INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

IT IS GENERALLY ACCEPTED THAT total body aerobic capacity [maximum oxygen uptake (O_2 max)] declines with age (26), independent of decreases in physical activity (15, 16, 24, 26). Decreased O_2 max with age is associated with declines in maximal cardiac output and muscle mass. Declines in muscle mitochondrial function may also be responsible (26). Although it is generally accepted that older adults have lower skeletal muscle aerobic capacity than younger adults (7, 15, 23, 26), Kent-Braun and Ng (16), using ³¹P-magnetic resonance spectroscopy (MRS), have recently reported data suggesting that the lower values found in older adults are mediated by reduced physical activity and disappear when older adults have activity patterns similar to younger adults. This is consistent with evidence that the aging-related decline in muscle function is more apparent in locomotor muscles than in nonlocomotor muscles (3, 13, 15, 17).

We are aware of little research concerning the relationship between muscle anaerobic capacity and age. Holloszy et al. (13) reported a reduced glycolytic capacity of superficial type IIb and deep IIa vastus lateralis of rat muscle, and Welle et al. (32) reported less abundant mRNA encoding enzymes of glucose metabolism in vastus lateralis of older humans. Neither study reported physical activity levels; therefore, it is possible that reductions in markers of glycolytic activity may have occurred as a result of age-related reductions in physical activity.

Most previous studies of which we are aware have demonstrated age-related declines in muscle oxidative enzyme activity or ³¹P-MRS muscle oxidative capacity by cross-sectional comparisons between groups in the 4th decade vs. the 7th or 8th decades (7, 15, 16, 23). In the only study of which we are aware that compared younger individuals, Rooyackers et al. (25) found a reduction in citrate synthase (CS) and cytochrome-c oxidase (COX) activity when comparing younger adults between 20 and 52 yr of age. More research is needed to determine whether muscle metabolic capacity changes between the 3rd and 5th decades of life.

One of the problems in determining whether muscle function decreases with age, even in individuals who maintain physical activity, is the difficulty in measuring physical activity. Questionnaires and activity monitors only indirectly measure physical activity and have limited validity for determining free-living physical activity. Doubly labeled water measurement of total energy expenditure (TEE) combined with indirect calorimetry measurement of resting or sleeping energy expenditure (SEE) can be used to measure free-living activity-related energy expenditure (AEE) accurately. To our knowledge, no studies have evaluated the potential age-related decrease in muscle function while adjusting for AEE.

Further research is necessary to determine whether exercise endurance, whole body aerobic capacity, and muscle metabolic capacity decrease with age, independent of decreases in physical activity. We have been able to do this in a group of premenopausal women who are participating in an ongoing study designed to identify metabolic factors that are predisposed to obesity. Therefore, the purpose of this study was to evaluate the relationship between age and exercise endurance, whole body oxidative capacity, and muscle metabolic capacity while adjusting for AEE in an understudied group of premenopausal women.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Eighty-three normal-weight, healthy premenopausal women (23-47 yr old) participated in this study. The subjects were selected from a larger ongoing study designed to measure metabolic factors that predispose women to obesity. None were taking any medication known to affect body composition, energy expenditure, or exercise performance. A subset of 48 women also had muscle metabolic capacity measured during exercise by using ³¹P-MRS. In addition, 18 of the women who underwent assessment of muscle metabolic capacity had muscle enzyme activity measured from muscle biopsies. The study was approved by the University of Alabama at Birmingham Institutional Review Board. All volunteers were screened and briefed about the experimental protocol, and informed consent was obtained before testing.

Women were weight stable for 1 mo before testing and were on a macronutrient-controlled diet for 2 wk preceding testing. All testing was performed in the follicular phase of the menstrual cycle (within 10 days of the start of menses). Exercise sessions on the treadmill and in the magnet were separated by at least 2 days.

Dual-energy X-ray absorptiometry. Bone mineral content and regional lean and fat tissue (trunk, arm, and leg) were determined by dual-energy X-ray absorptiometry (DPX-L, Lunar Radiation, Madison, WI). The scans were analyzed by using the Adult Software (version 1.33). The separation point between arms and trunk was the glenohumeral joint, and the separation point between the legs and trunk was on an oblique angle through the neck of the femur. Dual-energy X-ray absorptiometry lean tissue does not include estimates of bone mass or fat tissue mass, only soft lean tissue mass.

Measurement of SEE. Subjects spent 23 h in a whole room respiration calorimeter (3.38 m long, 2.11 m wide, and 2.58 m high). The calorimeter design characteristics and calibration have been previously described (30). Oxygen consumption and carbon dioxide production were continuously measured by a magnetopneumatic differential oxygen analyzer (Magnos 4G) and the NDIR industrial photometer differential carbon dioxide analyzer (Uras 3G, both Hartmann & Braun, Frankfort, Germany). The calorimeter was calibrated before each subject entered the chamber. The zero calibration was carried out simultaneously for both analyzers. The full scale was set for 0-1% for the carbon dioxide analyzer and for 0-2% for the oxygen analyzer.

