Ca2+ binding to cardiac troponin C in the myofilament ...

Eur. J. Biochem. 226, 597-602 (1994) 0 FEBS 1994

Ca2+binding to cardiac troponin C in the myofilament lattice and its relation to the myofibrillar ATPase activity Sachio MORIMOTO and Iwao OHTSUKI Department of Pharmacology, Faculty of Medicine, Kyushu University, Japan (Received July IS/September 13, 1994)

-

EJB 94 1052/2

The Ca2+-bindingproperties of troponin C in the intact myofilament lattice and their relation to the activation of ATPase were investigated with isolated porcine cardiac myofibrils. Ca2+binding, which is composed of two classes of binding sites with different affinities (classes 1 and 2), was clearly detected by a novel method for subtracting the large background activity of myofibrillar Ca2+binding. The classes 1 and 2 were equivalent stoichiometrically to the two high-affinity sites (sites I11 and IV) and a single low-affinity site (site 11) of troponin C. In the presence of ATP, positive cooperativity was observed in the Ca2+ binding of class-2 sites and the Hill equation parameters were in excellent agreement with those for the Ca*+-activatedmyofibrillar ATPase activity, which indicated that the activation of ATPase is a linear function of the Ca2+occupancy of site 11. In the absence of ATP, a marked increase in the affinity of only class-2 sites was observed while the cooperativity was lost. These results provide direct evidence that some feedback mechanism exists between myosin crossbridge attachment and the Ca2+binding to site I1 of troponin C, which may thus confer positive cooperativity on the Ca2+activation of myofibrillar ATPase activity.

Contraction in vertebrate striated muscle is regulated by CaZi through the regulatory proteins, troponin and tropomyosin, which are located in the thin filament [l-41. Troponin consists of three subunits termed troponin C , I and T (TN . C, TN . I and TN . T, respectively); TN . C is a Ca2+-binding subunit, TN . I is an inhibitory subunit and TN . T is a tropomyosin-binding subunit. Upon binding Ca2+to TN . C, the inhibitory action of TN . I on the actin-myosin interaction is reduced, and hence the contraction ensues. The biochemical properties of cardiac TN . C have been studied extensively using isolated TN . C [5-lo]. Cardiac TN . C has three Ca2+-bindingsites, a single low-affinity site in the N-terminal domain (site 11) and two high-affinity sites in the C-terminal domain (sites I11 and IV). A number of studies on the static and kinetic properties of Ca2+binding to isolated TN . C, or its complexes with other troponin subunits, indicate that the single low-affinity site, site 11, is responsible for the regulation of muscle contraction [7, 8, 111. However, little is known about the Caz+-binding properties of TN . C integrated into the myofibrillar structure, although the Ca2+-bindingproperties of TN . C are thought to be altered by the complex protein-protein interactions within the myofilament lattice of myofibrils [12-161. Some investigators have tried to measure Ca” binding to the isolated cardiac myofibrils or muscle fibers [14, 17-20], but the

large background levels of total bound Caz+ has prevented them from obtaining precise information on the Ca2+-binding properties of TN . C and their relation to the contractile response. Recently, we developed a method for measuring Ca2+binding to TN . C in myofibrils which has a higher accuracy than previous methods, by subtracting the background Ca2+ binding [16, 21, 221. Using this method, we were able to characterize the Ca2+-bindingproperties of TN . C to the cardiac myofibrils and their relation to the regulation of myofibrillar ATPase activity. The results show that the activation of ATPase is in direct proportion to Ca2+binding to site I1 of TN . C, and also provide evidence that the myosin crossbridge interaction with actin in the presence of ATP confers cooperativity on the Ca2+-bindingto site I1 and, therefore, causes the cooperative activation of myofibrillar ATPase.

MATERIALS AND METHODS

-

Preparation of the myofibrils and TN C Cardiac myofibrils and TN . C were prepared from the left ventricles of porcine heart as described by Solar0 et al. [23] and Tsukui and Ebashi [24], respectively. The prepared myofibrils were suspended in 50% (by vol.) glycerol, stored at -20°C for at least 2 weeks and used for the experiments.

