Muscle fiber activity of speed and gait

as a function

R. B. ARMSTRONG, P. MARUM, C. W. SAUBERT IV, H. J. SEEHERMAN, AND C. R. TAYLOR Departments of Biology and Health Sciences, Boston University, Boston 02215; and Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138

ARMSTRONG,

R. B., P. MARUM, C. W. SAUBERT IV, H. J. C. R. TAYLOR. Muscle fiber activity as a function of speed and gait. J. Appl. Physiol.: Respirat. Environ. Exercise Physiol. 43(4): 672-677, 1977. -These experiments were designed to test the hypothesis that locomotory gait transitions occur when some critical cross-sectional area of active muscle is reached as animals increase speed within a gait. We used glycogen depletion as evidence of muscle fiber activity and selected an experimental animal in which all muscle fibers in the locomotory muscles rely extensively on glycogen as a substrate -the lion. We found a high correlation between biochemically and histochemically determined rates of glycogen depletion (r = 0.906). Rates of glycogen depletion in the biceps femoris and triceps brachii muscles increased logarithmically with speed with no discontinuities at the gait transitions. However, we found large discontinuities both in the total cross-sectional area of muscle that showed depletion and in the rates at which the different types of fibers depleted glycogen at the trot-gallop transition. Our results indicate that 1) gait transition did not occur at a maximum tension level either for a particular type of fiber or the whole muscle, and 2) different configurations of motor units within an individual muscle may be recruited as an animal changes gait. SEEHERMAN,

AND

exercise; lion; glycogen comotion

depletion;

motor unit recruitment;

lo-

VARIETY OF LOCOMOTORY PATTERNS described for biped and quadruped birds and mammals can be divided into three basic gaits - the walk, the trot or run, and the gallop (found only in quadrupeds) -on the basis of measurements of the forces applied to the ground (4). Animals change from one gait to another as they increase speed in a manner somewhat analogous to shifting gears in an automobile. The change consistently occurs at about the same speed in an individual animal and the speed, the stride frequency, and the stride length at the trot-gallop transition vary in a regular manner with the animal’s size (12). It has therefore been suggested that the gait transition speeds represent equivalent speeds for different animals at which the force generated by active muscles exceeds the same critical point for optimal muscular efficiency. To move faster while maintaining this optimal efficiency the animal would change to a different gait, additional muscles would be recruited, and the force would be redistributed over a larger cross-sectional area of muscle (12, 15). The present experiments were designed to test this THE

hypothesis by determining both the cross-sectional area of muscles and the proportions of the fiber type populations within the muscles that are active as an animal increases speed within the gaits and changes from one gait to another. We have used glycogen depletion as evidence of muscle fiber activity. This approach has been used in a variety of animals including humans (6, 9, lo), rats (1, Z), bush babies (8), guinea pigs (7), and horses (13). Potential problems in using the technique (1, 3, 9) can be avoided by selecting an experimental animal in which all muscle fibers rely extensively on glycogen as a substrate during locomotion. Lions exceeded their maximum aerobic capacity at a slow trot (5) and must depend heavily on anaerobic glycolysis at all trotting and galloping speeds. For this reason we used lions as our experimental animals and focused on the changes in glycogen depletion patterns at the trot-gallop transition. MATERIALS

AND

METHODS

Animals Two littermate lions, one male and one female, were purchased at 4 mo of age. They were weaned at the time of purchase. The animals were immediately trained to run on a motor-driven treadmill at different speeds within the three gaits. These exercise sessionsoccurred 5 days each week and continued for 4 mo prior to the initiation of experiments. The lions were 8 mo old (the male and female weighed 33 and 26 kg, respectively) when the experiments commenced. They always wore a loose choke stick during both training and experimental sessions for protection of the investigators, but this did not impede their normal locomotory patterns. Also, the animals wore a face mask during exercise for Vo, determinations. The experiments were terminated when the lions were about 10 mo old because of the increased danger involved in handling them. At this time the male and female weighed 46 and 35 kg, respectively. The two lions were housed together in a cage with free access to an adjoining runway. They were fed commercial lion food (Zupreme) consisting of horsemeat with a nutrient supplement. Exercise Protocol Samples were collected from the animals at rest or after running to exhaustion at each of five speedsrepre-

