| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Endocrinology and Metabolism (M.J.S., G.A., H.P.S.), Medical Biochemistry (J.E.G., J.M.A., A.J.M.), and Clinical Chemistry, Laboratory of Endocrinology (M.T.A., B.C.V.), Academic Medical Center, 1105 AZ Amsterdam, The Netherlands; and Department of Internal Medicine, Hospital Gelderse Vallei (R.H.), 6710 HN Ede, The Netherlands
Address all correspondence and requests for reprints to: M. J. M. Serlie, Academic Medical Center, F5-169, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands. E-mail: m.j.serlie{at}amc.uva.nl.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: We studied the protective effect of pioglitazone on FFA-induced insulin resistance and the effects on intramyocellular glycosphingolipids.
Design: We studied glucose metabolism in the basal state and during a hyperinsulinemic euglycemic clamp by using stable isotopes. Studies were performed at baseline and after 4 months of treatment with pioglitazone. Patients were then studied on a third occasion during infusion of a lipid emulsion to increase plasma FFA to pretreatment levels. All studies were combined with muscle biopsies to measure intramyocellular ceramide and glycosphingolipids.
Patients: Patients were obese with poorly controlled type 2 diabetes mellitus.
Intervention: Patients were treated with 30 mg pioglitazone once daily.
Main Outcome Measure: The change in peripheral insulin sensitivity after treatment with pioglitazone and during the infusion of the lipid emulsion was the main outcome measure.
Results: Peripheral glucose uptake (Rd) increased significantly, but returned to baseline levels after increasing plasma FFA to pretreatment levels. Insulin-mediated suppression of FFA was increased significantly. Intramyocellular ceramide concentrations were higher during the hyperinsulinemic clamp after treatment with pioglitazone, but not in the basal state. The intramyocellular content of glycosphingolipids and plasma concentrations of ceramide and glycosphingolipids did not change.
Conclusions: Pioglitazone increases Rd and insulin-mediated suppression of plasma FFA, but does not protect patients with type 2 diabetes mellitus from FFA-induced insulin resistance. This effect of pioglitazone is not attained via a decrease in intramyocellular concentrations of ceramide or glycosphingolipids.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Recently, intramyocellular ceramide was found to be increased in obese patients with T2DM. The concentration was positively correlated with plasma FFA concentrations (6) and negatively correlated with whole body insulin sensitivity (7). Ceramide negatively influences insulin signaling by activation of PKC (4).
Not only ceramide itself, but also gangliosides (formed from ceramide) may influence insulin sensitivity, as was shown in GM3 synthase knockout mice, which show a normal phenotype compared with their littermate control mice, but have enhanced insulin sensitivity and are protected from obesity-induced insulin resistance (8).
Thiazolidinediones (TZDs) increase peripheral insulin sensitivity via a partially elucidated mechanism. One of the major mechanisms in their insulin-sensitizing action is probably via lowering of plasma FFA. The expression of peroxisome proliferator-activated receptor 
(PPAR) in skeletal muscle is low compared with the expression in adipose tissue (9). Therefore, one may hypothesize that the main insulin- sensitizing action of these drugs is by manipulating adipose tissue. Treatment with PPAR agonists induces proliferation of smaller, more insulin-sensitive adipocytes (10) and different adipose tissue distribution (11), thus resulting in lower plasma FFA concentrations (12). Furthermore, in a recent study, it was shown that TZDs can exert their glucose-lowering effects only in the presence of adipose tissue (13) and still maintain their antidiabetic potential in the absence of muscle PPAR expression (14).
Several animal studies have shown the protecting effect of TZDs on the development of insulin resistance induced by a high-fat diet or by acute elevation of plasma FFA (15, 16), but surprisingly, this was not seen in a study with obese normal glucose-tolerant humans (17). One of the explanations for the difference between these studies can be a species difference. Another explanation can be related to the study design. The plasma FFA concentrations in that study were high (707 ± 121 µmol/liter), potentially overruling the TZD effect. Besides, time of treatment with TZDs was only 6 wk, whereas the maximum effect of TZDs in humans is reached after 14 wk (18).
