Increased Ca2+ Affinity of Cardiac Thin Filaments Reconstituted with Cardiomyopathy-related Mutant Cardiac Troponin I*

To understand the molecular mechanisms whereby cardiomyopathy-related cardiac troponin I (cTnI) mutations affect myofilament activity, we have investigated the Ca2+ binding properties of various assemblies of the regulatory components that contain one of the cardiomyopahty-related mutant cTnI. Acto-S1 ATPase activities in reconstituted systems were also determined. We investigated R145G and R145W mutations from the inhibitory region and D190H and R192H mutations from the second actin-tropomyosin-binding site. Each of the four mutations sensitized the acto-S1 ATPase to Ca2+. Whereas the mutations from the inhibitory region increased the basal level of ATPase activity, those from the second actin-tropomyosin-binding site did not. The effects on the Ca2+ binding properties of the troponin ternary complex and the troponin-tropomyosin complex with one of four mutations were either desensitization or no effect compared with those with wild-type cTnI. All of the mutations, however, affected the Ca2+ sensitivities of the reconstituted thin filaments in the same direction as the acto-S1 ATPase activity. Also the thin filaments with one of the mutant cTnIs bound Ca2+ with less cooperativity compared with those with wild-type cTnI. These data indicate that the mutations found in the inhibitory region and those from the second actin-tropomyosin site shift the equilibrium of the states of the thin filaments differently. Moreover, the increased Ca2+ bound to myofilaments containing the mutant cTnIs may be an important factor in triggered arrhythmias associated with the cardiomyopathy.

To understand the molecular mechanisms whereby cardiomyopathy-related cardiac troponin I (cTnI) mutations affect myofilament activity, we have investigated the Ca 2؉ binding properties of various assemblies of the regulatory components that contain one of the cardiomyopahty-related mutant cTnI. Acto-S1 ATPase activities in reconstituted systems were also determined. We investigated R145G and R145W mutations from the inhibitory region and D190H and R192H mutations from the second actin-tropomyosinbinding site. Each of the four mutations sensitized the acto-S1 ATPase to Ca 2؉ . Whereas the mutations from the inhibitory region increased the basal level of ATPase activity, those from the second actin-tropomyosin-binding site did not. The effects on the Ca 2؉ binding properties of the troponin ternary complex and the troponin-tropomyosin complex with one of four mutations were either desensitization or no effect compared with those with wild-type cTnI. All of the mutations, however, affected the Ca 2؉ sensitivities of the reconstituted thin filaments in the same direction as the acto-S1 ATPase activity. Also the thin filaments with one of the mutant cTnIs bound Ca 2؉ with less cooperativity compared with those with wild-type cTnI. These data indicate that the mutations found in the inhibitory region and those from the second actintropomyosin site shift the equilibrium of the states of the thin filaments differently. Moreover, the increased Ca 2؉ bound to myofilaments containing the mutant cTnIs may be an important factor in triggered arrhythmias associated with the cardiomyopathy.
In experiments reported here, we have investigated the functional significance of mutations of cardiac troponin I (cTnI) 2 linked to cardiomyopathy. The Ca 2ϩ -dependent interaction of TnI with actin and TnC is one of the most important events in the regulation of striated muscle contraction. Ca 2ϩ binding to troponin C (TnC) triggers a series of conformational transitions among thin filament proteins that activate the thin filament (for review, see Refs. [1][2][3][4]. A current model for the regulatory mechanism of muscle contraction, originally derived from biochemical analysis, involves three states of the thin filament: blocked (B-state), closed (C-state), and open (M-state) (5)(6)(7). The three states have been defined to reflect different interactions between actin and the myosin head. The B-state, which the majority of the thin filaments occupy when the cytoplasmic [Ca 2ϩ ] is low and Ca 2ϩ is not bound to the regulatory site(s) on Tn (5,8,9), is stabilized by the interaction between TnI and actin. With Ca 2ϩ binding to the regulatory site(s) of TnC, a hydrophobic patch is exposed in the N-terminal regulatory site of TnC (10,11). In the case of cardiac TnC (cTnC), this structural change requires cTnI (12)(13)(14). The newly exposed hydrophobic patch interacts with the regulatory region of TnI (12,(15)(16)(17) and the C-terminal half of the TnI molecule moves away from actin (18 -22). In the C-state, non-force-generating cross-bridges bind weakly to actin. The M-state is associated with strong binding of force-generating crossbridges that induce further movement of Tm observed by the threedimensional reconstruction of electron micrograph images (7,23,24). Agreement with the model has also come from Förster resonance energy transfer measurements (25)(26)(27).