MUSCLE

FIBER

ACTIVITY

DURING

673

RUNNING

senting a walk, a slow trot, a fast trot, a slow gallop, and a fast gallop. These speeds covered a range of metabolic powers from 55% to 168% of vo, max. The percent of . vo 2max9 the number of experiments at each speed, running speed, duration of exercise, an istance covered at each speed are summarized in Table 1. The animals became exhausted after relatively exercise periods at high speeds and the distant ed were less, so it was not possible to normalize our experiments at the different speeds for total distance or power. Only one experiment was carried out on a particular day, and at least 2 days elapsed between experiments. Resting muscle glycogen concentrations showed small daily variation (Table Z), indicating that there was sufficient time between experiments for complete glycogen resynthesis to occur. Tissue Sampling

and Analysis

Biopsy samples (10-30 mg) were obtained from two muscles: the long head of triceps brachii and biceps femoris. These muscles were chosen because they represent a major locomotory muscle in the forelimb and hindlimb, respectively, that can be palpated an bly sampled with the biopsy technique. The samples were removed while the lions were under Bre aesthesia (7 mg/kg in a solution of 20 mglml tered intravenously through an indwelling jugular catheter). The lions were anaesthetized only once on a particular day. Muscle samples were processed within 2-8 min after the end of an exercise bout. The biopsy samples were immediately divided into two parts. One portion was quickly frozen in 1iqui.d NZ for subsequent analysis for glycogen with the anthrone technique (21) or the fluorometric procedure described by Lowry and Passonneau (14). For the fluorometric assay glycogen was hydrolyzed with MC1 and measured as glucose units. The remaining portion was mounted TABLE

1. Performance

data for the exercising

Rest Speed, km/h % Maximal aerobic power Exercise time, min Distance covered, km No. of expts Values

Walk

0

5.0

8

0

5

Slow trot 0.5

55

21.5

0

1.7

+

6.9

-+ 0.7

2.1

15.1 + I.3

49.0 -+ 0.1

+- 0.1

I

0.6

111

137

168

16.1 t 4.6

6.7

-+ 2.1

2.2

2

1.7 * 0.2

1.3

+

0.5

t 0.1

7

4

* SE.

4

vo, max for the lions

TABLE 2. Glycogen concentrations at rest and after exercise

0.4

0.5

1.1

t

0.3

?I 0.1

4

3

was 3.6 ml * g-’ ~h-’

of lion

Depletion

Mean glycogen concentrations in each muscle [G] for each sampling condition were computed. The mean rate of glycogen depletion in each muscle (R) for each exercise condition (e) was calculated by dividing the decrease in glycogen concentration by the time over which it occurred R,=-------

where

t = exercise

I31rest

+

II Gl e

time in min.

Rates of Glycogen

Depletion

uscke composition. The estimated proportion of a muscle (P,) represented by a given fiber type (f) was calculated by multiplying the average cross-sectional area of each fiber type (A,) by the percentage of fibers of that type within each muscle (“/of) and dividing this by the sum of cross-sectional area of all of the fiber types

0.2

A,* %f =

C,(A,.

%,)

l

loo

3

(20).

muscles

where f = FOG, FG, SO. Muscle staining intensity. The average staining in-= tensity of each fiber type in each biopsy sample (Iaf) was calcula.ted by dividing the sum of the staining intensities of fibers of each type (I,) by the number of fibers of

I af

(long

Values are means f SE in mmol of glucose units values significantly lower than resting values: 0.05.

