OBJECTIVE—Whole-body protein metabolism is abnormal in suboptimally controlled type 2 diabetes and obesity. We hypothesized that glycemia, insulin resistance, and waist circumference modulate these alterations in type 2 diabetes and, to a lesser extent, in individuals without type 2 diabetes.

RESEARCH DESIGN AND METHODS—In 88 lean and obese subjects without and 40 with type 2 diabetes on an inpatient protein-controlled isoenergetic diet for 7 days, whole-body protein turnover was measured using the fed-fasted 60-h oral 15N-glycine method. Nitrogen flux was determined from urinary 15N urea and protein synthesis, breakdown and net balance calculated. Indexes of diabetes control, resting energy expenditure (REE), and body composition were assessed.

RESULTS—Higher protein turnover in obese subjects was further increased, and net balance was lower in type 2 diabetes. Waist-to-hip ratio and ln homeostasis model assessment of insulin resistance (HOMA-IR) explained 40% of the variance in flux in type 2 diabetes; fat-free mass and lnHOMA-IR explained 62% in subjects without type 2 diabetes. Overall, fasting glucose explained 16% of the variance in net balance. In type 2 diabetes, net balance correlated negatively with fasting glucose in men and positively with hip circumference in women.

CONCLUSIONS—Kinetics of whole-body protein metabolism are elevated, and net balance is diminished in type 2 diabetes, independently of obesity. Elevated flux is associated with greater visceral adiposity, REE, and insulin resistance of glucose. In type 2 diabetic men, these alterations worsened with magnitude of hyperglycemia. In type 2 diabetic women, larger hip circumferences may protect against such alterations. Our findings suggest that dietary protein requirements may be greater in type 2 diabetes to offset a reduced net balance, aggravated as glycemia increases, especially in men.

It is well established that the diabetic state affects protein metabolism and that insulin plays a role in achieving net protein accretion. There is evidence for altered protein metabolism in type 1 diabetes (1,2), and we have reported accelerated kinetics of integrated fed-fasted whole-body protein metabolism in obese type 2 diabetic subjects with hyperglycemia compared with obese control subjects after adjusting for fat-free mass (FFM), sex, and age as confounding variables (36). Elevated kinetics were associated with lower nitrogen balance despite generous protein intakes and weight maintenance. We (3) and others (7) reported increased catabolism of myofibrillar proteins during hyperglycemia, but using 13C-leucine methodology in the postabsorptive state, others found no abnormalities in net protein balance or change from improvement in glycemia (8,9). It is noteworthy that such type 2 diabetic subjects had higher insulin levels than the control subjects, a criterion of insulin resistance of glucose metabolism. Our findings, in 60-h studies, had the potential of detecting defects in the postprandial as well as interprandial states.

Therefore, differing experimental designs among published studies likely explain the controversy over whether type 2 diabetes alters protein turnover. A key difference is that with the 13C-leucine method, the only data previously available were from the postabsorptive catabolic state and from hyperinsulinemic-euglycemic clamps that decrease catabolism and thereby substrate availability for synthesis (10,11). To address this problem, we have combined hyperinsulinemia with both aminoacidemia and glycemia clamped (12). We found resistance of protein metabolism to the action of insulin in type 2 diabetic (13) and obese (14) individuals. Examples of other design differences include absence of prior diet control (15), varied levels of diabetes control preceding hyperinsulinemic clamps (11), control groups of only obese subjects who themselves have insulin resistance of protein metabolism (11,15,16), and adjustment of kinetic data for body composition not being specified (7,10,15,16). The present study takes as many of these design variables as feasible into account to question 1) whether alterations of whole-body protein metabolism are related to the magnitudes of hyperglycemia and insulin resistance of glucose metabolism in type 2 diabetes and 2) to what extent similar alterations may be associated with insulin resistance of glucose and body fat distribution across a diverse adult population. We used the fed-fasted 15N-glycine method to assess whole-body protein metabolism in men and women with or without type 2 diabetes with ranges of body fat and its distribution and of insulin sensitivity of glucose. Our prior studies of a range of subjects (36) combined with unpublished data in subjects of other studies (1214,17,18) provided a large database to test these questions.