To study the protective effect of pioglitazone on FFA-induced insulin resistance in a highly prevalent group of patients using an experimental design that allows translation of results to daily patient care, we studied glucose metabolism in obese patients with T2DM before and after treatment for 4 months with pioglitazone and during infusion of a lipid emulsion to increase plasma FFA to pretreatment levels. To gain more insight into the possible underlying pathophysiological mechanism, we also studied the effects of pioglitazone on levels of ceramide and glycosphingolipids in skeletal muscle and plasma.
We hypothesized that treatment with pioglitazone would not protect obese patients with T2DM from FFA-induced insulin resistance. We also hypothesized that one of the insulin-sensitizing mechanisms of pioglitazone is through a decrease in ceramide content in skeletal muscle and/or by changing the intramyocellular glycosphingolipids.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We recruited overweight [body mass index (BMI) > 27 kg/m2] patients (n = 8) from three different hospitals in The Netherlands with moderately controlled T2DM (glycosylated hemoglobin > 7%) without diabetes-related complications (macrovascular or microvascular complications, congestive heart failure) and on oral medication only to treat their T2DM. If patients were on metformin therapy, this drug was stopped and substituted by sulfonylurea derivate monotherapy during the study with a washout and run-in period of 2 wk. This was done because of the evaluation of gluconeogenesis (GNG) after treatment with pioglitazone. Metformin decreases GNG (19), which would lead to a potential overestimation of the effect of pioglitazone on GNG in the metformin-treated patients. Patients were allowed to take statins (but not fibrates) and antihypertensive drugs. After recruitment, a physical examination was performed together with the medical history. All patients served as their own control. The dosage of pioglitazone (Lilly Nederland BV, Houten, The Netherlands) was 30 mg once daily.
We studied the subjects on three occasions: at baseline, after 4 months of treatment (T1), and on a third occasion (after 5 months of treatment, T2) with the same protocol, but together with infusion of a lipid emulsion for 6 h. To rule out possible confounding effects of duration of treatment on T1 and T2, the lipid infusion study was carried out in balanced assignment between T1 and T2. The patients visited the Academic Medical Center every month to control their plasma glucose, liver enzymes, renal function (creatinine), and the presence of congestive heart failure. If patients had severe hyperglycemia (glucose > 12 mmol/liter) with or without complaints, the sulfonylurea derivative was increased to maximum dose.
All studies were performed after an overnight fast at the metabolic unit of the Academic Medical Center. Patients consumed approximately 250 g of carbohydrates during 3 d before the study and were asked to refrain from vigorous exercise.
The medical ethical committee of the Academic Medical Center approved the study protocol, and all participants signed a written informed consent after the nature of the study was explained.
Isotope studies and hyperinsulinemic euglycemic clamp
[6,6 2H2]Glucose was used to measure endogenous glucose production (EGP) and insulin-stimulated peripheral glucose uptake based on the principle of the isotope dilution technique. Deuterated water (2H2O) was used to determine GNG. All patients were studied in the supine position. The day before the study, venous blood samples were taken for background 2H-enrichment of water and 2H-enrichment at the position of C5 of glucose. Patients then ingested 5 g 2H2O per kilogram of body water divided into five doses, which were ingested at 2000, 2030, 2100, 2130, and 2200 h. Body water was estimated to be 50% of body weight for women and 60% of body weight for men. The patients were allowed to drink water ad libitum but enriched with 0.5% 2H2O to maintain isotopic steady state.
