PHYSIOLOGICAL

REVIEWS

Vol. 52, Kc’o. 1, January Printed

1972

in U.S.A.

Dynamic Mammalian

Properties of Skeletal Muscles

R. I. CLOSE Department Introduction. Fiber Types. Historical Classification Contractile Mechanical Introduction

of Physiology,

Australian

National

University,

Canberra,

Australia

............................................................ ............................................................ ................................................. introduction. ........................................... and terminology. properties Properties.

130 130 131 134 138 138 138

............................

of different types of fiber. ....................................................

........................................................... ................................................ Series-elastic component. Length : tension relation of contractile material. Force: velocity properties of contractile component.

129

............................. .........................

140 145

Behavior

of series-elastic and contractile components in isotonic and isometric ....................................................... contractions Active state, time course of isometric twitch, and posttetanic potentiation....... Ontogenetic Differentiation of Fast and Slow Muscles. ......................... Growth Dynamic Other Kelation Speed Speed Neural

............................................................... .................................................... properties. .......................................... developmental changes.

Effects Neural Correlations

169 170

........................................................... properties of normal and cross-innervated

muscles .................. ......................... of nerve cross-union on properties of myosin. influences on noncontractile structures in muscle cells ................. Between Dynamic and Chemical Properties of Contractile Material.

Introduction Structure Relation Functional

I.

161 163 166 166 166

.............................. Between Size and Speed of Contraction ............. of contraction of homologous muscles of different species. of contraction of different muscles of same animal. .................... Control of Dynamic Properties. ......................................

Introductiorl Dynamic

Review

147 149 161

........................................................... .................................................... of myosin. between ATPase activity of myosin and intrinsic speed of shortening differences between fast and slow muscles. ....................... ........................................... of Some Major Problems

170 172 175 176 ..

177 177 177

.

181 182 183

INTRODUCTION

Several kinds of extrafusal fiber have been identified in mammalian skeletal muscles on structural, physiological, and chemical criteria. This review deals with some of the properties of focally innervated extrafusal fibers that have propagated twitch responses. The main aim is to describe action potentials and “all-or-none” the mechanical properties of various types of mammalian twitch muscle fiber in physical terms and in relation to the structural and chemical properties of the con129

130

R.

I. CLOSE

Volume 52

tractile material and other components of individual fibers. A previous review by D. h/I. Needham of these aspects of mammalian muscle appeared in Physiological Reviews in 1926 (301). Readers interested in the properties of slow-graded, multiply innervated mammalian muscle fibers should consult the recent reviews by Peachey (325) and Hess (204). The classifications of mammalian twitch fibers are described in the first part of this review together with some of the contractile properties of the principal types of fiber. The remainder deals with various physiological properties of fasttwitch and slow-twitch muscles of different mammals. Particular attention is given to descriptions of differences and changes in the characteristic dynamic properties of different muscles. These variations in dynamic properties are important because they provide opportunities to determine some of the relations between mechanical, chemical, and structural properties of the contractile material.

II.

FIBER

TYPES

A. Historical

Introduction

It has been known for a long time that different mammalian skeletal muscles range in color from near white to deep red and that in a given animal species some skeletal muscles contract more quickly than others. Kiihne (257) showed the color difference was a property of the muscle fibers and was not attributable to differences in blood content. In 1873 and 1874 Ranvier (342, 343) published the first results of experimental work that showed that certain red muscles in the rabbit contracted much more slowly than white muscles; this observation has been confirmed by many nvestigators, including Kronecker and Stirling (254), Fischer (140), Denny-Brown (93), and Cooper and Eccles (89). Shortly after Ranvier’s work it was shown that the color of a muscle is not always associated with a particular speed or fiber structure. For example, Meyer (289) found that some red muscles, such as rabbit masseter muscle, had histological characteristics more like those of white muscles than some other red ones, and Knoll (249) noted that some red muscles contract more rapidly than white muscles. In 1904 Paukul (324) reported the results of his investigations on the properties of many different muscles of the rabbit, and his records of twitch contractions show clearly that slow-twitch muscles such as soleus and ischio-tibialis are always red but that not all red muscles are slow. In particular, Paukul found that the time course of the twitch of the red masseter muscle is about the same as that of fast-twitch muscles such as the medial gastrocnemius. Ray Lankester (344) and Knoll (249) observed that red pigment is present in those muscles involved in almost continuous activity, and Ranvier and others reported that red muscles are more resistant to fatigue than white muscles (93). Early histological studies also revealed that mammalian skeletal muscles are usually heterogeneous. For example, Griitzner (177), Knoll (249), Schaffer (373), and others (52, 93) found that most mammalian muscles are made up of at least

January

1972

DYNAMIC

PROPERTIES

OF

MUSCLE

131

two kinds of fiber: the red fibers are usually thin and dark and contain many mitochondria and fat droplets, whereas the white fiber is thick, appears clear, and contains few mitochondria and fat droplets. The full extent of this heterogeneity has been revealed with new histochemical techniques developed during the last 20 years, described below. B. Classzjkation

and Terminology

Two histochemically distinct groups of fiber have been recognized by Dubowitz and Pearce (108) on reciprocal differences in the activities of oxidative and glycolytic enzymes and by Engel and his colleagues (126, 127) on the basis of differences in activities of myofibrillar adenosinetriphosphatase (ATPase) at pH 9.4; some exceptions to this generalization have been discussed by Bocek and Beatty (35). It has been suggested by others that classification of fibers into two types may be an oversimplification, and recent histological and histochemical work has revealed three principal types of fiber that differ in ultrastructure, the activities of oxidative and glycolytic enzymes, myoglobin content, and certain properties of myosin. The various systems of nomenclature and the principal morphological and histochemical properties of the three kinds of fiber are listed in Table 1; the terms white, intermediate, and red are used in this review to describe the three histochemical types. One other classification of fibers into three types on the basis of qualitative differences in pH lability of actomyosin ATPase has been proposed (180, 365, 443) but is not included in Table 1 because it is not compatible with the other systems when applied to fibers from different animal species. It has been necessary to simplify the descriptions of the fibers and Table 1 is intended only as a guide to the literature on the subject; it would be necessary to consult the original works to find all the details on which the classifications are based, particularly those concerning ultrastructure. The histochemical properties listed in Table 1 indicate that white fibers have high glycolytic, low oxidative, and high myofibrillar ATPase activities, that the intermediate fibers have low glycolytic, intermediate oxida tive, and low myofibrillar ATPase activities, and that the red fibers have intermediate glycolytic, high oxidative, and high myofibrillar ATPase activities. Histochemical identification of white fibers is relatively simple but confusion has resulted from what appears to be incorrect identification of intermediate and red fibers. In some instances the identification has been based on differences in overall intensity of staining for oxiclative enzymes rather than on the distribution of diformazan granules and mitochondria as described by Stein and Padykula (392) and Padykula and Gauthier (3 19). Identification of intermediate and red fibers on the basis of histochemical staining reactions for oxidative enzymes may be complicated by reciprocal changes in oxidative enzyme concentration and adaptive changes in the cross-sectional dimensions of muscle fibers (164, 168). The photomicrographs in Figure 1 of Yellin and Guth (443) show that intermediate and red fibers, identified on the basis of differences in distribution of succinic dehydrogenase, may have opposite intensities of overall staining in fibers of homolo-

132

R.

1. Morphological

TABLE

I.

und histochemical

CLOSE

Volume

types of fiber in mtlmmalian

52

skeletal muscle

Class$cat Ogata Stein

ions and Mori and Padykula

(309,

313) (392)

Engel (126, 127) Romanul (352) Padykula and Gauthier Kugelberg and Edstrom

(318) (256),

neman and Olson and Swett (316) Barnard et al. (24)

Hen-

(X)2),

White A II

Mcdiunt 13 I

Red

I White A

IIT I II ternlediate c

C II II Red B

Fast-twitch wh i 1~:

Slow-twitch in ternledi-

Fast-twitch red

IIB

I

Small

I n tcrincd

Olson

att’ Brooke

and

Morphological Mitochondrial

Kaiser

(43,

properties content

44)

(152,

319, 381) 2 line (152, 318, 319, Fiber diameter* (152, 313, 319, 392)

310,

315,

II.\

iate

(and

IIc?)

Large

318,

Neuromuscular 314, 319) Histochemical Distribution ase

junction

313,

Glycolytic

In terrricdiate

I,arge

In ternlediate

Broad Small

(3 1 1, 3 12,

Large

In tcrnlcdiate

Small

dehydrogen-

enzyme activities 352, 392) ATPase (1.52)

Even

(12!),

and simple

network

Even

T ,ow

network

Prcdominan subsarcolemrrraf

High

or intermediate I n terniediate

Low

tly

Intermediate high High

or

activities

(352) (129) Myoglobin Glycogen Myofibrillar (118,

and

co1nplcx

properties of succinic (392)

Oxidative 309, Mitochondrial

Narrow I%)!), f310,

381) 168,

content

(352)

content (160) ATPase at pH 123, 126, 127, 392)

(129) pH sensitivity

(42, 43, Formaldehyde brillar

High

Low

Intermediate

High Low Intermediate

Variable High

LOW High

T .ow

High

1 AOW

High

9.4

of rnyofibrillar 179, 36.5)

ATPase

sensitivity ATPase (179,

of 392)

been

that

rnyofi-

High High Acid

reported

fiber

labile,

;\cid

alkali stable Sensitive

_--__* It has

Variable

labile, alkali stable

Stable

--___~-__ diameter

High Acid

stable, alkali labile

does

not

always

.-show

this

pattern

(118,

316).

gous muscles of different species. In the light of these observations it appears that intermediate and red fibers of rat tibialis anterior muscle may have been incorrectly identified, one for the other, by Kugelberg and Edstriim (256); it is difficult to be certain about this point but their observation (2.56, p. 416) that fibers they designated B and C were the same as those described by Stein and Padykula (392) is

./muary

1972

I )YNAM

1 C PROPERTIES

OF

M USCTX

133

not consistent with their later conclusion (256, p. 421) that these two groups of fibers corresponded rcspcctivcly to type II and type III fibers of Romanul (352). An error of this kind would account for the unexpcctcd differences in glycogen con tent, phosphorylasc activity, distribution of succinic dehydrogenase, and the effects of repetitive activation reported to exist between either B or C fibers in tibialis anterior and solcus muscles (256). The overall intensities of stains for mitochondrial enzymes have been used to distinguish between different types of fibers in pig (88) and cat (202, 316) muscles, and the fibers reported to be intermediate and red may have been red and intermediate, respectively. In one investigation the intensity of staining of succinic dehydrogenasc of rat muscle has been determined photometrically (413), but again this does not determine differences in distribution of the enzyme. Unequivocal identification of most. fibers can be achieved by staining serial sections for myosin ATYase, phosphorylase, and oxidativc cnzymcs (24, 118, 392, 443). For example, intcrmcdiatc fibers may bc clearly distinguished from white and red fibers on the basis of intensity of stain for myosin ATPasc, and white fibers show a much lighter staining reaction for oxidativc cnzymcs than either intermediate or red fibers (Table 1). It has generally been found that fibers can bc divided into three groups on the basis of all the enzyme activities listed in Table I cxccpt myofibrillar ATPase. This may mean that myosin ATPasc act.ivity is the same in white and red fibers in vii-o, despite qualitative differcnccs in pH (43, 44, 180, 365, 443) and formaldehyde (179, 392) scnsitivitics, but in attempting to interpret this observation it is important to keep in mind the limitations of the histochemical method of esti1nating myosin ATPasc. For cxa mplc, t hc activity of this cnzymc is usually deterrnincd at pII 9-10 (320) and although this incrcascs the difference between activities of rnyosin ATPascs of fast and slow ~nuscles(364, 387), and hcncc the sclectivity of the histochcmical method, it may obscure diffcrcnccs in white- and redfiber myosins if these ha.vc diffcrcnt 1,l-I dependcncics; furthermore the reaction time and small diffcrcnccs in the hydrogen ion conccntra tion of the medium have profound effects on the in tcnsi ty of the stain for myofibrillar ATPase (43, 365). Another difficulty in intcrprctation arises bccausc stain intensity is usually estimated subjectively and this introduces uncertainties; for example, the staining intensity of myofibrillar ATPase of white, intcrmcdiatc, and red fibers of rat gastrocncmius shown in Figure 4 of Edgcrton and Simpson (1 18) were assessed as dark, light, and dark, respcctivcly, whcrcas dcnsitomctcr mcasurcmcnts on fibers in the published photomicrographs yicldcd mcan rela tivc stain intensities of 1.O, 0.58, and 0.79, respectively.‘ In view of these limitations and difhculties it seem that at this stage it is not possi blc to dccidc on t hc basis of the conventional histochcmical methods whether myofibrillar ATPasc activity is exactly the same in white and red fibers in vivo, but it is quite clear that it is higher in white and red fibers than in intermediate ones. It is possible that the method dcscribcd by Meijcr (286) for staining myosin ATPasc at pH 7.2 may bc useful in estimating the relative activitics of this enzyme in the three kinds of fiber in conditions comparable to those in viva.

134

R.

C. Contractile

I.

CLOSE

Volume

52

Properties of Dzyerent Types of Fz’ber

Table 2 shows the composition of some mammalian muscles frequently used in physiological studies; the mean diameters of fibers in the first seven muscles listed in Table 2 are available in the original papers. Most of the muscles listed are heterogeneous and rnost contain the three types of fiber. Several attempts have been made to determine the contractile properties of the different types of fiber, usually by comparing the contractile properties of motor units or the whole muscle with the histochemical properties of the muscle. In these investigations it has been assumed generally that each motor unit is composed of only one kind of fiber, and indirect evidence supports this vie N (127). It has been a simple matter to identify intermediate fibers as slow-twitch fibers because the slow soleus muscles of the cat and guinea pig are composed entirely, or almost entirely, of intermediate fibers (Table 2); fur th ermore 90 % of the motor units of rat soleus are slow-twitch units (78) and 80-90 % of the fibers are intermediate (Table 2). TABLE

2. Composition

of dij’erent muscles I

~_----

‘J{, of Total Animal

and muscle

Number

of Fibers

Ref White

Intermediate

Red ~

.-

Cat Flexor

digitorum

Flexor Medial Soleus

hallucis longus gastrocnemius

longus

316 316 179,

41.6

202 202, 239,

50.2 50.9 284

28.0 14.0 21.3 95-100

30.4 35.8 27.8

20 40 14

60 52 4

18 11*

41 42*

3 .8 80

4-w

Rat Diaphragm

152 152

Red semitendinosus White semitendinosus Tibialis anterior Extensor

9 82 41

152 256

digitorum

179, 365 118 119

longus

Soleus

47* 60

118 138 239

Guinea p ig Flexor digitorum longus Flexor hallucis longus Medial gastrocnemius “Red” Soleus

20

vastus

85-90

24

53.8 72.8 38.6

24 24 24

lateralis

118,

The original

correspondence papers

cross-sectional areas. fibers was determined fibers in 119.

50.2 78.2 0

192

100

------_ ~~-_______ the

35.5 9.7

17.7 0

24,42

Rabbit Thyroarytenoid

20 6 10-15

between this system of nomenclature is given in Table 1. *From published tPersona1 communication from V. on

the

preparations

used

for

estimating

and other systems used in photomicrographs of limited R. Edgerton; the number of red the

number

of intermediate

Ja?luclry

1972

DYNAMIC

PROPEKTIES

OF

MUSCLE

135

It has been more diflicult to characterize the contractile properties of red and white fibers. The time courses of isometric twitch and tetanic contractions have been recorded for motor units in various muscles of different animals. In no instance has a ~nuscle with the three types of fiber been found to contain three types of motor unit that could be distinguished by the speed of the contractile response. Only two kinds of unit (fast twitch and slow twitch) have been distinguished n cat gastrocnemius muscle (66, 439) and cat flexor digitorum longus (316) muscles &en though both these muscles contain three histochemical fiber types. Olson and Swctt (316) suggested that the fast units of cat flexor digitorum longus are made up of either mitochondria-rich or mitochondria-poor fibers and that slow units contain mitochondria-rich fibers similar to those found in the soleus muscle. It may be deduced from these observations that the fast motor units were made up of either white or red fibers and that the slow units comprised intermediate fibers. Olson and Swett subdivided the fast units into large units that had high isometric tetanic tensions and fatigued rapidly and small units that had low isometric tetanic tensions and fatigued slowly; they suggested that the large and small units were composed of white and red fibers, respectively. Indirect evidence that red fibers arc faster than intermediate fibers was obtained from rat soleus muscle (78) in which 10 % of the motor units had a mean isometric twitch contraction time about half that of the slow units and corresponded in size with the red-fiber component that has been identified on histochemical (118, 239, 392) and ultrastructural characteristics (374). Further evidence that red fibers may have relatively short twitch contraction times was obtained by Hall-Craggs ( 192), who showed that the rabbit thyroarytenoid muscle is composed entirely of red fibers and has a twitch contraction time of about 6.5 msec. A similar conclusion regarding the speed of contraction of white and red fibers can be drawn from the results of several investigations on the rat fast muscle, extensor digitorum longus. Rat extensor digitorum longus muscle is heterogeneous histochemically and is composed almost entirely of white and red fibers in approximately equal numbers, as indicated by histochemical (Table 2) and ultrastructural properties (374); in contrast, the isometric twitch contraction times of motor units of rat extensor digitorum longus are norrnally distributed with a relative standard deviation of only 10 % (78) and it may be deduced that the time course of the twitch of the red-fiber and white-fiber components of this muscle is virtually the same. Kugelberg and Edstrijm (256, p. 421) identified three types of fiber in rat tibialis anterior muscle as white, intermediate, and red through correspondence of properties with the three classes of fiber described by Romanul (352); it was shown above that the USC of A, B, and C to designate the fibers in their papers (121,122, 256) is rather confusing because their B and C fibers appear to correspond to the C and B fibers, respectively, of Stein and Padykula (392). Edstrijm and Kugelberg (122) identified fibers of motor units in rat tibialis anterior muscle by mapping the histochemical changes that occur in phosphorylase and glycogen after repetitive stimulation. They found the white-fiber units contained about 2 % red fibers and no intermediate fibers, the red-fiber units contained about 1 % white fibers and no intermediate ones, and the intermediate-fiber units were prob-

136

R.

I. CT,OSE

Volrr me 52

ably pure. These observations are consistent with the findings of other investigators (25, 117) that white and red fibers appear to bc partially interconvertible (see below). One of the most interesting observations of Edstrom and Kugelberg (12 1, 122) is that white-fiber units and red-fiber units were fast, with isometric twitch contraction times about 12.5 mscc at 35 C, and that white fibers fatigued rapidly, red fibers fatigued slowly, and intermediate fibers showed no evidence of fatigue. The problem has been approached in another way by Barnard et al. (24); they studied guinea pig ~nusclcs composed either entirely or predominantly of one kind of fiber and determined the contractile properties of the whole muscle and the ATPase activity of extracted myosin at pH 6.8. The slow soleus muscle is composed solely of intermediate fibers and has a long twitch contraction time and low myosin ATPase activity Their most interesting finding was that there is no statistically significant difference between the isometric twitch con traction times and myosin ATPase (pH 6.8) activities of the flexor hallucis longus and “red” vastus lateralis despite the fact that these two rnusclcs are composed predominantly of white and red fibers, respectively (Table 2). They concluded that white and red fibers are fast-contracting fibers containing myosin ATPasc with a relatively high specific activity and that in tcrrncdiate fibers contract slowly and contain rnyosin ATPasc of relatively low specific activity. Buchthal and Schmalbruch (5 1) recorded contractions of bundles of fibers in several different human ~u~-lcs and corrclatcd the time course of the isometric twitch with the mitochondrial content of fibers. They distinguished three types of fiber and concluded that fibers with low and intcrmcdiate mitochondrial cm tent had mean contraction times about 64 I~SCT, whereas l~litocholldria-rich fibers had a mean contraction time about 120 iuscc; they designated these fibers A, B, and C, respectively, on the basis of the intensity of Sudan black B staining, but again it is not clear whether their B and C fih*s correspond to those described by &ein of mi tochondria. and Padykula (392) on distribution Several investigations in to the effects of cxercisc on fiber composition and contractile properties of mammalian ru~3clcs are very interesting in connection with the mechanical properties of the three histochemical types of fiber. In 1959 muscle became darker van Linge (265) f ound that the red color of rat plantaris after training. Ten years later Edgerton et al. (1 17) sh owed that plantaris muscle of chronically exercised rats contained a larger proportion of red fibers than plantaris of sedentary animals, but there was no change in the proportion of fibers with low and high histochemical myosin ATPasc activity; their results are consisten t with Binkhorst’s (3 1) observation that the characteristics of isometric contractions of rat plan taris ~nuscle are unaltered after a period of training. The most likely explanation for all these results is that the red-fiber population increased and the number of white fibers decreased (117). This work was extended by Barnard, Edgerton, and Peter (25, 26), who invcstiga ted the effects of prolonged “low-in tensi ty” exercise on guinea pig gastrocncrnius muscle and found that the and red fibers, classified t)y NADH-diaphorase proportions of white, intermediate, staining, were 52 %, 9.5 %, and 38.5 ‘Xl, respectively, in control sedentary animals, and 40 % , 7 % , and 53 ‘i: , respectively, in exercised animals; the increase in the

137 proportion of red fibers after training was statistically significant. No significant alterations were observed in the proportions of fibers that had high and low histochemical myosin ATPasc activities, the biochemical properties of the calcium transporting system of isolated fragments of sarcoplasmic reticulum, or the characteristics of isometric contractions of muscles from control and trained annnals (25, 26). On the basis of these results Barnard ct al. (25) suggested that marnmalian fast-twitch muscles have the capacity to adapt functionally through interconversion of white and red fibers and that red fibers with well-developed aerobic and anaerobic energy-yielding systems arc used preferentially during prolonged sustained phasic activity. This evidcncc of interconversion of white and red fibers with respect to certain histochenlical propertics 1s consis tent with diffcrcnccs in mitochondrial content and fiber dianrcter (Table 1)>but. it is yucstionable whether it involves other morphological characteristics sue h as the form of the neuromuscular junction and the width of the % line (Ta hlc I). The available cvidcnce shows that the fibers of some heterogeneous skeletal muscles can be classified into three groups on the basis of their morphological and histochemical propcrtics but only two groups according to the tirnc course of the isometric twitch and kinetic propcrtics of rnyosin ATPase at neutral pH. Barnard et al. (24) (Table 1) h ave suggcstcd that the three fibers be called fasttwitch white, fast-twitch red, and slow-twitch in tcrrncdia tc, and this nomenclature seems suitable because it employs tcrrns currcn tly used by physiologists and histochemists. This simple classification must not bc intcrprctcd as meaning that all fibers of a particular metabolic type have the same contractile properties in different tnuscles of the same animal, and in this connection it may be noted that the isometric twitch contraction tirncs of red-fiber components of rat extensor digitorum longus and soleus muscles arc probably about 12 mscc and 18 msec, respectively, at 35 C (text above and rcf 78). Morcovcr a particular histochemical type of fiber has different speeds and myosin ATPasc activities in a given Inuscle in different species, the speed depending partly on the adult body si’ze (sect. VA). With these reservations in mind, the three principal types of fiber arc: n) Fast-twitch white fibers with relatively short isonlctric twitch contraction times, high c activity of myosin ATPase, a well-dcvclopcd glycolytic enzynlc system, low mitochondrial content and oxidativc activity; these fibers fatigue rapidly. 6) Fast-twitch red fibers with relatively short isonlctric twitch contraction times, high activity of rnyosin ATPase, modcra tcly dcvclopcd glycolytic system, high total oxidativc enzymc activity and many mitochondria; these fibers arc much more resistant to fatigue than white fibers. c> Slow-twitch intermcdiatc fibers with low speed of contraction and specific activity of myosin ATPasc, poorly developed glycolytic enzyme system, high mitochondrial content and oxidativc enzyme activities; these fibers show little or no fatigue. The properties of each fiber indicate its probable function. For example, white fibers would function as fast units for short-term powerful phasic activity, whereas red fibers would bc bcttcr adapted for sustained phasic activity. On the other hand, the intermcdia tc fi bcrs arc low-speed economical con trac tilt units suitable for sustained tonic activity. (Biochemical diffcrcnces between fast and

138

R.