Subjects and diet

Caucasian adult subjects (88 [31 men and 57 women] without and 40 [15 men and 25 women] with type 2 diabetes) were recruited and screened by medical history, physical examination, and laboratory investigation (3,12). Although recruitment was done over a period of years, type 2 diabetic and nondiabetic subjects were studied concurrently, minimizing potential temporal confounding. They were admitted to the clinical investigation unit of the Royal Victoria Hospital for 7 days, after written informed consent. The study protocols were approved by the institutional human ethics review board. Obesity was defined as BMI >30 kg/m2. They had no hypertension, hepatic, thyroid, cardiovascular, hematologic, renal, or pulmonary dysfunction. Smokers, individuals requiring insulin, or those with unusual protein intakes were excluded. Their only exercise was walking in or around the hospital. Three days before admission, oral antihyperglycemic agents were stopped. They received a weight-maintaining liquid formula diet (Ensure; Ross Laboratories, Montreal, Canada) based on their resting energy expenditure (REE), determined by the Harris-Benedict equation or indirect calorimetry and multiplied by a factor of 1.5–1.7 for physical activity. Protein intake was 15% of energy, at 0.9–1.3 g/kg body wt [1.73 ± 0.02 g/kg FFM/day]. Six equal meals per day were taken during the 15N-glycine protocol (3,4). Capillary glucose was measured six times daily in type 2 diabetic and three in other subjects (Accusoft Advantage; Roche Diagnostics, Laval, Canada). Daily weight and water intake were recorded. Daily 24-h urine collections were made, nitrogen balance was calculated on the last 3 days (3,4), and urine glucose was measured in type 2 diabetic subjects. Energy lost as glycosuria was added to the diet as 50% glucose polymer and 50% soy oil. Body composition was assessed by bioelectrical impedance analysis (RJL-101A Systems, Detroit, MI) (n = 99) using equations validated for nonobese (19) and obese (20) subjects. In others (n = 13 type 2 diabetic and 16 obese subjects without type 2 diabetes), lean body mass was estimated from daily urinary creatinine excretion (21). The two methods correlated with a slope of 0.83 and r2 of 0.933 (n = 9). Waist circumferences were measured at the minimal value between the xiphoid process and the iliac crest and hips at the maximal protuberance in the trochanteric region.

15N-glycine kinetic studies were done on days 5–7, as detailed (4,5). Briefly, 99 atom percent excess 15N-glycine (Isotec; Sigma-Aldrich, St. Louis, MO) was given at a rate of 0.5 mg 15N · kg body wt−1 · 24h−1 in 5 ml water every 3 h for 60 h to achieve steady-state isotopic enrichment in urine urea. Urine was collected every 3 h. Urea nitrogen was isolated, after removal of ammonia and other nitrogenous compounds with an ion-exchange resin, using urease (U 1500; Sigma-Aldrich) and the Conway procedure (22). Urea 15N enrichment was measured with a dual-inlet, double-collector isotope ratio mass spectrometer (Vacuum Generators, Micromass 602D; Winsford, U.K.) or by mass isotopomer distribution analysis of hexamethylenetetramine (formed by reacting ammonium released from urea with formaldehyde) measured by gas chromatography/mass spectrometry. This method, adapted from Yang et al. (23), permits the measurement of low isotopic enrichment of 15N in urine byproducts, through amplification on a secondary molecule. Correlation of 15N enrichment between the two methods was excellent with a slope of 0.93 and r2 of 0.97. Enrichment curves were drawn for each subject, with pretest background subtracted. The method requires that urea 15N enrichment reaches an isotopic steady state within 60 h, defined as the first plateau with at least four consecutive points (12 h) with a slope not different from zero and before 15N recycling (24). The rate of entry or exit of nitrogen (flux) into the metabolic nitrogen pool was calculated from the mean steady-state value, assuming that the fraction of the administered isotope excreted as urinary 15N urea is the same as the fraction of total amino nitrogen entering the metabolic pool excreted as urinary urea nitrogen (24). We used urinary 15N urea enrichment because, unlike ammonia as end product, urea does not underestimate the synthesis rate (22). Nitrogen intake was known from the controlled diet, and total urinary nitrogen was measured. Thus, protein synthesis and breakdown were calculated from the Picou Taylor-Roberts equation as follows (25): Q = I + B = S + E, where Q is nitrogen flux; I, nitrogen intake, B, protein breakdown; S, protein synthesis; and E, total urinary nitrogen excretion.

Analytical methods

Postabsorptive venous blood samples were drawn on days 1 and 3 of the 15N-glycine study and processed as detailed (36). Glucose, urea, lipid, electrolyte, and A1C determinations were made in the hospital's clinical laboratory. Free fatty acid concentrations were measured using the NEFA C test kit (Wako Chemicals, Richmond, VA), immunoreactive insulin by a single antibody, and charcoal precipitation radioimmunoassay using human standards and labeled hormone from Linco Research (St. Charles, MO), with no cross-reactivity with proinsulin. The homeostasis model assessment of insulin resistance (HOMA-IR) was calculated from fasting glucose and insulin concentrations (26). REE was measured by indirect calorimetry (Deltratrac; Sensor Medics, Yorba Linda, CA).

