KEY WORDS: Anaerobic muscular power, contractile proteins, myosin heavy chain isoforms, myosin light chain isoforms, training, transformation.

Myosin, the main contractile protein of muscle contraction, is composed of heavy chain (MHC) and light chain (MLC) isoforms. In adult human skeletal muscle MHC is expressed with slow (MHC I) and fast isoforms (MHC IIa and MHC IIx). Each MHC isoform has its distinct ATPase activity and properties of shortening velocity, wherefore the MHC profile is considered a feasible marker of fiber type diversity (Bottinelli, 2001; Pette and Staron, 2001).

Five distinct MLC isoforms are expressed in adult human skeletal muscle. The essential MLC isoforms include two fast (MLC1f and MLC3) and one slow (MLC1s) isoforms. The content of MLC3 isoform has shown to correlate with the shortening velocity of muscle fibers (Larsson and Moss, 1993). The regulatory part of MLC includes both slow and fast MLC isoforms (MLC2s, MLC2f). The expression of MLC2 isoforms has also been shown to be associated with the speed and velocity regulation of muscle contraction (Lowey et al., 1993).

Exercise training causes adaptive changes in myosin profile. In order to create significant changes on protein level, quite a long training period is necessary. A fast-to-slow transition in MHC isoform has been reported in previous studies, caused by specific types of resistance or weight training lasting 8-24 weeks (Adams et al., 1993; Campos et al., 2002; Harber et al., 2004; Sharman et al., 2001). A decrease of the proportion of MHC I isoform has been shown to occur after sprint or strength training (Andersen et al., 1994; Liu et al., 2003). We have not found any follow-up studies concerning MHC adaptation to long-term (> 1 year) exercise training in healthy people.

MLC adaptation to exercise training has been reported in several studies on laboratory animals (Ingalls et al., 1996; Wada et al., 2003; Wahrmann et al., 2001). Few data exist about MLC adaptation to exercise training in human skeletal muscles. Trappe and co- workers (2000, 2001) examined MLC profile during 12 weeks of resistance training in elderly subjects but no significant changes in MLC profile were found.

The purpose of the present study was to examine the adaptive changes of MHC and MLC isoforms in human vastus lateralis muscle in response to long-term strength and power training in previously untrained middle-aged men. Furthermore, we examined whether the changes in MHC and MLC isoforms would correlate with the changes in maximal anaerobic cycling power, vertical jumping height, and the total number of training sessions.

Subjects
Seventy-five middle-aged healthy voluntary male subjects entered this study. Before the intervention, they participated in the health examinations and measurements included in the study protocol. Of the subjects, 56 volunteered for long-term exercise training and 19 for acting as a control subject. A total of 45 men in the training group and 18 in the control group completed the program; however, only 23 men in the training group trained to a sufficient extent (i.e. completed > 2/3 of the program). Of them the final training group consisted of the 16 men whose muscle biopsies were obtained both before and after the exercise intervention and whose tissue samples were large enough for the MHC and MLC analyses. Their age, height and weight were 43.8 ± 5.8 yrs, 1.79 ± 0.05 m, and 83.1 ± 12.8 kg (before) and 83.6 ± 13.3 kg (after), respectively.

Correspondingly, muscle biopsies were obtained from 6 persons belonging to the control group both before and after the 54-week intervention. Their age, height and weight were 40.0 ± 8.5 yrs, 1.77 ± 0.04 m, and 69.7 ± 9.7 kg (before) and 72.4 ± 9. 2 kg (after), respectively. The characteristics of the both subject groups were quite similar with the voluntary subjects entered originally into the study (n = 75; their age, height and weight were 42.4 ± 6.7 yrs, 1.80 ± 0.06 m, and 80.4 ± 13.1 kg, respectively). All the participants were either previously untrained or engaged in recreational sports that some of them continued simultaneously with the study program.

All the subjects gave their written informed consent to participate in this study that was approved by the Ethical Committee of the Research and Development Centre of the Social Insurance Institution of Finland.

