The rate of tension development in isometric tetanic contractions of mammalian fast and slow skeletal muscle A. J. Buller and D. M. Lewis J. Physiol. 1965;176;337-354

The Journal of Physiology Online is the official journal of The Physiological Society. It has been published continuously since 1878. To subscribe to The Journal of Physiology Online go to: http://jp.physoc.org/subscriptions/. The Journal of Physiology Online articles are free 12 months after publication. No part of this article may be reproduced without the permission of Blackwell Publishing: [email protected]

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

This information is current as of March 5, 2008

This is the final published version of this article; it is available at: http://jp.physoc.org This version of the article may not be posted on a public website for 12 months after publication unless article is open access.

The Journal of Physiology Online is the official journal of The Physiological Society. It has been published continuously since 1878. To subscribe to The Journal of Physiology Online go to: http://jp.physoc.org/subscriptions/. The Journal of Physiology Online articles are free 12 months after publication. No part of this article may be reproduced without the permission of Blackwell Publishing: [email protected]

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

J. Phy8iol. (1965), 176, pp. 337-354 With 8 text-ffgure8 Printed in Great Britain

337

THE RATE OF TENSION DEVELOPMENT IN ISOMETRIC TETANIC CONTRACTIONS OF MAMMALIAN FAST AND SLOW SKELETAL MUSCLE

By A. J. BULLER AND D. M. LEWIS From the Physiology Department, King's College, London

(Received 8 June 1964) Following the demonstration by Buller, Eccles & Eccles (1960b) that the speed of contraction of mammalian skeletal muscles was at least partially determined by the motor nerve innervation, two central problems remained. First, what part or parts of the contractile machinery of the muscle are influenced by the motor innervation, and, secondly, how do the motoneurones bring about their influence upon the muscle fibres? The solution of the first of these two questions requires a more detailed study of the contractile mechanism of mammalian muscle than has hitherto been made, with particular attention to any differences which exist between fast and slow skeletal muscle. In the mammal, both the fast and slow skeletal muscles consist of twitch fibres, and both types of muscle are therefore comparable with the fast fibre system of the frog. Only very recently has there been a demonstration of the equivalent of the frog slow fibre system in mammalian muscles, and as yet such fibres have only been identified in extrinsic ocular muscles (Hess & Pilar, 1963). The present paper is concerned with an investigation into the rate of isometric tension development in mammalian fast and slow muscles following repetitive stimulation of their motor nerves. This study was a necessary preliminary to the understanding of the alterations which occur in the rate of tension development following operative cross union of the nerves to mammalian fast and slow muscle (Buller & Lewis, 1964, 1965b). A preliminary account of some of the experiments herein reported has already been published (Buller & Lewis, 1963 a). METHODS on cats The experiments were performed weighing between 1-8 and 2-8 kg anaesthetized with pentobarbitone sodium (Nembutal). An initial dose of 40 mg/kg was injected intraperitoneally, and anaesthesia was maintained by subsequent intravenous injections through a jugular cannula. After making an incision down the mid line of the calf, the muscles to be studied were dissected free from surrounding structures whilst still preserving their full blood supply. While other hind-limb muscles have been examined this paper will confine itself to a

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

338

A. J. BULLER AND D. M. LEWIS

comparison of the soleus and flexor hallucis longus muscle. This latter muscle is more correctly called the mesial head of the flexor digitorum loingus (Chin, Cope & Pang, 1962), since in the cat its tendon of insertion fuses with that of the larger flexor digitorum longus. In this paper, however, we shall retain the commoner terminology of flexor hallucis longus (F.H.L.). Soleus and F.H.L. muscles were chosen because they are the slowest and fastest contracting calf muscles respectively and because, in any one cat, their maximum tetanic tensions rarely differ by more than 30 O0. The individual motor nerves to the muscles to be used were carefully dissected to allow an ample length for stimulation. They were then cut centrally and the distal ends prepared for mounting on bipolar stimulating electrodes. The individual muscle tendons were very firmly tied to steel hooks which could be attached directly to the strain gauge. The length of tendon betweein the muscle and hook was intentionally kept short. This is important, because the 'available' length of tendon for a muscle such as F.H.L. is much greater than for a muscle like soleus, and the incorporation of a long length (- 2 cm) of tendon in the recording system introduces an appreciable distortion. The total compliance of the recording system (short length of tendon, steel hook and strain gauge) was typically 0 5 mm/kg, and the introduction of 2 cm of tendon changed this figure to 3 mm/kg. These figures indicate only that part of the total compliance over which the experimenter has some control. It takes no account of the intramuscular tendon or the series elastic elements within the muscle fibres. In order to stabilize the leg, steel twist drills were inserted into both ends of the tibia and then mounted in chucks which were magnetically anchored to a massive metal table upon which all the associated equipment was also located. The skin flaps formed by the primary incision were sewn back to metal supports, thus creating a pool of some 50-200 ml. capacity which was filled with warmed liquid paraffin. Once filled, the temperature of this paraffin pool was maintained between 36-5 and 37.50 C by means of heaters and even temperature distribution ensured by stirring. Care was taken to see that the muscles were always fully immersed in the paraffin. On those occasions when a tendon was out of the pool during recording it was covered with a wisp of cotton-wool soaked in paraffin. In addition the cat's body temperature was continually monitored and the reading used to regulate the current supplied to an electric blanket on which the cat rested. The device used was similar to that described by Krnjevic' & Mitchell (1961). In order to allow access for the strain gauge the posterior part of the calcaneum was excised and the cat's foot fixed in maximum dorsiflexion. The master stimulator (Digitimer, Devices Ltd.) allowed the programming of up to five pulses separated by individually variable intervals during each oscillograph sweep. Each pulse could be used to initiate a single stimulus, or any two pulses could be used to gate a tetanic train of pulses, the train starting synchronously with the first of the two pulses and stopping at the second. The frequency of the tetanic train was controlled by a separate unit, the fixed frequencies available ranging from 1 to 1000 pulses/sec. The calibration accuracy was determined by a 10 kc quartz crystal. All pulses whether single or repetitive were fed to the nerve through a transistorized stimulus isolation unit. For each input pulse this unit provided an earth-free output pulse variable in duration and intensity. The output impedance of the stimulus isolation unit was 500 ohms. The stimuli were applied to the nerve through platinum or silver-wire electrodes. Recording. (a) Mlechanical. Isometric tension was measured by means of either Statham or Langham Thompson unbonded wire strain gauge bridges. The particular transducer used dependedonthesizeof the cat and was either G 1 64 orG 1 80 (Statham) or UF 2 (Langham Thompson). These gauges had tension maxima of 1-9, 2-4 and 4-8 kg and measured unloaded natural frequencies of approximately 720, 840 and 1100 c/s respectively. All gauges were d.c. excited (15-20 V) and had measured non-linearities of less than + 1 % full scale. Calibrations were obtained by shunting one of the arms of the transducer bridge with precision resistors (0-1 % Alma). The accuracy of such calibrations was periodically checked by loading the various strain gauges with weights. The voltage output from the strain gauge

