The C Terminus of Cardiac Troponin I Is Essential for Full Inhibitory Activity and Ca2+ Sensitivity of Rat Myofibrils*

Although the C terminus of troponin I is known to be important in myofilament Ca2+ regulation in skeletal muscle, the regulatory function of this region of cardiac troponin I (cTnI) has not been defined. To address this question, the following recombinant proteins were expressed in Escherichia coli and purified: mouse wild-type cTnI (WT cTnI; 211 residues), cTnI-(1–199) (missing 12 residues), cTnI-(1–188) (missing 23 residues), and cTnI-(1–151) (missing 60 residues). The inhibitory activity of cTnI and the mutants was tested in myofibrils, from which cTnI·cTnC was extracted by exchanging endogenous cardiac troponin with exogenous cTnT causing the Ca2+ sensitivity of the myofibrils to be lost. Addition of increasing amounts of exogenous WT cTnI or cTnI-(1–199) to cTnT-treated myofibrils at pCa 8 caused a concentration-dependent inhibition of the maximum ATPase activity. However, cTnI-(1–188) and cTnI-(1–151) inhibited this activity to about 75% and 50% of that of the WT cTnI, respectively. We also formed a complex of either WT cTnI or each of the mutants with cTnC, reconstituted the complex into the cTnT-treated myofibrils, and measured the Mg2+-ATPase activity as a function of pCa. We found that the cTnI-(1–188)·cTnC complex only partially restored Ca2+ sensitivity, whereas the cTnI-(1–151)·cTnC complex did not restore any Ca2+sensitivity. Each cTnI C-terminal deletion mutant was able to bind to cTnC, as shown by urea-polyacrylamide gel-shift analysis and size exclusion chromatography. Each mutant also co-sedimented with actin. Our results indicate that residues 152–199 (C-terminal to the inhibitory region) of cTnI are essential for full inhibitory activity and Ca2+ sensitivity of myofibrillar ATPase activity in the heart.

The transition of heart muscle from diastole to systole involves a Ca 2ϩ -dependent process in which an inhibitory protein, cTnI, 1 plays a key role in switching on the reaction of myosin cross-bridges with actin. A current model for the switching mechanism has been derived largely from studies with fast skeletal TnI and is as follows. During diastole, cTnI interacts tightly with actin and contributes to the inhibition of the actin-cross-bridge reaction. During systole, with Ca 2ϩ binding to cTnC, there is an increased affinity of cTnI for cTnC, which results in weakening of cTnI binding to actin. This leads to changes in Tm position, resulting in activation of contraction through promotion of the transition of cross-bridges from blocked or weak binding states to strong force-generating states (1).
In the case of heart muscle, Ca 2ϩ signaling appears different from that of fast skeletal muscle, and little is known about regions of cTnI that are essential for the Ca 2ϩ switch. An anti-parallel interaction between cTnI and cTnC has been demonstrated (2). In the anti-parallel arrangement, an interaction of the N-terminal region of cTnI with cTnC acts as an anchor maintaining the two proteins in the correct spatial orientation. The C-terminal region of cTnI is thought to bind to the N terminus of cTnC, which contains the low affinity Ca 2ϩ -specific site (site II) (2,3). Ca 2ϩ binding to site II on cTnC is essential for the regulation of cardiac contraction (4 -6). Interestingly, it has been shown recently by heteronuclear, multidimensional NMR spectroscopy that upon transition from the apo-to the Ca 2ϩ -saturated states of the N-domain of cTnC, fewer hydrophobic residues are exposed in cTnC than in fsTnC (7,8). This hydrophobic region of cTnC is thought to bind to cTnI (9). The finding that the structure of the regulatory N-domain of cTnC is significantly more compact relative to fsTnC (7) may be related to differences in how cTnI and fsTnI interact with TnC and ultimately to differences in the regulation of cardiac versus skeletal muscle contraction (6,10).
