| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Endocrinology and Metabolism, Eberhard Karls Universität, Tubingen, Germany
Address all correspondence and requests for reprints to: Dr. Michael Stumvoll, Medizinische Universitätsklinik, Otfried Müller Strasse 10, D-72076 Tubingen, Germany. E-mail: michael.stumvoll{at}med.uni-tuebingen.de
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In addition to the classical fat deposits, however, muscle has been identified as a tissue containing relevant amounts of lipids (5). These were shown to be located not only extra- but also intramyocellularly (6, 7). In addition, a significant correlation between the intramyocellular lipid content, determined by magnetic resonance spectroscopy, and insulin resistance, demonstrated by decreased metabolic glucose clearance during hyperinsulinemic-euglycemic clamps, was found in normal glucose-tolerant subjects (6, 7). Moreover, we recently demonstrated that muscle lipolysis is sensitive to minute elevations of serum insulin levels (8). This suggests an important role for im lipids in glucose homeostasis.
The extent to which FFAs interfere with metabolic processes such as muscle glucose metabolism depends on the availability of FFAs. This is a direct function of lipolysis in various triglyceride pools. Insulin is the most potent hormonal inhibitor of lipolysis, and the diurnal variations in insulin levels essentially determine the rate of lipolysis under everyday conditions. Thus, the insulin regulation of lipolysis in a given triglyceride pool will ultimately determine the metabolic relevance of this pool for glucose metabolism and insulin resistance in skeletal muscle. Therefore, we extended our previous studies (8) and established insulin doseresponse characteristics for systemic, sc adipose tissue and muscle lipolysis in 13 lean, healthy volunteers using a combination of isotopic and microdialysis techniques during a 3-step hyperinsulinemic-euglycemic clamp.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
After having obtained approval of the protocol from the local
ethical committee and informed written consent, we studied 13 normal
weight, healthy volunteers (subject characteristics are in Table 1
). Microdialysis data for 4 subjects
have been previously reported (8). Before the study, all subjects had
their medical history taken and underwent a physical examination, a
routine blood test, and an electrocardiogram. In addition, an oral
glucose tolerance test was performed to exclude diabetes or impaired
glucose tolerance. Body fat mass was estimated using bioelectrical
impedance (RJL, Detroit, MI).
|
Volunteers were admitted to the Metabolic Research Unit at 1800 h on the day before the study, where they received a standard meal. At approximately 2100 h the microdialysis catheters were inserted in the anterior tibial muscle and the periumbilical sc adipose tissue (see below) and perfused overnight to stabilize baseline conditions. After a 12-h fast an antecubital vein was cannulated to permit infusion of insulin, glucose, and isotopes. A dorsal hand vein on the contralateral arm was cannulated retrogradely and placed under a heating device to permit sampling of arterialized blood. At 0600 h a primed continuous infusion of [2H5]glycerol (Cambridge Isotope Laboratories, Andover, MA; 1 µmol/kg, 4 µmol/min) was started. At 0800 h subjects received sequential insulin infusions at rates of 0.1, 0.25, and 1.0 mU/kg·min for 2 h at each rate. Blood was drawn every 5–10 min for determination of blood glucose, and a glucose infusion was adjusted appropriately to maintain the baseline glucose level. Arterialized blood samples were obtained at -20, -10, and 0 min before the start of the insulin infusion and at 100, 110, and 120 min of each 2-h insulin infusion rate for determination of plasma glycerol concentration and [2H5]glycerol enrichment, and serum FFA and insulin concentrations.
Microdialysis procedure
Interstitial glycerol, which is neither taken up nor metabolized to a major extent by adipose tissue and skeletal muscle, is generally accepted as a qualitative index of hydrolysis of lipids (9, 10, 11). Therefore, alterations in dialysate glycerol reflect changes in tissue lipolysis (9, 10). After application of local anesthesia, two custom-made double lumen microdialysis catheters (CMA 60, CMA, Stockholm, Sweden) with a molecular mass cut-off of 20 kDa were inserted in both the periumbilical sc adipose tissue and the tibialis anterior muscle. They were connected to the perfusion pumps (CMA 102 or 106) and perfused with modified Krebs-Henseleit buffer at 0.3 µL/min (11).
For estimation of tissue blood flow, a second catheter in each compartment was perfused with a perfusion fluid (3.0 µL/min) supplemented with 50 mmol/L ethanol to assess the ethanol outflow/inflow ratio (10, 11). Dialysate was collected simultaneously with the metabolic analyses. The ethanol outflow/inflow ratio was used as an indirect estimate of tissue blood flow, with lower ratios indicating a higher flow and vice versa.