I. CLOSE

Volume

32

slow muscles in regard to the energy expenditure during maintained contractions are dealt with in section VIID.) The remainder of the review deals principally with differences in the dynamic properties of fast-twitch white and red fibers on the one hand and slow-twitch intermediate fibers on the other. In most instances the physiological work has ken carried out on fairly pure fast-twitch or slow-twitch systems and for simplicity they are referred to either by the anatomical name of the muscle or more generally as fast and slow muscles.

III.

MECHANICAL

PROPERTIES

A. Introduction There is sufficient evidence available now to show that mammalian muscles in the fully active or inactive states have mechanical properties that cl osely resemble those of frog muscles. Mammalian muscles, like frog muscles (207), appear to contain a contractile component, a lightly damped series-elastic component, and a parallel-elastic component. The three-component model is useful in describing the properties of fast and slow mammalian muscles even though there are important quantitative differences between the two, particularly in their force : velocity properties. The first three parts of this section are devoted to a summary of the available information on the 1oad:extension relation of the series-elastic component, and the length : tension and force : velocity relations of the contractile material. The mechanical properties of fast and slow mammalian muscles appear to be qualitatively similar in the fully active or inactive steady states, but there are major differences in the twitch contractions of these muscles that appear to result from differences in excitation-contraction coupling and activation. The final part of this section deals with activation and the active state in twitch contractions. 23. Series-Elastic

ComjPonen t

Evidence for the existence of a lightly damped elastic component in series with the contractile component is based primarily on the biphasic response of a muscle when it is released during isometric contraction and allowed to shorten freely against a load that is less than the isometric tension at the time of release. The initial rapid phase of shortening is completed within a few milliseconds and has been attributed to rapid release of energy stored in the elastic component after the change in load. The second slow phase is the result of shortening of the contractile component in accordance with its force:velocity properties. The biphasic nature of the shortening: time curve after release is shown clearly in records from soleus and tibialis anterior muscles of the rat (429), rat gracilis anticus (1 l), and cat soleus (235), and the responses of these mammalian muscles are qualitatively similar to those of frog sartorius muscle (207, 2 11, 23 1, 432, 433) when released in comparable condi tions.

January

1972

TABLE

3. Load:extension

--

-

DYNAMIC

properties

Muscle

Ref

PROPERTIES

OF

MUSCLE

of series-elastic elements

Conditions of Measurement

Method

Normalized Compliance LO/P, 0.2 P,

--

.-11

~

--.--

-_____

Rat

gracilis anticus

Quick

release

17.5

322 283

Rat Cat

soleus tenuissimus

Quick

429 429

Rat

soleus

Rat

tibialis anterior

mated

f,, is optimum from published

C,

in

vitro

PO ___.-..-

Normalized Extension at PO W

0.15*

0.011

0.07

release

20 C, in

vitro

trolled release Quick release

37 C, in

vitro

0.09* 0.16*

0.015* 0.01*

0.065 0.06

38 C, in

situ

0.18*

0.012*

Quick

38 C,

0.6*

0.033*

0.03 0 .O48

Con

length of muscle graphs.

release

and

P, is maximum

in situ

isometric

tetanic

tension.

*Esti-

Three main methods have been developed for measurement and estimation of the load : extension properties of the series-elastic component of frog muscle and all these methods have been used for studying mammalian muscles. The three methods are a) the controlled-release method of Hill (2 1 1 ), b) the quick-release method described by Wilkie (231, 432, 433), and G) calculation of the load : extension curve from the force:rate relations for isotonic shortening and isometric tension development (207). Table 3 shows some of the properties of the series-elastic components of several nammalian muscles determined by quick- and controlled-release methods. The relative values for compliance at maximum isometric tetanic tension (PO) and 0.2 p0 show that the load : extension relation is nonlinear, and although the compliance always decreases with increase in load it is clear that there is considerable variation in the actual slopes of the curves that have been obtained. Furthermore, Parmley et al. (322) obtained a linear relation between compliance and load for rat soleus in vitro, whereas Joyce and Rack (235) found that the relation between stiffness of the series-elastic component and load was nonlinear for cat soleus in situ. The estimated maximum extension of the series-elastic component at PO (Table 3) was about the same in three muscles examined in vitro (11, 283, 322) but was considerably less in two muscles examined in situ (429); this difference may be due partly to the damping effect of surrounding tissues in situ, thereby reducing the extent of shortening of the elastic elements after release. In attempting to evaluate the properties of the series-elastic component that have been determined from measurements on whole muscle it is important to remember that the proportion of tendon in the total length of the muscle varies greatly from one muscle to another. For example, the length of the tendon that runs along the side of the muscle and in series with the muscle fibers is about one-half the overall length of rat extensor digitorum longus muscle and about one-third the length of the rat soleus ~nusclc (74). In other more complicated muscles such as tibialis anterior and gastrocncmius it may be different in different parts of the muscle. It rnay be noted that the maxi-

140

R. I. CLOSE

Volume

52

mum extension of the series-elastic component of rat muscles is about 7 */o of the optimum length (L,,) ( see sect. IIIC), and this total compliance is more than 3 times greater than the corresponding value (0.02 L,) for the elastic components of frog sartorius; this difference is probably largely due to differences in the amount of series tendon in these muscles but it is also possible that it results partly from differences in the elastic properties of tendons and structures within the muscle fibers. Further investigation is needed to clarify this point. The method of estimating the properties of the series-elastic component from the force:velocity relation and the relation between tension and the time derivative of isometric tension during a tetanic contraction (method c above) is based on the assumption that the contractile component elongates the series-elastic component in isometric contractions and that the force:velocity relation of the contractile component is the same in isotonic and isometric conditions. In other words, neglecting the properties of the recording apparatus, dx dx dt dP = dt’dP where .Y is the change in length of contractile or series-elastic components, P is the load, and t is time. This method has always yielded values for compliance and maximum extension of the series-elastic component that are higher than those obtained by the quick- or controlled-release methods. Wilkie (43 1) used this method and estimated that the maximum extension in the series-elastic component of human muscle is about 0.1 L, at PO . McCrorey et al. (283) obtained a value of 0.11 L, at P, for segments of cat tenuissimus muscle and also found that the calculated values for the compliance of the elastic component at any given load were about 2 times greater than those obtained by the quick-release method on the same muscle. This observation complements those of Jewel1 and Wilkie (231) and Parmley et al. (322) that showed differences between calculated and observed values for dP, ‘dl and dx/‘dt, respectively, when estimated from equation I above. A possible explanation for these differences is discussed in section IL,!?. Another characteristic feature of the series-elastic component of muscle is that it is lightly damped. Bahler (11) obtained a value of about 300 dynes/cm per set for viscous resistance of rat muscle, which is comparable to the range given by Woledge (436) for frog sartorius muscle. C. Length: This muscle. teristics must be

Tension Relation

of

Contractile Material

relation has not been investigated thoroughly in any mammalian skeletal This is rather surprising because it is one of the most important characof the contractile material and one of the most important conditions that clearly defined in any work on mechanical properties of muscle.

1. Length:

isometric

tetanic tension relation

The tension in an inactive whole muscle is zero at rest length and increases approsimatelv , exponentially with increase in length (17, 58, 154, 22 1, 281, 283,

January

197.2

DYNAMIC

PKOPEKTIES

OF

MUSCLE

141

341, 378, 416, 422), but in some muscles it increases linearly at high tensions (394, 429). The length: tension curve for the contractile material of active mammalian muscle has usually been obtained by subtracting the initial passive tension from the total tension during isometric contraction. It has generally been found that there is a well-defined optimum initial length (L,) f or maximum isometric tetanic tension that is between 100 % and 120 % of the resting length (4, 13, 14, 58, 154, 221, 283, 341, 378, 416, 422) and is always within the normal range of muscle lengths in situ. In most instances only part of the curve has been determined for any particular mammalian muscle and measurements have usually been restricted to lengths about the optimum or less, presumably because increase in length of some muscles beyond about 1.3 L, causes irreversible effects (14). Nevertheless there is sufficient information from different investigations on various muscles that provides a fairly reliable indication that the length: tetanic tension relation of the contractile material is essentially the same for mammalian and frog muscles. The best available results on the sarcomere length-tetanic tension relation are those of Gordon, Huxley, and Julian (175) ; they determined the whole of the sarcomere length : tension urve for the contractile material of frog fibers in which the sarcomerc length was held constant during measurements and were able to find a satisfactory interpretation for the length : tension relation in terms of the lengths of the interdigitating thick myosin and thin actin filaments and the slidingfilament theory of contraction. The lengths of thick and thin filaments of frog muscle are 1.6 p and 2.05 p, respectively (321). G or d on et al. (175) drew attention to three important features of the sarcomere length: tetanic tension curve. a> The tension increases in two linear stages, from zero at sarcomere length 1.27 p to about 0.84 PO at 1.67 p and to I’,, at about 2.0 ,u, with the inflection point at 1.67 ,Umarking a change in the slope of the length: tension curve. b) There is a plateau region in the length: tension curve between about 2.0 p and 2.25 k in which PO is constant and there is maximum overlap of the thick and thin filaments; the length of the plateau of tension corresponds to the length of the central portion of the thick filaments that is devoid of cross bridges. c) There is a region of linear decrease in isometric tetanic tension from maximum to zero with increase in sarcomere length between 2.25 p and about 3.65 ,u; tension in this range of sarcomere lengths is proportional to the degree of overlap of thick and thin filaments. The length of the thick myosin filaments appears to be the same (1.6 p) in frog muscle and in various rnarnmalian muscles, including rabbit psoas (69, 70, 228, 32 1, 382), rat leg ~nusclcs (321, 419), and mouse biceps brachii muscle (167). In contrast the thin actin filaments, including the Z band, have different lengths in glutaraldehyde-fixed muscles of diflerent animal species; for example, the length of thin filaments has been reported to be 2.05 p in frog muscle (321), 2.1-2.2 p in mouse biceps brachii (167), 2.24 ,u in rabbit psoas (321), 2.3 ,U in rat leg muscles (321), and 2.55 p in human muscle (321). It is not known whether the lengths of thin filaments are different in red, intermediate, and white fibers, but electron micrographs of these muscle fibers in rat semitendinosus muscle reveal no obvious differences in this respect (152). According to the sliding-filament theory, maxirnum tetanic tension should be developed only when there is maximurn overlap of thick and thin filaments. The

142

R. I. CLOSE

Volume

52

few available results indicate that this is probably true for rat extensor digitorurn long-usand soleus muscles. The optimum initial length for twitch contractions is 2.7,5 ,u for rat extensor digitorum longus and 2.84 p for rat soleus (80) at 35 C, and these optimum lengths for twitch contractions are about O-5 % greater than those for tetanic con tractions. Allowing a 10 % correction for the combined effects of differences in optimum initial lengths for twitch and tetanic contractions and cx t-ension of the series-elastic elements during con traction, the optimum sarcomere length during isometric tetanic contraction would be about 2.5 p for rat extensor digitorum longus and 2.55 p for rat soleus, and these values are approximately the same as the optimum sarcomere lengths predicted from filament lengths. The optimum initial sarcomere length for maximum isometric tetanic tension is about 2.8 p for mouse biceps brachii (167), 2.7 p for cat flexor hallucis longus (60), and 2.8 j.4 (340) or 3.1 p (60) for cat soleus. These optimum initial sarcomere lengths are about the same as the sarcomere lengths of mammalian muscles in situ (124,

167, 224). The shape of the length: tetanus tension curve below the optimum length is similar in frog and mammalian muscles. Rack and Westbury (340) have estimated the sarcomere length: tetanic tension relation of cat soleus in situ; they found (340, Fig. 3) that tension increased from zero at about 1.5 p to P, at 2.8 p and there was an inflection in the curve at 0.8-0.9 P, and 2.2-2.3 p. Virtually identical results have been obtained by Bahler et al. (14, Fig. 2) for isolated rat gracilis anticus muscle at 17.5 C with the inflection at 0.8 P, and 0.8 L, . Similar inflections are also apparent in the length: tetanus tension curves for dog tibialis anterior and rat triceps surae muscles (418, Fig. 18) at 0.7 P, and 0.8 P, , respectively, but it is not possible to determine at what fiber length these occurred because in this instance only the relative muscle lengths were given. Gordon et al. (175) pointed out that the inflection at 1.67 p sarcomere length in the curve for frog muscle corresponds almost exactly with the length at which the ends of the thick filaments meet the 2 bands. In this connection it is important to note that the inflection occurs at an initial sarcomere length of about 2.2 ,u in rat gracilis anticus and 2.3 ,u in cat soleus; allowing for an extension of the series-elastic elements of 10 % of the initial length, these values would be reduced to only about 2.0 p and 2.07 p, respectively, and at these lengths the ends of the thick filaments would be at least 0.2 ,u distant from the 2 band. Rack and Westbury (340) suggested that tension at short lengths may be limited by something other than end-to-end apposition of the thick filaments at the Z-band level and that at short lengths the process of activation may function less effectively. This interesting observation is consistent with the results of recent work by Taylor and Riidel (400), who found that the activation of the central fibrils of frog fibers appeared to be terrninated during shortening when the sarcomere length reached 1.6 p, possibly as a result of interruption of the inward spread of the coupling process in the transverse tubular system in the center of the fiber. Preliminary experiments (unpublished work of author) on extensor digitorum longus and soleus muscles of the rat have shown that isometric tetanic tension decreases approximately linearly with increase in length above L0 , and extrapolation

Januarv

1972

DYNAMIC

PROPERTIES

OF

MUSCLE

143

of this part of the curve indicates that in these muscles the intercept at zero tension usually occurs at an initial sarcomere length only slightly greater than the combined length of the thick and thin filaments. In contrast, Carlsen et al. (69, 70) reported ATP-induced shortening of glycerol-extracted fibers of rabbit psoas muscle that had been stretched to sarcomere lengths up to 0.5 p beyond the length at which c overlap of thick and thin filaments is lost. This apparent contradiction of the slidingfilament hypothesis has been reviewed by Peachey (325), who put forward several possible explanations in defense of the hypothesis. Other more recent challenges (228,382) to the sliding-filament mechanism of contraction have been reviewed by that the essential features of the hypothesis had not Sandow (368), who concluded been disproved. The length: tension curve may be determined also by allowing a muscle to shorten isotonically to its full extent under different loads. Bahler et al. (14) determined the length: tetanic tension curves from isotonic and isometric contractions of rat gracilis anticus muscle at 17.5 C and found that these were the same if the initial length was less than 1.2 L, ; Gordon et al. (175) obtained similar results on single frog fibers. On the other hand, Geffen (154) found that the peak of the length: tension curve obtained from isotonic contractions of gastrocnemius muscle was only 0.92 of the optimum length for isometric contractions and suggested that the complicated arrangement of fibers may make this muscle unsuitable for this type of comparison. The maximum isometric tetanic tension of fast and slow muscle decreases with decrease in temperature for contractions elicited by either indirect or direct stimulation. The maximum tension decreases by about 30 % with change in temperature from 35-37 C to 20 C in cat (64, 92), rat (85, 407, 416), and mouse (149) muscles, and the change in tension is linear in some muscles and nonlinear in others. 2. Length : isometric twitch tension relation Even less is known about the length: tension relation for twitch contractions of mammalian muscle than for tetanic contractions. The optimum initial length for maximum twitch tension is usually 5-10 % greater than that for the tetanus at 35 C, but in some instances the two optima are the same (13, 58, 74, 340). Lewis and Luck (262) made the interesting observation that some motor units in cat flexor hallucis longus had optirnal lengths for twitch contractions that were markedlv different from those for tetanic contractions despite the fact that the two length optima are about the same for whole flexor hallucis longus muscles (58). Another important observation, made by Bahler et al. (13) on rat gracilis anticus muscle, is that the isometric twitch contraction time is independent of length below the optimum length for tctanic contractions but increases with increase in initial length between the optimum lengths for tetanic and twitch contractions (13, Figs. I 9 and 10). These observations indicate that the length: tension curve for twitch contractions is determined not only by the degree of overlap of thick and thin filaments, as described for tetanic contractions by Gordon et al. (175), but also

144

R. I. CLOSE

VOlUI~lP 52

by other factors that may involve length-dependent changes in activation. Similar but more pronounced effects have been observed in the responses of some frog sartorius muscles at 20 C (82, 98), and it has been shown that there is a welldefined length-dependent increase in the degree of activation of the contractile material of these muscles during twitch contractions when the sarcomere length is increased from about 2.1 p (at which there is maximum overlap of thick and thin filaments) to about 2.8 p (82). It is not known whether mammalian fast and slow muscles have different length: twitch tension relations but activation of these muscles during twitch contractions differs in other respects, such as ternperature dependence (64, 85) and the aftereffects of repetitive stimulation (45, 60, 84). In view of the possibility that there may be differences in the length dependence of activation of fast and slow mammalian muscles it will be necessary to determine the sarcomere length during contractions in future comparative work on the activation of these muscles. The ratio of maximum isometric twitch and tetanic tensions is about O-2-0.3 in most limb muscles of adult mammals at 35-37 C and is usually slightly higher for slow muscles than for fast muscles (a list of rcfercnces to twitch: tetanus ratios in some mammalian muscles is given in the legend of Fig. 4 in ref 79). Fast and slow muscles differ in the tcmpcrature dependence of peak twitch tension. When the temperature is decreased from 35-37 C to 20 C the peak twitch tension is almost doubled in fast muscles of the rnouse and rat (85, 149, 407) whereas it undergoes little or no change in rat soleus muscle (85). Maximum isometric tctanic tension decreases with decrease in temperature; consequently there is an inverse relation between twitch: tetanus ratio and temperature in the range 20-37 C. Similar, but not identical, effects of temperature on contraction of cat fast and slow muscles have been described (64). 3. Inlrinsic

strength of contractile material

The maximum isometric tetanic tension at optimum length is often expressed as the total force developed or the tension per gram of muscle, but for comparative purposes it is necessary to express the tension in terms of the amount of contractile material operating in parallel within the muscle. The simplest comparative Incasure, and the one most commonly used, is the tension per unit cross-sectional area of muscle. It is generally assumed that the specific gravity of the tissue is unity and the cross-sectional area of the muscle is taken as the muscle mass divided by the fiber length. This is a reasonable approximation because the mean density of dog and rabbit muscles is about 1.06 g/ml (287), but it should be renlembcred thit this measure of area gives no information about the density of packing of myofibrils within individual fibers and that it includes the extracellular space, which is about 8-14 % of the total area in various rat skeletal ~nusclcs (91, 141, 2 17, 250, 261, 264) and may be as high as 25 % in some mouse muscles (166). It is unfortunate that in some instances (416, 429) the tension per unit cross-sectional area has been estimated from the muscle length instead of thejber length; this would usually lead to an overestimate of the intrinsic strength of the contractile material because,

January

1972

DYNAMIC

PROPERTIES

OF

MUSCLE

145

with few exceptions, the overall length of a mammalian limb muscle is greater than the lengths of its fibers. The intrinsic strength has been underestimated in some instances. For example, the value reported for cat tenuissimus muscle (283) is probably low because the measurements were made on short lengths of muscle cut from the whole muscle and in these preparations many of the muscle fibers must have been transected and inactive. Some reliable measurements of tetanic tension per unit cross-sectional area have been made on rat and mouse muscles at 35-37 C; these include 1.62 kg/cm2 for isolated rat diaphragm (163), 1.57 kg/cm2 and 2.15 kg/cm2 for soleus muscles of male and female mice, respectively (360), 2.9-3.0 kg/cm2 for extensor digitorum longus of female rats (21, SO), and 1.9-2.1 kg/cm2 for solcus muscles of female rats (2 1, 80). The difference between the maximum tensions for rat extensor digitorum longus and soleus muscles is probably due to extrinsic factors that influence activation; it is not attributable to differences in extracellular space (250) or the amount of contractile material per unit mass of muscle (2 1), and the available evidence indicates that there is no rnarked difference in the intrinsic strength of the contractile material of fast and slow muscles. The results obtained by Sexton and Gersten (380) are interesting in this connection because they found that the ATP-induced isometric tension per unit cross-sectional area of myofibrils in small bundles of glycerol-extracted fibers was 1.8 kg/cm2 for fast medial gastrocnemius muscle and 2.28 kg/cm2 for slow soleus muscle of the rat at 27 C; this difference was statistically significant and they thus concluded that slow muscle is intrinsically stronger than fast muscle. Nevertheless, the earlier work of Sexton (379), in which the same method was used, gave a mean value of 2.4 kg//cm2 for gastrocnemius myofibrils, and the results of the two series of measurements taken together suggest that the force developed by a given amount of contractile material acting in parallel is approximately the same in fast and slow muscles. The maximum isometric tension per unit cross-sectional area of myofibrils has been determined for tetanic contractions of whole muscles (200, 360) and ATP-induced contraction of glycerol-extracted fibers (reviewed in 379 and 391) of rabbit, cat, and rat muscles; the usual value is several kilograms per square centimeter but the estimates range from 1.0 kg/cm2 to 49.1 kg/cm2. It seems likely that this great variation is the result of differences in methods of preparing the I,rather than large differences material for histological examination in intrinsic strength of the contractile material of different muscles. Dr. R. W. Rowe has kindly informed me that his estimate of 49.1 kg/cm2 cross-sectional area of myofibrils (360) f or mouse soleus muscle is about 10 times too high and that in this instance the total cross-sectional area of myofibrils was estimated using a mean myofibril diameter of 0.3 ,u (165) whereas more recent measurements show that it is approximately 1.0 p (169). D. Force : Velocity Properties of Contractile Component The relation between movements is qualitatively mammals.

load and speed of shortening in afterloaded isotonic the same in fully activated skeletal muscle of frogs and

146

R. I. CLOSE

Volume

52

The force:velocity relation for shortening has been determined from aftcrloaded isotonic tetanic contractions of some muscles of adult mammals including human muscle (208, 333, 43 1); cat muscles such as quadriceps (136), soleus (355), diaphragm (277), and tenuissimus (283); sloth diaphragm muscle (277) ; rat muscles including diaphragm (346), gastrocnemius (402), tibialis anterior (429), gracilis anticus ( 14), extensor digitorum longus (74, 80), and soleus (74, 80, 429) ; and mouse extensor digitorum longus and soleus muscles (76). The force : velocity curves for all these mammalian muscles are adequately described by Hill’s equation (207)

(P + a)V = b(Po - P)

( 2>

where LT is the speed of shortening, POis the maximum isometric tension, P is the load, and a and b are constants. The force:velocity data for mammalian muscle do not contribute new information on the basic mechanism of contraction or on the way in which extrinsic factors, such as the load, determine the speed of shortening. Nevertheless, the results are valuable because they show important differences between fast and slow muscles, and in this connection the three most interesting characteristics are a) the intrinsic strength of the contractile material, i.e., POat V = 0 (discussed in sect. IIIC), b) the intrinsic speed of shortening, i.e., V at P = 0, and c) the general shape of the force : velocity curve as indicated by the ratio a/PO. In most of the published work on force: velocity properties of mammalian muscle the maximum speed of shortening at zero load is expressed as the speed of shortening of whole muscle either in absolute units or as muscle lengths per second. These measures of speed are of little use in estimating the properties of the contractile material unless they are converted to speed of individual fibers or sarcomeres. Moreover, the measure fiber lengths per second is only useful as a comparative measure of intrinsic speed of shortening when the muscles have the same sarcomere length at maximum overlap of thick and thin filaments. The best measure of speed of shortening is the shortening velocity of single sarcomeres expressed in absolute units of length and time, assuming that shortening is the result of sliding of filaments; if the speed of shortening is expressed in this way it is possible to compare the intrinsic speeds of different muscles with different sarcomere lengths, to estimate the rate at which thick and thin filaments move past one another during shortening, and to determine the relations between dynamic and enzymic properties of the contractile material. This method has been used on mouse, rat, and cat muscles and it has been established that a) the intrinsic speed of fast and slow muscles is inversely related to body size of adults of different species (75, 79), b) there are developmental changes in the force-velocity properties of fast muscles (74, 76, 83), c) in adult animals the fast muscles have a higher intrinsic speed of shortening than slow muscles (74, 76) and that this difference may be reversed by cross-uniting the motor nerves of these muscles (77, 80), and d) that the intrinsic speed of shortening of sarcomercs of normal and cross-innervated muscles is directly proportional to the specific activity of actin-activated ATPase of myosin (21); these points are discussed in sections IV, V, and VI.

