Altered Ca2' Dependence of Tension Development in Skinned ...

THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1986 by The American Society of Biological Chemists, Inc.

Vol. 261, No. 13, Issue of May 5, pp. 6096-6099,1986 Printed in U.S.A.

Altered Ca2’ Dependence of Tension Developmentin Skinned Skeletal Muscle Fibers following Modification of Troponin by Partial Substitution with Cardiac TroponinC* (Received for publication, August 2, 1985)

Richard L. Moss$, Michael R.Lauer, Gary G. Giulian, and Marion L. Greaser5 From the Departmentof Physiology, University of Wisconsin Schoolof Medicine and the §Department of Meat and Animal Science and the MuscleBiology Laboratory, College of Agriculture and Life Sciences, Universityof Wisconsin, Madison, Wisconsin53706

Binding of Ca2+ to the troponin C (TnC) subunit of troponin is necessary for tension development in skeletal and cardiac muscles. Tension was measured in skinned fibers from rabbit skeletal muscle at various [Ca”’] before and after partial substitution of skeletal TnC with cardiac TnC. Following substitution, the tension-pCa relationship was altered ina mannerconsistent with the differences in the number of low-affinity Ca2+-bindingsites on the two types of TnC and their affinities for Ca2+.The alterations in the tension-pCa relationship were for the most part reversed by reextraction of cardiac TnC and readdition of skeletal TnC into the fiber segments. These findings indicate that the type of TnC present plays an important role in determining the Ca2+dependence of tension development in striatedmuscle.

The regulation of tension development in striated muscles involves the binding of Caz+t o low-affinity sites on the TnCl subunit of the regulatory protein troponin, which is localized at regular intervals along the thin filament (seeRef. 1 for a review). Troponin in t u r n is b o u n d t o a second regulatory protein, tropomyosin, which in relaxed muscle acts either to sterically block the myosin-binding site on actin (2) or to block a kinetic step i n the cross-bridgeinteractioncycle, (3). possibly the ejection of Pi from the cross-bridge head Once Ca2+ has b o u n dt o TnC, tropomyosinundergoes a change in position relative tothe thin filament such that the myosin cross-bridges can then cyclically interact with actin of to produce tension and mechanical work. The amount be varied by changing tension developed bya muscle fiber can the a m o u n t of free Ca2+i n the myoplasm, which determines the numberof available cross-bridge binding sites on the thin filament. The relationship between isometric tension development and free Ca2+is much steeper in fast-twitch skeletal than in cardiac (4)or slow-twitch skeletal muscles (5). Such * This work wassupported by Grants HL25861 and AM31806 from the National Institutes of Health and by the School of Medicine and the College of Agricultural and Life Sciences of the University of Wisconsin. A preliminary report of this work has been published elsewhere (Lauer, M.R., Moss, R. L., and Greaser, M. L. (1984) Circulation 70, 11-277). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. j: Performed this work during the tenure of an Established Investigatorship from the American Heart Association and with funds contributed in part by the Wisconsin Affiliate. The abbreviations used are: Tn, troponin; S-TnC, skeletal troponin C; C-TnC, cardiac troponin C; SDS, sodium dodecyl sulfate; EGTA, [ethylenebis(oxyethylenenitrilo)]tetraaceticacid.

a difference in steepness, in terms of current models of contractile activation ( 6 ) , might be explained on the basis of greater cooperativity in fast-twitch muscle,perhaps resulting from end-to-end interactions of adjacent tropomyosin molecules (7). These interactions are thought to enhance Ca2+ binding in regions of the thin filament adjacent toan already activated troponin-tropomyosin complex. In addition, fasttwitch skeletal TnC (S-TnC) has two low-affinity Ca2+-binding sites (S), whereas cardiac TnC (C-TnC) (9-11) and slowtwitch skeletal TnC have only one (12). However, it is not clear whether this difference in the number of Ca2+-binding sites translates into a difference in the Ca2+ sensitivity of tensiondevelopment. To examine this possibility,experiments were performed to determine whether partial substitution of S-TnC by C-TnC alters the relationship between tension andpCa (ie.-log[Ca2+]) in single skinned fibers from rabbit psoas muscles. MATERIALS ANDMETHODS