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

ISOMETRIC TETANI OF MAMMALIAN MUSCLES

339

was amplified initially by a d.c. differential amplifier, subsequently by a d.c. single-sided amplifier, the output of which was then displayed on one beam of a double beam cathode ray oscilloscope (Tektronix 502). The frequency response of the amplifier system was flat within 3 db to 1 kc/s. The total noise of the system was typically such that an easily observed deflexion of the base line could be produced by a force of 0-1 % of the strain gauge maximum. Much more commonly, however, the gain of the amplifier was reduced so that larger tensions were recorded in the apparent absence of noise. By means of associated circuitry described in more detail elsewhere (Buller & Lewis, 1965a), the initial tension on the muscle and the peak active tension developed by the muscle during each contraction could be read directly from meters. In addition, provision was made for the electrical quasi-differentiation of the tension record. This was performed as illustrated diagrammatically in Fig. 1 using a

R

12C 10

I I1 ~ ~~ 6 -

~

~

~

~

~

~

4-

2

C

64 32 48 10 msec Volts Fig. 1. Left top. Block diagram of differentiator. K2W = Philbrick operational amplifier type K 2 W; C = differentiating condenser; R = feed-back resistor; V = in line readout voltmeter (Solartron L. M. 901). Left middle. Ramp input to differentiator and square wave output from differentiator. The dots have the same significance as in Fig. 1 A. Left bottom. Calibration of differentiator, plotting the slope of the ramp input measured in g/msec against the peak voltage of the output square wave measured in volts by voltmeter V. Right (A) upper beam. Typical fast isometric twitch from F.H.L. Lower beam, digital representation of some of the contraction characteristics (see Buller & Lewis, 1965 a). The left-hand group of three dots indicates the initial tension on the muscle, each dot representing 5 g. The next group of twenty-three dots (raised from the base line) each represent 1 msec and measure the time from the start of contraction until the time of peak tension development. The next group of twentythree dots each represent 1 msec and measure the time from peak tension until the instant of half decay. The final group of twenty-four dots measure the peak tension developed during the twitch, each dot representing 10 g. (B) Upper beam same twitch, lower beam differentiated tension record. (C) Two beams superimposed. Upper showing original twitch, lower electrically integrated record of the differentiated record of (B). Time scale 10, 50 and 100 msec. 22 Physiol. 176

