Tension development in highly stretched vertebrate muscle fibres A. M. Gordon, A. F. Huxley and F. J. Julian J. Physiol. 1966;184;143-169

This information is current as of March 25, 2008

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 25, 2008

This is the final published version of this article; it is available at: http://jp.physoc.org/cgi/content/abstract/184/1/143 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 25, 2008

J. Phy8iol. (1966), 184, pp. 143-169 With 17 text-ftgure8 Printed in Great Britain

143

TENSION DEVELOPMENT IN HIGHLY STRETCHED VERTEBRATE MUSCLE FIBRES

By A. M. GORDON,* A. F. HUXLEY AND F. J. JULIANt From the Department of Physiology, University College London

(Received 1 September 1965) SUMMARY

1. An apparatus is described by which the length of a selected part of an isolated muscle fibre can be held constant, giving isometric conditions, or alternatively its length can be measured while tension is held constant (isotonic). Control can be switched between length and tension so as to carry out afterloaded contractions with a shortening stop. 2. When a part of a fibre with uniform striation spacing is stretched so far that there is presumably no overlap of filaments, the tension developed during an isometric tetanus with this apparatus is very small (not more than 3-5 % of the tension developed at optimum length). 3. If the tendon ends are held stationary, a fibre with the same initial length develops a large amount of tension (order of 30-40 % of tension at optimum length) with a slow time course. This additional tension is due to shortening of the end parts of the fibre, where the striation spacing is smaller and overlap of filaments still exists. 4. The resistance to elongation of a part of a fibre where there is no overlap is only slightly increased on stimulation. 5. To a first approximation, the results are in good agreement with expectations based on the sliding filament theory. The development of detectable amounts of tension, and of a slight increase of stiffness, on stimulation, are however not expected on the simplest form of this theory; possible explanations are discussed. INTRODUCTION

It has long been known that the tension developed by a vertebrate skeletal muscle during an isometric tetanus decreases as the length of the muscle is altered in either direction from an optimum value (Heidenhain, * Postdoctoral Research Follow of the National Foundation. Present address: Department of Physiology and Biophysics, School of Medicine, University of Washington, Seattle, Washington, U.S.A. t Holder of Special Fellowship from the National Institute of Neurological Diseases and Blindness, U.S. Public Health Service. Present address: Department of Muscle Research, Retina Foundation, 20 Staniford Street, Boston, Mass., U.S.A.

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

144 A. M. GORDON, A. F. HUXLEY AND F. J. JULIAN 1864; Blix, 1895; Beck, 1922). The form of this relation between length and tension acquired a new interest when the sliding filament theory of muscular contraction emerged, because this theory offered a new prospect of interpreting the effects of changes in length. In particular, A. F. Huxley & Niedergerke (1954) suggested that the linear drop of tetanic tension found by Ramsey & Street (1940) with increase of length above the optimum might be explained if tension were proportional to the amount of overlap of the two sets of filaments, as would be expected if a relative force between the two filaments is generated at each of a uniform series of points in the region of overlap in each half-sarcomere. On this basis, the tension should drop to zero at the length where the two sets of filaments just cease to overlap each other. This could not be determined from Ramsey & Street's results because they did not measure the striation spacings in their fibres. A. F. Huxley & Peachey (1961) therefore specifically investigated this point. They found that isometric tetani did not give a clear result, because the fibres did not stretch uniformly. Fibres that were stretched enough so that there was no overlap in the main part of their length were found to have a considerable amount of overlap near to their ends, and during a tetanus in which the tendons were held stationary the ends were found to shorten greatly, stretching the middle part of the fibre and producing substantial amounts of tension. They therefore used isotonic tetani, with a microscope to detect whether a region in the middle of the fibre did or did not shorten. They found that there was no shortening when the striation spacing exceeded a critical value which they estimated as 3-52 ,t, agreeing closely with their estimate (3.52-3-54 ,u) from electron microscopy of the striation spacing at which overlap ceases. Analogous experiments by Podolsky (1964) on the stripped-fibre preparation of Natori (1954), stimulated by the application of solutions containing calcium, gave a value of about 3-65 ,u for the critical length, and a recent determination of filament lengths (Page & H. E. Huxley, 1963), with new precautions against shrinkage during preparation, gave 3'65 It as the spacing at which overlap ceases. The agreement between all these values is probably as good as can reasonably be expected from such measurements, and indicates that contraction does not occur if there is no overlap between the two sets of filaments. However, Carlsen, Knappeis & Buchthal (1961) re-determined the isometric tension-length curve, with measurement of striation spacings, %nd found large tensions at lengths where there is no overlap. Thus, at a spacing of 3 7 It the tension developed was 30-40 % of that developed at the optimum length, and the developed tension did not approach zero until the spacing reached 4 1-4-2 It or even more. They confirmed that the spacing was less at the ends of the fibre, but estimated that contraction

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

TENSION DEVELOPMENT IN STRETCHED MUSCLE 145 of the ends would account for only about one-sixth of the tension observed at spacings from 3-5 to 44 ,u unless the parts of the fibre with no overlap did contract, or at least underwent an increase of stiffness. The experiments described in this paper were undertaken in the hope of clearing up this discrepancy. The methods are described at some length because they are used also in other series of experiments which will be reported separately. Short accounts of the apparatus (Gordon, Huxley & Julian, 1963) and of the main results (Gordon, Huxley & Julian, 1964) have already appeared. METHODS

