two weeks, the rats co-administered with high-fat and dexamethasone developed severe hyperglycemia, hyperlipidemia and insulin resistance compared to rats treated either of them alone. High-fat fed rats treated with higher dose of dexamethasone were presented with severe hyperglycemia, insulin resistance and also severe glycosuria. The hyperlipidemia caused by high-fat feeding might have altered the transport and distribution of dexamethasone, probably by altering the physical state of membranes and transport proteins.

Conclusion

From the results obtained, it can be speculated that the altered lipid and cortisol metabolism could affect one another, forming a vicious cycle.

L-Carnitine inhibits protein glycation in vitro and in vivo: evidence for a role in diabetic management

Glycation-initiated changes in tissue proteins are suggested to play an important role in the development of diabetes-related pathological changes. The purpose of this study was to examine the anti-glycating effect of L-carnitine (CA) in vivo in the high-fructose diet-fed rat and to determine the potential of CA to inhibit in vitro glycation. Additionally the glucose-disposal efficiency of CA in the rat diaphragm was investigated. High-fructose diet (60 g/100 g diet)-fed rats were treated with CA (300 mg/kg/day i.p.) for 60 days. The effect of CA on glucose, fructose and fructosamine in plasma, methyl glyoxal and glycated haemoglobin in whole blood and skin and tail tendon collagen glycation were determined. The inhibitory effect of CA on the glycation of bovine serum albumin in vitro was compared with that of aminoguanidine (AG), a known antiglycation agent. Glucose utilisation induced by insulin in the control rat diaphragm was monitored in the presence and absence of CA. High-fructose feeding induced hyperglycaemia and glycation of haemoglobin and skin and tail tendon collagen. In CA-administered fructose-fed rats glycation was significantly reduced. In vitro glycation and accumulation of advanced glycation end products were mitigated by CA. CA was more effective than AG in inhibiting glycation in vitro. CA also enhanced the utilisation of glucose in the rat diaphragm. The findings of the study reveal that CA not only has antiglycation effect but also enhances glucose disposal in the rat diaphragm. These findings provide evidence for the therapeutic utility of CA in diabetes and associated complications.

Sphingomyelin/cholesterol ratio: an important determinant of glucose transport mediated by GLUT-1 in 3T3-L1 preadipocytes

Sphingomyelin pathway has been linked with insulin signaling through insulin-dependent GLUT-4 glucose transporter, but a relationship between sphingomyelin and the GLUT-1 transporter responsible for the basal (insulin-independent) glucose transport has not been clearly established. As GLUT-1 is mainly distributed to the cell surface, we explored the effects of changes in membrane sphingomyelin content on glucose transport through GLUT-1. The addition of exogenous sphingomyelin or glutathione (an inhibitor of endogenous sphingomyelinase) to the culture medium increased membrane sphingomyelin and cholesterol contents. Basal glucose uptake was enhanced and positively correlated to sphingomyelin (SM), cholesterol (CL) and SM/CL ratio. The exposure of 3T3-L1 preadipocytes to sphingomyelinase (SMase) significantly increased basal glucose uptake, membrane fluidity and decreased membrane sphingomyelin and cholesterol contents 60 min after SMase addition. There was no significant change in the abundance of GLUT-1 at the cell surface. The membrane sphingomyelin and cholesterol contents, fluidity and basal glucose transport returned to baseline levels within 2 h. The basal glucose uptake was negatively correlated with cholesterol contents and positively with SM/CL ratio. The SM/CL ratio might represent an important parameter controlling basal glucose uptake and a mechanism by which insulin resistance might be induced.

Adipocyte and erythrocyte plasma membrane phospholipid composition and hyperinsulinemia: a study in nondiabetic and diabetic obese women

BACKGROUND

The cell functions involved in the action of insulin--receptor binding, enzyme and transporter activities--are controlled by membrane properties. We have previously shown that the fasting plasma insulin (FPI) concentration and the homeostasis model assessment (HOMA) estimate of insulin resistance are associated with the sphingomyelin concentration in the erythrocyte membranes of obese women.