DYNAMIC

PROPERTIES

OF

MUSCLE

147

The greater the curvature or concavity of a force-velocity curve the lower the ratio a/P0 when the data are described by Hill’s equation (207). It has generally been found that Q/P, for mammalian muscles is in the approximate range 0.1% 0.30 and that it is higher for fast muscles of adults than for slow ones (74, 80, 429). Differences in alp, may be a special adaptation of individual fibers that is important in connection with the functioning of muscles either as phasic ones, suitable for rapid powerful movements, or as tonic muscles, suitable for sustained maintenance of tension with minimal expenditure of energy; this aspect is discussed in set tion VIID. The ternperature dependence of speed of shortening has not been investigated extensively. The Q-, for the intrinsic speed of shortening of neonatal rat extensor digitorum longus muscle in vitro is 1.68 in the temperature range 25-35 C (75). Comparable data for adult muscle are not available but indirect evidence suggests that the temperature coeficient is probably about the same in adult and neonatal fast muscles. For example, the maximum speed of shortening of fast muscles of adult Wistar rats is 7-8 fiber lengths per second for gracilis anticus (14, Fig. 5) at 17 5 C and 18.7 fiber lengths per second for extensor digitorum longus (74) at 35 C; this difference in speeds corresponds to an average Q10 of about 1.67. I?. Behavior of Series-Elastic and Contractile in Isotonic and Isometric Contractions

Components

There seems to be little doubt that mammalian muscle contains a contractile component in series with a lightly damped elastic component and that these two components arc largely distinct functionally but are not necessarily separate morphological entities. The simplest model for contraction of fully active muscle at maximum filament overlap is one in which the response is determined by the load : extension properties of the series-elastic component and the force :velocity properties of the contractile component, as described by equation I, and in which the characteristic relations of these two components are the same for isotonic and isometric contractions. It appears that this model is an oversimplification of the properties and behavior of reptilian, amphibian, and mammalian muscles. Several attempts have been made to test the model by comparing observed and predicted values for the three variables in equation I. Katz (243) and Jewel1 and Wilkie (231) found that the estimated values for rate of rise of tension in an isometric contraction (dP/dl) were considerably higher than the observed rates for fully active tortoise and frog muscles. Two complementary investigations have been carried out on mammalian muscle: one by McCrorey et al. (283) in which it was shown that the estimated compliance of the series-elastic component of cat tenuissimus muscle was twice as great as the observed values obtained by the controlled-release method (sect. IX&), and the other by Parrnley et al. (322) in which it was found that the estimated speed of shortening (dx/dt) of the contractile component of rat soleus muscle was less than that obtained from isotonic contractions. At the present time there is no satisfactory explanation for these discrepancies. One possible explanation that may

148

R. I. CLOSE

Volume 52

account for part of the discrepancy is that, at particular load and length, dx/dt of the contractile material as a whole is less in isometric con tractions than in isotonic contractions. In this connection it is interesting to note that the 1ength:velocity curves obtained by Parmley et al. (322) sh ow an almost instantaneous rise in dx/dt during the transition from isometric to isotonic contraction, with the maximum value corresponding to that predicted from equation I, whereas the full velocity of shortening was reached much later; similar curves have been obtained by Bahler and his associates (12, 14) in which the peak isotonic shortening velocity appears to occur after a much longer period of acceleration than that which would be required to accelerate the lever system if the muscle were in a fully effective contractile condition at the onset of shortening. Several observations indicate that at any instant in a tetanic contraction the speed of shortening of the contractile material depends on factors related to previous activity of the muscle as well as force, length, and temperature. For example, Rosenblueth and Rubio (355) showed that the speed of shortening of cat soleus muscle against a moderate load was lowered as the initial length was increased above the optimum length; similar but much smaller, effects have been observed in frog muscle shortening against light loads (175, 350). Bahler et al. (14) confirmed these observations on rat gracilis anticus muscle and showed that shortening against light loads is almost independent of initial length, whereas for moderate and heavy loads the velocity decreases markedly with increase in initial length (14) ; they also confirmed the observation made by many earlier workers that the speed of shortening and rate of tension development are dependent on the frequency and duration of stimulation. Bahler was able to quantitate these effects of stimulation on shortening velocity and to develop a computational method for determining time-independent length: velocity curves; one of the most interesting observations arising from this work was that for a given load these theoretical curves were independent of initial length. On the basis of these results Bahler et al. (14) concluded that shortening velocity of rat gracilis anticus is a function of load, length, temperature, and an additional variable related to the time from onset of stimulation. Other factors also influence the speed of shortening. Rosenblueth and Rubio (358) and Joyce and Rack (235) f ound that dx/dt or dp/dt of cat muscle are higher if a muscle reaches a given length and load isometrically rather than isotonically, even when the time from onset of stimulation and the frequency of stimulation are held constant (235). S imilar observations have been made on frog muscle in the transitions from isometric to isotonic contractions and vice versa (326). The effects of one kind of mechanical activity on another and the transitions from one kind of contraction to another are difficult to determine, and at the present time there is insuflicien t quantitative information available to include these effects in analyses of the behavior of whole muscle in any but the simplest experimental conditions. The observations just described are valuable because they reveal some hitherto unknown properties of the contractile material, but much more work is needed to characterize all the mechanical components of

January

1972

DYNAMIC

PROPERTIES

OF

149

MUSCLE

active muscle and to determine precisely the way they interact in different conditions. Several investigators have studied the behavior of mammalian muscle in more complex conditions and have described the effects of various mechanical interventions on the isometric and isotonic contractile responses to different kinds and frequencies of stimulation (235, 236, 323, 339); in some instances the. experimental conditions resembled those that exist during normal limb movemen t and in this respect the results have an important bearing on the function of mammalian muscle in situ. F. Actiue State, Time

Course of Isometric

Twitch,

and Fosttetanic

Potentiation

1. Active state Interpretation of the active state of mammalian muscle is based almost completely on methods developed for studying this property of amphibian muscles. In some instances the original interpretations of results obtained for frog muscles have been revised or the method has been found to yield equivocal results. It may therefore be helpful to review some interpretations of the results of experiments on the active state of frog muscle along with the results obtained for mammalian muscle. Gasser and Hill (15 1, 209) applied quick stretches at different times during isometric twitches of frog muscle at 0 C and found that an appropriate stretch, which presumably fully extended the series-elastic component, enabled the muscle to bear a load equal to the maximum isometric tetanic tension within about 40 msec after a single stimulus. Hill (209) proposed that, at a given temperature and length of contractile component, the time course of the isometric twitch is determined partly by the force:velocity properties of the contractile material and the 1oad:extension properties of the series-elastic component and partly by the time course of development and decline of the active state of the contractile component as a whole. Hill defined the intensity of the active state as the tension that the contractile component could develop, or the load it could bear, without lengthening or shortening, and it follows from this that the active-state curve after a single stimulus is simply the time course of the twitch response of the contractile material in the absence of the series-elastic component and in strictly isometric conditions. The observation of Hill (209) that frog sartorius behaves as though it is fully activated for a brief period after a single stimulus at 0 C was substantiated when it was shown that the relation between rate of development of tension and tension [(dP/dt) : P curve] for the isometric twitch was the same as that for development of tension in the fully active muscle after release during a tetanus; the (dP/dl) : P curves for the two responses coincided for a few milliseconds about 45 msec after the stimulus for the twitch (73). However, none of these observations excludes the possibility that in certain conditions some frog sartorius muscles are only partially activated at 0 C, and there is evidence that some frog muscles arc not fully activated at low temperature (37 1). Walker (415) studied the effects of quick stretch on contractions of rat gas-

150

K.

I. crx-xx

Volume 52

trocnemius muscle; one of the principal findings was that a stretch from suboptimum length to optimum length caused an increase in peak twitch tension that was qualitatively similar to that found by Hill (209) for frog muscle. It is important to note that Riiegg and his colleagues (362) have dernonstrated quick-stretch-induced mechanical activation of the contractile material of glycerol-extracted fibers of rabbit psoas and frog semitendinosus muscles, and this property of the contactile system would have to be taken in to account in any attempt to interpret the effects of quick stretch on contractions of whole muscles. Subsequent work by Hill revealed that the maximum velocity of shortening of frog muscle against very light loads is developed shortly after the onset of contractions at 0 C (212). Hill (212) p ointed out that if maximum velocity of shortening against very light loads is an indication of the fully active state then the transition from rest to full activity must be extremely rapid, but objections have been raised against this argument (73, 222) when it is applied to the whole contractile component of individual fibers. Furthermore, it has been reported that the intrinsic speed of shortening and the relative force:velocity relation of skinned muscle fibers are independent of pCa in the range 5.0-6.75 (335) even though the myosin ATPase activity (426, 427) and maximum isometric tension (20 1) are half-maximal at pCa 6.75 and maximal at pCa 5.0. Ritchie (346) f ound that the isotonic shortening velocity against minimum load was the same for twitch and tetanic contractions of rat diaphragm and concluded that the full intensity of the active state is reached shortly after the onset of contraction; this interpretation is the same as that of Hill for frog muscle (212), and in both instances the interpretation can be criticized because the results are equivocal (73, 222). MacPherson and Wilkie (275) d eveloped a method for comparing the time courses of the first derivatives [(dp/dt) : t] of the tension: time curves of an isometric twitch and either an isometric tetanus or the double response to two appropriately timed stimuli. They found that the earliest separation of the curves for twitch and double response of frog sartorius muscle was about 44 msec after the first stimulus at 0 C. MacPherson and Wilkie interpreted this result as meaning that after a single stimulus there is an abrupt change from zero to full activity at about the time of onset of the twitch contraction, followed by a plateau of full activity that ends about 44 msec after the stimulus with the onset of decline in the intensity of the active state (275, Fig. 1). Later investigations revealed that the intensity of the active state increases gradually to a peak about 40 msec (73) or more (231) after a single stimulus at 0 C and that in normal conditions there is no plateau on the active-state curve. It is important to keep in mind that the method of MacPherson and Wilkie determines the time at which there is separation of the active-state curves of two different types of responses; it does not rneasure the intensity of the active state. The method of MacPherson and Wilkie (275) has been used to determine the earliest point of separation on the tension-time curves of twitch and double responses of skeletal muscles of mouse (149), rat (302, 430), cat (34, 59, 283),

January

1972

DYNAMIC

PROPERTIES

OF

MUSCLE

151

dog (280), and man (96). The mean values reported for the period from onset of contraction to point of separation range from about 1.5 msec to 4.0 msec at 35-37 C, and there appears to be no correlation between the time of separation and the isometric twitch contraction time. However, the limits of resolution oi the times of onset and separation of contractions are determined largely by the sensitivity of the recording method, the compliance of the preparation, and the method of stimulation, all of which were different in the investigations listed above. Although the results obtained for different muscles are not comparable it is clear that in all the mammalian muscles examined the tension : time curves for the twitch and double response coincide for a short period after the onset of contraction. There are two main possible interpretations of results obtained using the method described by MacPherson and Wilkie (275). The first interpretation is that the muscle is fully active during the period between onset and separation of the contractions. In evaluating this view it should be clearly understood that coincidence of the tension : time curves merely means that the mechanical response of the muscle has not been altered by the second stimulus, and in the absence of additional information it does not provide a measure of the actual intensity of the active state. There is certainly no evidence that mammalian muscle fibers are fully activated at any stage during a normal twitch. Moreover, the observation that there is more rapid development of tension in potentiated twitches (below) from the onset of contraction, rather than from a time corresponding to the point of separation of curves obtained by the MacPherson and Wilkie method, is not consistent with the first interpretation. The second possibility is that some part of the muscle fiber involved in excitation-contraction coupling has a refractory period that determines the earliest time at which the contractile mechanism can be reactivated irrespective of the intensity of the active state. There are a number of observations consistent with this second interpretation. For example, Martensson and Skogland (280) found that the shortest interval between conducted muscle action potentials of directly stimulated dog laryngeal muscles was 1.5-2.0 msec shorter than the period of coincidence of tension : time curves of isometric twitch and double response. They point out that this difference may be due partly to an overestimate of the period of coincidence, resulting from inaccuracy in determination, and partly to delay of the second response due to refractoriness in the excitation-contraction coupling process during a period exceeding that of the refractory period of the muscle membrane; a similar suggestion has been put forward by Norris (302) for results obtained from massively stimulated rat limb muscle. A similar set of observations is available from two independent investigations on indirectly stimulated cat muscles. Eccles and O’Connor (116) determined the minimum separation time of conducted muscle action potentials and found that these ranged from about 2.5 msec to 3.2 msec for fibers of the cat soleus muscle and that there is a minimum value of 2.2 msec for the cat tibialis anterior muscle. Bullcr and Lewis (59) obtained miniresponses of about 2.9 mum separation times for single and double mechanical msec for cat soleus muscle and about 2. 4 msec for cat flexor hallucis longus muscle.

152

R.

I. CLOSE

Volume

52

The combined results from work on indirectly stimulated cat muscles (59, 116) suggest that the limiting property that determines the earliest time the muscle can respond to a second stimulus is probably the refractory period of the outer membrane of the muscle fiber. Desmedt and Hainaut (96) obtained the same result on indirectly stimulated human adductor pollicis muscle and showed that the intervals between muscle action potentials and the time from onset to separation of twitch and double contractions were identical for all stimulus intervals between about 1 and 5 msec; the minimum interval between indirectly elicited muscle action potentials of adductor pollicis was virtually the same as the absolute refractory period to direct stimulation of one of two groups of human muscle fibers (13 1). There is therefore very strong evidence that the refractory period of some part of the muscle determines the earliest time of separation of twitch and double-response contractions. According to this interpretation the curve relating the stimulus interval (abscissa) and the duration of the period from onset to separation of contractions (ordinate) should have slope = 0 for stimulus intervals less than the refractory period of excitation contraction coupling and slope = 1 at greater stimulus intervals. This type of curve is not obtained when indirect stimulation is used because there is a marked reduction in conduction velocity of the second nerve action potential at short stimulus intervals (116; Ridge cited in 59, 96); consequently the curve obtained by indirect stimulation has a negative slope at short stimulus intervals and a positive slope = 1 at stimulus intervals greater than the refractory period for reactivation of the contractile material (e.g., 59, Fig. 8; 96, Fig. 1). In 1954 Ritchie (347) reported that the onset of decline of tension at the end of an isometric tetanic contraction occurred about 35 msec after the last stimulus at 0 C, and he concluded that there is a true plateau on the active-state curve after the last stimulus in a tetanus at that temperature. This method has been used by Wilander (430) and Bleckmann et al. (34) and the results obtained for mammalian muscle were qualitatively similar to those for frog muscle. The time course of decline of the intensity of the active state was determined bv, Ritchie and Wilkie (348, 349), who devised methods for discharging the tension in the series-elastic component at different times after a single stimulus, or the last of a train of stimuli, and recording the time course of redevelopment of tension. These methods yielded a family of twitch-like curves for redeveloped tension, and the peaks of these curves were, by definition, points on the falling phase of the active-state curve. The quick-release method of Ritchie and Wilkie (348, 349) has been used by Wells (429), McCrorey et al. (283), and Jurna et al. (238) on muscles of various mammals, and the results have shown that the time course of decline of the active-state curve is more rapid for fast muscles than for slow ones, as would be expected from the general observation that there are marked differences in contraction time of these muscles but little or no difference in the twitch: tetanus ratio (75, 79). Furthermore, the results show that the behavior of fast or slow mammalian muscles after a quick release from one length to another length during relaxation in an isometric contraction is essentially the same as that for frog muscle at 0 C.

January

1972

DYNAMIC

PROPERTIES

OF

MUSCLE

153

The active states of frog muscles at low temperature (O-10 C) and either fast or slow mammalian muscles at 35 C differ in at least one important respect; this is shown most clearly by differences between the responses of these muscles to a second stimulus delivered during the contraction phase of an isometric twitch. The peak tension of a double isometric response of frog muscle to two stimuli at 10 C is greater than that of the twitch response but the time interval between the second stimulus and the peak of the double mechanical response is approximately equal to, or less than, the isometric twitch contraction time (196, Fig. 7); this is the kind of response one might expect to observe if the muscle were fully or nearly fully activated by a single stimulus and if the time course of decay of the active state were the same after a single stimulus and after the second stimulus of a pair of stimuli at short interval. In contrast, the peak tension of a double response of mammalian muscle at 35 C is 2-3 times higher than the peak twitch tension, and the time interval between the second stimulus and the peak of the double mechanical response is about 1.5 times greater than the isometric twitch contraction time; these observations were made first by Cooper and Eccles in 1930 (89) and their results on cat fast and slow muscles have been confirmed (67, 357). Similar results have been obtained for dog laryngeal muscles (280) and the extensor digitorum longus and soleus muscles of the rat (unpublished observations of the author). In mammalian muscle the peak tension and time course of the double response depend on the interval between stimuli (89) and the effect is maximal when the interval between the two stimuli just exceeds the refractory period for excitation of the muscle (116). The double response can be elicited by direct or indirect stimulation and the increment in peak tension is not due to recruitrnent of fibers or repetitive firing of individual fibers (116). It may be coneluded from these observations that the time course of decline of the active state after the second of two closely spaced stimuli is very much more prolonged than after a single stimulus at 35 C. One other point of interest in connection with the double response is that the ratio of peak tension of the largest double response to peak tension of the twitch appears to be inverselv correlated with twitch: tetanus ratio; it is about 2 for 1i mb muscles with twitch: tetanus ratios about 0.20.25, whereas it is about 3 for dog thyroarytenoid (280) and cat internal rectus (89) muscles with twitch: tetanus ratios about 0.1 or less. The qualitative differences between the double mechanical response of mammalian muscle at 35 C and that of frog muscle at 0 C do not necessarily mean that the muscles are intrinsically different. It is possible that mammalian muscle at about 15-20 C behaves like frog muscle at 0 C and that frog muscle at 20 C behaves like mammalian muscle at 35 C; the results obtained by Fulton (146) support this view. The results of other investigations on relaxation of cat (357) and rat (303) muscles show that there may be further prolongation of the active state that is proportional to some factor related to the duration of repetitive stimulation and the number of stimuli, and this effect appears to be similar to that found in frog muscle at low temperature (349). In summary, fairly reliable methods have been developed for study of the active state of vertebrate twitch muscle fibers to determine a) the time of divergence of the active-state curves of two different kinds of isometric response (275),

154

R. I. CLOSE

Volume

52

6) whether the maximum intensity of the active state after a single stimulus is equal to or less than the fully active state (73), and c) the time course of decline of the active state (348, 349). A reliable method for determining the time course of development of the active state in a twitch has not yet been devised (87, 120, 294) . In addition to experimental investigations on the active-state curve there have been several theoretical analyses of different muscle models that deal with various aspects of the active state of frog (237, 399) and mammalian (13) muscle. Bahler et al. (13) have proposed that the active contractile component of mammalian muscle develops force that is partly transmitted through the series-elastic component and partly dissipated internally as an inverse function of the velocity of the contractile component. They describe a method for calculating the time course of part of the active-state curve from the experimentally determined tension : time curve of the isometric twitch, the load : extension properties of the serieselastic component, the force:velocity relation of the contractile component, and a set of assumptions. According to their definition of the velocity: load characteristic of the internal load, none of the force developed by the contractile material is dissipated across the internal load in conditions of zero velocity such as occur at the peak of the isometric twitch. It has been pointed out above that the isometric twitch is length dependent and has an optimum length a little greater than that for tetanic contraction. It is therefore not surprising that the calculated time course of the active state should show a corresponding length dependence. Consequently the claim by Bahler et al. (13) that predictions concerning the length dependence of twitch contractions based on the theoretical active-state curve are verifiable experimen tally does seem to involve circular argument. 2. Time course of isometric twitch The properties of the system that regulates the activity of the contractile material during the twitch contraction-relaxation cycles have been thoroughly reviewed by Sandow (367, 368), Ebashi et al. (109, 1 lo), and Weber (423). It is now generally accepted that in the resting state the troponin-tropomyosin complex inhibits interaction between actin and myosin in such a way that it prevents hydrolysis of ATP and contraction. Excitation causes release of calcium ions from the sarcoplasmic reticulum into the sarcoplasm and transition from resting to active state in vivo is thought to involve the release of the contractile material from the preexisting inhibitory influence of the troponin-tropomyosin complex as a direct result of the binding of calcium ions to troponin. Relaxation is brought about by removal of calcium from the troponin-tropomyosin complex on the contractile material, and the final stage in the cycle is the uptake of calcium by the sarcoplasmic reticulum. Although it appears fairly certain that the main event in the contraction-relaxation cycle is an exchange of calcium between the sarcoplasmic reticulum and troponin, it has not been established which process determines the time course of decline of the active state. In any event it is to be expected that there is approximate matching of the kinetic properties of the

January

I972

DYNAMIC

PROPERTIES

OF

155

MUSCLE

contractile and sarcoplasmic reticular components to ensure proper functioning of the muscle fibers during sustained repetitive activity. Other aspects of the role of (:a2f in excitation-contraction coupling in mammalian skeletal muscles have been discussed by Buss and Frank (68), Steward and Levy (393), and Weber

(424). It was mentioned above that the time course of the active state is different in fast and slow muscles of a given species and is also different in homologous muscles of different species. One interesting aspect of the comparative physiology of the active state of mammalian muscle is the observation that the twitch: tetanus ratio, or the peak twitch tension per unit cross-sectional area of muscle, is about the same for fast and slow limb muscles of the same and different animal species despite marked differences in the isometric twitch contraction time (75, 79, 8 1). Furthermore, there are well-defined hyperbolic relations between the isometric twitch contraction time (T,) and either the intrinsic speed of shortening of sarcomeres (VT^“) (75, 79, 81) or the specific activity of myosin ATPase (d&/dt) (see sect. v and VI). That is, max

Vs

T,

= k,

( .3)

and

K

max

(4)

= k2 (dPi/dt) l

where kl and k2 are constants. It is notable that each of these constants has approximately the same value not only for fast and slow muscles of a particular animal species but also for homologous or similar muscles of different species; there are indications that T, and hence kl in equation 3 are inversely related to a/l’, (80) but this has not been investigated thoroughly. If it is assumed that the properties of the series-elastic component and the amount of activator reaching a given amount of contractile material during a twitch are about the same for fast and slow muscles of different mammals, then it follows that the speed of contraction and the time course of decline of the active state are coupled in some way. This relation has been discussed earlier (74, 75) and it was pointed out that, whatever the nature of the processes of excitation-contraction coupling, it is likely that the duration of the active state after a single stimulus is determined in one of two general ways. One possibility is that the contractile material may be allowed to operate for a certain time and that the duration is controlled by some extrinsic process (e.g., 214); in this case it is necessary to postulate that the intrinsic speed of shortening and the time course of the active state are coupled indirectly, either inherently, through evolutionary processes, or functionally,

speed of shortening

(74).

