.J Mol Cell Cardiol 22, 1117- 1124 ( 1990)
Slowly
Exchanging
Bo-Sheng
Pan, Kimberly
Calcium Binding Sites Unique Muscle Troponin C A. Palmiter,
Maria
Plonczynski,
to Cardiac/Slow
and R. John
Solar0
Department of Physiology and Biophysics, Universityof Cincinnati, College of Medicine, Cincinnati, OH 45267-0576, i!SA and Department of Physiology and Biophysics, CTniversity of Illinoi.r, Co&r qf Medicine at Chicago, Chicago, IL 60680, LTSA i Received 13 November 1989, accepted in revised form 20 April
I990)
B.-S. PAN, K. A. PALMITER, M. PLONCZYNSKI AND R. J, SOLARO. Slowly Exchanging Calcium Binditlg Sites Unique to Cardiac/Slow Muscle Troponin C. Journal of Molecular and Cellular Cardiologv (1990) 22, I 117-I 124. Evidence is presented for the existence of slowly exchanging Cal+-binding sites in troponin C CTNC of cardiac and slow twitch skeletal muscles. These sites were revealed in the course of experiments aimed at qkinnrd measuring the Ca ‘+-binding properties of TNC in the myofilament lattice. 4sCa bound to chemically muscle fibers or myofibrils of cardiac and soleus muscles was &ted by EGTA in a two-exponential timecoursr with a slow phase of a rate constant of about 2 x 1F4/s. The slow phase was not found in skinned librr or myofibrils of psoas. a fast skeletal muscle. However, skinned psoas fibers in which the native TNC was rcplaccd of soleus and cardiac preparations, indicating that the by C’TNC exhibited the slow 45Ca elution characteristic slowly-exchanging sites are located in CTNC. These sites are tentatively identified as the Ca’+--Mg” sltrs 01 CTNC on the basis that the slow phase was observed under conditions known to restrict Ca” binding tv the Ca’+ -Mg2 + sites. KEY WORDS:
Calcium;
Magnesium;
Calcium-binding
Introduction of the contractile activity of cardiac or slow, red myofilaments is essentially the same as that of fast, white striated muscle, but there are some interesting differences. In each of these types of muscle, the Ca2+ receptor is troponin C (TNC), but the variant of cardiac and slow, red muscle (CTNC) has three Ca 2+ binding dom ains instead of the four that are present in fast striated muscle STNC (Van Eerd and Takahashi, 1976; Wilkinson, 1980). Studies with isolated TNC or TNC in the TN complex showed that there are two sites with relatively high affinity for Ca2+ and Mg2+ in both STNC and CTNC, but whereas STNC possesses two sites of lower affinity for Ca2+ (Potter and Gergely, 1975), CTNC has but one low affinitv Ca2+-binding site (Holroyde et al., 1980). I; is apparent that only the lower affinity sites can exchange Cz2+ in the time frame of Ca2+
activation
proteins;
Troponin;
Heart:
+ 08 $03.00/O
Fast music.
the contraction-relaxation cycle of cardiac and fast striated muscle, and thus these sites appear to be the physiological “regulatory” sites (Robertson et al., 198 1; 1982). The slowly exchanging Ca2+-?vlg2+ sites of STNC have been characterized as “structural” sites. Although there have been no studies of the kinetics of Ca2+ exchange with tht Ca2+-MgZ t sites of CTNC, it has been generally assumed that the kinetics of thesr metal binding sites of STNC and CTNC are similar (Robertson et al., 1981). Yet, in the course of experiments in which we measured the Ca’+binding properties of skinned cardiac muscle preparations (Pan and Solaro, 1987!, we discovered that elution of 45Ca from cardiac fibers was much slower than expected if the kinetics of Ca2+ exchange of CTNC and STNC were similar. In work presented here. we show evidence that these slowly exchanging sites of skinned fibers are indeed on ‘I’NC
Please address all correspondence to: R. John Solaro, Department of Physiology :M.iC 901 J, College of Medicine at Chicago, Chicago, IL 60680. US.4. 0022-2828/90/101117
Slow muscle:
and Biophysics,
I‘ni~crtit~
,c‘r 1990 Academic
oi Illinois
Press Limited
1118
B.-S. Pan et al.
present in the myofilaments and are unique to the CTNC variant present in heart and red, slow myofilaments.
0.5 ml NaOH at 95°C and assayedfor protein using the Lowry procedure as modified by Pagani and Solar0 (1984). The elution solutions and the fiber digest were counted for 45Ca and 3H in a liquid scintillation specMethods trometer. To construct the time course of the Preparations elution, the 45Ca activity remaining in the Thin bundles of muscle fibers (0.5-1.0 mm in muscle preparation at a specified time point diameter) were dissectedfrom the soleusand was obtained by summing up the 45Ca activpsoasmusclesof rabbits and from the right ities eluted from the preparation after that ventricles of mongrel dog hearts as previously time point and the 45Ca activity in the fiber described (Blanchard et al., 1984; Pan and digest. Correction for unbound 45Ca was Solaro, 1987). The fiber bundles were chemi- made using the 3H activity eluted from the cally skinned by extraction at 4°C for 12 h in a preparation. relaxing solution containing 1o/o Triton The time course of elution of 45Ca from X-100, 10 mM EGTA, 5.5 mM ATP, 12 mM myofibrillar preparations was determined by creatine phosphate, 8.2 mM MgClz, 30 mM a filtration technique. The myofibrils (0.3 KCI, 60 mM imidazole, 60 PM phenylmethymg/ml) were preincubated for 60 min in a sulfonate fluoride, pH 7.0. The skinned fibers “binding” solution containing 60 mM KCl, 20 were stored at -20°C in a detergent-free mM imidazole (pH 7.0), 2 mM Mg2+, 20 PM relaxing solution with 50% glycerol and fur- total Ca2+, 3 @i/ml 45Ca, and appropriate ther dissectedto thinner bundles immediately quantity of EGTA to give a desired free before use. We (Pan and Solaro, 1987) and [Ca2+]. At zero time, the mixture was made others (Miller et al., 1985) have previously 13 mM in EGTA. To determine 45Ca bound to shown that this and similar skinning proce- myofibrils at various times, aliquots of the dures completely removed subcellular mem- myofibrils were filtered through Millipore filbranes. This was verified by transmission ters (HA 0.45 pm) and the filters were washed electron microscopy of the preparations, and with 50 mM KCl, dried and counted. enzyme marker assays, which showed no evidence for the presence of surface memResults brane, sarcoplasmic reticulum and mitochonWe first present results of experiments comdria (Pan and Solaro, 1987). Myofibrillar preparations were isolated as described by paring the time course of elution of 45Ca from Solar0 et al., 1971. skinned fibers of psoas, soleus and heart muscle that were pre-equilibrated with binding solution containing 45Ca and 3H-glucose Kinetics of e&ion of 45Ca from skinned jbers and (Fig. 1). The washout of 3H-glucose activity myojbrils in these preparations was essentially comA thin bundle (approximately 0.1 mm in pleted within 2 min (data not shown), and diameter) of skinned fibers was incubated for about the same time was required for elution 90 min in a “binding solution” containing 2 of 45Ca from psoas fibers. However, much mM Mg2+, 110 mM KCl, 60 mM imidazole longer times were required for complete re(pH 7.0), 1 mM D-glucose, 45Ca (6 &i/ ml) moval of fiber bound 45Ca in the caseof the and 3H-glucose (6 @i/ml) at 23°C. As deter- soleusand cardiac muscle. The relatively slow mined by atomic absorption spectroscopy, the time courseof elution of 45Ca from the soleus, free Ca2+ of the binding solution was 5-6 PM. is similar to what we have previously found in The preparation was gently blotted to remove skinned fibers from heart muscle (Pan and adsorbed liquid and then incubated sequenti- Solaro, 1987), and was evident whether the ally for specified periods in a seriesof vials binding was carried out in the presence or each with one ml of an elution solution of 10 absence of MgATP, or with the fibers unmM EGTA, 50 mM KCl, 2 mM MgC12, 1 mM loaded or held isometric. Moreover, in both glucose and 20 mM imidazole, pH 7.0. At the the cardiac and soleusfibers, further increases end, the muscle preparation was digested in in the EGTA concentration in the elution