Training protocol
The subjects were advised to exercise 3 times a week according to a special strength and power training program for a period of 54 weeks (Appendix). Simultaneously, they continued 0-2 times a week their previous recreational sports activities (including walking, jogging, cycling, skiing, swimming, and ball games, such as volleyball, badminton, tennis, or soccer). The training group carried out 93.5 ± 12.5% (range 75-118%) of the planned sessions for special training in the gym and 134.6 ± 35.2% (range 76-210%) of the recommended amount of training (3 times a week) including both all type of exercise sessions in the planned program (see Appendix) and the subjects' spontaneous recreational sports activities. The strength and power training program was individually adjusted to the muscular performance capacity of each subject.

The supervised program started with a lead-in strength-training period of 6 weeks, followed by a basic strength-training period of 4 weeks. The next phase consisted of progressive strength and power training [at 60-75% and 30-85% of 1-RM (repetition maximum), respectively] with stretching and elasticity exercises. For the first four weeks of this phase, the subjects exercised twice a week as consecutive training, together with power training sessions once a week as circuit training, after which they had one consecutive and two circuit training sessions a week during the next four weeks. Thereafter, the exercises were changed and the number of repetitions (varying from 4 to 12 rep.), and later on, also the number of sets, was increased (Appendix). At the beginning of the primary strength and power training phase, three circuits or sets with a pause of 2-3 min in between were performed; at the end of the training period the number of circuits was five. During the pauses the subjects performed warm-up (for muscles to be exercised next) and recovery stretching exercises (for muscles just used).

Power-type training and basic strength training were emphasized by turns, and the exercises focused mainly on the legs. The training group performed 7285 ± 1099 (range 5378-9868) femoral muscle exercises during the whole training period. In addition, various trunk and upper body exercises were performed. Special attention was paid to the velocity of muscle contraction in the exercises performed with small loads (power training) in order to induce training effects on the force and velocity characteristics of leg muscles, and especially on the fast-twitch fibers. After three months there was a two-week 'recovery phase' during which the subjects only carried out recreational outdoor sports 1-2 times per week. During the next 12 weeks (in the spring) the subjects continued the strength and power training program. In the summer time, the program included interval running, plyometric drills, shot put, discus, javelin, and circuit training.

Before the training intervention started, seven muscle groups were tested by using 30-second repetition tests. Individual training programs with 10 exercises were drawn up on the special training cards for each man in the training group. Later on, 10-second repetition tests or 1-RM were applied to determine the proper progression for training (additional weight and/or number of repetitions). New, individual training programs were given for each phase: the exercises were changed and loadings increased according to the program model.

Earlier recreational sports activity
Leisure-time physical activities after the school years were asked for and graded into five grades as follows: 0 = having no leisure-time physical activity during adult age, 1 = having 1-3 types of sports and being active < 5 years, 2 = having 1-3 types of sports and being active 5-10 years, 3 = having 1-3 types of sports and being active >10 years or having several types of sports and being active 5-10 years, 4 = having several types of sports and being active >10 years.

Maximal anaerobic cycling power
The maximal anaerobic power of leg muscles was measured using a bicycle ergometry modification of maximal anaerobic power test on a treadmill (Rusko et al., 1993). In this intermittently progressive test, the loading consisted of 5-10 successive 20-s cycling periods with a pause of 100 s between the periods. The test was continued until the subject could not keep the correct predetermined pace, which was 90 revolutions/min for men younger than 40 years and 86 revolutions/min for men older than 40 years. The increment in work rate for each consecutive cycling period was adjusted to increase the oxygen demand by 5-6 ml·kg^-1·min^-1.

Vertical jump
The explosive force of leg muscles was measured as vertical jumping height. The jump was performed from a static leg position with the knee angle of 100° and without a countermovement. The hands were held on the hips. A contact mat and reaction time equipment (Newtest powertimer^®, Finland) were used to measure the off-the-ground time from which the jumping height was calculated (Bosco et al., 1983).

Muscle biopsies
Muscle biopsies of 50-100 mg were taken under local anesthesia from the left vastus lateralis muscle, using the percutaneous conchotome technique, at the beginning and at the end of the training period. The muscle sample was frozen in liquid nitrogen and stored at -70°C for further analysis of heavy and light chain myosin isoforms.