We previously described the recovery rates of glycerol in our hands. At the flow rate of 0.3 µL/min, they were 96 ± 3% and 61 ± 10% in adipose tissue and the tibialis anterior muscle, respectively, and did not change over time or in response to insulin (8). Thus, in adipose tissue, nearly absolute tissue concentrations are measured, whereas in tibialis, anterior dialysate glycerol underestimates the interstitial levels. Thus, changes in dialysate glycerol concentrations induced by insulin in both compartments should qualitatively reflect changes in interstitial glycerol.
It may be argued that the interpretation of the microdialysis data is limited by the fact that there is glycerol uptake by muscle, but not by adipose tissue (12). This would suggest that the interstitial glycerol concentration in muscle is not as good a parameter for lipolysis as that in adipose tissue. Nevertheless, isotopically determined fractional extraction of glycerol across the forearm does not change during glucose-induced hyperinsulinemia of similar magnitude as in our third hyperinsulinemic step (12). This strongly indicates that muscle glycerol uptake is not regulated by insulin. Therefore, changes in interstitial glycerol concentrations should primarily reflect changes in glycerol release, i.e. lipolysis.
Analytical procedures
Blood glucose was determined using a bedside glucose analyzer (glucose oxidase method; YSI, Inc., Yellow Springs, OH). Serum insulin was measured with a microparticle enzyme immunoassay (Abbott, Wiesbaden, Germany), FFA with an enzymatic method (Wako Chemicals, Neuss, Germany), glycerol with an enzymatic method (Sigma, Deisenhofen, Germany), and ethanol with a colorimetric method (ALC, DuPont, Wilmington, DE). Dialysate concentrations of glycerol were measured in a CMA 600 bedside analyzer (provided by Roche, Mannheim, Germany) with commercially available test kits (CMA), which previously have been shown to correlate closely with conventional methods (10). [2H5]Glycerol enrichment was determined by gas chromatography-mass spectrometry using the trimethylsilyl derivative of glycerol. Electron impact ionization was applied, and the mass to charge ratios of 205 and 208 were monitored (13).
Calculations
The plasma rate of appearance of glycerol (Ra), used as an index for systemic lipolysis, was calculated according to the steady state equation: Ra = (ENRinf/ENRpl - 1) x F, where ENRinf is the isotopic enrichment of the infusate, ENRpl is the isotopic enrichment of plasma (both in atom percent excess), and F is the rate of the isotope infusion (micromoles per min). The interstitial glycerol concentration in the periumbilical adipose tissue, and the anterior tibial muscle was used as index for sc adipose tissue and im lipolysis, respectively. For glycerol Ra, the mean of the -20, -10, and 0 min and that of the 100, 110, and 120 min values of each step were used as steady state values. For interstitial concentration, the mean of the -40, -20, and 0 min and that of the 80, 100, and 120 min values of each step were used as steady state values.
The insulin sensitivity of each compartment (systemic, sc adipose tissue, and muscle) was assessed as the serum insulin concentration that effectively suppressed lipolysis in the respective compartment by 50% (EC50). The EC50 for each compartment was estimated by curve fitting a monoexponential function (lipolysis index = max + a x e-Ins x k) to the basal and the three steady state values of the respective indexes of lipolysis where max is maximal suppression, a and k represent the fitted parameters, and Ins is the serum insulin concentration. The EC50 for each compartment was calculated as EC50 = Ins0 + ln2/k, where Ins0 represents the serum insulin concentration at baseline. The infusion rate of exogenous glucose (GIR, micromoles per kg/min) necessary to maintain euglycemia during the third step was used to calculate an insulin sensitivity index (SI; micromoles per kg/min·pmol/L) for systemic glucose uptake: SI = GIR/Ins, where Ins indicates the steady state serum insulin concentration during step 3.
Statistical analysis
Data are given as the mean ± SEM unless otherwise stated. Changes over time, i.e. basally and at the three steps of the hyperinsulinemic-euglycemic clamp, and differences between compartments (muscle vs. adipose tissue) were calculated by a full factorial ANOVA with repeated measures design and the factors step and compartment. Direct comparisons between compartments at individual steps were made using paired, two-tailed Student’s t test. EC50 values were compared using paired t test after applying a Bonferroni transformation. The software package SPSS, Inc./PC+ (SPSS, Inc., Chicago, IL) was used for statistical analysis.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
)Blood glucose remained constant during the three clamp steps. Serum insulin, glycerol, and FFAs were in an apparent steady state at baseline and at the end of each step. Serum insulin increased to supraphysiological levels during step 3. During step 3 serum glycerol decreased by approximately 50%, whereas serum FFA was almost completely suppressed.
