Muscle Glycogen Content in Type 2 Diabetes Mellitus. Jing He and David E. Kelley

Transcription

1 Articles in PresS. Am J Physiol Endocrinol Metab (July 13, 2004) /ajpendo E R3 Muscle Glycogen Content in Type 2 Diabetes Mellitus Jing He and David E. Kelley From the Department of Medicine, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania. Running Title: Muscle glycogen in type 2 DM Key Words: Skeletal muscle; glycogen; insulin resistance; diabetes mellitus Address correspondence to: David E. Kelley; Professor of Medicine; 807N Montefiore- University Hospital; 3459 Fifth Ave; University of Pittsburgh; Pittsburgh, PA kelley@msx.dept-med.pitt.edu Abbreviations: T2DM: type 2 diabetes mellitus; OX-ENZ: oxidative enzyme activity; IMCL: intramyocellular lipid content; GLYC: glycogen; IR: insulin resistance; FPG: fasting plasma glucose; Ob: obesity 1 Copyright 2004 by the American Physiological Society.

2 ABSTRACT Muscle contains the largest reservoir of glycogen (GLYC), a depot that is closely regulated and with influence upon insulin sensitivity. The current study examines muscle GLYC in T2DM and obesity, and with respect to muscle fiber type, IMCL and mitochondrial function (OX-ENZ). There is increasing interest in the relation of IMCL and mitochondrial dysfunction with IR, yet the association with muscle GLYC has not been examined in regard to these parameters. Using a quantitative histological approach, specific to muscle fiber types, muscle GLYC, IMCL and OX-ENZ were assessed in vastus lateralis obtained by percutaneous biopsy in lean non-diabetic (L; n=16), obese non-diabetic (Ob; n=15), and T2DM volunteers (n=14). Insulin sensitivity was estimated using HOMA-IR. Muscle GLYC was reduced in T2DM; a deficit evident for type 2a fibers, yet minor in types 1 and 2b fibers. Low GLYC in T2DM correlated with fasting hyperglycemia. Also, in T2DM and Ob, there was significantly higher IMCL and lower OX-ENZ in all fiber types. The ratio of IMCL:OX-ENZ, especially for type 1 fibers, correlated strongly with IR. Similarly, a ratio of GLYC: OX-ENZ correlated with IR, particularly for type 2b fibers. This ratio tended to be higher in Ob and T2DM. In summary, there is decreased muscle GLYC in T2DM, yet a disproportional relationship of GLYC to OX-ENZ that is related to IR though not as robustly as IMCL:OX-ENZ. 2

3 INTRODUCTION The intertwined epidemics of obesity and type 2 diabetes (T2DM) have stimulated increased interest in understanding potential mechanisms for nutrient induced insulin resistance (IR) (24). A principal focus has been upon intra-myocellular lipid (IMCL) (25; 28; 32). Mitochondrial dysfunction in skeletal muscle has been described in T2DM, in those at risk for this disease, and in obesity (17; 22; 26; 30), and may relate to IR by disposing to IMCL accretion (16; 29). Furthermore, muscle oxidative enzyme capacity (OX-ENZ) may modulate the relation between IMCL and IR. Highly trained athletes manifest heightened insulin sensitivity and OX-ENZ, and maintain high IMCL (12), and recent studies, including interventions, indicate that OX-ENZ might be more important than IMCL in relation to IR (4; 15; 20; 38). Thus, it would seem that the relationship between IMCL and IR is more precisely delineated when considered in relation to muscle OX-ENZ. Whether there is an analogous interaction between OX-ENZ and glycogen content in skeletal muscle (GLYC) in modulating insulin sensitivity has not, to our knowledge, been examined. While IMCL is just a minute fraction of systemic lipid stores, skeletal muscle contains the largest reservoir of glycogen, approximately four-fold that of postprandial liver. Glycogen content (GLYC) can regulate insulin sensitivity of muscle (9; 11). Increases in muscle GLYC induced by over-feeding and inactivity rapidly induce IR while depletion of GLYC by activity rapidly improves insulin sensitivity, even in those with IR (3; 23; 27); effects that persist until GLYC is replete (11). It has been firmly established that insulin-stimulated rates of glycogen synthesis in muscle are markedly reduced T2DM (2; 8; 18; 19; 40; 44). This impairment has been 3