MHC and MLC determination
Muscle samples were pulverized in liquid nitrogen and homogenized in phosphate buffer according to Sugiura and Murakami, 1990. The protein concentration of the homogenates was determined by using the Bio-Rad Protein Assay Kit (Bio-Rad Laboratories, USA). Aliquots of protein were loaded on the gel after incubation of 10 min at 65°C in lysis buffer, containing 10% (v/v) glycerol, 5% (v/v) 2-mercaptoethanol, 2.3% SDS, 0.05% bromphenole blue, and 62.5 mM TRIS-HCl pH 6.8.

The MHC isoforms were separated by 5-8% gradient SDS-PAGE gel using 0.75 mm thick gradient separating gel and 3.5% stacking gel (Bär and Pette, 1988). Electrophoresis was performed using vertical slab gel system (Protean II Xi Bio-Rad). 1 g myofibrillar protein sample was loaded on gel. Electrophoresis took 24 h at 120V. The gels were silver stained by using the Bio-Rad Silver-Stain Plus Kit according to the manufacturer's instructions.

The MLC isoforms were separated by 12.5% one-dimensional SDS PAGE gel system according to Laemmli, 1970, except that the glycerol content in the separating gel was 10%. 10 μg myofibrillar protein sample was loaded on 1mm thick gel per well. Electrophoresis was carried out at a constant current (30mA) using mini-Protean II Bio-Rad Electrophoresis Cell. The gels were Coomassie blue R-250 stained. The positions of MLC isoforms on the gel were identified by their apparent molecular masses, in comparison with protein mobility of the kaleidoscope pre-stained standard marker proteins (Bio-Rad) and by reports in the literature. Staining reactions of MHC and MLC were quantified by a computer- based image analysis system and software (Image Master 1D, Amersham Pharmacia Biotech).

Statistical analysis
All results are presented as means ± SD. Differences between the variable means of the training and control groups were assessed using Kruskal- Wallis test, and changes within groups using Wilcoxon's matched-pairs signed-ranks test. Pearson's correlation coefficients or Spearman's rank correlation coefficients were calculated to evaluate the associations between the variables and their changes. If necessary for Pearson's correlation coefficients, logarithmic transformations were applied to correct skewed data distributions. Differences were considered significant at p < 0.05. Logical correlation coefficients are presented for evaluating reliability and accuracy of electrophoretic MLC analyses.

The mean values of maximal anaerobic cycling power and maximal jumping height at baseline and after the intervention are presented in Table 1. On the average, the control group was 3.8 years younger (ns) and weighed 13.4 kg less (p < 0.05) than the training group. At baseline, the groups also differed for the myosin heavy chain distribution: the proportion of MHC IIa was 17. 6 %-units higher and MHC IIx was 11.3 %-units lower in the control group (p < 0.05 for both; Table 2). During the training period, maximal anaerobic cycling power increased by 64 W (p < 0.001) and the maximal jumping height by 1.5 cm (p < 0.05) in the training group, but no significant changes were found in the control group (Table 1). However, the group by time effect was not significant. In the training group, the increase of the maximal jumping height correlated with the number of strength and power training sessions (r = 0.56; p < 0.05).

After 54 weeks of strength and power training the proportion of MHC I isoform remained the same (Table 2). The change of the proportion of MHC IIa isoform from 52.6 ± 12.2% to 59.4 ± 11.6% did not reach statistical significance (p = 0.070 for group by time; within training group p = 0.061) and neither did the change of the proportion of MHC IIx isoform from 18.1 ± 11.4% to 11.1 ± 9.1% (p = 0.104 for group by time; within training group p = 0.032) (Table 2). The degree of change of MHC IIx isoform correlated with the earlier recreational sports activity during adult age (r = 0.61; p < 0.05). No changes in MHC isoforms were found in the control group.

During the training period, the proportion of MLC1f changed from 23.9 ± 3.8% to 25.5 ± 2.0% and the proportion of MLC2f changed from 17. 8 ± 5.0% to 18.8 ± 5.4% in the training group. None of the changes in MLC isoforms reached statistical significance (Table 3). In the training group, the changes of MLC1f isoform correlated negatively with those of MLC1s isoform (r = -0.79; p < 0.05), and the changes of MLC1s isoform correlated positively with changes of MHC I isoform (r = 0.81; p < 0. 05) and negatively with changes in maximal anaerobic cycling power (r = -0.81; p < 0.05).

