Progressive Hyperglycemia across the Glucose Tolerance Continuum in Older Obese Adults Is Related to Skeletal Muscle Capillarization and Nitric Oxide Bioavailability

Muscle capillarization is impaired with advancing glucose intolerance, and plasma nitric oxide may provide a marker of this progression.

Abstract

Context:

Reduced tissue nutrient exposure may aid in the progression of glucose intolerance.

Objective:

The aim of the study was to examine peripheral tissue glucose disposal in relation to muscle capillarization and plasma nitric oxide bioavailability.

Design:

Participants were carefully matched for age, adiposity, and lipid status and stratified into normal (n = 20), impaired (n = 20), and type 2 diabetic (n = 20) glucose-tolerant groups.

Setting:

The study was conducted in an outpatient setting at a Clinical Research Unit.

Participants:

Older, obese men and women (n = 60; age, 65 ± 1 yr; body mass index, 32.7 ± 0.5 kg/m²) participated in the study.

Intervention:

We performed a cross-sectional study.

Main Outcome Measures:

Body composition, energy metabolism, aerobic fitness (maximum oxygen consumption), insulin sensitivity (glucose clamp), vastus lateralis muscle morphology, and plasma nitric oxide were assessed.

Results:

Although subjects were identical with respect to age, body composition, energy expenditure, and lipid status, insulin-stimulated glucose disposal and maximum oxygen consumption showed progressive decline with increasing glucose intolerance. Muscle fiber type composition and mitochondrial density were not different between groups. However, capillary density markedly declined with advancing glucose intolerance (1.86 ± 0.31, 1.70 ± 0.28, 1.42 ± 0.24 capillary/fiber; P < 0.05), a trend that was mirrored by fasting plasma nitric oxide concentrations (26.3 ± 3.6, 19.8 ± 2.3, 15.2 ± 2.1 μmol/liter; P < 0.05). Furthermore, skeletal muscle capillary density correlated with insulin sensitivity (r = 0.65; P < 0.001).

Conclusions:

Impaired muscle capillarization and reduced nutrient exposure to the metabolizing tissue may play a major role in the progression of insulin resistance across the glucose tolerance continuum, independent of age, adiposity, lipid status, and resting energy metabolism. These data also highlight plasma nitric oxide as a potential surrogate marker of these impairments and may be indicative of the progression toward type 2 diabetes.

Skeletal muscle insulin resistance is the key impairment that initiates oral glucose intolerance, and thus underlies the pathogenesis of type 2 diabetes mellitus (T2DM) (1). Although defective intramyocellular insulin signaling and nutrient partitioning are central to the development of skeletal muscle insulin resistance (2), it is acknowledged that tissue nutrient exposure and muscle capillarization are also important contributory variables (3). This indicates that hyperglycemia may arise not only from defective intracellular nutrient metabolism but also as a result of impaired delivery of nutrients to the metabolizing tissues.

Obesity is hallmarked by an underlying state of insulin resistance. Previous work has demonstrated that skeletal muscle blood flow (4), microvascular recruitment (5), and glucose delivery (6) are impaired in obese humans. Furthermore, muscle capillary blood volume (7) and permeability surface area (8) are also reduced in individuals with T2DM. The ubiquitous signaling molecule, nitric oxide, is a powerful modulator of arteriolar vasodilation and capillary bed perfusion (9). Experimental hyperinsulinemia stimulates nitric oxide production in vitro (10) and elevates plasma nitric oxide levels in healthy humans (11, 12). However, besides evidence that skeletal muscle nitric oxide synthase activity (13) and protein content (14) are reduced in T2DM patients, nitric oxide metabolism has not been assessed in relation to muscle perfusion in obesity-related disease. The findings mentioned above indicate that exposure of the metabolizing tissues (skeletal muscle) to nutrients in the local circulation may decrease as one advances through the glucose tolerance continuum due to impaired tissue perfusion as a result of reduced nitric oxide bioavailability.

