Troponin and Tropomyosin

Proteins That Switch on and Tune in the Activity of Cardiac Myofilaments

From the Department of Physiology and Biophysics, College of Medicine, University of Illinois at Chicago.

Correspondence to R. John Solaro, PhD, Department of Physiology and Biophysics (M/C 901), College of Medicine, University of Illinois at Chicago, 835 South Wolcott Ave, Chicago, IL 60612-7342.

	Abstract

Top Abstract Activation and the Structure... Crossbridges at Rest and... Molecular Interactions... Molecular Interactions... Tn-Tm and Tuning Myofilament... Concluding Remarks References

 
Abstract—We present a current perception of the regulation of activation of cardiac myofilaments with emphasis on troponin (Tn) and tropomyosin (Tm). Activation involves both a Ca²⁺-regulated molecular switch and a potentiated state, dependent on feedback effects of force-generating crossbridges. Recent developments in the elucidation of the structure and arrangement of the myofilament proteins offer insights into the molecular interactions that constitute the switching and potentiating mechanisms. Transgenic mice overexpressing myofilament proteins, in vitro studies of mutant myofilament proteins, multidimensional multinuclear nuclear magnetic resonance, and fluorescence resonance energy transfer offer important approaches to understanding the molecular signaling processes. These studies reveal special features of the cardiac myofilament proteins that appear specialized for the unique functions of the heart. An important aspect of these special features is their role in mechanical, chemical, and neurohumoral coupling processes that tune myofilament activation to hemodynamics and beating frequency. Understanding these processes has become essential to understanding cardiac pathologies such as heart failure, ischemia and reperfusion injury, stunning, and familial hypertrophic cardiac myopathies.

Key Words: heart • thin filament • myosin • actin • Ca²⁺ binding protein

	Activation and the Structure and Arrangement of the Myofilament Proteins

Top Abstract Activation and the Structure... Crossbridges at Rest and... Molecular Interactions... Molecular Interactions... Tn-Tm and Tuning Myofilament... Concluding Remarks References

 
Our review focuses on current concepts and questions relating to how contraction of cardiac myofilaments is switched on by Ca²⁺ and tuned to match the prevailing beating frequency and hemodynamic load. By tuning of myofilament activity we mean adjustments in the intensity and dynamics of myofilament contraction and relaxation independent of adjustments in membrane-regulated cellular Ca²⁺ flows. Tuning includes intrinsic modulation by coupling myofilament activity to mechanical state and extrinsic modulation by coupling to neurohumoral state and prevailing intracellular conditions.

One of the most striking features of the activation process in heart muscle is the complexity and extent of interlinked protein-protein interactions that are triggered by Ca²⁺ binding to the thin filament. Figure 1 illustrates the disposition and interactions among the thin-filament proteins in a fundamental structural unit during diastole and systole.^{1 2 3} The structures shown are based on data from a variety of approaches, including analysis of protein crystal structures,⁴ x-ray and neutron diffraction,⁵ cryoelectron microscopy,⁶ and multidimensional multinuclear nuclear magnetic resonance (NMR) spectroscopy.^{7 8} In Figure 1, we show one strand of the double-helical array of actin molecules that forms the thin filament. The actins are rather flat molecules, 5 nm across with 4 major domains, 2 of which are depicted in Figure 1. Tropomyosin (Tm), a double-stranded -helical protein, winds around the actin array. Cardiac (c) troponin (cTn) is a heterotrimer of the following distinct gene products: cTnC, the Ca²⁺ receptor; cTnI, an inhibitor of the actin-myosin reaction that shuttles between tight binding to actin and tight binding to Ca²⁺-TnC; and, cTnT, which binds to Tm, cTnI, and cTnC. cTnC exists in solution as a highly -helical dumbbell-shaped molecule comprising 2 globular domains joined by a central linker.^{7 8} Each of the globular domains contains 2 helix-loop-helix motifs that bind metals. In both fast skeletal (fs) TnC (fsTnC) and cTnC, there are 2 metal binding sites in the COOH-terminal domain that bind both Mg²⁺ and Ca²⁺ with relatively high affinity. These sites exchange these metals too slowly to switch activation of the myofilaments on and off and have a main role in anchoring cTnC tightly to the NH₂ terminus of cTnI.^{9 10} Regulatory Ca²⁺-binding sites reside in the NH₂-terminal domains in both variants, but there are 2 competent sites in fsTnC and only 1 in cTnC, owing to the substitution of an amino acid in the coordinating structure. The single site in the NH₂ domain of cTnC is relatively specific for Ca²⁺, binds Mg²⁺ weakly, and is able to exchange Ca²⁺ fast enough to regulate the diastolic/systolic transition.¹⁰ Figure 1 depicts cTnI as an elongated structure with a doughnut-shaped toroid at each end and a central spiral, making multiple contacts with cTnC. This shape for the cardiac variant is based on the model structure derived from small-angle x-ray and neutron-scattering data of Olah et al,⁵ who analyzed the fsTnC-fsTnI-4Ca²⁺ complex. The maximum linear dimension of fsTnI is 11.8 nm but may be longer in cTnI, which has 32 more amino acids. Striated muscle TnT is an asymmetric protein (Figure 1). In the case of fsTnT, it extends 19 nm.⁶ cTnT is most likely longer, inasmuch as there is an amino terminal extension making the protein 2 nm longer than its skeletal counterpart.² The unique features of the cardiac variants of cTn indicate a special role in the heart. Not only are cTnI and cTnT bigger and possibly longer proteins than fsTnI and fsTnT, but there are also important sites of phosphorylation in cardiac TnI and TnT that are not present in their skeletal counterparts.¹

View larger version (31K): [in this window] [in a new window]   Figure 1. Structure and arrangement of cardiac myofilament proteins in diastole and systole. Presumed shapes of the myofilament proteins are based on evidence from multinuclear mulitdimensional NMR, low-resolution small-angle x-ray diffraction and neutron scattering, and protein crystal analysis as described in the text. The crosshatched area in TnI depicts a highly basic inhibitory peptide. Tm, which extends over 7 actin monomers, is not depicted as the full-length molecule. Arrows indicate relative movements of Tm position in diastole and systole. C indicates COOH terminus; N, NH2 terminus. See text for further description.