Psoas muscles were dissected from adult male New Zealand rabbits. Bundles of fibers were stripped free while in relaxing solution, tied to glass capillary tubes, and then stored a t -22 “C in relaxing solution containing 50% (v/v) glycerol for several days before use (13). Individual skinned fibers were then pulled free and mounted in the experimental chamber (14) between a force transducer (Model 403, Cambridge Technology, Cambridge, MA) and a DC torque motor (Model 300s, Cambridge Technology). Sarcomere length in the relaxed fiber segments was adjusted to 2.5pmby changing overall segment length, so that sarcomere length during contraction was approximately 2.4 pm. The fiber segments were activated in solutions containing various concentrations of free calcium between 0.1 and 10 p ~which , are expressed aspCa values (i.e. -log[Ca*+]) in the present report. The relaxing and activating solutions were identical to those of Julian (15). At any given pCa, a steady tension was allowed to develop, a t which time the segment was rapidly (i.e. within 1 ms) slackened and was subsequently relaxed. The difference between the steady developed tension and the tension base line obtained immediately following the slack step was measured as totaltension. Active tension was calculated as the difference between total tension and the resting tension (usually less than 1 mg in weight) measured in the same segment while in relaxing solution. Tensions (P)at submaximally activating levels of calcium were expressed as a fraction of Po, the tension obtained during maximal activation a t pCa 5.49. Every third or fourth contractionwas performed a t pCa 5.49 in order to assess any decline in fiber performance (12). Tension-pCa relationships were obtained ( a ) in the untreated fiber segment and ( b ) following partial extraction of endogenous S-TnC by bathing the fiber for 90-120 min a t 15 “C in a solution containing 20mM Tris and 5 mM EDTA, pH 7.8 (16, 17), and subsequent reconstitution with bovine C-TnC. Reconstitution was accomplished by bathing the fiber segment for 10 min in relaxing solution containing 0.5-1.0 mg/ml CTnC and was followed by two 5-min washes in relaxing solution in order to remove excess C-TnC. In some experiments, a final tensionpCa relationship was determined after re-extraction of the C-TnC and reconstitution with S-TnC. The results of previous control experiments (18) demonstrated that fibers from which S-TnC was first

6096

6097

Contraction in Skeletal Muscle Substituted with Cardiac TnC extracted and then recombined had contractile properties that were virtually identical to control. The TnC content of the fibers a t each step of the experimental protocol was determined by performing SDS-polyacrylamide gel electrophoresis on short segments of the same skinned fiber (19). Each fiber was divided into two or three segments of approximately equal lengths. The first of these was dissolved in SDS-containing sample buffer (19). One of the remaining segments was tied between the force transducer and motor for use in the physiological measurements. In thoseexperiments in which the experimental fiber segment was finally reconstituted with C-TnC, an additional segment was tied at both ends to the motor arm so that it was exposed to the Same solutions as the experimental segment. Following the treatments to partially extract S-TnC and recombine C-TnC, the segment tied only to the motor arm was removed and then dissolved in sample buffer. Finally, following the extraction of C-TnC and recombination of sTnC and the subsequent determination of the relative tension-pca relationship, the experimental segment was dissolved in sample buffer. Once the staining and drying procedures were completed, the gels were scanned using a Bio-Med Instruments (Fullerton, CAI laserlight scanning densitometer (19). The relative amount of S-TnC or C-TnC present in any given fiber segment was taken as the ratio of the integrated areaof the appropriate peak to thesum of areas of the S-TnC and C-TnC peaks. Total TnC content ( i e . C plus s) of the fiber segments was also referenced to themyosin light chain1content of the same fiber segment. If total TnC determined inthis way varied by more than 10% between control and reconstituted segments, that fiber was excluded from the analysis of the proportions of S-TnC and C-TnC present in the fiber segments at each stage of the protocol. RESULTS AND DISCUSSION