Slow
Ca’
+-sites
Removal
on Cardiac
of
TNC
45Ca-Skrnned
20
1119
f 1bers
30 Time
(mln)
FIGURE 1. Time course of removal of bound Ca*+ from two prep arations of detergent treated fibers from (a), psoas (W) and cardiac (0) muscle. Initially Ca’+ was bound to the fibers at pCa 5.3 and then removed presence of 10 mM EGTA as detailed in the METHODS. See Table I for a summary of kinetic parameters.
solution or replacing EGTA with CaClz of equal or higher concentrations did not accelerate the removal of bound Ca2 + . As shown in Figure 1, in the caseof the soleusand cardiac preparations, the data exhibited a biphasic feature, and could be fitted with an equation of the form, F = AIe-k11 + A2e -k2t, where F denotes the fraction of 45Ca remaining bound (as percent of bound 45Ca at time zero), kr and k2 are rate constants, and A, and A,, the fractional amount of the two exponential components. The data from psoas muscle were fitted to a single exponential. Table 1 lists the rate constants obtained in the curve fitting analysis. For both cardiac and soleus fibers, the rate constant of the slow component was about 30-fold lower than the rate constant of the fast component, which was similar in magnitude to the single rate constant of the psoasfibers.
soleus in the
We also tested if the differences in the time course of exchange of bound Ca2+ are also a property of myofibrillar preparations. The detergent-extracted myofibrillar preparations are free of membranes (Solar0 et al., 1971; Pan and Solaro, 1987) and about 1 pm in diameter. Thus diffusional delays are negligible when eluting isotope. As shown by the results depicted in Figure 2, psoasand soleusmyofibrillar preparations demonstrated differences in the time course of removal of 4JCa similar to those obtained with the skinned fibers (Fig. 1). We conclude from the results shown in Figure 2 that the relatively slow exchange of Ca2+ from sites on soleusfibers reflects properties of myofilament proteins. Moreover, thesesitesare not presentin fast, white striated fibers. We next present results of studies, which provide evidence that the site(s) of the slow
TABLE I. Kinetic parametersof 45Cawashoutfrom skinnedfibers Muscle c;‘6, Cardiac Soleus Psoas
67.2 + 2.2 44.6 + 1.9
100.0* 1.04
(&
$) 0.032 + 0.003 0.027 f 0.003
32.8 i 1.6 55.3 * 1.5
0.031+ 0.001
The data in Figure 1 were fitted with the equation F = Ale-“’ and soleus) or F = Ae-“’ (psoas). See text for more details.
+ Aze-“’
(cardiac
h-z (SC’, 0.0013 * 0.0001 0.0013 + 0.0001
1120
B.-S. Removal
Pan
et
al.
of 45Ca-Myoflbrtls
20
40 Time
60
(mm)
FIGURE 2. Time course of removal of bound 45 Ca from two preparations of cardiac myofibrils (0 0) and from psoas myofibrils (A) in the presence of 13 mM EGTA. The data from cardiac myofibrils were fitted (broken line) to two exponentials (see legend to Table I) giving A, = 61.5%; K, = 0.025/s; A, = 38.5% K2 = 0.001/s.
exchange is CTNC, and not other site(s) such asthose on myosin light chain II and/or actin, which are known to bind Ca2+ (Bagshaw and Reed, 1977; Holroyde et al., 1980; Korn, 1982). CTNC would be expected to be the major site of binding under the conditions in which equilibrium Ca 2+ binding was carried out. This was verified by autoradiography of the preparations electroblotted onto nitrocellulose and incubated in 45Ca containing binding solution. In agreement with our previous results with skinned cardiac muscle (Pan and Solaro, 1987), the autoradiograms of soleus fibers showed that CTNC was the only site of 45Ca binding. This indicates that the slow Ca2+ sitesare on myofilament-CTNC, but to further assessthis possibility, we studied skinned psoasfibers in which the native STNC was replaced with CTNC. A bundle of skinned psoasfibers was cut transversely into two segmentsof equal length; TNC was extracted from one segment as decribed by Brandt et al., (1984) except that during extraction the fiber was stretched to 3.5 pm, a manipulation that greatly increased the amount of STNC extracted. Skeletal or cardiac TNC was reintroduced into the fibers by bathing the bundle for 30 min in a relaxing solution containing 1 mg/ml purified CTNC or STNC asdescribed by Brandt et al., (1984). Both segmentswere then exposed to binding
solution containing 45Ca for 60 min, and the time coursesof elution of bound Ca2+ measured (Fig. 3). As shown in Figure 3, removal and exchange of TNC was verified by analysis of the skinned fibers by alkaline urea polyacrylamide electrophoresis and silver staining (Oakley et al., 1980; Blanchard et al., 1984). The elution of 45Ca from psoasfibers with their native STNC replaced with CTNC had a slow time course similar to that characteristic of intact cardiac and soleus skinned fibers and myofibrils. Extracted psoas fibers reconstituted with its native STNC showed properties essentially the sameas those of the untreated preparations. An important question is whether the slowly exchanging sites on CTNC are the high affinity Ca2+-Mg2 + sites.To test this we did two types of experiments. In the first (Fig. 4), we measuredthe kinetics of the Ca’+-removal from preparations equilibrated with 45Ca at pCa 8 and pCa 7 in the presence of 2 mM Mg 2+. We (Pan and Solaro, 1987) have shown previously that cardiac myofilaments contain two classes of Ca2+-bindin sites which are equivalent to the Ca2+--Mg H+ and Ca2+-specific sites of CTNC (Robertson et al., 1982); under conditions essentially identical with those used here, their densities (denoted by n) and Ca’+-affinity constants (denoted by K) were Ici = 2.6 x 107/~ and
Slow
Ca’
+-sites
on Cardiac
CTnC-Substituted
psoas
TNC fibers
m-wa
1121
B
wTNC c
D
IO
20 Time
(mln)
FIGURE 3. The time course of removal ofa5Ca from psoas muscle fibers before ( l 0) and after ( W A j substitution of native STNC with cardiac or skeletal TNC. Inset. Alkaline-Urea PAGE of the skinned fibers. Lane A pure TNC; B, control fibers; C, TNC extracted fibers; D, fibers with CTNC substituted for STNC. The solid line represents the least square fit of the data to an exponential equation (See legend to Table 1) giving At = 5596; Kt = 0.074/s; ‘4s = 45”,,; K2 = 0.0036/s. See METHODS and text for details.