There are more than 25 genetic mutations of cTnI that have been reported to be related to cardiomyopathy. Most of the mutations are distributed in the C-terminal half of the molecule, which contains two or more actin-Tm interacting sites, i.e. the inhibitory region and the second actin-Tm site. There is ample evidence for the importance of the interactions of these regions with actin-Tm at low [Ca 2ϩ ]. For example, the deletion of the inhibitory region from fsTnI drastically impaired its ability to inhibit the actin-activated myosin ATPase activity (28). These interactions hold most of the thin filaments in the B-state at low [Ca 2ϩ ] (5,7,9). With Ca 2ϩ -bound to the regulatory site(s) of TnC, the population of the thin filaments in the B-state decreases and the C-state and the M-state predominate. If the cardiomyopathy-related mutations destabilize the B-state, the energy barrier of the transition from the B-to the C-states would be lower and thus the equilibrium of the thin filaments would shift toward the C-state. Hence the myofilaments show a sensitization to Ca 2ϩ . Yet so far there are no reports regarding the effects of cardiomyopathy-linked mutations of cTnI on the equilibrium of the thin filaments states.
Previous studies showed that the R145G mutation of cTnI sensitizes the myofilament activity to Ca 2ϩ and causes diastolic dysfunction. However, the molecular mechanism for the Ca 2ϩ sensitization remains unclear. There are at least two possible mechanisms to sensitize the myofilament to Ca 2ϩ : 1) by enhancing the Ca 2ϩ affinity of the regulatory site of the Tn and 2) by enhancing myosin binding to actin (3). These are not completely separate mechanisms, since there are reports that showed that the cross-bridge attachment to the thin filament increases the Ca 2ϩ binding affinity of the regulatory site(s) of TnC (29 -31).
In this study, we investigated four cardiomyopathy-related mutations: two (R145G and R145W) from the inhibitory region and two from (D190H and R192H) from the second actin-Tm binding region. These mutations are linked to either hypertrophic or restricted cardiomyopathy. Our data provided the first measurements of the Ca 2ϩ binding properties of various protein complexes with one of these cardiomyopathy-related mutant cTnI and demonstrated that these mutations increase the Ca 2ϩ binding affinity of the thin filament.

MATERIALS AND METHODS
Proteins-Recombinant human wild-type (wt) and mutant cTnCs in pET3d vector were expressed using BL21(DE3) cells. cTnC was extracted with 5% sucrose, 1 mM EDTA, 50 mM Tris/HCl, pH 8.0, and protease inhibitors. After centrifugation, the supernatant fraction was collected and CaCl 2 and MgCl 2 were added to final 5 mM and 1 mM, respectively, followed by the addition of ammonium sulfate to 60% saturation. After centrifugation, the supernatant fraction was applied to a phenyl-Sepharose column equilibrated with 1 M NaCl, 1 mM CaCl 2 and 50 mM Tris/HCl, pH 8.0. After washing with the same solution, the column was further washed with the same solution except that 0.2 mM CaCl 2 was used instead of 1 mM. Finally cTnC was eluted with 5 mM EDTA and 20 mM Tris/HCl, pH 8.0. The cTnC-containing fraction was dialyzed against 1 mM EDTA, 20 mM Tris/HCl, pH 8.0, and 1 mM DTT. Solid urea was added to the protein solution, and cTnC was separated on a QAE fast flow Sepharose column equilibrated with 6 M urea, 1 mM EDTA, and 20 mM Tris/HCl, pH 8.0. cTnC was eluted with a linear gradient of 0 -0.5 M NaCl. Recombinant mouse cTnIs were expressed and purified as described previously (32). Recombinant mouse cTnT with a myc-tag at the N terminus was expressed and purified with a combination of ammonium sulfate fractionation and a DEAE-Sepharose column chromatography as described (33). Tropomyosin was prepared from bovine left ventricular ether powder as described previously (34) and further purified by isoelectric point precipitation. Actin was prepared from bovine left ventricular ether powder (34). Myosin subfragment-1 (S1) was prepared by chymotryptic digestion of rabbit psoas muscle myosin and purified on a SP-Sephadex column as described (35).
