Older, obese, and sedentary individuals are at high risk of developing diabetes and cardiovascular disease. Exercise training improves metabolic anomalies associated with such diseases, but the effects of caloric restriction in addition to exercise in such a high-risk group are not known. Changes in body composition and metabolism during a lifestyle intervention were investigated in 23 older, obese men and women (aged 66 ± 1 yr, body mass index 33.2 ± 1.4 kg/m2) with impaired glucose tolerance. All volunteers undertook 12 wk of aerobic exercise training [5 days/wk for 60 min at 75% maximal oxygen consumption (V̇o2max)] with either normal caloric intake (eucaloric group, 1,901 ± 277 kcal/day, n = 12) or a reduced-calorie diet (hypocaloric group, 1,307 ± 70 kcal/day, n = 11), as dictated by nutritional counseling. Body composition (decreased fat mass; maintained fat-free mass), aerobic fitness (V̇o2max), leptinemia, insulin sensitivity, and intramyocellular lipid accumulation (IMCL) in skeletal muscle improved in both groups (P < 0.05). Improvements in body composition, leptin, and basal fat oxidation were greater in the hypocaloric group. Following the intervention, there was a correlation between the increase in basal fat oxidation and the decrease in IMCL (r = −0.53, P = 0.04). In addition, basal fat oxidation was associated with circulating leptin after (r = 0.65, P = 0.0007) but not before the intervention (r = 0.05, P = 0.84). In conclusion, these data show that exercise training improves resting substrate oxidation and creates a metabolic milieu that appears to promote lipid utilization in skeletal muscle, thus facilitating a reversal of insulin resistance. We also demonstrate that leptin sensitivity is improved but that such a trend may rely on reducing caloric intake in addition to exercise training.
- substrate oxidation
- insulin sensitivity
skeletal muscle oxidative capacity is impaired in obese individuals, predisposing them to further weight gain and insulin resistance (16), and increasing the risk of developing Type 2 diabetes (T2DM) and associated macrovascular complications (24, 35). Aging is also associated with impaired substrate utilization and insulin resistance, probably due to a sedentary lifestyle and elevated body fat causing impaired mitochondrial function (13, 22, 26, 30). Therefore, older, obese, impaired glucose tolerant (IGT) cohorts are at high risk of developing metabolic diseases.
Aerobic exercise training improves oxidative capacity and insulin sensitivity in younger and older obese and/or IGT/T2DM individuals (8, 14, 15, 20, 36). Caloric restriction interventions also exhibit similar improvements in insulin sensitivity; yet such findings are inconsistent, and the impact on substrate metabolism is less clear (27, 32). The main drawback of hypocaloric weight loss interventions is that fat-free mass, a major determinant of substrate utilization, in particular basal fat oxidation, is also often reduced (8, 15). Aging is associated with negative nitrogen balance, and therefore older adults are at increased risk for developing sarcopenia (25). Exercise promotes protein synthesis rates, which may help maintain fat-free mass in older individuals. Therefore, a weight loss program that includes exercise may be favorable to optimally improve insulin sensitivity and substrate metabolism in an older, obese, IGT population, thus reducing such a group's already high risk of developing macrovascular complications.
Impaired insulin action mainly occurs in obesity due to accumulation of intracellular fatty acid derivatives. However, adipocytokines are also documented to play a role. Circulating concentrations of leptin have been shown to be positively related to fat mass and inversely related to insulin sensitivity (33). Leptin is involved in satiety, via a hypothalamic signaling pathway (10), and in substrate metabolism, via direct peripheral tissue action (23). However, it has been demonstrated that leptin can increase fat oxidation (23), illustrating a paradox: it is possible that, in obesity, leptin resistance may arise (34). The relationship between leptin and fat oxidation was investigated in this study. In addition to leptin, another adipocytokine, adiponectin, has also been shown to be related to insulin sensitivity and substrate metabolism (37, 41). Therefore, the relationship between adiponectin and fat oxidation was also measured in this study.