Correspondence to I. Ohtsuki, Department of Pharmacology, Faculty of Medicine, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Extraction of TN C from the myofibrils Fukuoka, Japan 812 and reconstitution with TN C Fax: +81 092 632 2373. Abbreviations. TN . C, troponin C ; TN . I, troponin I; TN . T, Endogenous TN . C in cardiac myofibrils was extracted troponin T; LC2, myosin light chain 2; CDTA, truns-1,2-cyclohexacid anediamine-N,N,N‘,N‘-tetraaceticacid; pCa, -log,,([Caz’ ],&I)by . trans- 1,2-cyclohexanediamine-N,N,N’,Nf-tetraacetic Enzymes. Creatine kinase (EC 2.7.3.2); CAMP-dependent pro- (CDTA) treatment and the CDTA-treated cardiac myofibrils tein kinase and protein kinase C (EC 2.7.1.37). were reconstituted with purified cardiac TN . C as described

-

-

598 Table 1. CaZ+sensitive cardiac myofibrillar ATPase activity and relative contents of TN * C and LC2 in myofibrils before and after CDTA-treatment and after reconstitution with TN C. ATPase activities are expressed as mean 2 SE for three determinations. SDSPAGE of myofibrils was carried out by the method of Laemmli [40]. The gel stained with Coomassie Brilliant Blue R-250 was scanned with a dual-wavelength densitometer (Shimadzu CS9000) and the relative contents of TN . C and LC2 were obtained by normalizing the peak areas of TN . C and LC2 to the area of myosin light-chain 1. Myofibrils

ATPase activity

Relative content

pCa 8.0

TN'C

pCa 4.3

LC2

I h

9

E

2 +

0

2.0 -

A

1.5-

0

0

Untreated CDTA-treated Reconstituted with TN . C

. mg-'

A

a

0

0.5-

a o -

. .. 0

0.0

34.821.3 10.7 2 0.2

0.11 0.01

0.58 0.29

8.7 2 1.6

32.1 i- 2.1

0.12

0.38

previously [21, 251. Briefly, TN . C was extracted three times by incubating cardiac myofibrils for 15 min at 25°C in a solution containing 40mM Tris, pH 8.4, 5 mM CDTA, 15 mM 2-mercaptoethanol, 200 pM phenylmethylsulfonyl fluoride, 1 pg/ml pepstatin A and 0.6 mM NaN,. The CDTAtreated myofibrils were reconstituted with TN . C by incubating them for 15 min under ice-cold conditions in a solution containing an excess amount of purified TN . C, followed by intensively washing out either the unbound or nonspecifically bound TN . C by centrifugation and resuspension.

ATPase assay The myofibrillar ATPase activity was measured at 25 "C in 2 ml solution containing 1 mg/ml myofibrils, 20 mM MopsNaOH, pH7.0, 100mM KCI, 5 m M MgCI,, 2 m M ATP, 0.1 mg/ml creatine kinase, 10 mM creatine phosphate, 0.1 mM dithiothreitol, 10 mM glucose, 0.1 mM EGTA and varied amounts of CaCI,. The reaction was started by the addition of ATP, then stopped after 5 min incubation by the addition of 2 ml ice-cold 20% trichloroacetic acid containing 4% ascorbic acid. Liberated inorganic phosphate was measured by the method of Baginski et al. [26].

8

7

m a *

..

...*** 6

.

O 0

C

4

7.1 ?1.4 8.8 2 1.4

o o

A

1.0-

0

nmol Pi . min-'

0

A

A

v N

::

I

A

5

,I

4

-log ,,([Ca'+] free /M) Fig.1. Ca2+binding to cardiac myofibrils before (A)and after (0)TN C-extraction (CDTA treatment) and to the CDTAtreated myofibrils reconstituted with purified cardiac TN C (0) in the presence of ATP. Caz+ binding was measured n i t h the double-isotope technique using 45Caand ['H]glucose, as demibed in Materials and Methods.

-

-

incubated for 5 min at 25 "C and centrifuged at 3500 rpm for 10 min at 25 "C. After the careful removal of supernatant, the precipitate and 100 pl supernatant were mixed with 100 p1 0.1 M EGTA, pH 7.0, treated with 1.5 rnl Soluene-350 for 1.5 h at 65 "C, mixed with 10 ml Aquasol-2 containing 0.6% acetic acid, then 45Caand ,H were determined using an .\loka LSC-1000.

Calculation of the free Ca2+concentration The free Ca2+ concentration in the solution was ialculated with the aid of a computer using the absolute binding constants for multiple ionic equilibria as described previously [27].

Determination of the protein concentration The protein concentrations of TN . C and myofibrils were determined by the Bio-Rad protein assay method and the biuret method, respectively. The absolute amounts of proteins were determined by an amino acid analysis.