Mechanistic insight into the role of skeletal muscle capillary function and perfusion may aid in the development of treatment strategies for reversing hyperglycemia. Thus, we examined skeletal muscle capillarization and circulatory nitric oxide concentrations during basal and insulin-stimulated conditions in individuals representative of the whole glucose tolerance continuum. To control for factors that influence insulin resistance and skeletal muscle function, we studied men and postmenopausal women matched for age, body mass and adiposity, and lipid status. We hypothesized that with advancing glucose intolerance, impaired insulin-stimulated peripheral tissue glucose uptake would be related to a reduction in nutrient availability to the tissue, as indicated by decreased skeletal muscle capillarization and nitric oxide levels.

Subjects and Methods

Subjects

Following ethical approval from the Institutional Review Board, older, obese, and sedentary volunteers were recruited from the local community. These included normal glucose tolerant (Obese-NGT; n = 20), impaired glucose tolerant (Obese-IGT; n = 20), and newly diagnosed T2DM (Obese-T2DM; n = 20) individuals who were matched for age, body mass index (BMI), whole body and abdominal adiposity, and lipid status. Glucose tolerance was characterized by an oral glucose tolerance test (OGTT) (15). Full health and physical screening was performed to assess suitability for the study. Individuals were excluded if they were smokers, had undergone more than a 2-kg weight change in the last 6 months, engaged in a regular exercise program (>30 min of daily activity), and had evidence of hematological, renal, hepatic, or cardiovascular disease. Participants taking glucose-lowering compounds or any drug known to interfere with the outcome variables were safely titrated off their medications by their physician before testing. Subjects also underwent a resting electrocardiogram and an exercise stress test to identify any cardiovascular abnormality that could be exacerbated by a maximal exercise bout. All participants provided informed written consent before clinical testing.

Inpatient control period

All participants resided in the Clinical Research Unit at the Cleveland Clinic for 3 d while metabolic tests were performed. Each subject received a standard hospital diet isocaloric to their own metabolic needs [resting energy expenditure (REE) assessed by indirect calorimetry, adjusted for an activity factor related to a sedentary lifestyle (16)]. Macronutrients were provided as 55, 30, and 15% of calories derived from carbohydrate, fat, and protein, respectively.

Body composition

Body height and weight were determined using standard procedures. Whole body and abdominal adiposity were determined as previously described using dual-energy x-ray absorptiometry (Lunar iDXA, Madison, WI) (17) and computed tomography scanning (SOMOTOM Sensation 16 Scanner; Siemens Medical Solutions, Malvern, PA) (18), respectively.

Maximal aerobic capacity

A maximal exercise test was used to determine maximum oxygen consumption (VO₂max) during a graded treadmill exercise bout performed to exhaustion, as previously described (18).

Energy metabolism and insulin sensitivity

After an overnight fast, participants underwent REE measurements via ventilated-hood indirect calorimetry. Substrate oxidation rates were calculated according to Frayn (19), while overnight urinary nitrogen excretion was measured to correct for protein oxidation. After basal indirect calorimetry measurements, a 3-h, 40 mU/m² · min hyperinsulinemic euglycemic (90 mg/dl) clamp proceeded (20, 21). Mean space-corrected exogenous glucose infusion rates during the final 30 min of steady-state hyperinsulinemia are presented as peripheral tissue glucose disposal rates (GDRs). We have previously described these procedures in detail (17). In addition, at baseline and during the final 30 min of the clamp, percutaneous skeletal muscle (vastus lateralis) biopsies were taken under local anesthesia, using the Bergström needle technique.