These

antithetical

possibilities

must

be considered

in

156

R. I. CLOSE

Volume

52

is an exarnple of extrinsic control of the duration of activity. The amount of calcium that must be sequestered in vivo in the transition from full activation to complete relaxation has been estimated to be about 0.1-W pmoles/g of muscle (423), and measurements on the yield of fragments of sarcoplasmic reticulum isolated from whole muscles and their maximum calcium capacity indicate that the sarcoplasmic reticulum of fast and slow muscles has the capacity to reduce the concentration of free calcium ions in the sarcoplasm to a level below the threshold for contraction (110, 139, 423). Various kinetic studies on fragmented sarcoplasmic reticulum in vitro have shown that the maximum rate of uptake of Ca2f by fast muscle fragments is approximately 4-l 1 times greater than the rate for slow muscle fragments (139, 194, 385, 386, 441) at room temperature. Furthermore, there appears to be twice as much sarcoplasmic reticulum in fast muscles compared with slow muscles, at least as indicated by the yield of fragments of reticulum per unit weight of muscle (139, 194) and the volume of the reticulum as assessed from electron micrographs (271). These results, together with the kinetic data referred to above, indicate that fast muscle sarcoplasmic reticulum may accumulate calcium 8-22 times faster than slow muscle reticulum in vivo. As mentioned earlier, the twitch-tetanus ratio is about 0.2-0.25 for most mammalian limb muscle at 35 C and the isometric twitch contraction time is therefore a reasonably good indicator of the duration of the active state at about 0.2 PO to 0.25 P, . The point of interest here is that the isometric twitch contraction times of these muscles differ by a factor of about 3 or 4 (79), and this means that the relation between the duration of the active state in vivo (35 C) and the rate of uptake of calcium by fragments of sarcoplasmic reticulum in vitro (20-25 C) is not the same for fast and slow muscles : the temperature coefficient for rate of uptake of calcium is about 2 for fast and slow sarcoplasmic reticulum (385, 427). Despite this difhculty, there is a rough correlation between rate of uptake of calcium by fragments of sarcoplasmic reticulum in vitro and the intrinsic speed of contraction; the values for rate of calcium uptake (in pmoles Ca”f/nlg fragment of sarcoplasmic reticulum per minute) are 0.28 for human muscle (438), 1.8 and 0.16 for rabbit fast and slow muscle, respectively (385), 1.2 and 0.3 for guinea pig fast and slow muscle, respectively ( 139)) and 2.8 for mixed fast and slow muscle of the rat (438), all measurements having been made at room temperature and in similar conditions of stirring. These values are roughly inversely related to body size and consequently are directly related to muscle speed (see below). Nevertheless, it has been pointed out that the highest reported value for maxirnum rate of calcium uptake by fragments of sarcoplasmic reticulum of fast muscles in vitro is considerably lower than the rate that would correspond to the time course of decline of the active state in vivo (109, 1 lo), and the discrepancy would be even greater for slow muscle reticulum. This discrepancy does not disprove the hypothesis that the time course of decline of the active state is determined by the rate of uptake of calcium by the sarcoplasmic reticulum because it can be argued that a large proportion of the calcium transport sites may be inactivated in vitro. Furthermore, Weber (425) has reported that detergent-treated fragments of sarcoplasrnic reticulum of rabbit muscle do not accumulate calcium and that the

January

1972

DYNAMIC

PROPERTIES

OF

MUSCLE

157

rate of hydrolysis of ATP by transport enzyme, and presumably the rate of transport of calcium, was 2.5 times higher for solubilized fragments than the initial rate for intact vesicles. These results were interpreted as meaning that calcium present in the interior of in tact vesicles inhibits calcium uptake in vitro (425). Weber also pointed out that if rates of hydrolysis of ATP in the presence of detergents are representative of the initial rates without calcium inhibition, then the rate of uptake of calcium in predominantly fast muscle of the rabbit may be as high as 10-l 2 pmoles Ca2f/mg fragment of sarcoplasmic reticulum per minute in viva at 24 C (425). n -evertheless, it is still not clear whether these results are consistent with the hypothesis that the uptake of calcium by the sarcoplasmic reticu lum in vivo determines the time course of decrease in sarcoplasmic calcium ion concentration to sub threshold levels. Apparently it will be necessary to develop methods for determining the time course of change of concentration of calcium ions in the sarcoplasm, and the rate of uptake of calcium in vivo (e.g., 334), in order to obtain an accurate assessment of the role of the sarcoplasmic reticulum in relaxation. The second possibility is that there is an undiscovered Yntrinsic” control of the time course of decline of the active state of the contractile material, for a given amount of calcium released into the sarcoplasm. It is conceivable, for exc ample, that during contraction of intact fibers there is an exchange of calcium between sarcoplasm and a calcium binding site, such as troponin, and that the displaced calcium is in some form that prevents it being bound again to the contractile material but does not prevent it being accumulated in the sarcoplasmic reticulum. It is of interest to note that myofibrils undergoing syneresis in vitro contain bound calcium that is in equilibrium and readily exchangeable with the ionized calcium in the medium (426), but it should be clearly understood that the suggestion that this exchange leads to release of calcium in complex form in whole fibers is supposition. Even if such a turnover were shown to occur it is still the intri nsic speed of possible that there may be no causal connection between shortening and the time course of the active state, except through evol utiona .rY matching. Nevertheless, it is possible that there is a feedback mechanism, as yet undiscovered, whereby the rate of splitting of ATP by actomyosin, or another intrinsic property that is correlated with speed, determines the rate of disappearance of free calcium ions and the decline of the active state. If this occurred the muscle would continue to contract until the pCa had been raised to the threshold for contraction, then relaxation would commence; this behavior would be consisten t with the observations on other muscles that the calcium transient occurs in the contraction phase of the isometric twitch and that pCa of the sarcoplasm appears to be raised to the resting level at the peak of the twitch (8, 232, 345). It may be relevant that relaxation does not simply depend on removal of calcium from troponin but also depends on a relationship between the extent of calcium binding by troponin and ATP binding by myosin (424). The hypothesis that there is intrinsic control of the ac tivc state of the contractile material is based mainly on supposition and is merely an extension of ideas put forward about 6 years ago in discussions on the relation between intrinsic speed of shortening and the dura-

158

K. I. CLOSE

volume

52

tion of the active state (74, 75). Nevertheless, these ideas are worth considering as long as there is still difficulty in demonstrating that the time course of decline of the active state is determined by the rate of uptake of calcium by sarcoplasmic reticulum. AYIost of the papers dealing with characteristics of isometric twitch contractions are referred to in various sections of this review; some others containing useful information on twitches include references 32, 33, 46, 47, 227, 263, 288, 356, 369, and 370. 3. Posttetanic potentiation Repetitive stimulation of mammalian muscle leads to a transitory increase in the peak tension of the isometric twitch of adult fast muscle, but usually causes either no change or a small transitory decrease in the twitch of adult slow muscles. These effects reveal fundamental differences between fast and slow muscles that are important in connection with excitation-contraction coupling and the active state of these muscles. Maximum posttetanic potentiation (PTP) causes up to twofold increases in the peak twitch tensions of adult fast muscles of mammals at 35 C. There have been conflicting reports that the contraction and relaxation phases of the “stairpotentiated twitch are increased, decreased, or unaltered, case” or posttetanic but these differences are probably mainly attributable to differences in method and pattern of stimulation. Not all the mammalian muscle preparations that have been used are suitable for studying all aspects of PTP. The best preparation is an isolated single muscle fiber stimulated simultaneously at all points along its length, but this has not yet been employed for studying this property of mammalian muscle. Small extensor digitorum longus muscles from juvenile rats behave as though all the muscle fibers are activated in a normal twitch when the curarized muscle is stimulated massively in vitro, and this particular preparation has been useful for determining the relations between degree of potentiation, the the time course of disappearance of PTP, time course of potentiated twitches, and the frequency and number of conditioning stimuli (84). It has been found that there are two distinct phases of potentiation for a given frequency of stimulation (84, Fig. 5). In the initial phase of potentiation an increase in the number of stimuli increases the degree of potentiation up to a maximum with little or no change in the twitch contraction time and only a small decrease in half-relaxation time. In the second phase a further increase in the number of stimuli causes prolongation of the contraction and relaxation phases during c the period of potentiation. There has been considerable interest in the cause of PTP. It has been established that neither staircase potentiation nor PTP of peak twitch tension of whole fast muscles is the result of recruitment of muscle fibers (30, 45, 96) or repetitive firing of muscle fibers (30, 37, 45, 96, 383, 390) and that the potentiation may be induced by indirect stimulation of normal rnusclc or direct stimulation of curarized or denervated muscle (37, 45, 84, 86, 306) ; however, prolonged

January

1972

DYNAMIC

PROPERTIES

OF

MUSCLE

159

indirect stimulation of some slow muscles, such as cat soleus, causes repetitive firing of individual muscle fibers after a single nerve volley (see 38, 389, and 390 for references to earlier work). The rate of change of tension with respect to time (dY/dt) in an isometric tetanic contraction elicited during a period of PTP is higher at a given tension than the rate in an initial control tetanus, but only for a brief period’ shortly after the onset of contraction; in other respects the two tetanic contractions are virtually identical provided the muscle is stimulated at the optimum frequency (37, 96, 306). Consequently, PTP of peak twitch tension is not due to a transitory increase in the intrinsic strength of the contractile material. Furthermore it is very unlikely that PTP of peak twitch tension is due to increased stiffness of the series-elastic component because dP/dt in a maximally potentiated twitch is increased approximately twofold for all tensions during the twitch contraction phase up to about 0.35 P, to 0.4 P, at 35 C, whereas no comparable increase has been observed in tetanic contractions elicited during PTP. Nevertheless, direct measurements should be made to determine whether the properties of the series-elastic component are altered in any way during repetitive stimulation and during PTP. The other main possible interpretation is that PTP of the isometric twitch results from a change in the active state after a single stimulus. Two of the most interesting aspects of PTP of mammalian fast muscle are (1> that it is possible to almost double the peak twitch tension with virtually no change in the time course of the contraction phase and b) that there is divergence of the tension-time and (dp/dt) :P curves of control and potentiated twitches from the onset of contraction (37; 84, Fig. 3; 96, Fig. 3; 191, Fig. 1). If the properties of the series-elastic component are the same in normal and potentiated muscles, then it is probable that the intensity of the active state during the contraction phase is greater in twitches elicited during periods of PTP than in normal conditions. Rosenfalck (359) has pointed out that none of the observations on PTP of marnmalian muscle excludes the possibility that the contractile material is in the fully active state in both normal and posttetanic potentiated twitches at the end of the latent period and that the difference between the two contractions may be determined by differences in the time course of decline of the active state. Nevertheless, it is hard to imagine how in the course of an ordinary control twitch the active state could undergo a sudden change at the end of the latent period from maximum intensity of the active state to an intensity that would allow only about half-maximum rate of development of tension at very low loads. The alternative interpretation is that mammalian fast muscle fibers are only partially activated during normal twitch contractions at 35 C, and several investigators have emphasized the possibility that PTP results from an increase in the number of parallel contractile elements that are active at any instant during the contraction phase of the twitch (75, 84, 86, 96, 306). S ome of the possibilities that must be considered in this connection are that PTP results from n) an increase in the number of active myofibrils due to increased inward spread of excitation in the transverse tubular system, b) an increase in the number of active filaments within fibrils or the number of active cross bridges < between actin and myosin filaments

160

R. I. CLOSE

Volume 52

due to increase in the amount of ‘Cactivator” (presumably Ca2+) released from the sarcoplasmic reticulum, and c) a combination of a and 6. There is no evidence excluding the possibility that increased spread of excitation in the transverse tubules contributes to PTP in mammalian muscle, but one of the simplest explanations for the two stages of PTP in mammalian fast muscle at 35 C is that the potentiation is due to gradation of the amount of activator released in all regions of the fiber (84). According to this hypothesis, all the actin-myosin cross bridges of a fiber may contribute to tension development in the course of a normal twitch contraction of fast muscle at 35 C, but the amount of activator liberated after the action potential may be submaximal and therefore limit the average rate of cycling of each cross bridge at any particular load. The initial stage of PTP may then result from an increase in the amount of activator liberated after a single action potential, thereby raising the activator concentration throughout the fiber, with a corresponding increase in the average rate of cycling of individual cross bridges at a particular load and an increase in the number of bridges formed and tension developed at any given time in the twitch; the degree of activation or potentiation would increase with increased liberation of activator up to a maximum corresponding to the maximum rate of cycling of cross bridges for each load, and this would not necessarily alter either the intrinsic speed of shortening of the contractile material at zero load or the isometric twitch contraction time. This hypothesis is supported by the recent work of Podolsky and Teichholz (335) on skinned muscle fibers of the frog, in which they demonstrated that the intrinsic speed and the speed of isotonic shortening at any relative load pIP, are independent of pCa in the range 5.0-6.75, and also by the results of earlier work showing that the steady-state isometric tension in frog muscle (201) and the specific activity of myosin ATPase of mammalian muscle (426, 427) vary from maximum to approximately half-maximum values in the same range of calcium concentration. The second stage of PTP, in which prolonged stimulation increases the twitch duration, rnay result from further increase in the amount of activator liberated, thereby exceeding the amount that saturates and fully activates the contractile material. As a consequence it would probably take some tirne for the concentration of activator to fall below the saturation level, and this would result in delayed decline of the active state and an increase in the contraction time. Some of the possible causes of an increase in the amount of activator released have been discussed by several authors and these have been reviewed briefly (84). The temperature dependence of peak tension is not the same for normal and posttetanic potentiated twitches of isolated rat fast and slow muscles (85). The peak twitch tension of the slow soleus muscle is not potentiated after a tetanus and is virtually independent of temperature in the range 20-35 C. In contrast, the peak twitch tension of fast muscles such as gastrocnemius and extensor digitorum longus is increased approximately 1.8 times either as a result of decrease in temperature from 35 C to 20 C or maximum PTP at 35 C (85, 407, 414). Posttetanic potentiation of massively stimulated rat extensor digitorum longus muscle is progressively decreased with decrease in temperature and is virtually

January

1972

DYNAMIC

PROPERTIES

OF

MUSCLE

161

absent at 20 C; furthermore, the peak tension of posttetanic twitches is independent of temperature between 35 and 20 C (85, Fig. 2). There are at least two possible interpretations of the differences in activation of fast and slow muscles that have been revealed by studies of PTP. One interpretation is that slow fibers do not exhibit PTP because they are normally fully activated after a single stimulus at 35 C. The other interpretation is that the degree of activation is the same for fast and slow fibers in normal twitches at 35 C and that the differences in the aftereffects of repetitive stimulation are the result of qualitative differences in the properties of structures that are involved in excitation-con traction coupling. There are two observations suggesting that the second interpretation is the correct one : first, the peak twitch tension per unit cross-sectional area of muscle is virtually the same in normal and cross-innervated fast and slow muscles of the rat (2 1); second, both normal extensor digitorum longus and cross-innervated soleus muscle innervated by fast nerve show PTP, whereas it is absent in normal soleus and cross-innervated extensor digitorum longus muscle innervated by slow nerve (86). Elucidation of the qualitative differences in activation of fast and slow twitch muscles of mammals would contribute to a general understanding of excitation-contraction coupling. Other aspects of activation such as heat production, volume changes, and latency relaxation have not been studied thoroughly in mammalian fast- and slow-twitch fibers. Latency relaxation occurs in mammalian muscle and has been recorded in responses of cat tibialis anterior (174, Fig. 6) and rat gastrocnemius (430, Figs. l-4) and observed in responses of the peroneus digiti quinti (302), extensor digitorum longus, and soleus (unpublished observations of author) muscles of rats.

IV.

ONTOGENETIC

DIFFERENTIATION

OF

FAST

AND

SLOW

MUSCLES

A. Growth In attempting to interpret the dynamic properties of developing muscles it is necessary to take into account the rapid growth of muscles during late prenatal and early postnatal stages. Many of the main features of growth of muscle cells have been elucidated but little is known about the rnore subtle processes of control of protein and isoenzyme synthesis during development. Various aspects of histogenesis of rat and mouse muscles have been described by Kelly and Zacks (244), Banker ( 15), B er g man (29), and Hitchcock (2 16). The rat and mouse are very immature at birth and provide convenient material for studying the transition from myotube to myofiber because this change takes place mainly in the postnatal period in the hindlimb muscles of these animals. The general pattern of growth and differentiation found for rat and mouse limb muscles is probably the same as that for muscles of other mammals, although the changes occur during fetal stages of development in animals such as the guinea pig, which are fairly mature at birth. It is also important to note that maturation

162

R.

I. CLOSE

Volume 52

of’ muscle fibers occurs at different times in ontogeny in different muscles within the s’r~ne animal. For example, the transition from myotubc to rnyofiber is complctcd before birth in rat intercostal rrmsclcs (244) but is postnatal in limb muscles; furt hcrmore, the diaphragm muscle of newborn kittens has about the same speed of contraction as adult diaphragm muscle (161). There is also evidence that cephalic rrluscles develop earlier than caudal muscles in some respects (36). Longitudinal growth of rabbit skeletal muscle fibers has been studied by Crawford and his coworkers (4, 4a, 89a, 124a). One of their most important findings is that isornctric tetanic tension is maximal at the muscle length correspondins to the in situ length at any particular postnatal stage; moreover, longitudinal growth of the rnusclc does not lead to any marked change in the curve relating ~nusdc 1mgt.h as a fraction of the in situ length and isometric tetanic tension as a fraction of the maximum tension (4). The nmscle fiber length increases during normd postnatal dcvclopment but there is no significant change in sarcomere length (124a). These results indicate that increase in muscle fiber length is brought about by increase in the number of sarcomeres in each fiber, and they are consistcn t with observations made by other workers. For example, the lengths of t 1lick ad thin contractile filaments of sarcomeres appear to be identical in neonatal and adult mouse limb muscles (167); consequently, the sarcomere length at which there is maximum overlap of thick and thin filaments is unaltered throughout postnatal life. Increase in the length of muscle fibers or myofibrils at masinmm overlap of filaments is due entirely to increase in the number of sarcomercs in mouse (167) and rat fast and slow (74) muscles, and the available edcnce indicates that new sarcomeres are formed at the ends of fibers (247, 273, 274). C rawford and his colleagues also carried out an interesting series of expcri rncnts on rabbit ti bialis anterior in which it was shown that artificial increase or dccrcase in the in situ length and shortening distance of the muscle in young animals leads to proportional changes in the lengths of fibers in adult muscles &I, 8%). Thcsc results suggest that in the course of normal postnatal developmcnt the incrcasc in nurnbcr of sarcomercs in a fiber is determined to some extent by the longitudinal growth of neighboring tissues such as bone and tendon. It is clear that there is appropriate coordination of the growth of bone, tendon, and ~~~usc~lc tha t cnsurcs maximum overlap of thick and thin filarnents in the muscle in situ, but it is not known how elements of the muscle fiber obtain and utilize information concerning the number of sarcomeres that should be formed. Goldspink and his associates (167, 169, 17 1, 172) have shown that postnatal incrcasc in cross-set tional area of a muscle is due to growth of individual cells and not to an increase in the total number of cells in either myotube or myofiber stages. In rat and ~nousc fast and slow muscles there is a rapid decrease in the number of rnyotubcs from about 90 7:) of the total cell population at birth to zero ;It 2- -3 weeks postpartum (128, 361). Goldspink (169) reported that transverse growth is not synchronous in all fibers of a muscle, although it appears that there ;W stq)wisc increases in the cross-sectional area of individual muscle fibers and approximately twofold increases in the number of myofibrils at each step. Goldspink (169) has also shown that the distribution of myofibril diameters is bi-

January

1972

DYNAMIC

PROPERTIES

OF

MUSCLE

163

modal, with peaks at 0.5 ,u and about 0.9-L 1 p at all stages of postnatal development, and he found no very small myofibrils or any other evidence that there is de novo synthesis of myofibrils during postnatal growth; furthermore, electron micrographs of differentiating muscle show that many myofibrils about 1.0 p in diameter are partially divided longitudinally. On the basis of thcsc observations of myofibrils Goldspink ( 169) h as suggested that the increase in total number with increase in cross-sectional area of fibers is the result of longitudinal splitting of rnyofibrils when they reach a critical size -about 1.0 ,U diameter. Growth of myofibrils from 0.5 p to 1.0 ,u in diameter presumably involves synthesis of the principal contractile proteins on polyribosomes in the sarcoplasrn (205, 206, 259) with subsequent polymerization into thick myosin filaments and thin actin filaments that are laid down on the periphery of the myofibrils (296). The results of these investigations on growing muscle show that contractile material is added in series and in parallel within the muscle fibers. R. Dynamic Properties Most investigations into developmental changes in dynamic properties of skeletal muscle have been carried out on hindlimb muscles of different mammals. Histogenesis of striated muscle in mammals leads to the formation of limb muscles &at are uniformly slow at first (16). In the early stages of development the isometric twitch contraction times are about the same for all hindlimb muscles in a given animal species; subsequently the responses become faster and contraction times decrease to adult values along different courses leading to differentiation of fast and slow muscles (55, Fig. 3; 74, Fig. 3). This differentiation is completed within a few weeks; although the pattern of change is qualitatively similar in different mammals, the time course is shorter in small species than in large< ones. Furthermore, differentiation into fast and slow muscle takes place at different times during ontogeny depending on the maturity of the species at birth; it occurs mainly during prenatal stages in animals such as the guinea pig and sheep, partly during prenatal and postnatal stages in the cat, and mainly in postnatal stages of development in rodents such as the rat and mouse (7, 55, 60, 74-76, 83, $3, 253). I n addition to changes in the time course of isometric twitches there are changes in the twitch: tetanus ratio (60, 74, 75, 83), decrease in latent period (72), and differentiation of fast and slow muscles with respect to development of posttetanic potentiation of peak twitch tension (60,305, 306,308). These observations clearly indicate changes in some aspects of excitation-contraction coupling and the active state that may be related to concomitant changes in (5, 271, 375, 410, 417, 418, the biochemical (130, 220, 398) and morphologial 420, 42 1) propertics of the sarcoplasmic reticulum and transverse tubular system. However, at this stage it is not clear which of several interpretations of these observa tions is correct. In order to determine whether there are developmental changes in intrinsic ‘ speed and intrinsic strength of the contractile material it is necessary to express the speed and tension of the whole muscle or fiber in terms of unit amounts of

164

R.

I.