Currently, in older obese IGT individuals, it is unknown whether the combination of exercise training with a hypocaloric diet is more effective at improving factors associated with metabolic disease compared with a similar exercise regime with a eucaloric macronutrient intake. This study serves to investigate such differences. It was hypothesized that exercise plus caloric restriction would improve insulin sensitivity and substrate metabolism to a greater extent than exercise training alone as a result of greater weight loss, particularly fat mass.
Twenty-three older, obese, IGT (fasting plasma glucose >5.6 mmol/l and 2-h oral glucose tolerance test glucose concentration between 7.8 and 11.1 mmol/l) volunteers were recruited from the general population to take part in a 12-wk exercise training intervention, while consuming either a eucaloric or a hypocaloric diet. Medical screening excluded individuals with heart, kidney, liver, intestinal, and pulmonary diseases, or those taking medications for hypertension, diabetes, or other obesity-related conditions. All volunteers were sedentary, and they had been weight stable for 6 mo before the study. The study was approved by the Institutional Review Board, and all subjects provided written informed consent in accordance with our guidelines for the protection of human subjects.
Volunteers were divided into two body mass index (BMI)-matched groups: a eucaloric group (n = 12: 4 men, 8 women; age = 66 ± 1 yr; BMI = 34.7 ± 1.6 kg/m2) and a hypocaloric group (n = 11: 3 men, 8 women; age = 67 ± 1 yr BMI = 33.6 ± 1.3 kg/m2). Both groups undertook 60 min of supervised moderate-intensity aerobic exercise (treadmill walking/cycle ergometry/stationary rowing) at 75% of maximal oxygen uptake capacity 5 days a week (Monday to Friday) for 12 wk. Dietary records were collected for 3 days before the study, and individual nutritional counseling was provided weekly to monitor caloric intake. The eucaloric group were instructed to continue their typical dietary intake throughout the study, while the hypocaloric group were instructed to reduce their daily energy intake by ∼500 kcal. Several anthropometric and metabolic measures were taken pre- and postintervention as described below during a 3-day inpatient stay at the General Clinical Research Center.
Height was measured to the nearest 1.0 cm without shoes. Body weight was measured to the nearest 0.1 kg with the subject wearing their underclothing and a hospital gown. Waist circumference was measured midway between the lower rib margin and iliac crest to the nearest 1.0 cm. Body density was determined by hydrostatic weighing, and body fat mass was calculated using Siri equations as previously described (29).
Each subject performed an incremental treadmill exercise test to determine their maximal oxygen consumption (V̇o2max). The speed was set between 2 and 4 miles/h, and the incline of the treadmill was increased 2.5% every 2 min until fatigue. Exhaled air was collected, and concentrations of oxygen and carbon dioxide were measured using an electrochemical oxygen analyzer (model A S-3, Applied Electrochemistry,) and infrared carbon dioxide analyzer (model LB-2, Beckman). Maximum aerobic capacity (V̇o2max) was achieved when at least two of the following criteria were achieved: 1) a plateau in oxygen uptake 2) a heart rate within 10 beats/min of age-predicted maximum, and/or 3) a respiratory exchange ratio > 1.0. Because of the acute effects of exercise on insulin sensitivity, the time period between the V̇o2max test and the euglycemic clamp was always at least 48 h. Postintervention V̇o2max testing was performed the morning following the clamp procedure.
Basal substrate metabolism.
On day 2 of the inpatient stay, following a 12-h overnight fast, subjects were awakened at 0600 and taken by wheelchair to void and to be weighed, and then they reclined in a semidarkened, thermoneutral (22 ± 1°C) environment under a clear plastic hood (Brooks Instruments, Hatfield, PA) for 30 min for indirect calorimetry measurements, as previously described (40). The molar ratio of oxygen consumed to carbon dioxide produced was used to derive a measure of the relative amounts of substrate that were being oxidized [respiratory quotient (RQ)]. The Weir equation was used to calculate resting energy expenditure (39), and substrate oxidation rates were calculated according to Frayn (7). Timed urinary nitrogen excretion measurements were also made for estimates of protein oxidation rates (7). In brief, urine was collected from 0600 until completion of the calorimetry measures. Total volume and time of collection were recorded, and aliquots were analyzed for urea nitrogen (Roche Modular Diagnostics, Indianapolis, IN).