Long-term strength and power training improved the maximal anaerobic cycling power in all subjects in the training group, and the mean maximal jumping height in the training group. These changes were in line with a tendency of increased proportion of MHC IIa and a decreased proportion of MHC IIx isoforms. Similar adaptive responses of MHC IIx to regular exercise training have been shown in previous studies (e.g. , Campos et al., 2002; Sharman et al., 2001). Both endurance and strength and power types of physical activity decrease the proportion of the fastest human MHC isoform (Andersen et al., 1994; Carroll et al., 1998; Klitgaard et al., 1990; Staron et al., 1994; Tajsharghi et al., 2004). We found no studies in previous literature where exercise training would have caused an increase in the proportion of MHC IIx isoform. The proportion of MHC IIx isoform only increases after declined neuromuscular activity, such as detraining (Andersen and Aagaard, 2000), or during inactivity, such as immobilization (Talmadge et al., 2002) or long-term bed rest (Andersen et al., 1999; Hostler et al., 2001; Trappe et al., 2004).

In the present study, the training lasted for 54 weeks and decreased the proportion of MHC IIx by 7.0 %-units. Similar decreases of MHC IIx have been reported in other studies with shorter exercise interventions (Campos et al., 2002; Putman et al., 2004; Sharman et al., 2001). In the study with young adults, only six weeks of exercise training (combined strength and endurance training) resulted in a significant decrease in MHC IIx content (Putman et al., 2004).

The increase of MHC IIa and decrease of MHC IIx isoforms have been suggested to reflect a transition from the fast to the slower type of MHC; so also in the present study. A similar MHC transition has been found after heavy resistance training in young (Adams et al., 1993; Campos et al., 2002; Liu et al., 2003) and old subjects (Sharman et al., 2001). Alternative, bi-directional MHC transitions from MHC I and MHC IIx to MHC IIa have previously been demonstrated after sprint training (Andersen et al., 1994). In the present study, the mean percentage of MHC I remained the same during the training period, which was comparable with the results of previous studies with resistance or strength training interventions (Campos et al., 2002; Hostler et al., 2001).

MHC profile analysis and histochemical fiber type analysis are not fully comparable. It has been shown that MHC profile responds faster to exercise stimulus than mATPase fiber profile does (Staron et al., 1994) and that, for example, part of histochemically assessed IIB fibers contain MHC IIa isoforms (Sant`ana Pereira et al., 1995). Thus, the MHC profile analyses are more sensitive measures to show changes in functional properties of muscle fibers as a result of exercise training than the mATPase fiber profile analyses do.

In the present study, strength and power training lasting 54 weeks tended to increase the proportion of MLC1f in the vastus lateralis muscle in the training group. In addition, the negative correlation between the changes of MLC1f and MLC1s proportions (r = -0.79; p < 0.05) shows that these alterations were in the same direction in most of the subjects (6/8). Trappe and co-workers (2000, 2001) studied the effects of a 12-week resistance training intervention on MLC isoform proportions in subjects aged approximately 74 years and found no significant changes in single muscle fibers of m. vastus lateralis. Their exercise program, however, differed from our program: the exercises were performed with heavy loads (80% of 1 RM) and with slow speed which resulted in increased cell size, strength and contractile velocity in both slow- and fast MHC muscle fibers ('more pronounced in MHC I fibers') in men (Trappe et al., 2000) and only in MHC I muscle fibers in women (Trappe et al., 2001). In contrast, we used varying (30-85% of 1 RM) but mainly lower loads and high exercise tempo or high contractile velocity (power-type resistance training). Furthermore, the subjects in the study by Trappe et al., 2001 were older, and the training period was only a quarter of that in our study. Given that the duration of the training period in our study was much longer, that the subjects had rather high proportion of fast MHC muscle fibers in their vastus lateralis muscles, and that the training focused on the fast muscle fibers, the tentative changes in MLC isoforms can be assumed to be real. The small number of subjects and the fact that MHC and MLC fiber proportions were assessed using the muscle tissue homogenate instead of the single fiber method could be reasons for insufficient power of our study. Although the changes in the MLC isoforms during the long-term strength and power training were not statistically significant, they correlated with the changes in MHC profiles. The changes in MLC isoforms may possibly be associated with the transition of MHC isoforms as a result of the strength and power training and thus indicate improved muscle contraction. This, however, remains to be investigated in the future.