Principle of the method A part of the length of an isolated muscle fibre, within which the striation spacing was sufficiently uniform, was defined by two 'markers' (pieces of gold leaf) stuck to the fibre with grease. The length L of the part between these markers was measured continuously by a photoelectric device (the 'photo-electronic spot follower'). L could be held constant ('length control') by feeding from the output of this device to a moving-coil apparatus which acted as a servo motor, pulling on the tendon at one end of the fibre; stimulation in this condition produced an isometric contraction of the part of the fibre between the markers, and the tension was recorded by a transducer attached to the other tendon. Alternatively, the tension could be held constant ('tension control') by feeding from the tension signal to the servo motor; the contraction was then isotonic and the shortening of the part between the markers was signalled by the photoelectronic spot follower. A diode circuit allowed tension control to take over when tension reached a pre-set level, and length control to take over again when shortening reached a pre-set value, so as to make the part of the fibre between the markers undergo an afterloaded contraction with shortening stop. Di8section Twitch fibres, together with pieces of tendon at either end, were disected from the dorsal part of the semitendinosus muscles of frogs (Rana temporaria) stored before use in a moist environment at about 40 C. The muscle was dissected at room temperature in Ringer's solution containing tubocurarine (10-5 g/ml.), using knives made from pieces of stainless. steel razor blade. To avoid non-uniformities of striation spacing, special care was taken to remove capillaries and other adherent tissue, and to avoid stretching the fibres heavily. In most experiments the fibres were stored overnight in the usual Ringer solution at 40C before use the next day (cf. Ramsey & Street, 1940). There were no obvious differences between the results given by fibres used immediately after dissection and by fibres stored overnight. The fibre was transferred from the dissecting chamber to the experimental trough by carefully lifting it up out of the dissection solution adhering to a thin glass rod, which was then submerged in the Ringer's solution filling the experimental trough under microscopic observation. The fibre was separated from the rod by pulling on one tendon while rotating the rod in such a way as to prevent the fibre becoming twisted. 10 Physiol. 184 Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

146

~ ~ ~ n1

A. M. CORDON, A. F. HUXLEY AND F. J. JULIAN

The composition of the Ringer solution was: NaCl, 115 mm; KCI, 2-5 mm; CaCJ2, 1-8 mM; Na2HPO4, 2-15 mm; NaH2PO4, 0-85 mM.
T1).

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

154

A. M. GORDON,A. F. HUXLEY AND F. J. JULIAN

4. Connexion is through diodes 6 and 4: isotonic shortening under tension T2. Stage four can be eliminated by setting T2 at a value greater than the fibre can develop; the result is then an after-loaded contraction with shortening stop. Similarly, a simple isometric contraction is obtained if T1 is set at a high value. The diodes are of silicon (Mullard OA 202), which give a slope resistance of about 4000 Q with a current of 10 /zA; their back resistance is of the order of 108 Q. The AL and T inputs come from cathode followers with output impedance about 1000 Q; the output feeds to a cathode follower with high input impedance. The voltage L1 is provided by a floating HT battery; T1 and T2 are provided by potentiometer settings on the parallel cathode loads of the tension cathode follower. Pha8e advance. These blocks are R-C networks attenuating a steady signal to one half, and providing an adjustable amount of phase advance (time constant 0-1 msec) to reduce the lag due to motor inertia when the diode circuit switches over, and to improve the stability of the complete system. Recording C.R.O. This was the Tektronix 502 oscilloscope, with a 35 mm recording camera. Performance. Figure 3B shows the change (increase) of length, recorded by the photoelectronic spot follower, when a rectangular pulse was injected at the point marked 'Length Step' in Fig. 5. There is a slight overshoot, whose peak is reached about 3 msec after the start of the pulse. The response at the end of the pulse is less good because the fibre becomes slacker as it is shortened, so that the movement of the motor is transmitted to the markers with a delay. The overall stiffness of the servo can be obtained by multiplying together the sensitivity or gain factors associated with the elements in the loop. These are given in the preceding paragraphs and summarized in Fig. 5. Using the values in Fig. 5, it is seen that under 'length control' a steady error of 1 /4 in the distance between the markers causes a force of about 70 mg wt. at the tendon attached to the motor arm. This is to be compared with (a) static friction in the slider, etc., amounting to about 30 mg, or (b) the tension developed by the fibre, say 300 mg in a large fibre, or (c) the stiffness of the fibre. This last may be estimated on the basis that the isometric tension drops to zero with about 1 % sudden shortening; for a large fibre with 10 mm between the markers this is 300 mg for 100 ,u, or 3 mg/,. The servo is about twenty times stiffer than this in the steady state; at short times it was 2-5 times less stiff, leaving a factor of about eight over the stiffness of the fibre.

RESULTS

Procedure The procedure in experiments on highly stretched fibres was repeatedly modified so as to improve the uniformity of striation spacing in the part of the fibre between the markers. The procedure finally adopted was as follows. 1. The fibre was mounted as described under Methods, and its length adjusted by moving the 'motor stop' until the striation spacing, measured visually at one or two places near the middle of the fibre, was about 37 ,t. 2. The striation spacing was measured visually with great care at intervals of not more than 2 mm along the whole length of the fibre and 1 mm in the part of the fibre that we expected to use. The results were plotted out and positions for the markers were chosen, 6-10 mm apart, in such a way that the variations of striation spacing in the part of the fibre between the markers should be as small as possible. Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