16

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

340

A. J. BULLER AND D. M. LEWIS

Philbrick K2W operational amplifier connected as a differentiator. Switched input capacitors provided differentiating time constants of between 100 tusec and 1 msec, while high-frequency noise was reduced by adding series resistance to C (Fig. 1) and shunting R with a small capacitance. The advantages of using an operational amplifier as a differentiator rather than a simple input RC network followed by a voltage amplifier proved considerable (cf. Korn & Korn, 1956). The differentiator was followed by a peak-reading voltmeter and the peak voltage developed by the differentiator was displayed on an in-line readout voltmeter (Solartron L. M. 901). The accuracy of the differentiator was tested by electrically integrating the differentiated tension record (see Fig. 1). The linearity of the peak-reading device was measured by feeding voltage ramps equivalent to known rates of change of input voltage per second into the differentiator and measuring the square wave voltage developed by the differentiator (see Fig. 1). By suitable scaling the reading of the in-line readout voltmeter could be equated to changes of tension (measured in g/msec) at the input of any one of the strain gauges used. If required the output of the differentiator could be displayed on one beam of the oscilloscope. (b) Electrical. Action potentials from muscle were recorded either by means of belly tendon leads of silver wire or by means of bipolar concentric needle electrodes. In either case the signals were fed to a conventional a.c. coupled preamplifier (Tektronix 122) with 3 db points at 80 c/s and 10 kc/s. The output from the preamplifier was a.c. coupled (O I,uF and 1 MQ) into a Tektronix 502 oscilloscope. Time traces were derived from count-down circuits driven from a 10 kc/s crystal oscillator. Permanent records were obtained by photographing the face of the cathode ray tube on film (Ilford type 5B62) using a camera system with an optical reduction of approximately 1: 3. The dissection completed, the cat was set up and the pool formed. The animal was then left for half to one hour for the muscles to come into temperature equilibrium with the paraffin in the pool, since despite efforts to the contrary the muscles often cooled during the dissection. After this the motor nerve of the muscle to be studied was stimulated approximately once every 6 or 9 sec with shocks of 30 or 100 gsec duration and 2-5 V intensity (approximately 3 times that necessary to produce a maximal contraction). The lower repetition frequency was used particularly for fast muscles such as flexor hallucis longus. This reduced to a very low level the potentiating effects of previous stimuli on the mechanical responses. Considerable care was then taken to align correctly the strain gauge in the natural line of pull of the muscle, and to apply that amount of initial tension to the muscle which produced the maximal twitch response. The initial tension was adjusted by means of a micrometer drive coupled to the strain gauge. The extreme importance of care at this stage has been stressed by Buller, Eccles & Eccles (1960a) and in more detailed studies by Buller & Lewis (1963 b). Next, evidence was sought for any effect on the muscle twitch of a back response in the motor nerve fibres (Brown & Matthews, 1960; Buller & Lewis, 1963b). This was done by applying two maximal stimuli to the motor nerve separated by intervals ranging between 0 5 and 1*5 msec and noting whether any decrease in the size of the mechanical response occurred. If such an effect was observed (it appeared more commonly in large cats and in soleus) the muscle was excluded from the present study. Measurements were then commenced of either the effects of two stimuli at various intervals or the effects of short tetani (100-400 msec) at various frequencies. After a tetanus the motor nerve was stimulated by single shocks once every 9 sec until any alteration which had taken place in the twitch size (post-tetanic potentiation, Brown & von Euler, 1938) and/or rate of tension development had disappeared. This rarely took longer than 60 sec and was easily assessed since both these contraction characteristics could be directly read out on meters after each twitch. Usually the frequency of the tetanic stimuli was increased in steps up to a maximuim of 400-600pulses/sec and in afew cases to 1000 pulses/sec and then reduced through intermediate frequencies to ensure that no changes had taken place in the muscle as a result of the high-

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

ISOMETRIC TETANI OF MAMMALIAN MUSCLES

341

frequency activation. A similar procedure was adopted with the two-stimuli experiments. Occasionally the frequency changes or intervals were randomized. No differences were noted in the results obtained. At the conclusion of the experiment the muscles used were dissected out of the body, blotted 'dry' with filter paper and weighed. The over-all length of the muscle (but excluding any tendon) was also measured. Source8 of error. The finite time constant of the differentiating network and the peak voltage condenser (Fig. 1 block diagram) introduce errors in the reading of the maximum rate of change of tension. However, calculation shows that the maxrimum error produced with the highest rates of changeencountered was of theorderof -5 %. Amore important sourceof error was the occasional deterioration of the muscles following repeated tetani. The changes observed were effectively confined to the fast muscles studied and were evidenced as a reduced peak tension and decreased maximuim rate of tension rise during the twitch while the peak tetanic tension and the rate of risewerelittleaffected. Wehavenotinvestigatedthisphenomenon in detail, but have rejected results in which the peak twitch tension fell by 15 % or more during the course of the whole experiment. Apart from these occasional exceptions the muscles remained in excellent condition, and consistent responses could be obtained over many hours.

F.H.L

Soleus

83

12-5

32

50

25

50

80

125

10 msec

100 msec

Fig. 2. Genesis of tetanus for slow and fast mammalian muscles. On the left four tetani of soleus at the stimulus frequencies shown below each record. On the right four tetani of F.H.L. at the frequencies shown. Note the different sweep speeds for the two sets of records. The dots on the lower beam of each pair of records indicate the peak tension developed during the contraction shown above. Each dot is equivalent to 20 g. RESULTS

Figure 2 illustrates the classical genesis of tetanus experiment for both a slowly contracting skeletal muscle (soleus) and a fast contracting skeletal muscle (flexor hallucis longus). As has been pointed out many times previously (cf. Cooper & Eccles, 1930), the stimulation rate necessary to bring about apparent fusion of the mechanical responses is higher for the fast muscle (125 pulses/sec) than for the slow muscle (50 pulses/sec). The term apparent fusion frequency is used because, although no obvious oscillations of the tension record may be seen at the amplification used, Ritchie (1954) has rightly stressed that the apparent fusion frequency is 22-2 Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