|
and Fig. 1)Glycerol enrichments achieved an apparent plateau at baseline and at the end of each step. The glycerol rate of appearance in plasma was 1.35 ± 0.16 µmol/kg·min at baseline and 0.86 ± 0.15, 0.48 ± 0.07, and 0.38 ± 0.04 µmol/kg·min at the end of steps 1, 2, and 3, respectively. Based on fat mass (FM), the glycerol rate of appearance in plasma was 8.74 ± 1.3 µmol/kg FM·min at baseline and 5.56 ± 1.17, 3.19 ± 0.62, and 2.45 ± 0.33 µmol/kg FM·min at the end of steps 1, 2, and 3, respectively.
|
and Fig. 1)Tissue blood flow as estimated by the ethanol outflow/inflow ratio was higher in skeletal muscle (0.25 ± 0.04) than in adipose tissue (0.44 ± 0.03; P < 0.05). Hyperinsulinemia did not induce any change in the ethanol outflow/inflow ratio in any compartment (data not shown). Thus, the variation in dialysate glycerol concentration during the three-step glucose clamp should reflect alterations of its release by the different tissues. The absolute dialysate glycerol concentration was higher in adipose tissue at baseline and at all three steps during the clamp (P < 0.002). However, due to the recovery of about 60% in skeletal muscle, dialysate concentrations underestimate true interstitial glycerol concentrations. During the last 40 min of each step, a metabolic steady state for glycerol was reached in skeletal muscle and adipose tissue, as demonstrated by the slope of the last three measurements over time not being significantly different from zero (data not shown). The decrease in interstitial glycerol concentrations, expressed as percentage of the baseline, was 69 ± 4% in muscle and 70 ± 7% in adipose tissue (P = 0.82) during step 1, 51 ± 4% in muscle and 45 ± 7% in adipose tissue (P = 0.50) during step 2, and 53 ± 4% in muscle and 22 ± 2% (P < 0.001) in adipose tissue during step 3.
Insulin dose-response characteristics for glycerol Ra and muscle
and adipose tissue glycerol concentrations (Table 3 and Fig. 2)
The insulin dose-response curves for systemic, adipose tissue, and
muscle lipolysis are shown in Fig. 2. The insulin
EC50 was lowest for the decrease in the
interstitial muscle glycerol concentration, followed by that for the
decrease in plasma glycerol Ra, and was highest for the decrease in the
interstitial adipose tissue glycerol concentration. This indicates that
muscle lipolysis was most sensitive to insulin, followed by systemic
lipolysis, and was least sensitive in adipose tissue. As shown in Table 3, this order was preserved in every patient. The
EC50 for glycerol Ra was not different when
normalized by fat mass rather than by total body weight.
|
|
There were significant correlations between the insulin sensitivity index for glucose uptake and the EC50 for systemic lipolysis (linear: r = 0.78; P < 0.05; nonlinear: r = 0.93; P < 0.05), muscle lipolysis (linear: r = 0.67; P < 0.05; nonlinear: r = 0.85; P < 0.05), and adipose tissue lipolysis (linear: r = 0.83; P < 0.05; nonlinear: r = 0.95; P < 0.05).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The interpretation of the EC50 values in the two local compartments is somewhat limited by the fact that interstitial insulin concentrations were not measured. There is substantial evidence for an active transport system of insulin across the blood-tissue barrier. In the interstitium of muscle and fat, insulin concentrations were found to be approximately 50% of those in plasma (14, 15). As the transendothelial transport is likely to vary between individuals and tissues, we did not directly assess the cellular insulin sensitivity of muscle and fat cells. Nevertheless, our determinations of the suppression of glycerol release provide a reasonable index for the sum effect of circulating insulin, i.e. transendothelial transport plus intracellular metabolic action, on lipolysis. Therefore, we believe that the EC50 permits direct comparison of the tissue-specific net insulin sensitivity of lipolysis.
In a strict pharmacological sense, the use of EC50 values is appropriate to compare insulin sensitivity between tissues. In a more physiological sense, however, it is also important to interpret the percent decrease from baseline, which was not different between muscle and adipose tissue during the first two steps. It is only during the third, supraphysiological step, in which adipose tissue lipolysis continues to decrease, that the differences between the tissues become clearly evident. This, nevertheless, indicates that muscle exerts its full regulatory range of lipolysis within much lower insulin levels than adipose tissue.
