Single muscle fiber adaptations to resistance training in old (>80 yr) men: evidence for limited skeletal muscle plasticity

Subjects.

Six men were recruited from the local community for this investigation (age 82 ± 1 yr, height 173 ± 1 cm, weight 74 ± 4 kg). All subjects were over 80 yr of age (80–86 yr), nonexercising, nonobese (BMI 25 ± 1 kg/m²), and otherwise healthy. Each subject underwent a physical examination, which included medical history, blood and urine chemistries, resting and exercising electrocardiogram, and blood pressure before participating in any resistance training. Subjects were excluded if they had any acute or chronic illness; cardiac, pulmonary, liver, or kidney abnormalities; uncontrolled hypertension; insulin- or noninsulin-dependent diabetes; abnormal blood or urine chemistries; arthritis; a history of neuromuscular problems; or if they smoked tobacco. Additionally, these free-living men were not consuming any chronic medication (prescribed or over-the-counter). All subjects were given oral and written information about the experimental procedures and potential risks before giving written consent. All procedures conformed to the standards set forth by the Declaration of Helsinki, and these procedures were approved by the Institutional Review Boards of Ball State University and Ball Memorial Hospital.

Experimental design.

All subjects underwent testing for whole thigh muscle cross-sectional area, knee extensor (thigh) strength, vastus lateralis single muscle fiber physiology, and MHC distribution. These procedures were completed before and after the 12-wk PRT program. For this investigation, all of the methods and instrumentation employed for the resistance training program, whole muscle analysis, and single muscle fiber analysis were identical to our previous investigations in men a decade younger (64).

Whole muscle strength.

Bilateral isotonic one-repetition maximum (1-RM) of the knee extensor muscles was assessed using a seated-knee extension device (Cybex Eagle, Medway, MA). The 1-RM procedure was preceded by 10 min of warm-up exercise (self-selected workload, ∼50 W) on a cycle ergometer (Monark 828E, Vansbro, Sweden). The 1-RM was determined by incrementally increasing the amount of weight on the device with each successful lift until the subject was not able to maintain proper form and/or fully extend his legs at a given weight. The last weight successfully lifted was determined to be the 1-RM. The 1-RM procedure was conducted before, every 2 wk during, and after completion of the PRT program. The initial 1-RM procedure was preceded by three equipment and procedure familiarization sessions. These sessions were to help ensure that the initial 1-RM was not limited by a learning effect.

Whole muscle size.

Muscle cross-sectional area (CSA) of the right thigh was measured before and after PRT using computed tomography (CT) on a helical scanner (CTI helical scanner, General Electric, Milwaukee, WI), as described previously (63). Subjects were supine for 30 min before scanning to minimize the influence of fluid shifts on the CSA measurements (3). The anatomical midpoint of the femur was used as the point of measurement for pretraining and posttraining CT measurements. The CSA of the thigh minus the area of bone and subcutaneous fat was determined using computerized planimetry (NIH Image Program, v.1.61; National Institutes of Health, Bethesda, MD).

Progressive resistance training program.

Subjects performed 12 wk of a supervised PRT program designed to strengthen the quadriceps muscle group. Subjects performed bilateral isotonic leg extensions on a seated device (Cybex Eagle, Medway, MA) 3 days per week on nonconsecutive days (36 total sessions). Subjects performed three sets of 10 repetitions (3 × 10) at ∼70% of their 1-RM with 2-min rest between sets. The training was progressive in nature in that 1-RM was assessed every 2 wk, and the weight was adjusted accordingly to maintain an intensity of ∼70%. Each session was preceded by 10 min of warm-up (∼50–75 W) on a cycle ergometer (Monark 828E, Vansbro, Sweden). This entire protocol is identical to several previous investigations from our laboratory (16, 17, 60, 63, 64, 73, 74). All six men completed every training session (n = 36) for 100% compliance to the training program.