CLOSE

Volume

TABLE 4. Properties of extensor digitorum longus (EDL) muscles of rat at birth and 100 days bostbartum

and soleus (SOL)

E DL _______-Isometric

twitch

Twitch:

tetanus

Maximum

contraction

time,

msec

ratio

speed

65.0 0.6

of shortening

of a sarcomere,

19.3

52

56.5 0 . ,513 17.3

~-

SOL

EDL

36.0

12.5

0.25 18.2

0.19 42.7

Cl/= a/P,

(eq. 2 in From

ref

text) 74;

0.25 tenlperature

0.27

0.17

0.25

3.5 C.

material in series and parallel, respectively. In the case of speed of shortening, an estimate of the rate at which thick and thin filaments slide past each other of the during contraction can be obtained by dividin g the speed of shortening whole fiber by twice the number of sarcomeres in series along the fiber. The cornparable measure of tension, expressed in terms of the number of cross bridges per half-sarcomere, has not been used in the study of developing muscle. This lirnitation makes it impossible to interpret ontogenetic changes ii twitch: tetanus ratio and the maximum twitch and tetanic tensions per unit cross-sectional area of muscle. Nevertheless, the relation between the speed of shortening of sarcorneres and the load expressed as a fraction of the rnaxirnurn load (P/P,) has proved to be useful for comparing the force : velocity properties of tnuscles of different size and it provides a reliable estimate of the intrinsic speed of the contractile material at zero load. The force : velocity properties have been detertnined to define more precisely the ontogenetic changes that occur during differentiation of fast and slow muscles (74-76, 83 j. Rat and mouse muscles have been used for most of this work because the changes in dynamic properties of these muscles take place in the early postnatal stages; the extensor digitorum longus and soleus ~nusclcswere used because there is a simple arrangement of muscle fibers and there is no difIiculty estimating the properties of sarcomeres from measurements on whole muscles (74). The relations between speed of shortening of sarcomercs and the relative load (I’&) show that the force:velocity properties of the contracile material of mouse and rat extensor digitorurn longus and soleus muscles are virtually identical at birth; thereafter the speed of shortening of extensor digitorum longus is increased twoto threefold within 3 or 4 weeks postparturn, whereas soleus undergoes little or no change in this respect (74, Fig. 9; 76, Fig. 1; 75, Fig. 3). Subsequent growth of rat muscles up to 500 days postpartum is not associated with any major change in the intrinsic speed of shortening of these two ~nusclcs(79, Fig. 2) though there is a slight decrease in speed of contraction of extensor digi torum longus muscles of very old animals between 2 and 3 years of aqe (397). Some of the dynamic

January

1972

DYNAMIC

PROPERTIES

OF

MUSCLE

165

characteristics of extensor digitorum longus and soleus muscles of newborn and adult rats are listed in Table 4. The constant a/& [from Hill’s equation (207)] is about the same for both muscles at birth and for adult extensor digitorum longus, whereas there is progressive decrease in a/e, for soleus as the animal matures; this difference in a/PO and the shape of the force : velocity curve are discussed in section VIID. It should also be noted that although the pattern of change in force: velocity properties of fast and slow muscles is the same in animals such as the cat, rat, and mouse, the intrinsic speeds of shortening of homologous muscles of different animal species are inversely related to body size at any comparable stage in ontogeny; this relation between speed of shortening and body size is discussed in section VA. In 1964 it was shown that there is a correlation between ontogenetic changes in intrinsic speed of shortening of sarcomeres and enzymic properties of the contractile material (74). This obscrva Con was based on the results of independent investigations of biochcnnical and dynamic properties of developing rat muscle; de Villafranca (411) showed that the specific activity of Mg2f-activated ATPase of actomyosin (isolated from predominantly fast muscles) increases 2.8 times during the first 5 weeks after birth and this change is almost exactly proportional to the increase in intrinsic speed of shortening of sarcomcres in the same period (74, Table I). S imilar developrncntal changes in ATPase activity have been reported for actomyosin and myosin from muscles of various mammals (23, 203, 220, 330, 404, 405, 408). Trayer and Perry (405) made a comparative study of SOIW of the propertics of myosin of various species during ontogeny. They found that the specific activity of myosin ATPase incrcascd severalfold in muscles of rat, rabbit, and guinea pig and that in the guinea pig the change is completed before birth whereas in the rat it occurs during postnatal stages. Hydrodynamic measurements revealed no diffcrcnce in the size or shape of molecules of fetal and adult myosins (405). Neverthelcss, Trayer and Perry (405) pointed out that the difference between the cnzymic activity of fetal and adult myosin of fast muscles must be in the structure of the protein, and several observations support this view. For example, Traycr and Perry showed diffcrcnccs in succinylation of fetal and adult myosins that they suggested were due to differences in reactivity of the r-NH2 of the lysine residues (405). It has also been found that there is no 3-methylhistidine in myosins of rabbit fetal muscle (233, 403) or the adult cat soleus muscle (255), which are composed almost entirely of slow muscle fibers, whereas it is present in myosin of adult fast muscle (255, 403). Traycr and Perry (405) have also suggested that fetal and adult myosins correspond to different isoenzymes, or different combinations of isocnzymes, and that dcvclopmental changes involve control, perhaps at the gene level, of synthesis of individual isoenzymes to produce a myosin that has a specific ATPase activity appropriate to the activity of the tissue. There appears to bc no information available on the properties of myosin of slow muscles, such as solcus, at an early stage of development before differentiation of fast and slow muscles. The approximately threefold increases in specific activity of myosin ATYasc and intrinsic speed of shortening and the methylation

166

R. I. CLOSE

Volume

52

of histidine in myosin of fast muscles during development are comparable to the differences in the properties of mature fast and slow muscles. This suggests that the myosin of adult slow muscle is very similar in these respects, if not identical, to myosin of fetal or neonatal muscles at the onset of differentiation of fast and slow muscles.

C. 0 her Developmental

Changes

Some work has been carried out on a few propcrtics of the membranes of In rat and mouse muscles there is a marked increase in the developing muscle. resting membrane potential to adult levels within the first 2 or 3 weeks postowever, there appears to be no informapartum (36, 144, 195, 198, 199, 376). H tion available on membrane constants and the form of the action potential of fast and slow muscles during development, although these properties have been determined for various adult muscles (2, 3, 39, 49, 62, 125, 132-135, 226, 248, 269, 444, 447). Changes in the intracellular ionic composition and ace tylcholine sensitivity of the muscle membrane have been determined for some developing fast and slow muscles (27, 97, 145, 199, 229, 230, 246, 272, 409) and adult muscles (1, 71, 90, 100, 291, 384, 388). Histochemical (28, 99, 103, 104, 128, 304, 307, 435) and biochemical (184, to follow the developmental 197, 245, 260, 276) methods have been employed changes in concentration and activities of various sarcoplasmic, glycolytic, and oxidative enzymes of mammalian muscles.

V.

RELATION

BETWEEN

A. Speed of Contraction

SIZE

AND

SPEED

of Homologous

OF

CONTRACTION

Muscles

of Diferent L

Species

A. V. Hill has written several informative and very entertaining papers on the relation between the dimensions of animals and the dynamic properties of their muscles (148, 2 10, 2 13). One of the characteristics of locomotion in small mammals such as the mouse is the rapidity of limb movements compared with those of larger mammals. Hill focused attention on the observation that similar animals carry out similar movements, not in the same time but in times that are approximately proportional to their linear dimensions, and hc gave an explicit description of the principal factors that determine the relations between intrinsic speed of shortening and body size; a sumrnary of one of the examples used by Hill will illustrate some of the main points. If a large animal is 1000 times as heavy as a small one of similar shape, its linear dimensions will be approximately 10 times as great. That is to say, the length of a limb or muscle of the large animal will be 10 times greater than the homologous part of the small animal. During a con traction the force per unit cross-set tional area of muscle has the units F =

January

IlYNAMIC

1972

PROPEKTIES

OF

167

MUSCLE

p/z. t-“, where p is the density of the muscle, I is the length, and t is time for a particular movement. Assuming the densities of the muscles of large and small animals are the same, then the inertial stress (F) at corresponding ‘points in the long (n) and short (b) muscles will be in the ratio Za t,+ : Ir,=lb-l. The maximum speed of shortening of a muscle is determined by the intrinsic properties of its contractile material, but the limit to which these properties must conform is set either by the need for economy, as for slow muscles such as soleus that maintain posture, or by the strength of the contractile material and other body materials such as connective tissue, tendon, and bone. Hill emphasized that the intrinsic strength of the contractile material is about the same in all vertebrate limb muscles in contrast to the great range of their intrinsic speeds. Consequently, if the inertial stresses at corresponding points in the long and short muscles are to be the same, and near the limit set by the intrinsic strength of the contractile material, i.e., is to equal Zb then t, must be equal to 10 because Za equals 10 Zb . if ZU In other words, the actual speed of movement of the limb or muscle will be the same in both animals. However, as the frequency of limb movement will be 10 times greater in the small animal and the linear dimensions are 10 times greater in the large animal, the maximum running speed along flat ground will be approximately the same for the two animals. In contrast, the intrinsic speed in I~UScle lengths per second will be 10 times greater in the small animal and, in this example, it varies inversely as the cube root of the body weight. Hill collected data showing that similar terrestrial animals of different size are able to run about the same linear speed and jump the same height; for example, the hare, fox, wild donkey, and horse have maximum speeds about 2 l-22 m/see. These data clearly indicate that Hill’s interpretation is qualitatively correct. The body weights of animals of similar form tend to vary as the cube of the linear dimensions (4 1) . However, homologous muscles of different species may differ structurally and functionally according to the activity, behavior, and size of the animal, and it is not surprising that the intrinsic speed of shortening does not vary precisely as the inverse of the cube root of body weight. Nevertheless the intrinsic speed of shortening of sarcomeres of homologous muscles of different species is inversely related to body size, as shown in Table 5. One point of interest is that the intrinsic speed of a muscle does not decrease with increase in body size during postnatal growth (Table 5). At the onset of differentiation of limb muscles into fast and slow muscles the intrinsic speed is predetermined, presumably genetically, at a level appropriate for the functions that different muscles perform in the adult animal. In the case of slow muscles such as soleus, which maintain posture, the intrinsic speed remains the same throughout life and is appropriate for economical operation of the muscle during sustained activity. The intrinsic speed of fast limb muscles is increased two- to threefold during early development and this change presumably brings it to a level that is appropriate for the body size of the adult, thereby permitting fast phasic activity while maintaining inertial stresses within the limits set by the intrinsic strengths of the contractile material and connective tissues. There are some apparent exceptions to the generalization that intrinsic speed l

l

t,+

l

tb+

l

tb

l

168

R.

I.

CLOSE

Volume

TABLE 5. Relations between intrinsic speed of shortening of sarcomeres*, rate hydrolysis of adenosinetriphos@ute by myosin in presence of indicated activatorsJr, and approximate runge of udult body weights

Intrinsic Speed of Shortening of Sarcomeres, p/set

Mouse Mouse

EDL SOL

Rat Rat

EDL SOL

Cat Cat

fast SOL

EDL *From refs

= extensor 74, 76, 83.

A’I’Pase Muscle P;/min ___---

Activity Myosin, per m g ~~ -_

of Adult pmoles myosin --.-

Neonate

Adult

M$+-actinactivated

28.2 30.6 17.3

60.5 31.7

1.8 0.86

1 .:31 0 . El

19.3 22.8$

42.7 18.2 131g

1 .68 0.64 1.22

1 .24 0.58 0.73

12.7

13

0.45

0.44

digitorum tFrom

longus ref 19.

--._ _-____-__ _ muscle, SOL $Extensor

= soleus digitorum

52

of

Adult

Body

Wt,

kg

Ca2+-activated

muscle;

longus.

0.02-0.03 0.2-0.25 2.0-3.0

temperature

35 C.

SQuadriceps.

of shortening is inversely related to body size. It is well known that Rana temporaria and Bufo bufo have about the same body shape and size and that the muscles of the frog are intrinsically faster than those of the toad. The muscles of the sloth Choloepus hoflmanni have intrinsic speeds of contraction that are only about one-third of the speeds of homologous muscles of the domestic cat even though these two animals have about the same body weight (162, 277). However, the cat and the sloth are adapted to quite different habitats and locomotory movements. In this connection it is interesting to note that the total amount of skeletal muscle as a fraction of the body weight of the sloth is only about one-half of the amount present in other mammals (40). Furthermore, the intrinsic strength of the contractile material is about the same in sloth and cat muscles (162). Consequently it is not surprising, at least for the dynamic reasons discussed above, that sloth muscles are intrinsically slower than cat muscles. The large range of intrinsic speeds of shortening of similar muscles of species of different size has afforded opportunities for determining some of the relations between speed of shortening and the physical and chemical properties of the contractile material. One of the most interesting properties that has been examined is the relation between intrinsic speed of shortening and the specific activity of ATPase of actomyosin or myosin. Prosser (338) reviewed the results of early investigations and pointed out that speed of contraction in fast and slow muscles of invertebrates and vertebrates appeared to be directly related to actomyosin ATPase activity. Since then the relation between the two variables has been determined with greater precision for fast and slow muscles of various mammals (19-21, 63, 156, 179, 242, 297, 364, 377, 387, 440) ; the general conclusion reached was that the ‘cspeed” of isometric twitch contraction was directly related to the rate of hydrolysis of ATP by myosin, and in most instances workers in this field pointed out the possibility that the ATPase activity of myosin may be the rate-limiting step in the shortening process. The observation that the ratio of specific activity

January

DYNAMIC

1972

PROPERTIES

OF

169

MUSCLE

of ATPase of myosins of rabbit fast and slow muscles (20, 377) was the same as the ratio of the intrinsic speeds of shortening of sarcomeres of rat and mouse extensor digitorum longus and soleus muscles gave the first indication, albeit inconclusive, that the two variables arc proportional in mammals (75) ; this was confirmed by B&&y by measurement of the enzymic activity of myosins isolated from rat and mouse extensor digitorum longus and soleus muscles ( 19) ; the results are shown in Table 5. However, the comparative work of B&r&y (19) went much further than this and is a most valuable contribution because the results show clearly for the first time that the intrinsic speed of contraction is proportional to the specific activity of the Mg2+- and actin-activated ATPase of myosin for many different vertebrate and invertebrate muscles. The high degree of correlation between the two variables provides very strong evidence in support of the hypothesis that myosin ATPase is the rate-limiting step in the shortening process. Another important outcome of Bar&y’s work was the observation that the F-actin binding activity, as indicated by ATP sensitivity, and the intrinsic strength of the contractile material are about the same for various muscles, and hc pointed out the possibility that these two propertics may be causally conncctcd. B. Speed

of

Contractiori

o;f Dilj'erent .

Muscles

of &me

Animal

Hill (210, 213) also p ointed out that the intrinsic speed ( f contraction varies greatly from one muscle to another in any particular animal species. Muscles reiuired to move light structures have high intrinsic speeds of shortening whereas those required to move massive structures must contract more slowly. Table 6 shows the isometric twitch contraction times and twitch : tetanus ratios for different muscles of the cat. The results in Table 6 show that small muscles such as the 6. Characterhks

TABLE --

of twitch contractions

- --_________--.~-

-

-

muscles

Isorne tric Twitch Contraction Time, msec

Muscle __.

of cut

_

Internal rectus Inferior oblique Thyroarytenoid Posterior cricothyroid Cricothyroid Intercostal fast units Intercostal slow units Extensor digitorurn longus (hindlirnb) Gastrocnernius Flexor digitorurn longus

(215) (284) (21.5) (21.5) 63)

(60,

176, 434)

--_

___-_Measurements

in situ

at 35-38

C ; references

0.091

Ratio

(89)

0. 133

(215)

0.294 0.233

(215) (215)

0.24 0.185 0.26 0.22

. (60) (65) (60) (65)

6‘) (17;) i

22.5 18.0 7:3 . O--78.0

S oleus

( W) uw

7.5 -10.0 5.7-6.9 21 .o !LO-13.0 22.0 44.0 24.6 47.0 19.6

Twitch:Tetanus

in paren

theses.

K.

170

I. CLOSE

vdw?lc~ 52

extraocular and laryngeal muscles have shorter contraction times than the hindlimb muscles. Furthermore, the twitch : tetanus ratios of the internal rectus and thyroarytenoid muscles are less than one-half the usual values for lirnb muscles at 35-37 CA. The force :velocity properties of internal rectus and thyroarytenoid muscles have not been determined and it is not known whether the constant kr in equation 3 (sect. IIF) is the same for these small muscles and the large limb muscles. Very low values for contraction time and twitch : tetanus ratio have been reported for the thyroarytenoid muscle of the dog (258, 279, 280) and similar relations between muscle size and speed of contraction have been reported for 17 different muscles of man (48, 50, 54, 95, 111, 282, 383). A comparative study of the dynamic, histochemical, and biochemical properties of heterologous muscles would contribute to an understanding of the relation between intrinsic speed of shortening and duration of the active state.

VI.

NEURAL

CONTROL

OF

DYNAMIC

PROPERTIES

The trophic influences of nerve on contractile properties of muscle have usually been studied by observing the effects of various alterations in activity or neuromuscular connections. Some of the results of work in this field have been reviewed rethat there is as yet no convincing evicently (81, 331, 332), and it was concluded dence that the intrinsic properties of functional contractile material are altered by changes in activity pattern or work load brought about by denervation, tenotorlly, immobilization, exercise, or denerva tion of synergists. This conclusion was reached because in a few instances it has been shown that there is no alteration of the for-cc : velocity properties or characteristics of isometric contractions, but in other instances in which changes in twitch contractions have been observed there are either conflicting reports or inadequate information to allow a decision to be made regarding the nature of the change. This must not be taken to mean that the dynamic and chemical properties of the contractile material are immutable. On the contrary, there are two preliminary reports by Gutmann and associates (188, 189) that certain kinds of activity lead to reciprocal changes in the isometric twitch contraction time and the specific activity of Ca2+- activated ATPase of rnyosin of rat extensor digitorum longus. *Much more information has been obtained from observations on the effects of transection of the spinal cord and cross-union of the motor nerves of fast and slow muscles. About 13 years ago Eccles and his colleagues (115) drew attention to the fact that slow muscie fibers of adult cats are innervated by tonic motoneurons that usually discharge at about Z0-20/set, whereas fast muscle fibers are innervated by phasic motoneurons that usually have a higher rate of discharge between about 30/ set and 60/set. This correlation between frequency of repetitive discharge of motoneurons and speed of contraction of motor units indicated the possibility that neural elements determine some of the contractile properties during ontogenetic

January

1972

DYNAMIC

PROPERTIES

OF

MUSCLE

171

differentiation of fast and slow muscles. Eccles (112) described some of the definitive experiments that were carried out to test this possibility, and the work in his laboratory culminated in two very important papers on postnatal differentiation of fast and slow muscles of the cat (55; see sect. IVB) and neural influences on speed of muscle contraction (56). The principal findings were: a) postnatal differentiation of fast muscles is virtually unaffected by spinal cord transection, or by operative isolation of the lumbosacral spinal cord from all incoming impulses, whereas the normal course of development of slow muscles is greatly altered so that within a few weeks the time course of the isometric twitch response is nearly the same as that of a fast muscle; and h) operative cross-union of motor nerves to fast and slow muscles of juvenile and adult animals leads to reciprocal changes in isometric twitch time course with the result that the contraction of muscles formerly fast was slowed and the speed of isometric contraction of muscles formerly slow was increased. These results showed that differences in speed of response of fast and slow muscles are dependent to some extent on specific neural influences that are mediated through motor nerves and operate throughout the life of the animal. Two important questions remained unanswered at that time: the nature of the neural influences on speed of muscle contraction and the part (or parts) of the muscle fiber that are affected. Eccles and Buller (56, 61, 113, 114) h ave proposed two possible mechanisms of neural influence on speed of contraction. The first hypothesis is that the neural influence on speed is related in some specific way to the pattern of motor nerve impulses and that it is mediated either directly through activity imposed on the muscle or indirectly by causing the motoneurons to produce a specific substance that in turn finds its way to the muscle and influences the speed of contraction. :Ceveral attempts have been made to test this hypothesis by stimulating fast and slow muscles at different frequencies (e.g., 114, 363) ; some of the changes resemble those after nerve cross-union (8 1) but the results were inconclusive because equivocal criteria were used to assess changes in the dynamic properties of the muscles. The second hypothesis is that a specific trophic substance is produced in the motor nerve and is passed on to the muscle fiber that it innervates; production or secretion of this hypothetical substance may be dependent on synaptic bombardment of the rnotoneuron, impulse conduction down the nerve, and normal transmission at the neurobut the nature of the substance and its influence on dynamic muscular junction, properties of the muscle are not related to the characteristic patterns of activity of fast phasic and slow tonic motoneurons (56). Fex (137) has made several interesting observations in connection with the second hypothesis; he reported a decrease in the isometric twitch contraction time of normally innervated rat soleus muscle in the presence of fast nerve implants, despite the fact that these failed to make functional synaptic contact with the extrafusal muscle fibers. Furthermore, Fex and Sonesson (138) studied the histochemical properties of these soleus preparations and found that the number of fibers having high myofibrillar ATPase activity is increased as much as 5 times after successful implantation of the tibia1 nerve. These observations are valuable because they indicate that changes in the twitch response may be brought about by a neural influence, presumably a substance, that operates in the absence of neuromuscular connections. Neverthelcss, it should bc empha-

172

R.

I. CLOSE

Volume

52

sized that here, as in other instances mentioned above, a much more rigorous approach is necessary before it can be established that there is transformation of the contractile material or changes in the active state. The literature on mechanisms of neural influence on speed of contraction has been reviewed several times in the last few years (53, 8 1, 178).

B.

Dynnmic

Properties

of Normal

and Cross-lnneruated

Mudes

Some progress has been made during the last few years in deciding which parts of the muscle fiber are influenced by the nerve in its control of the dynamic properties of fast and slow muscle fibers. In this connection it may be recalled that the observa tions of Eccles and his colleagues (56) were on isometric contractions only and it was not possible to decide frorn their results whether the same kind of change was involved in normal development and the artificial conditions after nerve crossunion. Work has been undertaken in several laboratories to make more detailed comparisons of the properties of normal and cross-innervated muscles to determine whether the neural influence on speed affects the contractile material, thereby determining the force:velocity properties, or whether it alters the time course of the active state through an effect mediated at some extrinsic site such as the sarcoplasmic reticulum. *No information is available on the kinetic properties of the calcium transport system of the sarcoplasmic reticulum or the calcium-binding properties of the troponin-tropomyosin complex of cross-innervated fast and slow muscles, and the extent to which possible changes in these systerns contribute to changes in the isometric twitch after nerve cross-union is unknown. However, it has been shown that the difference in maximum calcium capacity of fragmented sarcoplasmic reticulum of fast and slow muscles is reversed after nerve cross-union (295). A considerable amount of information is available on the dynamic properties of the contractile material of cross-innervated muscles. The first significant contribution was made in 1965 (77) with measurements of the force-velocity properties of sarcomeres of normal self-innervated and cross-innervated extensor digitorum longus and soleus muscles of the rat, and these preliminary results were confirmed in a later investigation (80). The principal effects of nerve cross-union on the dynamic properties of these muscles are summarized in Table 7; the properties of self-innervated muscles have not been included because they are virtually the same as those of normal muscles (80). The first point to note in Table 7 is that the influences of nerve cross-union on the characteristics of isometric twitch responses of rat muscles are very similar to those originally described for cat muscles (56). There are four main findings from the physiological work on the effects of nerve cross-

and isometric

twitch

contraction

time of cross-innervated

muscles differ from con-

January

1972

DYNAMIC

PROPERTIES

OF

173

MUSCLE

TABLE 7. Pro;berties of normal and cross-innervated extensor digitorum longus (jW?DL, X-EDL) and soleus (N-SOL, X-SOL) muscles of rat at 35 C Ref

N-EDL

X-EDL

N-SOL

X-SOL ~~

Maximum

isometric

twitch

unit cross-sectional kg/cm2 Maximum isometric unit cross-sectional Twitch

kg/cm2 : tetanus

Isometric Intrinsic meres,

tension

area tetanic area

per

21

0.51

0.54

0.53

0.505*

tension per of muscle,

21

3.00

1.91

2.09

2.67

21 21, 80

0.17 12.5

of

muscle,

ratio

twitch speed &ec

contraction of shortening

Intrinsic speed of shortening fibers, mm/set a/PO (from eq. 2 in text); and actin-activated Mg2+ity of myosin myosin)

(pmoles

time, of of

ATPase Pi/min

msec sarco-

80

45.1

muscle

80

208.0

activ-

80 21

per

0.25 1.17

0.28

0.255

22.0 22.5

37.0 19.8

102 .o

101 .o

0.18 0.55

0.15 0.50

0.16 15.6 33.8

mg

* From

Table

5 in

reference

21.

trol values ana approacn tne values cnaracteristic ror tne muscle rormerly innervated by the nerve; c) the force : velocity relation is altered as indicated by changes in a/PO (the possible significance of this is discussed in sect. VIID); and d) the isometric twitch contraction times of normal and cross-innervated extensor digitorum longus and soleus muscles are inversely proportional to the intrinsic speeds of shortening of sarcomeres, and neither of these variables is correlated with the peak twitch tension per unit cross-sectional area of muscle. The hyperbolic relation between the isometric twitch contraction time and intrinsic speed of shortening of sarcomeres of cross-innervated muscles (equation 3) is the same as that found for a variety of normal muscles (74, 75, 79), and the same questions arise concerning the mechanism underlying the correlation between the two variables and the factors that determine the tirne course of decline of the active state (see sect. III). It has been pointed out previously (74, 75) that this pattern of change is to be distinguished from changes in isometric twitch contraction time that are accompanied by approximately proportional changes in twitch: tetanus ratio and result from changes in the duration of the active state with no alteration of force :velocity properties. Buller and his colleagues have carried out similar investigations on the dynamic characteristics of fast flexor hallucis longus and slow soleus muscles of cats. Their recent observations on cat fast muscles show that the long-term effects of nerve cross-union on the isometric contractions (61, 65) are essentially the same as those described above for rat muscles. Buller and Lewis (61) also found that the maximum rate of rise of tetanic tension is markedly increased in flexor hallucis longus muscles after innervation by soleus nerve; they suggested this change resulted from alteration of the force : velocity properties of flexor hallucis longus, and

174

K.