OBJECTIVES

(1) To study the distribution of phospholipid classes in the plasma membrane and their association with insulin resistance markers in the adipocyte, an insulin-sensitive cell in obese women. (2) To investigate the influence of diabetes in a small group of obese women treated by diet alone. (3) To compare the distribution of phospholipids in erythrocyte membranes in a subgroup of obese nondiabetic and diabetic women.

SUBJECTS

Subcutaneous fat biopsies were taken from the abdominal region of 19 obese non-diabetic and seven obese type 2 diabetic women. Erythrocyte membrane assessment was performed in a subgroup of 10 of the 19 obese nondiabetic and in the seven diabetic patients.

METHODS

The phospholipid composition of adipocyte and erythrocyte plasma membranes was analyzed by high performance liquid chromatography.

RESULTS

FPI was positively correlated with the adipocyte membrane contents of sphingomyelin (P < 0.001), phosphatidylethanolamine (P < 0.05), and phosphatidylcholine (P < 0.01) in the obese nondiabetic women. Similar correlations were obtained with HOMA. A stepwise multiple regression analysis indicated that sphingomyelin accounted for 45.6 and 43.8% of the variance in FPI and HOMA values as an independent predictor. There was a similar positive independent association between FPI and SM in the erythrocyte membranes of the studied subgroup. Diabetes per se did not influence the independent association between SM membrane contents and FPI in both cell types.

CONCLUSION

These results suggest a link between membrane phospholipid composition, especially SM, and hyperinsulinemia in obese women.

Adipocyte membrane phospholipids and PPAR-gamma expression in obese women: relationship to hyperinsulinemia

We have shown that membrane sphingomyelin (SM) is an independent predictor of the variance of fasting plasma insulin (FPI) concentrations and the homeostasis model assessment (HOMA) estimate of insulin resistance in obese women. The peroxisome proliferator-activated receptor-gamma (PPAR-gamma) is a key component in adipocyte differentiation that may also contribute to the sensitivity of cells to insulin. PPAR-gamma is activated by fatty acids, and the membrane composition may have an impact on the activity of PPAR-gamma and thus on the sensitivity of adipocytes to insulin. We investigated these possible links by determining the phospholipid contents of adipocyte membranes, the mRNA expression of PPAR-gamma, and the FPI and HOMA estimate of insulin resistance in obese women. The mRNA levels of tumor necrosis factor-alpha (TNF-alpha), which is suspected to play a role in insulin resistance and which downregulates PPAR-gamma expression, were also quantified. FPI and HOMA were strongly positively correlated with membrane SM (P < 0.005) and cholesterol (P < 0.005). PPAR-gamma mRNA levels were negatively correlated with FPI (P < 0.05) and HOMA (P < 0.05) and positively correlated with high-density lipoprotein (HDL) cholesterol (P < 0.05), membrane SM (P < 0.05), and cholesterol contents (P < 0.05). TNF-alpha mRNA levels were not correlated with membrane parameters. In stepwise multiple regression analysis, the variations in PPAR-gamma mRNA levels were mainly explained by HDL cholesterol (31.9%) and membrane SM (17.7%). Our study shows that the expression of PPAR-gamma, a major factor controlling adipocyte functions, the lipid composition of the membrane, and insulin sensitivity are probably closely associated in the adipose tissue of obese women.

Perinatal consequences of maternal-fetal Rh blood group interaction in diabetic pregnancy: a nonimmunological perspective

BACKGROUND

The recent discoveries about the structure of Rh protein that suggest a transport function and the recent observations of a positive correlation between Rh(D) protein and glycosylated hemoglobin levels in non-insulin-dependent diabetes mellitus prompted us to review our data on diabetic pregnancy to evaluate the perinatal consequences of maternal-fetal Rh blood group interactions in a metabolic perspective.

SUBJECTS AND METHODS

One hundred thirty-two women with gestational diabetes and 120 women with preexisting insulin-dependent diabetes mellitus were examined. Three hundred eighty-seven consecutive nondiabetic puerperae from the same population were considered control subjects.