TENSION DEVELOPMENT IN STRETCHED MUSCLE 155 3. The markers were attached to the fibre and their separation measured exactly by reading the stage vernier when each was in the centre of the field. 4. The height and width of the fibre and the striation spacing were measured visually at intervals of 2 mm along the part of the fibre between the markers. 5. The fibre was photographed by ordinary light microscopy at intervals of either 1 or 2 mm between and just outside the markers for subsequent more precise measurement of striation spacing. 6. Circulation of the cooling water was begun. 7. The photocell assembly was mounted on the microscope, the spots 'hooked' on to the markers, and the length signal calibrated, as described under Methods. 8. The servo control of fibre length was turned on. 9. The threshold for twitches was measured; it was sometimes necessary to shorten the fibre a little in order for the twitches to be detectable. 10. Recording of tetani was begun. The important features of the procedure were (a) that a uniform part of the fibre was selected, (b) that the fibre was not stimulated between making the measurements and turning on the servo, since a contraction might have caused the distance between the markers to change; (c) that the fibre was not stimulated tetanically at moderate lengths before the observations at great length, since such stimulation would be expected to increase the residual non-uniformity of striation spacing (Hill, 1953); and (d) that the fibre was stretched to just beyond the expected critical length before attaching the markers, since the grease on them might have prevented the parts of the fibre in contact with them from stretching like the rest. Tension in isometric and fixed-end tetani The lower trace of each pair of records in Fig. 7 A-D shows the tension developed during a tetanus by a fibre stretched to a length where one would expect there to be no overlap of thick and thin filaments. In Fig. 7 C-D, the servo control was switched off and the tendons were held stationary; a considerable amount of tension was developed slowly, in agreement with Ramsey & Street (1940), Huxley & Peachey (1961) and Carlsen et al. (1961). The upward deflexion of the upper traces indicates that the part of the fibre between the markers was being stretched. Fig. 7A and B, however, show the result of tetanic stimulation of the same fibres at the same initial length, under control by the regulator so that the length of the part of the fibre between the markers was held constant. The tension rise was very much smaller than when the tendons were held. The upward deflexion of the upper trace in this case means that Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

A. M. GORDON, A. F. HUXLEY AND F. J. JULIAN 156 the right-hand marker was moving towards the right. The right-hand tendon was held stationary by its attachment to the tension recorder, so the right-hand end of the fibre between the marker and the tendon must have been shortening. No doubt the left-hand end of the fibre, to the left of the left-hand marker, was also shortening but this could not be recorded. 28. viii. 63

20. ix. 63 0125

0 25

mm

Position

mm

-------

=

0-18

007=_Tension Tension 0-2 sec

kg/c m2 .......kg/cm 2 0-2 sec

A

B

47/s

Length

0i

,I 24,u 0-18 kg/cm2 ____.L..... C

.

04..... kg/cm2

_

-.1Tension

D if

-F-re^!kflfF j!ell IT, ; Stimuli V

F E Fig. 7. Tetanic stimulation of isolated fibres, with servo control of length (A and B) and with tendons held (C and D); A, C, E, experiment of 28 August 1963, 40 C, average striation spacing between the markers 3 92 ,u; B, D, F, experiment of 20 September 1963, 30 C, striation spacing 3-86 ,u. A and B (with servo control of a nearly uniform part of the fibre length): lower trace, tension; upper trace, position of right-hand marker (upward deflexion signals movement towards the right-hand tendon). C and D (with tendons held stationary), lower trace, tension; upper trace, change in distance between markers (increase upwards). In C, distance between markers is 6-1 mm, so that one grid square represents 0*39 % increase; in D, distance is 7-65 mm, and one grid square represents 0-62 % increase. E and F: time of occurrence of stimuli. Note difference of amplification between B and D. Sensitivity and speed figures refer to one grid square.

In Figs. 8 and 9 are reproduced the two sets of records of servo-controlled isometric contractions at varying degrees of stretch from which Fig. 7A and B are taken. The measurements of striation spacing in these same two fibres are shown in Figs. 10 and 1 1. At the shorter lengths in Fig. 9, and even at extreme degrees of stretch in some other experiments (e.g. Fig. 14), the tension continued to rise slowly throughout the tetanus. The evidence given in the next section makes it probable that this slow rise was due to the residual irregularity of striation spacing, so that overlap existed locally and would increase progressively as these regions shortened. On this basis, the measure of tension appropriate for relating to the striation spacings measured in the resting fibre will be that given by extrapolating the tension record back to the beginning of stimulation. Most of the records

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

TENSION DEVELOPMENT IN STRETCHED MUSCLE 157 contained a small component detected by this procedure, which developed at a speed similar to that of an isometric tetanus at the slack length of the fibre. The results were plotted against mean striation spacing, and two such graphs derived from the same experiments as Figs. 7-11 are reproduced in Figs. 12 and 13. The resting tensions are also shown in these 0-251 mm ZI

0-25 --1< i

JL iI ;

0-18

mm .ZILLL.L IL

kg/cm2

018

A, 3 92,u

k/

m2

2

0.18

i

.....

_

__

__

C, 3-74,

|

l

5

jy7l_

mm--

1Z 4 I i mmLLjjj K: 0-18 -I 0-25

kg/cm2

B, 3-85,

0

J _

-

kg/cm2

0 2 sec

0*5 05 Kmm fjJ IIi

.

- ._I,

0 25

__..__

__

F 0-2 sec E, 3-74# D, 3-64, Fig. 8. Tetanic responses of a nearly uniform part of the fibre length under servo control. Experiment of 28 August 1963, 40 C. Various striation spacings, as indicated on the figure. A is the same frame as Fig. 7A. Upper trace of each pair, position of right-hand marker; lower trace, tension. Note that amplification of both channels is half as great in D as in A, B, C and E. E, no stimulus, tension recorder moved by hand to the right, and other end of fibre is moved by the servo: note absence of tension change. F, times of stimuli. Sensitivity and speed figures refer to one grid square.