Ca²⁺ binding to TnC triggers a sequence of protein-protein interactions signaling a cyclic reaction between actin and myosin that results in force generation. A reflection of the cooperative nature of the process is how steeply steady-state myofilament force rises with increases in the surrounding Ca²⁺ concentration, especially in situ.^{1 11} In heart muscle, force increases from a resting level to full activation over but a fraction of a pCa (-log molar Ca²⁺ concentration) unit. This steep relation cannot occur by Ca²⁺ binding alone. There is a single regulatory Ca²⁺-binding site on cardiac TnC, and long-range cooperative interactions between the cTn molecules along the thin filament result in little cooperativity in the Ca²⁺ binding process.¹² There is, however, evidence for substantial cooperative interactions between bound crossbridges and the thin filament.¹¹ The result is that activation of a patch of the thin filament spreads; thereby, the force-generating binding of some crossbridges with the thin filament eases the further binding of crossbridges. However, despite the cooperative and rather explosive nature of the Ca²⁺-activation process, cardiac myofilaments at a basal level of physiological activity operate at 25% to 50% of their maximum capability.¹³ Partial activation of the myofilaments permits the heart to use a reserve of myofilament activation and crossbridge activity to meet varying hemodynamic demands.

	Crossbridges at Rest and Crossbridges at Work

Top Abstract Activation and the Structure... Crossbridges at Rest and... Molecular Interactions... Molecular Interactions... Tn-Tm and Tuning Myofilament... Concluding Remarks References

 
Elucidation of the structural biology of the actin-myosin interface has contributed to a surge in our understanding of the state of crossbridges in resting muscle and the molecular basis for how crossbridges work in active muscle. This information is essential to a clear understanding of how the actin-myosin reaction is activated and how the crossbridge– thin filament interaction modulates the activation.

Crossbridges at Rest: Blocked and Weakly Binding Crossbridges in the Diastolic State
At rest, the myofilaments are freely extensible; crossbridge cycling and force generation are shut down. At relatively low concentrations of Ca²⁺, the relaxed state is imposed on the actin-myosin reaction by the presence of Tn-Tm. Concepts as to how these thin-filament proteins act to maintain the diastolic state revolve around the following questions: Is crossbridge binding to actin blocked, or is there a weakly bound state? If present, does the block involve Tm alone or perhaps TnI and/or TnT?

A textbook view of the myofilaments in diastole is that crossbridges are blocked from binding to actin by steric hindrance involving Tm. The idea that there is a release of the actin-crossbridge reaction from steric inhibition associated with the position of Tm on the thin filament received support from analysis of time-resolved x-ray diffraction¹⁴ and 3-dimensional reconstructions.¹⁵ However, it is now clear that the steric blocking model, a durable, easy to understand, and didactically useful model of the diastolic state of the myofilaments, must be reconsidered. Steric block implies no crossbridge interactions in the relaxed state, yet there is considerable evidence that a substantial portion of the crossbridges bind weakly to the thin filament in the relaxed state.^{16 17} Early evidence^{16 17} indicated that (1) the ATPase cycle could occur without dissociation of the crossbridges from actin, and (2) the weakly bound non–force-generating crossbridges bind with rapid on-off kinetics. For example, a recent report¹⁸ of high-resolution x-ray diffraction patterns from skeletal muscle cells demonstrates the presence of an actin-based layer line, providing strong evidence for weak attachments of crossbridges in the relaxed state. The x-ray data were also interpreted as indicative of both detached and weakly attached crossbridges. The idea that these detached crossbridges represent a blocked state is supported by the findings after measurement of the time course of the actin-myosin interaction and thin-filament activation.¹⁹ A delay in the activation could not be easily fit without including a blocked population of crossbridges. Blocked crossbridges were visualized to constitute a substantial proportion of the population. Thus, we presently view the relaxed state to include both blocked and weakly attached crossbridges.

Crossbridges at Work: Molecular Interactions Generating Myofilament Work
Our understanding of the physical basis of the actin-crossbridge interaction was considerably advanced by reports providing models based on the fitting of images from cryoelectron micrographs using the atomic structures of F-actin and myosin S-1.²⁰ The mechanism by which striated muscle develops force and shortens is based on the idea that heads of myosins (the crossbridge or myosin S-1) protruding from the thick filament react with thin-filament actins in a reaction cycle that is powered by ATP.^{21 22} The binding surface for NH₂-terminal regions of myosin spans 2 actins and includes both the NH₂-terminal and COOH-terminal regions of actin.²⁰ A step in the cycle is one in which a nucleotide-free crossbridge is bound tightly to actin in a rigor link that involves extensive contacts between myosin and the thin filament. This interface is formed from a combination of complementary, hydrophobic, and ionic interactions, as well as from hydrogen bonding. Binding of nucleotide initiates a substantial change in the actin-myosin interface, resulting in dissociation and eventual generation of a power stroke powered by ATP hydrolysis.²²

Electrostatic contacts between the positively charged essential light chain of the myosin NH₂ terminus and the negatively charged residues in the COOH terminus of actin form a secondary region of contact between the crossbridges and the thin filament. Recent cross-linking²³ studies show clear evidence that the cardiac variant is able to bind to actin. Although the physiological role of essential light chain–actin binding remains unclear, there is strong evidence that the interaction may significantly modulate the crossbridge cycling rate²⁴ as well as the level of potentiation of the thin filament.²⁵ Both of these properties are likely to be important elements in determining the kinetics of back and forth transitions of myofilaments between diastolic and systolic states.

	Molecular Interactions Triggering Myofilament Activation

Top Abstract Activation and the Structure... Crossbridges at Rest and... Molecular Interactions... Molecular Interactions... Tn-Tm and Tuning Myofilament... Concluding Remarks References

 
Understanding the molecular mechanisms of signal transmission between thin-filament Ca²⁺ binding and promotion of crossbridges from relaxed to force-generating states is one of the main research objectives in the field of myofilament activation. We begin by describing what happens to actin and Tm when myofilaments are turned on. We will then summarize molecular interactions among Tn components that transduce the Ca²⁺-binding signal to the movement and state changes of actin and Tm. The regions of interaction between the thin-filament proteins are summarized in Figure 2.

View larger version (29K): [in this window] [in a new window]   Figure 2. Relation between sites of protein phosphorylation and sites of Ca2+-induced changes in the interaction between cTn units and actin-Tm. Double-headed arrows indicate Ca2+-sensitive reversible reactions. Lines indicate structural interactions. Substrates for PKA-dependent phosphorylation include Ser23 and Ser24 in the unique NH2-terminal region of cTnI. Substrates for PKC-dependent phosphorylation include Thr280 and Thr199 of TnT and Ser43, Ser45, and Thr144 of cTnI. The crosshatched area of cTnI is the inhibitory peptide. C indicates COOH terminus; N, NH2 terminus. See text for further description.