Fig. 1represents tensionrecords from a single fiber segment at various stages of the experimental protocol. Following 90 min of exposure to theEDTA-containing solution, maximum developed tension decreased by approximately 70% ( b versus a). The mean S-TnC content of the extracted fibers, determined by SDS-polyacrylamide gel electrophoresis (Fig. 2), was 40% of control levels. Bathing this same fiber in relaxing solution containing bovine C-TnC (0.8 mg/ml) for 10 min resulted in a recovery of Po to over 90% of its control value (Fig. IC).For reasons that are notyet clear, still longer soaks in the solution containing C-TnC or increasing the concentration of C-TnC did not result in furtherrecovery oftension but in many cases actually resulted in a decline from the maximum observed at approximately 10 min. On average, maximum Ca2+-activated tension following C-TnC recombination was 92% of the control values measured in the same fibers. At very low Ca2+concentrations (pCa 6.70 in Fig. 1, d and e ) , the developed tension was found to be greater with CTnC present when compared to control; however, at relatively high Ca2+concentrations (pCa 6.09 in Fig. 1, f and g), steady isometric tension was less with C-TnC present. Tension-pCa relationships areshown in Fig. 3 for one fiber after partial extraction segment under controlconditions (0), of S-TnC and reconstitution with C-TnC (O), and finally, after another period of extraction followed by reconstitution of the fiber with S-TnC (x). C-TnC substitution intothe fiber resulted in an increase in tension relative to control for pCa > 6.51 and a decrease in relative tensions in the range of pCa values between 6.51 and 5.0. Thus, in thepresence of C-TnC, there was a rightward shift of the upper part of the tensionpCa relationship and a leftward shift of the lower part. Also apparent from Fig. 3 is that subsequent recombination of STnCintothe fiber segment resulted in recovery of the tension-pCa relationship to near its original form; however, in most cases, recovery to theform of the control relationship was incomplete, a phenomenon which was correlated with incomplete removal of C-TnC (determined by SDS-polyacrylamide gel electrophoresis) prior to S-TnC recombination. Thus, for many fibers, the tension-pCa relationship following

i

I A 5.00

R

A 5.00

S R

s

A 6.70

S

A

n

6.70

A 6.09

s

A 5.00

S I

A 6.09

S R

R

S

R

FIG. 1. Tension records obtained from a single muscle fiber prior to manipulation of TnC content (a,d , and 0 , following partial extraction of TnC (b),and in the same fiber following recombination with C-TnC (c, e, and g).Once the fiber has been mounted in the experimental chamber, sarcomere length was adjusted to about 2.4 pm, as determined by light microscopy (14). The fiber was initially placed in relaxing solution containing 100 mM KCI, 2 mM EGTA, 4 mM ATP, 1 mM MgCl,, 10 mM imidazole, pH 7.00 (15). In each tension record, the fiber was transferred from relaxing solution to the indicated pCa at the time point designated A . When a steady active tension was developed, the fiber was rapidly slackened (S)by imposing a length step (complete within 1 ms) at the motor end. This was done to obtain an accurate zero-force base line. Active tension was then measured as the difference between this base line and thepeak tension prior to the step from an oscilloscope record at a fast sweep speed (note: the tension trace did not drop to base line on the slow chart records shown). Following the length step,the fiber was returned to relaxing solution (R). The tensions obtained from these records, relative to the tensions obtained a t pCa 5.00 in the control condition. were as follows.

pca'

Control

S-TnC extracted

5.00 6.70 6.09

1.00 (a) 0.05 ( d ) 0.82 (f)

0.30 ( b )

C-TnC recombined

0.95 ( e ) 0.10 ( e ) 0.65 (g)

S-TnC recombination likely represented a composite relationship, with contributions from both C- and S-TnC as was also the case for the initial C-TnC substitution (Fig. 2, lane 2). A convenient way to examine the changes in the tensionpCa relationship due to partialsubstitution with C-TnC is to use the Hill plot transformationof the raw data (20-22). Data from 12 fibers were used to construct the Hill plot in Fig. 4. These data are best fit bytwo straight lines, one line for P/Po > 0.5 and another for P/Po < 0.5. For P/Po < 0.5, the Hill coefficient, n, was 3.73 under control conditions and 2.44 after partial substitutionwith C-TnC. For P/Po > 0.5, n was calculated as 1.71 under control conditions and 0.86 after reconstitution with C-TnC. In addition, there were distinct shifts of the two portions of the Hill plot following substitution with C-TnC, and thesewere consistent with the shifts in the tension-pCa relationship noted previously (Fig. 3). Certainly, the apparent decrease in Ca2+sensitivity observed for

Contraction inSkeletal Muscle Substituted with Cardiac TnC

6098

123

a.

4 5

1.0-

0.e ao h 0.6 .8 e .- 0.4 -o u" 0.2 -

ul

A

e

L

'7.0

-

6.5

FIG. 3. Relative tension-pCa relationship obtained from a single psoas fiber. Data were firstobtained from an untreated

LC,-

LANE I

5.0

PC0

C-TnC LC,-

b

5.5

6.0

1

control segment (0).then from the same segment following a 90-min extraction of S-TnC andsubsequent recombination with C-TnC (O), and finally following a second 90-min extraction period followed by recombination with S-TnC (X). For each of these three conditions, submaximal tensions have been expressed as a fraction of the tension developed a t pCa 5.0 under the same condition.