On the final morning of the inpatient stay, a hyperinsulinemic euglycemic clamp was performed as previously described (5, 19). Following an overnight fast, subjects voided morning urine and were weighed. An 18-gauge polyethylene catheter was inserted into an antecubital vein for the infusion of glucose (20% dextrose) and insulin. A second 20-gauge polyethylene catheter was inserted in retrograde fashion into a dorsal hand vein, and the hand was warmed in a heated box (∼65°C) for sampling of arterialized venous blood. After collection of baseline blood samples (5 ml), a percutaneous biopsy of the vastus lateralis muscle was obtained using the Bergstrom needle technique, as previously described (19). A primed 40 mU·m−2·min−1 insulin infusion commenced at time = 0 min followed by a 2-h variable glucose infusion intended to maintain plasma glucose concentrations at 5 mmol/l. Arterialized blood samples (0.5 ml) were collected every 5 min for plasma glucose determination (Beckman Instruments, Fullerton, CA) so as to make continuous adjustments to the glucose infusion rate to maintain euglycemia. Insulin sensitivity was calculated as the mean glucose metabolized (clamp M) over the last 30-min period of the clamp, expressed as mg·kg fat-free mass·−1·min−1. A preclamp blood sample was also drawn for the measurement of plasma leptin by radioimmunoassay (Linco Research, St. Charles, MO) and of plasma adiponectin by ELISA (R&D Systems, Minneapolis, MN).
Skeletal muscle histology.
Muscle mounts (n = 14: 7 hypocaloric, 7 eucaloric) obtained from the vastus lateralis biopsy were stored at −70°C to be later analyzed for intramuscular lipid content (IMCL). IMCL was measured by a semiquantitative method using the Oil-Red-O (ORO) technique as previously described by Goodpaster et al. (9). In brief, 10-μm cryosections were mounted on glass slides, air-dried, and immersed in an ORO (Sigma, St. Louis, MO) solution (36 mg ORO, 12 ml 2-propanol, 8 ml distilled H2O) for 10 min. Slides were rinsed thoroughly in distilled H2O, allowed to dry, and then fixed under a coverslip using Glycergel mounting medium (DAKO, Carpinteria, CA), and sealed using an acrylic polymer. Slides were viewed under a light microscope using ×40 objective and bright-field settings (Olympus BX61, Melville, NY), and digital images were captured (Hamamatsu Orca-ER, Bridgewater, NJ). Images were viewed and analyzed in gray-scale format using NIH Image software (http://rsb.info.nih.gov/nih-image/). Quantification of IMCL was performed by identifying ORO-stained areas with staining intensity of <150 arbitrary gray-scale units. This was repeated on five fields of view (750 × 550 pixels; 1 pixel = 0.18 μm2) per subject, and the mean cross-sectional muscle fiber area and mean lipid-stained area were calculated. Intramyocellular lipid accumulation was normalized for changes in fiber surface area: lipid surface area divided by fiber area multiplied by 100 [= lipid accumulation index (LAI, %)].
Statistical analyses were carried out using Statview (SAS Institute, Cary, NC), and all data are expressed as means ± SE. Normality of each variable was assessed using the Kolmogorov-Smirnov test. Variables deviating from normal distributions (plasma leptin concentrations) were natural log-transformed (ln) before statistical analysis. Between-group (eucaloric vs. hypocaloric) comparisons for all variables were analyzed using two-way (group × time) repeated-measures ANOVA. Tukey post hoc tests were applied to significant group × time interactions. Bivariate correlations were used to examine relationships between pre- to poststudy changes (Δ) in basal fat oxidation, clamp M, and ln leptin with other variables. Additionally, separate bivariate analyses were also applied to variables at baseline, and to variables following the intervention (e.g., baseline leptin vs. baseline fat oxidation, and postintervention leptin vs. postintervention fat oxidation). This was carried out to identify changes in relationships as a result of the intervention. Statistical significance was accepted at P < 0.05.