RESULTS

In both gestational and insulin-dependent diabetes mellitus, an increased proportion of mother Rh(+)/newborn Rh(-) and a decreased proportion of mother Rh(-)/newborn Rh(+) joint phenotype has been observed. No deviation has been observed for joint phenotypes in which mother and newborn are similar [ie, Rh(+)/Rh(+) and Rh(-)/Rh(-)]. In the situation of mother Rh(+)/newborn Rh(-), there is a relatively lower rate of fetal loss and a decreased tendency to high birth weight. On the contrary, in pairs mother Rh(-)/newborn Rh(+) the fetus shows an increase of fetal loss and of tendency to high birth weight.

CONCLUSIONS

The results are compatible with the hypothesis that when the density of Rh protein in the mother is higher than that in the fetus, the conceptus is relatively protected against the toxic effect of glucose. In the opposite genotypic combination (ie, density of Rh protein higher in the fetus than in the mother), the fetus is relatively more susceptible to these effects.

Metformin modulates insulin receptor signaling in normal and cholesterol-treated human hepatoma cells (HepG2)

The effects of the biguanide anti-hyperglycemic agent, metformin (N,N'-dimethyl-biguanide), on insulin signaling was studied in a human hepatoma cell line (HepG2). Cells were cultured in the absence (control cells) or in the presence of 100 microM of a cholesterol derivative, hemisuccinate of cholesterol. Cholesterol hemisuccinate-treatment alters cholesterol and lipid content of HepG2 and modulates membrane fluidity. Cholesterol hemisuccinate-treatment induces a decrease in insulin responsiveness and creates an 'insulin-resistant' state in these cells. Exposure to 100 microM of metformin resulted in a significant enhancement of insulin-stimulated lipogenesis in control and cholesterol hemisuccinate-treated cells. In control cells, metformin altered glycogenesis in a biphasic manner. In cholesterol hemisuccinate-treated cells, metformin inhibited basal glycogenesis but restored insulin-stimulated glycogenesis. Hence, to understand the mechanism of metformin action, we analyzed early steps in the insulin signaling pathway, including insulin receptor autophosphorylation, mitogen-activated-protein kinase and phosphatidylinositol 3-kinase activities, in both control and cholesterol hemisuccinate-treated cells. Overall, the results suggest that metformin may interact with the insulin receptor and/or a component involved in the early steps of insulin signal transduction.

Effect of feeding thoroughbred horses a high unsaturated or saturated vegetable oil supplemented diet for 6 months following a 10 month fat acclimation

This study looked at the effect of feeding diets supplemented with either a predominantly saturated or unsaturated vegetable oil over a prolonged period to exercising horses. Eight Thoroughbred horses were assigned to 2 diet treatments and for 10 months were fed Timothy hay and oats, together with a fortified sweet feed supplemented with either a predominantly unsaturated (Un) or a saturated (S) vegetable oil so that approximately 19% DE (Digestible Energy) came from dietary fat and approximately 12% from either the Un or S source (AC). An increased amount of Un or S fortified sweet feed, replacing the oats, was then fed for a further 6 months (HF) so that approximately 27% DE came from fat and approximately 20% from the Un or S vegetable oil. Standardised incremental treadmill exercise (8-12 m/s) tests (STEP) and duplicate oral glucose tolerance tests (TOL) were carried out after 3, 6 and 9 months of the AC diet and after 3 and 6 months on the HF diet. There was no significant effect of dietary treatment or when the tests were undertaken (time) on the insulin or lactate responses to the STEP tests. Overall there was a significant (P < 0.05) effect of time and treatment on the glucose response, but there was no difference between treatments at the first and last tests or between the results for these tests or between the endAC and endHF tests. No significant effect of treatment or time was seen on the TOL glucose response (% change from Time '0') although there was a trend for the glucose concentrations to be lower and the insulin responses higher (nonsignificant) in the S treatment group. No significant effect of treatment on haematological parameters, monitored monthly, was found. Total protein and gamma glutamyl transferase remained within the normal range throughout. There was a significant effect of treatment (P < 0.05) on cholesterol and triglycerides with higher concentrations in the S group from the first (1 month) sample. Linoleic acid was the main fatty acid in all the 4 plasma lipid classes with slightly, but significant (P < 0.05), higher concentrations in Un for the cholesterol ester and phospholipid classes. There was no effect of time. Overall, the total resting plasma fatty acid content was significantly higher (P < 0.05) with S at the sample points (endAC and endHF). No adverse effects of feeding either diet on apparent coat condition or hoof appearance were seen apart from an apparent increase in the grease score. Many of the parameters assessed showed significant improvements with time (P < 0.05). In conclusion, no apparent adverse effects of feeding a diet supplemented with either an unsaturated or saturated vegetable oil for 6 months at approximately 20% DE after 10 months at approximately 12% DE were identified and there were no apparent disadvantages of feeding a saturated vegetable oil supplemented diet compared with an unsaturated one.