Regional differences in the regulation of adipose tissue lipolysis using different methodologies have been described previously in humans (reviewed in Ref. 1). It is generally accepted that visceral adipose tissue is more insulin resistant than sc adipose tissue (16, 17, 18, 19). Moreover, sc adipose tissue appears to be more resistant at abdominal than at gluteal or femoral sites (20, 21, 22). Thus, in the hierarchy for suppression of lipolysis, muscle appears to be the most insulin-sensitive compartment. The underlying cellular mechanisms for this observation are unclear at present, as few data regarding the specific molecular equipment involved in the regulation of lipolysis are available for muscle. Nevertheless, the concept of differential regulation of lipolysis is in principle supported by the demonstration of different phosphodiesterase subtypes (11) and different ß-adrenergic receptor subtypes (23) in muscle and adipose tissue, respectively, to be involved in inhibition and/or stimulation of lipolysis. In addition, differences in autonomic innervation between muscle and adipose tissue are likely to be involved.
Teleologically, im lipolysis, which is highly sensitive to small increases in insulin, will guarantee the postprandial switch to preferential utilization of glucose and ultimately the maintenance of normal glucose tolerance. In the postabsorptive state resting muscle predominantly uses FFAs as a metabolic fuel, from both the systemic circulation and lipolysis of im triglycerides. With the transition to the postprandial state, glucose is increasingly made available and becomes the preferred fuel in muscle. For this sequence to function smoothly it is essential that the accompanying rise in insulin shuts off lipolysis primarily in the muscle compartment. In this way the immediate substrate competition for oxidative pathways between FFAs and glucose (24) and FFA-induced impairment of glucose transport and storage (25, 26) are eliminated postprandially.
The second important finding is that the insulin EC50 for systemic lipolysis, which was of the same order of magnitude as that reported previously (27), ranged precisely between that for muscle and adipose tissue. At first glance, there appears to be a problem with respect to the quantitative contributions of local glycerol release to systemic Ra. One would expect that systemic glycerol Ra should be essentially attributable to glycerol release from adipose tissue, and that the two dose-response curves should be similar. Moreover, an assumed dose-response curve for visceral adipose tissue would be expected to the right of the curve for sc adipose tissue. The most likely explanation is the overestimation of the insulin concentrations by using circulating instead of interstitial levels, which resulted in an apparent rightward shift of the local curves. There are, however, a number of alternative explanations.
Firstly, the microdialysis technique in the basal state underestimates glycerol release into the vascular compartment more than that in the insulin-stimulated state, resulting in a false rightward shift of the microdialysis curves (28). Secondly, there may, in fact, be a greater contribution of the glycerol released from muscle than usually expected. Although it is not known to what extent muscle lipolysis and subsequent release of glycerol from muscle contribute to the systemic appearance of glycerol in plasma as determined isotopically, there is robust evidence from isotopic arteriovenous difference studies that there is significant release of glycerol from muscle (12). Assuming that blood flow to muscle and adipose tissue is similar and estimating whole body muscle mass to be at least 3 times that of fat, it can be calculated from the work of Coppack et al. (12) that release of glycerol into the systemic circulation from muscle may be as much as 50% of that from adipose tissue.
The compartmentalization of triglyceride deposits with respect to regulation of lipolysis is of immediate relevance for the concept of FFA-mediated development of insulin resistance. Not only increased im triglyceride deposits (6, 7) but also excessive lipolysis could be involved. The compartmental proximity of muscular lipolysis and glucose disposal (storage and oxidation) makes excessive FFA release within muscle itself a prime candidate for interfering with muscle glucose metabolism. This implies that not only endocrine mechanisms, i.e. FFAs from distant fat deposits delivered by systemic circulation, but also paracrine and/or autocrine mechanisms, i.e. within muscle itself, could be involved in the FFA-mediated insulin resistance of skeletal muscle.
The third interesting finding of this study is the correlation between the EC50 of suppression of lipolysis and the insulin sensitivity index for glucose disposal. About half of the variation in insulin-stimulated glucose disposal could be explained by changes in the suppression of lipolysis in each compartment. Either this indicates the causal relationship between increased FFA delivery and muscular insulin sensitivity (3) or it reflects common elements of the insulin signaling cascade. Intracellular signaling from the insulin receptor to glucose transport and glycogen synthesis, on the one hand, and hormone-sensitive lipase, on the other hand, share a number of signaling proteins (29, 30). The correlation between the two effector systems could thus reflect the interindividual variation in the intrinsic level of activity in the common signaling pathway.