4 directly imaged using NMR spectroscopy during insulin infusions (40), and following meal ingestion (6). There is no evidence that IR of muscle, in T2DM, is caused by increased muscle GLYC. Instead, most studies indicate that muscle GLYC is decreased or not different in T2DM. Roch-Norlund et al. (34), measured muscle GLYC in a large cohort of patients with diabetes mellitus. In those with type 1 DM, among a cohort of patients in generally poor metabolic control, a 65% reduction in muscle GLYC was observed, while in those with T2DM treated with diet or oral medications, muscle GLYC was reduced by approximately 20% (34). Interestingly, this was a non-significant reduction, as just one-quarter of T2DM had muscle GLYC below the normal range. A similarly mild deficit was observed by Carey using NMR despite a marked reduction in rates of postprandial glycogen synthesis in muscle (6). Consistent with the above findings, in several animal models of obesity, T2DM, and IR, muscle GLYC is equivalent to control animals, or only moderately reduced, despite prominent defects in activity of glycogen synthase and other components of the glycogen synthesis pathway (5; 10; 39; 43). Thus, there is an intriguing paradox, with marked IR in rates of glycogen accretion in T2DM yet only a moderate decrease in muscle GLYC. The current study was undertaken to examine muscle GLYC content in T2DM taking into account the potential effects of muscle fiber type, and to explore interactions amongst muscle GLYC, IMCL and OX-ENZ in relation to IR. These objective were addressed using a quantitative, fiber-type specific, histological method to assess muscle GLYC, IMCL and OX-ENZ. It has been recognized that mitochondrial diseases can dispose to GLYC accumulation in muscle in addition to accumulation of IMCL (41), and that increased GLYC disposes to increased fat balance (37). Thus, we undertook these 4

5 studies to test the novel hypothesis that considered with regard mitochondrial dysfunction in T2DM and Ob, muscle GLYC might contribute to nutrient sensing mechanisms that modulate IR (9; 24). METHODS Subjects: The clinical characteristics the 16 lean, non-diabetic (L), 15 obese, nondiabetic (Ob), and 14 research participants with T2DM are shown in Table 1. Data on IMCL and OX-ENZ had previously been reported on these individuals (14). In this study, to address our current hypothesis concerning muscle GLYC, the muscle tissue blocks were newly sectioned and analyzed for GLYC in conjunction with fiber type, IMCL and OX-ENZ, using the single fiber analytic approach previously described (14). Research participants were recruited by public advertisement and written, informed consent was obtained. A medical examination was performed prior to participation. Those on treatment with insulin were excluded and oral agents were discontinued four weeks prior to muscle biopsy. Those with T2DM had fasting hyperglycemia and higher HbA1c, and had more severe IR. T2DM and Ob were matched for BMI, and with respect to L, the Ob group had higher fasting insulin, higher FPG and were IR; the participants with T2DM were slightly older. Values for fasting insulin and plasma glucose were used to calculate HOMA-IR (FPG*Insulin/22.5) (33). This research was approved by the University of Pittsburgh Institutional Review Board and volunteers gave written informed consent prior to participation. Muscle Biopsy: The research participants fasted overnight and had been instructed to not exercise on the day prior to the muscle biopsy. Blood samples were 5

6 obtained for measurement of HbA1c and fasting plasma glucose and insulin. After bed rest for 30 minutes, a percutaneous muscle biopsy was performed in the vastus lateralis. Samples were immediately dissected free of adipose and connective tissue, embedded in Cryomatrix freezing compound (Shandon, Pittsburgh, PA.) and frozen in isopentane cooled in liquid nitrogen. Biopsy samples were maintained within sealed containers at 70 o C until microscopy studies. Histochemical analysis of muscle: Cryomatrix blocks from L, Ob, and T2DM subjects were mounted in parallel and transverse sections (8µm) were cut using a cryostat at -20 C, and then mounted on Superfrost/Plus slides (Fisher Scientific, Pittsburgh, PA) and examined for muscle glycogen content (PAS histochemical method; Sigma, St. Louis, MO), muscle fiber type, lipid and oxidative enzyme activity by using single fiber type quantitative image analysis as previously described (14). For each individual, fibers were analyzed for parameters of fiber type, GLYC, IMCL and OX-ENZ, divided approximately equally amongst type 1, type 2a and type 2b fibers. The mean value for GLYC (and for other parameters) within each fiber type was calculated and used to calculate group mean values. The within subject co-efficient of variation for assessment of GLYC was 6±2% for each fiber type. Images were captured by an optical microscope (Nikon, Microphot-FXL Japan) with a connected CCD Sony video camera (Sony Corporation, Japan) and image analyses was performed using Optimas (Media Cyberneics, L. P. Silver Spring, MD 20910), as previously described (14). Representative images are shown in Figure 1, including as a negative control a muscle section incubated in 1% amylase for 45 minutes at 37 o C to hydrolyze glycogen before proceeding to PAS staining. 6