Movement and State Changes of Actin-Tm
Detailed evidence provides a picture of actin and Tm structure and their protein-protein interactions as fluid and dynamic. Triggering of the actin-myosin reaction involves both a movement and rotation of Tm on the thin filament and a change in actin structure.^{26 27} The extensive and dynamic interactions of Tm on the thin filament require flexible protein-protein interactions. Tm is a nearly 100% -helical protein consisting of a dimer of 2 strands held together by hydrophobic side chains that wind around each other to form a coiled coil (Figure 1). The coiled coil intertwines as a polymer, overlapping a contiguous Tm by some 5 to 10 residues along the thin-filament actins at a ratio of 1 Tm per 7 actin monomers. As Tm winds around and moves on the actin helix, it opposes different faces of the actin monomers. Thus, it makes sense that the interaction of Tm with actin involves nonspecific ionic interactions of mobile side chains with actin.²⁷ Our experiments²⁸ with cardiac myofilaments from mice harboring transgenes expressing a skeletal isoform (ß-Tm) revealed that small specific differences in Tm charge are able to elicit large changes in myofilament activation. These data derived from studies on the intact myofilament lattice support the concept of the importance of charge in the function of Tm.

Transmission of the Ca^2+-Binding Signal From TnC to TnI
Initial critical events that trigger movement and state changes of Tm and actin are Ca²⁺ binding to cTnC and transmission of the Ca²⁺-binding signal to cTnI. As depicted in Figure 1, a consequence of the Ca²⁺-signaling process is that cTnI may move as much as 1.5 nm from its diastolic off state (tightly bound to actin) to its systolic state (tightly bound to cTnC).²⁹ Comparative structural evidence provides exciting new information on unique features of the molecular and atomic process by which Ca²⁺ binding to cTnC initiates the signaling cascade. The first determinations of the solution structure of the regulatory domain of fsTnC in the apoprotein- and Ca²⁺-bound states have been carried out with the use of multinuclear and multidimensional NMR.⁷ In the case of fsTnC, the binding of Ca²⁺ at the 2 helix-loop-helix motifs results in movements of helices such that an extensive hydrophobic patch of amino acids is exposed in an "open" configuration. In the case of cTnC, the events appear quite different and indicate that Ca²⁺ signaling in cTn may involve a stronger dependence on charge-charge interactions than fsTn.⁸ With Ca²⁺ binding to this single site, the NH₂-terminal region remains substantially closed with exposure of a much smaller area of hydrophobic patch than is the case with fsTnC. This would seem to follow from the differences in the number of metals bound to the cTnC and fsTnC NH₂-terminal region. Whether or not cTnI binding to cTnC induces the hydrophobic patch remains to be determined.

Mutational and structure-function analyses have clarified how various regions of cTnI react with cTnC and participate in the thin-filament Ca²⁺ signaling.¹ As shown in Figure 2, there is a far NH₂-terminal extension (residues 1 to 32) consisting of a stretch of amino acids unique to cTnI. Ser23 and Ser24 are sites for phosphorylation by protein kinase A (PKA). When these sites are phosphorylated, the affinity of cTnI for cTnC falls, as does the affinity of Ca²⁺ for cTnC and Ca²⁺ sensitivity of myofilament activation.^{10 30} Both Ser23 and Ser24 must be phosphorylated for the depression of Ca²⁺ sensitivity to be expressed.³¹ However, our experiments demonstrated that the pCa-ATPase activity of myofibrils reconstituted with recombinant cTnI lacking the 32–amino acid NH₂-terminal peptide was the same as the control activity.³² This result led us to the conclusion that phosphorylation of the sites in the peptide leads to a new state of cTnI.³²

A near NH₂-terminal region of cTnI (residues 33 to 80) is highly conserved and binds to cTnC.^{9 33} As discussed below, this near NH₂-terminal region of cTnI is likely also to contain sites of interaction with cTnT.^{34 35} The binding of cTnI_33–80 to the COOH terminus of cTnC was demonstrated by using selective [¹³C]methionine labeling combined with NMR spectroscopy.^{9 33} Thus, the molecular arrangement of cTnI and cTnC is antiparallel (Figures 1 and 2). The interaction between the NH₂ terminus of cTnI with cTnC is greatly weakened if metal is not bound to the COOH terminus of cTnC. Thus, as illustrated in Figures 1 and 2, an important role of metal binding to these slowly exchanging sites is to anchor the NH₂ terminus of cTnI to cTnC throughout systole and diastole. In more recent experiments,³⁵ we found that cTnI lacking the first 53 amino acids was unable to bind cTnC and restore Ca²⁺ sensitivity in reconstituted preparations. Thus, the NH₂-terminal cTnC binding domain on cTnI appears localized to within residues 33 to 53. Serines at positions 43 and 45 in this area are functionally significant as substrates for protein kinase C (PKC).³⁶ Phosphorylation of these sites resulted in a marked decrease in the maximum ATPase rate of reconstituted thin-filament myosin S-1 preparations with no change in Ca²⁺ sensitivity.³⁶ Phosphorylation of these sites also caused a decrease in the apparent affinity of myosin S-1 for the thin filament.³⁶ A third site of phosphorylation by PKC is located at Thr144, a cardiac-specific residue (Pro in skeletal TnI) located in the TnI inhibitory peptide (cTnI_129–150 and fsTnI_96–116, which is shown as a crosshatched region in Figure 2). The inhibitory peptide is downstream from this cTnC-cTnT binding region of cTnI containing a preponderance of positively charged amino acids.³⁷ The inhibitory peptide retains much of the inhibitory activity of the full-length cTnI, suggesting strongly that this region of cTnI interacts with actin at a site that either interferes with crossbridge binding or allosterically inhibits the actin-myosin reaction. Whether PKC-dependent phosphorylation of Thr144 in the inhibitory peptide is functionally significant is not clear. We³⁶ found no functional effects of Thr144 phosphorylation, whereas the results of Malhotra et al³⁸ indicate that phosphorylation of Thr144 may be of significance in the depression in the maximum ATPase rate after PKC-dependent phosphorylation of cTnI.

Although the inhibitory peptide is clearly of importance in the signaling cascade, our recent mutational analysis has identified a carboxyl-terminal region (cTnI_152–199) that is essential for activation.³⁹ Progressive truncation of cTnI demonstrated that cTnI_1–188 and cTnI_1–151 were able to inhibit ATPase activity of reconstituted myofibrils at pCa 8 to 75% and 50% of that of the wild-type cTnI, respectively. In addition, the cTnI_1–188-cTnC complex only partially restored Ca²⁺ sensitivity on reconstitution into myofibrils, whereas the cTnI_1–151-cTnC complex was ineffective in restoring Ca²⁺ sensitivity. These mutants retained both actin and cTnC binding activity.