Blood chemistry

As a marker of nitric oxide, plasma concentrations of nitrate and nitrite, stable end-products of nitric oxide metabolism, were measured using the Griess reaction (Cayman Chemicals, Ann Arbor, MI) (22). Fasting plasma triglycerides, cholesterol, and lipoprotein cholesterol subfractions (high-density lipoprotein, low-density lipoprotein, and very low-density lipoprotein) were analyzed by enzymatic assay on a high throughput clinical platform (Roche Modular Diagnostics, Indianapolis, IN). Plasma glucose was measured using an automated glucose oxidase assay (YSI 2300 Stat Plus; YSI Inc., Yellow Springs, OH). Plasma insulin and leptin were determined using RIAs (Millipore, Billerica, MA).

Muscle processing

Skeletal muscle biopsies were obtained in 40 individuals (16 Obese-NGT, 15 Obese-IGT, and nine Obese-T2DM). Connective tissue and adipose deposits were removed, and the samples were then mounted in Tragacanth gum and frozen in liquid nitrogen-cooled isopentane before being stored at −80 C. Serial 10-μm cross-sections were obtained using a cryostat (Leica CM 3050; Leica Microsystems, Bannockburn, IL) and Accu-Edge high profile blades (Sakura Finetek, Torrance, CA), and mounted on poly-l-lysine-coated glass slides allocated to each staining procedure.

Muscle fiber typing

A myofibrillar ATPase precipitation method was used to identify muscle fiber types, as described by Brooke and Kaiser (23). Stained sections were viewed under a light microscope (Olympus Model BX61; Olympus, Center Valley, PA). Images were captured using a digital camera (model ORCA-ER; Hamamatsu Photonics, Hamamatsu City, Japan) and quantified using NIH Image software (http://rsb.info.nih.gov/nih-image/). Fiber type percentage and the average fiber surface area were determined.

Muscle capillarization

Capillaries were identified using a periodic acid-Schiff stain after pretreatment with amylase, as described by Andersen (24), with the addition of a hematoxylin stain for nuclei identification. Sections were analyzed as described above. The number of capillaries per muscle fiber, the number of capillaries per square millimeter of surface area, and the average number of capillary contacts per fiber were determined. All images were corrected for background periodic acid-Schiff staining by quantifying a non-amylase-incubated section for each sample.

Muscle mitochondrial density

Mitochondrial density was determined via immunohistochemistry using an antioxphos complex IV subunit I mouse IgG_2a monoclonal primary antibody and an Alexa Fluor 488 Goat antimouse IgG_2a secondary antibody (Invitrogen Molecular Probes, Eugene, OR) as described by Shaw et al. (25). Fluorescence intensity was assessed under a fluorescence microscope (model BX61; Olympus, Center Valley, PA) at excitation/emission λ = 495/519 nm. Mitochondrial density was calculated as the percentage of the visible area stained corrected for background fluorescence using a section stained without primary antibody for each sample. Analysis was performed using Image Pro v5.1 (Media Cybernetics, Bethesda, MD).

Statistics

All data were analyzed using GraphPad Prism version 4.00 for Windows (GraphPad Software, San Diego, CA; www.graphpad.com). Data represent mean ± sem, and statistical significance was achieved if P < 0.05. To examine differences between group means, one-way ANOVAs were performed. Bonferroni's post hoc test was used to identify differences between specific group means. Additionally, correlations were examined, and Pearson's product moment correlation coefficients were calculated to identify significant relationships between variables.

Results

Subject characteristics

Table 1 demonstrates that the Obese-NGT, Obese-IGT, and Obese-T2DM groups showed no statistically significant differences with regard to age, body mass, body fat percentage, abdominal adiposity, circulatory leptin, fasting lipemia (triglycerides and cholesterol), and resting energy metabolism (energy expenditure and respiratory exchange ratio) (Table 1; all comparisons, P > 0.05). However, aerobic fitness, as measured by VO₂max during exhaustive exercise, was significantly lower in Obese-IGT and Obese-T2DM individuals when compared with the Obese-NGT group (Table 1; both comparisons, P < 0.05).

Table 1.