342

A. J. BULLER AND D. M. LEWIS dependent on the sensitivity of the recording system. If the apparently fused tetani of Fig. 2 were examined at higher amplification, oscillations at the stimulus frequency would certainly be visible. However, from the maximum frequencies illustrated (50 pulses/sec for soleus and 125 pulses/ sec for F.H.L.) any further increase in frequency produces negligible change in the peak tension developed by the muscle. Definition must here be made of peak tension since, while the fast muscles usually show a plateau of tension, the slow muscles often show a slow rise of tension for several hundred milliseconds after the initial faster rise. Peak tension will therefore be defined for the purposes of this paper as the maximum tension reached 400 msec after the first stimulus of a tetanic train, since by this time any residual climb in tension is small. While it is true that no increase occurs in the peak tension developed if the stimulation frequency exceeds 50 pulses/sec for soleus and 125 pulses/ sec for F.H.L. it is apparent from Fig. 3 that the rate at which the peak tension is developed is considerably enhanced by further increases in the stimulation frequency. Figure 3 shows superimposed isometric tension records of two tetani of soleus at stimulation rates of 50 pulses/sec and 310 pulses/sec and two tetani of F.H.L. at stimulation rates of 125 pulses/ sec and 500 pulses/sec. In order to study this change in the rate of tension development more closely differentiated records of the tension rise were employed (see Methods) and an illustration of such records obtained during tetanic stimulation are shown in the lower beam recordings of Fig. 4. The differentiated tension records serve in fact to increase the effective amplification of the recording system and clearly demonstrate changes in the rate of tension build up with time. In the lower oscillographic records of both the soleus tetanus at 50 pulses/sec and the flexor hallucis longus tetanus at 125 pulses/ sec obvious oscillations are apparent while with the tetani at 310 pulses/ sec and 500 pulses/sec the differentiated records are much smoother. Close examination of Fig. 4 will also show that in both the slow and fast muscle the maximum rate of change of tension (the peak of the differential record) is greater and occurs earlier with the higher frequency of stimulation. This latter point is illustrated in Fig. 5A, where the maximum rate of change of tension measured in g/msec is plotted against stimulus frequency. The filled circles show the maximum rate of rise of tension during a single twitch, the crosses show points observed with increasing stimulus frequency and the open circles points obtained with decreasing stimulus frequency. For any particular muscle the shape of the curve is extremely reproducible from animal to animal, though the absolute value in g/msec varies. In order to compare the maximum rate of tension development during

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

ISOMETRIC TETANI OF MAMMALIAN MUSCLES 343 isometric tetani in different animals use was made of a derived term-the percentage of the maximum tetanic tension (defined above) developed per msec (% P0/msec). It may be seen from Fig. 5B that slow and fast muscles F.H.L.

Soleus

500g . .. .. .. .. .. I kg . 1-5 kg 10 msec 10 msec Fig. 3. Left. Superimposed tetani of soleus muscle at stimulation frequencies of 50 pulses/sec (lower record) and 310 pulses/sec (upper record). Right. Similar records (at different sweep speed) for F.H.L. at stimulation frequencies of 125 pulses/sec (lower record) and 500 pulses/sec (upper record). Tension calibrations of 500 g, 1 and 1-5 kg. ....... ......... ......... .......

F.H.L.

-Soleus

125

50 _

500 g 1 kg 1-5 kg

310

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

....

....

....

....

500

....

....

.........

..

.

.

.

.

.

.

.

.

.

.

..

10 msec 10 msec Fig. 4. Paired recordings of isometric tension development and differentiated tension records for soleus (left) and F.H.L. muscle (right). The stimulation frequencies used are shown under each pair of records. Note the different sweep speeds for soleus and F.H.L. Different time constants of differentiation were used for the two muscles.

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

344 A. J. BULLER AND D. M. LEWIS typified by soleus and F.H.L. have clearly distinct maximum rates of rise, and that by plotting the time to peak tension of the isometric twitch response against the maximum rate of tension development in isometric tetani two distinct populations of points occur, filled circles representing F.H.L. muscles and crosses soleus muscles. 25

A Soleus

15

-3

~~~~0x

I

Ox

-

E

30 10

10~)

20

x Ox 2

t

50 100 200

-1 -2

5 -e

20

Stimulation frequency (pulses/see) Time to peak (msec) Fig. 5A. Plot of the maxrimum rate of tension development (ordinate) during tetani of the frequency shown (abscissa) for soleus anld F.H.L. The two filled circles indicate the maxrimum rate of tension development during a twitch. The crosses indicate the maximum rates measured during a series of tetani of increasing frequency, the open circles the maximulm rates observed with a series of tetani of decreasing frequency. The durations of the tetani were 200 msec for soleus and 100 msec for F.H.L. The two short arrows indicate the frequencies commonly quoted in the literature as the 'fusion frequencies' for slow and fast muscles, respectively. The longer arrows indicate the absolute refractory periods between the first and second stimuli in the two muscles used in this exrperiment. The ordinate scales to the right of the two graphs indicate the maxrimum rate of tension development in the two muscles at various frequencies of stimulation with the maximum rate observed in the twitch scaled as one. Fig. 5B. A plot of the time from the start of contraction to the development of peak tension in an isometric twitch (abscissa) against the maximum rate of tension development during an isometric tetanus (ordinate). Crosses indicate soleus muscles, filled circles F.H.L. muscles.

We were initially surprised at the high frequency of stimulation necessary to produce the maxsimulm rate of rise of tension in both soleus and flexor hallucis longus. While it may easily be shown that the large diameter motor axonLs supplying both soleus and F.H.L. muscle can conduct trains of impulses at such repetition rates, it was decided to make a direct attempt to record the muscle electrical activity in order to confirm that the muscle action potentials were following the stimulation frequency. Bipolar

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

ISOMETRIC TETANI OF MAMMALIAN MUSCLES 345 concentric needle electrodes were used, but because of the relatively long biphasic muscle action potentials it is often difficult to interpret the records with the highest stimulation frequencies. Movement of the muscle during the contraction also served to alter the recording conditions at the electrode tip, and to make difficult the interpretation of changes in amplitude of the action potentials. Figure 6 illustrates typical records obtained from soleus, the rate of change of tension being recorded on the upper beam and

1

50 1 msec

2 msec

2 mV

5 msec

3 msec

250 ..

...

...

.....

1 msec 10 msec Fig. 6A. Differentiated tension records (upper beam) and muscle action potentials (lower beam) from soleus muscle at stimulation frequencies of 50 and 250 pulses/ sec. (Action potentials retouched.) Fig. 6B. Differentiated tension records (upper beam) and muscle action potentials resulting from two maximal stimuli applied to the motor nerves at the intervals indicated. Soleus muscle.

the muscle action potentials on the lower beam. The responses to repetitive stimulation at 50 pulses/sec and 250 pulses/sec (the optimum frequency for this particular muscle) are shown on the left, and to two stimuli at varying intervals on the right. The absolute refractory period of the system, motor nerve-terminal nerve fibres-end-plate-muscle fibres as judged by muscle action potential records, and muscle tension records, was 1-61 msec for soleus (S.D. = 0.14) and 1 03 msec for F.H.L. (S.D. = 0-17). No attempt was made to elucidate the refractory periods of the individual components of the system.

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

A. J. BULLER AND D. M. LEWIS 346 The time at which the maximum rate of tension development occurred was also different in soleus and F.H.L. During a twitch the maximum was reached in 15-19 msec following the start of the contraction for soleus and 6-9 msec for F.H.L. During tetanic stimulation the time between the start of the tension rise and the peak rate of tension rise became progressively less with increasing frequency of stimulation, until with optimal frequencies of stimulation the time to peak rate of rise was within 1-2 msec of the time taken during a twitch. 24_ 20

0

-

*

xK

x

x

16

x

x dC$X?j C)~~~~cca 0 bo

.+.

+ +

+ + + 0