Structural analyses, including NMR spectroscopy⁹ and fluorescence resonance energy transfer,^{30 40 41} have been aimed at understanding the changes that occur when cTnC reacts with cTnI. Our studies⁹ using site-directed spin labeling together with NMR showed that isolated cTnC demonstrated a relatively compact conformation most likely resulting from flexibility in the central helix. However, cTnC adopted an extended conformation on binding to cTnI. Other studies using fluorescence resonance energy transfer have indicated that cTnI also becomes more extended in the cTnC-cTnI complex.^{40 41} Olah et al⁵ also concluded that Ca²⁺-saturated fsTnC in a complex with fsTnI has an extended conformation, as determined from studies using small-angle x-ray and neutron scattering.

Signaling Between TnI-TnT, TnC-TnT, and TnT-Tm
As the biggest and longest component of cTn, it is apparent that cTnT has the potential for the most extensive and versatile interactions with adjacent proteins of the thin filament.^{1 2 3} Multiple cTnT isoforms, which are generated by alternative splicing and contain variable NH₂-terminal regions, also give rise to versatility in the actions of cTnT.⁴² There are no detailed studies as yet on interactions of cTnT with other thin-filament proteins. Extrapolations from studies with fsTnT must be made with caution. Although there are extensive regions of homology among the various forms of cTnT and fsTnT, there are important regions of structural diversity, especially in the NH₂-terminal extension, which is highly charged and essentially absent in fsTnT.^{2 3 42} As illustrated in Figure 2 and based largely on studies with fsTnT, it is apparent that both the NH₂- and COOH-terminal halves of cTnT bind to actin-Tm with high affinity.^{1 2 3} The NH₂-terminal half of cTnT appears to anchor the cTn complex to Tm independently of the Ca²⁺ concentration surrounding the myofilaments, whereas the COOH-terminal half forms a complex with cTnC, cTnI, and Tm.^{43 44 45} The COOH-terminal end of fsTnT represented by residues 159 to 259 (TnT2) binds to the region containing Cys190 in Tm, and this interaction is weakened when Ca²⁺ binds to fsTnC.⁴³ In studies of the role of fsTnT in the regulation of myofilament activation, Schaertl et al⁴⁵ showed that TnT1 (residues 1 to 159) did not affect the rate of S-1 binding to actin in preparations reconstituted from myosin S-1 and regulated thin filaments. However, TnT1 was able to influence the size of the cooperative unit from 6 to 9 actins to 12 actins. This fits generally with earlier data showing that TnT1 binds at the region of overlap between contiguous Tm molecules. On the other hand, whereas TnT2 had no effect on cooperativity, it did impart Ca²⁺ sensitivity to the rate of S-1 binding to actin-Tm. Interestingly, in the absence of Ca²⁺, both TnT2 and fsTnT inhibited the initial binding of S1 to a similar extent. A delay in the rate of binding provided evidence for a blocked state of crossbridges.

PKC-dependent sites of phosphorylation on cTnT are also located in the COOH-terminal half of the molecule (Figure 2). Functionally significant sites of PKC-dependent phosphorylation are at Thr199 and Thr280 (in the bovine sequence).⁴⁶ It is this region of cTnT that interacts with cTnI and cTnC and with Tm (Figure 2) and is crucial to transmission of the Ca²⁺-binding signal to switching on and possibly potentiating the thin filament. When these sites were phosphorylated, there was a depression in the maximum ATPase rate of reconstituted thin-filament preparations activating myosin S-1 with no significant change in the pCa-ATPase activity relation.⁴⁶ To date there have been no detailed studies determining the relative significance of phosphorylation of the various PKC sites on cTnT in these inhibitory effects. Regions of interaction between cTnI and cTnT remain largely unknown. In the case of fsTnI, it was proposed that residues 57 to 107, which contain a heptad repeat, may interact with a similar segment in fsTnT, forming a coiled coil.⁴³ Chemical reactivity studies support this hypothesis by the demonstration that Cys and Lys residues in the fsTnI_40–98 region are affected by binding to fsTnT.⁴⁴ However, studies with a truncated mutant of fsTnI missing its first 57 amino acids (TnId57) have also identified sites in the NH₂-terminal region of fsTnI that react with the C terminus of fsTnT.³⁴ TnId57 retains binding to fsTnC but does not bind to fsTnT. When this mutant was reconstituted into myofilaments containing all the other thin-filament components except fsTnT, Ca²⁺ sensitivity was retained, presumably through the TnC-TnI interaction, but the maximum ATPase rate was lower than that for the native myofibrils. However, when the myofilaments were fully reconstituted by adding fsTnT, the ATPase rate was restored to that of the native preparations, even though interactions between fsTnI and fsTnT were disrupted. On the basis of these observations, Potter et al³⁴ proposed that fsTnC has a dual role in regulating myofilament activity: (1) release of the myofilaments from inhibition by TnI and (2) potentiation of myofilament activity by a TnC-TnT interaction. Whether cardiac myofilaments demonstrate the same regulatory device awaits further experiments. We^{32 35} have used truncated mutants to identify regions of interaction between cTnI and cTnT. cTnI missing the first 32 residues is able to react with its neighbors in the thin filament normally and can fully restore Ca²⁺ sensitivity when added with cTnC to myofilament preparations lacking native cTnI-cTnC. Thus, the unique NH₂-terminal extension of cTnI (at least in its dephosphorylated state) does not appear to interact with cTnT. We³⁵ also showed that cTnI_53–211 was able to bind to cTnT but could not restore Ca²⁺ sensitivity when reconstituted into myofilaments. However, removal of an additional 26 amino acids resulted in a form of cTnI (cTnI_80–211) that could no longer bind to cTnT but could partially restore Ca²⁺ sensitivity. These results indicate that a region in the near NH₂-terminal region of cTnI may be repulsive in its interaction with cTnT and thus may influence myofilament regulation differently from the fsTnT-fsTnI interaction.

Summary of Molecular Events Triggering the Diastolic/Systolic Transition
Our hypothesis (Figure 1) is that during diastole a substantial proportion of crossbridges is physically blocked from reacting with the thin filament, whereas the rest are in a weak binding state. Force-generating interactions between crossbridges and the thin filament are inhibited by the position of Tm and possibly cTnI and cTnT and by a Tn-induced depression of the reactivity of domains on both actin and Tm. In diastole, cTnT binds at its COOH-terminal end to a central region of Tm and at its NH₂-terminal end to overlapping Tm molecules. cTnI binds tightly at its NH₂-terminal end to the COOH terminus of cTnT and to the COOH terminus of cTnC. cTnI also interacts with the NH₂-terminal and possibly COOH-terminal regions of actin to inhibit the actin-myosin interaction through the cTnI inhibitory peptide and downstream sites (cTnI_152–199). In systole, Ca²⁺ binding to the regulatory site of cTnC induces a state of high affinity between the COOH terminus of cTnI and the NH₂ terminus of cTnC that extends both cTnC and cTnI. Associated with the interaction between cTnI_152–199 and cTnC, cTnI moves on the thin filament to a new position, and there is a release of the cTnI inhibitory peptide from its interaction with actin. The interaction between the COOH-terminal region of cTnT and Tm is weakened; Tm moves on the thin filament, as do the subdomains of actin. The movements and state changes of the thin-filament proteins release the blocked state and promote the rate of transition of weakly bound crossbridges to the force-generating state. The myofilaments are switched on.

