INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

ONE OF THE MOST CONSPICUOUS adaptations accompanying high-resistance training (HRT) is an increase in the cross-sectional area of the muscle cell (13). The cellular hypertrophy appears to extend to both major fiber types (type I and type II) and subtypes (IIA and IIB) (8, 21, 32, 33), although the magnitude of the increase appears to be fiber type specific (8, 14, 33). The increase in fiber size results in an increase in muscle force-generating potential (13).

Somewhat unclear is the role of HRT in muscle cell capillarization and oxidative potential, two properties that appear to be intimately related in promoting increased oxidative phosphorylation during exercise (11, 27). Earlier studies based on comparisons between weight lifters and age-matched, untrained controls have indicated that fiber hypertrophy occurs in the absence of parallel increases in oxidative potential (2, 18, 36) and capillarization (37). However, other studies (28, 30) have challenged these conclusions. The limited number of longitudinal studies, which have employed HRT for up to 20 wk, are inconclusive. Some studies have reported no changes in capillary density, expressed as either capillaries per fiber or per unit fiber area (16, 35, 38), whereas others (8, 21) have found increases in the number of capillaries per fiber, both type I and type II, without increases in capillaries per unit fiber area. In general, changes in mitochondrial potential have not been reported with HRT. Tesch et al. (35) and Wang et al. (38), using the maximal activity of citrate synthase in homogenates as a measure of oxidative potential, have found no change in either women or men after HRT. Such findings have also been confirmed at the single-fiber level, regardless of fiber type (24). These findings suggest that the absolute number and size of mitochondria increase on an absolute basis to maintain a constant relative potential. This hypothesis has been confirmed by Wang et al. (38) by using ultrastructural measurements. However, interpretation of the results of most studies is confusing because hypertrophy was not demonstrated (16, 24, 35).

A number of stimuli have been identified in promoting angiogenesis, and ischemia and hypoxemia have been recognized as most potent (6, 15). In addition, mechanical factors such as increased wall tension, increased sheer stress, and increased intravascular pressure appear to be important (12). What factors are involved in inducing capillarization with regular exercise is uncertain. It is known that prolonged, submaximal exercise training represents a major stimulus for angiogenesis and mitochondrial biogenesis (12), suggesting that sustained elevation of one or more of the stimulating factors may be critical. It has been hypothesized that the sustained elevation in blood flow which occurs during prolonged exercise and which is designed to meet the nutrient demands of a persistently contracting muscle may be of importance in inducing changes in stimuli necessary for capillary growth (12). If such is the case, few repetitions of heavy resistance exercise, typically employed for HRT, should have minimal effect on angiogenesis.

In this study, we hypothesized that the HRT program would lead to an increase in cellular cross-sectional area in the absence of an increase in capillary number. As a consequence, the capillaries subserving a defined cellular area would decrease. Moreover, the decrease in perfusion potential would be accompanied by a decrease in cellular oxidative potential. All of these changes would be manifested in the major fiber types and subtypes.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Subjects. Six healthy but untrained male volunteers (age 19.2 ± 0.48 yr; wt 81.0 ± 6.0 kg; means ± SE) completed all phases of the study. A substantially greater number (n = 11) were recruited, but with attrition and limitations in tissue availability a complete data set for investigating histochemical features was restricted to six subjects. Based on the information provided by questionnaire, none of the volunteers was "trained," i.e., none was involved in exercise on a regular basis or had been for at least 3 mo before the beginning of the study. Peak aerobic power, determined during a progressive cycle test to fatigue, was 45.6 ± 2.3 ml · kg¹ · min¹. As required, all experimental procedures and risks were described to each volunteer, and the study was approved by the Office of Human Research.

Experimental design. The training model used in this study consisted of performing three different exercises, specifically chosen for their emphasis on the quadriceps muscle. The exercises (parallel squats, incline leg presses, and leg extensions) were performed 3 times/wk with supervision. At least one day of rest was provided between each work session. Each training session included three sets of each exercise with each set performed for 6-8 repetitions maximum (RM). An approximate 2-min recovery period was provided between each set. For safety, a warm-up set of 10 repetitions at 50% of each subject's 6-8 RM was included for each exercise. The training extended for 12 wk with training intensity adjusted after 4 and 7 wk. On average, 10.1, 5.6, and 15.4 training sessions were performed during each training segment. During the first training segment, the weights (kg) used were 90.5 ± 9.1, 165 ± 10, and 40.9 ± 2.7 for the squats, leg presses, and leg extensions, respectively. For the final training segment, the loads (kg) were increased to 146 ± 12, 285 ± 23, and 63.6 ± 5, respectively, for each of the exercises. Before the training and at 4, 7, and 12 wk, samples of tissue were obtained from the vastus lateralis muscle using the technique of Bergström (3). The tissue obtained was oriented under a dissecting microscope, mounted, rapidly frozen in isopentane cooled to liquid nitrogen temperatures, and stored at 80°C until analysis. Two tissue samples were obtained from each leg during the study with the order randomized between legs after each training segment. All tissue samples were obtained at least 24 h after the last training session.