Statistical methods

Results are presented as means ± SE. Subject characteristics were analyzed by ANOVA, with the Bonferroni adjustment for multiple comparisons. The effects of diabetes and obesity were assessed by ANCOVA, with FFM and protein intake as covariates when found to have a significant predictive value from prior regression analysis (27). HOMA-IR values were log transformed to yield a normal distribution for analyses. Pearson's coefficient was used for all correlations, and when they required controlling for other variables, partial correlation was performed. Stepwise linear regression analysis was conducted with individual predictors of flux, synthesis, breakdown, and net balance found by simple correlations. The analyses were performed with SPSS 11.0 for Windows (SPSS, Chicago, IL). Significance was at P < 0.05.

Subject data are presented in Table 1. Diabetes duration was 4.7 ± 0.7 years. There were more women in the two groups of obese subjects. Mean age differed among groups, but each group included younger and older subjects. Weight, BMI, FFM, and percent body fat were greater in obese subjects but did not differ between obese groups. Body circumferences were greater in obese subjects, with no sex difference. Waist-to-hip ratio was greater in men than women and, for both sexes, greatest in type 2 diabetic patients. REE was lowest in nonobese and highest in type 2 diabetic subjects. Protein intake did not differ per kilogram FFM. Daily nitrogen balance was less in type 2 diabetic subjects and negative (P = 0.013 vs. 0). Fasting triacylglycerol, glycemia, A1C, free fatty acids, and daily mean capillary glucoses were elevated in type 2 diabetes. Fasting plasma insulin was lowest in nonobese and highest in type 2 diabetic subjects. HOMA-IR and lnHOMA-IR were highest in type 2 diabetic subjects.

Whole-body protein kinetics are shown in Fig. 1A–C as means adjusted for FFM and protein intake as covariates for flux and FFM as a covariate for synthesis and breakdown. Within each group, there were no sex effects, and an age effect was seen in type 2 diabetic subjects only. There was a diabetes effect on flux (Fig. 1A), synthesis, and breakdown (Fig. 1B), which were significantly greater, and on net balance (Fig. 1C), which was less in type 2 diabetic subjects than in the other groups. The diabetes effect on all variables remained significant when BMI was also used as a covariate to control for adiposity. There was an obesity effect on flux, synthesis, and breakdown, with rates greater in obese subjects without diabetes than in nonobese subjects; however, net balance did not differ. Flux, synthesis, and breakdown correlated positively with indexes of body fatness and circumferences, and REE, fasting triacylglycerol, glucose and insulin, lnHOMA-IR, and A1C (for flux and breakdown) (r = 0.236–0.431, all P < 0.021) controlled for FFM and protein intake. Net balance correlated negatively with triacylglycerol (r = −0.282, P = 0.002), glucose (r = −0.412, P < 0.001), A1C (r = −0.413, P < 0.001), and lnHOMA-IR (r = −0.255, P = 0.004).

In type 2 diabetic subjects, lnHOMA-IR and insulin, but not glucose, correlated with waist circumference and BMI. Flux, synthesis, and breakdown correlated negatively with age (r = −0.433 to −0.513, all P ≤ 0.006). In women with type 2 diabetes, but not in men, net balance correlated positively with hip circumference (Fig. 2A). In men with type 2 diabetes, but not in women, net balance correlated negatively with fasting glycemia (Fig. 2B).

Stepwise linear regression models were as follows (all P < 0.001). In type 2 diabetic subjects, waist-to-hip ratio and lnHOMA-IR explained 40% of the variance in flux and 43% of that in synthesis; in all the subjects without diabetes, FFM and lnHOMA-IR explained 62% of the variance in flux. In all subjects, fasting glucose was a negative predictor of 16% of the variance in net balance.

Our approach to whole-body integrated fed-fasted protein metabolism was deployed in a large sample of adults varying in body composition and especially in insulin sensitivity of glucose metabolism. The strategy was to safely seek, as marked, a diabetes effect as feasible in a short-term, closely supervised inpatient protocol. The subjects were nonetheless representative of community-dwelling, suboptimally controlled type 2 diabetes. They demonstrated significantly higher values of all components of whole-body kinetics of protein metabolism, with a lesser net balance. The obese subjects without diabetes had flux, synthesis, and breakdown rates intermediate between nonobese and type 2 diabetic subjects. The type 2 diabetic subjects had greater insulin resistance of glucose than the obese subjects, by HOMA-IR. These diabetes effects on whole-body protein metabolism were independent of age, sex, FFM, and body fatness.