Subject characteristics

	Obese-NGT	Obese-IGT	Obese-T2DM
n	20	20	20
No. of males/females	9/11	11/9	10/10
Age (yr)	64 ± 2	66 ± 2	65 ± 2
Body mass (kg)	94.2 ± 3.3	95.9 ± 3.4	93.1 ± 2.7
BMI (kg/m2)	32.0 ± 1.0	33.1 ± 0.7	32.8 ± 0.7
WC (cm)	112.5 ± 3.4	114.2 ± 2.5	112.7 ± 2.7
DXA fat (%)	41.3 ± 1.4	41.5 ± 1.5	40.5 ± 1.9
AAT (cm2)	435.3 ± 31.0	431.0 ± 32.3	399.5 ± 42.2
Leptin (ng/ml)	27.0 ± 5.7	23.2 ± 3.9	25.1 ± 5.3
TG (mg/dl)	184.4 ± 23.6	161.5 ± 15.8	173.5 ± 17.4
Chol (mg/dl)	206.2 ± 5.8	219.6 ± 7.2	196.2 ± 10.2
REE × 103 (kcal/kg · min)	11.8 ± 0.3	11.3 ± 0.4	11.9 ± 0.6
RER (a.u.)	0.83 ± 0.01	0.84 ± 0.02	0.86 ± 0.02
Resting Cox (mg/kg · min)	1.11 ± 0.13	1.15 ± 0.14	1.25 ± 0.12
Resting Fox (mg/kg · min)	0.54 ± 0.07	0.49 ± 0.06	0.62 ± 0.07
VO2max (ml/kg · min)	24.8 ± 0.7	22.4 ± 0.4a	21.1 ± 0.6a

Data represent mean ± sem. WC, Waist circumference; DXA, dual x-ray absorptiometry; AAT, abdominal adipose tissue; TG, fasting plasma triglycerides; Chol, fasting plasma total cholesterol; RER, respiratory exchange ratio; a.u., arbitrary units; C_ox, carbohydrate oxidation rate; F_ox, fat oxidation rate; VO₂max, maximal oxygen consumption during exhaustive aerobic exercise.

Group means were compared via one-way ANOVA.

^aP < 0.05 vs. Obese-NGT group.

Glucose tolerance

Table 2 indicates the progressive glucose intolerance across the three groups. The advancing impairment in peripheral tissue insulin-stimulated GDR across the three groups is shown. This trend was mirrored by impairments in nonoxidative glucose disposal (1.55 ± 0.33, 0.96 ± 0.20, and 0.30 ± 0.10 mg/kg · min for the Obese-NGT, Obese-IGT, and Obese-T2DM groups, respectively; P < 0.05). Fasting plasma glucose was significantly higher in the Obese-T2DM group compared with the IGT or NGT groups, whereas 2-h plasma glucose levels during OGTT were significantly elevated in Obese-IGT and Obese-T2DM individuals, compared with the Obese-NGT group (Table 2; group mean comparisons, P < 0.05). In addition, insulin secretory responses to oral glucose ingestion (ΔI/ΔG) and pancreatic β-cell function (ΔI/ΔG÷IR) were also progressively impaired across the glucose tolerance continuum (Table 2; P < 0.05).

Table 2.

Glucose tolerance

	Obese-NGT	Obese-IGT	Obese-T2DM
GDR (mg/kg/min)	2.90 ± 0.36	1.98 ± 0.17a	1.78 ± 0.26b
FPG (mg/dl)	98.6 ± 2.9	103.6 ± 2.4	128.0 ± 5.2b,c
2-h OGTT (mg/dl)	121.8 ± 3.4	160.7 ± 3.4a	235.5 ± 11.9b,c
FPI (μU/ml)	16.7 ± 1.5	16.0 ± 1.6	19.1 ± 2.0
ΔI/ΔG (a.u.)	34.5 ± 6.5	17.7 ± 2.2a	12.8 ± 2.5b,c
ΔI/ΔG÷IR (a.u.)	7.12 ± 1.59	2.41 ± 0.42a	1.29 ± 0.28b,c

Data represent mean ± sem. GDR, Glucose disposal rate during 40 mU/m² · min hyperinsulinemic euglycemic clamp; FPG, fasting plasma glucose; 2-h OGTT, plasma glucose 2 h after OGTT; FPI, fasting plasma insulin; ΔI, change in plasma insulin during first 30 min of OGTT; ΔG, change in plasma glucose during first 30 min of OGTT; IR, the reciprocal of the GDR per unit of insulin during the hyperinsulinemic clamp; a.u., arbitrary units.

