It is generally believed that the maximum shortening velocity (Vo) of a skeletal muscle fiber type does not vary unless a change in myosin heavy chain (MHC) isoform composition occurs. However, recent findings have shown that Vo of a given fiber type can change after training, suggesting the hypothesis that the function of myosin can vary without a change in isoform. The present study addressed the latter hypothesis by studying the function of isolated myosin isoforms by the use of the in vitro motility assay (IVMA) technique. Four young (age 23–29 yr, YO) and four elderly men (age 68–82 yr, EL) underwent a 12-wk progressive resistance training program of the knee extensor muscles and to one pre- and one posttraining biopsy of the vastus lateralis muscle. The significant increase in one-repetition maximum posttraining in both YO and EL indicated that training was effective. After training, MHC isoform composition showed a shift from MHC2X toward MHC2A in YO and no shift in EL. The velocity of sliding (Vf) of actin filaments on pure myosin isoforms extracted from single fibers was studied in IVMA. One hundred sixty IVMA samples were prepared from 480 single fibers, and at least 50 filaments were analyzed in each experiment. Whereas no training-induced change was observed in Vf of myosin isoform 1 either in YO or in EL, a significant increase in Vf of myosin isoform 2A after training was observed in both YO (18%) and EL (19%). The results indicate that resistance training can change the velocity of the myosin molecule.
- myosin isoforms
- velocity of shortening
it is generally believed that unloaded shortening velocity (Vo) of single skeletal muscle fibers mostly depends on their myosin heavy chain (MHC) isoform composition (10). On the basis of MHC isoform content, human muscle fibers can be classified in three pure fiber types (1, 2A, and 2X) and two hybrid fibers type (1–2A and 2AX) whose Vo increases in the order type 1 → 1–2A → 2A → 2AX → 2X (10). According to the MHC-based regulation of Vo, a given fiber type should have the same mean values of Vo regardless the muscle of origin (23) and the conditions in which the muscle works in the body (8).
Results from recent studies suggest that this might not always be true. Vo of both slow and fast fibers has been found to be lower in aging (14, 28) and higher in disuse or immobilization (41) than in control conditions. The latter findings have opened the interesting possibility that, besides the MHC isoform content, additional mechanisms can modulate Vo of muscle fibers (8). Age-related changes in Vo have been attributed to an alteration of the function of the myosin molecule itself (24), whereas in disuse no such change occurs (14) and the underlying mechanism might be related to an alteration of the geometry of the sarcomere (41).
Although less information is available as to the effects of exercise on Vo of single muscle fibers, significant changes have been observed in elderly subjects. For example, Vo of type 1 fibers, but not of type 2A fibers, was significantly higher in elderly elite endurance trained runners than in elderly sedentary subjects (43). In addition, Vo of both type 1 and 2A fibers was found to increase in elderly male subjects after a period of resistance training (40), although no change was observed in female elderly subjects (39). In young subjects, contradictory results have been obtained. No change in Vo was observed after resistance (42) or sprint (22) training, whereas an increase in Vo was observed after a decrease in swim training volume (37).
No study so far has addressed the mechanism underlying the observed training induced changes of Vo. The purpose of the present study was to assess whether a 12-wk progressive resistance training program can modulate the function of the myosin molecule. Because training might have a differential impact on young and elderly subjects, analysis was performed on both populations undertaking the same training regimen. To dissect the role played by the myosin molecule from a possible modulatory influence of other factors such as the geometry of the sarcomere and the role of the other myofibrillar protein isoforms, the velocity of sliding of unregulated actin on myosin (Vf) was studied by use of in vitro motility assays (IVMA). In IVMA, actin sliding velocity solely depends on the properties of myosin because the sarcomeric organization is lost and the other myofibrillar proteins are not present. The present findings indicate that high-resistance strength training can modulate the function of the myosin molecule.