Moreover, it is known that TnI contains an inhibitory region (residues 139 -150 in mouse cTnI) that binds to both TnC and actin, but not to both simultaneously (11,12). This region is believed to be largely responsible for the ability of TnI to inhibit actomyosin ATPase activity and constitutes a key part of the molecular switch turning on the thin filament. Although structural and modulatory functions have been identified within the unique cardiac-specific N-terminal region of cTnI, the functional role of its C-terminal region has yet to be delineated.
Our laboratory is using recombinant DNA technology to investigate structure-function relations of cardiac TnI. The C terminus of cTnI is highly conserved among the TnI isoforms and its binding to the N terminus of cTnC containing the low affinity Ca 2ϩ -specific site indicates it may be important for Ca 2ϩ -dependent regulation of cardiac muscle contraction. Although it has been shown that the C terminus of fsTnI appears to be important for Ca 2ϩ sensitivity in myofilaments in skeletal muscle (13,14), there is no evidence for such a role in cardiac muscle. In this study, we generated three deletion mutants of cardiac TnI to examine the function of the C-terminal domain of cTnI in the cardiac myofilament. Our results indicate that regions of cTnI C-terminal to the inhibitory region are essential for full inhibitory activity and Ca 2ϩ sensitivity of cardiac myofilaments.
The conditions for the PCR reaction were described previously (17). The PCR products were cloned directly into a pCR™ cloning vector (Invitrogen, San Diego, CA). Several clones were selected for DNA sequence analysis (18). The cTnI and mutant cTnI DNAs were excised from the cloning vector using NcoI and BamHI and isolated by agarose gel electrophoresis. After purification from gels, the fragment was ligated into the NcoI/BamHI site of the pET-3d expression vector (Novagen, Inc., Madison, WI). This DNA was used to transform BL21(DE3) cells (Novagen, Inc.) before expression of WT cTnI or the deletion mutants of cTnI.
A single transformant or frozen stock of transformed cells was used to seed 1 liter of LB medium containing 50 g/ml of carbenicillin. The culture was grown overnight at 37°C. The cells were centrifuged at 5000 ϫ g for 10 min, and the supernatant fraction removed. The cells were stored overnight at Ϫ80°C. The pellets were resuspended in ice-cold 20 mM Tris, pH 8.0, 6 M urea, 10 mM EDTA, 1 mM DTT, 1 mM phenylmethylsulfonyl fluoride (10 ml/liter of cell culture) and sonicated on ice for approximately 20 min. The cell lysate was centrifuged at 14,000 ϫ g for 15 min at 4°C and loaded onto a CM-Sepharose column equilibrated with Buffer B (20 mM Tris, pH 8.0, 6 M urea, 1 mM EDTA, 1 mM DTT). The WT cTnI and mutant cTnIs were eluted with a linear gradient of 0 -0.5 M NaCl in the presence of Buffer B. The WT cTnI and mutant cTnIs were further purified by affinity chromatography on an Affi-Gel-15 (Bio-Rad) cTnC column equilibrated with 50 mM Tris-HCl, pH 8.0, 6 M urea, 1 mM CaCl 2 , 1 mM DTT. The WT cTnI and mutant cTnIs were eluted with 50 mM Tris-HCl, pH 8.0, 6 M urea, 2 mM EDTA, 1 M NaCl, 1 mM DTT. The preparations of WT cTnI and the mutant cTnIs were analyzed on SDS-polyacrylamide gels (15%) (19). The calculated molecular weights of WT cTnI, cTnI-(1-199), cTnI-(1-188), and cTnI-(1-151) were 24,260, 22,925, 21,656, and 17,407, respectively.
Expression and Purification of cTnC-The contruction of a pET-3d plasmid construct containing the complete coding sequence for human cTnC and the expression and purification of the cTnC was performed as described by Pan and Johnson (20). Purity of the cTnC was examined by SDS-PAGE (15%).
Myofibrillar Preparation-Hearts were immediately excised from ether-anesthetized male Sprague-Dawley rats (225-250 g) and washed in saline. The atria were trimmed and myofibrils were prepared from the ventricles as described previously (21). Purified myofibrils were resuspended in Buffer A and total protein concentration was determined (22). Myofibrillar proteins were visualized on SDS-polyacrylamide gels (12.5%).