During the 12-wk intervention, the eucaloric group consumed a similar daily caloric load (1,901 ± 277 vs. 1,866 ± 231 kcal/day; P = 0.87) and macronutrient composition to their existing diet. The hypocaloric group consumed ∼33% fewer calories during the study compared with their existing diet (1,307 ± 70 vs. 1,964 ± 164 kcal/day; P = 0.008), during which there was also a shift toward reduced fat intake (30 ± 1 vs. 25 ± 1%) and increased carbohydrate intake (51 ± 2 vs. 57 ± 1%).
Both study groups exhibited significant weight loss, improvements in BMI, and reductions in body fat mass and waist circumference between baseline and post-study measures (Table 1), but, as expected, weight, BMI, and fat mass showed greater changes in the hypocaloric group. Fat-free mass did not change in either group (Table 1; P > 0.05).
V̇o2max increased similarly in both groups (Table 1; eucaloric: 10.6 ± 3.0% increase, P = 0.003; hypocaloric: 11.8 ± 5.7% increase, P = 0.04).
Glucose metabolized during the hyperinsulinemic euglycemic clamp was improved by similar amounts in both groups (Table 1; eucaloric: 55.1 ± 19.0% increase, P = 0.03; hypocaloric: 65.1 ± 14.4% increase, P = 0.001).
Leptin and adiponectin.
Preintervention plasma leptin concentrations were similar between groups (Table 1). Leptin concentrations decreased in both groups following the intervention, with a larger reduction seen in the hypocaloric group (31.6 ± 6.0% decrease vs. 12.2 ± 3.8% decrease). Analysis of plasma adiponectin concentrations did not reveal any effect of time or trial (Table 1; P > 0.05).
Following the exercise training intervention, intramuscular lipids (LAI, %) decreased by 25.9 ± 12.4 and 34.3 ± 17.6% in the eu- and hypocaloric groups, respectively (2-way ANOVA, effect of time P = 0.04; Table 2). No group differences were found (P = 0.66).
Basal substrate metabolism.
Table 3 indicates changes in basal substrate metabolism measured at rest following an overnight fast. The RQ was decreased equally in both groups (P = 0.03), whereas basal fat oxidation was significantly improved in the hypocaloric group only (38.4 ± 12.8% increase; P = 0.02). Figure 1 displays the change in basal carbohydrate and fat oxidation following the intervention in each group, illustrating a larger improvement in basal fat oxidation in the hypocaloric group (P = 0.03). Analysis of the change in basal fat oxidation between groups also revealed a statistically significant difference, showing a greater increase in the hypocaloric group (P = 0.05). No changes in energy expenditure or in carbohydrate and protein oxidation rates were found as a result of the interventions.
The majority of measured variables demonstrated no group×time interactions following analysis, therefore, because it appeared that changes with time were occurring independent of caloric consumption, data from the eucaloric and hypocaloric groups were pooled for correlation analyses. Table 4 indicates that changes in clamp M, leptin, and basal fat oxidation were mostly related to changes in body composition. Changes in clamp M were not found to be related to changes in leptin or basal fat oxidation, nor were changes in leptin found to be related to changes in basal fat oxidation. There was a significant association between the change in IMCL and the changes in basal fat oxidation (r = −0.53, P = 0.04; Table 4, Fig. 2) and circulating leptin (r = 0.44, P = 0.05; Table 4). There was also a trend for a correlation between the change in circulating adiponectin and basal fat oxidation (r = 0.44, P = 0.05). In addition, Fig. 3 illustrates that following the intervention a strong relationship developed between leptin and basal fat oxidation poststudy (r = 0.65, P = 0.0007) that was not present at baseline (r = 0.046, P = 0.84). In a separate analysis of eu- and hypocaloric groups, this poststudy relationship between leptin and basal fat oxidation was greatest in the hypocaloric group (hypocaloric vs. eucaloric: r = 0.79, P = 0.005 vs. r = 0.58, P = 0.08).