K-2 = 9.9 x 105/M with fz.1= 1.22 and n2 = 0.61 nmol/mg protein. Thus it is expetted that 98% of the total bound Ca’+ is on the high affinity Ca’+-Mg2+ sites at pCa 8 and 95% at pCa 7. As shown in Figure 4, under conditions in which Ca2+ is bound exclusively to the higher affinity Ca’+-Mg2+
Cardiac
sites, the slow exchange of Ca2+ is evident. A second type of experiment that we did was to equilibrate the preparations for 60 min in 40Ca (“cold” Ca2+) at pCa 7, and, after a brief rinse in Ca2+-free buffer, expose them to 45Ca for 10 min at pCa 5.3. The idea here was to fill the slowly exchanging sites with cold
DOG Myoflbrlls
0 pCa 8 0 pea 7
Time
FIGURE was bound
(min)
4. Time courses of elution of ‘sCa from the Ca*+-Mg*+ sates of myofibrillar preparations. initially to the myofibrils at pCa 7 (0) and pCa 8 (0) and then removed in the presence of 13 rnM EGTA.
Ca*;
B.-S. Pan et al.
1122 40Ca
Pre-Incubation-Soleus
fibers
(A)
00
v prelncubated
i
Time
45Ca
fibers
(mini
Pre-lncubatlon-Cordlac
myoflbrlls
100 preincubated myoflbrlis 0 control myoflbrlls l
80 26
: 3 z a
60
m 40
60 Time
80 (min)
FIGURE 5. Time courses ofelution of4sCa from A. Skinned soleus fibers and B. cardiac myofibrils briefly exposed to 4sCa after pm-incubation in 40Ca to fill the slow sites with “cold” Ca*‘. A. (7) normalized data from a soleus fiber in the binding solution bundle (0.15 mm) in diameter, which was preincubated with “cold” Ca2+ prior to incubation 45Ca (~3) normalized data form a control fiber bundle (0.15 mm in diameter) pre-incubated in 45Ca in containing place ofcold Ca ‘+ . B experiments as those in A., but done with cardiac myofibrillar preparations. (0) preincubated with cold Gas+ and (0) control. See methods and text for details.
Cazf and then fill the fast siteswith 45Ca. As shown in Figure 5(A), this manipulation indeed resulted in a significant reduction of the relative size of the slow component as compared to a control fiber incubated similarly but in the presenceof 45Ca. The data from the fibers pre-incubated with 40Ca gave A1 = 90.1(%), /cl = 2.10 x IO-‘/s and Az = 9.90(%), ks = 19.2 x 10W4/s;control fibers
gave A1 = 72.7(%), ki = 1.84 x IO-‘/s and As = 27.3(%), k2 = 8.45 x 10m4/s.Figure 5(B) shows that essentially the same kinetics were obtained when we applied the experimental approach described for results shown in Figure 5(A) to cardiac myofibrils. In this case [Fig. 5(B)] the myofibrils pre-incubated with 40Ca gave Ai = 89(%), kl = 3.15 x 10-‘/s and A2 = 10.6(%), k2 = 0.10 x
Slow
Ca’
+-sites
1O-4/s; control myofibrils gave Ai = 63(%), ki = 0.95 x 10w2/s and A2 = 36.2(:/,), kz = 2.3 x 1O-4/s.
Discussion
Our main finding is that CTNC in skinned fibers and myofibrils of cardiac and slow/red muscle has what appear to be unique metal binding sites that bind Ca’+ in the physiological range of free Ca2+ and exchange these Ca2+ ions much more slowly than previously thought. These sites were revealed in experiments in which we followed the time course of elution of 45Ca from skinned fibers and myofibrils. In the case of psoas fibers, the rate constant for elution of 45Ca was 3.15 x IO-‘/s, which gives a half-time of 20-30 s, a value that fits with estimates of diffusional delays and Ca2+ exchange with STNC. We picture the elution of 45Ca to consist of diffusion of the EGTA into the preparation, dissociation of 45Ca from the binding sites, binding of 4sCa to EGTA, and diffusion of 45CaEGTA out of the preparation. From values of ionic mobilities in muscle cells (Kushmerick and Podolsky, 1069), we estimate that diffusion of EGTA into and CaEGTA out of the preparation occurs with a diffusion coefficient (D) of about 3 x low6 rm2/s as estimated on the basis of molecular weight of EGTA and CaEGTA. The time required for the EGTA concentration in the center of a skinned muscle bundle to reach that of the elution solution would be approximately equal to r2/D, where r = radius of the preparation (Hill, 1928). For a fiber bundle with a radius of 100 pm, the diffusion equilibrium of EGTA from outside the fiber to its center should be reached in about 30 s. About the same time would be required for 45CaEGTA to diffuse out. Therefore, overall diffusion of EGTA and 45CaEGTA should take about a minute. In fact this accounts nearly entirely for the rate of 45Ca elution from the psoas fibers. Ca2 + exchange with STNC occurs with a rate constant of about 20/s for the low affinity, regulatory sites and 0.6/s for the high affinity sites (Robertson et al., 1982). Considering that the reaction of excess EGTA with 45Ca is essentially instantaneous and irreversible, the half-time for Ca2+-exchange would be about a second.