Removal of Troponin by Excess cTnT and Reconstitution with cTnI⅐cTnC
Complexes-This method removes endogenous cardiac troponin from the myofibrils by treatment with excess exogenous cTnT. The treated myofibrils contain the exogenous cTnT, but removal of cTnI and cTnC makes the myofibrils insensitive to Ca 2ϩ . The procedure was originally suggested by Hatakenaka and Ohtsuki (25) and was performed as described by Rarick et al. (26). Purified bovine cTnT was dissolved in Buffer D (20 mM MOPS, pH 6.5, 250 mM KCl, 5 mM EGTA, 5 mM MgCl 2 , 0.5 mM DTT, 0.1 g/ml pepstatin A). The cTnT, at a concentration of 0.4 mg/ml, was added to centrifuged rat myofibrils (4 mg) in a 1:10 weight ratio of cTnT:myofibrils. The volume was adjusted to 2.5 ml with Buffer D, and the mixture was incubated at 25°C for 60 min with gentle mixing every 2 min. The cTnT-treated myofibrils were centrifuged (2,000 ϫ g; 10 min), and the pellet was resuspended in 2.5 ml of Buffer C containing 2 mg of WT cTnI, 2 mg of cTnI mutants, or 4 mg cTnI⅐cTnC complex. This mixture was incubated at 25°C for 75 min with gentle mixing at 2-min intervals. The reconstituted myofibrils were centrifuged (2,000 ϫ g; 10 min) and the pellets resuspended in 2 ml of Buffer A, recentrifuged, and suspended in a final volume of 1 ml of Buffer A. The protein concentration was determined by the Lowry method (22).
ATPase Measurements-ATPase activity was determined by measuring inorganic phosphate release using a modification of the methods of White (27) and Dobrowolski et al. (28). Assays were carried out in 96-well microtiter plates at 30°C in an incubator. The assay conditions varied and are described in the figure legends. The total volume was 70 l, and reactions were initiated by the addition of 2 mM ATP. Reactions were stopped after 10 min by the addition of 25 l of 13.4% SDS, 0.12 M EDTA. Phosphate liberation was linear over the time course of the experiment. The amount of inorganic phosphate released was determined colorimetrically by the addition of 200 l of 0.5% FeSO 4 , 0.5% ammonium molybdate in 0.5 M H 2 SO 4 . After 20 min, the absorbance was measured at 650 nm with a microtiter plate reader (Bio-Tek Instruments, Winooski, VT). The total concentrations of buffers, salts, and ATP needed to give the desired concentrations of free Mg 2ϩ , free Ca 2ϩ , KCl, EGTA, MgATP 2Ϫ and ionic strength at pH 7.0 in the reaction mixtures were computed using constants summarized in Fabiato and Fabiato (29).
Statistics-Repeated measurements were at least n ϭ 3, and are presented as mean Ϯ S.E. Measured relationships between pCa and ATPase activity were fit to the Hill equation using nonlinear leastsquares regression to derive the pCa 50 and Hill coefficient (n) (Inplot curve fitting software, GraphPAD Software, Inc., San Diego, CA). Statistical differences of pCa 50 values were analyzed by a paired Student's t test with significance set at p Ͻ 0.05.

RESULTS
To localize protein domains in the C terminus of cTnI that are important for Ca 2ϩ regulation of cardiac muscle contraction, we generated three deletion mutants of mouse cardiac cTnI, which were compared with the WT cTnI in functional assays. The amino acid sequence of the mouse WT cTnI consists of 211 residues and is shown in Fig. 1 (upper panel). The WT cTnI and the cTnI C-terminal deletion mutants are shown schematically in Fig. 1 (lower panel). The cTnI-(1-199), cTnI-(1-188), and cTnI-(1-151) mutants have 12, 23, and 60 residues deleted from the C terminus, respectively. The cTnI-(1-151) mutant terminates 1 amino acid residue past the inhibitory region of cTnI.