At 0700 h, after the subjects had fasted overnight (for 12 h), a catheter was inserted into a distal vein of each arm or a dorsal vein of the hand. One catheter was used to sample arterialized blood with use of a heated hand box (60 C). The other catheter was used to infuse [6,6-2H2]glucose, a 20%-glucose solution and insulin. After a blood sample was taken for background enrichment of plasma glucose, a primed, continuous infusion of [6,6-2H2]glucose (>99% enriched; Cambridge Isotope Laboratories, Cambridge, MA) at a rate of 0.22 µmol·kg–1·min–1 (prime: 17.6 µmol/kg) was started. After 180-min equilibration time, three blood samples were drawn for measurements of isotopic enrichments of glucose, glucoregulatory hormones, total adiponectin, FFA, soluble TNF
-receptors I and II (sTNF-RI and RII), ceramide, and glucosylceramide. Thereafter, an insulin infusion (100 kU Actrapid per liter; Novo Nordisk Farma B.V., Alphen aan de Rijn, The Netherlands) was started at a rate of 60 mU·m body surface area–2·min–1. Plasma glucose concentrations were measured every 5 min with a Glucose Analyzer 2 (Beckman, Palo Alto, CA), and a 20%-glucose solution was infused at a variable rate to maintain euglycemia at 5.0 mmol/liter. [6,6-2H2]Glucose was added to the 20% glucose infusate to approximate the values for enrichment reached in plasma and to prevent negative peripheral glucose uptake (Rd) artifacts during the clamp. After 320 min of hyperinsulinemic euglycemia, five blood samples were drawn at 10-min intervals for measurements of the isotopic enrichment of glucose and earlier mentioned plasma parameters.
On T2, the same protocol was conducted for basal glucose kinetics, but at the start of the clamp, a concomitant infusion of Intralipid 20% (Fresenius Kabi Nederland B.V., Den Bosch, The Netherlands) without heparin was started at a variable rate to maintain plasma FFA at approximately 0.2 mmol/liter. This was achieved by measuring plasma FFA every 30 min and by adjusting the lipid infusion rate. After 320 min of insulin and Intralipid infusion, five blood samples were drawn for the same plasma measurements as described earlier.
Body composition
Body composition was measured with bioimpedance analysis (Maltron BF-906; Maltron International Ltd., Essex, UK).
Muscle biopsy
In the basal state and after 6 h of the hyperinsulinemic euglycemic clamp, muscle biopsies were taken from the musculus vastus lateralis under local anesthesia with Lidocaine 20%. Biopsies were performed using an automatic biopsy instrument (Pro-Mag I 2.5; Medical Device Technologies Inc., Gainesville, TX). The biopsies were washed with HEPES to reduce blood contamination and thereafter immediately frozen in liquid nitrogen.
Analytical procedures
Insulin and cortisol were determined with an Immulite 2000 system (Diagnostic Products Corporation, Los Angeles, CA). Insulin was measured by a chemiluminiscent immunometric assay: intraassay variation 47 pmol/liter 6%, 609 pmol/liter 3%; interassay variation 91 pmol/liter 4%, 120 pmol/liter 6%; detection limit 15 pmol/liter. Cortisol was measured by a chemiluminescent immunoassay: intraassay variation 89 nmol/liter 8%, 500 nmol/liter 7%; interassay variation 136 nmol/liter 8%, 1092 nmol/liter 7%; detection limit 50 nmol/liter. Plasma FFA were measured by an enzymatic method (NEFAC, Wako Chemicals, Richmond, VA). Intraassay variation was 0.22 mmol/liter 1%, 0.93 mmol/liter 1%; interassay variation 0.01 mmol/liter 15%, 0.48 mmol/liter 4%; detection limit 0.02 mmol/liter. Glucagon was determined with a 125I RIA (LINCO Research, St. Charles, MO). Intraassay variation was 71 ng/liter 10%, 147 ng/liter 9% and interassay variation was 84 ng/liter 5%, 192 ng/liter 7%; detection limit 15 ng/liter. Plasma adiponectin concentrations were measured by a RIA kit (LINCO Research). Intraassay variation was 3.0 µg/ml 2%, 8.5 µg/ml 7%; and interassay variation was 3.4 µg/ml 16%, 10.1 µg/ml 17%; detection limit 1.0 µg/ml. sTNF-RI and sTNF-RII were determined with an EASIA (Biosource Europe S.A., Nivelles, Belgium), sTNF-RI: intraassay variation 2.31 ng/ml 7%; interassay variation 1.76 ng/ml 6%, 26.80 ng/ml 9%; detection limit 0.05 ng/ml, sTNF-RII: intraassay variation 5.76 ng/ml 8%; interassay variation 3.25 ng/ml 9%, 18.70 ng/ml 7%; detection limit 0.1 ng/ml.