7 Statistic analysis: Data are expressed as mean ± SEM unless otherwise indicated. Two-way ANOVA was used to examine for differences on the basis of gender and group (L, Ob and Type 2 DM). Linear regression was used to examine the relation of GLYC, IMCL, OX-ENZ and HOMA-IR. A p value < 0.05 was considered significant. RESULTS Muscle GLYC: Muscle GLYC was lower in type 1 than types 2a and 2b fibers (p<0.001), regardless of group, as shown in Figure 2. In regard to L, Ob and T2DM groups, GLYC did not differ in type 1 fibers (p=0.2). Also, group differences for GLYC were not different for type 2b fibers (p=0.4). In type 2a fibers, there was nearly significant difference for lower GLYC in T2DM (p=0.06); a reduction of ~14% versus Ob, and of ~18% compared to L. In L, type 2a fibers accounted for the largest percentage of muscle fiber types (41%, 50%, and 9% for fiber types 1, 2a and 2b, respectively), the percentages were similar for Ob (34%, 56%, and 10%) and T2DM (39%, 52% and 9%). An overall value for muscle GLYC was calculated on the basis of the percentage of each fiber type and its respective GLYC. Muscle GLYC was lowest in T2DM (77.8±5.5, 94.6±4.6, and 99.5±4.4 AU; in T2DM, Ob and L, respectively; p<0.01), with significant differences between T2DM and L (p=0.02) and between T2DM and Ob (p=0.03). There was not an effect of gender on GLYC within any muscle fiber type (p=0.3 to 0.4; data not shown). The potential relation of muscle GLYC to fasting hyperglycemia was examined. There was a significant negative correlation for all 3 fiber types: type 1 (r=-0.37; p=0.01), type 2a (r=-0.50; p<0.001), and type 2b (r=-0.45; p<0.01), indicated that fasting 7

8 hyperglycemia was associated with lower GLYC, as previously has been reported by Carey et al (6). This negative correlation was particularly robust when taking into account overall GLYC (r=-0.57; p<0.01). The relation of muscle GLYC to HOMA-IR was also examined. There was a negative correlation of moderate statistical significance for type 2a (r=-0.31; p=0.04) and type 2b (r=-0.35; p=0.02), but the relation to GLYC in type 1 fibers was not significant (r=-0.21; p=0.17). The relationship between overall muscle GLYC and HOMA-IR was of borderline significance (r=-0.32; p=0.052). These findings indicate that reduced GLYC is associated with greater IR. Muscle GLYC in relation to OX-ENZ: There were strong group differences in OX-ENZ (p<0.01). OX-ENZ was lower in T2DM and Ob compared to L, for all fiber types, as shown in Table 2. Furthermore, OX-ENZ was negatively correlated with HOMA IR, though more strongly for oxidative fiber types: for type 1 (r=-0.52; p<0.001), type 2a (r=-0.38; p<0.05) and type 2b (r=-0.31; p=0.06). Values for the ratio of muscle GLYC normalized to OX-ENZ (GLYC:OX-ENZ) are shown in Table 2. These values differed significantly across fiber types: type 2b > type 2a > type 1 (p<0.001). There were not significant group differences for GLYC:OX- ENZ in type 1 fibers, however, the ratio of GLYC:OX-ENZ did correlate significantly with HOMA IR (r=0.68; p<0.001), and more strongly than OX-ENZ alone (as shown above). For type 2a and type 2b fibers, the lowest values for GLYC:OX-ENZ were in L (p<0.05). The ratio of GLYC:OX-ENZ for both type 2a and type 2b fibers was highest in Ob. HOMA-IR was greatest in T2DM, and therefore correlation with GLYC:OX-ENZ was not statistically significant across all groups. However, the association between 8