Our diet maintained weight and provided sufficient protein to ensure nitrogen equilibrium and thereby preserve lean tissue. However, in the hyperglycemic type 2 diabetic subjects, nitrogen balance was worse and net endogenous balance significantly less than in nonobese subjects by an average of 0.89 g N (5.6 g protein, 28 g lean tissue) per day. These data suggest that protein needs in type 2 diabetes may be increased with poor glycemic control. Type 2 diabetes thus mirrors chronic disease states associated with elevated protein turnover and rapid loss of body protein (28). These type 2 diabetes alterations can be corrected by restoration of normoglycemia with intensive insulin therapy (5,6) or with near-normal glycemia using oral antihyperglycemic agents and sufficient protein intake even with moderate energy restriction (3).

All protein kinetic parameters correlated positively and net balance negatively with fasting glucose, lnHOMA-IR, and A1C. The correlations with net balance remained significant even with nonobese data excluded from the analysis. Also, lnHOMA-IR was a significant predictor of the variance in flux and fasting glucose of net balance. This suggests a worsening of protein metabolism that is concurrent with that of glucose and, since A1C reflects metabolic state before the study, that it is possibly affected by the duration of hyperglycemia. We have reported such relationships between glucose and protein metabolism in obese women (14) and type 2 diabetic men (13), using the hyperinsulinemic, euglycemic, isoaminoacidemic clamp to simultaneously measure insulin resistance of both. Rates of glucose disposal correlated with the protein anabolic response to hyperinsulinemia (14). We had predicted a dose-response relationship between glycemic control and the derangement of protein metabolism (3), with a glucose threshold above which abnormal kinetics appear. In type 2 diabetic women as a group, no significant correlation between protein kinetics and glycemia was found, but in men net balance decreased significantly as hyperglycemia worsened (Fig. 2B). In obese men without diabetes, the lower boundary of the 95% CI for net balance (mean 2.2 g N/day), was 1.4 g N/day, a value that crosses the regression line using data from type 2 diabetic men (Fig. 2B) at a fasting glucose level of 8.2 mmol/l and crosses the regression line using data from all men (r = −0.555, P < 0.001) at 7.6 mmol/l. The same calculations in women, in whom the lower boundary for net balance was 1.9 g N/day, crosses the regression line (r = −0.354, P < 0.001) at a fasting glucose of 6.7 mmol/l. These concentrations may reflect glycemic thresholds above which protein anabolism is blunted.

HOMA-IR scores reflect the postabsorptive state, whereas measurements of protein kinetics and net balance are integrated over the day; still, they were correlated. These findings indicate that the magnitude of derangements in the metabolism of protein relates to that of insulin resistance of glucose. We consider lnHOMA-IR to be an appropriate surrogate for insulin resistance of glucose (26), since three subjects only had impaired fasting glucose, two subjects with type 2 diabetes had low fasting insulin levels, and none were on insulin therapy, factors claimed to limit its validity (29). Indeed, lnHOMA-IR correlated with lnM (glucose uptake/[insulin]) (r = −0.850, P < 0.001, n = 65) in our subjects in whom clamps were performed. The correlation was significant in the nonobese (n = 19), obese (n = 29), and type 2 diabetic (n = 17) groups (r = −0.508, −0.740, and −0.594, respectively; P ≤ 0.012).

We tested whether protein metabolism would worsen as waist circumference increased in type 2 diabetes. Although it correlated with flux, synthesis, and breakdown with all subjects included, waist circumference did not do so within the type 2 diabetic group. However, our type 2 diabetic women with larger hips had greater net balance. This may indicate that lower-body fat has a “protective” protein-sparing effect. There is evidence that abdominal obesity may be a marker of “dysfunctional adipose tissue” characterized by a relative inability for subcutaneous adipose tissue to store excess energy, which then accumulates at undesired sites (30). In our viscerally obese type 2 diabetic women, having larger hips may explain why we did not discern an effect of the magnitude of hyperglycemia. Women with upper-body obesity have been found to have blunted insulin-mediated decrease in protein breakdown compared with women with lower body obesity (31).