Muscle biopsy.

Muscle biopsies (4) were obtained from the vastus lateralis of each subject before the initial training session and after the last training session. Each muscle sample was sectioned into longitudinal sections and placed in cold skinning solution (see below) and stored at −20°C for later analysis of single muscle fiber physiology and single-muscle fiber MHC isoform analysis.

Skinning, relaxing, and activating solutions.

The skinning solution contained (in mM): 125 K propionate, 2.0 EGTA, 4.0 ATP, 1.0 MgCl₂, 20.0 imidazole (pH 7.0), and 50% (vol/vol) glycerol. The compositions of the relaxing and activating solutions were calculated using an interactive computer program described by Fabiato and Fabiato (8). These solutions were adjusted for temperature, pH, and ionic strength using stability constants in the calculations (18). Each solution contained (in mM): 7.0 EGTA, 20.0 imidazole, 14.5 creatine phosphate, 1.0 free Mg²⁺, 4.0 free MgATP, KCl, and KOH to produce an ionic strength of 180 mM and a pH of 7.0 (32). The relaxing and activating solutions had a free [Ca²⁺] of pCa 9.0 and pCa 4.5, respectively (where pCa = −log Ca²⁺ concentration).

Single muscle fiber physiology experiments.

On the day of an experiment, a 2- to 3-mm muscle fiber segment was isolated from a muscle bundle and transferred to an experimental chamber filled with relaxing solution where the ends were securely fastened between a force transducer (model 400A; Cambridge Technology, Lexington, MA) and a direct current torque motor (model 308B; Cambridge Technology) as described by Moss (45). The force transducer and the torque motor were calibrated before each experiment. The instrumentation was arranged so that the muscle fiber could be rapidly transferred back and forth between experimental chambers filled with relaxing or activating solutions. The apparatus was mounted on a microscope (Olympus BH-2, Japan), so that the fiber could be viewed (×800) during an experiment. Using an eyepiece micrometer, sarcomeres along the isolated muscle segment length were adjusted to 2.5 μm, and the fiber length (59) was measured. All single muscle fiber experiments were performed at 15°C.

Unamplified force and length signals were sent to a digital oscilloscope (Nicolet 310, Madison WI), enabling muscle fiber performance to be monitored throughout data collection. Analog force and position signals were amplified (Positron Development, Dual Differential Amplifier, 300-DIF2; Ingelwood, CA), converted to digital signals (National Instruments, Austin, TX), and transferred to a computer (Micron Electronics, Nampa, ID) for analysis using customized software. Servo-motor arm and isotonic force clamps were controlled using a computer-interfaced force-position controller (Positron Development, Force Controller, 300-FC1, Ingelwood, CA).

For each single muscle fiber experiment, a fiber with a compliance (calculated as fiber length divided by y-intercept) greater than 10%, and/or a decrease in peak force of more than 10% was discarded and not used for analysis. The within-fiber test/retest of a single muscle fiber in our lab for the measurements of size, force-power relationships, peak force, and contractile velocity were less than 1%. The coefficient of variation for the force transducer and servo-mechanical lever mechanism during the timeframe of this investigation was 0.5% and 0.6%, respectively.

Single muscle fiber analysis.

The analysis of single muscle fibers included a measurement for diameter, peak force (Po), maximal unloaded shortening velocity (Vo), and power parameters. Detailed descriptions and illustrations of these procedures have been previously published by our laboratory (59, 64).

Single muscle fiber diameter.

A video camera (Sony CCD-IRIS, DXC-107A, Tokyo, Japan) connected to the microscope and interfaced to a computer allowed viewing on a computer monitor and storage of the digitized images of the single muscle fibers. Fiber diameter was determined from a captured computer image taken with the fiber briefly suspended in air (<5 s). With the assumption that the fiber forms a cylindrical cross section when suspended in air, the fiber width is equal to the fiber diameter. Fiber width (diameter) was determined at three points along the segment length of the captured image using NIH public domain software (Scion Image, release Beta 4.0.2, for Windows). The average of these three measurements was the determined single muscle fiber diameter.