Effect of pravastatin on erythrocyte rheological and biochemical properties in poorly controlled Type 2 diabetic patients

AIMS

The rheological properties of erythrocytes are impaired in diabetes mellitus, especially because of changes in their membrane lipid composition. In hypercholesterolaemic patients, lowering plasma cholesterol is associated with an improvement of the erythrocyte rheological parameters. The aim of this study was to investigate the relationship between erythrocyte deformability, plasma lipids, lipid membrane composition and cytosolic cations in poorly controlled Type 2 diabetic patients and to test the effects of a cholesterol-lowering treatment on these parameters.

METHODS

We compared 37 poorly controlled Type 2 diabetic patients with 26 controls. In 22 of the diabetic patients who showed an impairment in erythrocyte deformability (filtration index >10.5 on the Hanss' haemorheometer), a double-blind randomized trial compared the effect of the inhibitor of HMG CoA reductase pravastatin 20 mg per day for 4 months vs. placebo on the erythrocyte parameters.

RESULTS

Compared with controls, diabetic patients had higher filtration index (FI), erythrocyte sodium and calcium contents and lower free cholesterol-phospholipids ratio in erythrocyte membranes. Erythrocyte sodium content correlated positively with the FI and the membrane free cholesterol-phospholipids ratio. In the pravastatin-treated group (11 patients), fibrinogen decreased significantly, FI reached a normal value (<10) in six patients. Four of the five other patients who still had abnormal FI after 4 months of treatment had either a high plasma triglycerides (> or =4.60 mmol/l) or a high plasma fibrinogen (> or =4 g/l) level at baseline. Only two of the 11 placebo-treated patients achieved a normal FI.

CONCLUSION

These data suggest that in poorly controlled Type 2 diabetic patients there is a link between the chemical composition and the rheological properties of erythrocytes. Erythrocyte deformability may be improved by lowering plasma cholesterol with a statin.

Diabetes mellitus: polymorphonuclear leukocyte (PMN) filtration parameters and PMN membrane fluidity after chemotactic activation

The goal of this research was to determine leukocyte rheology at baseline and after chemotactic activation in type I and type II diabetics. In 19 normal subjects, 21 type I diabetics, and 16 type II diabetics at baseline and after in vitro chemotactic activation (prolonged for 5 and 15 minutes) with two stimulating agents (4-phorbol 12-myristate 13-acetate [PMA] and N-formyl-methionyl-leucyl-phenylalanine [fMLP]), we evaluated polymorphonuclear (PMN) filtration parameters (using a St. George filtrometer [Carri-Med, Dorking, UK] and considering the initial relative flow rate [IRFR] and the concentration of clogging particles [CP]) and PMN membrane fluidity (obtained by marking PMNs with the fluorescent probe 1-(4-[trimethylamino]phenyl)-6-phenyl-1,3,5-hexatriene (TMA-DPH). At baseline, there was a difference between normals and type I and II diabetics for PMN membrane fluidity only. After activation in normals and diabetics of both types, a significant variation was present in PMN filtration parameters (IRFR and CP) at both 5 and 15 minutes. In normals, no variation was present in PMN membrane fluidity after activation with PMA or fMLP. After PMN activation, only in type I diabetics was a significant decrease in PMN membrane fluidity present at both 5 and 15 minutes. After PMN activation with either PMA or fMLP in comparison to basal values, only the mean variation (delta%) of the IRFR was significantly different between normals, type I diabetics, and type II diabetics at both 5 and 15 minutes. From the data obtained, it is evident that after activation, the PMN filtration pattern shows a specific behavior in diabetics of both types, while PMN membrane fluidity changes only in type I diabetics. The latter finding may be the basis of a metabolic pattern present in PMNs of this type, revealed after in vitro activation.