I. CLOSE

Volume

52

a recent preliminary report (57) confirms this interpretation. Innervation of cat soleus muscles by nerves of fast muscles leads to a decrease in isometric twitch contraction times to about one-half the normal value and a decrease in twitch: tetanus ratio from about 0.22 to about 0.14 (65); these changes are virtually the same as those described (Table 7) for rat soleus muscle innervated by extensor digitorum longus nerve. In contrast, the maximum rate of rise of tetanic tension of cat crossinnervated soleus muscle was increased only about 1.2 times (53, 61, 65) and the intrinsic speed of shortening of the whole muscle was virtually unaltered (57). On the basis of these observations it was suggested that changes in cat soleus muscle after innervation by fast nerve may be due mainly to increased rate of decline of the active state with little or no change in the force:velocity properties (53, 57, 58, 61, 65). There have been several reports that cross-innervated fast and slow muscles of cat and rat are not completely transformed after nerve cross-union. Table 7 shows that changes in isometric twitch contraction time and intrinsic speed of shortening of sarcomeres indicate incomplete transformations of cross-innervated extensor digitorum longus to a slow muscle and cross-innervated soleus to a fast muscle, and L there are statistically significant differences between mean values for the cross-innervated and normal muscles innervated by a particular nerve (80). On the other hand, the maximum speed of shortening of whole fibers and hence the speed of shortening of the whole muscles (cf., 74, Fig. 2) were virtually the same for all muscles innervated by a given nerve. In other words there was complete transformation of cross-innervated muscles with respect to the speed of shortening of whole muscle fibers. At this stage it is not clear which measure of intrinsic speed should be used to assess the degree of transformation. Despite this difficulty in interpretation the observations bring additional interest to the problem because they raise the possibility that the neural influence determines the speed of shortening of whole fibers rather than merely altering the intrinsic speed of shortening of the contractile material to a predetermined level. It is important to note that fibers of normal solcus are longer than fibers of normal extensor digitorum longus, that the number of sarcomeres in soleus fibers is increased 1.25 times after innervation by extensor dig-i torum longus nerve, and that the sarcomere length is about the same in all these muscle fibers (80). Consequently, the intrinsic speed of shortening of sarcomeres of cross-innervated soleus would not have to be transformed completely if the neural influence determined the speed of shortening of the whole fiber. Similarly, the shorter fibers of cross-innervated extensor digitorum longus could be altered to shorten at the same speed as the longer fibers of normal soleus without complete transformation of the intrinsic properties of the contractile material. It should not be too difficult to test this possibility by cross-innervating fast and slow muscles with different fiber lengths. These results also pose the interesting question of whether muscles that are partially transformed with respect to intrinsic speed of shortening of sarcomeres have individual fibers that contain either a mixture of fast and slow contractile material or a contractile material that is continuously variable with respect to speed within certain limits (80, 351). Other effects of nerve cross-union on the characteristic contractile properties

January

1972

DYNAMIC

PROPERTIES

OF

MUSCLE

of fast and slow muscles include reciprocal changes in the temperature (65, 218) and posttetanic potentiation of isometric twitch contractions C. Efects of Nerve Cross-Union

175 dependence (38, 86).

on Pro/w-ties of Myosin

Differences in the dynamic properties of fully active contractile material of normal and cross-innervated muscles revealed that changes in speed after nerve cross-union are brought about by a neural influence that affects the contractile material itself, thereby determining the intrinsic speed of shortening and the force: velocity properties (77, 80). The next stage in the investigation was to find out in what way the contractile material is modified. Histochemical studies provided the first evidence that the enzymic properties of the contractile material were determined by specific neural influences; Dubowitz (106) and Karpati and Engel (240) showed that the activity of myofibrillar ATPase decreased in cross-innervated fast muscle and increased in cross-innervated slow muscle. The first attempt to study the biochemical properties of the contractile material of cross-innervated muscles was made by Dubowitz in 1967 (106) but he was unable to demonstrate changes in the specific activity of myosin ATPase. The histochemical evidence showed that changes in myofibrillar ATPase were restricted to small groups of fibers in his cross-innervated muscles, and he suggested the extent of reinnervation of muscles by alien nerves may have been too small to lead to a biochemically detectable change in myosin extracted from homogenates of the whole muscle; in retrospect it does seern likely that the muscles he used were largely selfinnervated. The next investigation was undertaken by Buller, Mommaerts, and Seraydarian (63). They obtained the first reliable biochemical evidence that the specific activities of myosin and myofibrillar ATPase of fast muscle (flexor digitorum 101qus) decrease after innervation by slow soleus nerve, to become nearly the same as those of normal slow muscle. However, they did not find consistent changes in the enzymic properties of the contractile material of soleus muscle after innervation by flexor digitorurn longus nerve; the Ca2+- activated ATPase activity of myosin of cross-innervated solcus increased about 1.4 times but they found virtually no change in the EDTA-activated myosin ATPase or Mg2+-activated myofibrillar *4T Pase. Samaha, Guth, and Albers (366) obtained results similar to those of Buller of Ca2+-activated ATPase activity of myosin of et al. (63) f rom measurements cross-inncrvatcd fast and slow muscles of the cat, although in this instance the specific activity of Ca2+ -activated ATPase activity of myosin of cross-innervated soleus muscle was incrcascd 1.73 times; in view of this result it is rather surprising tha t a corresponding increase in the intrinsic speed of shortening of cross-innervated cat solcus muscle has not been observed (57). Guth and his collca\gucs took the invcstigation a stage further by determining the effects of nerve cross-union on the qualitative differences between myosins of fast and slow n~usclcs. It had been shown

176

K.

I. CLOSE

Volume

52

earlier that ATPase of fast-muscle myosin is relatively alkali stable and acid labile, whereas slow muscle myosin has the opposite pH dependence, and that each type of myosin contains a pair of electrophoretically distinct protein subunits that are released by incubation with p-chloromercuriphenylsulfona te (364, 387). These qualitative differences between myosins of fast and slow muscles were reversed after nerve cross-union and these important observations pose additional interesting questions concerning the control of synthesis of qualitatively different proteins (366). The relation between the maximum rate at which myosin and actin filaments slide past one another during contraction (one-half the intrinsic speed of shortening of sarcomeres listed in Table 7) and the specific activity of myosin and actornyosin ATPase has been determined for normal, self-innervated, and cross-innervated extensor digitorum longus and soleus muscles of the rat (2 1). Reinnervation of each of these muscles by its own nerve had no effect on either the dynamic or enzymic properties of the contractile material. On the other hand, Barany (21) found that the ATPase activities of myosin and actomyosin of cross-innervated extensor digitorum longus muscles decrease to the levels of those of normal soleus and the ATPase activities of the contractile material of cross-innervated soleus muscles approach those of normal extensor digitorum longus (some of the results are summarized in Table 7). These results showed for the first time the direct proportionality between the intrinsic speed of shortening of sarcomeres and actin-activa ted ATPase activity of myosin of normal and cross-innervated fast and slow muscles of mammals, and this finding provides strong evidence in support of the hypothesis that neural influences determine the fundamental dynamic properties of the contractile material through an effect on the ATPase site of myosin. Other effects of nerve cross-union on the properties of myosin determined by Barany include a) reciprocal changes in the rate of superprecipitation of myosin-actin mixtures that are directly proportional to the changes in intrinsic speed of shortening and 6) reversal of the characteristic curves relating myosin ATPase activity and pH and reversal of the rates of incorporation of 1 -fluoro-2 ,4-dini trobenzene into myosins of fast and slow muscles (21). The pH-profile curve and the ATP-induced dinitrophenylation reaction revealed that the structure of myosin of cross-innervated extensor digitorum longus was altered to resemble that of normal soleus and the structure of myosin of cross-innervated soleus was modified to that of normal extensor digi torum longus. It is important to note that there was no systematic variation in the yield of myosin from normal, self-innervated, and cross-innervated muscles. Furthermore, Bar5ny’s results were obtained for about 60 % of the total myosin of the muscle isolated as pure myosin and for nearly all the remaining myosin isolated in the form of actomyosin; consequently the conclusions that can be drawn from his work are based on mean values for the enzymic activities and other properties of virtually all the myosin molecules in cross-innervated and normal muscles (2 1).

D. Neural Influences on Noncontractile

Structures in Muscle Cells

It has generally been found that cross- union of the motor nerves of fast and his tot hemical ( 1OS- 107, slow muscle fibers leads to a reversal of their characteristic

January

DYNAMIC

I972

PROPERTIES

OF

MUSCLE

177

182, 240, 337, 35 1, 353, 354, 442) and biochemical (173, 18 1, 182, 336, 337) properties and reciprocal changes in intracellular ionic composition (219), the myoglobin content (285), and the contractural response to caffeine (190, 290). These and other aspects of trophic neural control of physiological and biochemical properties have been reviewed by Buller (53), Denny-Brown (94), Guth (178), Gutmann (183, 185-187), Mark (282), and Thesleff (401).

VII.

CORRELATIONS OF

BETWEEN

CONTRACTILE

DYNAMIC

AND

CHEMICAL

PROPERTIES

MATERIAL

A. Introduction This section deals briefly with some differences in biochemical properties of contractile material that appear to be correlated with differences in dynamic properties of skeletal muscles. This topic has been reviewed by Riiegg (361 a) in a very interesting paper on the relations between speed of contraction, properties of myosin, and the “holding economy” of various smooth and striated muscles. Other physicochemical properties of contractile proteins and the biochemical aspects of contraction have been reviewed recently (102, 159, 193, 241, 328, 329, 368, 446). The available evidence indicates several impor tan t differences between myosin of fast and slow muscles from the same species and between myosins of different species. On the other hand the properties of actin and the thin contractile filaments seem to be less varied and in most respects are identical in different muscles. It is of interest in this connection that differences in ATPase activity of actin-activated my&n of different muscles appear to be determined solely by the myosin component and are independent of the source of actin (19, 2 1). This shows that differences in enzymic activity of the contractile material in steady-state conditions are not due to differences in actin but it does not exclude the possibility that other properties of the contractile material are important in determining kinetic properties in transitory states such as activation and relaxation during a twitch contraction. For example, the affinity of fast-muscle troponin for calcium ions is 2 times greater than that for slow-muscle troponin (110, Table 3), and this may be one of the factors that determine the time course of twitch contractions of fast and slow muscles. B. Structure of Myosin Myosins isolated from muscles of different animal species (328) and from fast and slow muscles of one species at different times during ontogeny (405) have the same general shape, size, and molecular weight. Nevertheless, it is probable that there are structural differences between fast and slow myosins because these molecules differ in 3-methylhistidine content, rate of proteolysis, properties of low-molecularweight protein components, and enzymic properties. It was pointed out (sect. IVB) that 3-methylhistidine has not been found in myosin of fetal limb muscle (233, 403) or adult soleus muscle (255). In contrast,

178

R. I. CLOSE

Volume 52

there are about 1.3-1.7 residues of 3-methylhistidine per mole of myosin from longissimus dorsi (403) and psoas (255) muscles of adult rabbits and from flexor hallucis longus muscle of adult cats (255). Kuel and Adelstein (255) pointed out that all these adult fast muscles contain some slow fibers and that the presence of less than stoichiometric amounts of 3-methylhistidine in the myosin extracted from these muscles is probably attributable to the presence of slow-fiber myosin, with no 1tie thylhis tidine, mixed with fast-fiber myosin containing one residue of 3-methylhistidine per heavy chain, i.e., two residues per myosin molecule. Perry and his associates (233, 234) have carried out some experiments designed to determine the role of 3-methylhistidine in myosin; they showed that photo-oxidation of Inethylene blue-sensitized myosin decreased the content of 3-methylhistidine to 0.1 pmole/ mole with no reduction in ATPase and actin-combining activities of myosin, and they suggested that this amino acid is not directly involved in the biological activities of myosin. Proteolytic enzymes such as trypsin split the myosin molecule into heavy mole) and light (about 1.5 X lo5 g/mole) meromyosins at a (about 3.5 X lo5 g/ point about midway along the tail of the molecule (155, 223, 270, 292, 293, 445). The rate of hydrolysis appears to be directly correlated with the specific activity of myosin ATPase and intrinsic speed of contraction, being higher for rabbit muscle myosin than canine muscle myosin (297, 299) and higher for fast-muscle myosin than for slow-muscle myosin for a given animal species (156). These observations raise the possibility that similar myosin molecules may differ structurally within a segment of the molecule that is rnost sensitive to proteolysis. This segment may correspond to the postulated flexible linkage (reviewed in 193, 225) between movable heavy merornyosin (HMM), which is thought to project from the thick filament, which is bonded into the backbone of the thick and the light meromyosin (LMM), myosin filament. It is not known whether differences in the mechanical properties of the linkages between HMM and LMM contribute to differences in mechanical properties of the contractile material of fast- and slow-twitch muscles. Further proteolysis of HMM by papain yields two components, and the one designated subfragment I has the full ATPase and actin-binding activities of myosin (298). In addition to these major components a number of small protein components can be separated from myosin by various means; the literature on these low-molecularweight proteins (LMP) has been reviewed by Gibbons (159), Katz (241), and Young (446). The LMP fraction appears to be homogeneous in hydrodynamic studies but it is composed of several electrophoretically distinct components. The number and nature of these components seem to depend on the source and method of extraction of myosin (327) and the methods of dissociation and fractionation of the components (see below). The amount of LMP associated with myosin has usually been found to be between 10 % and 15 % by weight (142, 143, 153, 158, 266, 268, 406, 428), although a high value of 20 % LMP has been reported for succinylated myosin (3 17) and a low value of about 5 % LMP in highly purified preparations of myosin from which 5’-AMP deaminase, adenylate kinase, and nucleic acids had been removed (150); the mean value for the 10 reports is about 12 % LMP by weight. Average molecular weight of all the unfractionated LMP components of

Jmuclry

1972

DYNAMIC

PROPERTIES

OF

MUSCLE

179

myosin has been reported as 32,000 (143), 24,000 (142), and 20,000 (102, 158). Locker and Hagyard (266) separated three LMP components by gel electrophoresis and obtained molecular weights of 17,000, 19,000, and 20,000. Trotta et al. (406) used high pH to dissociate papain-prepared subfragment I of HMM into a light component about 1.8 X lo4 g/ mole and a heavy component about 8.6 X lo4 g/mole; since the total molecular weight of the two is the same as the parent subfragment I, they suggested there is 1 mole of light component per mole of subfragment I. Variations in the estimates of the amounts and molecular weights of LMP associated with myosin inevitably lead to difficulties in interpretation of the results in terms of stoichiometric relations between the various components of myosin and the possible role of the LMP. Despite this difficulty it has been established that U) all Illyosin preparations that have been examined contain LMP, 6) LMP is tightly bound to HMM (18, 102, 150, 158, 266) and some exclusively to subfragment I (252, 406), and c) removal of all LMP by various methods results in total lossof ATPase and actin-binding activities of myosin. It has not been established unequivocally that any part of the LMP component is a real subunit of the myosin molecule and not merely a contaminant that is always present in myosin that has been isolated by the usual methods of extraction employed at this time. Nevertheless, there are several very interesting observations that support the view that at least some parts of LMP are real subunits of myosin and that some of these may have important roles in determining some of the characteristic enzymic and actinbinding properties of myosin of different muscles. For example, isoleucine is the Cterminal residue of myosin, HMM, papain-prepared subfragment I of HMM, and the alkali-dissociated LIz/IP, whereas this amino acid is not an end group of the residual heavy alkali-dissociated component of myosin (158, 251, 372, 406). Atits ATPase and actintempts to remove all LMP from myosin while preserving binding activities have been unsuccessful (143, 150, 158). Alkali dissociation of LMP occurs between pH 9.5 and 10.5, and there is a concomitant change in C$+activated ATPasc of myosin (presumably from fast muscle) from full activity to zero, with complete loss of actin-binding ability (102, 158). In sirnilar investigations Dreizen and Gershman (10 1) showed that Ca2+- activated ATPase activity varies with the ratio of LMP to residual heavy myosin component, and Samaha et al. (364) found that “closely associated” proteins of fast-muscle myosin are released by p-chloromercuriphenylsulfonate at rates and in amounts inversely proportional to the inhibition of ATPase activity. Some attempts to restore myosin ATPase activity by adding LMP components to LMP-free myosin have been unsuccessful (150, 364) but in other instances partial restoration of biological activities has been achieved. Frederiksen and Holtzer (143) dissociated LMP frorn myosin at pH 11 and showed that neither the LMP nor the residual heavy component of myosin had ATPase or ac tin-binding activities; using unseparated mixtures of the two components they found that backtitration to neutral pH lead to recombination of the two with resulting recovery of full actin-binding ability and more than 50 % enzymic activity. Stracher et al. (395, 396) have made preliminary reports that there is complete loss of myosin ATPasc in the dissociating medium containing 4 M LiCl and 2 rnM ATP

180

R. I. CLOSE

Volume 52

but up to 60 % of the enzyrnic activity was restored when the LiCl was removed within 1 hr, and similar observations have been made by Dreizen and Gershman (101); furthermore, Stracher and his coworkers found that addition of previously separated LMP and heavy components in stoichiometric amounts lead to recovery of about 40 % of the original enzyme activity. These observations clearly suggest an important role of the light components in the biological activity of myosin. It was mentioned above that the number and electrophoretic mobilities of LMP components depend on the methods of dissociation and separation of LMP, as well as the source and method of extraction of myosin. There have been several reports that rabbit muscle myosin treated in various ways, including high pH, yields three electrophoretically distinct LMP components (102, 142, 150, 158, 266, 268, 428). Weeds (428) made the interesting observation that one of the three alkali-LMP components could be selectively dissociated from rnyosin by treatment with 5,5’-dithio-&s-(2 nitrobenzoic acid) (DTNB) and that, on regeneration of the thiol groups in DTNB-treated myosin, full ATPase activity was restored; similar observations were made by Gazith et al. (153). Other investigators used different methods of dissociation and obtained four major bands on electrophoresis (157, 327, 364, 366) but in two instances it was concluded that two of the four LMP componen ts were not essential for biological activity of rnyosin. Samaha et al. reported that two of the four components that they separated had the same electrophoretic mobilities as troponin and tropornyosin (364) and that these two components were identical in fast and slow muscles of the cat (366). Perrie and Perry (327) were able to prepare rnyosin with full ATPase activity in which one or the other of two LMP components was absent and two other LMP components were always present. All these observations suggest that there are only two main electrophoretically distinct LMP components that are essential for ATPase and actin-binding activities but the differences between these two LMP of fast and slow muscles have not been clearly defined. Samaha et al. (366) reported that these two LMP components of fast and slow muscles of cat differ in electrophoretic mobility, and Locker and Hagyard (268) obtained different electrophoretic patterns for LMP or rabbit fast and slow muscles. On the other hand, Perrie and Perry (327) found that the two LMP components that are essential for the biological activities of myosin have identical rnobilities in fast and slow myosins and that the principal difference is the amount of one of these LMP components present in the two types of myosin. Nevertheless, it seems clear that fast and slow rnyosins differ either qualitatively or quantitatively in LMP content, and in view of the probable role of these components it is possible that they determine the difference between ATPase activity of fast and slow myosins in a given animal species. The electrophoretic pattern and mobilities of LMP components are different for adult muscles of some species (267), but it is important to note that Figure 5 in the paper by Perrie and Perry (327) shows virtually no difference in the electrophoretic properties of LMP of myosin from rat and rabbit muscles despite differences in intrinsic speed of contraction of the muscles of these closely related species. There have been two main reports on the electrophoretic properties of LMP of myosin of developing muscle and in both instances distinct differences in electro-

January

1972

DYNAMIC

phoretic patterns of LMP ontogeny (267, 327).

PROPERTIES

components

have

OF

181

MUSCLE

been observed

C. Relation Between A TPase Activity of Myosin and Intrinsic

at different

Speed

of

stages in

Shortening

It has been shown above that there are proportional changes in intrinsic speed of shortening of sarcomeres and myosin ATPase activity during ontogenetic differentiation of fast and slow muscles and after nerve cross-union and that the two variables are directly proportional for a variety of muscles from different animal species (sect. IV-VI). This high degree of correlation between the two variables is consistent with the hypothesis that the maximum rate at which thick and thin filaments slide past one another during contraction is limited by the rate of hydrolysis of ATP by myosin at individual enzyme sites. Further support for this view has been obtained by studying the mechanical behavior of simple models formed from contractile material. Superprecipitation of actomyosin is probably the simplest model that exhibits the essential features of contraction of whole fibers. B&Any and his coworkers (22) showed that actin-activated ATPase of myosin is 3-4 times higher for cat muscles than for homologous muscles of the sloth and that in comparable experimental conditions the rate of superprecipitation of cat actomyosin is about 5 times higher than sloth actomyosin. These differences in kinetic properties of myosin are correlated with the four- to fivefold difference in isometric twitch contraction times of homologous muscles of the two species. In contrast, both the actin-binding capacity and the relation between degree of superprecipitation and myosin:actin ratio are virtually the same for cat and sloth myosin. Barany (21) has also shown that myosin of normal, self-innervated, and cross-innervated rat extensor digi torum longus and soleus muscles, combined with F-actin, superprecipitated at rates proportional to the intrinsic speeds of shortening of sarcomeres and specific activities of actinactivated ATPase of myosin. Similar results have been obtained by Furukawa and Peter (147) in their study of properties of natural actomyosin from normal and dystrophic human muscle; they found that the rate of superprecipita tion of actomyosin of dystrophic muscle was much lower than that for normal muscle and it was correlated with, but not proportional to, decreased specific ATPase activity of rnyosin of the pathologic muscles. Glycerol-extracted fibers have also been used to investigate differences in the contractile machinery of fast and slow muscles. Sexton and Gersten (380) found that the average initial rate of tension development in small bundles of glycerolextracted fibers of rat medial gastrocnemius muscle is 1.6-l. 7 times greater than for similar preparations of rat soleus fibers and that these rates are correlated with the speeds of contraction of the two muscles; furthermore, the maximum tension per unit cross-sectional area of the bundles of fibers is about the same for the two muscles (sect. IrrC). It appears that there has been no systematic comparative study of the kinetic properties of models of many different muscles with a wide range of speeds. An in-

182

R.