Labeling of cTnC-Single Cys residue of mutant cTnCs, cTnC(C35S), and cTnC(C85S) were labeled with 1.5-fold excess amount of IAANS in the presence of 1 M NaCl, 5 mM MgCl 2 , and 20 mM HEPES, pH 7.4. The reaction was quenched by the addition of DTT, and the excess IAANS was removed by dialysis and a desalting column. Labeling yield was determined using ⑀ 326 nm ϭ 27,000 M Ϫ1 for IAANS.
Reconstitution of the Tn Complex and the Thin Filament-Equimolar amount of Tn components were combined in a solution containing 6 M urea, 1 M NaCl, 5 mM MgCl 2 , 0.1 mM CaCl 2 , 1 mM DTT, and 20 mM Tris/HCl, pH 8.0. The protein mixture was dialyzed against the same solution without urea, then NaCl concentration was reduced to 0.3 M and finally to 0.1 M. After dialysis, the protein mixture was clarified by a brief centrifugation and then applied to a Resource-Q (1 ml, Amersham Biosciences) column equilibrated with 0.1 M NaCl, 5 mM MgCl 2 , and 20 mM Tris/HCl, pH 8.0. The Tn complex was eluted with a linear gradient of 0.1-0.5 M NaCl in the same solution (36).
To obtain the reconstituted thin filaments, we used two different methods (1). We mixed actin:Tm to 6:1 molar ratio, followed by the addition of Tn to make actin:Tm:Tn ϭ 6:1:1 molar ratio (2). In another method, actin-Tm was first sedimented. The resultant pellet was suspended with 0.1 M NaCl, 5 mM MgCl 2 , and 20 mM MOPS, pH 7.0 and incubated with the excess amount of the Tn complex (1.5:1 ϭ Tn:Tm). The reconstituted thin filaments were sedimented at 18,000 ϫ g at 4°C for 50 min. The data obtained from each of the preparations were indistinguishable from each other.
Actin-activated Acto-S1 ATPase Measurements-We modified the micro-ATPase assay developed by Dobrowolski et al. (37). The typical reaction conditions were 6 mM actin, 0.5 mM myosin S1, 1 mM Tm, and 1.2 mM Tn in 50 mM NaCl, 5 mM MgCl 2 , 20 mM MOPS, pH 7.0, and various concentration of CaCl 2 at 25°C. Free Ca 2ϩ concentration was calculated using the WINMAXC Version 2.10 or SLIDERS Version 2.00 programs (38). A reaction was initiated by the addition of final concentration of 1.0 mM ATP. ATPase activity was determined from a time course of inorganic phosphate liberation up to 10 min. Every 2 min, a 10-ml aliquot was removed and the reaction was terminated by 90 ml of 0.2 M perchloric acid. The amount of released phosphate was determined using the malachite green method (39). In all figures and tables, the rate for S1 alone has been subtracted from the measured rates.