0-25 ~

0125 m0m 0

kg/cm2

A,

014

-

3-86,u

A, 386u02 0 2 sec

~U

'

2-5

mm

kg/cm2 2-5

_

----350 36ukgcm kg/cM2

D, 3-56,u 25

mm

mm

25

-

0-69__

kg/cm2

-

G,

~

I-

014

,

kg/cm2

B~~,3.77#e __

_

_

_

_

_

_

_

_

mm

035

kg/CM2 k

~

_1-2

mm

3-03#L-k/m

mm

2-5

C, 3 -67# _

_

_

_

_

_

_

_

_

mm

069 kg/cm2 kg/

E, 3451u _

_

_

_

_

_

_

_

_

_

__2_

_

69

-

kg/CM2

0.25

F, 3 24# _

_

mm

-

-0-14

H,

2-85u

J, 3 77#

0-2 sec

Fig. 9. As Fig. 8, but another experiment (20 September 1963, 30 C). Note changes of amplification. A is the same frame as Fig. 7 B. Times of stimulating shocks as in Fig. 7F. J: bodily displacement of the whole fibre, as in Fig. 8E. Sensitivity and speed figures refer to one grid square.

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

158 A. M. GORDON, A. F. HUXLEY AND F. J. JULIAN figures. It will be seen that the results deviate only slightly from the ideal case of a linear fall up to a striation spacing a little over 3-6 /t, and zero tension at greater degrees of stretch. The corner is somewhat rounded, but the greatest deviation from the ideal result is only about 3-5 % of the tension developed by the fibre at its optimum length. L. end

L. marker

R. marker

R. end

4-0

.50

38

0

0

00

0

8

3-6 0~~~~

1915 mm

0

2

4

6

8

10

12

14

16

18

20

Distance along fibre (mm) Fig. 10. Measurements of striation spacing in fibre of 28 August 1963. Open circles, visual measurements before markers were attached; filled circles, visual measurements after markers were attached; open rectangles, range of photographic measurements, with horizontal line at mean (twenty-one measurements on each frame between the markers; six on each of the two frames outside the markers). a, at position with much connective tissue adhering to fibre; b, at position of end-plate. The shaded rectangles indicate the positions where the markers were attached, and the vertical lines show the positions of the extreme ends of the fibre. Abscissa: reading of vernier on horizontal movement of mechanical stage. L. end