The 15N-glycine method used cannot differentiate between the effects of insulin and amino acids on protein metabolism or elucidate the mechanisms underlying the results. Amino acids have protein anabolic effects in whole-body studies in healthy subjects (32) and share with insulin the intracellular mammalian target of rapamycin (mTOR) pathway to stimulate skeletal muscle protein synthesis (33). Assessment of whole-body insulin sensitivity in the fed part of the feeding-fasting cycle requires a separate technique for quantifying it.

All parameters of protein kinetics were highly correlated with REE, and REE was a predictive factor of flux and synthesis in type 2 diabetes (data not shown). This is not surprising, as the energy requirement of protein turnover may account for ∼20% of REE (34). We have reported greater REE in hyperglycemic type 2 diabetic subjects, with glycemia as an independent predictor of REE (35). Others have found the same with poor diabetes control and attributed it to greater protein turnover (8).

Mean age differed among groups, but each group was composed of younger and older subjects, which may explain the absence of an age effect on protein kinetics. Age did not correlate with kinetic parameters, controlling for FFM. Others have found a significant linear decline in postabsorptive whole-body protein turnover by advancing decades but no significant difference when comparing groups of middle-aged to younger or older subjects (27). However, within the type 2 diabetic group, age was negatively correlated with flux, synthesis, and breakdown and was a significant predictor of flux in men. We have reported a decrease in flux with age (18). This effect was attributed mostly to changes in body composition with aging. In the present study, there were trends of FFM to decrease with age in the type 2 diabetic group and for the older men with type 2 diabetes to be leaner.

In conclusion, protein flux, synthesis, and breakdown are elevated, while net balance is diminished in type 2 diabetes. While in men with type 2 diabetes fasting glycemia relates to net balance, this is not the case in women, in whom lower body fat appears to be protective. We have also shown that abnormalities in protein metabolism and insulin resistance of glucose coexist in obesity (14) and that insulin sensitivity of protein metabolism is blunted in type 2 diabetes (13). For these reasons, there is need to define protein requirements for type 2 diabetes in relation to concurrent obesity, weight loss, and level of glycemic (metabolic) control.

Figure 1—

Whole-body protein kinetics in nonobese and obese subjects with or without type 2 diabetes. Data are nitrogen (N) in grams per day adjusted means ± SE. Results were obtained by ANCOVA among groups, with FFM and protein intake as covariates for flux (Q) and FFM for synthesis (S) and breakdown (B) and by ANOVA for net balance (S-B). ▪, flux; , synthesis; □, breakdown; , net balance. *P < 0.05 vs. corresponding values of the other two groups. C: *P < 0.05 vs. other two groups.

Figure 1—

Whole-body protein kinetics in nonobese and obese subjects with or without type 2 diabetes. Data are nitrogen (N) in grams per day adjusted means ± SE. Results were obtained by ANCOVA among groups, with FFM and protein intake as covariates for flux (Q) and FFM for synthesis (S) and breakdown (B) and by ANOVA for net balance (S-B). ▪, flux; , synthesis; □, breakdown; , net balance. *P < 0.05 vs. corresponding values of the other two groups. C: *P < 0.05 vs. other two groups.

Close modal
Figure 2—

Simple linear correlations between net protein balance (S-B) in grams per day and hip circumferences in centimeters in 18 women with type 2 diabetes (A) and fasting serum glucose in 14 men with type 2 diabetes (B).

Figure 2—

Simple linear correlations between net protein balance (S-B) in grams per day and hip circumferences in centimeters in 18 women with type 2 diabetes (A) and fasting serum glucose in 14 men with type 2 diabetes (B).