Ca 2ϩ Binding Measurements-The steady-state fluorescence measurements were carried out using a model 2000 -4 Spectrofluorometer equipped with two 814 PMT photon-counting detectors (Photon Technology International) with a cell holder containing a thermostat and a magnetic stirrer. The Ca 2ϩ binding was monitored by fluorescence emission of IAANS attached at either Cys-35 or Cys-84 of cTnC. The fluorescence emission intensity change observed was assumed to be due to the direct Ca 2ϩ binding to the regulatory site of cTnC in the protein complexes. The solution conditions were 50 mM NaCl, 5 mM MgCl 2 , 1 mM EGTA, and 20 mM MOPS, pH 7.0. The titration was carried out at 25°C, and the free Ca 2ϩ concentration was calculated using the WIN-MAXC Version 2.10 or SLIDERS Version 2.00 programs. The following equation was used to analyze the titration curve, where ⌬F i is the total fluorescence signal change after ith addition of stock Ca 2ϩ solution, X i is the free Ca 2ϩ concentration after the ith addition, and n H and K are the Hill coefficient and the association constant for a Ca 2ϩ -binding site, respectively. ⌬F max is the maximum fluorescence change.
The Choice of Cardiomyopathy-related Mutations-Arg-145 is located in the middle of the inhibitory region and the R145G mutation is the most studied among cTnI cardiomyopathy-related mutations (40 -48). Thus we chose R145G as a "standard." The mutation at the same position Arg-145 to Trp was reported to cause restrictive cardiomyopathy (RCM) (49), which is characterized by impaired filling in the ventricles without increased wall thickness. Thus different mutations at the same position are associated with different phenotypes. To elucidate if functional differences exist at the level of myofilaments among difference diseases, we chose R145W. Tripet et al. (50) reported that, besides the inhibitory region, fsTnI contains another actin-Tm-binding site at its C-terminal part (residues 140 -148 of fsTnI). Ramos (51) reported a different second actin-Tm site for fsTnI. Ramos (51) found that the very C-terminal part of fsTnI is involved in the binding of fsTnI to the thin filament. For cTnI, Rarick et al. (52) found that the cTnI segment residues 188 -198 contains the second actin-Tm site. Peptide array experiments supported the latter finding. 3 Thus we consider this region as a second actin-Tm-binding site of cTnI. Among the region residues 188 -198 of cTnI, two mutations, D190H and R192H, were reported to link to RCM (49). We chose these mutations to investigate whether there is a functional difference of the myofilament activity between the preparations containing mutations from the inhibitory region and those from the second actin-Tm sites.
Effects of the Mutations on the Acto-S1 ATPase Activity-To assess functional effects of mutant cTnIs, we measured acto-S1 ATPase activity in a reconstituted system. Fig. 1A shows the pCa-actin activated S1 ATPase activity relationship of reconstituted preparations with wt-TnI or either of the two mutations from the inhibitory region of cTnI. Fig. 1B shows the data for wt-TnI and mutations from the second actin-Tm site of cTnI. In Table 1, we summarized the parameters obtained from the acto-S1 ATPase measurements.
As shown in Fig. 1 and Table 1, all the mutations retained the Ca 2ϩdependent regulatory property for actin-activated S1-ATPase activity. The most notable departure from the wild-type cTnI was the impaired inhibitory action of TnI mutants with R145G or R145W at low [Ca 2ϩ ]. Whereas wt-cTnI and mutants from the second actin-Tm site shut down actin activated S1-ATPase almost completely at low [Ca 2ϩ ], R145G and R145W retained 34 -40% of the maximum activity even at low [Ca 2ϩ ]. In all cases, the maximum activities were indistinguishable from the maximum activity with wt-TnI. All four mutants sensitized the preparations to Ca 2ϩ and induced a leftward shift of acto-S1 ATPase activity to nearly the same extent (⌬pCa 50 ϭ ϳ0.3, where pCa 50 indicates pCa at 50% of maximal change) compared with wt-TnI. The Hill coefficient (n H ) was 1.8 for wt-TnI, whereas it was 1.1-1.3 for mutant cTnIs. These observations indicate that the mutations in the inhibitory region and the second actin-Tm site of cTnI both affected the distribution of the thin filament states but differently. Differences between the mutations that link to HCM and those linked to RCM were not clear from these experiments. It should be mentioned that recently Gomes et al. (53) reported that the R192H mutation of cTnI also impaired the ability of cTnI to inhibit actomyosin ATPase activity at low [Ca 2ϩ ]. Our results are consistent with the observation that the C-terminal truncation of cTnI did not affect the basal level of ATPase and unloaded filament velocity (54).
