STRUCTURAL KINETICS OF CARDIAC TROPONIN C MUTANTS LINKED TO FAMILIAL HYPERTROPHIC AND DILATED CARDIOMYOPATHY IN TROPONIN COMPLEXES

The key events in regulating cardiac muscle contraction involve Ca 2+ binding to and release from cTnC (troponin C) and structural changes in cTnC and other thin filament proteins triggered by Ca 2+ movement. Single mutations L29Q and G159D in human cTnC have been reported to associate with familial hypertrophic and dilated cardiomyopathy, respectively. We have examined the effects of these individual mutations on structural transitions in the regulatory N-domain of cTnC triggered by Ca 2+ binding and dissociation. This study was carried out with a double mutant or triple mutants of cTnC, reconstituted into troponin with tryptophanless cTnI and cTnT. The double mutant, cTnC(L12W/N51C) labeled with 1,5-IAEDANS at Cys51, served as a control to monitor Ca 2+ -induced opening and closing of the N-domain by Förster resonance energy transfer (FRET). The triple mutants contained both L12W and N51C labeled with 1,5-IAEDANS, and either L29Q or G159D. Both mutations had minimal effects on the equilibrium distance between Trp12 and Cys51-AEDANS

The key events in regulating cardiac muscle contraction involve Ca 2+ binding to and release from cTnC (troponin C) and structural changes in cTnC and other thin filament proteins triggered by Ca 2+ movement. Single mutations L29Q and G159D in human cTnC have been reported to associate with familial hypertrophic and dilated cardiomyopathy, respectively. We have examined the effects of these individual mutations on structural transitions in the regulatory N-domain of cTnC triggered by Ca 2+ binding and dissociation. This study was carried out with a double mutant or triple mutants of cTnC, reconstituted into troponin with tryptophanless cTnI and cTnT. The double mutant, cTnC(L12W/N51C) labeled with 1,5-IAEDANS at Cys51, served as a control to monitor Ca 2+ -induced opening and closing of the N-domain by Förster resonance energy transfer (FRET). The triple mutants contained both L12W and N51C labeled with 1,5-IAEDANS, and either L29Q or G159D. Both mutations had minimal effects on the equilibrium distance between Trp12 and Cys51-AEDANS in the absence or presence of bound Ca 2+ . L29Q had no effect on the closing rate of the N-domain triggered by release of Ca 2+ , but reduced the Ca 2+ -induced opening rate. G159D reduced both the closing and opening rates. Previous results showed that the closing rate of cTnC N-domain triggered by Ca 2+ dissociation was substantially enhanced by PKA phosphorylation of cTnI. This rate enhancement was abolished by L29Q or G159D. These mutations alter the kinetics of structural transitions in the regulatory N-domain of cTnC that are involved in either activation (L29Q) or deactivation (G159D). Both mutations appear to be antagonistic toward phosphorylation signaling between cTnI and cTnC.
Contraction of cardiac muscle is activated by the binding of Ca 2+ to the Ca 2+ -binding subunit troponin C (cTnC) of the trimeric troponin complex. cTnC, the other two troponin subunits (troponin I (cTnI) and troponin T (cTnT)) and tropomyosin form the regulatory system of the contractile apparatus. These proteins are located in the thin filament. Contraction occurs when the myosin head in the thick filament interacts with actin causing the two filaments to slide past each other. The troponin complex in the thin filament regulates the actin-myosin interaction. Ca 2+ binding to the single site in the N-domain of cTnC initiates the activation process. cTnC has two globular regions, an N-terminal domain and a Cterminal domain. The N-domain has only one Ca 2+ binding site (site II) and the C-domain contains two binding site (binding sites III and IV). Ca 2+ binding to site II in the presence of cTnI induces an open conformation of the N-domain to expose a hydrophobic patch in the domain (1-3). Once exposed, the hydrophobic patch binds strongly with the regulatory region of cTnI. This strong interaction pulls cTnI away from actin and relieves the inhibition of actomyosin ATPase and muscle contraction. This regulation process is further finetuned with phosphorylation of cTnI serine 23 and serine 24 by protein kinase A (PKA) (4). This phosphorylation provides a mechanism to modulate heart activity to meet momentary hemodynamic demands (5). The modulation mechanism of PKA phosphorylation involves alterations in the Ca 2+ -cTnC interaction, thin filament regulation and actin-myosin interaction (6,7). Phosphorylation of Ser23/Ser24 of cTnI decreases myofibrillar Ca 2+ sensitivity (8)(9)(10)(11)(12)(13) and increases the rate at which Ca 2+ dissociates from cTnC (14). These changes may lead to a faster relaxation by increasing the rate of thin filament deactivation. This increase of relaxation rate is important for proper heart function because it allows adequate time for diastolic filling of the ventricles despite the raised heart rate during sympathetic stimulation. Since the transduction mechanism of PKA phosphorylation within thin filaments may involve phosphorylation-induced global conformational changes in cTnI (15)(16)(17)(18), or phosphorylation-induced direct impact on the cTnC-cTnI interaction, particularly the interaction between the N-terminus of cTnI and the N-domain of cTnC (19)(20)(21)(22)(23)(24)(25), any perturbation in the transduction mechanism can lead to functional impairment of heart function. One such perturbation is caused by cardiomyopathy-related mutations of thin filament proteins associated with hypertrophic cardiomyopathy (HCM) and dilated cardiomyopathy (DCM).