Close modal
Table 1—

Subject data

NonobeseObese subjects without type 2 diabetesObese subjects with type 2 diabetes
Sex (male/female) 18/20 13/37 15/25 
Age (years) 29 ± 2* 40 ± 2† 48 ± 1‡ 
Age range (years) 18–65 19–61 33–65 
Weight (kg) 62 ± 1* 102 ± 3† 102 ± 3† 
BMI (kg/m221.4 ± 0.3* 36.9 ± 0.9† 37.7 ± 1.0† 
FFM (kg) 49 ± 1* 55 ± 2† 57 ± 2† 
Body fat (%) 20 ± 1* 45 ± 1† 44 ± 1† 
Waist circumference (cm), men (n = 40) 78 ± 1* 115 ± 4†¶ 125 ± 4† 
Waist circumference (cm), women (n = 62) 69 ± 1§ 108 ± 2¶ 117 ± 3† 
Hip circumference (cm), men (n = 40) 92 ± 1* 116 ± 3† 123 ± 5†§ 
Hip circumference (cm), women (n = 62) 95 ± 1‡ 127 ± 3§ 128 ± 3§ 
Waist-to-hip ratio, men (n = 42) 0.85 ± 0.01* 0.97 ± 0.03† 1.04 ± 0.02‡ 
Waist-to-hip ratio, women (n = 70) 0.73 ± 0.01§ 0.84 ± 0.01* 0.93 ± 0.02† 
REE (kcal/day) 1459 ± 36* 1749 ± 46† 1910 ± 51‡ 
Protein intake (g/day) 83 ± 2* 96 ± 2† 97 ± 2† 
Protein intake (g · kg FFM−1 · day−11.69 ± 0.03* 1.76 ± 0.04* 1.73 ± 0.04* 
Nitrogen balance (g/day) 0.6 ± 0.1* 0.7 ± 0.2* −0.5 ± 0.2† 
Fasting triacylglycerol (mmol/l) 1.0 ± 0.1* 1.9 ± 0.2† 2.9 ± 0.2‡ 
Fasting glucose (mmol/l) 4.5 ± 0.1* 5.0 ± 0.1* 13.0 ± 0.7† 
Daily mean capillary glucose (mmol/l) 4.7 ± 0.1* 4.9 ± 0.1* 14.0 ± 0.6† 
A1C (%) 5.2 ± 0.1 (n = 18)* 5.5 ± 0.2 (n = 20)* 8.2 ± 0.3 (n = 35)† 
Fasting free fatty acids (μmol/l) 607 ± 34* 754 ± 36* 988 ± 66† 
Fasting insulin (pmol/l) 71 ± 4* 174 ± 12† 224 ± 20‡ 
HOMA-IR 2.0 ± 0.1* 5.5 ± 0.4† 17.5 ± 1.5‡ 
LnHOMA-IR 0.63 ± 0.04* 1.57 ± 0.08† 2.72 ± 0.09‡ 
NonobeseObese subjects without type 2 diabetesObese subjects with type 2 diabetes
Sex (male/female) 18/20 13/37 15/25 
Age (years) 29 ± 2* 40 ± 2† 48 ± 1‡ 
Age range (years) 18–65 19–61 33–65 
Weight (kg) 62 ± 1* 102 ± 3† 102 ± 3† 
BMI (kg/m221.4 ± 0.3* 36.9 ± 0.9† 37.7 ± 1.0† 
FFM (kg) 49 ± 1* 55 ± 2† 57 ± 2† 
Body fat (%) 20 ± 1* 45 ± 1† 44 ± 1† 
Waist circumference (cm), men (n = 40) 78 ± 1* 115 ± 4†¶ 125 ± 4† 
Waist circumference (cm), women (n = 62) 69 ± 1§ 108 ± 2¶ 117 ± 3† 
Hip circumference (cm), men (n = 40) 92 ± 1* 116 ± 3† 123 ± 5†§ 
Hip circumference (cm), women (n = 62) 95 ± 1‡ 127 ± 3§ 128 ± 3§ 
Waist-to-hip ratio, men (n = 42) 0.85 ± 0.01* 0.97 ± 0.03† 1.04 ± 0.02‡ 
Waist-to-hip ratio, women (n = 70) 0.73 ± 0.01§ 0.84 ± 0.01* 0.93 ± 0.02† 
REE (kcal/day) 1459 ± 36* 1749 ± 46† 1910 ± 51‡ 
Protein intake (g/day) 83 ± 2* 96 ± 2† 97 ± 2† 
Protein intake (g · kg FFM−1 · day−11.69 ± 0.03* 1.76 ± 0.04* 1.73 ± 0.04* 
Nitrogen balance (g/day) 0.6 ± 0.1* 0.7 ± 0.2* −0.5 ± 0.2† 
Fasting triacylglycerol (mmol/l) 1.0 ± 0.1* 1.9 ± 0.2† 2.9 ± 0.2‡ 
Fasting glucose (mmol/l) 4.5 ± 0.1* 5.0 ± 0.1* 13.0 ± 0.7† 
Daily mean capillary glucose (mmol/l) 4.7 ± 0.1* 4.9 ± 0.1* 14.0 ± 0.6† 
A1C (%) 5.2 ± 0.1 (n = 18)* 5.5 ± 0.2 (n = 20)* 8.2 ± 0.3 (n = 35)† 
Fasting free fatty acids (μmol/l) 607 ± 34* 754 ± 36* 988 ± 66† 
Fasting insulin (pmol/l) 71 ± 4* 174 ± 12† 224 ± 20‡ 
HOMA-IR 2.0 ± 0.1* 5.5 ± 0.4† 17.5 ± 1.5‡ 
LnHOMA-IR 0.63 ± 0.04* 1.57 ± 0.08† 2.72 ± 0.09‡ 

Data are means ± SE. Groups were compared by ANOVA, with post-hoc Bonferroni test. Data with a common symbol are not different from one another, within each variable and including both men and women where data are shown by sex. HOMA-IR: [insulin (μU/ml) × glucose (mmol/l)]/22.5. lnHOMA-IR, log-transformed HOMA-IR values.