	Molecular Interactions Sustaining and Potentiating Myofilament Activation

Top Abstract Activation and the Structure... Crossbridges at Rest and... Molecular Interactions... Molecular Interactions... Tn-Tm and Tuning Myofilament... Concluding Remarks References

 
There is universal agreement that the change in the state of the crossbridges that occurs with the diastolic/systolic transition is triggered by Ca²⁺ binding to cTnC, yet what happens next to sustain and to modulate myofilament activation is not clear. Major questions are the following: Is Ca²⁺ alone able to fully activate the myofilaments, or does full activation involve a potentiated state requiring the feedback effects of strongly bound crossbridges? What are the roles of long- and short-range cooperative effects on activation? How many actins are under the control of one Tn?

Prominent models of the activation process include (1) the classical 2-state model in which Ca²⁺ activation removes a steric block and in which crossbridges react with actin in an all-or-none fashion,¹⁴ (2) an alternative 2-state model in which Ca²⁺ regulates a kinetic step in the turnover of crossbridges resulting in a graded activation mechanism,⁴⁷ and (3) a 3-state model^{19 48} in which Ca²⁺ acts directly to determine how many crossbridges are in a blocked population ("off state") and indirectly as a cofactor to determine the ability of weakly bound crossbridges ("closed state") to induce an "on state" of actin-Tm leading to strong crossbridges. These models have made important contributions to our perception and understanding of the activation process, but they fall short in terms of accounting for important observations in the literature, especially cooperative interactions within and between functional units.

However, there are appealing alternative 3-state models and multistate models^{2 11 49} that incorporate near-neighbor cooperative interactions between bound crossbridges and Tn and cooperative interactions between crossbridges and Tn-Tm along the thin filament and that are able to accommodate both steric and kinetic effects of activation. These models stem in part from the early work of Bremel et al,⁵⁰ who reported that rigor crossbridge binding to the thin filament induced a Tm-dependent potentiated state of actin as measured by the ATPase rate. It is now generally accepted that tightly bound crossbridges increase the affinity of TnC for Ca²⁺, the affinity of the thin filament for myosin, and the kinetics of transition of weak crossbridges to the force-generating state.^{47 51} Strongly bound crossbridges appear to induce a new (potentiated) state of actin-Tm by distortion of actin, induction of movements in Tm beyond that produced by Ca²⁺ alone, and alteration of the affinity of Tn binding to the thin filament.^{2 12} These actions of strongly bound crossbridges spread activation on the thin filament to near-neighbor units, most likely through interactions between contiguous actins and Tms along the thin filament. A recent model formulated by Campbell⁴⁹ also suggested that the kinetics of crossbridge turnover could be affected by variations in the strength of the feedback interactions between force-bearing crossbridges and activation. These newer models suggest that myofilaments can be significantly activated by Ca²⁺ alone, which fits with the large structural changes¹⁵ of the thin filament induced by Ca²⁺, but also indicate that full activation requires the reaction of strongly bound crossbridges. Resting myofilaments could be composed of blocked (off) or weakly bound crossbridges (closed), as suggested in the theory of Geeves and Lehrer.^{19 48} Moreover, these hypotheses also consider that activation of force generation could be determined by the rate of transition of crossbridges from an off state to a blocked state to an open state and that these rates change with activation. A related issue is how many actins are under the control of one Tn. The consensus from a number of studies^{1 2 11 19} explicitly addressing this question is that the number may be at least 12 to 14 actins, which is at least twice the number expected from the stoichiometric ratio of 1 Tn to 7 actins.

In summary, we think that during basal physiological states after the release of Ca²⁺ into the myofilament space, only a fraction of the Tn sites for Ca²⁺ are occupied. These Ca²⁺-Tn complexes switch on a fraction of functional units on the thin filament. The fraction could be quite small in that 1 Ca²⁺-Tn complex may turn on as many as 14 actins. The Ca²⁺-Tn complex thereby releases blocked and weakly bound crossbridges from their inhibited state. In turn, crossbridge binding potentiates thin-filament activation beyond that produced by Ca²⁺ alone. This model accounts for the highly cooperative dependence of force on Ca²⁺ in the face of weakly cooperative binding of Ca²⁺ to cTnC. This mechanism also has the thermodynamic advantage of amplifying the Ca²⁺ signal to the myofilament. Compared with a mechanism by which activation would depend on myofilament Ca²⁺ binding in a linear manner, this mechanism reduces the energetic cost of Ca²⁺ transport by the sarcoplasmic reticulum.

	Tn-Tm and Tuning Myofilament Activity to Hemodynamic Demands

Top Abstract Activation and the Structure... Crossbridges at Rest and... Molecular Interactions... Molecular Interactions... Tn-Tm and Tuning Myofilament... Concluding Remarks References

 
Our hypothesis is that activation and potentiation of myofilament activity are coupled to prevailing hemodynamic demands and beating frequency. Central questions in testing this hypothesis are as follows: Is modulation of the activation and potentiation of myofilament activity an important determinant of the dynamics of contraction and relaxation of the heart? If so, what are the cellular and molecular mechanisms that allow the myofilaments to "sense" changes in venous return, blood pressure, and heart rate? How do they affect the dynamics of contraction and relaxation of the heart?