Measurement of TEE. Free-living TEE was measured over 14 days of controlled diet and energy balance conditions by using the doubly labeled water technique. The previously described protocol (11) has a theoretical error of <5%. Samples were analyzed in triplicate for H₂¹⁸O and ²H₂O by isotope ratio mass spectrometry at the University of Alabama at Birmingham, as previously described (12). When all samples for deuterium and oxygen-18 were reanalyzed in seven subjects, values of TEE were in close agreement (coefficient of variation = 4.3%), as previously described (12). CO₂ production rates were determined by using a fixed assumption for the dilution space ratio (1.0427) with Eq. R2 of Speakman et al. (29), and energy expenditure was calculated with Eq. 12 of de Weir (8) by using a mean value for the dietary food quotient of 0.88 obtained from the foods provided.

Assessment of physical AEE. Physical AEE was estimated by subtracting SEE from TEE after reducing TEE by 10% to account for the thermic response to meals. SEE was used instead of resting energy expenditure to estimate AEE, because SEE encompassed a much longer period of assessment and had a 45% lower standard deviation than resting energy expenditure.

O_2 max. In the morning and after an overnight fast, O_2 max was determined by indirect calorimetry on a treadmill by using a modified Bruce Protocol to exhaustion. Volumes of O₂ and CO₂ were measured continuously by open-circuit spirometry and analyzed by using a Sensormedics metabolic measurement cart (model 2900, Yorba Linda, CA). Heart rate was monitored by a Polar Vantage XL heart rate monitor (Polar Beat, Port Washington, NY). The highest O₂ uptake, respiratory exchange ratio (RER), and heart rate achieved within the last 2 min of exercise were recorded as O_2 max, maximum RER, and maximum heart rate. The three criteria for achievement of O_2 max were 1) a leveling or plateauing of O_2 max; 2) RER > 1.1; and 3) maximum heart rate within 10 beats of age-predicted maximum.

³¹P-MRS. ¹H-magnetic resonance images (MRIs) and ³¹P-MRS were collected on the right calf muscle with a 4.1-T whole body imaging and spectroscopy system. Subjects were studied on 2 separate days. A series of resting calf muscle MRIs were collected on the first day to measure maximum cross-sectional area (CSA) of the gastrocnemius and soleus muscles. The images were collected by using a torroid coil with the following protocol: repetition time = 1,000 ms, echo time = 14.5 ms, 256-mm field of view, and 5-mm slice thickness with a slice separation of 10 mm. The CSA of the gastrocnemius and soleus muscle group was determined by manually drawing the area around both muscles from the MRIs of each slice. The maximal CSA was subsequently used to normalize force output between individuals. In addition, the subjects practiced performing isometric plantar flexion. Maximal isometric plantar flexion was measured three different times across 2 days, with the highest force defined as maximum plantar flexion force output.

Muscle biopsy procedure. Muscle biopsies were obtained on a subset of 18 subjects. Samples were removed under local anesthesia (1% lidocaine) from the left lateral gastrocnemius by percutaneous needle biopsy by using a 5-mm Bergstrom biopsy needle under suction, as our laboratory has previously described (2). Portions used for enzyme assays were snap frozen in liquid nitrogen. Samples were mounted on cork with Tissue-Tek optimum cutting temperature mounting medium (Miles, Elkhart, IN) oriented cross sectionally by using a dissecting microscope and were quickly frozen in liquid nitrogen-cooled isopentane. All samples were stored at 80°C until analysis. Because the lateral gastrocnemius has a thin muscle belly, it is possible that some biopsies may have included deeper muscle (e.g., soleus). We attempted to avoid this by entering the muscle at a specific location and angle intended to isolate the gastrocnemius.

Myofiber histochemistry. Muscle blocks were sectioned (10 µm) in a cryostat microtome cooled to 22°C. Myofibers were classified as type I or type II by metachromatic dye-ATPase histochemistry by using methods described previously (22) and modified in our laboratory. Metachromasia was revealed by 0.1% toluidine blue after acid preincubation (pH 4.4) and incubation in 0.15% ATP disodium salt (pH 9.4). Type I myofibers were colored turquoise. Type II myofiber subtypes ranged in color from light gray (type IIa) to violet (type IIx). For this analysis, we did not differentiate the type II subtypes. Microscope views were captured by a color digital camera interfaced with a 500-MHz desktop personal computer. Myofiber CSA by type and fiber-type distribution was determined by using Mocha (Jandel Scientific) image analysis software. To assess the cross-sectional quality of individual fibers, each myofiber was traced, and its shape factor (4 · area/perimeter²) was computed. The average CSA of 25-50 fibers with the highest shape factors (>0.60) for each fiber type was determined and used for analysis. Myofiber-type distribution was determined from 154 ± 56 myofibers in each sample.