Measurements of Caz+binding to TN * C in myofibrils Caz+binding to myofibrils was measured in the presence or absence of ATP. In either case, the free Mg2+ concentration and the ionic strength were kept at 2.2 mM and 0.17 M, respectively. The amounts of Ca2+bound to TN . C in myofibrils were determined by subtracting the amounts of Ca2+ bound to the CDTA-treated (TN . C-extracted) myofibrils from those bound to the TN . C-reconstituted myofibrils. Ca2+ binding in the presence of ATP was measured at 25°C in 5 ml of a solution which had the same composition as that for the measurement of myofibrillar ATPase activity, except for the presence of 0.3 pCi/ml 45Ca and 0.6 pCi/ml [3H]glucose. The measurement in the absence of ATP was carried out at 25 "C in 5 ml solution containing 1 mg/ml myofibrils, 20 mM Mops/NaOH, pH 7.0, 161 mM KCI, 2.2 mM MgCl,, 0.1 mM dithiothreitol, 10 mM glucose, 0.3 pCi/ml 45Ca, 0.6 pCi/ml ['Hlglucose, 0.1 mM EGTA and varied amounts of CaCI,. The solutions containing myofibrils were

RESULTS The ATPase activity of the CDTA-treated myofibrils showed no Ca" dependence and was inhibited over the entire Ca2+concentration range 10-8-10-4 M, and SDS/PAGE showed that almost all of the TN . C and approximately 50% of the myosin light chain 2 (LC2) in cardiac myofibrilr were selectively extracted by the CDTA treatment, and that the reincorporation of TN . C was nearly stoichiometric (Table 1). Since it was confirmed in a previous study that the removal of approximately 50% of the LC2 has no effect on the Ca2+regulation of the cardiac myofibrillar ATPase [27, 281, all experiments described have been carried out with partially LC2-extracted myofibrillar preparations. Fig. 1 shows the representative Caz+-binding data in the presence of ATP. With an increase in the free Ca'+ concentration, the TN . C-extracted (CDTA-treated) cardiac myofibrils

599 1.0,

h

cn

0.8

1

-

2 0

E

2

+v

Table 2. Ca2+-bindingparameters of TN * C in cardiac myofibrils. The Ca2+-binding capacity ( N ) , apparent Ca2+-binding constant ( K ) and Hill coefficient (n,J are obtained by fitting the data in Figs 2 and 4B to the Eqn (1) with a weighted non-linear leastsquares method. ~

Class 1

Conditions

0.6 .

-~

N

8

z

-

0.2

-

N2

ntll

pmoVg X107 M-’

+ ATP

3 0

rn

Ki

Nl 0.4

Class 2

0.54 0.67

- ATP

8

7

5

6

-log ([Ca”]

free

nH2

ImoVg X1O6 M-‘

0.9 1.0

2.57 2.94

K2

0.20 1.03

0.27 0.37

1.4 1.o

4

/M)

Fig.2. CaZ+binding to TN C in cardiac myofibrils in the presence of ATP. Caz+binding to TN . C in myofibrils was obtained from three sets of data, as shown in Fig. 1, by subtracting the Ca2+ binding to TN . C-extracted myofibrils from the binding obtained with the myofibrils reconstituted with TN . C. The data points represent the meant- SE. The solid curve represents the weighted nonlinear least-squares fit of the data to a model with assuming cooperativity as described by Eqn(1); the best fit was obtained at the weighted sum of squares of deviations of the data from the fitting function 01‘)= 0.71. The dashed curve represents the weighted nonlinear least-squares fit of the data to a model without assuming cooperativity, which is described by Eqn (l), setting nHI= 1 and Nl = 2N2; the best fit was obtained at N, = 2N2 = 0.50 FmoVg, K, = 2.89X107M-’, K, = 3.43X105 M-’, x2 = 2.46. The extra sum of squares principle F test [14] showed that the goodness of fit was improved by assuming cooperativity; F,,,, = 9.86 > F(0.5%, 3, 12) = 7.23.

were observed to bind a significant amount of Ca2+ without saturation. Reconstitution with purified cardiac TN . C markedly increased the amount of bound Ca2+over the entire concentration range of free Ca2+; the TN . C-unextracted control myofibrils bound an even greater amount of Ca2+ than the TN . C-reconstituted myofibrils, which is consistent with the observation that CDTA treatment also extracts another Ca2+binding protein, LC2 (Table 1). The increased amount of bound Ca” (Fig. 2), which should be attributed to Ca2+binding to TN . C, became linear in the pCa (-logl,([Ca2+],d M) range 5.0-4.0 and had an inflection point in the pCa 6.0 region, which indicated that two classes of Ca2+-bindingsites exist, and each with a different affinity. Thus, the following equation was fitted to the data by means of a weighted nonlinear least-squares method, assuming that there are two classes of binding sites with cooperativity : N,(K,[Ca2+]P