Analytical procedures. Histochemical properties were determined on cross sections of tissue cut in a cryostat maintained at 20°C. Muscle fiber-type identification and typing was based on the procedure of Brooke and Kaiser (5) as modified by Staron et al. (29). This modification allows identification of type I fibers as well as several type II subtypes: IIA, IIB, IIAB, and IIC. On average, ~200 fibers from each tissue sample were used to determine fiber-type distribution. Staining for fiber capillarization was performed on 8-µm-thick cross sections using a lectin (Ulex europeaus I ) technique described by Parsons et al. (22). In addition, the oxidative potential of each fiber was determined on cross sections (10 µm) stained for succinic dehydrogenase (SDH) activity. Determinations of SDH activity were based on the procedure of Pette (23) using a reaction medium containing 100 mM sodium phosphate buffer (pH 7.6), 1.0 mM sodium azide, 0.2 mM 1-methoxy-5-methylphenazinium methyl sulfate, 50 mM succinate, and 1.5 mM nitro blue tetrazolium (NBT). Nonspecific reduction of NBT was determined by incubating several sections in the above medium with succinate. The reaction was allowed to run for 10 min at 25°C. SDH measurements were based on a single end point (7). For quantification, the optical density (OD) measured from the blank, representing the mounting medium, glass slide, and coverslip just outside the section being measured, was subtracted from the OD measure in the fiber. In addition, we also subtracted the change in OD due to the nonspecific reduction of the reaction indicator NBT (4). The value for this nonspecific reduction was obtained on separate pieces of tissue that had been stained under conditions as nearly identical as possible. The OD measured on 25 fibers of each type, where possible, was obtained using an image analysis system (Image-Pro Plus; Media Cybernetics, Silver Spring, MD). Fiber-type differences in SDH activity, obtained before training, were very similar to those obtained by others using a similar procedure and similar sex (24). Fiber type-specific areas were obtained from the SDH stain using the same fibers selected for the OD measurements. The image analysis system, equipped with a video monitor, digitizing tablet, and tracing cursor, was used for area measurements. Serial sections permitted the identification of fiber area and fiber capillarization in the same fiber, which could then be identified as to type.

Capillarization was determined using two approaches, i.e., at the level of the individual fiber and in a defined field of a fixed, cross-sectional area. At the level of the individual fiber, the capillarization indexes included the number of capillaries in contact with each fiber type (I, IIA, IIAB, and IIB) or capillary contacts (CC), and the number of capillary contacts per fiber-type area. Capillary contacts per fiber were determined by counting the number of capillaries around each individual fiber, subdivided by type, and then computing the mean (9, 21, 25). The number of capillaries per fiber area was estimated by dividing the number of capillaries in contact with each fiber by the area of the fiber to which they were adjacent and computing the mean number of capillaries per fiber area for each fiber type (21, 25).

The indexes that were calculated from a defined cross-sectional area of the tissue sample were the number of fibers or fiber density (FD), the total number of capillaries or capillary density (CD), the number of capillaries around the fiber (CAF), and the sharing factor (SF). Both FD and CD were obtained by simply counting the total number of fibers and the total number of capillaries, respectively, in the set cross-sectional area (21, 25). The CAF index was obtained as the ratio CD/FD (9, 21, 25). The determination of SF (fibers/capillary) was based on the quotient obtained by dividing the number of capillaries around the fiber calculated from the defined field into the average number of capillary contacts (CC/CAF) (9, 25). For all of these measurements, discrete regions with clear cross-sections free of artifacts were used. The number of fibers and capillaries on the borders were counted and then halved.

Statistical procedures. To determine the effects of training time (0, 4, 7, 12 wk) and fiber-type (type I, IIA, IIB, IIAB) distribution, area, and capillarization, a two-way ANOVA for repeated measures was employed. Where only a single measure was obtained, for example with capillary density, without regard to fiber-type differences, a one-way ANOVA was used. Where significance was indicated, post hoc analysis with the use of the Newman-Keuls technique was performed to compare specific means. A 95% level of confidence was accepted for all comparisons.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Muscle fiber-type distribution (%) for the vastus lateralis muscle before training was 47.2 ± 3.1, 28.4 ± 4.8, 7.7 ± 1.3, and 15.1 ± 2.1 for type I, IIA, IIAB, and IIB, respectively (Table 1; see Fig. 3). With the exception of the type IIB fibers, training was without effect in altering the percent composition. For the type IIB fibers, a pronounced reduction ranging from 34 to 45% was observed. The reduction was fully manifested by 4 wk of training. Although type IC and IIC fibers were identified, these represented <1% of the fiber pool and consequently were not included in the analyses.