[6,6-2H2]Glucose enrichment and 2H-plasma water enrichment and 2H-enrichment on the C5 position of glucose were measured as described earlier (20). [6,6-2H2]Glucose enrichment was as follows (tracer to tracee ratio): intraassay variation 1.42, 1, 3.46, and 0.5%; and interassay variation 1.47 and 1%, 3.56 and 1%; detection limit 0.04%. Plasma water enrichment was as follows (tracer to tracee ratio): intraassay variation 0.51 and 6%, 0.34 and 5% and interassay variation 0.51 and 6%, 0.32 and 5%; detection limit 0.04%. C5 enrichment was as follows (tracer to tracee ratio): intraassay variation 0.26 and 8% and interassay variation 0.29 and 6%, detection limit 0.04%.
Glycosphingolipid and ceramide measurements
Glycosphingolipids and ceramide in plasma and muscle biopsies were measured with a HPLC method by a modification of the method described by Taketomi et al. (21). Muscle biopsies were weighed and homogenized in 300 µl water by sonification. A total of 1 nmol C18-sphinganine was added to 50 µl plasma or muscle homogenates as internal standard. Lipids were extracted according to Folch et al. (22). The lipids were hydrolyzed in borosilicate glass tubes (Schott GL14) (12 x 100 mm) with polytetrafluoretheen-lined screw caps in 0.5 ml of freshly prepared 0.1 M NaOH in methanol, using the CEM microwave Solids/Moisture System SAM-155 oven, equipped with a rotating Teflon tray with 36 tube holes, 60 min at 85% of maximum power. Deacylated glycosphingolipids and sphingoid bases were derivatized with 25 µl o-phthalaldehyde reagent as described by Merrill et al. (23) with a slight modification. O-phthalaldehyde-derivatized sphingoid bases and lyso-glycosphingolipids were separated using an HPLC system (Waters Associates, Milford, MA) with a Altima BDS C18 3µ, 150 x 4.6 mm reverse phase column and a methanol to water ratio of 88:12 (wt/wt as eluent). All samples were run in duplicate, and in every run, two reference samples were included. Coefficient of variation was as follows: interassay 4%, intraassay less than 14% (Groener, J. E., B. J. Poorthuis, S. Kuiper, M. T. Helmond, and J. M. Aerts, unpublished observations).
Calculations and statistics
EGP (Ra) and Rd were calculated using the modified form of the Steele equations as described previously (24). Fractional GNG was calculated by dividing the deuterium enrichment of the C5 position of glucose by deuterium enrichment of body water. GNG was calculated by multiplying fractional GNG with endogenous glucose production.
The index of hepatic insulin resistance was calculated by multiplying basal endogenous glucose production with fasting plasma insulin (25). Patients were their own controls; therefore differences between baseline, T1, and T2 were calculated using the nonparametric Wilcoxon Signed Ranks test. Data are presented as medians (ranges). Multivariable linear regression analysis was performed to identify independent variables of Rd at baseline. The correlation between these variables is expressed as Spearman’s rank correlation coefficient (
). The absolute change in muscle ceramide, glucosylceramide, and lactosylceramide after treatment was correlated to the absolute change in Rd with multivariable analysis. This was done in six patients because insufficient material was obtained in two subjects.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
) and in poor metabolic control (Table 2). BMI and body weight increased significantly after 4 months of treatment (Table 1). Lean body mass remained stable, but fat percentage increased (Table 1).
|
|
EGP in the basal state did not change statistically significantly (P = 0.07), and GNG was not affected by treatment with pioglitazone (Table 2
). The hepatic insulin resistance index did not change (Table 2). During the hyperinsulinemic euglycemic clamp, Rd increased significantly with pioglitazone [12 (8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24) µmol/kg·min vs. 21 (9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27) µmol/kg·min (mean increase of 42%), P = 0.01; Fig. 1 and Table 3]. EGP during the clamp did not change significantly (P = 0.09), but there was a trend for an increase in EGP during the hyperinsulinemic euglycemic clamp during infusion of Intralipid (Table 3). The infusion of Intralipid resulted in a decrease in Rd to pretreatment levels (Fig. 1 and Table 3). There was no statistically significant change in fasting plasma FFA (Table 2). However, pioglitazone treatment resulted in an increase of insulin-induced suppression of plasma FFA (Table 3). The infusion of Intralipid resulted in an increase of plasma FFA to baseline levels (Table 3).