B=C 1 + ( K ~[ c a 2 + ] p’

(1)

where B is the total amount of bound Ca2+as a function of the free Ca2+ concentration ([Ca”]), and N,, K,, and nH3, are the Ca2+-bindingcapacity, the apparent binding constant (reciprocal of the free Ca’+ concentration required for half-maximal binding) and the Hill coefficient of each class of binding site, respectively. The solid curve in Fig. 2 represents the best fit, and the obtained Ca2+-binding parameters are summarized in Table 2. The apparent Ca2+-bindingconstant of the class-1 binding sites (Kl = 2.6X107 M-I) is two-orders larger than that of the class-2 binding sites (K2 =

I

8

7

6

5

4

-log,, ([Ca**lfree IM) Fig. 3. A comparison of Ca*+-activatedcardiac myofibrillar ATPase with Ca2+binding in the presence of ATP to class 1 and 2 sites of cardiac TN * C in myofibrils. The ATPase activities of the CDTA-treated myofibrils reconstituted with purified cardiac TN . C (0) and the TN . C-unextracted control myofibrils (A)were measured under the same conditions as those used for the Ca2+-binding measurements in the presence of ATP, and was compared with the Ca2+-binding curves of the class-1 sites (curve 1) and class-2 sites (curve 2) of TN . C in myofibrils, which were computed using the values of the Ca2+-binding parameters listed in Table 2.

2.0X105 M-I). Ca2+ binding to the class-2 sites, but not to the class-1 sites, shows a positive cooperativity with a Hill coefficient of 1.4. The Ca2+-bindingcapacities of classes 1 and 2 (Nl and N2, respectively) are in the ratio 2: 1, which is equivalent to the stoichiometry of the high-affinity sites (sites 111 and IV) and the low-affinity site (site 11) of isolated cardiac TN . C. A simpler model without assuming cooperativity, which is described by Eqn (l), setting nH, = 1, did not provide a statistically significantly worse fit, but the Ca”binding capacities of classes 1 and 2 were in the ratio 4 :3. The cooperative model described by Eqn (1) provided a statistically significantly better fit than the non-cooperative model when the ratio NJN, was set to 2 at a level of significance of 0.5%. These results indicate that the two high-affinity sites have a two-order higher affinity for Ca2+ than the single low-affinity site, while also suggesting that Ca2+binding to the low-affinity site, but not to the high-affinity sites, is cooperative when cardiac TN . C is integrated into the intact myofilament lattice of myofibrils. Fig. 3 compares the Ca2+bindings of class-1 and class-2 sites with the Ca2+-activatedATPase activity of the myofi-

600 2.5 brils reconstituted with purified cardiac TN . C. The Ca2+A binding curve calculated using the binding parameters for class 2, listed in Table 2, passed through the data points of 2.0 myofibrillar ATPase activity, while the calculated curve for 2 class 1 almost reached a plateau before the activation of ATPase. The Hill coefficient and apparent Ca2+-binding 0 I constant for the ATPase activity obtained from three inI dependent experiments were 1.5? 0.1 X lo5 M-' and N+ 2.54 ? 0.08X lo5 M-' ( m e a n t SE), respectively, which was 8 1.0 in close agreement with those values for Ca2+binding to the '0 C sites of class 2. The results indicate that the myofibrillar 0 a 0 0 b ATPase activity is activated in direct proportion to the Ca2+ m 0.5 n b binding to site I1 of TN . C in cardiac myofibrils. Fig. 4 shows the typical Caz+-bindingdata in the absence of ATP. The Ca2+binding to the TN . C-extracted myofibrils was similar to that in the presence of ATP (Figs 1 and 4A), but the increased amounts of bound Caz+after TN . C reconstitution in the pCa range 7.0-5.0 were much greater than those in the presence of ATP. The Ca2+ binding to TN ' C 1.2 obtained by substraction is monophasic, and it is still unclear 16 on visual inspection as to whether multiple classes of Ca2+binding sites are present (Fig. 4B). However, the computer curve fitting using Eqn(1) revealed the presence of two classes of binding sites, each with a different affinity for Ca2+ (Table 2). As in the presence of ATP, classes 1 and 2 were stoichiometrically equivalent to the high-affinity and low0.6affinity sites of isolated cardiac TN . C, respectively. The apparent Ca2+-binding constant of class 2 (Kz = 1.0 X106 M-I) was much larger (fivefold) than that in the pres0.4 3 ence of ATP (K2 = 2.0X105 M-'), while the binding conm" 0.2 stants of class 1 were nearly the same. In contrast to in the presence of ATP, neither of the two classes of the Ca2+-binding sites showed any cooperativity. Curve fitting with a sim0.0 pler model in which it is assumed that there is only a single 8 7 6 5 4 class of binding sites without cooperativity provides a statistically significantly worse fit. These results indicate that the attachment of myosin crossbridges to actin in the rigor state enhances the Caz+ binding to site I1 of cardiac TN . C five- Fig.4. Caz+ binding to cardiac myofibrils and TN * C in the absence of ATP. Ca2+binding was measured as described in Materifold, but cannot confer any cooperativity on Ca2+binding. als and Methods. (A) Ca2+ binding to TN . C-extracted (CDTA-