There is substantial evidence that kinetic properties of the myofilaments affect the dynamics of cardiac contraction and relaxation. Two main kinetic steps could be involved: (1) the rate of crossbridge cycling and (2) the kinetics of activation and potentiation of thin-filament activity, including the rates of Ca²⁺ binding and release from Tn. Shifts in populations of slow and fast ventricular myosin isoforms are well known and generally accepted to be an important determinant of the rates of contraction and relaxation of the heart. However, there is also evidence that the kinetics of thin-filament processes are also able to affect contraction/relaxation dynamics. For example, we have found in a transgenic mouse model in which ß-Tm has specifically replaced the native -Tm in the myofilaments that there is an increase in myofilament Ca²⁺ sensitivity.²⁸ Myocardium from these same animals demonstrates a reduced rate of relaxation, as reflected in a change in -dP/dt in a working heart preparation.⁵² This result provides indirect evidence that the kinetics of contraction and relaxation could be affected by changes in myofilament activation and deactivation kinetics in addition to changes in crossbridge kinetics. Interestingly, there is also evidence that changes in TnC Ca²⁺-binding kinetics can influence the rate of crossbridge force development. Regnier et al⁵³ reported that an increase in the affinity of Ca²⁺ for binding to fsTnC induced by calmidazolium, which acts specifically on TnC,⁵⁴ also resulted in an increase in the kinetics of crossbridge attachment. Prominent mechanisms by which the steps in cardiac myofilament activation and potentiation processes could be coupled to prevailing hemodynamic state and beating frequency include (1) coupling by the neurohumoral state, as signaled through protein kinase/phosphatase pathways and protein phosphorylation, and (2) coupling by the prevailing mechanical strain, as signaled by length and load.

Modulation of Myofilament Activation by Protein Phosphorylation
One mechanism by which the myofilaments participate in the signaling processes that occur with hemodynamic changes is by protein phosphorylation. Myofilament proteins are substrates for PKA, PKC, Ca²⁺-calmodulin–dependent protein kinases, and kinases similar to casein kinase.¹ In vitro studies have identified many of the functional changes associated with phosphorylation of these myofilament proteins (see Solaro and Van Eyk¹ for review), yet the exact impact of these functional alterations on the inotropic state and dynamics of contraction and relaxation remains poorly understood. A main problem in addressing this question is that manipulation of the level of phosphorylation of myofilaments, eg, by ß-adrenergic agonists in beating hearts, involves changes in phosphorylation of other functionally significant proteins. Moreover, multisite phosphorylation of a particular protein also complicates mechanistic studies.

One of the potentially most significant pathways for signaling is phosphorylation of the myofilaments by PKA. PKA phosphorylates both cTnI and C protein, which to our knowledge are the only 2 myofilament proteins that are phosphorylated by PKA in vitro and by ß-adrenergic stimulation of the heart in situ.¹ The proposal that cTnI phosphorylations may be related to the relaxant effect of catecholamines arose from data showing that PKA-dependent phosphorylation of cTnI reduced the myofilament sensitivity to Ca²⁺ and increased the "off rate" for Ca²⁺ exchange with cTnC.¹⁰ By exchanging native cTnI with a mutant cTnI lacking the PKA phosphorylation site, we^{1 55} showed that phosphorylation of cTnI by PKA is both necessary and sufficient to induce the reduction in the Ca²⁺ sensitivity of myofilament activity. Moreover, in myocytes of mutant mouse hearts in which phospholamban was knocked out, we found that the relaxant effect of catecholamines was retained.⁵⁶ Whether this was due to cTnI phosphorylation is not clear, but it is apparent that the only other PKA-dependent process that could account for this effect is cTnI phosphorylation. Unfortunately, recent experiments using a photolabile Ca²⁺ chelator were not able to resolve the issue of the role of cTnI phosphorylation in the increased rate of cardiac relaxation. Zhang et al⁵⁷ reported that myofilaments containing phosphorylated cTnI relax faster than dephosphorylated controls after release of the chelator, diazo-2. However, similar experiments by Johns et al⁵⁸ did not confirm these results.

There is also conflicting evidence on whether PKA-dependent phosphorylation of the myofilaments affects crossbridge cycling.^{59 60 61 62} Some laboratories report an increase in crossbridge cycling with PKA-dependent phosphorylation,^{59 60} whereas others find no effect.^{61 62} To our knowledge, there is no clear evidence that phosphorylation of C protein is able to affect either the rate of crossbridge cycling or the kinetics of the thin-filament activation processes in the myofilament lattice. It is apparent that some of these controversies will be resolved through the use of mutagenesis and transgenesis. These powerful approaches provide important experimental tools to study isolated and intact preparations in which specific changes in the myofilament proteins have been generated.

Length and Load Dependence of Myofilament Activation
Feedback control of myofilament activation by crossbridge connections provides a mechanism by which cardiac function may be tuned to prevailing hemodynamic conditions that determine preload and afterload.^{1 11 63} Variations in the numbers of crossbridges reacting with the thin filament occur as a function of sarcomere length (filament overlap and interfilament spacing) and as a function of the velocity of shortening. Length-dependent myofilament activation is an important determinant of the length-tension relation and is currently the dominant theory for the cellular basis of Starling's law.^{1 11 63} Experiments in intact preparations and detergent-extracted bundles of heart cells demonstrated that the pCa-force relation at relatively short sarcomere lengths was shifted to the right of the relation obtained at longer sarcomere lengths.⁶³ The length dependence of myofilament activation is a general property of striated muscle but is especially prominent in heart cells that operate at submaximal activation. The consequence is that the force falls more steeply as sarcomere length decreases from an optimum of 2.2 µm than would be expected from the change in filament overlap. Why the relative force that myofilaments develop becomes less sensitive to Ca²⁺ as sarcomere length decreases remains controversial. One hypothesis is that cTnC is the "length sensor."⁶⁴ This theory is based on experiments in which exchange of cTnC with fsTnC endowed bundles of detergent-extracted heart cells with a diminished length dependence of activation, presumably characteristic of fast skeletal muscle fibers.^{64 65} Complementary studies in fast skeletal fibers in which fsTnC was exchanged with cTnC showed the converse effect but required measurements at a low unphysiological ionic strength.⁶⁵ However, these findings have not been confirmed in studies comparing force developed by single cardiac cells obtained from wild-type mice and transgenic mice overexpressing fsTnC in the heart at physiological ionic strengths.⁶⁶ These studies showed that the wild-type myocytes had the same length dependence of activation as did the transgenic myocytes. Moreover, Moss et al⁶⁷ reported that exchange of cTnC into psoas skeletal muscle fibers did not alter the length dependence of Ca²⁺ sensitivity of tension development.

Another issue (apart from the issue of whether cTnC is the length sensor) is whether there is a difference in length-dependent activation between fast skeletal and heart myofilaments in the first place. There is a need for well controlled comparisons taking into account levels of protein phosphorylation and profiles of protein isoforms. For example, after a reduction in sarcomere length from 2.2 µm to 1.8 µm, rat heart myofilaments demonstrate a change in pCa₅₀ of 0.18 pCa units⁶⁸; mouse heart myofilaments, a change of 0.12 pCa units⁶⁶; and fast skeletal muscle, a change of 0.24 pCa units.⁶⁹ In a study comparing length dependence of Ca²⁺ sensitivity in fast versus slow skeletal myofilaments, it was shown that for a similar length change, the change in pCa₅₀ was 0.24 pCa units for fast myofilaments and 0.12 pCa units for slow myofilaments. However, over this same range of length changes, another study reported a change in pCa₅₀ of 0.05 pCa units for fast skeletal myofilaments and a similar change of 0.13 pCa units for both cardiac and slow skeletal myofilaments.⁶⁵ Until such comparisons are made with fast, slow, and cardiac myofilaments under well controlled conditions, it is difficult to make definitive conclusions regarding the mechanisms for the magnitude of the length dependence of activation.