L. marker

R. marker

Xo 38

R. end

a

0 ~36o~~~~~~0o

~~~0

C

0

0

-4--~~~~~~~~--765

-~3.4

mm

0

30

20.1 mm

4a

0

2

4

6

8

10

12

14

16

18

20

Distance along fibre (mm) Fig. 11. As Fig. 10, for fibre of 20 September 1963. Each photograph measured at six positions. a, pair of measurements at position of end-plate.

Origin of slow tension rise Huxley & Peachey (1961) attributed the slow development of substantial amounts of tension in highly stretched fibres to the shortening of the ends, where the filaments, still overlapped, with corresponding stretch of the

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

TENSION DEVELOPMENT IN STRETCHED MUSCLE 159 0-5 0-4 Resting tension b 0-3 bo

.'0 e 0-2 C) D

I

0-1

0

(2) xl

(4)

3-9 3-7 4-0 3-8 3-6 Striation spacing (i4) Fig. 12. Graph of tension against striation spacing for fibre of 28 August 1963. Crosses, maximum tension developed in servo-controlled tetani; filled circles, tension in same tetani extrapolated back to start of stimulation as shown in inset of Fig. 13; means from two to four tetani as indicated. Duration of tetanus, 1 sec at 20 shocks/sec. Vertical bars: range of resting tensions, which varied considerably according to sequence of length changes. Tensions related to cross-sectional area at striation spacing of 2-1 /t. Isometric tension at spacing of 2-72 ,t was 1-70 kg/cm2, which corresponds to about 2-6 kg/cm2 at the optimum length.

3-5

x

_~~~__

1 5 I-

2.Y

0- -7 -=.--

I-0

-

x

0o ._

x

C

to

0-5

F

x 0

6 0

a

I

I

I

28

29

30

31

II

1

I30

3-2

0

a

0

3-3

3-4

35

36

3-7

3-8

I

3-9

Striation spacing (,u) Fig. 13. As Fig. 12, for experiment of 20 September 1963. Only one measurement of resting tension at each length, shown by open circles. Tetanus, 1 sec at 20 shocks/ sec. Tetanic tension at optimum length about 2-8 kg/cm2. Inset shows method of extrapolating back to start of tetanus. Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

160 A. M. CORDON, A. F. HUXLEY AND F. J. JULIAN middle part of the fibre. This explanation is strongly supported by the observation that the amount of tension developed is reduced to a small fraction when the length of a nearly uniform part of the fibre was held constant. If this is correct, it suggests that the small slow rise of tension that was sometimes found even with the servo arrangement described here (e.g. Fig. 14), in which the effect of shortening of the ends of the fibre is eliminated, may have been due to the residual irregularities of striation spacing within the part of the fibre between the markers. This interpretation is supported by the experiment illustrated in Fig. 14, which shows

0m5

mm~~~iLiiiLI ___

--33%J

0~~~~~~ m

015

kg/cm2 0_15 -e 005 sec A kg/cm

1-245 -

0.15

[ 25 ~~~~~0*31

_

kg/c 2

4-1 _____sec

u 2A

kg/cm D ~~~~~E Fig. 14. Servo-controlled tetani at the same length (mean striation spacing 3-83 it), with various procedures in between. Experiment of 6 August 1963; 5° C. Lower trace, tension; upper trace, position of right-hand marker except in C where it shows AL. A, the first tetanus undergone by this fibre. B, 5 min later, with no stimulation or change of length since A. Between B and C, fibre released to 8 = 3-73 , stretched to 3.93 ,u and released to 3 83 1z, without stimulation. Between C and D, fibre released to 8 = 2-8 ,A and stretched again to 3-83 ALL, three times over, without stimulation. Between D and E, fibra given two tetani at 8 = 3*52 A (note reduced amplification of tension trace in E). F, stimuli on same time scale. Sensitivity and speed figures refer to one grid square.

a series of tetani, all under servo control at the same length, with an average striation spacing of 3 81 /t. The treatment of the fibre between these successive tetani is described in the legend, and comparison of the tension records shows that the slow component of tension rise is greater in the second of two tetani with the muscle left alone in between (compare B with A) or with a tetanus at shorter length in between (compare E with D), while the slow rise was decreased if the fibre was pasively released and stretched in between (compare C with B, and D with C). The first two procedures would be expected to increase, and the last to diminish, any irregularity of striation spacing.

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

TENSION DEVELOPMENT IN STRETCHED MUSCLE 161 Resistance to stretching As mentioned in the Introduction (p. 144), Carlsen et al. (1961) concluded that sarcomeres stretched to a length where there is no overlap must either be capable of developing substantial tensions or else must undergo a large increase (order of fivefold) in their resistance to stretch. The results presented on pp. 155-160 show that only a very small amount of tension is developed so long as the length of the part of the fibre in question is held constant. The possibility of an increase in stiffness on stimulation was examined in experiments of two types. Manual stretch. In the first, the rise of tension during a tetanus with the tendons held stationary was compared with the rise of tension when the same fibre was stretched, without stimulation, so that the part between the markers was extended by an equal amount. Records obtained in two such experiments are shown in Fig. 15. A and B show the tension developed in fixed-end tetani (lower traces) and the increase in separation of the 20. ix. 63

28. viii. 63

24,

;

47'I ,.

0.18

_r.

l...l.

t...

.. | ..

..1..

0.14

kg/cm1

kg/cm2C'

-I

0-2 sec

A

I T1

1

1

i

B

0-2 sec

1

1

2441u .IJ 1¶ LLLLI 47,u Iv ....

0k18 kg/CM2

..

...

..

.. ..

I

4

14

=

1 1 1

1

,1 ..

1

-!7:

kg/cm2 C

0-2 sec

D

02 sec

....... l.li.~~~~~~~,.I

....i,-,Ir-,,,,, I, i,, ..m .

E

F 02 sec 0-2 sec Fig. 15. Comparison of tension increase during fixed-end tetani (upper frames) and during passive stretch of the same fibre without stimulation (lower frames). Upper trace: change in separation between the markers; lower trace: tension. A, C, E, experiment of 28 August 1963; mean striation spacing between the markers 3-92p at beginning of the records; temperature 40 C. Distance between markers 6-1 mm, so that one square of deflexion of the length trace represents 0 39 % stretch. B, D, F, experiment of 20 September 1963; initial mean striation spacing 3-86 p; temperature 30 0. Distance between markers 7-65 mm, so that one square of deflexion of the length trace represents 0.62 % stretch. Sensitivity and speed figures refer to one grid square. 11 Physiol. 184

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

162 A. A. GORDON, A. F. HUXLEY AND F. J. JULIAN markers (upper traces) due to shortening of the end parts of the fibre. In C and D, the same fibres are stretched, without stimulation, by moving the 'motor stop' manually, and the same quantities, tension and separation of the markers, are recorded on the two traces. The manual stretch was carried out, as nearly as could be judged visually on the oscilloscope, at the same speed as the elongation during the tetanus. 07

-

A

x

, x

-~~~~

iB020

0-5 04

004-006 008 010 0-12 /02

+1

03~~~

n

C~~~~~~~~~~~-

001

002

00

/ (

0~~~~~~~~~~~

03

0w

co 0.2

+ '

00

0

0 0-02 0-04 0-06 0-08 0-10 0-12 Increase in striation spacing (iu)

0~

0-03 0.02 0 0.01 Increase in striation spacing (u)

Fig. 16. Graphs of tension increase against increase of length between the markers during fixed-end stimulation (crosses and full lines) and during passive stretch (circles and broken lines). A, experiment of 18 March 1963, temperature 60 C, initial striation spacing 3-88 ,u. +, tetanus of 2 sec duration (recorded only up to 0-5 sec); *, passive stretch; x, tetanus of 2 sec duration, in that order. B, experiment of 20 September 1963, temperature 30 C, initial striation spacing 3-86 ,u. +, tetanus of 1 sec duration (Fig. 15B); 0, passive stretch, at rather irregular speed; *, passive stretch repeated (Fig. 15D). Tensions referred to cross-sectional area at 2-1 ,t striation spacing. Cross on tension axis shows tension in servocontrolled isometric tetanus at same length, extrapolated as in inset of Fig. 12.

In Fig. 16 the rise of tension is plotted against increase of distance between the markers for two experiments of this kind. It is seen that the points from each pair of records fall fairly well on a straight line, the tension dropping below the line as the speed of stretch decreases at the end. The lines referring to passive stretch of resting fibres pass through the origin, while those from fixed-end tetani have a finite intercept on the tension axis which in most cases agrees well with the tension measured at the same length in servo-controlled tetani. The slopes of the lines, which are a measure of the resistance to stretch, are collected in column 8 of Table 1. In most experiments, the slope from a fixed-end tetanus was about 50 % larger than that from passive stretch. This suggests that the resistance

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

TENSION DEVELOPMENT IN STRETCHED MUSCLE 163 o0

:-

o

X

r. 0 .-

0L

12 12 x

=4 0

C 14

5r 4

c: 1e4

=

=s

CZ

l

=

1:

Cs

-

vt4

=

(o

ll

4

(::

m

C

m

m

C 0

2C

'

CD

CB Lb 4 6XCB

no

XD

=

C,

0

=

C".

-4

(=

=

0

x 0

0

00c

=

=

0o]- C0

N 00

00

C)

-.

CD* 01

=s (A

'-_t

N0

0-

0.

A

41

~ .I -

o;

N

0 =,

>-

;D

oI

X

o

2.0

9 4

711 0

0: -

41

_

C! 01

01

.0~~

-~

01

0

"

C; ~~~

0e

CD "~~~~~0

"0 "0 0

12 4 - 4t0

oo g Cl 0;

10

o~ ,, 2015co

e

.0

bO

0 ~~0

0 ~~

~~

0

.0

C)

e

Qp > p > 41 00~~~~

n

00

10

01 CS

2

_IC:t >_

. m~ ~ ~ ~1

O N

41

P1 4

12

12

0x,Q C) 0

0 0

11 n

L

0

-

N

o

2

CA

4Z Il4.d C)~ 1l)

01

t-

U

N

0

C0 e

Ce 00e r

0

0) O0 r2

0,*C,

410

0. 0

'Q

O2c O

m

3

cb c2

Wi