9 GLYC:OX-ENZ and HOMA-IR was statistically significant among L and Ob with respect to type 2a (r=0.39; p<0.05) and type 2b (r=0.50; p<0.01). Muscle IMCL and IMCL:OX-ENZ ratio: IMCL was higher in T2DM and Ob compared to L in each fiber type (p<0.01). IMCL was correlated with HOMA IR for type 1 fibers (r=0.32; p=0.04), but the correlation with type 2a and 2b fibers was not statistically significant. There was not a significant correlation between IMCL and GLYC within any of the three muscle fiber types, values for r ranged from 0.05 to We next assessed the ratio of IMCL:OX-ENZ for each fiber type and the potential association with HOMA-IR. In muscle from L volunteers, the ratio of IMCL:OX-ENZ is strikingly similar regardless of fiber types and approximately half to a third of the values in Ob and T2DM, differences that were highly significant for each fiber type (p<0.001), as shown in Table 2. The ratios of IMCL:OX-ENZ were strongly correlated with HOMA IR for type 1 (r=0.79; p<0.001), and type 2a (r=0.64; p<0.001), but more modestly in relation to type 2b fibers (r=0.38; p<0.05). Because both GLYC:OX-ENZ and IMCL:OX-ENZ correlated with HOMA IR, these relationships were further assessed using step-wise regression analysis. For type 1 fibers, for all volunteers or for L and Ob volunteers separately, after IMCL:OX-ENZ, inclusion of GLYC:OX-ENZ did not add significantly to the association with HOMA-IR. A similar pattern was observed for type 2a fibers among L and Ob non-diabetic volunteers. However, for type 2b, or glycolytic fibers, an opposite pattern emerged wherein GLYC:OX-ENZ ratio was more strongly correlated with HOMA-IR than IMCL:OX-ENZ, and inclusion of IMCL:OX-ENZ did not add significantly after adjusting for GLYC:OX-ENZ. 9

10 DISCUSSION The current study was undertaken to examine the relation of muscle content of GLYC in T2DM and obesity to IR and hyperglycemia, and with specific regard to muscle fiber type, IMCL and OX-ENZ. A diminished rate of insulin-stimulated glycogen formation in skeletal muscle is highly characteristic of IR in T2DM and obesity (18; 19; 40; 44). The mechanisms that account for IR of glycogen synthesis include impaired glucose transport, elevated plasma fatty acids and IMCL, impaired insulin signaling, and reduced activation of glycogen synthase (1; 7; 18; 36; 44). Because of the severe impairment in rates of insulin-stimulated glycogen formation in skeletal muscle in T2DM and obesity, it seems logical to postulate an equally severe reduction in muscle GLYC. Yet, prior studies indicate a relatively modest reduction in muscle GLYC in T2DM (6; 22; 34), which in itself is intriguing Therefore, we re-examined this issue, taking into account fiber type, IMCL and OX-ENZ, factors not assessed in the cited studies. One of the main findings of the current study is that muscle GLYC is lower in T2DM. Overall, there is approximately a 20% reduction in muscle GLYC in T2DM compared with muscle from lean, non-diabetics. These assessments of GLYC were based upon careful analysis of a large number of muscle fibers ( fibers from each fiber type) in each biopsy sample, and furthermore, there were relatively large numbers of volunteers in each of the groups of lean, obese and T2DM volunteers. The T2DM group was approximately a decade older than the obese and lean controls and this may have influenced the findings for GLYC and OX-ENZ. Nonetheless, our findings are highly consistent with prior observations on GLYC in muscle in T2DM (6; 22; 34). 10

11 The deficit of muscle GLYC in T2DM was not symmetric across muscle fiber types. There was not a statistically significant deficit in GLYC in type 1 or 2b muscle fibers. A deficit was observed more clearly in type 2a muscle fibers, which account for nearly 50% of muscle fibers in vastus lateralis and therefore have considerable impact upon overall muscle GLYC. A significant negative correlation was found for muscle GLYC and fasting hyperglycemia, an observation that is also consistent with a prior report by Carey et al (6). We further observed a negative correlation, albeit of borderline statistical significance, between muscle GLYC and IR. These findings are consistent with the concept that the principal pathophysiology of T2DM, namely IR and impaired insulin secretion, reduces rates of glycogen synthesis in muscle in T2DM. Yet, the findings, while consistent with prior work, should not in our opinion be tacitly accepted as being entirely straight-forward. We posit that the deficit in muscle GLYC in T2DM seems less pronounced than might be anticipated from the severe reductions in rates of insulinstimulated glycogen synthesis that are generally reported. Also, a deficit was not found for type 1 or type 2b fibers. This differs from the pattern observed for IMCL, which is increased in all fiber types in T2DM and Ob, and by similar amplitude. Furthermore, the reductions in OX-ENZ in T2DM and Ob, are quite proportionate across fiber types (14). Therefore, to consider our findings on muscle GLYC from these additional perspectives, we examined muscle GLYC in relation to muscle OX-ENZ and IMCL. Considerable research in recent years has focused upon the relationship of IMCL to IR, and recent progress has served to emphasize the role of muscle OX-ENZ in modulating the relationship between IMCL and IR (13). Intervention studies of physical activity in obese and lean sedentary individuals suggest that increased OX-ENZ might be 11