In any case, we believe that whereas cTnC may participate in a special way in the length dependence of activation, particularly if the isoforms of its neighbors on the thin filament change, it seems less likely that TnC is the length sensor in striated muscle. More complicated mechanisms appear involved, as revealed by studies comparing length-dependent activation of cardiac and slow skeletal myofilaments, which contain the same isoforms of TnC, myosin heavy chain, and myosin light chain. Both muscle types demonstrate length-dependent activation, but only cardiac muscle preparations show a change in TnC Ca²⁺ affinity with length.^{70 71}

A unifying hypothesis for length-dependent activation is that a change in sarcomere length at constant volume involves a change in interfilament spacing that modulates the ability of crossbridges to react with thin filaments at the same Ca²⁺ ionic concentration.⁷² Experiments showing an increased Ca²⁺ sensitivity in skinned fibers with shrinkage of the lattice by osmotic compression provided support and a basis for this theory.⁷³ A likely mechanism for the change in force with interfilament spacing is a change in the rate of transition (k_tr) of crossbridges from weak to strong binding states.⁶⁹ McDonald et al⁶⁹ reported a fall in k_tr when sarcomere length was reduced from 2.3 to 2.0 µm in both fast and slow skeletal muscle cells. These changes could be mimicked by osmotic compression of the cells at constant sarcomere length of 2.0 µm. Thus, lateral spacing of the myofilaments associated with length changes appears to be the main determinant of the length dependence of k_tr. Yet, the magnitude of the shift in Ca²⁺ sensitivity was significantly greater in fast-twitch than in slow-twitch cells, as was the effect of length change on the change in k_tr. The mechanism for the differential effects of length on activation between these fiber types is likely to be due to differences in the ability of strong crossbridges to produce feedback effects and cooperatively affect myofilament activation.⁶⁹ Structural and functional differences among thin-filament proteins, especially TnI, TnT, and/or Tm, may determine the ability of strong crossbridges to influence TnC Ca²⁺ binding or to potentiate (cooperatively activate) the myofilaments and, therefore, influence the probability that crossbridges will make the transition from the blocked or weak binding state to the strong binding state as interfilament spacing is changed. Akella et al,⁶⁸ for example, reported that shifts in cTnT isoform population in myofilaments from the diabetic rat heart were associated with a change in the length dependence of activation. Control myofilaments contained 10% of the shorter and less acidic isoform TnT3 compared with 30% in the diabetic myofilaments. This relatively small change in TnT isoform population was associated with doubling of the length-induced changes in Ca²⁺ sensitivity in the diabetic versus the control myofilaments. Moreover, Komukai and Kurihara⁷⁴ have presented data suggesting a role for TnI phosphorylation in the length dependence of activation. This occurred because the desensitizing effect of isoproterenol treatment was larger at shorter muscle lengths that at the maximal length in ferret papillary muscle preparations in which intracellular Ca²⁺ was measured by the aequorin technique. Thus, after stimulation of the preparations with isoproterenol, the length-dependent change in the Ca²⁺-tension relation was amplified. Although it is not clearly defined in these studies, it seems likely that the state of phosphorylation of cTnI rather than C protein was responsible for this effect.

	Concluding Remarks

Top Abstract Activation and the Structure... Crossbridges at Rest and... Molecular Interactions... Molecular Interactions... Tn-Tm and Tuning Myofilament... Concluding Remarks References

 
Contraction and relaxation are so vital to survival that they are exquisitely controlled to accommodate the many-fold increases in the heart rate and venous return that occur with exercise, all the while maintaining optimal changes in end-diastolic volume. Apart from the essential switch in which Ca²⁺ binding to cTnC turns on activity of the myofilament contractile machine, it is clear that there are layers of mechanisms all related to tuning the contractile state of the heart to the venous return and heart rate. The impressive and extensive array of pumps, channels, and exchangers that regulates the flows of Ca²⁺ to and from the myofilaments gives testimony to this statement. In the present review, though, we have emphasized the mechanisms that modulate changes in the myofilament response to Ca²⁺. These mechanisms are significant in the intrinsic control of the heart by Starling's law and extrinsic control by neurohumoral mechanisms. Although ischemia, reperfusion injury, stunning, and heart failure are not stressed in this review, it is important to note that they appear likely to involve changes in the myofilament response to Ca²⁺ as well as altered myofilament structure.^{75 76 77} Isoform shifts of myofilament proteins also modulate the response to Ca²⁺ in physiological and pathological states.⁴² There is also solid emerging evidence indicating that familial hypertrophic cardiomyopathy is a "sarcomeric disease" genetically linked to missense mutations in thick- and thin-filament proteins.⁷⁸ One of the latest additions to this list is a mutation in the functionally significant regions of the COOH terminus of cTnI.⁷⁹ Targeting the myofilaments for the action of inotropic agents also remains an important objective in the development of experimental tools and inotropic agents.⁸⁰ It is clear that the understanding and investigation of myofilament regulation offer an excellent example of the confluence of information on the atomic, molecular, and macromolecular structure and the integrated biology, pharmacology, and pathology of a vital organ.

	Acknowledgments

 
Work described in this review was supported in part by National Institutes of Health research grants HL-22231 and HL-49934 to Dr Solaro and by fellowship grant F32 HL-09009 to Dr Rarick. The authors gratefully acknowledge their many fine colleagues, who have contributed to the science summarized here.

Received January 6, 1998; accepted June 16, 1998.