Familial hypertrophic cardiomyopathy (FHC) and DCM are the most common forms of primary cardiac diseases. Mutations associated with FHC and DCM have been identified in 10 different genes encoding cardiac sarcomeric proteins, including titin, myosin binding protein-C, βcardiac myosin heavy chain, the essential and regulatory myosin light chains, α-cardiac actin, αtropomyosin, cTnT, cTnI and cTnC (26,27). A single mutation of G159D and a double mutation E59D/D75Y in human cTnC have been reported to associate with DCM (28) and mutation L29Q has been identified to associate with FHC (29). Within cTnC, leucine 29 is located at the junction of helix A and the nonfunctional Ca 2+ binding loop I, and is important for the integrity of helix A and the pseudo Ca 2+ binding site I (30). Several lines of evidence have indicated that this region involves interactions with the cardiac specific N-terminal extension of cTnI where the two PKA phosphorylation sites are located, and these interactions are PKA phosphorylation dependent (19,23,24,31,32). Therefore, substituting a nonpolar residue with a polar residue in this region may potentially interrupt the cTnC-cTnI interaction and the Ca 2+ -induced structural transition of the cTnC N-domain. In an in vitro peptide array experiment, replacing Leu29 with glutamine weakened the interaction between cTnI and cTnC (32). Glycine at position 159 of cTnC is located at the third position from to the Cterminus. The C-domain of cTnC is generally considered to play a structural role in troponin since the exchange of bound Mg 2+ with Ca 2+ in the C-domain is too slow to participate in muscle regulation (33). However, there is evidence that the interaction of this region of cTnC with cTnI has a significant impact on the force of cardiac myofilament (34). Considering the anti-parallel structural arrangement of cTnC and cTnI in troponin, the C-domain of cTnC may interact with the N-terminal segment of cTnI where multiple PKA and PKC phosphorylation sites are located. Thus, replacing glycine with an acidic residue at position 159 may modify the interaction between this region of cTnC and cTnI.
In this study we investigated the effects of L29Q and G159D mutations in cTnC on the structural dynamics and kinetics of Ca 2+ -induced structural transition in the N-domain of cTnC, using reconstituted troponin preparations and a combination of FRET equilibrium and kinetic measurements. Our rationale is that, in addition to the potential structural effects, these mutations may also have a significant impact on the dynamics of the Ca 2+ -induced N-domain conformational changes of cTnC. Kinetic experiments showed that both mutations significantly altered the kinetics of structural transition and eliminated the inotropic effect of PKA phosphorylation of cTnI on the transition. These findings suggest an antagonistic role of these cTnC cardiomyopathy-related mutations of cTnC in cardiac thin filament regulation.