on Cardiac
TNC
1123
Based on the analysis given above, the total time between the moment the preparation comes into contact with elution solution and essentially complete elimination of 45Ca from the skinned fiber preparation should be one to two minutes. The time course of elution of 45Ca from fast skeletal skinned fibers and myofibrils fits reasonably well into this time frame. On the other hand, the slow component of 45Ca exchange from soleus and cardiac fibers is not explained by the above predictions. The rate constant for Ca2+ exchange of the high affinity Ca2+-Mg2+ sites of CTNC has been estimated to be about 0.3/s as computed from the equilibrium Ca2+ binding properties of CTNC and an assumed diffusional “on” rate constant (Robertson P/ al., 1982). This is much faster than the rate constant for exchange determined from the results presented here. Yet it seems clear that the slow component of exchange is indeed a property of Ca’+-binding to CTNC and not due to a slow diffusional process or Ca2+-binding to other myofilament proteins such as actin or myosin, which are known to possess very slowlv exchanging divalent metal binding sites iBagshaw and Reed, 1977; Korn 19823. The strongest evidence for this conclusion comes from results of our studies with cardiac and soleus myofibrils and psoas skinned fibers with CTNC substituted for STNC. Results presented here are important with regard to experiments aimed at direct measurement of CTNC Ca2+-binding in myofilaments. For example, an important current issue is whether or not CTNC Ca’+-binding depends on muscle force or length (Allen and Kentish, 1985; Hoffman and Fuchs, 19873. An obvious problem that the existence of the sIow Ca2+ sites raises is that very long elution times are required to remove 45Ca, and this reduces the possibility of making repeated measureon the same functionment of Ca 2f-binding ally stable preparation, say at different sarcomere lengths or with and without pharmacological agents. As outlined above one approach we’ve considered would be to fill the slow sites with 40Ca and to adjust the time of exposure to 45Ca so that the faster exchanging sites are the main sites of binding of 45C:a. What is the molecular location and functional significance of the slowly exchanging sites of CTNC? Evidence presented herr and
1124
B.-S. Pan et al.
previous work strongly indicate the identity of the slowly exchanging sites with the Ca’+-Mg’+ sites of CTNC. Moreover, it seemsreasonable to think these sitesserve as an important feature of structure of the CTNC-thin filament complex. We think that this difference in the Ca’ +-Mg2 + sites of CTNC may be related to the relative difficulty of removal of CTNC from skinned heart fibers by soaks in buffered EDTA compared with removal of STNC from skinned fibers of fast skeletal muscle (Fig. 3). It has been our experience that removal of CTNC required much longer incubation times in which there is loss not only of CTNC but also of myosin light chains, which also contain a slowly exchang-
ing Ca2+-site (Bagshaw and Reed, 1977). In any case, it is clear that the exchange properties of these slow sites must be taken into account in experiments on skinned fiber preparations, which aim to make direct measurements of the Ca’+-binding properties of CTNC. The kinetics of thesesitesmust also be taken into account in terms of schemesdescribing the Ca2+ flows in muscle cells containing CTNC. Acknowledgements
This work was supported in part by National Institutes of Health Grants HL 22231 and HL 22619 (IIIB).
References ALLEN DG, KENTISH JC (1985) The cellular basis of the length-tension relation in cardiac muscle. J Mel Cell Cardiol 17: 82 l-840. BAGSHAW CR, REED GH (1977) The significance of the slow dissociation of divalent metal ions from myosin regulatory light chains. FEBS Lett 81: 386-390. BLANCHARD EM, PAN B-S, SOLARO RJ (1984) The effect of acidic pH on the ATPase activity and troponin Ca*+ binding of rabbit skeletal myofilaments. J Biol Chem 259: 3181-3186. BRANDT PW, DIAMOND MS, SCHACHAT FH (1984) The thin filament of vertebrate skeletal muscle co-operatively activates as a unit. J Mol Biol 180: 379384. HILL AV (1928) The diffusion of oxygen and lactic acid through tissues. Proc Roy Sot B 1041 93-96. HOFFMAN PA, FUCHS F (1987) Effect of length and cross-bridge attachment on Ca *’ binding to cardiac troponin C. Am J Physio1253: C90-C96. HOLROYDE MJ, ROBERTSON SP, JOHNSON JD, SOLARO RJ, POTTER JD (1980) The Ca*+ and Mg*+ binding sites on cardiac troponin and their role in the regulation of myofibrillar adenosine triphosphatase. J Biol Chem 255: 11688811672. KORN ED (1982) Actin polymerization and its regulation by proteins from non-muscle cells. Physiol Rev 62: 672-737. KCSHMERICK MJ, PODOLSKI RJ (1969) I onic mobility in muscle cells. Science 166: 1297-l 298. MILLER DJ, ELDER HY, SMITH GL (1985) Ultrastructural and X-ray microanalytical studies of EGTA- and detergenttreated heart muscle. J Muscle Res Cell Motility 6: 525-540. OAKLEY BR, KIRSCH DR, MORRIS NR (1980) A simplified ultrasensitive silver stain for detecting proteins in polyacrylamide gels. Anal Biochem 105: 361-363. PACANI ED, SOLARO RJ (1984) Methods for measuring functional properties of sarcoplasmic reticulum and myofibrils in small samples of myocardium. In: Methods in Pharmacology, Vol 5, A. Scwartz (Ed.), New York, Plenum, pp 49-61. PAN B-S, SOLARO RJ (1987) Calcium-binding properties of troponin C in detergent-skinned heart muscle fibers. J Biol Chem 262: 7339-7349. POTTER JD, GERCELY J (1975) The calcium and magnesium binding sites on troponin and their role in the regulation of myofibrillar adenosine triphosphatase. J Biol Chem 250: 46294633. ROBERTSON SP, JOHNSON JD, POTTER JD (1981) The time course of Ca*+ exchange with calmodulin, troponin, parvalbumin, and myosin in response to transient increases in Ca’+. Biophys J 34: 559-569. ROBERTSON SP, JOHNSON JD, HOLROYDE MJ, KRANIAS EG, POTTER JD, SOLARO RJ (1982) The effect of troponin I phosphorylation on the Ca *+-binding properties of the Ca’+ -regulatory site of bovine cardiac troponin. J Biol Chem 257: 260-263. SOLARO RJ, PANG DJ, BRICGS FN (197 1) Purification of cardiac myofibrils with Triton X-100. Biochim Biophys Acta 245: 259-262. VAN EERD JP, TAKAHASHI K (1976) Determination of the complete amino acid sequence of bovine cardiac troponin C. Biochemistry 15: 1171-1175. WILKINSON JM (1980) Troponin C from rabbit slow skeletal muscle and cardiac muscle is the product of a single gene. Eur J Biochem 103: 179-188.