|
|
No differences were found in basal plasma concentrations of glucoregulatory hormones after treatment. Insulin concentrations during the clamps were not different (Table 3). Total adiponectin increased significantly [4 (1.5–7) µg/ml vs. 7.1 (4, 5, 6, 7, 8, 9, 10, 11, 12, 13) µg/ml, P = 0.01]. Adiponectin levels did not change after 6 h of infusion with Intralipid [7.1 (3.7–13.1) µg/ml vs. 7.1 (2.8–11.4) µg/ml, P = 0.29)]. Plasma sTNF-RI and sTNF-RII concentrations were not changed after 4 months of treatment.
Glycosphingolipids and ceramide
Plasma concentrations of ceramide and glucosylceramide did not change after treatment (data not shown). The intramyocellular concentrations of ceramide in the basal state did not change. Ceramide was significantly higher during the hyperinsulinemic clamp after treatment. The absolute change in ceramide concentration between the basal state and the clamp before and after treatment was not different. Glucosylceramide and lactosylceramide did not differ (Table 4). From one patient, the amount of muscle biopsy was too little to perform a reliable analysis.
|
Multivariate analysis identified plasma FFA as only independent variable in predicting Rd (r = –0.83; Fig. 2). There was no significant correlation between ceramide, glucosylceramide, and lactosylceramide concentrations in muscle and Rd.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We did not find a change in basal EGP after treatment with pioglitazone. This is in line with a recent systematic review that showed that, in placebo-controlled trials, there is no change in basal EGP from baseline after treatment with TZDs (29). Insulin-mediated suppression of EGP was not enhanced by pioglitazone, but this could be due to the high insulin concentrations used in our study. There was a trend for an increase in EGP after infusion of Intralipid while on treatment, thus suggesting that pioglitazone cannot protect the liver from FFA-induced increase in endogenous glucose production.
Adiponectin concentrations were significantly increased after treatment, which is a well-known effect of TZDs in general (30). Adiponectin is suggested to be a mediator in the insulin sensitizing effect of TZDs because it stimulates muscle fatty acid oxidation, leading to decreased intramyocellular triglycerides and decreases EGP (31). However, recent data in adiponectin knockout ob/ob mice indicate that adiponectin is not required for pioglitazone to ameliorate skeletal muscle insulin resistance (32). We found no significant correlation between adiponectin levels and Rd. Because we measured full-length adiponectin, we cannot rule out an increase in the HMW complex of adiponectin upon treatment with pioglitazone, which is suggested to be correlated to the insulin-sensitizing effect of TZDs (33). However, in our study, the increase in plasma adiponectin did not protect either skeletal muscle or the liver from a FFA-induced decrease in Rd or increase in EGP, respectively.
In insulin-resistant obese patients, intramyocellular ceramide content is increased and correlates positively with plasma FFA (6) and negatively with whole body insulin sensitivity (7). Data on the effect of TZDs on muscle ceramide levels, (only obtained in rodents) are conflicting. Treatment of mice with troglitazone for 1 d decreased intramyocellular ceramide levels by 76% together with a decrease in its precursor palmitoyl CoA (34). However, another study with rosiglitazone in rats showed a 100% increase in intramyocellular ceramide levels despite an improvement in glucose tolerance (35). Furthermore, the increase in plasma FFA in animal models of high-fat diet induced insulin resistance does not always result in higher levels of ceramide in skeletal muscle despite reduced peripheral insulin sensitivity (1). Data for the exact role of intramyocellular ceramide and glycosphingolipids in peripheral insulin resistance in humans are lacking. In our study, treatment for 4 months with pioglitazone did not result in a decrease of intramyocellular ceramide or other glycosphingolipids in the basal state. During the hyperinsulinemic clamp, intramyocellular concentrations of ceramide were even higher after treatment despite lower plasma FFA, and a significant increase in peripheral glucose disposal. This is surprising because an earlier report suggested that a relatively small increase in ceramide is sufficient to interfere with normal insulin signaling (36). We conclude from these findings that the insulin-sensitizing effect of pioglitazone is not attained via a decrease in intramyocellular ceramide, glucosylceramide, or lactosylceramide.