EXPERIMENTAL PROCEDURES
Protein Preparations -A recombinant tryptophanless cTnT mutant (W239F/W289F) and a recombinant tryptophanless cTnI mutant (W192F) were generated as in a previous report (35). Both proteins were expressed in BL21(DE3) cells (Invitrogen) under isopropyl-1-thio-Dgalactopyranoside induction. The expressed proteins were purified as previously described (36). Recombinant single cysteine and single tryptophan mutants, cTnC(L12W/N51C), cTnC(L12W/N51C/L29Q), and cTnC(L12W/ N51C/G159D), were generated from a cTnC cDNA clone from chicken slow skeletal muscle as previously reported (1). Protein purification and modification of the single cysteine residue with IAEDANS ((iodoacetamidoethyl)aminonaphthalene-1-ulfonic acid)) were performed with the procedures described in the previous report (35). The identities of all mutant proteins and labeled proteins were verified using electrospray mass spectrometric analysis. The biochemical activity of the labeled cTnC and tryptophanless cTnI and cTnT mutants was tested by Ca 2+ -dependent regulation of acto-S1 ATPase activity. Measurements were performed at 30 o C in 60 mM KCl, 5.6 mM MgCl 2 , 2 mM ATP, 30 mM imidazole (pH 7.0), 1 mM DTT and either 500 µM CaCl 2 for the +Ca 2+ state or 1 mM EGTA for the -Ca 2+ state. The protein concentrations used were 4.2 µM F-actin, 0.6 µM Tm, 0.6 µM Tn and 0.5 µM S1. The amounts of inorganic phosphate released were determined colorimetrically according to the reported procedure (37). Absorption was measured at 630 nm with a Beckman Du-640 spectrophotometer. Troponin complex was prepared by incubating 6 µM of cTnC mutants and 8 µM tryptophanless cTnI and 8 µM of tryptophanless cTnT on ice for 30 minutes before measurements were carried out. The final buffer contained 50 mM Mops, pH 7.0, 1 mM DTT, and 0.2 M KCl, plus various amounts of Ca 2+ or EGTA as required. All fluorescence measurements were carried out at 10.0 o C. PKA Phosphorylation of cTnI-Recombinant tryptophanless cTnI was phosphorylated by the catalytic subunit of PKA, using a cTnC affinity column as previously described (19). Briefly, purified cTnI mutant was loaded on a cTnC affinity column equilibrated in 50 mM KH 2 PO4 at pH 7.0, 500 mM KCl, 10 mM MgCl 2 , 0.5 mM DTT, and 125 units PKA/mg cTnI. ATP was added to the column to initiate the reaction. After 30 min at 30 °C, the column was washed with a buffer containing 50 mM Mops at pH 7.0, 500 mM KCl, 2 mM CaCl 2 and 0.5 mM DTT. Phosphorylated cTnI was eluted with a buffer containing 6 M urea, 10 mM EDTA, 0.5 mM DTT, and 50 mM Mops (pH 7.0). The extent of phosphorylation was quantified by both mass spectral measurements and treatment of the sample with alkaline phosphatase, followed by determination of inorganic phosphate using the EnzChek Phosphate Assay kit (Molecular Probes) (38). Phosphorylation of the two PKA sites in cTnI was >90%. The same amount of nonphosphorylated and phosphorylated tryptophanless cTnI was used for in the preparation of reconstituted troponin. Fluorescence Measurements-Steady-state measurements were carried out on an ISS PCI photon-counting spectrofluorometer equipped with a micro titrator at 10 ± 0.1 °C (1). FRET was used in titration experiments to monitor Ca 2+ induced cTnC N-domain opening. For Ca 2+ titration, 1.0 mL of the reconstituted complex (1 µM of labeled cTnC mutant) was in a buffer containing 50 mM Mops, pH 7.0, 1 mM DTT, 2 mM EGTA, 5 mM MgCl 2 , and 0.2 M KCl. The fluorescence intensity of the donor (tryptophan) was monitored at 340 nm with an excitation of 295 nm. In a typical titration experiment, up to 90 data points were acquired after successively injecting aliquots of 5 µl of a Ca 2+ -EGTA buffer. Free [Ca 2+ ] was calculated by an in-house program, using stability constants given by Fabiato (39). For both emission spectra and titration measurements, the donor (tryptophan) fluorescence of a donor-only protein (cTnC(L12W/N51C)) was first determined, followed by determination of the donor fluorescence of the corresponding donor-acceptor protein (cTnC(L12W/N51C) AEDANS ) at the same protein concentration and identical conditions. All measurements were background subtracted. The background was obtained under identical experimental conditions by collecting background fluorescence from an identical sample solution in which cTnC was omitted. The procedures previously described were used to convert titration data to FRET efficiency and inter-site distances (40).