Slowly
Exchanging
Bo-Sheng
Pan, Kimberly
Calcium Binding Sites Unique Muscle Troponin C A. Palmiter,
Maria
Plonczynski,
to Cardiac/Slow
and R. John
Solar0
Department of Physiology and Biophysics, Universityof Cincinnati, College of Medicine, Cincinnati, OH 45267-0576, i!SA and Department of Physiology and Biophysics, CTniversity of Illinoi.r, Co&r qf Medicine at Chicago, Chicago, IL 60680, LTSA i Received 13 November 1989, accepted in revised form 20 April
I990)
B.-S. PAN, K. A. PALMITER, M. PLONCZYNSKI AND R. J, SOLARO. Slowly Exchanging Calcium Binditlg Sites Unique to Cardiac/Slow Muscle Troponin C. Journal of Molecular and Cellular Cardiologv (1990) 22, I 117-I 124. Evidence is presented for the existence of slowly exchanging Cal+-binding sites in troponin C CTNC of cardiac and slow twitch skeletal muscles. These sites were revealed in the course of experiments aimed at qkinnrd measuring the Ca ‘+-binding properties of TNC in the myofilament lattice. 4sCa bound to chemically muscle fibers or myofibrils of cardiac and soleus muscles was &ted by EGTA in a two-exponential timecoursr with a slow phase of a rate constant of about 2 x 1F4/s. The slow phase was not found in skinned librr or myofibrils of psoas. a fast skeletal muscle. However, skinned psoas fibers in which the native TNC was rcplaccd of soleus and cardiac preparations, indicating that the by C’TNC exhibited the slow 45Ca elution characteristic slowly-exchanging sites are located in CTNC. These sites are tentatively identified as the Ca’+--Mg” sltrs 01 CTNC on the basis that the slow phase was observed under conditions known to restrict Ca” binding tv the Ca’+ -Mg2 + sites. KEY WORDS:
Calcium;
Magnesium;
Calcium-binding
Introduction of the contractile activity of cardiac or slow, red myofilaments is essentially the same as that of fast, white striated muscle, but there are some interesting differences. In each of these types of muscle, the Ca2+ receptor is troponin C (TNC), but the variant of cardiac and slow, red muscle (CTNC) has three Ca 2+ binding dom ains instead of the four that are present in fast striated muscle STNC (Van Eerd and Takahashi, 1976; Wilkinson, 1980). Studies with isolated TNC or TNC in the TN complex showed that there are two sites with relatively high affinity for Ca2+ and Mg2+ in both STNC and CTNC, but whereas STNC possesses two sites of lower affinity for Ca2+ (Potter and Gergely, 1975), CTNC has but one low affinitv Ca2+-binding site (Holroyde et al., 1980). I; is apparent that only the lower affinity sites can exchange Cz2+ in the time frame of Ca2+
activation
proteins;
Troponin;
Heart:
+ 08 $03.00/O
Fast music.
the contraction-relaxation cycle of cardiac and fast striated muscle, and thus these sites appear to be the physiological “regulatory” sites (Robertson et al., 198 1; 1982). The slowly exchanging Ca2+-?vlg2+ sites of STNC have been characterized as “structural” sites. Although there have been no studies of the kinetics of Ca2+ exchange with tht Ca2+-MgZ t sites of CTNC, it has been generally assumed that the kinetics of thesr metal binding sites of STNC and CTNC are similar (Robertson et al., 1981). Yet, in the course of experiments in which we measured the Ca’+binding properties of skinned cardiac muscle preparations (Pan and Solaro, 1987!, we discovered that elution of 45Ca from cardiac fibers was much slower than expected if the kinetics of Ca2+ exchange of CTNC and STNC were similar. In work presented here. we show evidence that these slowly exchanging sites of skinned fibers are indeed on ‘I’NC
Please address all correspondence to: R. John Solaro, Department of Physiology :M.iC 901 J, College of Medicine at Chicago, Chicago, IL 60680. US.4. 0022-2828/90/101117
Slow muscle:
and Biophysics,
I‘ni~crtit~
,c‘r 1990 Academic
oi Illinois
Press Limited
1118
B.-S. Pan et al.
present in the myofilaments and are unique to the CTNC variant present in heart and red, slow myofilaments.
0.5 ml NaOH at 95°C and assayedfor protein using the Lowry procedure as modified by Pagani and Solar0 (1984). The elution solutions and the fiber digest were counted for 45Ca and 3H in a liquid scintillation specMethods trometer. To construct the time course of the Preparations elution, the 45Ca activity remaining in the Thin bundles of muscle fibers (0.5-1.0 mm in muscle preparation at a specified time point diameter) were dissectedfrom the soleusand was obtained by summing up the 45Ca activpsoasmusclesof rabbits and from the right ities eluted from the preparation after that ventricles of mongrel dog hearts as previously time point and the 45Ca activity in the fiber described (Blanchard et al., 1984; Pan and digest. Correction for unbound 45Ca was Solaro, 1987). The fiber bundles were chemi- made using the 3H activity eluted from the cally skinned by extraction at 4°C for 12 h in a preparation. relaxing solution containing 1o/o Triton The time course of elution of 45Ca from X-100, 10 mM EGTA, 5.5 mM ATP, 12 mM myofibrillar preparations was determined by creatine phosphate, 8.2 mM MgClz, 30 mM a filtration technique. The myofibrils (0.3 KCI, 60 mM imidazole, 60 PM phenylmethymg/ml) were preincubated for 60 min in a sulfonate fluoride, pH 7.0. The skinned fibers “binding” solution containing 60 mM KCl, 20 were stored at -20°C in a detergent-free mM imidazole (pH 7.0), 2 mM Mg2+, 20 PM relaxing solution with 50% glycerol and fur- total Ca2+, 3 @i/ml 45Ca, and appropriate ther dissectedto thinner bundles immediately quantity of EGTA to give a desired free before use. We (Pan and Solaro, 1987) and [Ca2+]. At zero time, the mixture was made others (Miller et al., 1985) have previously 13 mM in EGTA. To determine 45Ca bound to shown that this and similar skinning proce- myofibrils at various times, aliquots of the dures completely removed subcellular mem- myofibrils were filtered through Millipore filbranes. This was verified by transmission ters (HA 0.45 pm) and the filters were washed electron microscopy of the preparations, and with 50 mM KCl, dried and counted. enzyme marker assays, which showed no evidence for the presence of surface memResults brane, sarcoplasmic reticulum and mitochonWe first present results of experiments comdria (Pan and Solaro, 1987). Myofibrillar preparations were isolated as described by paring the time course of elution of 45Ca from Solar0 et al., 1971. skinned fibers of psoas, soleus and heart muscle that were pre-equilibrated with binding solution containing 45Ca and 3H-glucose Kinetics of e&ion of 45Ca from skinned jbers and (Fig. 1). The washout of 3H-glucose activity myojbrils in these preparations was essentially comA thin bundle (approximately 0.1 mm in pleted within 2 min (data not shown), and diameter) of skinned fibers was incubated for about the same time was required for elution 90 min in a “binding solution” containing 2 of 45Ca from psoas fibers. However, much mM Mg2+, 110 mM KCl, 60 mM imidazole longer times were required for complete re(pH 7.0), 1 mM D-glucose, 45Ca (6 &i/ ml) moval of fiber bound 45Ca in the caseof the and 3H-glucose (6 @i/ml) at 23°C. As deter- soleusand cardiac muscle. The relatively slow mined by atomic absorption spectroscopy, the time courseof elution of 45Ca from the soleus, free Ca2+ of the binding solution was 5-6 PM. is similar to what we have previously found in The preparation was gently blotted to remove skinned fibers from heart muscle (Pan and adsorbed liquid and then incubated sequenti- Solaro, 1987), and was evident whether the ally for specified periods in a seriesof vials binding was carried out in the presence or each with one ml of an elution solution of 10 absence of MgATP, or with the fibers unmM EGTA, 50 mM KCl, 2 mM MgC12, 1 mM loaded or held isometric. Moreover, in both glucose and 20 mM imidazole, pH 7.0. At the the cardiac and soleusfibers, further increases end, the muscle preparation was digested in in the EGTA concentration in the elution