Effects of the Mutations on the Ca 2ϩ Binding Properties-To elucidate the molecular mechanisms that cause the shift of the Ca 2ϩ sensitivity of ATPase activity and that of the equilibrium of the thin filament states by the mutations introduced into cTnI, we measured the Ca 2ϩ binding properties of cTn, cTn-Tm, and the reconstituted thin filament with one of the recombinant cTnIs. Ca 2ϩ binding to a single regulatory site of cTnC in various complexes was monitored by the intensity change of the fluorescence emission from 1-anilino-8-naphthalenesulfonate moiety attached to a single Cys residue of mutant cTnC.
We labeled Cys-35 of cTnC(C84S) with IAANS for the measurements of Ca 2ϩ binding to a single regulatory site of the Tn complex. Details of fluorescence properties of IAANS at Cys-35 of cTnC in the ternary complex were characterized previously (55,56). Upon Ca 2ϩ binding to the regulatory site of cTnC, the fluorescence intensity decreased to 46 -48% of that in the absence of bound Ca 2ϩ (Fig. 2A). Thus the Ca 2ϩ binding was easily monitored. Representative data are shown in Fig. 3 and Ca 2ϩ binding properties are summarized in Table 2. Wild-type cTn binds Ca 2ϩ with a K d ϭ 2.0 ϫ 10 Ϫ7 M (pCa 50 ϭ 6.69), which is consistent with the previously determined values (57,58). The cTn complexes with the mutations in the inhibitory region, particularly R145W mutation, bind Ca 2ϩ slightly weaker compared with wt-Tn (Table 2), i.e. in a direction opposite to that from pCa-ATPase relationship. This indicates that the mutations from the inhibitory region of cTnI slightly destabilize the Ca 2ϩ -bound state of cTn complex. This is consistent with the previous observation that R145G mutation reduces the affinity of cTnI for cTnC in the presence of Ca 2ϩ (47,59). The mutations from the second actin-Tm sites had essentially no effect on the Ca 2ϩ binding affinity of the Tn complex. These data indicate that either 1) these mutations have little or no effect on the Tn structure, or 2) these mutations disturb the Tn structure in the presence and absence of Ca 2ϩ to the same extent. All the Tn complexes bind Ca 2ϩ with a Hill coefficient (nH) ϳ1.0, which suggests the validity of our measurements and calculation of [Ca 2ϩ ].
Ca 2ϩ binding to the Tn-Tm complex was also measured as with the Tn complex ( Fig. 4 and Table 2). Cys-35 of cTnC(C84S) was labeled with IAANS. Ca 2ϩ binding induced a decrease of the fluorescence intensity to 55-58% of that without Ca 2ϩ bound at the single regulatory site ( Fig. 2A). Wild-type Tn-Tm complex bound Ca 2ϩ with a K d ϭ 2.0 ϫ 10 Ϫ7 M (pCa 50 ϭ 6.70) ( Table 2). Thus Tn-Tm binds Ca 2ϩ with almost the same affinity of the Tn complex, which is consistent with the previous study with crayfish Tn and Tn-Tm (60). The Tn-Tm complex with cTnI(R145W) again bound Ca 2ϩ with a weaker affinity (pCa 50 ϭ 6.60) compared with Tn-Tm with wt-cTnI. As before, other mutants did not significantly affect the Ca 2ϩ binding properties of Tn-Tm complex. The Hill coefficients for the Ca 2ϩ binding to the Tn-Tm complexes were about 1.0, which indicates the head-to-tail interaction of Tm molecules is weak or does not play a significant role in the cooperative behavior in the absence of actin under our experimental conditions.