Fluorescence intensity decays were measured in the time domain using an IBH 5000U fluorescence lifetime system equipped with a 295 nm LED as the light source. Fluorescence intensity decays of the donor from the donor-alone and donor-acceptor samples were collected with a time-correlated single photon counting system associated with the IBH 5000U under identical experimental condition and were corrected for background signals using an identical solution in which either cTnC(L12W/N51C) or cTnC(L12W/ N51C) AEDANS was omitted. We previously found that background correction was critical for distance calculation with tryptophan as the energy donor. Even a small amount of contaminating tryptophan-contained protein would contribute by guest on March 24, 2020 http://www.jbc.org/ Downloaded from significantly to the observed fluorescence intensity and consequently skew the calculated titration curves. Background subtracted decays data were used to calculate intersite distances between donor and acceptor as in previous studies (16,41).

Stopped-Flow
Measurements-Kinetic measurements were carried out at 10 of donor tryptophan fluorescence intensity was first determined from a donor-only sample (unlabeled cTnC mutants), followed by determination of the time-dependent fluorescence intensity for the corresponding donoracceptor sample (AEDANS labeled cTnC mutants). Similar to steady-state measurements, all measurements were corrected by subtracting background fluorescence of the sample in which cTnC was omitted. Eight to 10 kinetic tracings were collected for each set of donor only and donor-acceptor samples and the averages of each set of samples were used to calculate the timedependent FRET efficiency, from which the timedependent FRET distance was calculated (40).

RESULTS
To examine whether mutants cTnI and cTnT, and labeled cTnC have the same regulatory function as wild-type proteins, Ca 2+ regulation of the actin-activated S1-ATPase activity assay was carried out. The results are summarized in Table 1. The ATPase activity of S1 in the presence of actin, but in the absence of troponin and Tm was taken as 100%. The Ca 2+ activation was 0.718 for the control preparation containing wild-type cTn. The Ca 2+ activations for the other two preparations containing mutants of cTnI and cTnT, and labeled cTnC mutant were similar to that of the control, suggesting that the effects of cTnI and cTnT mutations and labeling of cTnC on the Ca 2+ regulatory activity were negligible.
To verify the stability and stoichiometry of the mutant troponin complex reconstituted with labeled cTnC, tryptophanless cTnI and cTnT, SDS-PAGE and native gel analysis were exploited. The troponin complexes were reconstituted by incubating labeled cTnC mutants, tryptophanless cTnI and cTnT in a molar ratio of 1.0 : 1.2 : 1.2 on ice for 30 minutes, and then dialyzing against the working buffer containing 50 mM Mops(pH 7.0), 1 mM DTT, and 0.1M KCl in the presence of 5 mM Mg 2+ . The samples were centrifuged at 10,000 x g for 10 minutes to remove all insoluble extra cTnI and cTnT before native and SDS-PAGE analysis. Electrophoresis analysis showed that both wild-type and mutant troponin complexes existed as a single complex with the correct stoichiometry (the molar ratio of cTnC:cTnI:cTnT was 1:1:1, based on density analysis using Bio-Rad Quantity One Software) in the samples (Fig. 1). Multiple experiments were performed on the samples prepared within two weeks and no protein degradation was observed by electrophoresis analysis.

Ca 2+ -Induced cTnC N-domain Opening
Residue 12 of cTnC is located at the Nterminal end of the helix A and residue 51 is located in the linker between helices B and C. The distance between these two sites is expected to increase upon Ca 2+ binding to the N-domain of cTnC in the presence of cTnI. The AEDANS labeled cTnC(L12W/N51C) was used to monitor this structural transition. Steady-state measurements showed that in the troponin complex the fluorescence intensity of donor tryptophan in a cTnC(L12W/N51C) sample at 340 nm was insensitive to Ca 2+ binding to the regulatory site of cTnC N-domain ( Fig. 2A). This donor fluorescence intensity was quenched by 50% when the acceptor AEDANS was attached to Cys51 indicating a high efficiency of FRET and suggesting a close proximity between the two residues. The second emission band at 480 nm is the acceptor fluorescence in cTnC(L12W/  Fig. 2A). These reciprocal changes indicate a Ca 2+ -induced reduction of FRET and an increase in the distance between the two sites. The magnitude of the distances was quantified from data obtained from time-resolved fluorescence measurements (Fig.  2B). As expected, Ca 2+ had little effect on the intensity decay of the donor from the donor-only sample (Fig. 2B), but the presence of Ca 2+ significantly slowed down the donor decay curve. These changes of the intensity decay suggest a Ca 2+ -induced decrease in FRET and an increase in the inter-site distance. The data were analyzed in terms of a distribution of the distances. The mean distances between residues 12 and 51 derived from the distribution of distances are given in Table 2. With control cTnC, the FRET distance in the apo-N-domain was 22.6 Å. This distance compares favorably with the C α -C α distance (20.4 Å) between the two residues in the NMR cTnC structures (42). In the Ca 2+ -saturated state in which the N-domain contains a bound Ca 2+ , the FRET distance increased by 6.3 Å to 28.9 Å. The corresponding distance from NMR structures (3) is 29.5 Å, and from the crystal structure of the core domain of cardiac troponin (43) is 29.4 Å. It is noted that the FRET distances reported here for control cTnC in cTn were 3.7 and 3.1 Å longer for the apo and Ca 2+ -bound states, respectively, than in our previous report (1). This discrepancy arises from failure in the previous work to correct for background signals contributed from a small amount of tryptophan-containing contamination. The magnitude of the Ca 2+ -induced change in distance was very similar in both studies.