This study was supported by grants from the Canadian Institutes of Health Research (MOP-94059 and MOP-77562 to R.G.), salary awards (to R.G.) and studentships (to S.P.) from the McGill University Health Centre Research Institute, fellowships (to J.A.M.) from Fonds de la Recherch en Santé and to S.C. from the Canadian Diabetes Association.

We thank Mary Shingler, Josie Plescia, Ginette Sabourin, Concettina Nardolillo, Madeleine Giroux, and Paul Meillon for their assistance. All authors contributed to study design, analysis, and interpretation. None have financial agreements that might cause conflicts of interest.

1.
Bennet WM, Connacher AA, Jung RT, Stehle P, Rennie MJ: Effects of insulin and amino acids on leg protein turnover in IDDM patients.
Diabetes
40
:
499
–508,
1991
2.
Nair KS, Ford GC, Halliday D: Effect of intravenous insulin treatment on in vivo whole body leucine kinetics and oxygen consumption in insulin-deprived type I diabetic patients.
Metabolism
36
:
491
–495,
1987
3.
Gougeon R, Styhler K, Morais JA, Jones PJ, Marliss EB: Effects of oral hypoglycemic agents and diet on protein metabolism in type 2 diabetes.
Diabetes Care
23
:
1
–8,
2000
4.
Gougeon R, Pencharz PB, Marliss EB: Effect of NIDDM on the kinetics of whole-body protein metabolism.
Diabetes
43
:
318
–328,
1994
5.
Gougeon R, Pencharz PB, Sigal RJ: Effect of glycemic control on the kinetics of whole-body protein metabolism in obese subjects with non-insulin-dependent diabetes mellitus during iso- and hypoenergetic feeding.
Am J Clin Nutr
65
:
861
–870,
1997
6.
Gougeon R, Marliss EB, Jones PJ, Pencharz PB, Morais JA: Effect of exogenous insulin on protein metabolism with differing nonprotein energy intakes in type 2 diabetes mellitus.
Int J Obes Relat Metab Disord
22
:
250
–261,
1998
7.
Marchesini G, Forlani G, Zoli M, Vannini P, Pisi E: Muscle protein breakdown in uncontrolled diabetes as assessed by urinary 3-methylhistidine excretion.
Diabetologia
23
:
456
–458,
1982
8.
Welle S, Nair KS: Failure of glyburide and insulin treatment to decrease leucine flux in obese type II diabetic patients.
Int J Obes
14
:
701
–710,
1990
9.
Staten MA, Matthews DE, Bier DM: Leucine metabolism in type II diabetes mellitus.
Diabetes
35
:
1249
–1253,
1986
10.
Luzi L, Petrides AS, De Fronzo RA: Different sensitivity of glucose and amino acid metabolism to insulin in NIDDM.
Diabetes
42
:
1868
–1877,
1993
11.
Halvatsiotis P, Short KR, Bigelow M, Nair KS: Synthesis rate of muscle proteins, muscle functions, and amino acid kinetics in type 2 diabetes.
Diabetes
51
:
2395
–2404,
2002
12.
Chevalier S, Gougeon R, Kreisman SH, Cassis C, Morais JA: The hyperinsulinemic amino acid clamp increases whole-body protein synthesis in young subjects.
Metabolism
53
:
388
–396,
2004
13.
Pereira S, Chevalier S, Morais JA, Marliss EB, Gougeon R: Type 2 diabetes–associated insulin resistance of protein anabolism.
Diabetes
57
:
56
–63,
2008
14.
Chevalier S, Marliss EB, Morais JA, Lamarche M, Gougeon R: Whole-body protein anabolic response is resistant to the action of insulin in obese women.
Am J Clin Nutr
82
:
355
–365,
2005
15.
Halvatsiotis PG, Turk D, Alzaid A, Dinneen S, Rizza RA, Nair KS: Insulin effect on leucine kinetics in type 2 diabetes mellitus.
Diabetes Nutr Metab
15
:
136
–142,
2002
16.
Denne SC, Brechtel G, Johnson A, Liechty EA, Baron AD: Skeletal muscle proteolysis is reduced in noninsulin-dependent diabetes mellitus and is unaltered by euglycemic hyperinsulinemia or intensive insulin therapy.
J Clin Endocrinol Metab
80
:
2371
–2377,
1995
17.