Slow
Ca’
+-sites
Removal
on Cardiac
of
TNC
45Ca-Skrnned
20
1119
f 1bers
30 Time
(mln)
FIGURE 1. Time course of removal of bound Ca*+ from two prep arations of detergent treated fibers from (a), psoas (W) and cardiac (0) muscle. Initially Ca’+ was bound to the fibers at pCa 5.3 and then removed presence of 10 mM EGTA as detailed in the METHODS. See Table I for a summary of kinetic parameters.
solution or replacing EGTA with CaClz of equal or higher concentrations did not accelerate the removal of bound Ca2 + . As shown in Figure 1, in the caseof the soleusand cardiac preparations, the data exhibited a biphasic feature, and could be fitted with an equation of the form, F = AIe-k11 + A2e -k2t, where F denotes the fraction of 45Ca remaining bound (as percent of bound 45Ca at time zero), kr and k2 are rate constants, and A, and A,, the fractional amount of the two exponential components. The data from psoas muscle were fitted to a single exponential. Table 1 lists the rate constants obtained in the curve fitting analysis. For both cardiac and soleus fibers, the rate constant of the slow component was about 30-fold lower than the rate constant of the fast component, which was similar in magnitude to the single rate constant of the psoasfibers.
soleus in the
We also tested if the differences in the time course of exchange of bound Ca2+ are also a property of myofibrillar preparations. The detergent-extracted myofibrillar preparations are free of membranes (Solar0 et al., 1971; Pan and Solaro, 1987) and about 1 pm in diameter. Thus diffusional delays are negligible when eluting isotope. As shown by the results depicted in Figure 2, psoasand soleusmyofibrillar preparations demonstrated differences in the time course of removal of 4JCa similar to those obtained with the skinned fibers (Fig. 1). We conclude from the results shown in Figure 2 that the relatively slow exchange of Ca2+ from sites on soleusfibers reflects properties of myofilament proteins. Moreover, thesesitesare not presentin fast, white striated fibers. We next present results of studies, which provide evidence that the site(s) of the slow
TABLE I. Kinetic parametersof 45Cawashoutfrom skinnedfibers Muscle c;‘6, Cardiac Soleus Psoas
67.2 + 2.2 44.6 + 1.9
100.0* 1.04
(&
$) 0.032 + 0.003 0.027 f 0.003
32.8 i 1.6 55.3 * 1.5
0.031+ 0.001
The data in Figure 1 were fitted with the equation F = Ale-“’ and soleus) or F = Ae-“’ (psoas). See text for more details.
+ Aze-“’
(cardiac
h-z (SC’, 0.0013 * 0.0001 0.0013 + 0.0001
1120
B.-S. Removal
Pan
et
al.
of 45Ca-Myoflbrtls
20
40 Time
60
(mm)
FIGURE 2. Time course of removal of bound 45 Ca from two preparations of cardiac myofibrils (0 0) and from psoas myofibrils (A) in the presence of 13 mM EGTA. The data from cardiac myofibrils were fitted (broken line) to two exponentials (see legend to Table I) giving A, = 61.5%; K, = 0.025/s; A, = 38.5% K2 = 0.001/s.
exchange is CTNC, and not other site(s) such asthose on myosin light chain II and/or actin, which are known to bind Ca2+ (Bagshaw and Reed, 1977; Holroyde et al., 1980; Korn, 1982). CTNC would be expected to be the major site of binding under the conditions in which equilibrium Ca 2+ binding was carried out. This was verified by autoradiography of the preparations electroblotted onto nitrocellulose and incubated in 45Ca containing binding solution. In agreement with our previous results with skinned cardiac muscle (Pan and Solaro, 1987), the autoradiograms of soleus fibers showed that CTNC was the only site of 45Ca binding. This indicates that the slow Ca2+ sitesare on myofilament-CTNC, but to further assessthis possibility, we studied skinned psoasfibers in which the native STNC was replaced with CTNC. A bundle of skinned psoasfibers was cut transversely into two segmentsof equal length; TNC was extracted from one segment as decribed by Brandt et al., (1984) except that during extraction the fiber was stretched to 3.5 pm, a manipulation that greatly increased the amount of STNC extracted. Skeletal or cardiac TNC was reintroduced into the fibers by bathing the bundle for 30 min in a relaxing solution containing 1 mg/ml purified CTNC or STNC asdescribed by Brandt et al., (1984). Both segmentswere then exposed to binding
solution containing 45Ca for 60 min, and the time coursesof elution of bound Ca2+ measured (Fig. 3). As shown in Figure 3, removal and exchange of TNC was verified by analysis of the skinned fibers by alkaline urea polyacrylamide electrophoresis and silver staining (Oakley et al., 1980; Blanchard et al., 1984). The elution of 45Ca from psoasfibers with their native STNC replaced with CTNC had a slow time course similar to that characteristic of intact cardiac and soleus skinned fibers and myofibrils. Extracted psoas fibers reconstituted with its native STNC showed properties essentially the sameas those of the untreated preparations. An important question is whether the slowly exchanging sites on CTNC are the high affinity Ca2+-Mg2 + sites.To test this we did two types of experiments. In the first (Fig. 4), we measuredthe kinetics of the Ca’+-removal from preparations equilibrated with 45Ca at pCa 8 and pCa 7 in the presence of 2 mM Mg 2+. We (Pan and Solaro, 1987) have shown previously that cardiac myofilaments contain two classes of Ca2+-bindin sites which are equivalent to the Ca2+--Mg H+ and Ca2+-specific sites of CTNC (Robertson et al., 1982); under conditions essentially identical with those used here, their densities (denoted by n) and Ca’+-affinity constants (denoted by K) were Ici = 2.6 x 107/~ and
Slow
Ca’
+-sites
on Cardiac
CTnC-Substituted
psoas
TNC fibers
m-wa
1121
B
wTNC c
D
IO
20 Time
(mln)
FIGURE 3. The time course of removal ofa5Ca from psoas muscle fibers before ( l 0) and after ( W A j substitution of native STNC with cardiac or skeletal TNC. Inset. Alkaline-Urea PAGE of the skinned fibers. Lane A pure TNC; B, control fibers; C, TNC extracted fibers; D, fibers with CTNC substituted for STNC. The solid line represents the least square fit of the data to an exponential equation (See legend to Table 1) giving At = 5596; Kt = 0.074/s; ‘4s = 45”,,; K2 = 0.0036/s. See METHODS and text for details.