Single muscle fiber peak force (po).

The output of the force and position transducers was amplified and sent to a microcomputer via a Lab-PC+ 12-bit data acquisition board (National Instruments). Resting force was monitored, and then the fiber was maximally activated in pCa 4.5 solution. Peak force was determined in each fiber by computer subtraction of the force at baseline from the peak force in pCa 4.5 solution.

Single muscle fiber shortening velocity.

Fiber shortening velocity (Vo) was measured by the slack test technique, as described by Edman (7). The fiber was fully activated in activation solution and then rapidly released to a shorter length, such that force fell to baseline. The fiber shortened, taking up the slack, after which force began to redevelop. The fiber was then returned to its original length. The duration of unloaded shortening, or time between onset of slack and redevelopment of force, was determined by computer analysis. Four different activation and length steps [150, 200, 250, and 300 μm; each ≤ 15% of fiber length (FL)] were used for each fiber. The slack distance was divided by the fiber length and plotted as a function of the duration of unloaded shortening. Fiber Vo (in FL/s) was then equal to the slope of this linear graph. Only experiments in which r² was greater than or equal to 0.98 were included for analysis.

Single muscle fiber power.

Submaximal isotonic load clamps were performed on each fiber for determination of force-velocity parameters and power. Each fiber was fully activated in pCa 4.5 solution and then subjected to three isotonic load steps. This procedure was performed at various loads so that each fiber was subjected to a total of 15–18 isotonic contractions. Force and shortening velocity data points derived from the isotonic contractions were fit using the hyperbolic Hill equation (24). Fiber peak power was calculated from the fitted force-velocity parameters (Po, V_max, and a/Po, where a is a force constant and V_max is the y-intercept). Absolute power (μN·FL⁻¹·s⁻¹) was defined as the product of force (μN) and shortening velocity (FL/s). Normalized power (W/l) was defined as the product of normalized force and shortening velocity.

Single muscle fiber MHC determination.

Following the single-muscle fiber contractile physiology measurements, each fiber was solubilized in 1% SDS sample buffer and stored at −20°C until assayed (15, 74). To determine the MHC composition, fibers were run on a Hoefer SE 600 gel electophoresis system that consisted of a 3.5% (wt/vol) acrylamide stacking gel with 5% separating gel at 4°C. Following gel electophoresis, the gels were silver stained for MHC identification, as described by Giulian et al. (15).

Statistical analysis.

Whole muscle strength was analyzed using a one-way repeated-measures ANOVA using SPSS ver. 14.0 for Windows. Whole muscle size, single muscle fiber contractile function, and MHC distribution were analyzed using a paired t-test. For each single muscle fiber parameter (diameter, Po, Vo, power) within a subject, a mean for each fiber type was determined. The mean value was used to represent all fibers of that fiber type within the given individual. Significance was set at P < 0.05. All data are reported as means ± SE.

Collectively, the hybrid (I/IIa and IIa/IIx) fibers constituted a reasonable proportion of the muscle, but when partitioned into the individual subtypes, there were insufficient data for pretraining to posttraining comparisons. As a result, they were not included in single-muscle fiber contractile function analysis. The pretraining to posttraining comparisons for the single-muscle fiber physiology experiments are focused to the MHC I and IIa fiber types.

Whole muscle strength and size.

Bilateral knee extensor strength, as measured by one-repetition maximum, increased (P < 0.05) 23 ± 4 kg (56 ± 4 to 79 ± 7 kg; 41%) after the 12-wk resistance-training program. There was an increase (P < 0.05) in 1-RM strength every 2 wk from the previous 2-wk strength testing session except for the period between weeks 4 and 6, when the increase in strength did not reach statistical significance (Fig. 1). Whole muscle thigh cross-sectional area increased (P < 0.05) 3 ± 1 cm² (120 ± 7 to 123 ± 7 cm²; 2.5%) from pretraining to posttraining.