Group means were compared via one-way ANOVA: ^a P < 0.01 vs. Obese-NGT group;

^bP < 0.01 vs. Obese-NGT group;

^cP < 0.01 vs. Obese-IGT group.

Skeletal muscle morphology

Figure 1 indicates representative photomicrographs of the histochemical stains. Muscle fiber type composition (fiber number and cross-sectional surface area) was not different between groups (P > 0.05), whereas mitochondrial density was slightly but nonsignificantly reduced in the Obese-T2DM subjects (Table 3; P = 0.25). Meanwhile, the capillary to fiber ratio, the number of capillaries per square millimeter of surface area (capillary density), and the mean number of capillary contacts per fiber were all found to be decreased with advancing glucose intolerance (Table 3; all P < 0.05).

Fig. 1.

Representative photomicrographs of skeletal muscle morphology. Vastus lateralis skeletal muscle biopsies were obtained, and histomounts were stained using an amylase-periodic acid Schiff histochemical stain (A), and an ATPase histochemical stain (B). I, Type I muscle fibers; II, type 2 muscle fibers. Blue staining indicates nuclei; purple staining indicates capillaries. The inset bar represents 50 μm.

Table 3.

Skeletal muscle morphology

	Obese-NGT	Obese-IGT	Obese-T2DM
Type I fibers (%)	44.9 ± 4.4	45.2 ± 3.2	46.7 ± 4.5
Type II fibers (%)	55.1 ± 4.4	54.8 ± 3.2	53.3 ± 4.5
FSA type I (μm2)	3009 ± 148	3161 ± 162	3082 ± 241
FSA type II (μm2)	2721 ± 189	2655 ± 203	2799 ± 222
C/F ratio	1.86 ± 0.31	1.70 ± 0.28	1.42 ± 0.24a,b
CD (mm−2)	471 ± 22	422 ± 20c	400 ± 18a
CC/F ratio	4.28 ± 0.08	4.11 ± 0.06c	3.99 ± 0.08a,b
MD (% of area stained)	15.1 ± 1.7	14.9 ± 1.9	13.9 ± 1.1

Data represent mean ± sem. Skeletal muscle histochemical procedures were performed in 40 individuals (16 Obese-NGT, 15 Obese-IGT, and nine Obese-T2DM). FSA, Skeletal muscle fiber cross-sectional surface area; C/F, skeletal muscle capillary to fiber ratio; CD, capillaries per square millimeter of muscle (capillary density); CC/F, capillary contacts per fiber; MD, mitochondrial density; a.u., arbitrary units.

Group means were compared via one-way ANOVA: ^a P < 0.05 vs. Obese-NGT group;

^bP < 0.05 vs. Obese-IGT group;

^cP < 0.05 vs. Obese-NGT group.

Plasma nitric oxide (nitrate/nitrite)

Although peripheral plasma nitric oxide (nitrate/nitrite) concentrations did not respond to hyperinsulinemia in any of the obese groups, there was a progressive decline in basal and insulin-stimulated nitric oxide in relation to decreasing glucose tolerance (Fig. 2, A and B; P < 0.05).

Fig. 2.