The mouse WT cTnI and each cTnI C-terminal deletion mutant were expressed in Escherichia coli and purified as described under "Experimental Procedures" to Ͼ99% purity (Fig.  2). The gel in Fig. 2 shows that the recombinant mouse WT cTnI (lane 3) migrated slightly slower than native cardiac TnI (lane 2) purified from bovine tissue, possibly because of the lack of acetylation of the N-terminal methionine.
We examined the binding of each cTnI C-terminal deletion mutant to cTnC by visualization of cTnI⅐cTnC complex formation on urea polyacrylamide gels and size exclusion chromatography. As shown in Fig. 3, WT cTnI and each C-terminal deletion mutant of cTnI formed a binary complex with cTnC on urea polyacrylamide gels in the presence of Ca 2ϩ , but not in the presence of EGTA. We could not detect any differences between WT cTnI and each cTnI mutant in their ability to bind to cTnC in the denaturing conditions used for the urea PAGE analysis. This suggests that the structural interaction between the N terminus of cTnI and the C terminus of cTnC is the major contributor to formation of the cTnI⅐cTnC complex. We also tested whether WT cTnI and each cTnI C-terminal deletion mutant formed a complex with cTnC under non-denaturing conditions by size exclusion chromatography. At physiological salt concentrations, after passage through a HPLC gel filtration column, WT cTnI and each cTnI C-terminal deletion mutant formed a stable binary complex with cTnC (data not shown).
We also examined the ability of WT cTnI and the C-terminal deletion mutants of cTnI to bind to F-actin. F-actin, cTnI or cTnI deletion mutants, and Tm were combined, and the binding of cTnI or the cTnI deletion mutants to F-actin was assayed by centrifugation in an Airfuge. The supernatant fractions and pellets were analyzed by SDS-polyacrylamide electrophoresis.
WT cTnI and each cTnI C-terminal deletion mutant co-sedimented with F-actin or F-actin-Tm (data not shown). This was only a qualitative technique, but it indicated that there was binding to F-actin even after deletion of up to 60 residues from the C terminus of cTnI.
The ability of the cTnI C-terminal deletion mutants to bind to F-actin-Tm suggested that they retained inhibitory activity. We directly tested their inhibitory capabilities by adding WT cTnI or the cTnI C-terminal deletion mutants to rat myofibrils lacking endogenous cTnI and cTnC. The cTnI⅐cTnC complex was extracted by modification (26) of the method by Hatakenata and Ohtsuki (25), which uses excess exogenous bovine cTnT to displace the troponin complex from the myofibrils. Fig.  4, which shows SDS-PAGE analysis of control and cTnT-extracted myofibrils, demonstrates a decrease in the levels of endogenous cTnI and cTnC, and the presence of bovine cTnT without obvious alteration of other myofilament protein levels (Fig. 4, lanes 1 and 2). In Fig. 4, we also show profiles of cTnT-extracted myofibrils after reconstitution with WT cTnI or each cTnI C-terminal deletion mutant complexed with cTnC (lanes 3-6). It is difficult to visualize cTnC with Coomassie Blue staining, but in a previous publication (26), we have used alkaline urea polyacrylamide gel electrophoresis to show the removal and restoration of cTnC under these conditions.
We analyzed the effect of C-terminal deletion of cTnI on its ability to inhibit the Mg 2ϩ -ATPase activity of myofibrils lacking cTnI⅐cTnC. Fig. 5 shows the effect of increasing concentrations of the various forms of cTnI on the unregulated ATPase activity of the myofibrils. In the absence of added cTnI, as expected, the ATPase rate was independent of Ca 2ϩ and similar to that obtained in native myofibrils at maximal activity (pCa 4.9). WT cTnI reduced this activity to 25% of maximum with the inhibition reaching a plateau at 1.0 M WT cTnI. The cTnI-(1-199) mutant had an inhibitory effect similar to WT cTnI. However, the inhibitory capacity of the cTnI-(1-188) mutant was reduced to 75% of that of WT cTnI and reached a plateau at a concentration of 1 M. Inhibition by the cTnI-(1-151) mutant was about 50% of that of WT cTnI and attained a plateau at 2 M. Even a concentration of 12 M could not increase the inhibitory capacity of this mutant beyond 45% of that of WT cTnI (data not shown). Thus, loss of 23 residues at the C terminus of cTnI results in some loss of inhibitory activity that is further decreased after removal of an additional 37 residues.