Single muscle fiber diameter and contractile properties.

We successfully studied 142 fibers pre- (MHC I = 98; MHC IIa = 44) and 125 fibers post- (MHC I = 83; MHC IIa = 42) resistance training for diameter and contractile function. For each individual, 23 ± 1 were studied before and 21 ± 2 studied after resistance training. Single muscle fiber diameter, Po, and Po/CSA did not change as a result of the 12-wk progressive resistance training in either MHC I or MHC IIa muscle fibers (Table 1). Contractile speed (Vo and V_max) did not change in MHC I fibers with resistance training. MHC IIa fiber Vo increased 6.5% (P < 0.05), while MHC IIa V_max was similar before and after resistance training (Table 2). Single muscle fiber power (absolute and relative to fiber size) was unaltered with the resistance training program (Table 3). The coefficient of variation between fibers was similar before and after resistance training for MHC I and IIa muscle fibers: diameter (17–19%), Po (34–43%), Po/CSA (14–18%), Vo (18–22%), and power (41–52%).

Single-fiber MHC isoform distribution.

A total of 2,215 (pre = 1,202; post = 1,013) single muscle fibers were isolated and studied for MHC composition. For each individual, 200 ± 19 isolated muscle fibers were analyzed before and 169 ± 6 fibers were analyzed after training. No single muscle fibers, analyzed either before or after training, expressed only the MHC IIx isoform. Additionally, no single muscle fiber analyzed coexpressed all three human MHC isoforms (MHCI/IIa/IIx) or a combination of MHC I and IIx isoforms (MHCI/IIx). The percent distribution of the MHC isoforms is presented in Table 4. No change in MHC isoform distribution was evident with the progressive resistance-training program.

Baseline comparisons.

The 82-yr-old men in the current study had 10% less muscle mass (120 ± 7 vs. 134 ± 8 cm²) compared with our previous study in 74-yr-old men. Interestingly, both groups of men had nearly identical 1-RM strength (56 ± 4 vs. 53 ± 5 kg) prior to beginning the resistance-training program. Similar leg muscle strength with less muscle mass would point to potential differences in muscle fiber recruitment, fiber type profile, or the intrinsic properties of the muscle fibers. In support of the latter, the 82-yr-old and 74-yr-old men had a similar fiber type profile with noted differences in single muscle fiber contractile properties before the training program. Generally, the 82-yr-old men's MHC I and IIa fibers were slightly larger, stronger, and faster than the 74-yr-old men, contributing to more powerful muscle fibers, which were more pronounced in the MHC IIa muscle fibers. A comparison between these two studies and the current available literature on normalized power is shown in Table 5. Normalized power was chosen as a variable since it represents the most comprehensive profile of single muscle fiber function, as this measure incorporates size, strength, and speed. Data from these 14 studies, representing 20 different groups of individuals reveal an interesting trend that contrasts the typical dogma of poor fiber quality with age. In old individuals, MHC I and IIa normalized power is comparable to younger individuals and highly trained athletes. Equally interesting is the MHC I and IIa normalized power from spinal cord injury patients, which is on the upper end of the spectrum for what has been reported in human muscle. Collectively, these data suggest that chronic perturbations that induce significant whole muscle atrophy over a period of several years maintain or even slightly improve the power output of the remaining MHC I and IIa muscle fibers when normalized for cell size. The quantitative and qualitative regulation of individual muscle fibers in a challenging environment that promotes muscle cell loss (37) is a complex paradigm that warrants further research.

Whole muscle adaptation.