We tested the effects of the two HCM-related single cTnC mutations, L29Q and G159D, on the conformation of the cTnC regulatory domain using troponin reconstituted with nonphosphorylated and PKA phosphorylated cTnI. The results are also listed in Table 2. These mutations had negligible effects on the distance between residues 12 and 51 of cTnC when determined in either the Mg 2+ -or Ca 2+ -saturted state. The distance remained essentially the same in each of the two states, regardless of whether the cTnI was phosphorylated by PKA.

Equilibrium Conformation of the cTnC N-Domain
Two sets of Ca 2+ titration curves were obtained with reconstituted troponin using the FRET marker cTnC(L12W/N51C) AEDANS to determine the distances between residues 12 and 51 in cTnC at intermediate Ca 2+ concentrations between pCa 6.4 and 4.0. One set of curves were obtained with samples reconstituted with (1) cTnC marker, cTnI and cTnT (control) and (2) cTnC marker containing cTnC mutation L29Q or G159D, cTnI and cTnT. The other set of curves were obtained with samples reconstituted with (1) cTnC marker, phosphorylated cTnI, and cTnT (control) and (2) cTnC marker containing either cTnC mutation L29Q or G159D, phosphorylated cTnI, and cTnT. These titration curves were monophasic and are shown in Fig. 3A and 3B. The values of pCa 50 and the Hill coefficients from the curves are given in Table 3. Mutation L29Q decreased Ca 2+ sensitivity by 0.1 pCa units, and G159D increased Ca 2+ sensitivity by 0.1 units. The Hill coefficient was increased by 10% and 20% by these two mutations. Overall, the two cTnC mutations appear to have minimal effects on the equilibrium conformations of the cTnC regulatory N-domain.
In the absence of the cTnC mutations (control sample), cTnI phosphorylation decreased the pCa 50 from 5.55 to 5.38. This small loss in Ca sensitivity (∆pCa 50 -0.17) was accompanied by a 30% decrease in the Hill coefficient. The presence of L29D or G159D had negligible effects on Ca 2+ sensitivity (∆pCa 50 ≤ -0.04). However, the presence of L29Q reduced the Hill coefficient by 20% (2.02 to 1.61), while G159D had no effect on the Hill coefficient. The main effect of L29Q appears to be reduction of the Ca 2+ sensitivity of the N-domain and G159D an enhancement of the Ca 2+ sensitivity. Both mutations inhibited the ability of PKA phosphorylated cTnI to reduce Ca 2+ sensitivity.

Kinetics of Structural Transition in cTnC Nterminal Domain
We established the magnitude of the FRET distance between W12 and C51 AEDANS in cTnC reconstituted into troponin in different by guest on March 24, 2020 http://www.jbc.org/ Downloaded from biochemical states, and the effects of the two cTnC mutations on the distances in these states. The FRET distances define the global conformations of the cTnC N-domain. We next performed stopped-flow measurements to investigate the kinetics of conformational transitions between these states. Shown in Fig. 4 are representative FRET-based kinetic tracings. Tracing 1 is for Ca 2+ -induced N-domain opening, obtained by mixing Ca 2+ with a donor-acceptor sample of troponin. Tracing 2 is the baseline for this mixing experiment. Curve 3 is from an inverse experiment in which the donor-acceptor sample was mixed with an EGTA solution to monitor domain closing, and its baseline is shown as tracing 4. In both tracings, a portion of the initial FRET signal was lost within the mixing time (1.5 ms) of the instrument. There was a 10% signal loss in the tracing triggered by Ca 2+ dissociation (curve 3), and a 45% signal loss in the tracing initiated by Ca 2+ binding (curve 1). Although the lost signals could not be time-resolved, they provided evidence for very fast initial structural transitions occurring in both domain opening and closing.