FIGURE 1. Effect of cTnI mutations on Ca 2؉ -dependent actin-activated actoS1-ATPase activity.
A, the Ca 2ϩ -dependent actin-activated actoS1-ATPase activity of reconstituted system with either wt-cTnI (closed circles), cTnI(R145G) (closed squares), or cTnI(R145W) (closed triangles) as a function of pCa. Each data point represents the mean Ϯ S.E. of four to eight measurements. B, the Ca 2ϩ -dependent actin-activated actoS1-ATPase activity of reconstituted system with wt-cTnI (closed circles), cTnI(R192H) (closed squares), or cTnI(D190H) (closed triangles) as a function of pCa. Each data point represents the mean Ϯ S.E. of four to eight measurements. The rate for S1 alone has been subtracted from the measured rates.
Whereas the cardiomyopathy-related mutations of cTnI had either no or little opposite effect on the Ca 2ϩ binding to the Tn complex and the Tn-Tm complex, they affected the Ca 2ϩ binding to the thin filament in the same direction as ATPase activity. Ca 2ϩ binding to the reconstituted thin filament was monitored by the fluorescence emission intensity change of IAANS attached to Cys-84 of mutant cTnC(C35S) (61). Ca 2ϩ increased the fluorescence intensity about 22% (Fig. 2B). The mutations introduced into cTnI did not affect the extent of fluorescence change, suggesting that the fluorescence intensity of IAANS attached to Cys-84 of cTnC reports the Ca 2ϩ binding to the regulatory site, not the states of the thin filaments. The fluorescence intensity of IAANS on Cys-84 of cTnC in the Tn or Tn-Tm complex does not change signifi-cantly (ϳ5% increase) upon binding Ca 2ϩ to the regulatory site as shown in Fig. 2B. Therefore a slight excess amount of the Tn or Tn-Tm complex in our thin filament preparation did not interfere our measurements. The apparent Ca 2ϩ dissociation constant of the thin filament with wt-cTnI was 2.0 ϫ 10 Ϫ6 M (Fig. 5; Table 2). Thus the thin filament binds Ca 2ϩ about 10-fold weaker than the Tn complex or the Tn-Tm complex. Our finding of 10-fold weaker affinity of the thin filaments compared with the Tn complex is in excellent agreement with the previous Ca 2ϩ binding measurements (60,62). Also, unlike in the Tn-Tm complex, the cTnC in the thin filament binds Ca 2ϩ cooperatively (n H ϭ 1.58) as reported previously (29, 61, 63, 64), suggesting actin is required for the cooperative Ca 2ϩ binding to the protein assembly. All the mutants

TABLE 1 Summary for the effect of cTnI mutations on the Ca 2؉ -dependent actin-activated actoS1-ATPase activity
Data are presented as mean Ϯ S.E. Unpaired t tests were carried out for the maximum and basal activities. Note that the maximum ATPase rates with mutant cTnIs are not significantly different from the rate with wt-cTnI, whereas the basal level of ATPase rate with R145G and R145W mutations are significantly different from the rate with wt-cTnI.   tested in this study affected Ca 2ϩ binding to the reconstituted thin filaments significantly ( Table 2) and, importantly, in the same direction as the ATPase activity. The R145G mutation had the largest effects: the thin filament with the cTnI(R145G) mutation binds Ca 2ϩ with K d ϭ 7.5 ϫ 10 Ϫ7 M, which corresponds to ⌬⌬G ϭ Ϫ0.58 Kcal. The R192H mutation had the smallest effects (⌬⌬G ϭ Ϫ0.20 Kcal). These observations indicate that all the mutations studied in this report either stabilize the Ca 2ϩ -bound form of the thin filaments or destabilize the Ca 2ϩ -free form of the thin filaments. Since 1) none of the mutations destabilize the Ca 2ϩ -free state of Tn or Tn-Tm and (2) the inhibitory region and the second actin-Tm site interact with actin in the absence of Ca 2ϩ , our Ca 2ϩ binding data strongly indicate that all the mutations impair the interaction between cTnI and actin in the absence of bound Ca 2ϩ at the regulatory site of cTnC. It should be mentioned that Tm may share the interface for cTnI binding with actin as suggested previously (65). The apparent affinities of the thin filaments for Ca 2ϩ from this measurement are slightly higher than those obtained from the acto-S1 ATPase activity measurements. This is likely to be due to the IAANS labeling, since it was shown previously that IAANS labeling of Cys-84 of cTnC slightly sensitizes myofilament activity to Ca 2ϩ (66). Ca 2ϩ binding to the thin filament could be measured by the fluorescence emission intensity change of IAANS attached to Cys-35 of mutant cTnC(C84S) (Fig. 2A). Previous studies (e.g. Ref. 64), however, indicated that the IAANS labeling of Cys-35 significantly increases the affinity for Ca 2ϩ of the thin filaments. We found that the dissociation constant thus obtained for the thin filament was almost identical to that of the Tn complex, which is not consistent with other studies (60,62). Therefore, in this study, we only used the data obtained from IAANS attached to Cys-85 of cTnC(C35S) for the thin filaments.