E

Y

Y

d

DISCUSSION A single low-affinity Ca2+-bindingsite, site 11, of cardiac TN . C is thought to be a trigger site of cardiac muscle contraction, based on the observations that the activation of myofibrillar ATPase and the isometric contraction of muscle fibers occur in approximately the same free Ca2+ concentration ranges when Caz+binds to site I1 of TN . C in isolated troponin and muscle fibers, respectively [7, 141. Recently, this idea was directly demonstrated by the study using the technique of site-directed mutagenesis, which showed that cardiac TN . C lost its regulatory function upon inactivation of the Ca2+-binding site I1 [29]. However, it is still unclear as to what kind of function correlates Ca2+binding to site I1 of TN . C with the contractile response. While Ca2+binding of isolated cardiac TN . C or its complexes with other troponin subunits to site I1 are not cooperative [7], the contractile response of the cardiac muscle is cooperative [14, 25, 301, which suggests that the contractile response is not a simple function of Ca2+binding to site 11. The cooperative feature of Caz+regulation has often been explained by theoretical models which assume the nearest-neighbor protein-protein interactions in the thin filaments as a source of cooperativity [31-331. Previous experiments measuring the Ca2+

treated) myofibrils ( 0 )and the CDTA-treated myofibrils reconstituted with purified TN . C (0).(B) Caz+ binding to TN . C in myofibrils obtained from three sets of data, as shown in (A), by subtraction. The data points represent the mean2SE. The solid curve represents the weighted non-linear least-squares fit of the data to the Eqn (1); the best fit was obtained at ,y2 = 1.50. The dashed curve represents the weighted non-linear least-squares fit of the data to a model for a single class of binding sites without cooperativity which is described by the equation, B = NKICaz+]/(l NK[CaZ+]); the best fit was obtained at N = 0.91 pmol/g, K = 1 3 7 x 1 0 ' M-', x2 = 18.2. Fcalc = 27.8>F(0.5%, 4, 10) = 7.34.

+

binding to cardiac skinned muscle fibers failed to address this problem, probably because of the difficulty in distinguishing Ca2+binding to TN . C by directly fitting a theoretical equation to the Ca2+-binding data with large background levels [14]. In the present study, Ca2+ binding to TN . C in isolated cardiac myofibrils was evaluated by subtracting the amount of Ca2+ which was bound to TN . C-extracted (CDTAtreated) myofibrils from that bound to myofibrils reconstituted with purified cardiac TN . C. Most of TN . C molecules have previously been shown to be extracted by the CDTA treatment [21, 251. In this study, the CDTA-treated myofibrils actually bound a very small amount of Ca2+in the pCa range