The FRET signals displayed in curves 1 and 3 in Fig. 4 were converted to intersite distances. Figure 5 shows the distance plot for domain closing triggered by Ca 2+ dissociation. This distance tracing was biphasic and was fitted to a bi-exponential function, yielding two rate constants, 153 s -1 (0.85) and 22 s -1 (0.15). The fast phase had a fractional amplitude of 0.85 and the slow phase had a fractional amplitude of 0.15. These amplitudes corresponded to a decrease of 5.4 Å in the fast phase and a decrease of 0.95 Å in the slow phase. It is not entirely clear what structural changes were associated with the distance changes. We would suggest that the fast phase with a large distance change may be the key step for regulation and the slow phase with a small distance change may reflect structural relaxation following the major structural change.
The effects of cTnC mutations L29Q and G159D on the kinetics of cTnC N-domain closing triggered by Ca 2+ dissociation are shown in Fig.  6A. L29Q had little effect on the fast phase, but it increased slightly the rate of the slow phase without affecting the fractional amplitudes of the two phases. G159D decreased the rate of the fast phase from153.7 s -1 to 122.1 s -1 and increased the rate of the slow phase from 22.0 s -1 to 32.1 s -1 .
Although the total distance decrease was not affected by this mutation, the fractional amplitudes were changed. These results are summarized in Table 3. Figure 5B shows the tracings for domain opening initiated by Ca 2+ binding. A large signal was lost in all tracings within the mixing time of the instrument. Excluding the lost signal, the tracings were well fitted to a two-exponential function. The two recovered rate constants are listed in Table 4. Also listed in Table 4 are the rate constants determined in the presence of cTnC mutations. L29Q decreased the rate of the fast phase by 17% and lowered the rate of the slow phase by 27%. G159D decreased the rate of the fast phase by 43% and reduced the rate of the slow phase by 34%.
The two rate constants for closing the Ndomain of cTnC in troponin were enhanced when cTnI was replaced by PKA-phosphorylated cTnI. The 21% rate enhancement of the fast phase (153.7 to 184.5 s -1 ) was essentially abolished in the presence of L29Q. G159D also eliminated any rate enhancement by phosphorylated cTnI. Figure  6 summarizes these results for the rate of the fast kinetic phase. The results for both the fast and slow phases are listed in Table 4 for comparison with results obtained from non-phosphorylated cTnI. In contrast to domain closing, phosphorylated cTnI reduced the rates of domain opening for both kinetic phases (Table 5).

DISCUSSION
The Ca 2+ -induced open conformation of the regulatory N-domain of cTnC is a key feature in muscle regulation. In a previous study we developed a FRET-based marker, cTnC(L12W/ N51C) AEDANS , to study the transition between the closed and open conformations of the N-domain (1). This marker enables us to monitor structural transitions that occur in the N-domain resulting from Ca 2+ binding (activation) and Ca 2+ dissociation (deactivation) in cardiac troponin. In the present study, we used the marker to investigate the effects of cardiomyopathy-related mutations in cTnC, L29Q and G159D, on these transitions in the cTnC N-domain. We also investigated how the mutations modified the effects of cTnI phosphorylation on these structural transitions.
In control samples of reconstituted troponin, the FRET distance between residue 12 and 51 in by guest on March 24, 2020 http://www.jbc.org/ Downloaded from cTnC increases by 6.3 Å upon binding Ca 2+ at the single regulatory site in the N-domain. This change has been reported in previous studies (1,3). The two cTnC mutations have negligible effects on the distance observed in both the inactive and active state. Although they also have negligible effects on the Ca 2+ sensitivity of the transition between the two states (within 0.1 pCa 50 units), both mutations enhance the cooperativity of Ca 2+ binding. PKA phosphorylated cTnI in the control troponin complex has no effect on the FRET distance, and the cTnC mutations do not alter the distance. However, phosphorylated cTnI in the control complex reduces the Ca 2+ sensitivity (pCa 50 shifting from 5.55 to 5.38) and the cooperativity of Ca 2+ binding in the transition from the inactive to the active state. The two cTnC mutations alter the observed effects of phosphorylated cTnI in two ways: (1) eliminating a change in Ca 2+ sensitivity and (2) increasing cooperativity of Ca 2+ binding.