Two groups of subjects were enrolled: four young men, 23–29 yr old, with no previous record of muscular disease or traumatic lesions, and four healthy [as defined by Greig et al. (21)] elderly men, 68–82 yr old. Both the young and elderly subjects were not involved in any regular exercise training before being enrolled in the study and most performed recreational activities, i.e., walking regularly and gardening for the elderly and occasional sport activities for the young subjects.
The study was approved by the Ethics Committee of the Royal Free Hospital (London, UK) and by the ethical committee of the University of Pavia and conformed to the standards set by the Declaration of Helsinki (last modified in 2000). After subjects were fully informed of the goal of the experiments and of the risks involved in the biopsy procedure, written, informed consent was obtained.
All subjects performed a 12-wk progressive training program of the knee extensor muscles. The progressive training program consisted of three sessions per week in which four sets of six repetitions of lifting and lowering a weight equivalent to 80% of the maximum amount of weight that they could lift (1-RM) were performed. Two minutes recovery was given between sets. 1-RM was retested every 2 wk, and the training load was adjusted accordingly to ensure that the training was progressive.
Muscle biopsies (∼100 mg) were obtained from the vastus lateralis muscle by using the needle technique (7) with applied suction. All biopsies were taken at midfemur length, defined as the distance from the anterior inferior iliac spine to the middle of patella. Biopsy samples were taken 1 wk before initiation of the training program and 24 h after the final training session. Each muscle sample was divided into several fiber bundles, placed in cold skinning solution with 50% of glycerol, and then stored at −20°C for later analysis. After muscle sampling all experiments were completed over a 30-day period. On each day of experiment, one fiber bundle was removed from the freezer. The fiber bundles were used for 1) the analysis of isolated myosin isoforms in IVMA and 2) electrophoretic analysis of the distribution of MHC isoforms to assess whether a shift in muscle phenotype occurred after training.
To be able to compare the actin sliding velocity on pure myosin isoforms, an approach based on extraction of myosin from single pure fibers was used (13). Because single fibers mostly contain only one type of MHC isoform, they are a convenient source of pure myosin isoforms. However, because of the short length of the fibers in a needle biopsy sample, one fiber segment does not provide sufficient myosin to be used in IVMA. Therefore, 1) single fibers were dissected, 2) SDS-PAGE was performed (9) to identify MHC isoform content, 3) three to four fibers shown to contain the same MHC isoform were pooled, 4) myosin was extracted from pooled fibers, and 5) myosin was loaded in IVMA. To maintain myosin function in IVMA, all the above procedures had to be performed within 48 h after the removal of a fiber bundle from the freezer.
As can be seen in Fig. 1, three bands corresponding to MHC1, MHC2A, and MHC2X could be separated by SDS-PAGE in the area of migration of MHCs. MHC isoform distribution of the vastus lateralis muscles was determined by densitometric analysis of the SDS-PAGE gels, performed according to Bottinelli et al. (9). Each sample to be loaded in the gels was obtained by pooling together transverse sections (1-mm length) of all the fiber bundles dissected from the biopsy to include segments of all fibers of the biopsy. The SDS-PAGE analysis of each sample was repeated three times, and densitometry was performed at three different but consistently chosen positions along the bands in each run. Variability among the three repeated measurements of the same sample was small (∼2–3%). The three repeated measurements were averaged. Therefore, at the end of the analysis, each sample provided a single value of MHC distribution that was averaged with the other values of the same group of subjects to assess the mean MHC isoform distribution.
To obtain pure myosin isoforms to be used in IVMA the following procedure was used. Single fibers were manually dissected from each biopsy sample, chemically skinned for 1 h in skinning solution containing 1% Triton X-100, and cut into two further segments. The shorter segment (∼2 mm long) was characterized for MHC isoform composition as described by Bottinelli et al. (9). A small number of hybrid and pure 2X fibers were found and excluded from this study. Thus the comparisons between young and older subjects and pretraining to posttraining were limited to the MHC1 and MHC2A isoform-containing fibers.