Fast gallop

12.0

Rates of Glycogen

‘8%)

are means

Triceps brachii head) Biceps femoris

____-~Fast trot Slow gallop

70

Biochemical

Histochemical

lions

Condition Variable

on a specimen holder and frozen in 2-methylbutane cooled with liquid N,. Serial sections were subsequently cut from this sample in a cryostat at -2O”C, mounted on coverslips, and stained for diphosphopyridine nucleotide-diaphorase (DPNH-diaphorase) (16) and myofibrillar adenosine triphosphatase (ATPase) (17) activities, and for glycogen with the periodic acid-SchifYs (PAS) stain (18). Muscle fibers were classified as fast-twitch oxidative glycolytic (FOG), fast-twitch glycolytic (FG), or slowtwitch oxidative (SO) from the DNPH-diaphorase and ATPase stains, using the system proposed by Peter et al. (19). Fiber populations were estimated by classifying and counting 50-100 fibers from each sample in randomly chosen fascicles distributed throughout the crosssectional area of the sample. Fiber diameters were measured from the sections stained for DPNH-diaphorase by use of a light microscope with a micrometer eyepiece. Staining intensities for glycogen in the fibers were estimated from the PAS sections by use of a light microscope and were subjectively rated as dark, moderate, light, or negative. These ratings were subsequently assigned values of 4, 3, 2, and 1, respectively, for mathematical and statistical treatment.

=

-4 >I f ---

(3

nf

per kg wet muscle weight. Postexercise * P < 0.01; t P . (See MATERIALS AND METHODS for explanation of histochemical units).

RESULTS

Rates of Glycogen

Depletion

as a Function

of Speed

The rate at which glycogen was depleted from the two muscles increased logarithmically with speed (Fig. 1, A and B). There were no discontinuities at the gait transition points. The relationships between log depletion rate and speed were similar for both the biochemical and histochemical measurements. A high correlation (r = 0.906) with a slope of 1 was found between the two determinations (Fig. 2) indicating it was valid to use histochemical measurements for estimating rates of glycogen depletion in our experiments. The mean values of glycogen concentration in the muscles after each experiment are presented in Table 2.

r = 0906 y = 0986x + 0112 1: 95 % Conf ldence 3

FIG.

glycogen logarithms TABLE

Cross-Sectional Areas of Muscle Showing Glycogen Loss and Rates of Depletion of Fiber Types as a Function of Speed

~--

L/r

’ 1.0

1 15 LOG (Blochemcal

20 25 Glycogen Depletion Rate)

2. Logarithms of histochemically depletion rates as plotted of biochemically determined

in

determined whole muscle Fig. 1B are regressed on rates (Fig. 1A).

3. Fiber composition

of the lion muscles -___ -_____ Fiber

Characteristic

so Triceps

Both the triceps and biceps femoris muscles possessed three distinct fiber types: SO, FOG, and FG. The FG fibers were the largest and most numerous, accounting for more than 55% of the cross-sectional area of both muscles (Table 3). SO fibers constituted about 30%, and FOG fibers the remaining 10-E%. Although no discontinuities were observed in the glycogen depletion rates of the whole muscles during transitions in gait, there were marked discontinuities both in the total cross-sectional area of the muscle that showed glycogen loss (Fig. 3) and in the rates at which the different types of fibers depleted glycogen (Fig. 4). Approximately 12% of the cross-sectional area of both muscles showed glycogen loss after walking. In triceps muscle this increased with increasing speed and reached 60% at a fast trot. Then the cross-sectional area of the muscle showing loss decreased to 28% as the lion switched from a fast trot to a slow gallop, even though it was moving at a higher speed relative to the tread. As the animal continued to increase its galloping speed, the

Diameter, Populations, Estimated sections*

pm % proportion

Diameter, Populations, Estimated sections*

pm

of muscle

cross-

brachii

of muscle

cross-

are means

f SE.

* Calculated

FG

30.36 + 1.07 19.30 + 3.30 15.6

38.97 2 1.15 42.00 k 2.30 55.9

27.08 + 1.23 17.40 + 3.20 11.3

37.17 t 0.82 47.10 + 3.00 57.4

femoris

31.59 -+ 0.66 35.50 -t 4.80 31.3

% proportion

Type

FOG

(long head)

28.96 + 1.15 38.70 + 2.40 28.5 Biceps

Values

Inter vat

as described

in text

by Eq. 2.