In summary, based on EC50 values, lipolysis of im triglycerides appears to be more sensitively regulated by insulin than that of adipose tissue. The insulin EC50 of systemic lipolysis ranged precisely between those of muscle and adipose tissue. However, when expressed as the percent suppression from baseline, muscle and adipose lipolysis decreased similarly during physiological hyperinsulinemia. By exerting its fully regulatory range, regulation of muscle lipolysis could contribute to the fine-tuning of glycerol (and FFA) turnover within diurnal insulin variations. This may be a prerequisite for the preferential utilization of glucose as opposed to FFAs by muscle in the postprandial state. Moreover, the correlation between insulin’s effect on muscle lipolysis and glucose disposal suggests either a causal relationship between FFA delivery and insulin sensitivity or common elements of the insulin signaling cascade to both effector systems. Finally, inadequate suppression of im lipolysis resulting in excessive availability of FFAs may represent a novel mechanism contributing to the pathogenesis of impaired glucose disposal, i.e. insulin resistance, in muscle.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received December 8, 1999.
Revised June 27, 2000.
Accepted June 28, 2000.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
isozyme of protein kinase C by phosphatidylinositol
3,4,5-trisphosphate. J Biol Chem. 268:13–16.This article has been cited by other articles:
 |
|
 |
|
  M. O. Weickert, M. Mohlig, J. Spranger, C. Schofl, C. V. Loeffelholz, R. L. Riepl, B. Otto, and A. F. H. Pfeiffer Effects of Euglycemic Hyperinsulinemia and Lipid Infusion on Circulating Cholecystokinin J. Clin. Endocrinol. Metab., June 1, 2008; 93(6): 2328 - 2333. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Dimitriadis, P. Mitrou, V. Lambadiari, E. Boutati, E. Maratou, E. Koukkou, M. Tzanela, N. Thalassinos, and S. A. Raptis Glucose and Lipid Fluxes in the Adipose Tissue after Meal Ingestion in Hyperthyroidism J. Clin. Endocrinol. Metab., March 1, 2006; 91(3): 1112 - 1118. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Qvisth, E. Hagstrom-Toft, S. Enoksson, E. Moberg, P. Arner, and J. Bolinder Human Skeletal Muscle Lipolysis Is More Responsive to Epinephrine Than to Norepinephrine Stimulation in Vivo J. Clin. Endocrinol. Metab., February 1, 2006; 91(2): 665 - 670. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Berglund Adipose tissue, skeletal muscle, and insulin resistance across ethnicities--systems biology in action Am. J. Clinical Nutrition, December 1, 2005; 82(6): 1153 - 1154. [Full Text] [PDF]
|
|
|
|
|
|
V. Qvisth, E. Hagstrom-Toft, S. Enoksson, R. S. Sherwin, S. Sjoberg, and J. Bolinder Combined Hyperinsulinemia and Hyperglycemia, But Not Hyperinsulinemia Alone, Suppress Human Skeletal Muscle Lipolytic Activity in Vivo J. Clin. Endocrinol. Metab., September 1, 2004; 89(9): 4693 - 4700. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. C Gannon, F. Q Nuttall, A. Saeed, K. Jordan, and H. Hoover An increase in dietary protein improves the blood glucose response in persons with type 2 diabetes Am. J. Clinical Nutrition, October 1, 2003; 78(4): 734 - 741. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Hagstrom-Toft, V. Qvisth, I. Nennesmo, M. Ryden, H. Bolinder, S. Enoksson, J. Bolinder, and P. Arner Marked Heterogeneity of Human Skeletal Muscle Lipolysis at Rest Diabetes, December 1, 2002; 51(12): 3376 - 3383. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Moberg, S. Sjoberg, E. Hagstrom-Toft, and J. Bolinder No apparent suppression by insulin of in vivo skeletal muscle lipolysis in nonobese women Am J Physiol Endocrinol Metab, August 1, 2002; 283(2): E295 - E301. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Sjostrand, S. Gudbjornsdottir, A. Holmang, L. Strindberg, K. Ekberg, and P. Lonnroth Measurements of Interstitial Muscle Glycerol in Normal and Insulin-Resistant Subjects J. Clin. Endocrinol. Metab., May 1, 2002; 87(5): 2206 - 2211. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Stumvoll, H. G. Wahl, S. Jacob, A. Rettig, F. Machicao, and H. Haring Two novel prevalent polymorphisms in the hormone-sensitive lipase gene have no effect on insulin sensitivity of lipolysis and glucose disposal J. Lipid Res., November 1, 2001; 42(11): 1782 - 1788. [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 |