Myosin light chain (MLC) isoform separation and identification was performed as previously described in detail (9). Three essential MLC isoforms (MLC1s, MLC1f, and MLC3f) and two regulatory MLC isoforms (MLC2s and MLC2f) could be separated on 10–20% linear gradient gels (Fig. 2). MLC analysis could not be performed on the same fibers used for myosin extraction because the amount of protein left after MLC identification was not sufficient for IVMA analysis and vice versa. A population of single muscle fibers was therefore dissected from the same biopsies just for MLC analysis.
The remaining segments (∼10 mm long) of at least 3 “pure” (containing the same type of MHC isoform) fibers were pooled together and utilized for extraction and purification of myosin as described by Canepari et al. (13). From each biopsy sample at least 10 IVMA samples were prepared so that 480 single fibers were studied.
The myosin sample was put as a drop on a coverslip with nitrocellulose, which was then used to construct the flow cell with a 30-μl channel. The IVMA was carried out as described by Anson et al. (3). The Vf on myosin samples was determined at 25°C, 50 mM ionic strength, and pH = 7.2. The analysis of the velocity of each sample was done as described by Canepari et al. (13). For each myosin sample the velocities of ∼50 filaments were measured, and their distribution was characterized according to parametric statistics.
All data were expressed as means ± SE. Statistical significance of the differences was assessed by two-way ANOVA for repeated-measures analysis followed by the Student-Newman-Keuls test. A probability of less than 5% (P < 0.05) was considered to be significant.
Because analyses were performed on young and elderly subjects and pre- and posttraining, the effect of both training and aging on MHC isoform distribution and myosin function could be studied. In terms of in vivo function, the elderly subjects were able to lift (∼40%) less than their young counterparts before training (Fig. 3A); however, after the 12-wk training program the weight that could be lifted was significantly increased in both young and older subjects (Fig. 3A). This increase was not significantly larger in young subjects (63 ± 13%) than in elderly subjects (40 ± 7%). Although the extent of the response to training varied from subject to subject, all young and all elderly subjects improved their 1-RM.
Strength Training and MHC Isoform Composition
Pretraining, the MHC isoform composition of vastus lateralis muscle of young and elderly subjects was not significantly different (Fig. 4). Strength training significantly shifted MHC isoform composition toward MHC2A in young subjects with a corresponding decrease in MHC2X content, whereas no significant training induced change was observed in elderly subjects (Fig. 4).
Strength Training and Actin Sliding Velocity
A very large number of single muscle fibers (n = 480) were dissected from the biopsy samples, individually typed, and pooled together in groups of two to three fibers on the basis of their MHC isoform content to enable extraction of a sufficient amount of pure myosin isoforms to load in in vitro motility assays. As expected on the basis of previous findings, Vf on myosin 1 was significantly lower than Vf on myosin 2A both in young and elderly subjects (14) (Fig. 5). When Vf on myosin 1 and 2A from the young subjects was compared with that from the older subjects before the initiation of the training program, no significant difference was observed between the two groups. As a result of the training program, no change was observed in Vf of myosin 1 in either the young or elderly subjects. However, a significant increase in Vf was observed in myosin 2A in both the young (18%) and elderly (19%) subjects.
To assess whether the differences in Vf could be due to a modulatory role of MLC isoforms, a total of 60 2A fibers (15 from both young and elderly subjects, pre- and posttraining) were studied as for MLC content. No differences were observed either in the regulatory MLC isoform (MLC2s and MLC2f) content or in the essential MLC isoform (MLC1s, MLC1f, and MLC3f) content pre- and posttraining: 2A fibers invariably expressed only MLC2f, MLC1f, and MLC3f.
The aim of the present investigation was to study the effects of strength training on myosin function in young and older men. To reach this goal we employed the IVMA technique, which enabled us to study myosin function in the absence of the sarcomere structure and of the other sarcomeric proteins that might modulate the properties of the molecule. The findings show a change in myosin function after training suggesting that exercise activity is able to modulate shortening velocity acting directly at the level of the myosin molecule.