~~~~0

12

x

x

x

x

0

0 0

0

0

8

00

0 0

0000~0

0

4

, t4

+

+

44

+,

.

.

200 100 50 10 20 Time in millisecond Fig. 7. Soleus muscle. The maximum rate of tension development during stimulation with two (0), three (x), four (0) and five (+) stimuli respectively. For explanation see text. The dotted line shows the maximujm rate of tension development during a tetanus at a frequency of 310 pulses/sec. The first stimulus occurred at time 0. The larger arrows show the times of the fixed second, third and fourth stimuli when used. The smaller arrows indicate the time of arrival of the second, third, fourth, fifth and sixth stimuli of a tetanic train at a frequency of 310 pulses/

1

2

5

Sec.

When it became apparent that the time to the peak reading of the differentiated tension record at the optimum stimulation rate (typically 310 pulses/sec for soleus and 600 pulses/sec for F.H.L.) was similar to the time to peak reading of the differential tension record of a twitch an attempt was made to find the minimum number of stimuli, and their temporal relations, which would produce the maximum rate of tension rise observed during tetani. Figure 7 illustrates one such experiment for soleus. The rate of rise of tension in g/msec is plotted against time. The dotted line

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

ISOMETRIC TETANI OF MAMMALIAN MUSCLES 347 at the top of the graph indicates the maximum rate of rise of tension observed during a tetanus (310 pulses/sec). At t = 0 a maximal stimulus was applied to the motor nerve and this was followed at various intervals by a second stimulus. The maximum rates of tension rise observed as a result of these two stimuli separated by the intervals shown are plotted as

o xXX X d~~~~~

604 04 0

40_ 4

-

X

0

x

XXX X

7

X/S

>e/S

5 X~~

.0

3 -40

XX

x

10~os

20

X(

* -H.L

xSolu Ev~~~~~

2

Soleus ~~~~~~~~~x 1 2 3 5 7 10 Stimulus interval (msec)

E-i

Fig. 8. Top row. Soleus muscle. Superimposed differentiated tension records resulting from one (solid trace) and two stimuli (modulated trace) at the intervals shown by the stimulus markers at the beginning of the sweep. Time marker: each dot represents 0*1 msec, the columns of ten dots therefore being 1 msec apart. Middle section. The maximum rate of tension developed by either the soleus ( x) or F.H.L. muscle (0) following two stimuli separated by the interval shown on the abscissa. For the ordinate an arbitrary scale has been used and the rate of tension development by F.H.L. halved so that the results from both soleus and F.H.L. muscles could be plotted close together. Bottom section. The time of first separation of the differentiated tension records following one and two stimuli measured from the beginning of the mechanical response (ordinate) plotted against the time interval between the paired stimuli (abscissa). This graph is a plot of the information obtained from the type of records illustrated at the top of this figure. The interrupted line indicates a slope of one.

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

348 A. J. BULLER AND D. M. LEWIS open circles. It is seen that following the absolute refractory period of the total system (nerve-end-plate-muscle) there is a rapid increase in the maximum rate of tension rise as the interval between the two stimuli is lengthened. This is followed by a plateau of some 3 msec and then a slow decline. Having plotted the curve of open circles two stimuli were applied to the nerve with a fixed interval of 2*5 msec. A third stimulus was then added at varying intervals after the second of the two fixed stimuli. The maximum rates of rise of tension in the contraction caused by the three stimuli are plotted in Fig. 7 as crosses, the time scale indicating the time of third stimulus (the first two occurring at times 0 and 2-5 msec). A similar but smaller rise in the rate of tension development is apparent. This is followed by a comparable plateau and decline. The filled circles indicate the responses to three fixed stimuli at times 0, 2-5 and 5-5 msec and a fourth stimulus at the interval shown on the abscissa. The + symbols indicate the responses to four fixed stimuli at times 0, 2-5, 5-5 and 10 msec and to a fifth stimulus at the interval shown. It may be seen that in this example five stimuli optimally timed were adequate to reach the maximum rate of tension development observed in a tetanus. Increasing the number of stimuli used above five did not increase the maximum rate of rise of tension. In other experiments the maximum rate of tension development was reached with three to six stimuli, and similar results were obtained with F.H.L. Since it was apparent from experiments of the type illustrated in Fig. 7 that in soleus two stimuli separated by only 2*5 msec were having a marked influence on the rate of rise of tension during the mechanical response of the muscle it seemed pertinent to examine in more detail the time of separation of the mechanical responses resulting from one and two stimuli. As has been shown by Macpherson & Wilkie (1954), and stressed earlier in this paper, differentiation of the tension record increases the resolution by effectively increasing the amplification of the recording system. The upper part of Fig. 8 illustrates superimposed traces of the differentiated isometric tension records of a soleus muscle following one and two stimuli applied at the intervals shown by the added markers on the upper beam. The oscilloscope sweep containing the response to two stimuli has been intensity modulated (and therefore appears dotted) in order to