K-2 = 9.9 x 105/M with fz.1= 1.22 and n2 = 0.61 nmol/mg protein. Thus it is expetted that 98% of the total bound Ca’+ is on the high affinity Ca’+-Mg2+ sites at pCa 8 and 95% at pCa 7. As shown in Figure 4, under conditions in which Ca2+ is bound exclusively to the higher affinity Ca’+-Mg2+
Cardiac
sites, the slow exchange of Ca2+ is evident. A second type of experiment that we did was to equilibrate the preparations for 60 min in 40Ca (“cold” Ca2+) at pCa 7, and, after a brief rinse in Ca2+-free buffer, expose them to 45Ca for 10 min at pCa 5.3. The idea here was to fill the slowly exchanging sites with cold
DOG Myoflbrlls
0 pCa 8 0 pea 7
Time
FIGURE was bound
(min)
4. Time courses of elution of ‘sCa from the Ca*+-Mg*+ sates of myofibrillar preparations. initially to the myofibrils at pCa 7 (0) and pCa 8 (0) and then removed in the presence of 13 rnM EGTA.
Ca*;
B.-S. Pan et al.
1122 40Ca
Pre-Incubation-Soleus
fibers
(A)
00
v prelncubated
i
Time
45Ca
fibers
(mini
Pre-lncubatlon-Cordlac
myoflbrlls
100 preincubated myoflbrlis 0 control myoflbrlls l
80 26
: 3 z a
60
m 40
60 Time
80 (min)
FIGURE 5. Time courses ofelution of4sCa from A. Skinned soleus fibers and B. cardiac myofibrils briefly exposed to 4sCa after pm-incubation in 40Ca to fill the slow sites with “cold” Ca*‘. A. (7) normalized data from a soleus fiber in the binding solution bundle (0.15 mm) in diameter, which was preincubated with “cold” Ca2+ prior to incubation 45Ca (~3) normalized data form a control fiber bundle (0.15 mm in diameter) pre-incubated in 45Ca in containing place ofcold Ca ‘+ . B experiments as those in A., but done with cardiac myofibrillar preparations. (0) preincubated with cold Gas+ and (0) control. See methods and text for details.
Cazf and then fill the fast siteswith 45Ca. As shown in Figure 5(A), this manipulation indeed resulted in a significant reduction of the relative size of the slow component as compared to a control fiber incubated similarly but in the presenceof 45Ca. The data from the fibers pre-incubated with 40Ca gave A1 = 90.1(%), /cl = 2.10 x IO-‘/s and Az = 9.90(%), ks = 19.2 x 10W4/s;control fibers
gave A1 = 72.7(%), ki = 1.84 x IO-‘/s and As = 27.3(%), k2 = 8.45 x 10m4/s.Figure 5(B) shows that essentially the same kinetics were obtained when we applied the experimental approach described for results shown in Figure 5(A) to cardiac myofibrils. In this case [Fig. 5(B)] the myofibrils pre-incubated with 40Ca gave Ai = 89(%), kl = 3.15 x 10-‘/s and A2 = 10.6(%), k2 = 0.10 x
Slow
Ca’
+-sites
1O-4/s; control myofibrils gave Ai = 63(%), ki = 0.95 x 10w2/s and A2 = 36.2(:/,), kz = 2.3 x 1O-4/s.
Discussion
Our main finding is that CTNC in skinned fibers and myofibrils of cardiac and slow/red muscle has what appear to be unique metal binding sites that bind Ca’+ in the physiological range of free Ca2+ and exchange these Ca2+ ions much more slowly than previously thought. These sites were revealed in experiments in which we followed the time course of elution of 45Ca from skinned fibers and myofibrils. In the case of psoas fibers, the rate constant for elution of 45Ca was 3.15 x IO-‘/s, which gives a half-time of 20-30 s, a value that fits with estimates of diffusional delays and Ca2+ exchange with STNC. We picture the elution of 45Ca to consist of diffusion of the EGTA into the preparation, dissociation of 45Ca from the binding sites, binding of 4sCa to EGTA, and diffusion of 45CaEGTA out of the preparation. From values of ionic mobilities in muscle cells (Kushmerick and Podolsky, 1069), we estimate that diffusion of EGTA into and CaEGTA out of the preparation occurs with a diffusion coefficient (D) of about 3 x low6 rm2/s as estimated on the basis of molecular weight of EGTA and CaEGTA. The time required for the EGTA concentration in the center of a skinned muscle bundle to reach that of the elution solution would be approximately equal to r2/D, where r = radius of the preparation (Hill, 1928). For a fiber bundle with a radius of 100 pm, the diffusion equilibrium of EGTA from outside the fiber to its center should be reached in about 30 s. About the same time would be required for 45CaEGTA to diffuse out. Therefore, overall diffusion of EGTA and 45CaEGTA should take about a minute. In fact this accounts nearly entirely for the rate of 45Ca elution from the psoas fibers. Ca2 + exchange with STNC occurs with a rate constant of about 20/s for the low affinity, regulatory sites and 0.6/s for the high affinity sites (Robertson et al., 1982). Considering that the reaction of excess EGTA with 45Ca is essentially instantaneous and irreversible, the half-time for Ca2+-exchange would be about a second.