HRT increased fiber area (Table 2). This effect, which was not specific to fiber type, was observed by 7 wk of training and persisted through the final 5 wk of training. After subjects were trained for 12 wk, fiber areas, when averaged over all fiber types, were larger than at 4 wk of training. When comparisons were made between the fiber types, type IIA fibers were observed to possess the largest area while type I fibers were the smallest. No difference was observed between the type IIAB and type IIB fibers in area. Training did not alter the differences between the fiber types in area.

Training also had a significant effect in increasing CC (Table 3; see Fig. 3). This effect, which was not specific to fiber type, was not observed until the 12th week of training. In general, type I and type IIA fibers possessed the largest number of capillary contacts per fiber, followed by the type IIAB fibers, which were intermediate, and the type IIB fibers, which were the lowest. When the changes in fiber-type area were considered, and the number of capillary contacts were expressed relative to fiber area, no effect of training was observed (Fig. 1). Type I fibers possessed the highest capillary contacts-to-fiber area ratio compared with the other types. No differences in this ratio were found between the type IIA, type IIAB, and type IIB fibers.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Capillary contacts-to-fiber area ratios in untrained males during high-resistance training (HRT). Values are means ± SE; n = 6. Only a main effect (P < 0.05) of fiber type was found: type I > type IIA = type IIAB > type IIB.

Additional indexes of fiber capillarization based on a defined field are presented in Table 4. Only one index, namely CAF, was altered by training. In the case of CAF, a 33% increase was observed, but not until the 12th week of training.

SDH activity, determined by microphotometry and expressed in absolute units, was observed to be highest in type I fibers, followed by type IIA and type IIB fibers, and lowest in type IIB fibers (Figs. 2 and 3). The oxidative potential was not affected by HRT, regardless of fiber type and training duration.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Fiber-type oxidative succinic dehydrogenase (SDH) activity in untrained males during HRT. Values are means ± SE; n = 6. Only a main effect (P < 0.05) of fiber type was found: type I > type IIA = type IIAB > type IIB. OD, optical density.

View larger version (118K): [in this window] [in a new window]   Fig. 3.   Cross sections of muscle samples extracted from vastus lateralis muscle at 0, 4, 8, and 12 wk of training. Serial sections stained for myofibrillar ATPase (A), SDH (B), and capillarity (C). Muscle tissue obtained from a single subject. See METHODS for further analytic details.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

We have been able to confirm, as hypothesized, that HRT would result in an increase in fiber cross-sectional area. This finding is not surprising given the many previous studies (13) that have found fiber hypertrophy with HRT. Our findings on the time course response are also consistent with earlier work (31), namely that increases in fiber area are a delayed response to training, only observable after several weeks of training. In this study, significant increases were not observed until the seventh week of training. Our HRT program failed to produce a differential increase in area between the different fiber types. By 12 wk of training, increases in area between fiber types ranged between 14 and 24%. Although similar findings have been reported previously (10, 14, 21), exceptions have been noted (8, 33). Several studies have reported a differential hypertrophy between fiber types with the increase in size more pronounced in the type II fiber subtypes (8, 10, 14, 33). The differences between the fiber-specific response between studies may well reflect the specifics of the different HRT programs (8, 14), including the duration of the training. Indeed, the relatively brief nature of previous training programs may be an important factor in the inability to observe hypertrophy (16, 24). It must be emphasized that the objective of our study was not to examine the efficacy of different HRT programs in inducing fiber hypertrophy. Rather, we were simply interested in selecting a program that would result in hypertrophy based on previous published reports. This was necessary so that we could examine the relationship between hypertrophy, capillarization, and oxidative potential.

HRT clearly resulted in an increase in the number of capillaries in contact with each fiber. However, this response was more delayed than the increases in fiber area and not observed until the 12th week of training. Similar to the fiber size changes, the magnitude of the effect was not differentiated by fiber type. As with HRT effects on fiber area, the delayed increase in capillarization may well explain the failure of a previous study (16) to observe changes because this program was shorter in duration than the present program, and of others (8, 21) in which increases in angiogenesis have been observed. Even with extended training, increases in capillary number are not always a consistent finding (38).