pCa < 6.51 is in keeping with the lower affinity for Ca2+at the Ca"-specific site on C-Tnc ( K = 2 X lo' M") uersus the sites on S-TnC ( K = 2 X 10' M-') (9). The biphasic form of the control Hill plot can be interpreted as reflecting cooperativity in cross-bridge binding (6). Cooperative mechanisms for which there is experimental evidence include effects of bound cross-bridges to increase the affinity ofCa'+ binding by TnC (23) andend-to-endinteractions involving tropomyosin to affect cross-bridge binding along the thin filament (7).In this regard, Walsh et al. (24) have recently reported that the cooperative response of the regulated actomyosin ATPase to increasing concentrations of Ca'+ was unaffected when the overlapping regions of adjacent tropomyosins were removed. Whereas this resultcertainly suggests that the cooperative response of the ATPase does not involve end-to-end interactionsof tropomyosin, Tawada et al. (25) came to the opposite conclusion based on otherwise similar experiments in which superprecipitation of actomyosin was observed. Given these contradictory findings and

LANE 3

and recombination with S-TnC; lone 4. C-TnC standard; lane 5, STnC standard. Lanes 1 3 each contain the equivalent of a 0.5-mm length of fiber. The total acrylamide content in the stacking gel was 3.5% (w/v) and 12% in the separating gel. Otherwise, the details of sample preparation,electrophoresis, silver staining, and densitometry were identical to the methods of Ciulian et of. (19). 6. densitometric scans of lanes 1 3 shown in a. The actual areas (in arbitrary units) of the peaks corresponding to the myosin light chains (LC),TnI, and TnC were as follows (note that theload in lane 2 was approximately 20% less than in l a n e s I and 3 as judged from the areas under the mvosin lieht chain Deaks). Lone 2 Lune 3 Lune I (contm')

LC1

FIG. 2. a, sodium dodecyl sulfate-polyacrylamide 295 gel of segmenta of the same single fiber obtained a t various stages of the experimental protocol. Lane 1, control segment; lane 2, fiber segment obtained following partial extraction of S-TnC and recombination with C124 of C-TnC TnC; lane 3, fiber segment obtainedfollowing re-extraction

235 TnI

S-TnC C-TnC LC, 107 LC8

492 315 137

(C-TnC recombined)

(S-TnC recombined)

390

496

58

154

78 591 125

43 1

510

Contraction in Skeletal Muscle Substituted with Cardiac TnC 1.q-

I-

1.0

I

v'

2

6099

form of the tension-pCa relationship is affected by the TnC that is present withinthe thinfilament and appearsto depend strongly on the number of low-affinity Ca2+-binding siteson TnC and theirrelative affinities for Ca2+.Second, the degree of cooperativity within the thin filament differs depending upon the type of TnC bound. Possible explanations for this effect may involve a decrease in interactions between bound cross-bridges and thelow-affinity Ca2+-bindingsite on C-TnC or a reduction in end-to-end interactionsamong tropomyosin molecules. Further experiments will be required to determine the relative importance of these two mechanisms in the Ca2+ regulation of tension development in striatedmuscles. Acknowledgment-We are grateful to James Graham for his assistance and toSusan Krey for preparation of the manuscript.

'

Log [Ca"]