12 more important than IMCL per se in assessing effects of training upon IR (4; 15; 20; 38). Muscle OX-ENZ is diminished (17), and related to severity of IR in obesity and T2DM (42), and likely caused by under-expression of nuclear genes encoding oxidative enzymes in T2DM (22; 26). Mechanistically, reduced OX-ENZ might dispose to IMCL accumulation in muscle. Too, OX-ENZ capacity might modulate the relationship of IMCL with IR. Considered as separate parameters, IMCL and OX-ENZ were correlated with HOMA-IR; higher IMCL and lower OX-ENZ each related to IR and these associations were most robust for type 1 fibers. Nevertheless, for all fiber types, and most especially for type 1 fibers, the correlation with IR was even stronger for the ratio of IMCL:OX-ENZ. We posit that the ratio of IMCL:OX-ENZ provides an index of proportionality between stored lipid and the maximal capacity of a myocyte to oxidize substrate and in this regard, provides an index potentially germane to nutrient sensing in modulating insulin sensitivity. Might a similar perspective pertain to muscle GLYC and IR in T2DM and Ob? Mott et al. (23), showed that even very slight increases in the absolute concentration of muscle GLYC induced marked IR in human volunteers. GLYC depletion, caused by physical activity rapidly though transiently improves IR (3; 27). Thus, there is certainly a conceptual basis for examining muscle GLYC as exerting feed-back upon IR. Despite lower GLYC in T2DM, the more striking deficit in OX-ENZ yielded values for the ratio of GLYC:OX-ENZ that were similar among lean, obese and T2DM for type 1 fibers and marginally increased in T2DM for type 2a and 2b muscle fibers. In obesity, this ratio in type 2a and 2b fibers is even more elevated. Higher values of GLYC:OX-ENZ were associated with more severe IR. An interpretation we would like to put forward for 12

13 consideration is that increased GLYC:OX-ENZ, like increased IMCL:OX-ENZ, serves as an index of intra-myocyte nutrient surplus relative to cellular capacity for substrate oxidation. Certainly this concept remains to be more rigorously tested. For example, we would speculate that elevated GLYC:OX-ENZ and IMCL:OX-ENZ may denote disposition for accumulation of malonyl-coa, acyl-coa, hexosamine or ceramides, nutrient induced signals that have been shown to modulate IR (21; 35). In comparing the relative strength of association of these indices with IR, for type 1 muscle fibers, the relation between IR and IMCL:OX-ENZ was considerably stronger than that of GLYC:OX-ENZ. This applied as well to type 2a fibers. Given that type 1 and type 2a fibers comprise a large majority, this would indicate that of the two indices IMCL:OX-ENZ is more important for IR. For type 2b fibers, the GLYC:OX-ENZ was a stronger correlate of IR than was IMCL:OX-ENZ. Indeed, for type 2b fibers, neither IMCL nor OX-ENZ had significant independent correlation with IR. These patterns, wherein lipid stores are better correlated with IR in oxidative fibers and glycogen stores are more strongly correlated with IR in glycolytic fibers are consistent with the known metabolic dispositions of oxidative and glycolytic fibers (31). In summary, in the current study we assessed muscle GLYC in T2DM and obesity and found a significant though modest deficit in T2DM. However, considered in relation to OX-ENZ, there is correlation of increased GLYC:OX-ENZ with IR. We postulate that evaluating lipid and glycogen storage in relation to capacity for oxidative metabolism might be a useful approach for examining nutrient induced IR in T2DM and obesity. 13