~~~~~~~

12

Cam

II

01-

S

124z

0Q0

0"

-

x

.

-I

Ht O

12"e 0 rn 00cz cS1) O

00,

(

4

4-0

o

EH

0;C0DO

12=

41

0

0

0:

c2

00

= ._

t

cXj

01

C0

11-2 Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

164 A. M. GORDON, A. F. HUXLEY AND F. J. JULIAN of a fibre to stretch is increased in this proportion as a result of stimulation, but three other factors account for much of the difference. 1. In cases where two stimulated records (18 March 1963) or two resting records (20 September 1963) were made on the same fibre, the slope of the line was 10-20 % less in the second than the first. In all cases except one the stimulated record was taken before the resting, so that this progressive decline in the slope would make the value for the resting case too low for comparison with the stimulated. In the one experiment (18 March 1963) where a second stimulated record was taken after the resting one, the slope of the resting line was intermediate between the values for the two stimulated ones, being very close to their mean. 2. Insomecases (6August 1963, 28August 1963)therateofelongationwas appreciably less in the unstimulated stretch than in the fixed-end tetanus. 3. In the experiments of 5 March 1963 and 6 August 1963 there was a considerable slow rise of tension even in a servo-controlled isometric tetanus at the same length, probably due to the fact that the striation spacing showed wide variation even in the part of the fibre between the markers (these experiments were done before the full procedure described on pp. 154-155 had been worked out). If this tension rise is subtracted from that observed in the fixed-end tetanus, the slopes of the stimulated and resting lines become approximately equal. It is not possible to say with any precision how much of the difference between the stimulated and resting slopes is due to these factors, but it seems unlikely that the resistance to stretch at speeds of this order is increased by more than 20-30% as a result of stimulation, and it is possible that the effect is much smaller. Square-pulse stretch. The second type of experiment consisted in subjecting the fibre to a nearly instantaneous increase of length, through the servo control, and recording the tension, both with the fibre at rest and during a servo-controlled isometric tetanus. Records from three such experiments are shown in Fig. 17. The tension rise is greater during stimulation than at rest; measurements on the records are collected in Table 2. The measure of stiffness tabulated in columns 7 and 8 is directly comparable with that of Table 1, column 9. For each experiment which appears in both tables, the value derived from the manual stretch is repeated in column 9 of Table 2, and it will be seen that it lies between those for the initial rise and the final level of the square pulse stretch in every case except one of the values for the experiment of 20 September 1963, where the pulse duration was less than in the other experiments. In all cases, the rise of tension is greater when the stretch was applied during stimulation than at rest, and there is a suggestion that the effect was larger as regards the initial rise than for the level at the end of the Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

TENSION DEVELOPMENT IN STRETCHED MUSCLE 165 pulse, though the figures for the latter are more uncertain because of the necessity for interpolating for the tension change that would have occurred in the absence of the stretch. The increases are about 1-3 times for the end of the pulse and about 1.5 times for the initial rise, except at the striation spacing of 3-77 ,tt on 20 September 1963, when the increase was 2*5-3 times. Passive stretch 6. viii. 63

s=4-004

During tetanus

l!-

1m25m mm~~~~~~~~~ ~~mmPosition h+ I

0-31

kg/cm2

.

20. ix. 63 ° 31%;

0-31

0 5 sec

A

to

:44.

kg/cm2

Tension

B

0.31%

Length

s=386p kg/c2 _____ 10-07 . C 0-2 sec kg/cm2 D 20. ix. 63 kg/cm2

0.31%

%

s=3-806ic

0.31% ~~~~kg/cM2

0-2 sec

.