We next tested the ability of cTnC to reverse the inhibitory effect of both WT cTnI and the cTnI C-terminal deletion mutants. The molar ratios of cTnC to cTnI tested varied from 0.5 to 4, and measurements were made in the absence and presence of Ca 2ϩ . As Fig. 6 illustrates, addition of cTnC at a 1:1 molar ratio to cTnI in the presence of Ca 2ϩ reversed the inhibition caused by both WT cTnI and the cTnI-(1-199) mutant. The inhibition caused by cTnI-(1-188) was released by cTnC in the presence of Ca 2ϩ with restoration to within 20% of that for WT cTnI. However, the inhibition of maximum ATPase activity caused by cTnI-(1-151) was not released by the addition of cTnC (in the presence of Ca 2ϩ ) even at a molar ratio of 4:1 (cTnC:cTnI).
We also formed a complex of WT cTnI or cTnI C-terminal deletion mutants with cTnC before reconstitution into cTnTextracted myofibrils, and measured the Mg 2ϩ -ATPase activity as a function of pCa. Fig. 7 shows that under these experimental conditions, the Ca 2ϩ sensitivity was restored to that observed for control myofibrils with WT cTnI⅐cTnC and cTnI-(1-199)⅐cTnC. On the other hand, reconstitution with the cTnI-(1-188)⅐cTnC complex only partially restored Ca 2ϩ sensi-tivity (data not shown), and we observed a significant leftward shift in the pCa 50 (Fig. 7). Data for the cTnI-(1-151)⅐cTnC complex is not shown in Fig. 7 because after reconstitution with this complex, no Ca 2ϩ sensitivity was observed. These results indicate that in cardiac TnI, residues 152-188 (C-terminal to the inhibitory region) are essential for Ca 2ϩ sensitivity of the myofibrils.

DISCUSSION
Our results provide the first evidence that the C-terminal region of cardiac TnI, downstream from the inhibitory region, is essential for the Ca 2ϩ -dependent regulation of cardiac myofilament activation. The functional significance of the C-terminal domain of cTnI fits with evidence of an anti-parallel alignment between cTnI and cTnC in which the C-terminal region of cTnI binds to the N terminus of cTnC. This insight makes an important contribution to our objective to determine unique aspects of the detailed mechanism by which Ca 2ϩ switches on cardiac myofilaments.
Our findings suggest a need to rethink the relative contribution of the central inhibitory region of cTnI to the regulation of force. In previous studies, the primary focus has been the inhibitory region of cTnI. An 11-amino acid peptide corresponding to the sequence of the inhibitory region of cTnI has the capacity to inhibit ATPase activity in vitro (12,30,31), although not to the same extent as full-length cTnI. Surprisingly, cardiac fiber bundles reconstituted with the cTnI inhibitory peptide are able to undergo sequential contraction-relaxation cycles (32). The importance of the inhibitory peptide in skeletal muscle is also supported by the inability of mutant fsTnI, missing the inhibitory region, to inhibit ATPase activity in vitro (33). However, our data indicate that in addition to the inhibitory region, two additional sites in cTnI, located C-terminal to the inhibitory region, are essential for the expression of maximum inhibition.