I. CLOSE

Volume

52

vestigation along these lines would probably yield interesting and valuable information, and it is encouraging to find that the rates of superprecipitation and ATPase activity of actomyosin are about the same for muscles of the rabbit and the horseshoe crab (19,20,412) despite possible differences in organization of the contractile material of these muscles. D. Functional

Dt@erences Between Fast and Slow Muscles

The general observation that there are differences in the kinetic properties of fast and slow myosin ATPases with little or no difference in the intrinsic strength of the contractile material is consistent with the roles of fast muscles in phasic movements and slow muscles in the sustained tonic activity required for maintenance of posture. There are at least two observations that give some insight into the functional significance of differences in intrinsic properties of the contractile material of fast and slow muscles. These are the differences between the two types of muscle in a) the efficiency of the contractile material in maintaining isometric tension (170) and 6) the shape of the force : velocity curve (74,80,429; C. Kean cited in 53). Goldspink et al. (170) measured the ATP content of resting and stimulated 2 ,4-dinitrofluorobenzenetrea ted muscles of the hamster, the difference between the two being the amount of ATP hydrolyzed by myosin, and they determined the relative efficiency of different muscles in terms of the amounts of isometric tetanic tension maintained and the ATP hydrolyzed in a given time. The results obtained for different hamster muscles (170, Table 3) are not strictly comparable because they did not take into account possible differences in muscle fiber lengths. In rnaking estimates of this kind it is important to bear in mind that, as a first approximation, the force developed by a particular type of contractile material is proportional to the cross-sectional dimensions of the fiber, assuming equal densities of packing of fibrils, and independent of fiber length, whereas the total amount of ATP hydrolyzed by myosin during a contraction will be proportional to the volume of the fiber; consequently, the measure of efficiency used by Goldspink et al. (170) is inversely related to the muscle fiber length. If one assumes that the average muscle fiber and thick filament lengths are about the same for hamster extensor digitorum longus and soleus, as they are in the rat and mouse, then it is clear from the results of Goldspink et al. (170, Table 3) that the rate of hydrolysis of ATP during an isometric tetanic contraction of extensor digitorunr longus is approximately twice that for soleus, whereas the maximum tension per unit cross-sectional area of muscle is about the same for the two muscles. In this sense the soleus is much more efficient than extensor digitorum longus in maintaining tension, and it was suggested (170) that this difference in efficiency is probably determined by differences in the duration of the period of attachment of actin and myosin in each cycle of formation and breakage of individual cross-linkages. Goldspink et al. (170) also concluded that extensor digitorum longus is 2.5-3 times more eficient in maintaining tension than either the diaphragm or biceps brachii muscles, but in the absence of information on fiber lengths their results are open to the alternative interpretation that the in-

January

DYNAMIC

1972

PROPERTIES

OF

MUSCLE

183

trinsic properties of the contractile material are about the same in these three muscles and that the fibers of biceps and diaphragm muscles are much longer than those of extensor digitorum longus. Katz (243) and Woledge (437) h ave shown that the force : velocity curve is much more curved for tortoise muscle than frog muscle, and although both curves could be fitted by Hill’s equation (207) the ratio a/P0 was about 0.25 for frog muscle and 0.07-o. 11 for tortoise muscle. Woledge (437) also showed that tortoise muscle is more efficient than frog muscle in converting free energy into work, as indicated by a higher ratio of work : (work + heat), and he hypothesized that increased mechanical efficiency is associated, presumably through some causal connection, with a more curved force : velocity curve. Woledge also discussed several possible alterations in the contractile material that would increase both the efficiency of the muscle and the curvature of the force : velocity curve, and there seems to be no reason why these arguments should not apply to mammalian muscles that also differ in a/p0 . For example, fast phasic muscles such as adult rat extensor digitorum longus and cat flexor hallucis longus have a/E, about 0.25-0.3 whereas a/p0 for rat and cat soleus muscles is about 0.15-O. 17 (53, 74, SO). Th e g enerality of Woledge’s thesis that there is an inverse relation between a/& and efficiency in conversion of free energy to mechanical work would appear to be contradicted by the recent work of Awan and Goldspink (9) on hamster muscles in which they reported that the ratio of work done to creatine phosphate utilized during isotonic contractions is higher for fast biceps brachii muscles contracting against loads of 5 g than for slow soleus muscles contracting against 2 g; however, it is difficult to interpret these observations because the mechanical efficiency varies with load and it was not made clear in their preliminary report (9) whether the ratio of isotonic load to maximum isometric tetanic tension was the same for both muscles. The observations of Awan and Goldspink (9) appear to corroborate the earlier work of Nakamura et al. (300) in which it was reported that isolated mydfibrils of rabbit “red” limb muscle are less efficient in shortening than myofibrils of the fast-twitch psoas muscle. Nakamura et al. (300) reported that the amount of ATP hydrolyzed by myofibrils during shortening from sarcomere length 1.8-l .O p was 1.6 times greater for myofibrils of muscle than myofibrils of the psoas; these results are very interesting but “red” again it is not possible to evaluate thern because the ‘cred” muscle was not named and it is not clear whether the muscle was fast or slow.

VIII.

REVIEW

OF

SOME

MAJOR

PROBLEMS

Mammalian muscle provides very valuable material for investigating certain aspects of activation and con traction. With regard to activation it is important to recognize that there is no evidence that marnmalian muscle fibers are fully activated during normal twitch contractions. From this viewpoint it is clear that determination of the maximum intensity of the active state is one of the major problems in the study of dynamics of twitch contractions. At this stage it seems unlikely that any significant advance toward an understanding of this aspect of activation will be

184

R. I. CLOSE

Volume

52

achieved until a method is developed for direct measurement of the time course of change of either the intensity of the active state or some related variable such as the sarcoplasmic concentration of free calcium ions. A development along these lines, together with information on the relations between speed of shortening, load, and sarcoplasmic calcium concentration, would make it possible to give a more detailed description of some of the main events in twitch contractions. The relations between factors that affect the time course of decline of the active state are virtually unknown. The problem is to determine the extent to which the active state is influenced by intrinsic properties of the contractile material and by extrinsic factors such as the amount of calcium released as activator and the rate of uptake of calcium by the sarcoplasmic reticulum. It is known that fragments of sarcoplasmic reticulum accumulate calcium in vitro but it has not yet been demonstrated that the rate of uptake of these ions is high enough to account for the rapid decline of the active state in vivo. On the other hand, there is a well-defined inverse relation between duration of the active state and intrinsic properties of the contractile material such as maximum speed of shortening and myosin ATPase activity. The relations between these variables are nearly identical for fast and slow lirnb muscles of the same and different species and are virtually unaltered after crossunion of the nerves of fast and slow muscles despite marked changes in intrinsic speed of shortening. These observations show that the intrinsic speed of the contractile material and the time course of decline of the active state are coupled in some way, and this introduces the possibility that there is a causal connection between the two variables. There are, therefore, at least two main hypotheses that must be tested; i.e., for a given amount of calcium released as activator, the duration of the active state is determined by either the rate of accumulation of calcium by the sarcoplasmic reticulum or the intrinsic speed of the contractile material. More information on the relations between the active state and the kinetic properties of both the contractile material and the sarcoplasmic reticulum could be obtained by studying the properties of heterologous muscles, such as m.edial rectus, thyroary tenoid, and hindlimb muscles, which differ markedly in relative peak twitch tension as well as isometric twitch contraction time. It is probable that these differences in twitch responses are determined by differences in intrinsic speed of shortening, the amount of ionic calcium released as activator, the rate of uptake of calcium by the sarcoplasmic reticulum, or a combination of these variables. It appears that the most widely accepted interpretation of this type of difference in twitch response is that it is due primarily to differences in rate of uptake of calcium by the sarcoplasmic reticulum. The alternative interpretation is that differences in normal isometric twitch contraction time are determined bY d ifferences in intrinsic speed of the contractile material and that differences in relative peak twitch tension are proportional to differences in the relative amounts of calcium liberated as activator. The possibility that the relation between intrinsic speed of the contractile material and isometric twitch contraction time is the same for medial rectus, thyrarytenoid, and hindlimb muscles could be tested experimentally, and the result would provide valuable information on the generality of the hypothesis in question. A comparative investigation of the physiological and biochemical properties of

January

DYNl4MIC

1972

PROPERTIES

OF

185

MUSCLE

muscles that differ markedly in relative peak twitch tension and speed would contribute toward an understanding of the factors determining the time course of decline of the active state. There is a wide range of intrinsic speeds of shortening of the contractile material of different mammalian muscles. It has been shown that the intrinsic speed of shortening is proportional to the specific activity of myosin ATPase, and this strongly supports the hypothesis that hydrolysis of ATP by myosin is the rate-limiting step in the shortening process. The next stage in the investigation is to find out what properties of the myosin molecule determine the kinetic properties of the ATPase site. This is essentially a biochemical problem but certain physiological observations are relevant. For example, the limb muscles of a given species appear to have the same intrinsic speed of shortening before ontogenetic differentiation of fast and slow muscles. The intrinsic speed of shortening of slow-twitch muscles such as soleus, and presumably soleus myosin ATPase, is unchanged during differentiation of fast and slow muscles but it is inversely related to the adult body size of the on these size-dependent differences are developlnental species. Superimposed changes that lead to two- to threefold increases in intrinsic speed of shortening and myosin ATPase activity of fast muscles of the hindlimb. This pattern raises the possibility that more than one factor may contribute to differences in ATPase activity of different myosins from hindlimb muscles. It is conceivable that one kind of difference in molecular structure of myosin may determine the size-dependent differences in ATPase activity, whereas another structural modification may act as a multiplier of specific activity of ATPase in fast muscle myosin. I thank

Professor

D. R. Wilkie

and Dr.

A. R. Luff

for helpful

comments

on the

manuscript

REFERENCES 1. ALBUQUERQUE, E. X., AND R. J. McISAAC. Early development of acetylcholine receptors on fast and slow mammalian skeletal muscle. Life Sci. (Oxfad) 8 : 409416, 1969. 2. ALBUQUERQUE, E. X., AND R. J. McISAAC. Fast and slow mammalian muscles after denervation. i?x/ll. Neural. 26 : 183-202, 1970. 3. ALBUQUERQUE, E. X., AND S. THESLEFF. A comparative study of membrane properties of innervated and chronically denervated fast and slow skeletal muscles of the rat. rlc!a Pi~ysiol. &and. 73: 471480, 1968. 4. ALDER, A. B., G. N. C. CRAWFORD, AND R. G. EDWARDS. The growth of the muscle tibialis anterior in the normal rabbit in relation to the tensionlength ratio. Proc. Roy. Sot. (London), Ser. fi 148: 207216, 1958. 4a.ALDER, A. B., G. N. C. CRAWFORD, AND R. G. EDWARDS. The effect of limitation of movement on longitudinal muscle growth. Froc. Koy. Sot. (London), Ser. R 150: 554-562, 1959. 5. ALOISI, M., AND A. MARGRETH. The question of glycolytic enzymes localization in the skeletal muscle fiber and its bearing upon certain aspects of muscle diseases. In: Exploratory Concepts in Muscular Dys-

6.

7.

8.

9.

10.

11.

12.

trophy and Related Disorders, edited by A. T . Milhorat. Amsterdam : Excerpta Med. Found., 1967, p. 305-3 17. ANDERSEN, P., AND T . A. SEARS. The mechanical properties and innervation of fast and slow motor units in the intercostal muscles of the cat. J. Yhysiof. (London) 173 : 114-l 29, 1964. ANGGARD, L., AND D. OTTOSON. Observations on the functional development of the neuromuscular apparatus in fetal sheep. Exptl. Neural. 7 : 294-304, 1963. ASHLEY, C. C., AND E. B. RIDGWAY. On the rela tionships between membrane potential, calcium transient and tension in single barnacle muscle fibres. J. I’hysiol. (London) 209: 105-130, 1970. AWAN, M. Z., AND G. GOLDSPINK. Energy utilization by mammalian fast and slow muscle in doing external work. Biochim. Biophyskta 2 16 : 229230, 1970. BACH-Y-RITA, P., AND F. ITO. In vivo studies on fast and slow muscle fibers in extraocular mtlscles. J. Gen. Physiol. 49 : 1177-l 198, 1966. BAHLER, A. S. Series elastic component of mammalian skeletal muscle. Am. J. Fhysiol. 213 : 15601564, 1967. BAHLER, A. S. Modeling of mammalian skeletal

186

13.

14.

15.

16.

17.

18. 19.

20.

21.

22.

23.

24

25.

26.

27.

28.

29.

30.

31.

R. I. CLOSE

muscle. IEEE Trans. Zjio-Med. Eng. 15: 249-257, 1968. BAHLER, A. S., J. T . FALES, AND K. L. ZIERLER. The active state of mammalian skeletal muscle. J. Gen. Physiol. 50 : 2239-2253, 1967. BAHLER, A. S., J. T . FALES, AND K. L. ZIERLER. The dynamic properties of mammalian skeletal muscle. J. Gen. Physiol. 51 : 369-384, 1968. BANKER, B. Q. A phase and electron microscopic study of dystrophic muscle. II. The pathological changes in the newborn Bar Harbor 129 dystrophic mouse. J. Neuropathol. ExptZ. Neural. 27 : 183-209, 1968. B,4NU, G. Recherches physiologiques SUI Ze de’veloppement neuromusculaite ches Z’homme et Z’animal. Paris: Maretheux, 1922. BANUS, M. G., AND A. ZETLIN. The relation of isometric tension to lengths in skeletal muscle. J. Cellular Camp. Physiol. 12 : 403-420, 1938. BARANY, K., AND I-1. OPPENHEIMER. Succinylated meromyosins. Nature 213 : 626-627, 1967. B.ARANY, M. ATPase activity of myosin correlated with speed of muscle shortening. J. Gen. Z’hysiof. 50: 197-216, 1967. BARANY, M., K. BARANY, T . RECKARD, AND A. VOLPE. Myosin of fast and slow muscles of the rabbit. Arch. Biochem. Biophys. 109 : 185-191, 1965. BARANY, M., AND R. I. CLOSE. The transformation of myosin in cross-innervated rat muscles. J. Z’hysiol. (London) 213 : 455-474, 197 1. BARANY, M., T . E. CONOVER, L. H. SCHLISELFELD, E. GAETJENS, AND M. GOFFART. Relation of properties of isolated myosin to those of intact muscles of the cat and sloth. European J. Biochem. 2 : 156-164, 1967. BARANY, M., A. F. TUCCI, K. BARANY, A. VOLPE, AND T . RECKARD. Myosin of newborn rabbits. Arch. Biochem. Biophys. 111 : 727, 1965. BARNARD, R. J., V. R. EDGERTON, ‘I’. FURUKAWA, AND J. B. PETER, Histochemical, biochemic:ll and contractile properties of red, white and intermediate fibers. Am, J. Physiol. 220: 410-414, 1971. BARNARD, R. J., V. R. EDGERTON, AND J. B. PETER. Effect of exercise on skeletal muscle. I. Biochemical and histochemical properties. J. &pi. Physiol. 28 : 762-766, 1970. BARNARD, R. J., V. R. EDGERTON, AND J. B. PETER. Effect of exercise on skeletal muscle. II. Contractile properties. J. A/@. PhysioZ. 28: 767-770, 1970. BASSO, N., I?. BLAQUIER, I. J. DE LA RIVA, AND A. C. TAQUINI. Changes in the water and electrolyte distribution in the growing rat. Acta Z’hysiol. Latinoam. 18 : 19-30, 1968. BEATTY, C. H., G. M. BASINGER, AND R. M BOCEK. Differentiation of red and white fibers in muscle from fetal, neonatal and infant rhesus monkeys. .I. Histochem. Cytochem. 15 : 93-l 03, 1967. BERGMAN, R. A. Observations on the morphogenesis of rat skeletal tissue. Bull. Johns IIopkins 110s~. 110: 187-201, 1962. BERNHARD, C. G., U. S. VON EULER, AND C. R. SKOGLUND. Post-tetanic action potentials in mammalian muscle. Acta Physiol. Stand. 2 : 284-288, 1941. BINKHORST, R. A. The effects of training on some isometric contraction characteristics of a fast muscle. Axh. Ges. Physiol. 309 : 193-202, 1969.

Volume

52

32. BISCOE, T . J. Th e isometric contraction characteristics of cat intercost: muscle. J. I’hysiol. (London) 164: 189-199, 1962. 33. BISCOE, T . J., AND ii. TAYLOR. The effect of admixture of fast and slow muscle in dctcrmining the form of the muscle twitch. .&fed. BioZ. L’ng. 5: 473479, 1967. 34. BLECKMANN, I., I. JURNA, AND W. RUMMEL. Vergleichende Untcrsllchungen von Methoden zur Bestimmung der Plateaudauer des aktiven Zustandes am Warmbliitermuskel. Verhalten des Soleus und Tibialis anterior der Katze unter Sympathicomimctica und Coffein. .tlrch. Ges. I’hysiol. 277 : 422433, 1963. 35. BOCEK, R. M., AND C. H. BEATTY. Glycogen synthetase and phosphorylase in red and white muscle of rat and rhesus monkey. J. Histochem. Cytothem. 14: 549-559, 1966. 36. BOETHIUS, J. Resting membrane potential in neck and leg muscles of young rats. Acta Physiol. Stand. 75 : 253-254, 1969. 37. BOWMAN, W. C., A. A. GOLDBERG, AND C. RAPER. A comparison between the effects of a tetanus and the effects of sympathomimetic amines on fast- and slow-contracting skeletal muscles. Brit. J. Pharmacol. Chemotherap. 19 : 464-484, 1962. 38. BOWMAN, W. C., A. A. J. GOLDBERG, AND C. RAPER. Posttetanic and drug-induced repetitive firing in the soleus muscle of the cat. Brit. J. Pharmacol. Chemotherap. 35 : 62-78, 1969. 39. BOYD, I. A., AND A. R. MARTIN. Membrane constants of mammalian muscle fibres. .I. Physiol. (Londdon) 147 : 450-457, 1959 40. BRITTON, S. W. Form and function in the sloth. Quart. Rev. BioZ. 16 : 13-34, 190-207, 1941. 41. BRODY, S., Bioenergetics and Growth. New York : Reinhold, 1945. 42. BROOKE, M. H., AND K. K. KAISER. Some comments on the histochemical characterization of muscle adenosine triphosphatase. J. Histochem. Cytochem. 17: 431432, 1969. 43. BROOKE, M. l-l., AND K. K. KAISER. Muscle fiber types: how many and what kind? Arch. Neurof. 23 : 369-379, 1970. 44. BROOKE, M. H., AND K. K. KAISER. Three “ myosin adenosine triphospha tase” sys terns : the nature of their pH lability and sulfhydryl dependence. J. Histochem. Cytochem. 18 : 670-672, 1970. 45. BROWN, G. L., AND U. S. VON EULER. The after effects of a tetanus on mammalian muscle. J. Physiol. (London) 93 : 39-60, 1938. 46. BROWN, M. C., AND P. B. C. MATTHEWS. The effect on a muscle twitch of the back-response of its motor nerve fibres. J. Z’hysiof. (London) 150 : 332-346, 1960. 47. BRUST, M. Combined effects of nitrate and caffeine on contractions of skeletal muscles. Am. J. Physiol. 208: 431435, 1965. 48. BRUST, M., AND 11. W. COSLA. Contractility of isolated human skeletal muscle. Arch. Phys. Med. Rehabil. 48 : 543-555, 1967. 49. BRYANT, S. H. Cable properties of external intercostal muscle fibres from myotonic and nonmyotonic goats. J. Physiol. (London) 204: 539-550, 1969. 50. BUCHTI-IAL, F., AND 1-l. SCHMALBRUCH. Spectrum of contraction times of different fibre bundles in the brachial biceps and triceps mtiscles of man. 1Vature 222 : 89, 1969.

January

1972

DYNAMIC

PROPERTIES

51. BI:(:f l’rllAT+ F., ANI) 11. SCITM.41,BRUCT-T. Contraction times and fibre types in intact human muscle. .,lcta I’hysiol. &and. 79 : 435-452, 1970. 52. BLJLLtZRD, Ii. H. Histological as related to physiological and chemical differences in certain muscles of the cat. Johns IIopkins Ilo@. Rep. 18: 323-328, 1919. _ control of the contractile 53. BULLER, A. ,J. Th e nellral mechanism in skeletal mllscle. Endeauour 29: 107-l 11, 1970. R. ED.54. BULLER, A. .J., A. C. DORNHORST, IYARDS, D. KERR, AND R. F. WHELAN. Fast and slow muscles in mammals. il’cllure 183 : 15 16-l 5 17, 1959. 55. BUT~LI’R, A. J., J. C. ECCLES, ANI) R. M. ECCLES. 1)ifferentiation of fast and slow muscles in the cat hind liltIf>. .J. i’hysioi. (London) 150: 399-416, 1960. 56. BULLER, A. J., J. C. ECCLES, AND R. M. ECCLES. Interactions between motoneurones and muscles in rcspcct of the characteristic speeds of their rcsponscs. J. /‘hysiol. (London) 150 : 417-439, 1960. 57. BULTXR, A. J., C. J. C. KEAN, AND K. W. RANATUNGA. ‘I‘he force :velocity characteristics of cat fast and slow-twitch skeletal muscle following crossinnervation. J. I’hysiol. (London) 213 : 66-67P, 1971. 58. BULTXR, A. .J., AND D. M. LEWIS. Factors affecting the differentiation of mammalian fast and slow muscle tihres. In : The cflect of Use and Disuse on Neuromuscular Functions, edited by E. Gutmann and P. Hnik. Prague : Czech. Acad. Sci., 1963, p. 149-l 59. 59. BULLER, A. J., AND D. M. LEWIS. The rate of tension development in isometric tetanic contractions of mammalian fast and slow skeletal muscle. J. i’hysiol. (1,ondon) 176 : 337-354, 1965. 60. BULTJER, A. ,J., AND D. M. LEWIS. Further observations on the differentiation of skeletal muscles in the kitten hind limb. J. Physiol. (London) 176 : 355370, 1965. 61. BI JT,T,ER, A. J., AND I). M. I,EWIS. Further observations on mammalian cross-innervated skeletal muscle. J. Fhysiol. (London) 178 : 343-358, 1965. (;2. BUT,LER, A. J., D. M. LEWIS, AND R. M. A. P. RIDGE. Some electrical characteristics of fast twitch and slow twitch skeletal muscle of the cat. J. physiol. (London) 180 : 29P, 1965. 63. BULLT’R, A. J., W. F. IT. M. MOMMAERTS, AND K. SERAYD-4RIAN. Enzymic properties of myosin in fast and slow twitch muscles of the cat following cross-innervation. J. Physiol. (London) 205 : 581-597, 1969. 64. BULLER, A. J., K. W. RANATUNGA, AND J. M. SMTTl-T. The influence of temperature on the contractile characteristics of mammalian fast and slow twitch muscles. J. Physiol. (London) 196: 82P, 1968. ti5. BULLER, A. J., K. W. R.4NATUNGA, AND J. SMITH. Tnfltlence of temperature on the isometric myograms of cross-inncrva ted mammalian fast twitch and slow twitch skeletal mtlscles. Nature 218 : 877-878, 1968. ri6. BURKE, R. E. Motor unit types of cat triceps surae muscle. J. Physiol. (London) 193 : 141-160, 1967. 67. BlJRKE, R. E., P. RUDOMIN, AND F. E. ZAJAC. Catch property in single mammalian motor units. bScience 168 : 122-l 24, 1970. 68. f3USS, W. C., AND G. B. FRANK. Calcium and excitation-contraction coupling in mammalian skeletal muscle. Arch. Intern Pharmacodyn. Therap. 181 : 15-26, 1969. 69. CARLSEN, I:., 1:. FUCI IS, AND G. (;. KNAPPEIS.

OF

MUSCLE

187

Contractility and ultrastructure in glycerol-extracted muscle fibers. I. The relationship of contractility to sarcomere length. J. cell Biol. 27 : 25-34, 1965. 70. CARLSEN, F., 1~. FUCHS, AND G. (:. KNAPPETS. Contractility and ultrastrlIcture in glycerol-extracted fibers. II. Ultrastructure in resting and shortened fibers. J. Cell Biol. 27 : 35-46, 1965. 71. CASSENS, R. G., W. G. HOEKSTRA, E. C:. FAT,TIN, AND E. .J. BRTSKEY. Zinc content and subcellular distriblltion in red vs. white porcine skeletal muscle. ~ltn. J. Physiol. 212 : 688-692, 1967. 72. CHAPLIN, E. R., G. W. NELL, AND S. M. WALKER. Excitation-contraction latencies in postnatal muscle fibers. Exptl. Neurof. 29 : 142-l 51, 1970. 73. CLOSE, R. Th e pattern of activation in the sartorius muscle of the frog. J. G‘en. Physiof. 46 : l-18, 1962. 74. CLOSE, R. D ynamic properties of fast and slow skeletal muscles of the rat during development. J. Physiol. (London) 173 : 74-95, 1964. 75. CLOSE, R. Th e relation between intrinsic speed of shortening and duration of the active state of muscle. J. Physiol. (London) 180 : 542-559, 1965. 76. CLOSE, R. F orce : velocity properties of mouse muscles. Nature 206 : 7 18-719, 1965. 77. CLOSE, R. Effects of cross-union of motor nerves to fast and slow skeletal muscles. Nature 206: 831-832, 1965. 78. CLOSE, R. Properties of motor units in fast and slow skeletal muscles of the rat. J. Physiol. (London) 193 : 45-55, 1967. 79. CLOSE, R. Dynamic properties of fast and slow skeletal muscles of mammals. In : Exploratory Concepts in Muscular I>ystrophy and Related Disorders, edited by A. T . Milhorat. Amsterdam: Excerpta Med. Found., 1967, p. 142-14-g. 80. CLOSE, R. Dynamic properties of fast and slow skeletal muscles of the rat after nerve cross-union. J. Physiol. (London) 204 : 33 l-346, 1969. 81. CLOSE, R. Neural influences on physiological properties of fast and slow limb muscles of mammals. In: Contractility of Muscle Cells and Related Processes. Symp. Sot. Gen. Physiol., in press. . 82. CLOSE, R. Length dependence of isometric twitch contractions of frog sartorius muscle. In: I’roc. Australian Physiol. Yharmacol. Sot. 2: 28-29, 1971. 83. CLOSE, R., AND J . 1:. Y. HOlI. Force:velocity properties of kitten mllscles. J. I’hysiof. (London) 192 : 815-822, 1967. 84. CLOSE, R., AND .J. F. Y. TlOII. The after-effects of repetitive stimulation on the isometric twitch contraction of rat fast skeletal mllscle. J. Yhysiol. (London) 197: 461477, 1968. 85. CLOSE, R., AND ,J. F. Y. HOH. Influence of temperature on isometric contractions of rat skeletal muscles. Nature 217 : 1179-l 180, 1968. 86. CLOSE, R., AND J. F. Y. HOH. Post-tetanic potentiation of twitch contractions of cross-innervated rat fast and slow muscles. Nature 221 : 179-181, 1969. 87. CLOSE, R. I. Development of contractile activity in frog striated muscle (PhD Thesis). Urbana, Ill. : Univ. of Illinois, 1960. 88. COOPER, C. C., R. G. CASSENS, L. L. KASTI’NSCHMIDT, AND E. J. BRTSKEY. Tiistochemical characterization of muscle differentiation. Develop. Bioi. 23: 169-184, 1970. 89. COOPER, S., AND .J. C. ECCLES. The isometric responses of mammalian muscles. J. Physiol. (London) 69 : 377-385, 1930.