area showing loss of glycogen increased again to about 45% at the fastest galloping speed we used in our experiments. Discontinuities in amounts of glycogen depletion of the FG fibers were responsible for the dramatic alterations in the proportions of the triceps muscle showing glycogen loss. Significant (P < 0.05) rates of glycogen loss were observed in the FG fibers during fast trotting and fast galloping in triceps muscle, but not at the slow gallop. All of the glycogen loss during slow galloping occurred in the two high oxidative fiber types. The rates of glycogen depletion were similar for FOG and SO

MUSCLE

FIBER

CTIVITY

DURING

675

RUNNING DISCUSSION

q n

Triceps Brachu M Beeps Femorts M

L

I

20

IO SPEED (km x hr-‘1

FIG. 3. Proportions of muscle cross-sectional areas showing glycogen loss after running as a function of speed. These values were calculated from percentages of fibers of each type showing any glycogen loss extrapolated to whole muscle cross-sectional area as per Eq. 2. Depicted range represents maximum and minimum proportions of muscle showing depletion.

Triceps

t

Brachil

M

1

I

0

I

IO SPEED

20

(km x hr-‘1

FIG. 4. Logarithms of histochemically determined glycogen depletion rates within the FOG (m), FG (A), and SO (0) fiber populations of muscles as a function of speed. These values were calculated as per Eq. 6. Ordinate units are log (1,000 x histochemical units x mine1 + 1). Values represent means rt SE.

fibers at all speeds except walking, where SO fibers experienced a faster utilization rate. The depletion patterns observed in biceps femoris muscle were considerably different than those for triceps muscle (Fig. 3). The proportion of biceps femoris muscle experiencing glycogen loss was fairly constant throughout trotting (=19%), was elevated slightly during slow galloping (=29%), then declined to the walking level during fast galloping (= 12%). The glycogen depletion rates in the fibers in biceps femoris muscle were also quite different (Fig. 4). The rates of depletion in SO fibers were sim ilar for the t wo muscles. Howev mer, in the two fast-tw itch fiber types, it appears that the glycogen depletion rates in biceps fe moris muscle were relatively high at the slower speed in each gait but decli .ned as the lions increased speed within the gait.

The absence of any discontinuities in the whole muscle glycogen depletion rates at the gait transition points is consistent with the absence of discontinuities in the function relating To2 and speed that has been observed in many quadruped mammals (22). oo,, however, increases linearly with speed, whereas rates of glycogen depletion increase logarithmically. This would seem to indicate a progressively greater reliance on glycogen as a substrate as speed increases. In the lion, voZrnax was exceeded at a slow trot and the transition from trot to gallop occurred at a metabolic power of = 128% of Vo, max (20). Thus the observed discontinuities in the rates of glycogen depletion at the trot-gallop transition could not be explained by an abrupt change in the reliance on glycogen as a substrate. In a number of recent investigations glycogen depletion has been used to estimate prior activity of muscle fibers in men (6, 9, 10) and animals (1, 2, 7, 8, 13). These experiments have generally demonstrated that low-tomoderate muscular activity is primarily supported by fibers with high oxidative capacities and/or slow contractile properties. With increasing muscular effort there appears to be a sequential recruitment of fibers with lower oxidative potentials and faster contractile characteristics. There have been no major conflicts between these findings and the information that has been obtained from in situ experiments concerned with the mechanisms that govern motor unit recruitment (3,12). None of these studies was designed to examine alterations in fiber recruitment within and between gaits, so the discontinuities in FG fiber utilization that we found in the present experiments have not previously been noted. Considerable caution must be observed in using glycogen depletion to estimate previous muscle fiber activity. The problems implicit in this type of interpretation have previously been described in considerable detail (1, 3, 9). A number of these concerns were circumvented in this particular study through the use of lions as experimental subjects. Because the gait change from trot to gallop occurs well above the animals’ maximal aerobic power, the complicating factor of major alterations in substrate utilization and blood flow in the muscles were minimized. Also, all of the fiber types in lion muscles have a relatively low oxidative capacity and there is not a large metabolic difference among SO, FOG, and FG fibers. Even though approximately half of the lion triceps muscle is composed of fibers classified as FOG or SO (Table 2), the muscle possesses about the same aerobic potential as pure FG muscle samples from rat gastrocnemius muscles (personal observation). Since all of the fibers in lion muscles possess low capacities for oxidative metabolism, large differences in substrate preference among the fibers at supramaximal exercise intensities probably did not exist. The relatively low aerobic capacities of the lions made them ideal subjects for these experiments and it seems reasonably safe to talk about populations of active fibers in the muscles of these animals. In this study, we tested the hypothesis that the trot-