From our analysis of the effect of C-terminal deletions on the ability of cTnI to inhibit ATPase activity (Fig. 5), we deduce that the inhibitory region inhibits only about 50% of the ATPase activity. Two additional sites, between residues 152-188 and residues 189 -199, contribute approximately 25% each to the inhibition of ATPase activity, as shown by reconstitution with mutants cTnI-(1-188) and cTnI- , respectively (Fig.  5). One interpretation of these data is that there are two additional actin-Tm binding sites located in the C-terminal domain of cTnI. Farah et al. (13) and Triplet et al. (34) also postulate the presence of an additional actin-Tm binding site located on the C-terminal side of the inhibitory region in fsTnI (residues 136 -148). The second putative actin-Tm binding site in cTnI, comprising 10 amino acids (cTnI residues 189 -199; fsTnI residues 157-167), does not appear to function in skeletal muscle. On the other hand, based on evidence that there is complete absence of inhibitory activity with removal of the inhibitory region (residues 104 -120) in fsTnI (13,33), we also cannot exclude the possibility that binding of the inhibitory region to actin-Tm induces conformational changes in actin-Tm that expose sites, which can then bind to actin-Tm binding sites located in the C terminus of cTnI.
In a model of control of thin filament activation, Lehrer (35) suggested that cTnI may be involved in stabilizing a "blocked state" of the thin filament by sterically blocking the actinmyosin interaction, and that Ca 2ϩ induces movement of cTnI away from actin. TnI would need to be in an elongated config-uration and span at least several actin monomers to be involved in a blocking function. In a model of Ca 2ϩ -saturated fsTnI-fsTnC complex derived from small-angle x-ray scattering data, fsTnI adopts an extended conformation in the presence of fsTnC and Ca 2ϩ that is about 115 Å long (36). An actin monomer is estimated to be about 40 Å in diameter (37); thus, TnI may span 2-3 monomers of actin. Residues C-terminal to the inhibitory region may be critical for extension of TnI to stabilize a blocked state of the thin filament, especially in the case of cTnI, which has an additional 32 amino acids at the N terminus.
In addition to the importance of residues C-terminal to the inhibitory region of cTnI in expression of maximal inhibition of ATPase activity, this C-terminal region of cTnI appears to be essential for Ca 2ϩ -dependent regulation of cardiac myofilament contraction. We observed that the the cTnI-(1-188) mutant had impaired ability to regulate Ca 2ϩ sensitivity in myofibrils and the cTnI-(1-151) mutant had lost the ability to regulate Ca 2ϩ sensitivity (Fig. 6). This loss of Ca 2ϩ -dependent control of the myofilament after deletion of portions of the C terminus of cTnI implies an alteration in the interaction of cTnI and cTnC, most likely through removal of a binding site for cTnC in the C-terminal region of cTnI. Evidence, derived largely from studies on fast skeletal myofilaments, indicates multiple sites of interaction between fsTnI and fsTnC (13,38,39). Experiments have revealed that fsTnC may bind C-terminal to the inhibitory region of fsTnI. A fsTnI peptide, fsTnI 96 -148, which extends about 30 residues C-terminal to the inhibitory region (96 -114), had tighter binding to fsTnC than the inhibitory peptide, fsTnI 96 -114 (40). Kobayashi et al. (41) found that residues 132-141 of fsTnI (corresponding to cTnI 166 -174 ) cross-linked to Cys-12 in the N terminus of fsTnC. Moreover, an N-terminal part of fsTnC (sTnC 46 -78 ) not only cross-linked to the inhibitory region, fsTnI 96 -114 , but to residues 122-152 of fsTnI (cTnI 156 -186 ). It has also been demonstrated that in the presence of Ca 2ϩ , Cys-133 of fsTnI (cTnI 167 ) moves 0.7 nm toward Cys-98 in the central helix of fsTnC (42, 43) and 1.5 nm away from actin (44). This latter finding underscores the importance of the C-terminal part of TnI in the Ca 2ϩ -switch mechanism. In our experiments, evidence for a cTnC binding site C-terminal to the inhibitory region comes from the finding that cTnC was not able to overcome the inhibition of maximal ATPase activity by the cTnI-(1-151) mutant in the presence of Ca 2ϩ (Fig. 6).