188

R. I. CLOSE

89a.CRAWFORD, G. N. C. Experimentally induced hypertrophy of growing voluntary muscle. F’roc. Roy. Sot. (London), Ser. B 154: 130-138, 1961. 90. CREESE, R. Sodium exchange in rat muscle. Nature 201: 505-506, 1964. 91. CREESE, R., J. L. D’SILVA, AND S. E. E. HASHISH. Inulin space and fibre size of stimulated rat muscle. J. Physiol. (London) 127 : 525-532, 1955. 92. CULLINGHAM, P. J., A. R. LIND, AriD R. J. MORTON. The maximal isometric tetanic tensions developed by mammalian muscle in situ, at different temperatures. Quart. J. Exptl. I hysiol. 45 : 142-156, 1960. 93. DENNY-BROWN, D. E. The histological features of striped muscle in relation to its functional activity. hoc. Roy. Sot. (London), Ser. B 104 : 37 I-41 1, 1929. 94. DENNY-BROWN, D. E. Experimental studies pertaining to hypertrophy, regeneration and degeneration. Res. Pub/. Assoc. Res. Neruous Mental Disease 38: 147-196, 1960. 95. DESMEDT, J. E. Th e isometric twitch of human muscle in the normal and the dystrophic states. In: Exploratory Concepts in Muscular Dystrophy and Related Disorders, edited by A. T . Milhorat. Amsterdam: Excerpta Med. Found., 1967, p. 224-231. 96. DESMEDT, J. E., AND K. HAINAUT. Kinetics of myofilament activation in potentiated contraction : staircase phenomenon in human skeletal muscle. Nature 217 : 529-532, 1968. 97. DIAMOND, .J., AND R. MILEDI. A study of foetal and new-born rat muscle fibres. J. Fhysiol. (London) 162 : 393-408, 1962. 98. DOI, Y. Studies on muscular contraction. I. The influence of temperature on the mechanical performance of skeletal and heart muscle. J. Physiol. (London) 54 : 218-226, 1920. 99. DORN, A. Studien zur Skeletmuskelentwicklung heim Meerschweinchen. II. Topochemische Untersuchungen einiger Oxidoreduktasen. Acta Histochem. 33 : 362-393, 1969. 100. DRAHOTA, Z. The ionic composition of various types of striated muscles. 1 hysiol. Bohemoslov. 10: 160-166, 1961. 101. DREIZEN, P., AND L. C. GERSHMAN. Relationship of structure to function in myosin. II. Salt denaturation and recombination experiments. Biochemistry 9 : 1688-l 693, 1970. 102. DREIZEN, P., L. C. GERSHMAN, P. P. TROTTA, AND A. STRACHER. Myosin. Subunits and their interactions. J. Gen. Physiol. 50, Suppl. 85 : 1967. 103. DUBOWITZ, V. Enzyme histochemistry of skeletal muscle. Part I. Developing animal muscle. J. Neural. Neurosurg. Psychiat. 28 : 516-5 19, 1965. 104. DUBOWITZ, V. Enzyme histochemistry of skeletal muscle. Part 2. Developing human muscle. J. Neural. Neurosurg. Psychiat. 28 : 519-524, 1965. 105. DUBOWITZ, V. Pathology of experimentally reinnervated skeletal muscle. J. Neural. Neurosurg. Psychiat. 30 : 99-l 10, 1967. 106. DUBOWITZ, V. Cross-innervated mammalian skeletal muscle : histochemical, physiological and biochemical observations. J. I hysiol. (London) 193 : 481-496, 1967. 107. DUBOWITZ, V., AND D. L. NEWMAN. Change in enzyme pattern after cross-innervation of fast and slow skeletal muscle. Nature 214 : 840-841, 1967. 108. DUBOWITZ, V., AND A. G. PEARSE. A compara-

Volume

52

tive histochemic;ll study of oxidative enzyme and phosphorylase activity in skeletal muscle. Histochemie 2: 105-l 17, 1960. 109. EBASHI, S., AND M. ENDO. Calcium ion and muscle contraction. Progr. Biophys. Mol. Biol. 18 : 123183, 1968. 1 IO. EBASHI, S., M. ENDO, AND I. OHTSUKI. Control of muscle contraction. Ouart. Rev. Biophys. 2 : 351384, 1969. 111. EBERSTEIN, A., AND .J. GOODGOLD. Slow and fast twitch fibers in human skeletal muscle. :lnc. J. Physiol. 2 15 : 535-541, 1968. 112. ECCLES, J. C. Problems of plasticity and organization at the simplest levels of mammalian central nervous system. Ferspectives Rio/. Med. 1 : 379-396, 1958. 113. ECCLES, J. C. The effects of nerve cross-union on muscle contraction. In: Exploratory Concepts in Muscular Dystrophy and Related Disorders, edited by A. T . Milhorat. Amsterdam : Excerpta Med. Found., 1967, p. 151-160. 114. ECCLES, J. C., R. M. ECCLES, AND W. KOZAK. Further investigations on the influence of motoneurones on the speed of muscle contraction. J. Physiol. (London) 163 : 324-339, 1962. 115. ECCLES, J. C., R. M. ECCLES, AND A. LUNDBERG. The action potentials of the alpha motoneurons supplying fast and slow muscles. J. Fhysiol. (London) 142 : 275-291, 1958. 116. ECCLES, J. C., AND W. J. O’CONNOR. Responses which nerve impulses evoke in mammalian striated muscles. J. f hysiol. (London) 97 : 44-102, 1939. 117. EDGERTON, V. R., L. GERCHMAN, AND R. CARROW. Histochemical changes in rat skeletal muscle after exercise. Exptl. Neural. 24 : 110-I 23, 1969. 118. EDGERTON, V. R., AND D. R. SIMPSON. The intermediate muscle fiber of rats and guinea pigs. J. Histochem. Cytochem. 17 : 828-838, 1969. 119. EDGERTON, V. R., AND D. R. SIMPSON. Dynamic and metabolic relationships in the rat extensor digitorum longus muscle. Exptf. Neurof. 30: 374-376, 1971. 120. EDMAN, K. A. P. The rising phase of the active state in single skeletal muscle fibres of the frog. :lcta i-hysiol. Stand. 79: 167-173, 1970. 12 1. EDSTRGM, L., AND E. KUGELBERG. Properties of motor units in the rat anterior tibia1 muscle. Flcta Physiol. Stand. 73 : 543-544, 1968. 122. EDSTRGM, L., AND E. KUGELBERG. Histochemical composition, distribution of fibers and fatiguability of single motor units. J. Newof. Newoswg. Psychiat. 31 : 424-433, 1968. 123. EDSTRGM, L., AND B. NYSTR6M. Histochemical types and sizes of fibers in normal human muscles. Acta Neural. &and. 45 : 257-269, 1969. 124. ELLIOT, G. F., J. LOWY, AND C. R. WORTHINGTON. An X-ray and light-diffraction study of the filament lattice of striated muscle in the living state and in rigor. J. Mol. Rio/. 6 : 295-305, 1963. 124a.ELLIOTT. D. H., AND G. N. C. CRAWFORD. The thickness and collagen content of tendon relative to the cross-sectional area of muscle during growth. Proc. Roy. Sot. (London), Ser. B 162 : 198-202, 1965. 125. ELMQVIST, D., T . R. JOHNS, AND S. THESLEFF. A study of some electro-physiological properties of human intercostal muscle. J. Physioi. (London) 154: 602-607, 1960.

January

1972

DYNAMIC

PROPERTIES

126. ENGEL, W. K. The essentiality of histo- and cytochemical studies of skeletal muscle in the investigation of neuromuscular disease. h;purolq,oy 12 : 778-784, 1962. 127. ENGEL, W. K. Selective and nonselective susceptibility of muscle fiber types. Arch. KeuroL. 22 : 97-117, 1970. 128. ENGEL, W. K., AND G. KARPATI. Impaired neonatal skeletal muscle maturation following neurectomy. Develop. Biol. 17 : 713-723, 1968. 129. EVERSOLE, L. R., AND S. M. STANDISH. Histochemical demonstration of muscle fiber types. J. Hzstochem. Cytochem. 18 : 591-593, 1970. 130. FANBURG, B. L., D. B. DRACHMAN, D. MOLL, AND S. I. ROTH. Calcium transport in isolated sarcoplasmic reticulum during muscle maturation Nature 218: 962-964, 1968. 131. FARMER, T . W., F. BUCHTHAL, AND P. ROSENFALCK. Refractory period of human muscle after passage of a propagated action potential. Neurophysiol. Electroencejdzalog. Clin. 12 : 455-466, l960. 132. FERRONI, A., AND D. BLANCHI. Potenziali intracellulari della fibra muscolare striate di mammifero in soluzioni iposodiche. Boll. Sot. Ital. Biol. Sper. 40: 21 I-213, 196+. 133. FERRONI, A. AND D. BLANCHI. VelocitB di depolarizzazione della singola fibra muscolare a diverse concentrazioni sodiche. Boll. Sot. Ital. Biol. Sper. 40 : 1106-l 109, 1964. 134. E’ERRONI, A., D. BLANCHI, AND E. MEDA. Controllo de1 voltaggio della fibra muscolare striata de1 mammifero. Boil. Sot. Ital. Biol. Sper. 44 : 16431646, 1968. 135. FERRONI, A., L. CEDRINI, AND 1). LOVISOLO. Relazione corren te-vol taggio della fi bra muscolare striata de1 mammifero, in condizioni di equilihrio stabile. I. ConducibilitB delle membrana per il K+ e Cl-. Boll. Sot. Ztai. Biol. ,Sj!w. 44 : 1646-l 649, 1968. 136. FENN, W. O., AND B. S. MARSH. Muscular force at different speeds of shortening. J. Fhysiol. (London) 85: 277-297, 1935. 137. FEX, S. “Trophic” influence of implanted fast nerve on innervated slow muscle. Physiol. Bohemoslov. 18: 205-208, 1969. 138. FEX, S., AND B. SONESSON. Histochemical observations after implantation of a ‘fast’ nerve into an innervated mammalian ‘slow’ skeletal muscle. 77 : l-10, 1970. 139. FIEHN, W., AND .J. B. PETER. Properties of the fragmented sarcoplasmic reticulum from fast-twitch and slow-twitch muscles. J. Ciin. Znwst. 50: 570-573, 1971. 140. FISCHER, I I. Zur Physiologie der quergestreiften Muskeln der Sgugetiere. Arch. Ges. Physiol. 125: 541-583, 1908. 141. FLEAR, C. ‘I’. G., R. F. CRAMPTON, AND D. M. MATTHEWS. An in vitro method for the determination of the inulin space of skeletal muscle with observations on the composition of human muscle. Cfin. S’ci. 19 : 483-493, 1960. 142. FRIIARSON, N., AND J, THOMAS. The lowmolecular-weight components of bovine skeletal myosin. Biochem. J. 1 18 : 57-58P, 1970. 143. FREDERIKSEN, 1). W., AND A. HOLTZER. The sllbstructure of the myosin molecule. Production and properties of the alkali suhtlnits. Biochemistry 7 : 39353950, 1968.

,4nat.

Acta

OF

MUSCLE

189

144. FUDEL-OSIPOVA, S. I., AND 0. A. MARTYNENKO. Early ontogenic development of membrane potential of muscle fibres in rats. Federation Proc. 23 : T28-T30, 1964. 145. FUDEL-OSIPOVA, S. I., AND 0. A. MARTYNENKO. Change in the water-salt composition of rat muscles during ontogenesis. Biophysics 10 : 879885, 1965. 146. FULTON, J. F. The influence of shortening on the size of the action current and the duration of the mechanical response of skeletal muscle. Am. J. Physiol. 75 : 235-260, 1925. 147. FURUKAWA, T., AND J. B. PETER. Superprecipitation and adenosine triphosphatase activity of myosin B in Duchenne muscular dystrophy. Neurology 21: 920, 1971. 148. FURUSAWA, K., A. V. HILL, AND J. L. PARKINSON. The dynamics of “sprint” running. I-rot. Koy Sot. (London), Ser. B 102: 2942, 1927. 149. GABEL, L. I’., C. CARSEN, AND E. VANCE. Active state of muscle and the second and third derivatives of twitch tension. Am. J. Fhysiol. 214 : 1025-1030, 1968. 150. GAETJENS, E., K. BARANY, G. BAILIN, H. OPPENHEIMER, AND M. BARANY. Studies on the low molecular weight protein components m rabbit skeletal myosin. Arch. Biochem. Biophys. 123 : 82-96, 1968. 151. GASSER, H. S., AND A. V. HILL. The dynamics of muscular contraction. Proc. Roy. Sot. (London), Ser. B 96: 398437, 1924. 152. GAUTHIER, G. F. On the relationship of ultrastructural and cytochemical features to color in mammalian skeletal muscle. 2. L?ellforsch. Mikro.rkop. hat. 95: 462482, 1969. 153. GAZITH, .J., S. HIMMELFARB, AND W. F. HARRINGTON. Studies on the suhtmit strllcture of myosin. J. Bid. Chem. 245 : 15-22, 1970. 154. GEFFEN, L. B. Optimum length for contraction of rat circulated limb muscles. Arch. Intern. Physiol. 72 : 825-834, 1964. 155. GERGELY, J. Studies on myosin-adenosinetriphosphatase. J. Biol. Chem. 200: 543-550, 1953. 156. GERGELY, J., D. PRAGAY, A. F. SCHOLZ, J. C. SEIDEL, F. A. SRfiTER, AND M. M. THOMPSON. Comparative studies on white and red muscle. In: Molecular Biology of Muscular Contraction, edited by S. Ebashi, F. Oosawa, T . Sekine, and Y. Tonomura. Amsterdam : Elsevier, 1965, p. 145-l 59. 157. GERSHMAN, L. C., AND P. DREIZEN. Relationship cf structure to function in myosin. I. Subunit dissociation in concentrated salt solutions. Biochemistry 9: 1677-1687, 1970. 158. GERSHMAN, 1~. C., I’. DREIZEN, AND A. STRACHER. Subunit structure of myosin. II. Heavy and light alkali components. 1‘70~. Natl. Acad. Sci. U.S. 56 : 966-973, 1966. 159. GIBBONS, I. R. The biochemistry of motility. Ann. Rev. Biochem. 37 : 521-546, 1968. 160. GILLESPIE, C. A., D. R. SIMPSON, AND V. R. EDGERTON. High glycogen content of red as opposed to white skeletal muscle fibers of guinea pigs. J. Histochem. Cytochem. 18 : 552-558, 1970. 161. GLEBOVSKII, V. D. Contractile properties of respiratory muscles in fully grown and neonate animals. Fi~tiol. Zh. SSSR 47 : 470-480, 1961. 162. GOFFART, M., 0. HOLMES, AND %. M. BACQ.

190

R. I. CLOSE

Some mechanical properties of skeletal muscle in the sloth (ChoZoe@s hoflmanni Peters). Arch. Intern. l’hysiol. 70: 103-106, 1962. 163. GOFFART, M., AND J. M. RITCHIE. The effect of adrenaline on the contraction of mammalian skeletal muscles. J. I llysiol. (London) 116 : 357-371, 1952. 164. GOLDSPINK, Cr. Biochemical and physiological changes associated with the postnatal development of biceps bra&ii. Camp. Biochem. l.hysioZ. 7 : 157-168,

182.

183. 184.

1’362.

165. GOLDSPINK, G. Cytological basis of decrease in muscle strength during starvation. Am. J. Physiol. 209: 100-104, 1965. 166. GOLL>Sl’lNK, G. An attempt at estimating cxrrafiber fluid in small skeletal muscles by a simple physic;d ~~~erl~ucl. Curl. J. 1 /ly~iul. 1.4 : 765. 775, 1966. ib7. GOLDSPINK, G. SdhXJlkICI~C hgth during postnatal growth of mammalian muscle fibres. J. Cell Sci. 3 : 539-548, 1968 168. GOLDSPINK, G. Succinic dehydrogenase content of individual muscle fibres at different ages and stages of growth. Life Sci. (Oxford) 8 : 791-808, 1969. 169. GOLDSPINK, G. The proliferation of myofibrils during muscle fibre growth. J. Cell Sci. 6: 593-604, 1970. 170. GOLDSPINK, G., R. E. LARSON, AND R. E. DAVIES. The immediate energy supply and the cost of maintenance of isometric tension for different muscles in the hamster. Z. Vergleich. Physiol. 66: 389397, 1970. 171. GOLDSPINK, G., AND R. W. D. ROWE. The growth and development of muscle fibres in normal and dystrophic mice. In: Research in A4uscuZur Dystrophy, I’roc. 4th Symp. London : Pitman, 1968, p. 116-13 1. 172. GOLDSPINK, G., AND R. W. D. ROWE. Studies on postembryonic growth and development of skeletal muscle. II. Some physiological and structural changes that are associated with the growth and development of skeletal muscle fibres. Proc. Roy. Irish Acad. Sect. B 66: 85-99, 1968. 173. GOLISCH, G., D. PETTE, AND IT. PICIILMAIER. Metabolic differentiation of rabbit skeletal muscle as induced by specific innervation. European J. Biothem. 16 : 110-l 16, 1970. 174. GijPFERT, H., AND H. SCHAEFER. Die mechanische Latenz des Warmblutermuskels, nebst Beoiiber die Muskelzuckung und den bachtungen Aktionsstrom. Arch. Ges. Physiol. 245 : 60-71, 1941. 175. GORDON, A. M., A, F. HUXLEY, AND F. J. JULIAN. The variation in isometric tension with sarcomere length in vertebrate muscle fibres. J. Physiol. (London) 184 : 170-l 92, 1966. 176. GORDON, G., AND C. G. PHILLIPS. Slow and rapid components in a flexor muscle. Quart. J. Exptl. Physiol. 38 : 35-45, 1953. 177. GRgT%NER, P. Zur Anatomie und Physiologie der quergestreiften Muskeln. Hec. ZooZ. Suisse 1 : 665-684, 1884. 178. GUTH, I,. ‘Trophic’ influences of nerve on muscIe. Physiol. Rev. 48: 645-687, 1968. 179. GUTII, L., AND F. J. SAMAHA. Qualitative differences between actomyosin ATPase of slow and fast mammalian muscle. ExptZ. Neural. 25 : 138-152, 1969. 180. GUTH, L., F. J. SAMAHA, AND R. W. ALBERS. The neural regulation of some phenotypic differences between the fiber types of mammalian skeletal muscle. Exptl. Neural. 26 : 126-135, 1970. 181. GUTH, L., AND P. K. WATSON. The influence of

185.

186.

Volume

52

innervation on the soluble proteins of slow and fast muscles of the rat. ExptZ. Neural. 17 : 107-l 17, 1967. GUTH, L., 1’. K. WATSON, AND W. C. BROWN. Effects of cross-reinnervation on some chemical properties of red and white muscles of rat and cat. Ex/)tZ. NeuroZ. 20: 52-69, 1968. GUTMANN, E. ‘I-ire Ilenercatrd 12luscZe. Prague : Czech. Acad. Sci., 1962. GUTMANN, 1:. Differentiation of metabolism in muscles of different function during ontogcnesis. Pro,qr. Drain Rcs. 22: 566-574, 1967. GUTMANN, 1:. Development and maintenance of neurotrophic relations between nerve and muscle. In: Ciba Found. Symj). Growth of Nervous System, edited by G. E. !4’. Nrolstcritrolrrrc Mid 21. U’ClonnOT. London : ChlIrc.hill, 1968, p. 233-243. GTJTMANN, I?. The trnphic ftlnction of the nrrve

cell. Sfirntin (MiZon) 104. l-20, 1Wl 187. GUTMANN, E. Open questions in the study of the ‘ trophic’ function of the ncrvc ccJ1. In : 7’0). Probl. Psychiat. Neural. Current Research in Neurosciences, edited by Il. T . Wycis. Base1 : Karger, 1970, vol. 10, p. 54-61. 188. GUTMANN, I’., I. HAJEK, AND I’. HORSKg. Effect of cxcessivc use on contraction and metabolic properties of cross-striated muscle. J. I /rysioZ. (London) 203 : 46--47P, 1969. 189. GUTMANN, E., I. HAJEK, I’. HORSE;+, I.

190.

191.

192.

193. 194.

195.

196.

197.

198.

199.

200.

SYROV%!, V. ROHLfeEK, AND M. $Ti!?PITAKLAUCO. Effect of repeated long lasting direct muscle stimulation on contraction and enzyme properties of muscles in organ c\tltures. I’hysiologia Bohemoslov. 17 : 461, 1968. GUTMANN, E., AND V. HANZLfKOVA. Effects of accessory nerve supply to muscIe achieved by implantation into muscle during regeneration of its nerve. Physiologia Hohemoslov. 16 : 244-250, 1967. HAINAUT, K., AND J. E. DESMEDT. Staircase potentiation in normal human muscle : instantaneous relation between force and speed of the twitch. Arch. Intern. I’hysiol. 76 : 554-556, 1968. HALL-CRAGGS, E. C. 13. The contraction times and enzyme activity of two rabbit laryngeal muscles. J. Anat. 102 : 241-255, 1968. HANSON, .J. Keccnt X-ray diffraction studies of muscle. Chnrt. ZI’CC. lliophys. 1 : 177-216, 1968. IIARIGAYA, S., Y. OGAWA, AND H. SUGITA. Calcium binding activity of microsomal fraction of rabbit ret1 m~lsclc. J. Biochem. (Tokyo) 63 : 324-331, 1968. HARRIS, .J. B., AND A. R. LUFF. Muscle fibre membrane potentials in fast and slow muscles of developing mice. J. Physiol. (London) 200 : 124-l 25P, 1969. HARTREE, W.. AND A. V. HILL. The nature of the isometric twitch. J. I’hysiol. (London) 55 : 389-411, 1921. HARTSHORNE, D. ,J., AND S. V. PERRY. :1 chromatographic and electrophoretic study of sarcoplasm from adllltand foetal-rabbit muscles. niothem. J. 85: 171-177, 1962. HAZELWOOD, C. I?., AND B. I,. NICllOLS. In vitro investigation of resting muscle membrane potential in prewcanling and weanling rat. Nature 213: 935-936, 1967. HAZELWOOD, C. F., AND B. I,. NICHOLS. The in vivo mllsclc resting potential in the developing rat. Johns Ilopkins Med. J. 123 : 198-203, 1968. MET~ANDI~R, I