14 ACKNOWLEDGEMENTS We would like to gratefully acknowledge the contributions of the research staffs of the University of Pittsburgh General Clinical Research Center and Obesity and Nutrition Research Center, and of the collaboration of colleagues and staff at the Structural Biology Imaging Center. Most of all, we would like to acknowledge the cooperation of the research volunteers. This research was supported by grant support from NIH-NIDDK (DK49200), by the University of Pittsburgh General Clinical Research Center (5 M01RR00056), and by the University of Pittsburgh Obesity and Nutrition Research Center (P30DK462). REFERENCES 1. Boden G, Chen X, Ruiz J, White JV and Rossetti L. Mechanisms of fatty acidinduced inhibition of glucose uptake. Journal of Clinical Investigation 93: , Bogardus C, Lillioja S, Mott D, Reaven G, Kashiwagi A and Foley J. Relationship between obesity and maximal insulin-stimulated glucose uptake in vivo and in vitro in Pima Indians. Journal of Clinical Investigation 73: , Bogardus C, Thuillez P, Ravussin E, Vasquez B, Narimiga M and Azhar S. Effect of muscle glycogen depletion on in vivo insulin action in man. Journal of Clinical Investigation 72: ,

15 4. Bruce C, Anderson M, Carey A, Newman D, A B, AD K, G C and Hawley J. Muscle oxidative capacity is a better predictor of insulin sensitivity than lipid status. Journal of Clinical Endocrinology & Metabolism 88: , Bryson J, Phuyal J, Swan V and Caterson I. Leptin has acute effects on glucose and lipid metabolism in both lean and gold thioglucose-obese mice. American Journal of Physiology (Endocrinol Metab) 277: E417-E422, Carey P, Halliday J, Snaar J, Morris P and Taylor R. Direct assessment of muscle glycogen storage after mixed meals in normal and type 2 diabetic subjects. American Journal of Physiology (Endocrinol Metab) 284: E688-E694, Czech MP and Corvera S. Signaling Mechanisms That Regulate Glucose Transport. Journal of Biological Science 274: , DeFronzo R, Jacot E, Jequier E, Maeder E, Wahren J and Felber J. The effect of insulin on disposal of intravenous glucose. Diabetes 30: , Flatt JP. Dietary fat, carbohydrate balance, and weight maintenance: effects of exercise. American Journal of Clinical Nutrition 45: , Furnsinn C, Brunmair B, Meyer M, Neschen S, Furtmuller R, Roden M, Kuhnle H, Nowotny P, Schneider B and Waldhausl W. Chronic and acute effects of thiazolidinediones BM and BM on rat skeletal muscle glucose metabolism. British Journal of Pharmacology 128: , Garcia-Roves P, Dong-Ho H, Song Z, Jones T, Hucker K and Holloszy J. Prevention of glycogen supercompensation prolongs the increase in muscle GLUT4 after exercise. American Journal of Physiology (Endocrinol Metab) 285: E729-E736,

16 12. Goodpaster B, He J, Watkins S and Kelley D. Skeletal muscle lipid content and insulin resistance: evidence for a paradox in endurance-trained athletes. Journal of Clinical Endocrinology and Metabolism 86: , Goodpaster BH and Kelley DE. Role of muscle in triglyceride metabolism. Curr Opin Lipido 9: , He J, Watkins S and Kelley DE. Skeletal muscle lipid content and oxidative enzyme activity in relation to muscle fiber type in Type 2 diabetes and obesity. Diabetes 50: 50: , Helge J and Dela F. Effect of training on muscle triacylglycerol and structural lipids: a relation to insulin sensitivity? Diabetes 52: , Kelley D, Goodpaster B, Wing R and Simoneau J. Skeletal muscle fatty acid metabolism in association with insulin resistance, obesity and weight loss. American Journal of Physiology (Endocrinol Metab) 277: E1130-E1141, Kelley D, He J, Menshikova E and Ritov V. Dysfunction of mitochondria in human skeletal muscle in type 2 diabetes mellitus. Diabetes 51: , Kelley D, Mokan M and Mandarino L. Intracellular defects in glucose metabolism in obese patients with noninsulin-dependent diabetes mellitus. Diabetes 41: , Lillioja S, Mott DM, Zawadzki JK, Young AA, Abbott WG and Bogardus C. Glucose Storage Is A Major Determinant of in Vivo "Insulin Resistance" in Subjects with Normal Glucose Tolerance. Journal of Clinical Endocrinology and Metabolism 62: ,