distinguish it. The lower part of Fig. 8 illustrates the results obtained from one such experiment on each of soleus (crosses) and F.H.L. (filled circles) muscles. The interval between the two stimuli (abscissa) is plotted against the time after the start of contraction at which the two differentiated mechanical records separate. The two graphs have a similar shape, but because of the longer absolute refractory period of soleus muscle the left limb of its graph

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

349 ISOMETRIC TETANI OF MAMMALIAN MUSCLES is displaced approximately 1 msec to the right compared with that of F.H.L. The minimum time of separation is also increased by approximately 0 5 msec. For both soleus and F.H.L. muscles, the slope of the relation between stimulus interval and the time of separation of the mechanical responses approximated to the expected value of one provided the stimulus interval exceeded 4 msec. Finally, it is pertinent to state that we have attempted to reproduce all the results described above on curarized muscle. However, in vivo we have so far found it impossible to stimulate directly such large muscles as soleus and F.H.L. and obtain on increasing the stimulus strength an unequivocal response equivalent to that obtained with a maximal motor nerve volley. It is certainly possible to stimulate directly through electrodes between the origin and tendon of the muscle and arbitrarily adjust the stimulus strength so that the tension developed by the muscle is similar to the tension obtained after a maximal stimulus to the nerve before curarization. However, in our experience such stimuli, even if of short duration (< 1 msec), cause some muscle fibres to discharge repetitively. Our direct stimulation experiments have therefore been confined to levels of submaximal stimulation at which repetitive firing of muscle fibres is negligible. With a single stimulus of such intensity some muscle fibres which do not respond by contracting are subliminally stimulated and if a second stimulus is applled sufficiently rapidly (< 2 msec) these fibres will be recruited to the second mechanical response. However, we have found this effect interferes very little at tetanic stimulation frequencies below about 500 pulses/sec. Under these conditions we have confirmed that the maximum rates of tension rise during isometric contractions in curarized fast and slow muscle remain different and the values obtained are similar to those obtained using indirect stimulation. DISCUSSION

The experiments described above quantify the differences which exist between the maximum rate of rise of tension during isometric tetani of mammalian fast and slow skeletal muscles. While for many years it has been apparent that the rise of tension during an isometric 'fused' tetanus of fast muscle is more rapid than the rise of tension during a tetanus of slow muscle, the marked influence of the rate of stimulation (see Fig. 5A) on both types of muscle has not been appreciated. It is apparent that if quantitative comparisons are to be made between normal muscles and muscles which have been cross-innervated (Buller et al. 1960b) careful attention must be paid to this parameter. These results also serve to

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

A. J. BULLER AND D. M. LEWIS emphasize the care which must be exercised in the use of the term 'fusion frequency'. If used to define the frequency at which the individual contractions produced by the successive stimuli of a tetanic train can no longer be identified, the term is as much a measure of the sensitivity of the recording system as of the muscle's performance (Ritchie, 1954). While it is often desirable to compare the 'apparent fusion' frequencies of two muscles this should only be done after due consideration of the recording conditions for each muscle. That the results described for the two types of mammalian skeletal muscle are due to differences of the contractile mechanisms and not to differences of the motor-nerve fibres or end-plates may be inferred from the fact that similar results were obtained by direct stimulation of curarized muscles. In addition, Professor Sandow has informed us that he has now observed a comparable effect of tetanic stimulation frequency on the rate of rise of isometric tension in frog sartorius muscles using 'all over' (massive) stimulation (cf. Mostofsky & Sandow, 1951). From our results with tetanic stimulation it is impossible to decide whether the maximum rate of tension development of which the contractile machinery is ultimately capable is ever reached. In Fig. 5A, which was obtained with indirect stimulation, there is no plateau to either curve, and the fall off in the maximum rate of tension development with the highest stimulation frequencies corresponds very precisely with what is to be expected from the total refractory periods of the two systems. In considering the limitations imposed by the refractory period due allowance must be made for the lengthening of the absolute refractory period which occurs with successive stimuli. This is apparent in Fig. 7, where the absolute refractory period of the system was 17 msec following the first stimulus, 2*5 msec following the second stimulus and 3-4 msec following the third stimulus. With direct stimulation using submaximal stimuli as described under Methods, recruitment of additional muscle fibres occurs if the separation between successive stimuli becomes very short (< 2 msec) and therefore the results obtained with the highest tetanic frequencies are again ambiguous. In either case, however, the steady increase observed in the maximum rate of tension development with increasing stimulation frequency suggests that if a plateau of active state (H-ill, 1949) occurs in the sarcomeres of these muscles following each stimulus its duration cannot exceed approximately 3 msec in soleus or 2 msec in F.H.L. The results from the two stimulus experiments are in complete accord with this view. As explained for the frog sartorius muscle at 00 C by Macpherson & Wilkie (1954), if a plateau of active state occurs following a single stimulus then, until the end of that plateau, the muscle will behave indistinguishably whether or not a second stimulus is applied. By noting 350