The main finding of this study is that a 12-wk aerobic exercise training intervention improves body composition (weight, fat mass, waist circumference), aerobic fitness (V̇o2max), leptinemia, intramuscular lipid accumulation (IMCL), and insulin sensitivity (clamp M) in older, obese, previously sedentary, IGT individuals, independent of total daily caloric consumption; but that improvement in basal fat oxidation is dependent on caloric restriction in addition to exercise. In addition, the changes in IMCL were related to improvements in basal fat oxidation and leptinemia. Furthermore, although changes in basal fat oxidation were not found to be related to changes in insulin sensitivity, a positive relationship between basal fat oxidation and plasma leptin developed postintervention, suggesting an improvement in leptin sensitivity. This trend was more pronounced when caloric restriction was undertaken in addition to exercise training. Similar findings in younger obese cohorts have been published by Goodpaster et al. (8) and He et al. (11), who found a relationship between fat utilization and insulin sensitivity; therefore, our work suggests a possible age-related dissociation of these two variables.
Weight loss via reductions in fat mass lowers systemic free fatty acid appearance into the circulation, thus reducing free fatty acid availability to skeletal muscle tissue (2). In insulin-resistant muscle, as found in older, obese, IGT individuals, where basal fat oxidation is impaired, this will limit the accumulation of intramuscular lipids and other fatty acid derivatives known to interfere with insulin signaling (21). IMCL content has been described to be elevated in insulin-resistant patients (38), yet it appears that these two variables may disassociate following exercise training (4). In this study, we demonstrate reduced IMCL postintervention, which, although not directly related to insulin sensitivity, was associated with changes in fat oxidation and circulating leptin. It has been suggested that decreased IMCL with increased rates of basal fat oxidation reduces reliance on plasma free fatty acids and increases utilization of intramuscular lipid pools (31). Thus, from our data, we can speculate that reductions in IMCL in our subjects following exercise training may be due to an increase in IMCL utilization. These changes occur independently of caloric intake, suggesting that exercise is the key component of the intervention.
The addition of an exercise stimulus to caloric restriction regimes also upregulates skeletal muscle capillarization and mitochondrial density (12), and it activates enzymes involved in fatty acid transport and oxidation (17, 18). Therefore, it seems likely that exercise training plus caloric restriction would lead to greater improvements in substrate metabolism and insulin sensitivity than exercise training alone. However, this present study indicates that, despite larger changes in body composition and leptinemia following a hypocaloric exercise training regime, improvements in insulin sensitivity appear to be driven by exercise per se, independent of caloric intake. Alternatively, it is possible that insulin sensitivity reached a peak improvement in these individuals, one that can be achieved by caloric restriction or by exercise. To clarify a potential “ceiling effect” in these older obese subjects, the addition of a hypocaloric nonexercising cohort would be useful. Also, compared with a eucaloric exercise training regime, greater improvements in basal fat oxidation were seen in our hypocaloric exercise group, despite similar improvements in V̇o2max. To explain this finding, it is possible that the improvements in body composition, insulin sensitivity, and leptin seen in the eucaloric group were below a threshold that would extrapolate to an improvement in whole body substrate utilization. However, because insulin sensitivity changed equally between groups, it is unlikely that this contributes to the different changes seen in substrate oxidation. This concept therefore merits further attention in future studies.