Effect of Aging
The data show no clear effect of aging on both MHC composition and Vf. There is some evidence that denervation is more evident for fast motor units, at least in some peripheral muscles (11, 30). However, the evidence for a shift toward an increase in the proportion of slow fibers in major muscle groups such as the quadriceps, as would be expected on the basis of such a selective denervation, has not been unequivocally demonstrated (4, 31). Some recent works have even shown a shift toward a faster phenotype (14, 18, 27). Such lack of consistency has been attributed to the fact that in aging muscle phenotype can be affected by several factors (disuse, hormonal changes, malnutrition, partial denervation of muscle fibers) and that such factors are bound to play a different role in different subject groups (14). Collectively these results indicate that skeletal muscles in aging are unlikely to have a distinctive MHC isoform composition. Consistent with such a hypothesis, in our laboratory’s previous study (14) elderly subjects were sedentary and had a faster muscle phenotype than young controls whereas in the present study elderly subjects were active and had a phenotype similar to young subjects.
The lack of a significant aging related effect on Vf is not consistent with recent findings that showed lower Vf of myosins of elderly subjects (14, 25). Such inconsistency might find an explanation in a variable impact of aging on active vs. inactive (or less active) subject groups and/or on the role of neuromuscular activity in modulating velocity of muscle fibers and isolated myosin. As discussed above, the elderly subjects enrolled in this study were healthy and reasonably active, whereas the subjects enrolled in our previous study (14) were sedentary. Indeed, two very recent studies have reported contradictory results on the impact of aging on Vo of single muscle fibers studying different subject groups: Trappe et al. (38) have shown no difference in Vo, whereas D’Antona et al. (14) have shown lower Vo in single fibers from elderly subjects.
Effect of Strength Training
The exercise regimen employed resulted in an increase in the strength of the knee extensor muscles in both young and older subjects as determined by their 1-RM, with the increases being similar to the values reported in the literature (40, 44).
In the young subjects, the exercise regimen resulted in an increase in the relative proportion of the muscle occupied by MHC2A isoforms in the vastus lateralis muscles, this being at the expense of MHC2X isoforms. Previous studies have reported a similar modulation of 2A and 2X isoform expression after resistance training (2, 17, 32). On the contrary, the exercise regimen did not significantly affect MHC isoform composition of vastus lateralis of elderly subjects, suggesting that skeletal muscle is less responsive to resistance training in elderly than in young subjects. The latter finding is in agreement with several previous observations on the effect of resistance training on fiber-type distribution of elderly subjects determined by ATPase histochemistry (16, 19). A recent study (44), using an approach based on individual fiber dissection and characterization on the basis of MHC isoform content, showed that training has some effect on MHC expression, i.e., the proportion of hybrid fibers was found to decrease. The technique used in this study, based on electrophoretic separation of MHC isoforms from whole muscle samples, is now the approach of choice to assess the overall MHC isoform distribution of skeletal muscles, but it cannot give information on coexpression of MHC isoforms in the same fiber.
Several previous studies have suggested that exercise training can affect unloaded shortening velocity (Vo) of single skeletal muscle fibers in elderly subjects, although contradictory results have also been reported. A cross-sectional study has suggested that endurance training increases Vo of type 1 but not type 2A fibers in elite master runners (43). Trappe et al. (40) showed a resistance-training related increase in Vo of both slow- and fast-twitch muscle fibers in elderly men, but not in elderly women (39). In young subjects, no changes in Vo have been observed after resistance (42) or sprint training (22), whereas an increase in Vo was observed after swim taper, i.e., a change in training volume (37). Notwithstanding some inconsistencies, collectively the latter findings indicate that Vo of single muscle fibers can indeed change with no change in myosin isoform content at least in some subject groups and some experimental conditions. The variability in the response of Vo to training could depend not only on the different age but also on the different exercise habits and training background of the subject groups before the study.