on Cardiac
TNC
1123
Based on the analysis given above, the total time between the moment the preparation comes into contact with elution solution and essentially complete elimination of 45Ca from the skinned fiber preparation should be one to two minutes. The time course of elution of 45Ca from fast skeletal skinned fibers and myofibrils fits reasonably well into this time frame. On the other hand, the slow component of 45Ca exchange from soleus and cardiac fibers is not explained by the above predictions. The rate constant for Ca2+ exchange of the high affinity Ca2+-Mg2+ sites of CTNC has been estimated to be about 0.3/s as computed from the equilibrium Ca2+ binding properties of CTNC and an assumed diffusional “on” rate constant (Robertson P/ al., 1982). This is much faster than the rate constant for exchange determined from the results presented here. Yet it seems clear that the slow component of exchange is indeed a property of Ca’+-binding to CTNC and not due to a slow diffusional process or Ca2+-binding to other myofilament proteins such as actin or myosin, which are known to possess very slowlv exchanging divalent metal binding sites iBagshaw and Reed, 1977; Korn 19823. The strongest evidence for this conclusion comes from results of our studies with cardiac and soleus myofibrils and psoas skinned fibers with CTNC substituted for STNC. Results presented here are important with regard to experiments aimed at direct measurement of CTNC Ca2+-binding in myofilaments. For example, an important current issue is whether or not CTNC Ca’+-binding depends on muscle force or length (Allen and Kentish, 1985; Hoffman and Fuchs, 19873. An obvious problem that the existence of the sIow Ca2+ sites raises is that very long elution times are required to remove 45Ca, and this reduces the possibility of making repeated measureon the same functionment of Ca 2f-binding ally stable preparation, say at different sarcomere lengths or with and without pharmacological agents. As outlined above one approach we’ve considered would be to fill the slow sites with 40Ca and to adjust the time of exposure to 45Ca so that the faster exchanging sites are the main sites of binding of 45C:a. What is the molecular location and functional significance of the slowly exchanging sites of CTNC? Evidence presented herr and
1124
B.-S. Pan et al.
previous work strongly indicate the identity of the slowly exchanging sites with the Ca’+-Mg’+ sites of CTNC. Moreover, it seemsreasonable to think these sitesserve as an important feature of structure of the CTNC-thin filament complex. We think that this difference in the Ca’ +-Mg2 + sites of CTNC may be related to the relative difficulty of removal of CTNC from skinned heart fibers by soaks in buffered EDTA compared with removal of STNC from skinned fibers of fast skeletal muscle (Fig. 3). It has been our experience that removal of CTNC required much longer incubation times in which there is loss not only of CTNC but also of myosin light chains, which also contain a slowly exchang-
ing Ca2+-site (Bagshaw and Reed, 1977). In any case, it is clear that the exchange properties of these slow sites must be taken into account in experiments on skinned fiber preparations, which aim to make direct measurements of the Ca’+-binding properties of CTNC. The kinetics of thesesitesmust also be taken into account in terms of schemesdescribing the Ca2+ flows in muscle cells containing CTNC. Acknowledgements
This work was supported in part by National Institutes of Health Grants HL 22231 and HL 22619 (IIIB).
References ALLEN DG, KENTISH JC (1985) The cellular basis of the length-tension relation in cardiac muscle. J Mel Cell Cardiol 17: 82 l-840. BAGSHAW CR, REED GH (1977) The significance of the slow dissociation of divalent metal ions from myosin regulatory light chains. FEBS Lett 81: 386-390. BLANCHARD EM, PAN B-S, SOLARO RJ (1984) The effect of acidic pH on the ATPase activity and troponin Ca*+ binding of rabbit skeletal myofilaments. J Biol Chem 259: 3181-3186. BRANDT PW, DIAMOND MS, SCHACHAT FH (1984) The thin filament of vertebrate skeletal muscle co-operatively activates as a unit. J Mol Biol 180: 379384. HILL AV (1928) The diffusion of oxygen and lactic acid through tissues. Proc Roy Sot B 1041 93-96. HOFFMAN PA, FUCHS F (1987) Effect of length and cross-bridge attachment on Ca *’ binding to cardiac troponin C. Am J Physio1253: C90-C96. HOLROYDE MJ, ROBERTSON SP, JOHNSON JD, SOLARO RJ, POTTER JD (1980) The Ca*+ and Mg*+ binding sites on cardiac troponin and their role in the regulation of myofibrillar adenosine triphosphatase. J Biol Chem 255: 11688811672. KORN ED (1982) Actin polymerization and its regulation by proteins from non-muscle cells. Physiol Rev 62: 672-737. KCSHMERICK MJ, PODOLSKI RJ (1969) I onic mobility in muscle cells. Science 166: 1297-l 298. MILLER DJ, ELDER HY, SMITH GL (1985) Ultrastructural and X-ray microanalytical studies of EGTA- and detergenttreated heart muscle. J Muscle Res Cell Motility 6: 525-540. OAKLEY BR, KIRSCH DR, MORRIS NR (1980) A simplified ultrasensitive silver stain for detecting proteins in polyacrylamide gels. Anal Biochem 105: 361-363. PACANI ED, SOLARO RJ (1984) Methods for measuring functional properties of sarcoplasmic reticulum and myofibrils in small samples of myocardium. In: Methods in Pharmacology, Vol 5, A. Scwartz (Ed.), New York, Plenum, pp 49-61. PAN B-S, SOLARO RJ (1987) Calcium-binding properties of troponin C in detergent-skinned heart muscle fibers. J Biol Chem 262: 7339-7349. POTTER JD, GERCELY J (1975) The calcium and magnesium binding sites on troponin and their role in the regulation of myofibrillar adenosine triphosphatase. J Biol Chem 250: 46294633. ROBERTSON SP, JOHNSON JD, POTTER JD (1981) The time course of Ca*+ exchange with calmodulin, troponin, parvalbumin, and myosin in response to transient increases in Ca’+. Biophys J 34: 559-569. ROBERTSON SP, JOHNSON JD, HOLROYDE MJ, KRANIAS EG, POTTER JD, SOLARO RJ (1982) The effect of troponin I phosphorylation on the Ca *+-binding properties of the Ca’+ -regulatory site of bovine cardiac troponin. J Biol Chem 257: 260-263. SOLARO RJ, PANG DJ, BRICGS FN (197 1) Purification of cardiac myofibrils with Triton X-100. Biochim Biophys Acta 245: 259-262. VAN EERD JP, TAKAHASHI K (1976) Determination of the complete amino acid sequence of bovine cardiac troponin C. Biochemistry 15: 1171-1175. WILKINSON JM (1980) Troponin C from rabbit slow skeletal muscle and cardiac muscle is the product of a single gene. Eur J Biochem 103: 179-188.