-

0-14 E

osiio

Tension

Length

=

Tension

F

Fig. 17. Tension increase on extension of highly stretched fibres at rest (left) or during isometric servo-controlled tetanus at same length (right). Lower trace: tension; upper trace: in A and B, position of right-hand marker, in C, D, E and F, distance between markers. A and B, experiment of 6 August 1963, 50 C, striation spacing 400 /%. C-F, experiment of 20 September 1963, 30 C: C and D, striation spacing 3*86 ,u; E and F, spacing 3*77 ,u. Sensitivity and speed figures refer to one grid square. DISCUSSION

Origin of tension in fixed-end tetani Our results confirm that a tension rise of about the magnitude seen by Carlsen et al. (1961) does occur during a tetanus in a fibre held with the ends stationary, even when it is stretched so far that presumably there is no overlap of filaments in the main part of the length of the fibre. We found, however (Fig. 7), that this tension was reduced to a small fraction when we excluded the effects of shortening in the ends of the fibre where substantial overlap of filaments existed, and also (Figs. 15-17, and Tables 1 and 2) that, if stimulation causes an increase in the resistance to stretching, this increase is of a smaller order of magnitude than the five- or sixfold rise suggested by Carlsen et al. These results appear to us to show that the large tensios recorded in fixed-end tetani in these highly stretched fibres are Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

X

A. M. GORDON, A. F. HUXLEY AND F. J. JULIAN

166

12-

7 C. ._

S C.

C.

CZ

0

--4. Ct C)

040

S

2.

C

-

*

bT 0

r-Q

_l

M

0co C9 01t0101- 000 't c0 " 01 CM M Lf

-4

N

---

0.

P0 0

W

C;

R

;a

M 01

I

-

017 r-

_-

1-f

.4.

0=0)

W0212

S4

C.

0t1< ---

OCo

010. C

e

e

COee

O

00000000000000

CO CO

O G

O O

0

0

O

z

O -

C)

-.

4+ 00

H

00

d0

00

41s13!3°1°:l~.

00 C4 an

0

0

C C C91e

°5eCO °"~ ^ 9 CC.°

S;

0

.

t

:

s C.oo

0.

0 Cq 0M- 100 00 0 M M " 01 Nt C 10 C 12- 10 CO 10 011

T

C

21

C C_

0

O I

It

- M

I.,

.

0 a4~~~~~~~~~~~~~~~~4 ,>e ~ ~ ~

-.00

I

ce

212~ ~ 4._

".lf, Z

LI to

Ca

C)

e.

- 0P

0CO0+.c H;X1

0

CI IO M I_,CD

I OC co CS1 MD Cq

0" f_~

0s

C.

0

0:

+ t10W

= C0

4Z

CO

M CD CO

OO

CO CO

--

M

00

C;O

1

C

r--r-r-

----

C CO

CO CO

r- r-

CO-~CO

r

CO CO

CO-.CO

CO

CO

C.

0H

10

C

O

.4.

,o

0 = > E, , g

X o_

0

EH,

b

X

_; C.21

0 2s 21

o

t

0

-

M

CO

6

C1

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

02

TENSION DEVELOPMENT IN STRETCHED MUSCLE 167 due to shortening of the ends of the fibres, as suggested by A. F. Huxley & Peachey (1959, 1961) without any large change of properties in regions where there is no overlap of filaments. Objections to the sliding filament theory based on tension development in highly stretched fibres with the tendons held stationary are therefore not valid. Our results do, however, suggest that tension development and stiffness increase of an altogether smaller order of magnitude do occur; these are discussed in the next section. The point where there is a direct disagreement between our results and those of Carlsen et al. (1961) is that we found that the tension rise in a fixed-end tetanus is almost completely accounted for (within a factor of 1 5 x ) by the stretching of parts of the fibre with no overlap, while Carlsen et al. claimed that the tension rise expected from this stretching (assuiming no change of properties on stimulation) was only 15-20% of that observed. A large part of this discrepancy is in the values for the resistance to stretch of the unstimulated fibre. Column 10 of our Table 1 is expressed in the same units as column 2 of Carlsen et al.'s table IV; our values for unstimulated fibres (for stretch at about the speed which occurred in the fixed-end tetani) are about five times larger than Carlsen et al.'s for the static stiffness, while the multiplying factor used by Carlsen et al. to convert the static figure to the 'semidynamic' conditions in a tetanus was only 2. The corresponding values for stiffness in the stimulated fibres can be obtained from Carlsen et al.'s table IV by dividing each figure in their column 3 by the product of the figures in the same row in columns 1 and 4; the results are a little larger-averaging about 1-35 times-than those for stimulated fibres in coluimn 10 of our Table 1. This may well be a real difference, since we took extreme precautions to avoid irregularities of striation spacing in the part of the fibre between the markers, and any irregularities sufficient to produce overlap would cause an increase in the apparent stiffness on stimulation.