Kinetic results show multiple steps in both opening and closing of the cTnC N-domain. With control samples, domain opening induced by Ca 2+ binding is described by three steps. One very fast step accounting for about one-half of the total distance change occurs within the mixing time (1.5 ms) of the instrument and could not be resolved. The other two steps are relatively slow with rates of 72.0 and 5.9 s -1 . In a previous study, using a single environmental sensitive florescence probe attached to Cys35 to monitor domain opening, we observed only two transitions (44). The two previously observed kinetic steps likely correspond to the two slow steps observed in the present FRET experiment. It is noted that detection of the third kinetic step of structural transition was made possible with a two-probe FRET sensing system. This transition was not detected from the signal of a single probe. The Ndomain opening results from reorientations of two helices (helices B and C) away from the central helix and an increase of the interhelical angle (3). The open structure accommodates the binding of the regulatory region of cTnI. The FRET-detected Ca 2+ -induced multiple transitions may be related to these reorientations and insertion of the regulatory region into the open domain. Upon Ca 2+ dissociation, reversal of the domain structure occurs in two kinetic steps: 153.7 s -1 and 22 s -1 . The fast step accounts for about 85% of the total distance change and the slow step accounts for about 15% of the change. Although these kinetic steps may reflect movements of the helices, it is not possible to relate the steps to specific structural alterations in the present study.
The mutation L29Q has little effect on the rates of structural transitions associated with domain closing triggered by Ca 2+ dissociation, but decreases the rates associated with Ca 2+ -induced domain opening. On the other hand, G159D affects the rates of both opening and closing of the domain. These effects become more pronounced when cTnI is phosphorylated by PKA. During sympathetic stimulation, the activation of βadrenocepters stimulates the cAMP/PKA pathway. This activation increases contractile force, raises heart rate and fastens relaxation of the myocardial cells. The phosphorylation-induced acceleration of heart rate and relaxation rate is an important modulation mechanism for proper heart function against an increase work load. The kinetic results suggest that this modulation mechanism may be impaired by the presence of the cardiomyopathyrelated cTnC mutations.
NMR studies have suggested that the cardiac specific N-terminal extension of cTnI directly interacts with the N-domain of cTnC in a PKA phosphorylation-dependent manner (19,(21)(22)(23)31,32). The non-functional Ca 2+ binding site I of cTnC and the residues 19-21 of cTnI have been implicated in this interaction (13,19,22,31,45). Without phosphorylation this interaction stabilizes the open N-domain conformation of cTnC in the troponin complex, and phosphorylation weakens this interaction and shifts the closed-open conformation equilibrium of the N-domain toward the closed state (46). L29Q located in the region of the non-binding loop I may increase the polarity of the region and this change may affect the interaction of this region with cTnI. A recent peptide array study showed that the mutation L29Q completely abolished the interaction between the cTnI N-terminal segment and cTnC in the L29 region (32). Although L29Q does not cause profound structural effects on the open Ndomain conformation, it alters Ca 2+ sensitivity and the rates of conformational transitions. Since residue G159 is third residue from the C-terminus, it is not surprising that the mutation G159D has little effect on the equilibrium conformation of the N-domain. It is surprising that the mutation affects  (47). Since the region of G159 is known to interact with the C-terminal end of the N-terminal extension of cTnI, the effect of G159 we observed here may be due to global conformational changes in cTnI and cTnT as previously reported (15).
G159D of cTnC is a DMC-related mutation. It has been suggested that DMC mutations of troponin proteins generally cause a decrease in myofilament Ca 2+ sensitivity (48)(49)(50). However, there are several studies that do not reach the same conclusion. For example, recent studies showed the presence of G159G mutation of cTnC in muscle fibers produced no change in myofilament Ca 2+ sensitivity when tension force and ATPase activities were measured (51). In our study with reconstituted troponin complex, G159D mutation slightly increased Ca 2+ sensitivity (0.1 pCa unit) of the N-domain conformational change of cTnC. This change may have resulted from alteration in the interaction between the C-domain of cTnC and cTnI caused by the mutation. Further study is needed to predict the relationship between G159D mutation and Ca 2+ sensitivity of the reconstituted actomyosin.