REFERENCES 1. Leavis, P. C., and Gergely, J. (1984) Crit. Rev.Biochem. 16,235305 2. Haselgrove, J. C. (1973) Cold Spring Harbor Symp. Quant. Biol. 37,341-352 3. Chalovich, J. M., and Eisenberg, E. (1982) J. Biol. Chem. 257, 2432-2437 4. Kerrick, W.G.L., Malencik, D. A., Hoar, P. E., Potter, J. D., the apparently lower degree of cooperativity in the ATPase Coby, R. L., Pocinwong, S., and Fischer, E. H. (1980) pflugers response to Ca2+compared to tension and Ca2+,it remains to Arch. 386,207-213 be shown whether nearest neighbor interactions of tropomy5. Kerrick, W.G. L., Secrist, D., Coby, R., and Lucas, S. (1976) osin can account for the cooperativity apparent in the Ca" Nature 260,440-441 activation of tension development. In addition, evidence has 6. Hil1,yT. L. (1983) Biophys. J. 44, 383-396 7. Wegner, A. (1979) J. Mol. Biol. 131,839-853 been presented suggesting that there may be direct interactions between the troponin-?' of one functional group and the 8. Potter, J. D., and Gergely, J. (1975) J. Biol. Chem. 250, 46284633 tropomyosin of the adjacent functional group (26). 9. Potter, J. D., Johnson, J. D., Dedman, J. R., Schreiber, W. E., The idea of cooperativity within the thinfilament has been Mandel, F., Jackson, R. L., and Means, A. R. (1977) in Calciumincorporated by Hill and co-workers (6, 27) into models of Binding Proteins and Calcium Function (Wasserman, R. H., Ca2+activation of contraction. Asymmetry in simulated tenCorradino, R. A., Carafoli, E., Kretsinger, R. H., MacLennan, D. H., and Siegel, F. L., eds) pp. 239-250, Elsevier/Northsion-pCa relationships of fast-twitch muscle, corresponding Holland Biomedical Press, Amsterdam to thebiphasic nature of our Hill plots, was achieved in Hill's (6) model by imposing the constraint that no movement of 10. Van Eerd, J.-P., and Takahashi, K.(1975) Biochem. Biophys. Res. Commun. 64,122-127 tropomyosin could occur unless both low-affinity sites on TnC 11. Leavis, P. C., and Kraft, E. L. (1978) Arch. Biochem. Biophys. were occupied with Ca2+.With this is mind, changes in the 186,411-415 form of the Hill plot following C-TnC reconstitution lead to 12. Wilkinson, J. M. (1980) Eur. J. Biochem. 103, 179-188 relatively straightforward interpretations. ForP/Po> 0.5, the 13. Julian, F. J., Moss, R.L., and Waller, G. S. (1981) J. Physiol. (Lord.)311,201-218 value of n following C-TnC substitution (n = 0.86) is approximately one-half that obtained under controlconditions (n = 14. Moss, R. L. (1979) J.Physiol. (Lord.)2 9 2 , 177-192 15. Julian, F. J. (1971) J. Physiol. (Lond.)218, 117-145 1.71). This result is consistent with the finding that C-TnC 16. Cox, J. A., Comte, M., and Stein, E. A. (1981) Biochem. J. 195, contains only one low-affinity Ca2+-bindingsite as compared 205-211 to two on S-TnC (9-11). For P/Po < 0.5, the Hill coefficient 17. Zot, H. G., and Potter, J. D. (1982) J. Bid. Chem. 257, 76787683 following C-TnC substitution (n = 2.44) was smaller than that measured under control conditions (n= 3.73). Since the 18. Moss, R.L., Giulian, G. G., and Greaser, M. L. (1985) J. Gen. Physiol. 86,585-600 overall steepness of the Hill plot may reflect cooperativity 19. Giulian, G.G., Moss, R. L., and Greaser, M. L.(1983) Anal. among Ca2+-bindingsites along the thinfilament (28), as well Biochem. 129,277-287 as between cross-bridge binding and Ca2+binding (23), the 20. Hill, A. V. (1913) Biochem. J. 7,471-480 present resultsuggests that partialsubstitution of C-TnC into 21. Moss, R. L., Swinford, A. E., and Greaser, M. L. (1983) Biophys. J. 43,115-119 fast-twitch fibers produces a decrease in such cooperativity. The decrease in steepness with C-TnC may, at least in part, 22. Cornish-Bowden, A., and Koshland, D. E., Jr. (1975)J. Mol. Biol. 95,201-212 simply reflect the need to bind only one Ca2+(versus two in R. D., and Weber, A. (1972) Nature New Biol. 238,97fast-twitch muscle) in order to induce movement of the asso- 23. Bremel, 101 ciated tropomyosin. 24. Walsh. T. P., Trueblood, C. E., Evans..~R.. and Weber., A. (1984) . , In summary, the present results have shown that partial J. Mol. Biol. 182, 265-269 substitution of C-TnC for S-TnC in skinned fast-twitchmus- 25. Tawada, Y., Ohara, H.,Ooi, T., and Tawada, K.(1975) J. Biochem. (Tokyo) 78,65-72 cle fibers from the rabbit causes a reversible change in the form of the tension-pCa relationship to one which is more 26. Brauer, M., and Sykes, B. D. (1984) Biophys. J. 45, 239a T.L., Eisenberg, E., and Greene, L. E. (1983) Proc. Natl. characteristic of skinned cardiac or slow-twitch skeletal mus- 27. Hill, Acad. Sei. U. S. A. 80, 60-64 cle. In relation to Ca2+regulation of contraction, there are at 28. Grabarek, Z., Grabarek, J., Leavis, P. C., and Gergely, J. (1983) least two important implications of this finding. First, the J. Biol. Chem. 258, 14098-14102

FIG. 4. Hill plotof tension-pCa data before (0)and after (0) partial substitution with C-TnC. The straight lines were tit to the data using least-squares regression analysis. The error bars indicate one standard deviation ( n = 12). P, represents relative tension (P/Po).

'