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

ISOMETRIC TETANI OF MAMMALIAN MUSCLES 351 the earliest time at which a second stimulus can produce a difference in the mechanical response from that seen following a single stimulus the duration of the plateau may be estimated. The essential difference between their experimental technique and ours is that we have used indirect stimulation. Initial assumptions have therefore to be made that the time interval between the application of the second stimulus to the nerve and the initiation of the muscle action potential remains the same as that following the first stimulus, and that the conduction velocity of the muscle action potential remains constant. With these assumptions it may be seen from Fig. 8 that the plateau of active state cannot last longer than approximately 3*3 msec in the sarcomeres of soleus muscle and 2-75 msec in the sarcomeres of F.H.L. In fact recent experiments (Ridge, unpublished observations) suggest that, contrary to the assumption made above, there is some delay in the passage of the second nerve action potential at very short stimulus intervals, so that the figures of 3-3 and 2-75 msec quoted above probably represent maximum estimates. In addition, Dr Miledi has informed us that he has observed curves comparable in shape to our Fig. 8 when studying the responses of both single nerve fibres and single muscle fibres to two stimuli. He attributes the shape to the interaction of the strength/latency relation and the relative refractory period of the tissue. If such an interpretation is true in our experiments it appears that we cannot define a plateau by this method, either because one does not exist following a single stimulus (cf. Walker, 1951) or because its duration is shorter than the limits set by other factors. That the plateau of active state in individual sarcomeres of mammalian muscle at 370 C might be as short as 2 msec is suggested by consideration of the Qlo for active state obtained in frog muscle by Sandow & Mauriello (1953) and the relation between duration of active state and duration of absolute refractory period in frog muscle described by Ramsey (1960). Our estimate of the duration of the plateau of active state in fast skeletal muscle is considerably shorter than that made by Norris (1960) using directly stimulated rat peroneus digiti quinti muscle at 370 C. The difference could be due to the greater sensitivity obtained by our use of differentiated tension records. The present results cast some doubt on the experimental evidence brought forward by Bowman, Goldberg & Raper (1962) in support of their contention that repetitive stimulation of mammalian fast muscle alters the force/velocity relation of the contractile element. These authors pointed out that in their experiments the rate of rise of tension in a posttetanic twitch was greater than in a preceding maximal conditioning tetanus. They argued that since during a maximal tetanus there was (supposedly) a constantly maintained plateau of active state the greater r

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

352 A. J. BULLER AND D. M. LEWIS rate of tension development observed in the post-tetanic twitch could not be attributed to a prolongation of its plateau of active state, as had earlier been suggested for frog muscle by Ritchie & Wilkie (1955). However, Bowman et al. (1962) used a stimulation frequency of only 120 pulses/sec to produce their 'maximal' tetanus. This frequency, as illustrated in Fig. 4, certainly does not produce a fused contraction of fast muscle and would not be expected to give a maintained plateau of active state. Using higher tetanic stimulation frequencies (500-600 pulses/sec) we have never observed a post-tetanic twitch to have a higher rate of tension development than the preceding conditioning tetanus. While our observations obviously still allow an effect of repetitive stimulation on the force-velocity relation the results of Bowman, Goldberg and Raper do appear open to alternative interpretation. Finally, a comparison may be made between our differentiated tension records and those of Sandow (1961). This author has used massive stimulation of the frog sartorius muscle and thus obtained simultaneous activation of all the sarcomeres. HIis records show a small initial negativity attributable to latency relaxation, and a hump on the rising phase which is attributed to the end of the plateau of active state. With the indirect stimulation which we have used the finite conduction velocity along the muscle fibres serves to blur out the latency relaxation of individual sarcomeres and we have seen no initial negativity in our differentiated tension records. The relatively slow conduction velocity along the muscle fibres also means that following a single stimulus there will be no 'muscle plateau of active state', even if a plateau of the value estimated above is occurring in the individual sarcomeres. This is because the terminal sarcomeres of a muscle fibre will only reach their plateau of active state at a time when the level of active state in the initially activated sarcomeres is decaying owing to the conduction time along the fibre exceeding the plateau time. However, in spite of the expectation that we should not see a hump on the rising phase of our differentiated records we have frequently seen one, particularly when studying fast muscles, occurring at a time comparable to the hump described by Sandow. While we have not investigated this finding in detail we are inclined to think that it may be due to minor inequalities of initial tension between different parts of the muscle. This would equate the finding to the much greater effects produced by initial tension discrepancies within the gracilis muscle described by Eccles & Iggo (1961).