Residual tension development in fibres at extreme length Although at spacings above 3*65 /t, where no overlap of filaments is to be expected, the tensions we recorded in servo-controlled isometric tetani were of a smaller order of magnitude than are observed when the tendons are held stationary, we did, nevertheless, record detectable tensions up to striation spacings of 3*8 or 3 9 ,c. These records always included a rapid component with roughly the same time course as in a tetanus at the optimum length, and sometimes also a slow component which rose throughout the period of stimulation employed by us. The slow component may reasonably be attributed to the presence of residual irregularities of striation spacing, so that overlap still existed at some points between the markers; such parts Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

168 A. M. GORDON, A. F. HUXLEY AND F. J. JULIAN of the fibre would be expected to shorten slowly and cause a corresponding rise of tension as suggested by Hill (1953), the effect being similar to the rise of tension due to shortening of the ends of the fibre in a tetanus with the tendons held stationary, which is likewise slow. Evidence strongly supporting this interpretation was presented on p. 160, and irregular shortening of the kind required for this explanation was observed directly under the microscope by A. F. Huxley & Peachey (1961, p. 157 and P1. 2). It seems to us, however, that any tension rise due to effects of this kind would necessarily have a slow time course, and that therefore the relatively fast component cannot be explained in this way. Further evidence for the reality of the fast component is provided by graphs such as are shown in Fig. 16, plotting tension increment against increase in separation of the markers. For unstimulated fibres the lines pass satisfactorily through the origin, but for stimulated ones they extrapolate back to an amount of tension about equal to that which appeared in a servo-controlled tetanus at the same length. Several possible explanations for this fast component suggest themselves and are listed below, but there does not appear to be any evidence at present on which to reach a firm conclusion about its origin. 1. Irregular positioning of filaments within a fibril, or existence of filaments of greater length than normal. The occasional overlap that might exist from these causes would not be expected to increase slowly and progressively if the fibril structure is rigid enough. 2. The thick filaments are tapered at their ends (H. E. Huxley, 1957, 1963). Although it is not evident in the published pictures of H. E. Huxley (1963), it is possible that the number of bridges per unit length may decrease as the filament gets thinner. The tapering appears to extend over about 01 #tt at each end of the filament; its effects would therefore be likely to spread over about 0-2 /t of striation spacing. 3. There may be a length change in one or other of the filaments, associated with activation, though if this were the explanation it is not clear why the tension rise should disappear at striation spacings above about 3-9 It. The small increase of stiffness which is suggested by the observations described on p. 164 may be associated with the same cause as the residual tension increase. REFERENCES

BECE, 0. (1922). Die gesamte Kraftkurve des tetanisierten Froschgastrocnemius und ihr physiologisch ausgenutzter Anteil. Pflflger Arch. ges. Physiol. 193, 495-526. BiNKs, J. R. (1965). Influence of osmotic strength on cross-section and volume of isolated single muscle fibres. J. Physiol. 177, 42-57. Brix, M. (1895). Die Lfinge und die Spannung des Muskels. Dritte Abhandlung, Skand. Arch. Physiol. 5, 150-172; Vierte Abhandlung, Skand. Arch. Phy8iol. 5, 173-206.

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

TENSION DEVELOPMENT IN STRETCHED MUSCLE 169 CARLSEN, F., KNAPPEIS, G. G. & BUCHTHAT, F. (1961). Ultrastructure of resting and contracted striated muscle fiber at different degrees of stretch. J. biophy8. biochem. Cytol. 11, 95-117. GORDON, A. M., HuxLEY, A. F. & JuiAN, F. J. (1963). Apparatus for mechanical investigations on isolated muscle fibres. J. Phy8iol. 167, 42-44P. GORDON, A. M., HuLEY, A. F. & JumN, F. J. (1964). The length-tension diagram of single vertebrate striated muscle fibres. J. Phy8iol. 171, 28-30P. HEIDENEAIN, R. (1864). Mechani8che Lei8tung, Wdrmeentwicklung und Stoffumsatz bei der Muskelthdtigkeit, p. 93. Leipzig: Breitkopf und Hartel. Hn.TT, A. V. (1953). The mechanics of active muscle. Proc. R. Soc. B, 141, 104-117. Hu EY, A. F. & NIEDERGERKE, R. (1954). Interference microscopy of living muscle fibres. Nature, Lond., 173, 971-973. HuXLEY, A. F. & PEACHEY, L. D. (1959). The maximum length for contraction in striated muscle. J. Phy8iol. 146, 55-56 P. HuXLEY, A. F. & PEACHEY, L. D. (1961). The maximum length for contraction in vertebrate striated muscle. J. Physiol. 156, 150-165. HuXLEY, H. E. (1957). The double array of filaments in cross-striated muscle. J. biophy8. biochem. Cytol. 3, 631-648. HuXLEY, H. E. (1963). Electron microscope studies on the structure of natural and synthetic protein filaments from striated muscle. J. molec. Biol. 7, 281-308. NATORI, R. (1954). The property and contraction process of isolated myofibrils. Jikeikai med. J. 1, 119-126. PAGE, S. & HuLEY, H. E. (1963). Filament lengths in striated muscle. J. cell Biol. 19, 369-390. PALME.R, J. F. & READ, G. L. (1962). A general purpose low-level d.c. amplifier suitable for biological use. J. Phy8iol. 161, 35-37P. PODOLSKY, R. J. (1964). The maximum sarcomere length for contraction of isolated myofibrils. J. Physiol. 170, 110-123. RAMSEY, R. W. & STREET, S. F. (1940). The isometric length-tension diagram of isolated skeletal muscle fibers of the frog. J. cell. comp. Phy8iol. 15, 11-34.

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008

Tension development in highly stretched vertebrate muscle fibres A. M. Gordon, A. F. Huxley and F. J. Julian J. Physiol. 1966;184;143-169 This information is current as of March 25, 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/misc/ Permissions.shtml

Reprints

Information about ordering reprints can be found online: http://jp.physoc.org/misc/r eprints.shtml

Downloaded from jp.physoc.org at BIUSJ (Paris 6) on March 25, 2008