Biochemical analysis. Muscle samples were kept at 80°C and shipped on dry ice to Laval University and assayed for enzyme activity, as previously described (27). Briefly, small pieces (~10 mg) of unmounted muscle were homogenized in a glass-glass Duall homogenizer with 39 volumes (weight/volume) of ice-cold extracting medium (0.1 M Na-K-phosphate, 2 mM EDTA, pH 7.2), and maximal (V_max) activity levels of CS (EC 4.1.3.7), COX (EC 1.9.3.1), 3-hydroxyacyl-CoA-dehydrogenase (EC 1.1.1.35), phosphofructokinase (PFK; EC 2.7.1.11), glyceraldehyde-3-phosphate dehydrogenase (EC 1.2.1.12), CK (EC 2.7.3.2), and glycogen phosphorylase (Phos; EC 2.4.1.1) were measured spectrophotometrically at 30°C. Values of the enzyme activities are expressed in units of micromoles of substrate per minute per gram of wet weight tissue (µmol · min¹ · g¹). Determination of these enzyme activities has been shown to be reproducible (10, 28).

Statistics. Zero-order Pearson product correlation was used to determine the relationship between age and muscle variables. Multiple regression procedures were used to determine partial correlations between age and each of the muscle variables after adjusting for AEE. Alpha was set at 0.05, and SPSS for Windows 8.0 (SPSS) was used in all analyses.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Means and SDs for all variables are contained in Table 1. The mean age of the 83 subjects was 34 yr, with a range of 23-47 yr. Subjects described themselves as sedentary, and the relatively low mean O_2 max of 31.4 ml O₂ · kg¹ · min¹ supports this contention. Table 2 shows the relationship among age and performance and muscle variables without adjusting for AEE. Table 3 shows the relationships after adjusting for AEE. Endurance time on the treadmill, O_2 max, and strength were all significantly negatively related to age, even after adjusting for AEE and lean tissue. O_2 max was also adjusted for body weight in a separate analysis. The body weight adjustment would be considered to be the best measure of aerobic capacity relative to functional demands, and the lean tissue adjustment would be considered a better measure of aerobic capacity relative to body size potential. O_2 max was significantly negatively related to age, whether body weight or lean tissue was used as an additional adjusting variables with AEE.

³¹P-MRS. Both maximal recovery rate of ADP and maximal anaerobic glycolytic rate were significantly negatively related to age, whether or not they were adjusted for AEE. Although maximal CK activity was significantly negatively related to age before adjustment, it was not significantly related to age after adjusting for AEE.

Muscle enzyme activity. The percentage of type I muscle fiber in the gastrocnemius was positively related to age, whether unadjusted or adjusted for AEE. In addition, CS, glyceraldehyde-3-phosphate dehydrogenase, PFK, and Phos activity all were significantly and negatively related to age, both adjusted and unadjusted for AEE. Although not significant, CK and -hydroxyacyl dehydrogenase activity both approached significance with age (3-hydroxyacyl-CoA-dehydrogenase = 0.36, P = 0.09 and CK = 0.39, P = 0.06). COX was not significantly related to age. Table 3 also contains regression equations for estimating each of the significant partial correlates with age. Estimated decreases in muscle function for 10-yr increases in age are substantial and clinically relevant, varying from 6.6% for O_2 max adjusted for lean tissue and AEE to 42.6% for CS activity adjusted for AEE (Table 3).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Even in this relatively young group of premenopausal women, both MRS and biopsy results showed that muscle metabolic capacity of skeletal muscle was inversely related to age, independent of variations in their levels of free-living AEE. This was the case for markers of glycolytic and oxidative metabolism, suggesting that both anaerobic and aerobic work capacity decline with age. This decline in muscle metabolic capacity occurred, despite no significant negative relationship with age in plantar flexor muscle CSA. A number of studies have shown reduced aerobic capacity (7, 13, 15) and reduced glycolytic capacity (13) or less abundant mRNA for encoding enzymes involved in glucose metabolism (32) in muscle of older adults. To our knowledge, this is the first study to demonstrate that the age-related decrease in muscle metabolic capacity 1) is apparent as early as between the 3rd and 5th decades of age and 2) is independent of variations in AEE. These results can be generalized to women 23-47 yr of age. Similar research is needed for men of this age.