DISCUSSION
Data described here are the first to report the Ca 2ϩ binding properties of the cTn complex, the cTn-Tm complex, and the reconstituted thin filament with cTnI(R145G). These data demonstrated the novel finding that the mutation actually increases the Ca 2ϩ binding affinity of the regulatory site of cTnC in the thin filament. Ca 2ϩ sensitization is, at least, due to the destabilization of Ca 2ϩ -free state of the thin filament.

TABLE 2 Summary for the effect of cTnI mutations on the Ca 2؉ binding properties for the different protein complexes
The numbers are expressed as the mean values Ϯ S.E. from four to nine experiments. Unpaired t tests were carried out for the dissociation constants K d and Hill coefficients (nH). Note all the mutants studied increase the Ca 2ϩ affinity of the thin filaments compared with the thin filaments with wt-cTnI. Also the thin filaments with mutant cTnI bind Ca 2ϩ with less cooperativity.   The same mechanism is accounted for the Ca 2ϩ sensitization of the myofilament activity caused by other mutations found in the inhibitory region and the second actin-Tm site. We did not find a difference between HCM-and RCM-related mutations. We, however, found that the cardiomyopathy-related mutations in the inhibitory region and those from the second actin-Tm site differently affect the equilibrium of thin filament states.
In resting conditions (diastole), TnI interacts with actin-Tm through at least two regions to turn off the thin filaments. One of these sites is the inhibitory region. The minimum inhibitory region (residues 104 -115 of rabbit fsTnI and residues 137-148 of human cTnI) is rich in basic amino acids, and its sequence is highly conserved. Grand et al. (67) were the first to show that Arg residues in the inhibitory region are involved in the interaction with actin. They titrated the peptide corresponding to the inhibitory region of fsTnI with actin and observed the perturbation of the proton NMR signals from Arg side chains. Using a series of synthetic peptides, van Eyk and Hodges (68) concluded that amino acid residues Arg-112, Arg-113, and Arg-115 of the minimum inhibitory region of rabbit fsTnI are important for the inhibition of acto-S1 ATPase activity: when one of these Arg residues was replaced by Gly, the synthetic peptide, corresponding to residues 104 -115 of rabbit fsTnI, significantly impaired its ability to inhibit ATPase activity, compared with the peptide with wild-type sequence. Although both of these studies were carried out using short peptides, their observations are consistent with our finding that the mutations found in one of these Arg residues, Arg-112, which corresponds to Arg-145 of hcTnI, into Gly or Trp result in the impaired interactions of the Tn complex with actin-Tm. Recently Patchell et al. (69) demonstrated that the addition of the peptide, which corresponds to the hcTnI minimum inhibitory region, to actin-Tm resulted in the dissociation of actin-binding peptides derived from myosin from the actin-Tm. They concluded that the inhibitory region of TnI interacts with actin in the absence of Ca 2ϩ in a way that causes the structural change in actin to prevent stable myosin-association. Arg-145 is likely to be one of the key residues that induce such conformational transitions in actin molecules, since both Gly and Trp mutations of this position impaired the inhibitory activity of TnI significantly.