676

gallop transition occurs at a point where the active cross-sectional area of muscles exceeds some critical point for optimal muscular efficiency. This point could be maximum tension level. However, our data indicate that the gait transition did not occur as a result of the muscles attaining a maximal tension level, either on a whole muscle basis or within specific fiber populations. During fast trotting only 60% of the cross-sectional area of the triceps muscle experienced glycogen loss, and in biceps femoris muscle only about 13% of the muscle was depleted. Under no condition did all of the fibers within a population show loss of glycogen. In fact, the greatest proportional contribution was observed in the FOG fibers in triceps muscle after fast trotting, at which time 69% of the population displayed loss of glycogen. It must be emphasized that we have looked at only two muscles in this study. The possibility exists that other muscles or fiber types within other muscles may reach a maximum tension level at the gait transition, necessitating the recruitment of additional muscle groups. The lack of any discontinuity in the rate of glycogen depletion of the whole triceps muscle as the lion changed from a trot to a gallop and the decrease in crosssectional areas of active muscle at first seems paradoxical. Reference to Fig. l.Z?shows that rates of glycogen loss were indeed overestimated from the histochemistry at the fast trot and underestimated during the slow gallop, but the data also indicate that fewer fibers showed faster rates of depletion during the slow gallop than during the fast trot. Thus it appears that the slower fibers may be active for a proportionally longer period of the stride. It is interesting to note that when an animal switches from a trot to a gallop the period of a repeating cycle doubles, since during a trot there are two steps per stride while during a gallop there is only one (4).

ARMSTRONG

ET

AL.

Our findings are interesting in light of the concept of orderly recruitment of motor units in active muscles (12). According to this concept one might expect to find all SO and/or FOG fibers losing glycogen before seeing any FG depletion. However, during fast trotting and fast galloping, some fibers within all three fiber types were completely devoid of glycogen while other fibers showed no depletion. These results suggest that different configurations of motor units may be recruited to meet specific muscular contractile requirements without conforming to a sequential recruitment order dependent on fiber type (i.e., some FG fibers may be recruited before all of the SO fibers are activated). The reason for the different glycogen depletion patterns in triceps brachii and biceps femoris muscles is not readily apparent. Both muscles cross two joints, and both function as extensors at the first joint (shoulder and hip, respectively), although at the second joint, triceps muscle serves as an extensor, and biceps femoris muscle, a flexor (elbow and knee, respectively). EMG recordings from dogs running at different gaits (unpublished observations) reveal no discontinuities at the gait transition in either muscle, and both muscles were active during the propulsive phases in their respective limbs. The major difference between the two muscles in the dogs was that biceps femoris underwent both shortening and lengthening contractions, whereas triceps muscle was active only during shortening. No data have been reported on the forces generated by these muscles during locomotion. Until this type of information is available, explanation of the contrasting glycogen depletion patterns will be difficult. This work AM18123-02 Received

was supported and AM18140-02.

for publication

by National

27 December

Institutes

of Health

Grants

1976.