Progressive reduction in plasma nitric oxide levels across the glucose tolerance continuum. Older, obese men and women were divided into three age-, BMI-, adiposity-, and lipid status-matched groups according to their state of glucose tolerance: normal glucose tolerance (Obese-NGT; n = 20), impaired glucose tolerance (Obese-IGT; n = 20), and T2DM (Obese-T2DM; n = 20). Plasma nitrite/nitrate concentrations, a marker of nitric oxide production (NOx), were measured in the basal (A) and insulin-stimulated (B) states. NOx levels were also progressively decreased with advancing glucose intolerance but unresponsive to hyperinsulinemia in each group. Significant differences between means were identified by one-way ANOVA (*, P < 0.05).

Correlation analyses

Insulin-stimulated glucose disposal was not found to be related to mitochondrial density (r = 0.12; P = 0.51; n = 40). However, glucose disposal was related to VO₂max (r = 0.41; P < 0.01; n = 60) and showed a significant positive relationship with skeletal muscle capillary density (r = 0.65, P < 0.001; Fig. 3) and a nonsignificant positive trend with plasma nitric oxide concentrations (r = 0.34; P = 0.17).

Fig. 3.

The relationship between insulin sensitivity and skeletal muscle capillary density. Skeletal muscle histochemical procedures were performed in 40 individuals (16 Obese-NGT, 15 Obese-IGT, and nine Obese-T2DM). Regression analyses revealed that insulin-stimulated glucose disposal (GDR, mg/kg · min) was significantly related to skeletal muscle capillary density (CD, mm⁻²; r = 0.65; P < 0.001).

Discussion

We demonstrate for the first time that peripheral tissue insulin sensitivity is related to skeletal muscle capillarization and nitric oxide bioavailability in obese individuals, representative of the whole glucose tolerance continuum. Due to the homogeneity of our study group with respect to age, body composition, and lipid status, these data support the view that reduced tissue nutrient exposure may aid in the progression of glucose intolerance. These novel findings also highlight plasma nitric oxide as a potential biomarker of these impairments during the development of T2DM in obese individuals.

Several factors can influence insulin sensitivity and metabolic function. Age, physical activity, body weight, adiposity, and circulating lipid levels have each been shown to be independently associated with our primary outcome variables (insulin sensitivity and muscle perfusion) (4–6, 26–28), and therefore each one may play a role in the onset of glucose intolerance. In this study, we prospectively matched our three groups for these variables so that any differences in skeletal muscle morphology or nitric oxide would be a function of glucose tolerance per se. By using this approach, we additionally found that basal energy expenditure, respiratory exchange ratios, substrate oxidation rates, and skeletal muscle fiber type composition were also matched between the three groups. This suggests that the decrease in VO₂max and insulin-stimulated glucose disposal seen with advancing glucose intolerance is unlikely to be a result of differences in intracellular nutrient metabolism. When interpreted in conjunction with our finding that muscle capillary density and nitric oxide bioavailability was progressively decreased with advancing glucose intolerance, it is likely that the group differences in aerobic fitness and insulin sensitivity arise as a result of reduced exposure of the metabolizing tissues to nutrients in the systemic circulation.

In their classical study, Laakso et al. (4) demonstrated decreased insulin sensitivity of arterioles in obese individuals. This observation has since been confirmed by several groups (29–31). In addition, lipid-induced insulin resistance decreases the responsiveness of the muscle microvasculature to insulin (26). Reduced insulin-mediated capillary recruitment lessens the perfusion of the skeletal muscle, limiting the potential for nutrient uptake and metabolism, but the mechanism by which this arises is not fully understood. Recently, capillary recruitment by insulin was shown to be blocked by the nitric oxide synthase inhibitor, N-monomethyl-L-arginine (32, 33), indicating a possible nitric oxide-dependent mechanism of endothelial insulin resistance. Although both skeletal muscle perfusion and blood flow have been shown to be reduced in human obesity (4, 5, 34), our current investigation demonstrates that both skeletal muscle perfusion and plasma nitric oxide levels are increasingly impaired with advancing glucose intolerance, irrespective of the degree of adiposity. This finding prompts a novel hypothesis that endothelial insulin resistance may be an underlying cause of oral glucose intolerance, rather than a consequence of the diabetic state. However, further work will be required to confirm this statement.