The loss of inhibitory function after deletion of either 23 (cTnI-(1-188)) or 60 (cTnI-(1-151)) residues (Fig. 5) suggests a weakened interaction of cTnI with actin-Tm, which would promote the availability of actin-Tm for reaction with myosin. A weakened cTnI-actin-Tm interaction could also explain the apparent increase in Ca 2ϩ sensitivity seen with the cTnI-(1-188) mutant (Fig. 7). The evidence for a proposed single cTnC binding domain downstream from the inhibitory region is based on the data presented in Fig. 6. Although inhibition of Mg 2ϩ -ATPase activity was reduced in myofibrils reconstituted with the cTnI-(1-188) mutant, addition of cTnC and Ca 2ϩ almost completely restored the Mg 2ϩ -ATPase activity. However, addition of cTnC and Ca 2ϩ to myofibrils reconstituted with the cTnI-(1-151) mutant could not reverse the inhibition of the Mg 2ϩ -ATPase activity (Fig. 6). These results indicate the presence of a Ca 2ϩ -dependent cTnC binding domain within residues 152-188. Thus, this region contains both actin-Tm and cTnC binding domains. The existence of these additional actin-Tm and cTnC binding sites C-terminal to the inhibitory region add to the complexity of thin filament protein interactions and may also be an important component of the Ca 2ϩ regulation of thin filament activation.  -(1-188)⅐cTnC. Ca 2ϩ -dependent Mg 2ϩ -ATPase activity of the control (q) myofibrils as well as myofibrils reconstituted with WT cTnI⅐cTnC (OE), cTnI-(1-199)⅐cTnC (f), and cTnI-(1-188)⅐cTnC (ࡗ) was measured. The final assay conditions were the same as in Fig. 5 except that CaCl 2 was varied to obtain the range of pCa values. Values of ATPase activity at pCa 4.9 for each myofibril preparation were taken as 100% after subtracting the basal rate. The pCa 50 for myofibrils reconstituted with WT cTnI⅐cTnC (OE; 6.21 pCa units) and with cTnI-(1-199)⅐cTnC (f; 6.27 pCa units) were not significantly different. The pCa 50 for myofibrils reconstituted with cTnI-(1-188)⅐cTnC (ࡗ; 6.44 pCa units) was significantly different from that of myofibrils reconstituted with WT cTnI⅐cTnC (p ϭ 0.032).
Although our study has focused on the role of the C-terminal domain of cTnI in Ca 2ϩ regulation of cardiac myofilaments, it is also important to emphasize that cTnI contains an unique N-terminal extension that undergoes covalent modification by protein kinase A, which decreases the pCa 50 for activation of the myofilaments. The mechanism involves phosphorylation of cTnI at Ser 23,24 and an increase in the rate of dissociation of Ca 2ϩ from the N terminus of cTnC (45). This effect appears to involve global changes in the cTnI molecule that signal an altered interaction between the C-terminal domain of cTnI with cTnC and/or actin-Tm. Dong et al. (46) have demonstrated this global change by using fluorescence resonance energy transfer to determine a decrease in mean distance between Nand C-terminal regions of cTnI induced by protein kinase A phosphorylation. Moreover, Chandra et al. (47) have demonstrated that phosphorylation of the N terminus of cTnI by protein kinase A is able to depress Ca 2ϩ binding to an Nterminal fragment of cTnC (cTnC-(1-89). Thus, covalent modifications at the N terminus of cTnI are sensed by the C terminus of cTnI, which appears able to modulate the interaction of the N-terminal domain of cTnC with Ca 2ϩ .
In conclusion, our results contribute new information to the detailed understanding of the mechanism by which cTnI participates in the Ca 2ϩ switch of the heart, and are important in the context of both cardiac physiology and patho-physiology. There is now evidence that mutations in the C-terminal region of the cTnI molecule are causal in familial hypertrophic cardiomyopathy in the Japanese population (48). Our results suggest that such mutations in cTnI could have severe effects on the control of cardiac myofilaments by Ca 2ϩ . Moreover, the region of cTnI missing in the cTnI-(1-151) mutant, which interacts with the N terminus of cTnC, could be important with regard to rational design of pharmacological agents useful in heart failure.