17 20. Malenfant P, Tremblay A, Doucet E, Imbeault P, J-A S and Joanisse D. Elevated intramyocellular lipid concentration in obese subjects is not reduced after diet and exercise training. American Journal of Physiology (Endocrinol and Metab) 280: E632- E639, Marshall S, Bacote V and Traxinger R. Discovery of a metabolic pathway mediating desensitization of the glucose transport system: role of hexosamine biosynthesis in the induction of insulin resistance. Journal of Biological Chemistry 266: , Mootha V, CM L, Eriksson K-F, Subramanian A, Sihag S, Lehar J, Puigserver P, Carlsson E, Ridderstrale M, Laurila E, Houstis N, Daly M, Patterson N, Mesirov J, Golub T, Tamayo P, Spiegelman B, Lander E, Hirschhorn J, Altshuler D and Groop L. PGC-1alpha responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nature Genetics 34: , Mott DM, Lillioja S and Bogardus C. Overnutrition Induced Decrease in Insulin Action for Glucose Storage: In Vivo and in Vitro in Man. Metabolism 35: , Obici S and Rossetti L. Minireview: Nutrient sensing and the regulation of insulin action and energy balance. Endocrinology 144: , Pan DA, Lillioja S, Kriketos AD, Milner MR, Baur LA, Bogradus C, Jenkins AB and Storlien LH. Skeletal muscle triglyceride levels are inversely related to insulin action. Diabetes 46: , Patti M, Butte A, Crunkhorn S, Cusi K, Berria R, Kashyap S, Miyazaki Y, Kohane I, M C, Saccone R, Landaker E, Goldfine A, Mun E, DeFronzo R, Finlayson J, Kahn C and Mandarino L. Coordinated reduction in genes of oxidative metabolism 17

18 in humans with insulin resistance and diabetes: potential roles of PGC1 and NRF-1. Proceedings of the National Academy of Science USA 100: , Perseghin G, Price T, Falk-Petersen K, Roden M, Cline G, Gerow K, Rothman D and Shulman G. Increased glucose transport-phosphorylation and muscle glycogen synthesis after exercise training in insulin-resistant subjects. New England Journal of Medicine 335: , Perseghin G, Scifo P, DeCobelli F, Pagliato E, Battezzati A, Arcelloni C, Vanzulli A, Testolin G, Pozza G, DelMaschio A and Luzi L. Intramyocellular triglyceride content is a determinant of in vivo insulin resistance in humans: a 1H-13C nuclear magnetic resonance spectroscopy assessment in offspring of type 2 diabetic patients. Diabetes 48: , Petersen K, Befoy D, Dufour S, Dziura J, Ariyan C, Rothman D, DiPietro L, Cline G and Shulman G. Mitochondrial dysfunction in the elderly: possible role in insulin resistance. Science 300: , Petersen K-F, Dufour S, Befroy D, Garcia R and Shulman G. Impaired mitochondrial activity in insulin-resistant offspring of patients with type 2 diabetes. New England Journal of Medicine 350: , Pette D and Staron RS. Cellular and molecular diversities of mammalian skeletal muscle fibers. Rev Physiol Biochem Pharmacol 116: 1-76, Phillips DW, Caddy S, Llic V, Frayn KN, Borthwick AC and Taylor R. Intramuscular triglycerides and muscle insulin sensitivity: evidence for a relationship in nondiabetic subjects. Metabolism. 45: ,

19 33. Radziuk J. Insulin sensitivity and its measurement: structural commonalities among the methods. Journal of Clinical Endocrinology & Metabolism 85: , Roch-Norlund AE, Bergstrom J, Castenfors H and Hultman E. Muscle Gycogen in Patients with Diabetes Mellitus. Acta Med. Scand. 187: , Ruderman NB, Saha AK, Vavvas D and Witters LA. Malonyl-CoA, fuel sensing, and insulin resistance. American Journal of Physiology 276: E1-E18, Saltiel A. The molecular and physiological basis of insulin resistance: emerging implications for metabolic and cardivascular disease. Journal of Clinical Investigation 106: , Schrauwen P, Van Merken Lichtenbelt WD, Saris WM and Westerterp KR. Fat balance in obese subjects: role of glycogen stores. American Journal of Physiology (Endocrinol and Metab) 276: E1027-E1033, Schrauwen-Hinderling V, Schrauwen P, Hesselink M, van Engelshoven J, Nicolay K, WH S, Kessels A and Kooi M. The increase in intramyocellular lipid content is a very early response to training. Journal of Clinical Endocrinology & Metabolism 88: , Sherman W, Katz A, Cutler C, Withers R and Ivy J. Glucose transport: locus of muscle insulin resistance in obese Zucker rats. American Journal of Physiology (Endocrinol and Metab) 255: E374-E382, Shulman GI, Rothman DL, Jue T, Stein P, DeFronzo RA and Schulman RG. Quantitiation of muscle glycogen synthesis in normal subjects and subjects with noninsulin-dependent diabetes by 13 C nuclear magnetic resonance spectroscopy. New England Journal of Medicine 322: ,