601 8.0-5.0 (Figs 1 and 4A). Reconstitution with TN . C markedly enhanced the Ca2+ binding to myofibrils, and this allowed us to evaluate, with a high accuracy, Ca2+ binding to TN . C in myofibrils by subtraction. The evaluated Ca2+ binding clearly shows that there are two classes of binding sites which are attributable to the low-affinity and high-affinity Ca2+-bindingsites of cardiac TN . C. The Ca2+ affinities of the two classes are in close agreement with those reported for CaZ+binding to the restrained or unrestrained skinned cardiac muscle fibers under similar conditions of ionic milieus [14]. The most significant finding of the present study is that Ca2+binding to a single low-affinity site (site 11) of cardiac TN . C integrated into the myofilament lattice of myofibrils is cooperative in the presence of ATP and in direct proportion to the activation of myofibrillar ATPase activity. Since positive cooperativity for a single Ca2+-binding site is, in principle, impossible, the present study indicates that an intermolecular interaction between TN . C molecules along the thin filament exists or that coupling between Ca2+ binding to the single binding site and the attachment of myosin heads to actin, as discussed in a theoretical model which has previously been presented for Ca2+ binding to skeletal TN . C in skeletal myofibrils [16] exists. In contrast to the ATPase activity of cardiac myofibrils regulated by cardiac TN . C, we have previously shown that the ATPase activity of cardiac myofibrils regulated by skeletal TN . C is not a simple function of the Caz+ binding to the low-affinity sites of TN . C; Ca2+binding to the two low-affinity sites of skeletal TN . C occurs in a 5 -6-times lower Ca2+concentration range compared to the activation of myofibrillar ATPase activity [21]. Further work will be needed to clarify that this is because both the two low-affinity sites or only one particular site of skeletal TN . C is occupied by Ca2+ to activate the cardiac myofibrillar ATPase activity. In ATP-free solutions, the Caz+ binding to the low-affinity site was found to be enhanced approximately fivefold. This is comparable to the approximately 3 -4-fold enhancement reported for Ca2+ binding to the low-affinity site of TN . C in skinned cardiac muscle fibers [14], which further substantiates the fact that only the low-affinity site of cardiac TN . C is affected by the attachment of myosin crossbridges to actin. Furthermore, the cooperativity in Ca2+ binding to the low-affinity site was lost in the absence of ATP, which suggested that Ca2+binding to the low-affinity site of cardiac TN . C is cooperatively enhanced by a dynamic interaction of myosin crossbridges with actin in the presence of ATP but not by the static rigor crossbridge attachment. When the ATPase activity and the Ca2’ binding are measured in the presence of ATP, myofibrils contract freely at high Ca2+ concentrations and the geometry of myofilaments is considered to be altered. This raises the possibility that the observed positive cooperativities in the ATPase activity and the Ca2+ binding to the low-affinity site in the presence of ATP might be attributable to the super-contraction of myofibrils which occurs only in the presence of ATP. However, it may be expected that the shortening of myofibrils during the measurements with increased CazCconcentrations leads to a decrease rather than increase in cooperativity, since the Caz+ sensitivity of isometric tension development in skinned cardiac muscle fibers has been shown to be reduced at a shorter sarcomere length [34]. This may also account for the relatively smaller Hill coefficient value for the cardiac myofibrillar ATPase activity (n, = 1.5) compared to the reported

values for the isometric force generation of skinned cardiac muscle fibers (n, = 2.7-3.0) [34]. In previous studies, approximately 50% of the endogenous LC2 was shown to be lost in the CDTA-treated cardiac myofibrils and the partial removal of LC2 had no significant effect on the Ca2+-activationprofile of ATPase, which indicated that LC2 does not play an important role in the regulatory mechanism of cardiac myofibrillar ATPase activity [27, 281. However, it was reported that the maximum velocity of shortening in single skinned skeletal muscle fibers [35] and the sliding velocity of actin filaments on the skeletal-myosincoated coverslip in a motility assays [36] were reduced by the removal of LC2. Further investigations are still required to determine whether or not the removal of LC2 from myosin changes the relationship between the Ca’+ binding to TN . C and the cardiac myofibrillar ATPase activity. It is known that cardiac TN . I and TN . T can be phosphorylated in vitro by several kinds of protein kinases. Phosphorylation of cardiac TN . I by CAMP-dependent protein kinase decreases the Ca2+ sensitivity of ATPase activity of native actomyosin [37] or myofibrils [17] without affecting basal or maximum ATPase activities, and also decreases the CaZ+-binding affinity of TN . C [38]. In contrast, phosphorylation of cardiac TN . I or TN . T by protein kinase C inhibits the ATPase activity of reconstituted actomyosin or myofibrils without affecting the Ca2+sensitivity [39]. Although we have not checked on the state of phosphorylation of TN . I and TN . T in the myofibrils used in this study, present experiments have been carried out using the same myofibrillar preparation at the same time so that, even if TN . I or TN . T had been phosphorylated, the state of phosphorylation would not vary from preparation to preparation. Further study using the method described in this paper for measuring Ca2+binding to TN . C in myofibrils may shed some light on the molecular mechanism of modulation, by protein kinases, of CaZ+regulation of cardiac muscle contraction. Finally, the findings presented here not only verified the general view that the Ca2+-bindingsite I1 of cardiac TN . C is responsible for the regulation of contraction under physiological conditions, but also demonstrated that the activation of cardiac myofibrillar ATPase activity is a simple (linear) function of the occupancy of site I1 by Ca2’. The data also provide strong evidence that either a feedback mechanism or synergism exists between the Ca2+ binding to site I1 of cardiac TN . C and myosin crossbridge attachment, which may cause the cooperative Ca2+regulation of cardiac muscle contraction.