Chevalier S, Marliss EB, Morais JA, Lamarche M, Gougeon R: The influence of sex on the protein anabolic response to insulin.
Metabolism
54
:
1529
–1535,
2005
18.
Morais JA, Gougeon R, Pencharz PB, Jones PJ, Ross R, Marliss EB: Whole-body protein turnover in the healthy elderly.
Am J Clin Nutr
66
:
880
–889,
1997
19.
Goran MI, Khaled MA: Cross-validation of fat-free mass estimated from body density against bioelectrical resistance: effects of obesity and gender.
Obes Res
3
:
531
–539,
1995
20.
Kushner RF, Schoeller DA, Fjeld CR, Danford L: Is the impedance index (ht2/R) significant in predicting total body water?
Am J Clin Nutr
56
:
835
–839,
1992
21.
Forbes G: Body composition. In
Modern Nutrition in Health and Disease.
7th ed. Shils M, Young V, Eds. Philadelphia, Lea & Febinger,
1988
, p.
533
–556
22.
Pencharz PB, Steffee WP, Cochran W, Scrimshaw NS, Rand WM, Young VR: Protein metabolism in human neonates: nitrogen-balance studies, estimated obligatory losses of nitrogen and whole-body turnover of nitrogen.
Clin Sci Mol Med
52
:
485
–498,
1977
23.
Yang D, Puchowicz MA, David F, Powers L, Halperin ML, Brunengraber H: 15N enrichment of ammonium, glutamine-amide and urea, measured via mass isotopomer analysis of hexamethylenetetramine.
J Mass Spectrom
34
:
1130
–1136,
1999
24.
Waterlow JC, Golden MH, Garlick PJ: Protein turnover in man measured with 15N: comparison of end products and dose regimes.
Am J Physiol
235
:
E165
–E174,
1978
25.
Picou D, Taylor-Roberts T: The measurement of total protein synthesis and catabolism and nitrogen turnover in infants in different nutritional states and receiving different amounts of dietary protein.
Clin Sci
36
:
283
–296,
1969
26.
Wallace TM, Levy JC, Matthews DR: Use and abuse of HOMA modeling.
Diabetes Care
27
:
1487
–1495,
2004
27.
Short KR, Vittone JL, Bigelow ML, Proctor DN, Nair KS: Age and aerobic exercise training effects on whole body and muscle protein metabolism.
Am J Physiol Endocrinol Metab
286
:
E92
–E101,
2004
28.
Biolo G, Antonione R, Barazzoni R, Zanetti M, Guarnieri G: Mechanisms of altered protein turnover in chronic diseases: a review of human kinetic studies.
Curr Opin Clin Nutr Metab Care
6
:
55
–63,
2003
29.
Kang ES, Yun YS, Park SW, Kim HJ, Ahn CW, Song YD, Cha BS, Lim SK, Kim KR, Lee HC: Limitation of the validity of the homeostasis model assessment as an index of insulin resistance in Korea.
Metabolism
54
:
206
–211,
2005
30.
Despres JP, Lemieux I: Abdominal obesity and metabolic syndrome.
Nature
444
:
881
–887,
2006
31.
Jensen MD, Haymond MW: Protein metabolism in obesity: effects of body fat distribution and hyperinsulinemia on leucine turnover.
Am J Clin Nutr
53
:
172
–176,
1991
32.
Castellino P, Luzi L, Simonson DC, Haymond M, DeFronzo RA: Effect of insulin and plasma amino acid concentrations on leucine metabolism in man. Role of substrate availability on estimates of whole body protein synthesis.
J Clin Invest
80
:
1784
–1793,
1987
33.
Kimball SR, Jefferson LS: New functions for amino acids: effects on gene transcription and translation.
Am J Clin Nutr
83
:
500S
–507S,
2006
34.
Welle S, Nair KS: Relationship of resting metabolic rate to body composition and protein turnover.
Am J Physiol
258
:
E990
–E998,
1990
35.
Gougeon R, Lamarche M, Yale JF, Venuta T: The prediction of resting energy expenditure in type 2 diabetes mellitus is improved by factoring for glycemia.
Int J Obes Relat Metab Disord
26
:
1547
–1552,
2002

Published ahead of print at http://care.diabetesjournals.org on 5 October 2007. DOI: 10.2337/dc07-1268.

A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.