20 41. Simon DK and Johns DR. Mitochondrial disorders: clinical and genetic features. Annual Review of Medicine 50: , Simoneau J-A, Veerkamp JH, Turcotte LP and DE K. Markers of capacity to utilize fatty acids in human skeletal muscle: relation to insulin resistance and obesity and effects of weight loss. FASEB J 13: , Thorburn A, Andrikopoulos S and Proietto J. Defects in liver and muscle glycogen metabolism in neonatal and adult New Zealand obese mice. Metabolism 44: , Thorburn AW, Gumbiner B, Bulacan F, Wallace P and Henry RR. Intracellular Glucose Oxidation and Glycogen Synthase Activity Are Reduced in Non-insulindependent (Type II) Diabetes Independent of Impaired Glucose Uptake. Journal of Clinical Investigation 85: , FIGURE LEGENDS Figure 1. Panels A-C: Reprentative PAS staining in skeletal muscle biopsy samples of vastus lateralis obtained from a lean research volunteer (A), an obese research volunteer (B) and from a research volunteer with T2DM (C). Magnification is x40; I: type I fiber, II: type II fiber Panel D reveals PAS staining in skeletal muscle from a lean research volunteer while Panel E is a section from the same biopsy sample after pretreatment with amylase to digestion glycogen and thus is shown as a negative control. Magnification: x10. 20

21 Figure 2. Glycogen content within type 1, type 2a and type 2b muscle fibers are shown for vastus lateralis muscle obtained by percutaneous biopsy from lean nondiabetic, obese non-diabetic and type 2 diabetic volunteers. The units of glycogen content are arbitrary absorbance units measured from PAS staining. There were no significant differences across groups in glycogen content for any of the 3 fiber types. Figure 3. Glycogen content within type 1, type 2a and type 2b muscle fibers is expressed relative to respective values for oxidative enzyme (SDH) activity. As previously described, tissue is vastus lateralis muscle obtained by percutaneous biopsy from lean non-diabetic, obese non-diabetic and type 2 diabetic volunteers. The ratio of GLYC:OX-ENZ is higher in type 2a and type 2b fibers than in type 1 fibers (p<0.001), and in type 2a and type 2b fibers is lower in muscle from lean compared to obese volunteers (p<0.05). 21

22 Table 1. Clinical Characteristics of Research Volunteers. Lean (9 F / 7 M) Obese (5 F / 10 M) Type 2 DM (8 F / 6 M) Age (yrs) 35±2 * 42±2 * 52±2 Body Mass Index (kg/m 2 ) 23.5±0.6 * 32.0± ±1.7 HbA1c 5.2±0.3% * 5.3±0.3% * 8.0±0.2% Fasting Glucose (mg/dl) 86±2 * 96±2, * 213±18 Fasting Insulin (µu/ml) 7.0±0.7 * 11.8± ±4.8 HOMA-IR 1.5±0.2 * 2.8±0.3, * 8.4±2.7 p<0.05; Obese versus Lean volunteers. * p<0.05; Type 2 DM versus Lean or Obese. 22

23 Table 2. Muscle SDH and Oil Red O. Lean Obese Type 2 DM SDH Type ±2.3 * 40.7± ±3.5 Type 2a 40.2±2.4 * 25.1± ±3.0 Type 2b 31.2±2.9 * 18.8± ±2.5 Glycogen : SDH Type ± ± ±0.40 Type 2a 2.72± ± ±0.52 Type 2b 3.49± ± ±0.86 Oil Red O Type ±1.1 * 16.7± ±2.0 Type 2a 9.5±1.1 * 11.4± ±2.2 Type 2b 7.0± ± ±1.2 Oil Red O : SDH 23

24 Type ±0.02 * 0.44± ±0.12 Type 2a 0.25±0.03 * 0.63± ±0.16 Type 2b 0.25±0.04 * 0.81± ±0.13 p<0.05; Obese versus Lean volunteers; * p<0.05; Type 2 DM versus Lean or Obese. 24

25 25 E R3

26 26 E R3

27 27 E R3