REFERENCES 1. Ebashi, S., Kodama, A. & Ebashi, F. (1968) Troponin. I. Preparation and physiological function, J. Biochem. (Tokyo) 64, 465-477. 2. Ebashi, S., Endo, M. & Ohtsuki, I. (1969) Control of muscle contraction, Quart. Rev. Biophys. 2, 351-384. 3. Ebashi, S., Ohtsuki, I. & Mihashi, K. (1972) Regulatory proteins of muscle with special reference to troponin, Cold Spring Harbor Symp. Quant. Biol. 37, 215-223. 4. Ohtsuki, I., Maruyama, K. & Ebashi, S. (1986) Regulatory and cytoskeletal proteins of vertebrate skeletal muscle, Adv. Prot. Chem. 38, 1-67. 5. Leavis, P. C. & Kraft, E. L. (1978) Calcium binding to cardiac troponin C, Arch. Biochem. Biophys. 186, 411 -415. 6. Kohama, K. (1979) Divalent cation-binding properties of slow skeletal muscle troponin in comparison with those of cardiac and fast skeletal muscle troponins, J. Biochem. (Tokyo) 86, 811-820.

7. Holroyde, M. J., Robertson, S. P., Johnson, J. D., Solaro, R. J. & Potter, J. D. (1980) The calcium and magnesium binding sites on cardiac troponin and their role in the regulation of myofibrillar adenosine triphosphatase, J. Biol. Chem. 255, 11688- 11693. 8. Johnson, J. D., Collins, J. H., Robertson, S. P. & Potter, J. D. (1980) Fluorescent probe study of Ca2+binding to Ca2+-specific sites of cardiac troponin and troponin c , J. B id . Chem. 255, 9635-9640. 9. van Eerd, J.-P. & Takahashi, K. (1976) Determination of the complete amino acid sequence of bovine cardiac troponin C, Biochemistry 15, 1171-1180. 10. Kobayashi, T., Takagi, K., Konishi, S., Morimoto, S. & Ohtsuki, I. (1989) Amino acid sequence of porcine cardiac muscle troponin C, J. Biochem. (Tokyo) 106, 55-59. 11. Robertson, S. P., Johnson, J. D. & Potter, J. D. (1981) The timecourse of Caz exchange with calmodulin, troponin, parvalbumin, and myosin in response to transient increases in Caz+, Biophys. J. 34, 559-569. 12. Bremel, R. D. & Weber, A. (1972) Cooperation within actin filament in vertebrate skeletal muscle, Nature 238, 97-101. 13. Zot, H. J., Guth, K. & Potter, J. D. (1986) Fast skeletal muscle skinned fibers and myofibrils reconstituted with N-terminal fluorescent analogues of troponin C, J. Biol. Chem. 261, 15 883- 15 890. 14. Pan, B.-S. & Solaro, R. J. (1987) Calcium-binding properties of troponin C in detergent-skinned heart muscle fibers, J. Biol. Chem. 261, 15883-15890. 15. Guth, K. & Potter, J. D. (1987) Effect of rigor and cycling crossbridges on the structure of troponin C and on the Caz+affinity of the Caz+-specific regulatory sites in skinned rabbit psoas fibers, J. Biol. Chem. 262, 13627-13635. 16. Morimoto, S. (1991) Effect of myosin cross-bridge interaction with actin on the Caz+-binding properties of troponin C in fast skeletal myofibrils, J. Biochem. (Tokyo) 109, 120- 126. 17. Holroyde, M. J., Howe, E. & Solaro, R. J. (1979) Modification of calcium requirements for activation of cardiac myofibrillar ATPase by cyclic AMP dependent phosphorylation, Biochim. Biophys. Acta 586, 63-69. 18. Solaro, R. J. & Riiegg, J. C. (1982) Stimulation of Ca2+binding and ATPase activity of dog cardiac myofibrils by AR-L 115BS, a novel cardiotonic agent, Circ. Res. 51, 290-294. 19. Hofmann, P. A. & Fuchs, F. (1987) Effect of length and crossbridge attachment on Ca2+binding to cardiac troponin C, Am. J. Physiol. 253, C90-C96. 20. Hofmann, P. A. & Fuchs, F. (1987) Evidence for a force-dependent component of calcium binding to cardiac troponin C, Am. J. Physiol. 253, C541 -C546. 21. Morimoto, S. & Ohtsuki, I. (1989) Caz+binding to skeletal muscle troponin C in skeletal and cardiac myofibrils, J. Biochem. (Tokyo) 105, 435-439. 22. Morimoto, S. (1991) The effect of Mgz+on the Ca2+binding to troponin C in rabbit fast skeletal myofibrils, Biochim. Biophys. Acta 1073, 336-340. 23. Solaro, R. J., Pang, D. C. & Briggs, F. N. (1971) The purification of cardiac myofibrils with Triton X-100, Biochim. Biophys. Acta 245, 259-262. 24. Tsukui, R. & Ebashi, S. (1973) Cardiac troponin, J. Biochem. (Tokyo) 73, 1119-1121. 25. Morimoto, S. & Ohtsuki, I. (1987) Ca2+-and Sr2+-sensitivityof the ATPase activity of rabbit skeletal myofibrils: effect of the