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

ISOMETRIC TETANI OF MAMMALIAN MUSCLES

353

SUMMABRY

1. The responses of soleus (slow) and flexor hallucis longus (fast) muscles of the cat have been examined following both repetitive and double stimulation through the motor nerves. 2. The maximum rate of rise of tension during tetanic stimulation increases with increase of stimulation frequency to approximately 300 pulses/sec in soleus and 600 pulses/sec in F.H.L. 3. Two-stimuli experiments show that the earliest observable separation of the mechanical response following two stimuli from that following one stimulus occurs at approximately 3-5 msec after the beginning of contraction in soleus and approximately 2 msec in F.H.L. 4. These results suggest that if a plateau of active state is reached in the sarcomeres following a single stimulus its duration is shorter than 3-5 msec in soleus and 2 msec in F.H.L. We are indebted to Miss M. O'Vens for constant assistance and to Mr G. F. Stonard for expert photographic help. We would also like to thank the Medical Research Council for a grant in support of this work. REFERENCES

BOWMAN, W. C., GOLDBERG, A. A. J. & RAPER, C. (1962). A comparison between the effects of a tetanus and the effects of sympathomimetic amines on fast and slow contracting mammalian muscles. Brit. J. Pharmacol. 19, 464-484. BROWN, G. L. & VON EULER, U. S. (1938). The after effects of a tetanus on skeletal muscle. J. Physiol. 93, 39-60. BROWN, M. C. & MATTHEWS, P. B. C. (1960). The effect on a muscle twitch of the back response of its motor nerve fibres. J. Physiol. 150, 332-346. BULLER, A. J., ECCLES, J. C. & EccLEs, R. M. (1960a). Differentiation of fast and slow muscles in the cat hind limb. J. Physiol. 150, 399-416. BULLER, A. J., ECCLES, J. C. & EccLEs, R. M. (1960b). Interactions between motoneurones and muscles in respect of the characteristic speeds of their responses. J. Physiol. 150, 417-439. BULLER, A. J. & LEWIS, D. M. (1963a). The rate of rise of tension in isometric contractions of mammalian fast and slow muscles. J. Phy8iol. 169, 29P. BULLER, A. J. & LEWIS, D. M. (1963b). Factors affecting the differentiation of mammalian fast and slow muscle fibres. In Proceedings of symposium on The Effect of Useand Disuse on Neuromuscular Functions, ed. GUTMANN, E. & HNIK, P. Prague: Czechoslovak Academy of Science. BULLER, A. J. & LEWIS, D. M. (1964). The rate of rise of tensioin in isometric tetani of cross innervated mammalian skeletal muscles. J. Physiol. 170, 67P. BULLER, A. J. & LEWIS, D. M. (1965 a). Further observations on the differentiation of skeletal muscles in the kitten hind limb. J. Phy8iol. 176, 355-370. BULLER, A. J. & LEWIS, D. M. (1965 b). Further observations on mammalion cross innervated skeletal muscle. J. Physiol. (In the Press.) CHIN, N. K., COPE, M. & PANG, M. (1962). Number and distribution of spindle capsules in seven hindlimb muscles of the cat. In Symposium on Muscle Receptors, ed. BARKER, D. Hong Kong: University Press. COOPER, S. & ECCLES, J. C. (1930). The isometric responses of mammalian muscles. J. Physiol. 69, 377-385. ECCLES, R. M. & IGGO, A. (1961). The double twitch of gracilis muscle. J. Physiol. 159, 500-506.

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

354

A. J. BULLER AND D. M. LEWIS

HESS, A. & PILAR, G. (1963). Slow fibres in the extraocular muscles of the cat. J. Physiol.

169, 780-798. HILL. A. V. (1949). The abrupt transition from rest to activity in muscle. Proc. roy. Soc. B, 136, 220-228. KORN, G. A. & KORN-, T. M. (1956). Electronic Analog Computers. New York: McGraw-Hill. KRNJEVIY6, K. & MITCHELL, J. F. (1961). A simple and reliable device utilizing transistors for the maintenance of a constant body temperature. J. Physiol. 158, 6P. MACPHERSON, L. & WILKIE, D. R. (1954). The duration of the active statein a muscle twitch. J. Physiol. 124, 292-299. MOSTOFSKY, D. & SANDow, A. (1951). High current dual pulse Physiologic Stimulator. Electronics (Nov). NORRIs, F. H. (1960). Active state plateau and latency in mammalian striated muscle. Amer. J. Physiol. 200, 667-671. RAMSEY, R. W. (1960). Some aspects of the biophysics of muscle. In Structure and Function of Muscle, vol. II, ed. BOIJRNE, G. H. London: Academic Press. RITCHIE, J. M. (1954). The duration of the plateau of full activity in frog muscle. J. Physiol. 124, 605-612. RITCHIE, J. M. & WILKIE, D. R. (1955). The effect of previous stimulation on the active state of muscle. J. Physiol. 130, 488-496. SANDow, A. (1961). Energetics of muscular contraction. In Biophysics of Physiological and Pharmacological Actions. New York: American Association for the Advancement of Science. SANDow, A. & MAI,RIELLO, G. E. (1953). Fusion frequency, the kinetics of active state in muscular tetanus. Fed. Proc. 12, 123. WALKER, S. M. (1951). Tension and extensibility changes in muscle suddenly stretched during the twitch response. Amer. J. Physiol. 164, 238-247.

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008

The rate of tension development in isometric tetanic contractions of mammalian fast and slow skeletal muscle A. J. Buller and D. M. Lewis J. Physiol. 1965;176;337-354 This information is current as of March 5, 2008 Updated Information & Services

including high-resolution figures, can be found at: http://jp.physoc.org

Permissions & Licensing

Information about reproducing this article in parts (figures, tables) or in its entirety can be found online at: http://jp.physoc.org/mis c/Permissions.shtml

Reprints

Information about ordering reprints can be found online: http://jp.physoc.org/mis c/reprints.shtml

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 5, 2008