Holloszy et al. (13) have shown that older rats have decreased activity of enzyme markers of respiratory capacity in the primarily type I soleus muscle, whereas Barazzoni and Nair (3) have shown a reduction in COX activity in all tissue except heart in older rats. Houmard et al. (15) have shown that CS activity is reduced in gastrocnemius but not vastus lateralis in older adults (60 yr of age) compared with younger adult humans (30 yr of age). Using ³¹P-MRS, Conley et al. (7) have recently shown that, after maximal electrical stimulation of the vastus lateralis muscle, the time constant for PCr is elevated in older adults (mean age: 69 yr), suggesting that oxidative capacity is reduced up to 50% compared with a group of younger adults (mean age: 39 yr).

None of these studies evaluated activity patterns. It has been hypothesized that decreased muscle oxidative capacity may occur because of reduced physical activity patterns in older adults. Kent-Braun and Ng (16) have recently attempted to test this hypothesis by evaluating muscle oxidative capacity in young and old subjects who have similar activity patterns. Using the ³¹P-MRS-determined PCr time constant to measure oxidative capacity of the tibialis anterior muscle, they found no difference in muscle oxidative capacity between a group of younger (mean age: 34 yr) and older subjects (mean age: 76 yr) who were matched for triaxial accelerometer-measured physical activity. Accelerometer counts do not measure energy expended in physical activity directly. In addition, evaluation of an older age group and different muscle group, tibialis anterior vs. plantar flexors, may have contributed to differences in results between the Kent-Braun and Ng study and the present study.

In another study, Proctor et al. (24) found that the vastus lateralis muscle CS activity was similar in older and younger aerobically trained men. Although this study also suggests that the age-related decrease in muscle oxidative capacity may be prevented by maintaining high physical activity levels, it is difficult to compare groups of trained and untrained subjects cross sectionally. Differences in physical activity did not affect the age-related decline in muscle oxidative capacity found in the present study; however, the women in this study were relatively sedentary and did little or no regular training. The study of Proctor et al. also compared older subjects (21-30 vs. 51-62 yr of age) and different muscle groups than those used in this study. Only speculation can be made concerning what impact high-intensity training would have on the age-related decline in muscle oxidative capacity in this group of relatively young sedentary women.

One hypothesis for metabolic decline during aging is that oxidative damage to mitochondrial DNA may affect expression of mitochondrial enzymes and other mitochondrial proteins by altering gene transcription (26). In fact, there is now strong evidence that mitochondrial DNA damage increases with age in skeletal muscle (20). Transcription rate has not been examined in muscle of older persons; however, mRNAs for mitochondrial proteins have been reported to be reduced in skeletal muscle of older persons (32).

Most previous studies of which we are aware have demonstrated age-related declines in muscle oxidative enzyme activity or ³¹P-MRS muscle oxidative capacity by cross-sectional comparisons between groups in the 4th decade vs. the 7th or 8th decades (7, 15, 16, 23). We report an age-related decline in muscle oxidative capacity across a younger age continuum (23-47 yr). Rooyackers et al. (25) also found a reduction in CS and COX activity when comparing younger adults between 20 and 52 yr of age.

We are aware of no ³¹P-MRS studies that compared the capacity of ATP production from the CK reaction and anaerobic glycolysis between young and old subjects. Consistent with our findings of a negative relationship of age with ³¹P-MRS-derived anaerobic glycolysis and with muscle fiber PFK, glyceraldehyde-3-phosphate, and Phos activity, Holloszy et al. (13) reported that the glycolytic capacity of superficial type IIb and deep IIa muscle fibers was reduced in vastus lateralis of older rats, and Welle et al. (32) reported less abundant mRNA for encoding enzymes involved in glucose metabolism in vastus lateralis of older humans. Neither study reported measures of physical activity. Thus our data extend these findings to suggest that the age-related decrease in muscle glycolytic capacity is independent of changes in physical activity.

Because other markers of muscle oxidative capacity (e.g., recovery rate of ADP and CS) were independently related to age, it was surprising that COX was not also independently related to age. However, we have previously reported that COX is much more weakly related to ³¹P-MRS-derived measures of muscle metabolic capacity than CS activity (19). This finding may be explained on the basis that COX may not be as good a biochemical marker of overall oxidative capacity as CS (19). Whereas both enzymes correlate well with changes in oxidative metabolism induced by exercise training, COX is generally not thought to be rate-limiting in the electron transport chain. CS activity appears to be a better marker of whole tissue mitochondrial density (14) and mitochondrial content (31).

In summary, these data suggest that, at least in sedentary women of premenopausal age, both oxidative and glycolytic capacity decrease with age even when taking into account normal variation in physical activity levels. These results are consistent with the finding that exercise training only partly counteracts many other aspects of aging (9), suggesting that both biological aging and physical activity may be involved. This does not suggest that exercise training will not improve oxidative and glycolytic capacity in aging muscle. In fact, it suggests that exercise training in older adults with reduced biological potential may be even more important for maintaining muscle metabolic capacity at levels that will allow adequate muscle function for a high quality of life.