In previous studies, the shortest exercise period shown to affect the MHC isoform profile (results from the triceps brachii muscle) has been six weeks of strength training with maximum contractions and training 3 times a week (+17. 3% in MHC IIa and -13.9% in MHC IIx isoforms) and with a combination of three types of training (once a week per type): strength training with maximum contractions, ballistic exercises, and stretch-shortening movements (+15% in MHC IIa and -9% in MHC I isoforms) (Liu et al., 2003). Training effects are perhaps detected earlier in the upper arm muscles than in the postural muscles, such as the vastus lateralis muscle.

Recently, Kyröläinen and co-workers (2005) demonstrated that, during a 15-week power training period, drop jump increased in young, recreationally active men but there was no significant change in the MHC isoforms and muscle fiber proportions. This may be due to the high initial training status and the fact that the subjects continued their previous endurance-type sport activities (cycling, walking and ball games) for 6 hours a week on the average. In the present study, the strength and power training lasted for 54 weeks, but we only observed an increase of 6.8 %-units in MHC IIa and a decrease of 7.0 %-units in MHC IIx isoforms. In both studies, the subjects exercised 3 times a week, as targeted in our study. Strength training with maximum contraction causes a stronger stress for skeletal muscles, leading to muscle hypertrophy, than the varying exercises with varying relative loads carried out in our training program. The training focused more on developing the speed of muscle contraction than on increasing the maximal strength of muscles. It is also possible that the adaptive changes in skeletal muscle structure are not as extensive in middle-aged participants as in young participants, e.g., as a result from a smaller margin for MHC IIx to change due to lower baseline proportions (Short et al., 2005) and perhaps also a lower synthesis rate of MHC in the older people (Hasten et al., 2000). An additional reason for low training effects can be that that the power type strength training was not carried out equally successfully by all the subjects. This suggestion was supported by the correlation coefficient of r = 0.61 between the amount of earlier recreational sports activity and the reduction of MHC IIx isoform reflecting the idea that the experienced subjects were more skilled and could perform exercises more effectively. However, in the both studies jumping performance increased significantly, perhaps due to neural adaptations (Häkkinen et al., 1985; Häkkinen and Häkkinen, 1995; Moritani and De Vries, 1980), but changes in MHC profiles did not reach statistical significance. That might be interpreted so that the type and/or amount of training was not proper for subjects and training included to many varying elements influencing on MHC profiles with opposite manner.

The present study raised a number of interesting questions that still remain open. In the future, studies should be designed with larger populations and with more homogeneous study groups. Furthermore, to get more reliable knowledge about training-induced adaptations in muscle tissue, to make correct interpretations, and to draw appropriate conclusions from the results, it is important to estimate the protein intake during the intervention. Similarly, it is critical to evaluate the effects of the current training status, the exercise background and the age of the study group on the variables studied already before fixing the study design. Otherwise, it is very difficult to confirm any changes reliably. In addition, it is essential to have knowledge of the extent of simultaneous physical activities of the subjects so as to ensure a positive nitrogen balance and sufficient recovery time for the muscles to develop their structures and energy capacity during the intervention period.

The control group was small and heterogeneous for age, maximal anaerobic cycling power, and proportion of MHC IIx. Unfortunately, only few of the control subjects were willing to give samples of their muscle tissue. These matters and the size difference between the training and control groups reduce the reliability of comparisons between the groups to some extent.

• A long-term strength and power training program seemed to decrease the proportion of MHC IIx isoform in previously untrained middle-aged men.
• The degree of change of MHC IIx isoform correlated with the amount of earlier recreational sports activity.
• The changes of MLC isoforms were associated with the transition of MHC isoforms. Whether this means improved speed and coordination of muscle contraction remains to be investigated in the future.