In summary, using a FRET-based conformational marker we have investigated the effects of two cardiomyopathy-related cTnC mutations, L29Q and G159D, on the equilibrium conformations of the cTnC regulatory N-domain under conditions that simulated the relaxed state and the activated state of troponin and in the presence of PKA phosphorylated cTnI. We also investigated the kinetics of the transitions of the N-domain conformations between the two states. The main finding is that the mutations have significant effects on the kinetics of opening and closing of the conformation of the regulatory Ndomain, although they have negligible effects on the equilibrium conformations. We previously showed that PKA phosphorylation of cTnI enhances the rate of N-domain closing triggered by Ca 2+ dissociation. This rate enhancement is eliminated by both cTnC mutations. Taken together the results suggest that the two cardiomyopathy-related cTnC mutations exert their effects mainly by modifying the structural dynamics of the cTnC regulatory N-domain during activation and deactivation. They appear to be antagonistic toward the effect of phosphorylation signaling from cTnI to cTnC. Integrity of the reconstituted troponin complexes was examined by native PAGE (panel A) and SDS-PAGE (panel B) analysis. Lane 1, HMW protein standard (Amersham Biosciences); Lanes 2 and 4, troponin samples reconstituted with wild type cTnC, cTnI and cTnT; Lanes 3 and 8, troponin reconstituted with labeled cTnC, tryptophanless cTnI and tryptophanless cTnT; Lane 5, 6 and 7, isolated wild type cTnC (18 KDa), cTnI (24 KDa) and cTnT (34 KDa), respectively; Lane 9, 10 and 11, labeled cTnC, tryptophanless cTnI and tryptophanless cTnT, respectively. Panel A: The native PAGE (8% resolving and 4% stacking) showed both reconstituted wild-type and mutant troponin complexes were resolved in a single band (lanes 2 and 3) with a molecular weight of ~76 KDa. Panel B: The SDS-PAGE (18% resolving and 4% stacking) showed that under denatured condition both reconstituted wildtype (lane 4) and mutant (lane 8) troponin complexes were separated into three bands which were identical to the subunits of cTnC, cTnI and cTnT, respectively. Density analysis of these protein bands with Bio-Rad Quantity One Software suggested that both the troponin complexes reconstituted with wild type proteins and mutant proteins consist of cTnC, cTnI and cTnT at molar ratio of 1:1:1. No significant degradation products were found on both gels.   FRET-based Ca 2+ titration of the distance between residues 12 and 51 of cTnC(L12W/N51C) AEDANS in reconstituted troponin. (A) Normalized distance change vs. pCa for troponin containing non-phosphorylated cTnI, in the absence of cardiomyopathy-related cTnC mutations (black), in the presence of cTnC mutation L29Q (red), and in the presence of cTnC mutation G150D (green). (B) Normalized distance change vs. pCa for troponin containing nonphosphorylated cTnI (solid black) and phosphorylated cTnI (open black), and phosphorylated cTnI in the presence of cTnC mutation L29Q (red) and in the presence of G159D (green). The curves were fitted with the Hill equation to obtain values of pCa 50 and the Hill coefficient. These parameters are listed in Table 2.  FRET-based stopped-flow tracings of distance change between residues 12 and 51 of cTnC(L12W/N51C) AEDANS in reconstituted troponin triggered by Ca 2+ dissociation. (A) The data were fitted with a two-exponential function and a single-exponential function. The two best-fitted tracings are shown in the figure. From the two-exponential fit, the fitted rate for the fast phase is 153.7 s -1 accounting for a decrease of 4.8 Å and the rate for the slow phase is 22.0 s -1 accounting for an additional decrease of 1.5 Å. (B) The residual plot for the biexponential fit shown in (A). (C) The residual plot for a single-exponential fit of the data shown in (A). The two residual plots indicate that the two-exponential fit is more satisfactory.  Comparison of the effects of cardiomyopathy-related cTnC mutation on the rate of closing of cTnC Ndomain induced by Ca 2+ dissociation. Shown here are the rates of the fast kinetic phase. Phosphorylated cTnI enhances the rate by 21%. This enhancement is abolished by either L29Q or G159D. * Numbers in parentheses give the respective lower and upper 68% (one standard deviation) confidence estimates of the fitted mean value. ** p-cTnI, phosphorylated cTnI.  * k fast and k slow are the rates of the fast and slow phases of the cTnC N-domain closing, respectively. A fast and A slow are the percentage of total distance change associated with the fast and slow phases, respectively. ± confidence limits are standard deviation (SD) (n=2). ** p-cTnI, phosphorylated cTnI.