From our review of resistance training studies that have focused on skeletal muscle in old individuals, it is apparent that a wide range in strength gains were achieved (7 to 174%). This variance is likely the result of differing training protocols and subject profiles. In the current investigation, we observed a 41% increase in knee extensor strength after 12 wk, which is in the median range of the previous reports. Of the more than 50 published resistance-training studies with an aging focus, three papers are most closely related to the current study (9, 34, 64). Previously published data from our laboratory, using the identical resistance training protocol and equipment as the current study, reported strength increases of 50% in men (64), who were approximately a decade younger. Two previous studies have included resistance-trained individuals exclusively over 80 yr and reported a 174% (9) and 134% (34) increase in leg strength. These studies differ from the current study in that the majority of subjects were institutionalized, considerably more frail, and were a mix of both men and women. When the one-legged strength improvements reported from the Fiatarone et al. (9) and Kryger and Anderson (34) studies are doubled to represent a bilateral 1-RM, the absolute amount of weight lifted increased by ∼25 kg and ∼21 kg, respectively. This compares favorably with the 23-kg increase in 1-RM after 12 wk of PRT in the current study. Thus, despite the weaker muscles at the onset of the training program of the institutionalized patients, the absolute amount of strength gain in individuals over 80 yr across these studies was nearly identical. The differential relative strength gains among these studies can be attributed to the strength output at the beginning of the study.

After 12 wk of resistance training, the thigh CSA from these old men increased 3.0 cm² (2.5%). This is considerably less than the 8.2 cm² (7%) increase in thigh CSA that we reported previously in 74-yr-old men (17). These results provide an indication that the magnitude of hypertrophy in response to resistance training may be slightly diminished as men age from their 70s into their 80s. Further support that the whole muscle hypertrophic response gradually decreases with age (60→80+ yr) is illustrated from the literature. Frontera et al. (13) reported a 12 cm² (∼10%) increase in muscle CSA among 60–72 y old men after 12 wk of resistance training. Fiatarone et al. (9) and Kryger and Anderson (34) report similar increases in the percent CSA (∼10%) with a comparable resistance-training program among very old (>80 yr) individuals. However, the absolute gain in muscle CSA was only 2.7 cm² (34) and 4.6 cm² (9), which is very similar to the 3.0 cm² increase in CSA from the 82-yr-old men in the current investigation. When examined on the basis of muscle CSA gains in absolute terms, these studies support a progressive decrease in the hypertrophic response to resistance training at the whole muscle level with age (80+ yr < 70 yr < 60 yr).

Single muscle fiber adaptations.

The unique aspect of this investigation was the single muscle fiber contractile function measurements in men over 80 yr before and after resistance training. Contrary to our hypothesis, no improvement was observed in single muscle fiber contractile function with the resistance-training program. This was a surprising and interesting finding in the context of the related literature on this topic. We have identified 64 published papers that have examined single-muscle fiber contractile function in humans. Of these studies, 24 have incorporated an intervention consisting of sports training responses (n = 8) (10, 21, 23, 38–40, 56, 61), unloading paradigms (n = 11) (11, 36, 57, 58, 62, 65–68, 71, 75), resistance training in young adults (n = 1) (69), and resistance training in old adults (n = 4) (12, 16, 60, 64). Of these human studies, 22 (92%) have reported alterations in contractile function with an intervention, and all five resistance-training studies have shown an improvement in contractile function. The exceptions are two sprint-training studies in young individuals (23, 38). Collectively, the consistent finding of alterations in single muscle fiber contractile function with a multitude of muscle perturbations provides strong support for the interpretation of limited muscle plasticity at the cell level with resistance training in men over 80 yr.