Reduced fat mass leads to improvements in plasma adipocytokine concentrations. Such adipose-derived molecules have direct effects on insulin signaling and substrate utilization (23, 33). Here we demonstrate reduced postintervention circulating leptin concentrations. Further to this, our data demonstrate that a state of leptin resistance existed in our older, obese, IGT cohort, shown by the lack of correlation between leptin and basal fat oxidation (Fig. 3). However, as a result of the exercise training interventions, leptin sensitivity was improved, particularly in the hypocaloric group (hypocaloric vs. eucaloric: r = 0.790, P = 0.005 vs. r = 0.579, P = 0.08), suggesting that exercise plus caloric restriction is required to increase fat oxidation via a leptin-mediated pathway in older, obese, IGT individuals. Leptin has been shown to upregulate fat oxidation via an AMP-activated protein kinase-related pathway (23). Other data also indicate that leptin fails to upregulate fat oxidation in obese humans compared with age- and gender-matched lean individuals, providing further evidence of a defect in leptin signaling in obesity (34). To our knowledge, no data on the effects of exercise training on leptin-induced fat oxidation exist in older obese IGT humans; however, it has been shown that an exercise training regime can upregulate acetyl-CoA carboxylase activity (a leptin-sensitive enzyme) and fat oxidation in lean, middle-aged individuals (31). These observations suggest a direction for future work. In this study, we demonstrate that the change in circulating leptin is related to the improvement in IMCL. This is of particular mechanistic interest, because the improvement in leptin sensitivity as a result of the lifestyle intervention may upregulate AMPK-related fat oxidation pathways, increasing intramuscular lipid utilization. Adiponectin is another adipocytokine implicated with insulin sensitivity and substrate oxidation. Analysis showed no effect of exercise or caloric consumption upon circulating plasma adiponectin. This finding strengthens the evidence that changes in plasma leptin concentrations as a result of diet and exercise are involved in the changes in levels of intramyocellular lipid and basal fat oxidation. However, the currently available data on exercise training, diet, and adiponectin, although limited, are quite conflicting (1, 3, 28). This is an area that certainly warrants further investigation when considering adiponectin's potential effect on substrate utilization and insulin sensitivity (37, 41).
A limitation of the present study is that we did not study a hypocaloric nonexercising group. This would further control the study by isolating the effects of weight loss per se on the measured metabolic variables. Additionally, the majority of the volunteers were women, and the small number of men recruited precluded any sex-difference analyses. Although subjects were postmenopausal, which would rule out the effects of estrogen on body composition and substrate metabolism, sex differences have been reported with leptin responses and action (6). Future work should attempt to establish potential sex differences in aging to ascertain optimal sex-specific lifestyle interventions. Despite this, clear improvements in body composition, insulin sensitivity, and basal fat metabolism were seen, factors that are of great importance to an old, obese, IGT group in reducing their risk of developing cardiovascular disease. In addition, weight loss was not associated with loss of fat-free mass, suggesting that regular aerobic exercise may have a beneficial impact on the state of sarcopenia and osteopenia that develops in such age groups. Maintaining metabolic function and mobility via the preservation of muscular and skeletal strength is of particular importance to an older, obese group; thus regular exercise prescription is sensible. In addition, it is important to note that 11 (3 eucaloric, 8 hypocaloric) of the 23 volunteers were no longer classified as IGT (fasting plasma glucose > 5.6 mmol/l) following the intervention. Fasting plasma glucose is a strong predictor of diabetes progression, and so, this study shows that just 12 wk of lifestyle intervention is enough to greatly reduce disease risk in older, obese individuals, and to normalize glucose tolerance in ∼50% of participants.
The major take-home message of this study is that regular moderate-intensity aerobic exercise is the driving force behind improvements in insulin sensitivity when older, obese, and previously sedentary IGT individuals follow an exercise and diet regime. Despite greater changes in body composition following a calorie-restricted exercise intervention, such changes did not extrapolate to greater changes in insulin sensitivity compared with a eucaloric exercise intervention. The loss of body fat mass in addition to the maintenance of fat-free mass is of importance to older cohorts. Such changes were demonstrated in this study. We also document that older, obese, IGT individuals exhibit leptin resistance with respect to basal fat oxidation, and that exercise training improves leptin sensitivity independent of caloric intake. This is a novel insight into the understanding of the onset of metabolic disease in obese sedentary groups. Further work in this area is warranted to identify the mechanistic pathways.
This research was supported by National Institute on Aging Grant AG-12834 (to J. P. Kirwan) and General Clinical Research Center Grants MO1 RR-10732, RR-00080, and RR-018390.
The authors thank the volunteers for their outstanding dedication and effort. We acknowledge the nursing/dietary staff of the General Clinical Research Center and the technical/engineering staff and students who helped with the implementation of the study and assisted with data collection.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society