We did not measure gangliosides and, therefore, cannot rule out any effect of pioglitazone on these parameters. We cannot rule out a slight decrease in intramyocellular glucosylceramide or lactosylceramide that was not detected with our method of analysis.
Finally, we also measured ceramide and glucosylceramide in plasma and found no effect of TZD treatment. Whether these plasma concentrations are a reflection of production by the liver, skeletal muscle, or other organs is not known. Also, a small amount of glycosphingolipids in the systemic circulation may be derived from dietary sources (37). It is not known whether plasma ceramide and glucosylceramide concentrations reflect intracellular ceramide metabolism. We found increased levels of plasma ceramide in obese otherwise healthy subjects compared with lean subjects (Serlie, M. J. M., A. J. Meijer, J. E. Groener, A. Poppema, M. T. Ackermans, J. M. F. G. Aerts, and H. P. Sauerwein, unpublished observation), but are not aware of any study addressing the issue of plasma concentrations in relation to intracellular concentrations.
In conclusion, we conducted a study with pioglitazone for 4 months in overweight patients with moderately controlled T2DM to evaluate the protective effects of pioglitazone on FFA-induced insulin resistance. Rd and insulin-mediated suppression of plasma FFA both were increased significantly after treatment with pioglitazone. We found that, by increasing plasma FFA to pretreatment levels, the pioglitazone-induced increase in Rd was completely abolished. Therefore, lowering plasma FFA is crucial in the insulin sensitizing effect of pioglitazone. Furthermore plasma FFA was the only independent variable of Rd. Measurements of plasma ceramide and glucosylceramide and intramyocellular ceramide and glycosphingolipids did not show any significant change in the basal state. There was a significantly higher concentration of intramyocellular ceramide during the hyperinsulinemic clamp after treatment with pioglitazone. This indicates that the insulin-sensitizing effects of TZDs are not attained via a decrease in ceramide or glycosphingolipids. Finally, our results indicate that dietary interventions aimed at lowering plasma FFA may be necessary to achieve the best effect of pioglitazone on peripheral insulin sensitivity.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online October 24, 2006
Abbreviations: BMI, Body mass index; EGP, endogenous glucose production; FFA, free fatty acid; GNG, gluconeogenesis; PPAR, peroxisome proliferator-activated receptor ; Rd, peripheral glucose uptake; sTNF-RI, soluble TNF-receptor I; sTNF-RII, soluble TNF-receptor II; T2DM, type 2 diabetes mellitus; TZD, thiazolidinedione.
Received July 13, 2006.
Accepted October 16, 2006.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
activation play an essential role in palmitate-induced insulin resistance in rat L6 skeletal muscle cells. Biochem J 382(Pt 2):619–629 (PPAR) deficiency and PPAR agonist improve insulin resistance. J Biol Chem 276:41245–41254-deficient mice develop increased adiposity and insulin resistance but respond to thiazolidinediones. J Clin Invest 112:608–618[CrossRef][Medline] deficiency and TZD treatment on insulin resistance associated with age and high-fat feeding. Am J Physiol Endocrinol Metab 284:E618–E626This article has been cited by other articles:
 |
|
 |
|
  W. L. Holland and S. A. Summers Sphingolipids, Insulin Resistance, and Metabolic Disease: New Insights from in Vivo Manipulation of Sphingolipid Metabolism Endocr. Rev., June 1, 2008; 29(4): 381 - 402. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. N. van der Crabben, G. Allick, M. T. Ackermans, E. Endert, J. A. Romijn, and H. P. Sauerwein Prolonged Fasting Induces Peripheral Insulin Resistance, Which Is Not Ameliorated by High-Dose Salicylate J. Clin. Endocrinol. Metab., February 1, 2008; 93(2): 638 - 641. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