REFERENCES 1. ARMSTRONG, R. B., C. W. SAUBERT IV, W. L. SEMBROWICH, R. E. SHEPHERD, AND P. D. GOLLNICK. Glycogen depletion in rat skeletal muscle fibers at different intensities and durations of exercise. Pfluegers Arch. 352: 243-256, 1974. 2. BALDWIN, K., J. S. REITMAN, R. L. TERJUNG, W. W. WINDER, AND J. 0. HOLLOSZY. Substrate depletion in different types of muscle and in liver during prolonged running. Am. J. Physiol. 225: 1045-1050, 1973. 3. BURKE, R. E., AND V. R. EDGERTON. Motor unit properties and selective involvement in movement. Exercise Sport Sci. Rev. 3: 31-81, 1975. 4. CAVAGNA, G. A., C. R. TAYLOR, AND N. C. HEGLUND. Mechanical work in terrestrial locomotion: two basic mechanisms for minimizing energy expenditure. Am. J. Physiol. In press. 5. CHASSIN, P. S., C. R. TAYLOR, N. C. HEGLUND, AND H. J. SEEHERMAN. Locomotion in lions: energetic cost and maximum aerobic capacity. Physiol. ZooZ. 49: l-10, 1976. 6. COSTILL, D. L., P. D. GOLLNICK, E. D. JANSSON, B. SALTIN, AND E. M. STEIN. Glycogen depletion pattern in human muscle fibers during distance running. Acta PhysioZ. Stand. 89: 374-383, 1973. 7. EDGERTON, V. R., D. R. SIMPSON, R. J. BARNARD, AND J. B. PETER. Phosphorylase activity in acutely exercised muscles. Nature 225: 866-867, 1970. 8. GILLESPIE, C. A., D. R. SIMPSON, AND V. R. EDGERTON. Motor unit recruitment as reflected by muscle fiber glycogen loss in a prosimian (bush baby) after running and jumping. J. NeuroZ. Neurosurg. Psychiat. 37: 817-824, 1974.

9. GOLLNICK, P., R. B. ARMSTRONG, C. W. SAUBERT IV, W. L. SEMBROWICH, AND R. E. SHEPHERD. Glycogen depletion patterns in human skeletal muscle fibers during prolonged work. Pfluegers Arch. 244: 1-12, 1973. 10. GOLLNICK, P., J. KARLSSON, K. PIEHL, AND B. SALTIN. Selective glycogen depletion in skeletal muscle fibers of man following sustained contractions. J. PhysioZ., London 241: 59-68, 1974. 11. HEGLUND, N. C., C. R. TAYLOR, AND T. A. MCMAHOW. Scaling stride frequency and gait to animal size: mice to horses. Science 186: 1112-1113, 1974. 12. HENNEMAN, E., AND C. B. OLSEN. Relations between structure and function in the design of skeletal muscles. J. Neural. 28: 581-598, 1965. 13. LINDHOLM, A., H. BJERNELD, AND B. SALTIN. Glycogen depletion pattern in muscle fibers of trotting horses. Acta Physiol. Stand. 90: 475-484, 1974. 14. LOWRY, 0. H., AND J. V. PASSONNEAU. A FZexibZe System of Enzymatic Analysis. New York: Academic, 1972, p. 174-177. 15. MCMAHON, T. A. Using body size to understand the structural design of animals: quadrupedal locomotion. J. AppZ. Physiol. 39: 619-627, 1975. 16. NOVIKOFF, A. B., W. SHIN, AND J. DRUCKER. Mitochondrial localization of oxidative enzymes: staining results with two tetrazolium salts. J. Biophys. Biochem. Cytol. 9: 47-61, 1961. 17. PADYKULA, H. A., AND E. HERMAN. The specificity of the histochemical method of adenosine triphosphatase. J. Histochem. Cytochem. 3: 170-195, 1955.

MUSCLE

FIBER

ACTIVITY

DURING

RUNNING

18. PEAIUE, A. G. E. Histochemistry - Theoretical and Applied. Boston, Mass.: Little, Brown, 1961, p. 832. 19. PETER, J. B., R. J. BARNARD, V. R., EDGERTON, C. A. GILLESPIE, AND K. E. STEMPEL. Metabolic profiles of three types of skeletal muscle in guinea pigs and rabbits. Biochemistry 11: 2627-2633, 1972. 20. SEEHERMAN, H. J., C. R. TAYLOR, AND G. M. 0. MALOIY. Maximum aerobic power and anaerobic glycolysis during running in

677 lions, horses, and dogs (Abstract). Federation Proc. 35: 797,1976. 21. SEIFTER, S., S. DAYTON, B. NOVIC, AND E. MUNTWYLER. The estimation of glycogen with the anthrone reagent. Arch. Biothem. 25: 191-200, 1950. 22. TAYLOR, C. R. The energetics of terrestrial locomotion and body size in vertebrates. In: Biodynamics of Animal Locomotion. London: Academic. In press.