26. 27.

28.

29.

+

30. 31. 32. 33. 34.

35.

36. 37.

38.

39.

40.

complete substitution of troponin C with cardiac troponv C, calmodulin, and parvalbumins, J. Biochenz. (Tokyo) 101, 291 -301. Baginski, E. S., Foa, P. P. & Zak, B. (1967) Determination of phosphate: study of labile organic phosphate interference, Clin. Chim. Acta 15, 155-158. Morimoto, S. & Ohtsuki, I. (1988) Effect of substitution of troponin C in cardiac myofibrils with skeletal troponin C or calmodulin on the Ca2+-and SrZ+-sensitiveATPase activity, J. Biochem. (Tokyo) 104, 149-154. Nakamura, Y., Shiraishi, F. & Ohtsuki, I. (1994) The effect of troponin C substitution on the Ca2' -sensitive ATPase activity of vertebrate and invertebrate myofibrils by troponin Cs with various numbers of Ca2+-binding sites, Comp. Biochem. Physiol. 108B, 121-133. Putkey, J. A., Sweeney, H. L. & Campbell, S. T. (1989) Sitedirected mutation of the trigger calcium-binding sites in cardiac troponin C, J. Biol. Chem. 264, 12370-12378. Blanchard, E. M. & Solaro, R. J. (1984) Inhibition of the activation and troponin calcium binding of dog cardiac myoi'ibrils by acidic pH, Circ. Res. 55, 382-391. Tawada, Y. & Tawada, K. (1975) Co-operative regulation nechanism of muscle contraction: inter-tropomyosin co-operation model, J. Them Biol. 50, 269-283. Hill, T. L. (1983) Two elementary models for the regulation of skeletal muscle contraction by calcium, Biophys. J. 44, 383396. Shiner, J. S. (1986) A theoretical analysis of binding to the Ca2+-specific sites on troponin incorporated into thin filaments, Biophys. J. 50, 601 -611. Gulati, J., Sonnenblick, E. & Babu, A. (1990) The role of troponin C in the length dependence of Ca'+-sensitive force of mammalian skeletal and cardiac muscles, J. Physiol. 441, 305 -324. Moss, R. L., Giulian, G. G. & Greaser, M. L. (1983) Effects of EDTA treatment upon the protein subunit composition and mechanical properties of mammalian single skeletal muscle fibers, J. Cell Biol. 96, 970-978. Lowey, S., Waller, G. S. & Trybus, K. M. (1993) Skeletal muscle myosin light chains are essential for physiological speeds of shortening, Nature 365, 454-456. Yamamoto, K. & Ohtsuki, 1. (1982) Effect of phosphor>lation of porcine cardiac troponin I by 3'5'-cyclic AMP-dependent protein kinase on the actomyosin ATPase activity, J. Biochrm. (Tokyo) 91, 1669-1677. Robertson, S. P., Johnson, J. D., Holroyde, M. J., Kranias, E. G., Potter, J. D. & Solaro, R. J. (1982) The effect of troponin I phosphorylation on the Caz+-binding properties of the Caz+regulatory site of bovine cardiac troponin, J. Biol. Chem. 257, 260-263. Noland, T. A. & Kuo, J. F. (1993) Protein kinase C phosphorylation of cardiac troponin I and troponin T inhibits Ca'. -stimulated MgATPase activity in reconstituted actomyosin and isolated myofibrils, and decreases actin-myosin interactions, J. Mol. Cell Cardiol. 25, 53-65. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227, 680-685.