cTnI mutations from the inhibitory region affected the basal level of acto-S1 ATPase activity; those from the second actin-Tm site did not. Thus it is apparent that the mutations from the inhibitory region and those from the second actin-Tm site affect the equilibrium of the thin filament differently. Yet all of the mutations studied here increased the affinities of the thin filaments for Ca 2ϩ . These observations can be interpreted as follows: whereas the mutations from the inhibitory region shift the equilibrium of the thin filaments from B-to C-and M-states, those from the second actin-Tm site shift the equilibrium of the thin filaments from B-to C-states, but not to M-state. In other words, the release of the inhibitory region from actin-Tm is necessary for the transition of the thin filaments from "off "-states (i.e. B-and C-states) to "on" state (M-state). Thus the second actin-Tm site may serve to increase the local concentration of TnI at the actin filament and thus promotes the effective interaction of the inhibitory region with actin-Tm or to transmit the structural change induced by the interaction with the inhibitory region with actin-Tm surface to nearby actin molecule, rather than inhibit the actin-myosin interaction actively. The detailed analysis of the C-terminal region of fsTnI using a series of deletion mutations; Ramos (51) found that the inhibitory activities of deletion mutants co-related with the length of the deletion from the C terminus of fsTnI: the longer deletion, the less inhibition of ATPase activity. This is consistent with the report by Rarick et al. (49) with cTnI. The second actin-Tm site may be relatively scattered along the TnI sequence. Also this explains why the different regions of TnI have been reported as a second actin-Tm site from different laboratories. Recently Murakami et al. (70) determined the structure of the C-terminal domain of fsTnI and modeled it into a three-dimensional cryo-EM map of the thin filament. Their model structure indicates the extensive interaction between actin and the C-terminal domain of TnI. Their model also indicated that the side chain of Asp-190 is directly involved in the interaction with the amino acid residue(s) in the DNase I loop of actin. On the other hand, the side chain of Arg-192 is not directly involved in the interaction with actin, which is consistent with our observation that R192H mutation did not cause a large increase of Ca 2ϩ binding affinity of the thin filaments. Recently Yumoto et al. (71) reported that one of the RCM-related mutations found in the C-terminal mobile domain of cTnI (K178E) induced a subtle localized structural perturbation around the mutated residue in an isolated C-terminal part of cTnI (residues 129 -210). A reversal of the charged state of this position could cause a decreased affinity for actin without disrupting a mobile domain structure significantly. It should be mentioned that the C-terminal mobile domain seems to interact with the core domain of the Tn complex, possibly the N-terminal domain of TnC in the presence of Ca 2ϩ (70,72), although physiological meaning of this interaction remains to be solved. More detailed study is needed to clarify the functional role of the second actin-Tm site of TnI.
Our observation that mutant cTnIs induce increased Ca 2ϩ binding to thin filaments has implications with regard to induction of arrhythmias associated with the sudden death in cardiomyopathies. There is compelling evidence that myofilament bound Ca 2ϩ is released by local mechanical or ischemic damage of the myocardium (73). These Ca 2ϩ ions have been demonstrated to induce Ca 2ϩ waves that trigger arrhythmic activity. Increases in extracellular Ca 2ϩ , which increases Ca 2ϩ binding to thin filaments, exacerbated the induction of arrhythmias. Our results show that increased Ca 2ϩ binding to thin filaments occurs with incorporation of mutant TnIs into thin filament proteins, and it is apparent under conditions of mechanical non-uniformity or ischemia, which are both likely to occur with disease, that this increased bound Ca 2ϩ may lead to an increase in Ca 2ϩ released from the myofilaments thereby amplifying the threat for induction of arrhythmias.