Despite the different time course that was observed between the changes in fiber size and the changes in fiber capillarization, capillarization as expressed by capillary contacts-to-fiber area ratio was not altered with HRT regardless of the period of training. Our findings demonstrate that modest increases in fiber area can occur in HRT without compromising the fiber area served by a capillary regardless of fiber type. Although previous longitudinal studies have reported similar findings (8, 21, 38), there is also evidence that HRT can compromise the capillary-to-fiber area ratio (8). These discrepancies may be explained by the type of HRT program. Training using concentric contractions only, compared with training using concentric and eccentric contractions similar to HRT with free weights, does not increase capillary-to-fiber area ratios. Capillarization was also examined from another perspective, namely by examining capillary density in a defined cross-sectional area of tissue. This procedure conveys the advantage of examining the changes from the perspective of a defined field without regard to fiber-type distribution. By using this procedure, neither the total number of capillaries nor the number of fibers was significantly altered. However, trends were clearly evident toward a decrease in the number of fibers and an increase in the number of capillaries, as might be expected. It is probable that the magnitude of the hypertrophy was insufficient to reduce the number of fibers occupying a given area. A more sensitive index, namely CAF, was increased by 12 wk of training. The change in this index was not accompanied by significant changes in SF, indicating that, on average, HRT was without effect in changing the number of fibers sharing one capillary. An increase in this index could conceivably compromise nutrient supply to specific fibers.

It should be emphasized that the measures of capillarization examined are only indexes that provide for the description of muscle capillaries relative to fiber areas and types. These properties may not provide the best estimate of the potential for blood-to-tissue exchange in skeletal muscle since they do not account for the configuration of the capillary network (20). A new property, described as the capillary-to-fiber perimeter ratio, has been proposed that incorporates the effects of capillary tortuosity and branching and may provide a better functional measure (9, 20).

Oxidative potential, as measured by SDH, remained unchanged throughout the training regardless of the fiber type. Only one previous study (24) has examined oxidative potential in the various fiber types in response to HRT and arrived at the same conclusions. However, in the study by Ploutz et al. (24), HRT failed to result in fiber hypertrophy. These results are also supported by others using determinations of maximal enzyme activities of representative mitochondrial enzymes in homogenates prepared from whole muscle samples (35, 38). Ultrastructural studies have provided inconsistent results with either no change (16, 38) or a decrease (17, 19) in the percentage of fiber volume occupied by mitochondria reported. At present, the reasons for the differences between the ultrastructural studies are not clear. However, the muscle group selected for training may be involved. The work performed by MacDougall et al. (17, 19) was based on the triceps brachii, whereas others have examined the quadriceps (16, 38).

As with other studies (8, 24, 31, 33, 38), we have found reductions in type IIB fiber percentage with HRT. This effect was observed very early in training and before changes in fiber area and capillarization. It is generally believed that the type IIB fibers, which are composed of myosin heavy chain IIb (MHCIIb) (34) undergo a sequential transformation in expression to type IIAB (composed of MHCIIa and MHCIIb), then type IIA (composed of MHCIIa), and ultimately ending up as type I fibers expressing MHCI. In this study, the increases in type IIA, type IIAB, and type I fiber percentages were not significant. The failure of HRT to elicit an increase in type I fiber percentage is a consistent finding (8, 14, 24, 31, 33). Although increases in type IIA fibers have been reported (1, 8, 33) such is not always the case (31), conceivably because the relatively low percentage of type IIB are transformed between type IIA and type IIB.

Of particular interest is the implication of the transformation of the type IIB fibers to the other fast subtypes regarding the changes in other properties. The transformation in itself, given the smaller fiber area, lower capillary counts, and oxidative potential, could have the effect of altering these properties in the type IIA and type IIAB fibers independent of actual changes in these properties. The magnitude of this effect remains to be determined.

In summary, using a 12-wk program of HRT, we demonstrated that modest fiber hypertrophy, typical of that observed with many HRT programs, can occur without compromising either the capillarization or the oxidative potential of the different fiber types. Accordingly, the hypothesis that we have proposed, namely that decreases in both of these indexes would occur with HRT-induced hypertrophy, must be rejected. The specific mechanisms eliciting the absolute increase in both angiogenesis and mitochondrial biogenesis with HRT remain to be determined but would appear to be associated with the elevated metabolic rate and blood flow, and the cellular hypoxemia that occur (6, 15, 20, 26). Our findings do not exclude the possibility that other manipulations of HRT may result in fiber hypertrophy accompanied by either increases or decreases in capillarization and oxidative potential. These studies remain to be done.