The limited muscle plasticity among the very old is further supported by Singh et al. (53) and compared with the improved cellular adaptations observed in our previous resistance training studies among men nearly a decade younger (64) (see Figs. 2 and 3). Singh et al. (53) reported no change (or a slight decrease) in muscle fiber size with a similar exercise program in very old (83 ± 3 yr) frail individuals. Our previous group of 74-yr-old men had substantial improvements in MHC I contractile function, with less improvement in MHC IIa function (64). When these data are coupled with single muscle fiber responses to resistance training in young men (69), a muscle fiber-specific continuum of adaptation is apparent. With resistance training, young individuals improve MHC I and IIa single-muscle fiber function; individuals in their 70s lose a large portion of the ability to increase MHC IIa contractile function but retain the ability of MHC I fibers to improve; individuals in their 80s do not improve MHC I or IIa single muscle fiber function. These data suggest that the cellular adaptive potential is gradually reduced with age in a fiber-specific manner.

The pretraining MHC profile from old men in the current study was composed of ∼25% hybrid MHC muscle fibers. In aging muscle, a relatively high proportion of single muscle fibers coexpressing multiple MHC isoforms have been previously observed (31, 34, 74). A high proportion of hybrid muscle fibers are not exclusive to aging muscle, as increases in hybrid fibers have been reported with a decrease in physical activity (41, 62). Conversely, individuals with high physical activity levels (i.e., athletes) typically have very few hybrid muscle fibers (22, 30, 56). Thus, there is strong support that changes in physical activity patterns can alter the single muscle fiber hybrid population. Using the identical resistance training protocol as the current study, we have previously shown that inactive young (73) and old (74) individuals significantly decrease the amount of hybrid muscle fibers after 12 wk of training. While both the young (∼25 yr) and the old (∼74 yr) had similar decreases in hybrid muscle fibers (∼20%), the posttraining profile for pure MHC phenotypes showed a distinct pattern of adaptation that paralleled the single muscle fiber contractile function changes highlighted above. That is, the 20-yr-old individuals increased the MHC IIa phenotype with training, while the 74-yr-old individuals increased the MHC I phenotype with training. In the current study, the static nature of the single-muscle fiber MHC profile with resistance exercise also mirrors the single muscle fiber contractile function findings and provides further evidence that the cellular adaptations to resistance exercise are extremely limited in the very old. This is also supported by the recent study by Kryger and Andersen (34) who reported no change in MHC profile with resistance exercise in very old (85–97 yr) men and women.

Perspectives and Significance

In the current study, whole muscle strength increased with resistance training, which most likely translated to improved functional status of the very old men (25). The fact that we observed a minor increase in whole muscle size with no alterations in single muscle fiber contractile function or MHC profile suggest that the strength gains observed in these old men following the resistance training program were primarily the result of improvements to the nervous system (i.e., muscle fiber recruitment, synchronization of motor units, etc.). In young healthy individuals, initial strength gains during a resistance-training program are due mostly to neural adaptation, with muscle hypertrophy becoming the dominant factor beyond 4 wk of training (42, 51). This time-course of neural and hypertrophic adaptation to resistance training is altered in old individuals with a greater reliance upon the neural component for strength gains (43). The finding of limited remodeling in single muscle fiber function or structure supports the idea that the physiological regulation for muscle hypertrophy is attenuated distal to the nerve in the skeletal muscle of individuals over 80 yr. This hypothesis is further strengthened by the pedigree of single-muscle fiber research in humans showing improvements in myocellular function with a resistance exercise intervention. Thus, the capacity for men over 80 yr to gain strength with resistance training is severely limited by the absence of a myocellular adaptive response. Potential molecular mechanisms from studies integrating aging, muscle growth, and resistance exercise are beginning to emerge that point to an imbalance of anabolic and catabolic regulation that are most likely acting in concert to alter the adaptive process to exercise (14, 19, 28, 29, 47, 48, 72). Future research should focus on progressive aging (60 vs. 70 vs. >80 yr) and muscle fiber type-specific mechanisms that regulate skeletal muscle growth and function at the molecular level. The rapidly expanding population over 80 yr and the increasing burden on our health care system necessitate a more complete understanding of the etiology of sarcopenia to develop targeted therapies to promote muscle health.