Nitric oxide metabolism has been sparsely studied in relation to obesity and hyperglycemic states such as type 2 diabetes. Reduced nitric oxide synthase muscle content and activity have been presented in diabetic patients (13, 14), but to our knowledge the current study is the first to indicate reduced plasma nitrate/nitrite levels in line with increasing peripheral tissue insulin resistance. Nitrates/nitrites are stable degradation products of nitric oxide that, collectively, are a useful marker of endogenous nitric oxide production (22). Previous work has demonstrated elevations in plasma and urinary nitric oxide levels during experimental hyperinsulinemia in lean healthy humans (11, 12). In contrast, we saw no alterations in plasma nitric oxide under insulin-stimulated conditions in our subjects—further evidence for endothelial insulin resistance in obese individuals. In support of this, Weiss et al. (35) demonstrated failure of oral glucose ingestion to raise plasma nitric oxide in overweight and mildly insulin-resistant humans. In addition, analysis of skeletal muscle capillaries during insulin stimulation (data not shown) failed to show any insulin-mediated effects, perhaps indicative of capillary resistance to insulin-mediated vasodilation. However, we interpret this latter result with caution because a more sensitive immunohistochemical method is required to adequately detect and visualize small changes in capillary surface area (36). Nonetheless, we hereby present novel evidence showing that the progression of glucose intolerance is related not only to a decline in skeletal muscle capillarization, but also to nitric oxide bioavailability. Furthermore, this indicates that plasma nitric oxide may be a useful biomarker of the progressive development of T2DM in obese subjects.

It is known that nitric oxide plays a key role in pancreatic function (37, 38). Therefore, with regard to the onset of diabetes, there is the likelihood that nitric oxide bioavailability affects insulin secretion from pancreatic β-cells. In the current study, we examined indices of insulin secretion after oral glucose ingestion and found a progressive decrease in insulin secretion (ΔI/ΔG) as well as a decrement in β-cell function (ΔI/ΔG÷IR), across the glucose tolerance continuum. However, no relationships with nitric oxide bioavailability were identified. That said, only indirect markers of β-cell function were examined; thus, future work examining specific insulin secretory mechanisms across the range of glucose tolerance would also be prudent.

Although mitochondrial respiration has been shown to be suppressed in the skeletal muscle of patients with T2DM (39), when normalized to DNA content or citrate synthase activity (markers of mitochondrial density), oxidative phosphorylation appears not to be different between BMI-matched normal glucose tolerant and T2DM individuals (40). Mitochondrial function may therefore influence the impairments in oxygen consumption and glucose uptake seen as one progresses through the glucose tolerance continuum. Although we found a slightly lower mitochondrial density in our Obese-T2DM group, reduced insulin-stimulated glucose disposal was only related to skeletal muscle capillary density, and not mitochondrial density. Furthermore, the trend between glucose disposal and circulatory nitric oxide concentrations in our data provides additional support of the notion that tissue nutrient availability may play a central role in the progression of glucose intolerance.

Impaired tissue capillarization is common in T2DM and is central to the advancement of cardiovascular anomalies that arise in untreated diabetic individuals (3). Mechanistic insight into the role of skeletal muscle capillary function and perfusion may therefore aid in the development of treatments for patients exhibiting cardiometabolic risk factors. This study provides evidence to suggest that impaired capillarization of skeletal muscle and reduced nitric oxide bioavailability may aid in the progression toward diabetes, independent of age, body mass, adiposity, fasting leptin, lipemia, and resting energy metabolism. Although a lean or young control group was not studied, these findings support the hypothesis that in older obese individuals, reduced nutrient exposure to skeletal muscle may be an underlying cause of progressive hyperglycemia across the glucose tolerance continuum and that plasma nitric oxide may serve as an insightful biomarker of such defects.