This study shows that resistance training can result in an increase in Vf on myosin 2A obtained from both young and elderly subjects in an in vitro motility assay. Such a finding represents the first demonstration that training can affect the properties of the myosin molecule itself. Interestingly, this is also the first example of an increase, and not a decrease as in aging (14), of the intrinsic velocity of myosin with no change in isoform. The lack of effect of training on Vf of myosin 1 suggests that the training regime preferentially affected fast fibers consistently with previous observations indicating preferential hypertrophy of fast fibers after this kind of training (1).
The results suggest that a change in the properties of myosin itself is a likely determinant of the increase in Vo of single muscle fibers after training reported in several works (37, 40, 43). It cannot be ruled out that other, still unknown, factors might be involved in the training-induced increase in Vo as well. Among the possible factors, we can consider changes in MLC isoform composition, existence of yet-unknown MHC isoforms, posttranslational modifications of myosin, modulatory role on myosin function of other myofibrillar proteins (troponins, titin, myosin binding protein C) and of changes in the geometry of the sarcomere. However, because in IVMA isolated myosin works in the absence of sarcomere structure, the role of other myofibrillar proteins and of the geometry of the sarcomere in modulating Vf can be ruled out (12, 13). Moreover, it is very unlikely that changes in MLC composition can be involved in the observed changes of Vf. No training-induced effect on MLC content in a subpopulation of single muscle fibers from the same biopsies was found, consistent with previous results (44). Previous findings failed to show an impact of MLCs in modulating Vo in human muscle fibers (14, 29), although a significant effect of MLC isoform composition on Vo has been demonstrated in single muscle fibers from rat (9) and to some extent rabbit (36). Finally, the increase in Vf of myosin 2A after training is very unlikely because of undetected coexistence of MHC isoforms or the existence of yet unknown MHC isoforms. Some evidences have suggested that more than one MHC1 isoform exist, at least in small mammals (15, 20, 26). Such findings open the possibility that two MHC1 isoforms with different functional properties can be expressed in type 1 fibers in variable proportions and can be responsible for the changes in Vo of this fiber type in several conditions, including after training (8). However, in this study training-induced changes were observed in myosin 2A, and no evidence has been reported that an extra fast MHC isoform could exist either in small mammals or in humans. It is very unlikely, therefore, that a still-unknown MHC isoform could comigrate with MHC2A in SDS-PAGE gels, be loaded in IVMA, and affect Vf.
The most likely hypothesis to explain the change in Vf after training is a posttranslational modification of myosin. Phosphorylation of regulatory MLCs has been shown to affect force at submaximal levels of activation (35) but not velocity. On the contrary, deamination (5) and glycation (6, 33) have been suggested to be able to modulate velocity. However, notwithstanding some recent work (34), the role of such posttranslational modifications of myosin in modulating the function of the molecule is still unsettled.
The modulatory role of exercise on myosin function confirms the hypothesis that, besides the well-known mechanisms of muscle plasticity, based on regulation of MHC isoform expression, additional mechanisms exist (8). The mechanism based on MHC isoforms is well known to determine large differences in contractile and energetic properties of muscle fibers, i.e., type 1 fibers from human muscles have 10-fold lower Vo and 3- to 4-fold lower ATPase activity than type 2X fibers (10). The change in velocity of a given myosin isoform observed in this study is more subtle and appears as a mechanism able to determine a finer regulation of the function of the molecule than a shift in its isoform.
This work has been supported by European Union contract Pan European Network for Ageing Muscle (QLK6-CT-2000-00417) to S. Harridge and R. Bottinelli and by Cofinanziamento Progetti di Rilevante Interesse Nazionale 2002, Ministero dell’Istruzione, dell’Università e della Ricerca, Italy, to R. Bottinelli and The Wellcome Trust, UK to S. Harridge.
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 © 2005 the American Physiological Society