	References

Top Abstract Activation and the Structure... Crossbridges at Rest and... Molecular Interactions... Molecular Interactions... Tn-Tm and Tuning Myofilament... Concluding Remarks References

 

1. Solaro RJ, Van Eyk J. Altered interactions among thin filament proteins modulate cardiac function. J Mol Cell Cardiol. 1996;28:217–230.[Medline] [Order article via Infotrieve]
2. Tobacman LS. Thin filament-mediated regulation of cardiac contraction. Annu Rev Physiol. 1996;58:447–481.[Medline] [Order article via Infotrieve]
3. Farah CS, Reinach FC. The troponin complex and regulation of muscle contraction. FASEB J. 1995;9:755–767.[Abstract]
4. Holmes KC, Popp D, Gebhard W, Kabsch W. Atomic model of the actin filament. Nature. 1990;347:44–49.[Medline] [Order article via Infotrieve]
5. Olah GA, Rokop SE, Wang C-LA, Blechner SL, Trewhella J. Troponin I encompasses an extended troponin C in the Ca²⁺-bound complex: a small-angle X-ray and neutron scattering study. Biochemistry. 1994;33:8233–8239.[Medline] [Order article via Infotrieve]
6. Milligan RA, Flicker PF. Structural relationships of actin, myosin, and tropomyosin revealed by cryo-electron microscopy. J Cell Biol. 1987;105:29–39.[Abstract/Free Full Text]
7. Gagne SM, Tsuca S, Li MX, Smillie LB, Sykes BD. Structure of the troponin C regulatory domains in the apo and calcium-saturated states. Nat Struct Biol. 1995;2:784–789.[Medline] [Order article via Infotrieve]
8. Spyracoupoulos L, Li MX, Sia SK, Gagne SM, Chandra M, Solaro RJ, Sykes BD. Calcium-induced structural transition in the regulatory domain of human cardiac troponin C. Biochemistry. 1997;36:12138–12146.[Medline] [Order article via Infotrieve]
9. Kleerekoper Q, Howarth JW, Guo X, Solaro RJ, Rosevear PR. Cardiac troponin I induced conformational changes in cardiac troponin C as monitored by NMR using site-directed spin and isotope labeling. Biochemistry. 1995;34:13343–13352.[Medline] [Order article via Infotrieve]
10. Robertson SP, Johnson JD, Holroyde MJ, Kranias EG, Potter JD, Solaro RJ. The effect of troponin I phosphorylation on the Ca²⁺-binding properties of the Ca²⁺-regulatory site of bovine cardiac troponin. J Biol Chem. 1982;257:260–263.[Free Full Text]
11. Moss RL. Ca²⁺ regulation of mechanical properties of striated muscle: mechanistic studies using extraction and replacement of regulatory proteins. Circ Res. 1992;70:865–884.[Abstract/Free Full Text]
12. Pan BS, Solaro RJ. Calcium-binding properties of troponin C in detergent-skinned heart muscle fibers. J Biol Chem. 1987;262:7839–7849.[Abstract/Free Full Text]
13. Kavaler F, Morad M. Paradoxical effects of epinephrine on excitation-contraction coupling in cardiac muscle. Circ Res. 1966;18:492–497.[Abstract/Free Full Text]
14. Kress M, Huxley HE, Faruqi AR, Hendrix J. Structural changes during activation of frog muscle studied by time resolved X-ray diffraction. J Mol Biol. 1986;188:325–342.[Medline] [Order article via Infotrieve]
15. Lehman W, Vibert P, Uman P, Craig R. Steric-blocking by tropomyosin visualized in relaxed vertebrate muscle thin filaments. J Mol Biol. 1995;251:191–196.[Medline] [Order article via Infotrieve]
16. Stein LA, Schwarz RB Jr, Chock PB, Eisenberg E. Mechanism of actomyosin adenosine triphosphatase: evidence that adenosine 5'-triphosphate hydrolysis can occur without dissociation of the actomyosin complex. Biochemistry. 1979;18:3895–3909.[Medline] [Order article via Infotrieve]
17. Chalovich JM, Chock PB, Eisenberg E. Mechanism of action of troponin-tropomyosin: inhibition of actomyosin ATPase activity without inhibition of myosin binding to actin. J Biol Chem. 1981;256:575–578.[Abstract/Free Full Text]
18. Xu S, Malinchik S, Gilroy D, Kraft T, Brenner B, Yu LC. X-ray diffraction studies of cross-bridges weakly bound to actin in relaxed fibers of rabbit psoas muscle. Biophys J. 1997;73:2292–2303.[Abstract/Free Full Text]
19. Geeves MA, Lehrer SS. Dynamics of the muscle thin filament regulatory switch: the size of the cooperative unit. Biophys J. 1994;67:273–282.[Abstract/Free Full Text]
20. Rayment L, Holden H, Whittaker M, Yohn CB, Lorenz M, Holmes KC, Milligan R. Structure of the actin-myosin complex and its implications for muscle contraction. Science. 1993;261:58–65.[Abstract/Free Full Text]
21. Hibberd MG, Trentham DR. Relationships between chemical and mechanical events during muscular contraction. Annu Rev Biophys Biophys Chem. 1986;15:119–161.[Medline] [Order article via Infotrieve]
22. Irving M, Piazzesi G. Motions of myosin heads that drive muscle contraction. News Physiol Sci. 1997;12:249–254.[Abstract/Free Full Text]
23. Andreev OA, Borejdo J. Interaction of the heavy and light chains of cardiac myosin subfragment-1 with F-actin. Circ Res. 1997;81:688–693.[Abstract/Free Full Text]
24. Morano I, Ritter O, Bonz A, Timek T, Vahl CF, Michel G. Myosin light chain-actin interaction regulates cardiac contractility. Circ Res. 1995;76:720–725.[Abstract/Free Full Text]
25. Rarick HM, Opgenorth TJ, von Geldern TW, Solaro RJ. An essential myosin light chain peptide stimulates cardiac myofibrillar ATPase activity. J Biol Chem. 1996;271:27039–27043.[Abstract/Free Full Text]
26. Squire JM, Al-Khayat HA, Yagi N. Muscle thin filament structure and regulation: actin subdomain movements and the tropomyosin shift modelled from low angle X-ray diffraction. J Chem Soc Faraday Trans. 1993;89:2717–2726.
27. Lorenz M, Poole KJV, Popp D, Rosenbaum G, Holmes KC. An atomic model of the unregulated thin filament obtained by X-ray fiber diffraction on oriented actin-tropomyosin gels. J Mol Biol. 1995;246:108–119.[Medline] [Order article via Infotrieve]
28. Palmiter KA, Kitada Y, Muthuchamy M, Wieczorek DF, Solaro RJ. Exchange of ß-tropomyosin for -tropomyosin in hearts of transgenic mice induces changes in thin filament response to Ca²⁺, strong cross-bridge binding, and protein phosphorylation. J Biol Chem. 1996;271:11611–11614.[Abstract/Free Full Text]
29. Tao T, Gong B-J, Leavis PC. Calcium-induced movement of troponin-I relative to actin in skeletal muscle thin filaments. Science. 1990;247:1339–1341.[Abstract/Free Full Text]
30. Chandra M, Dong W-J, Pan B-S, Cheung HC, Solaro RJ. Effects of protein kinase A phosphorylation on signaling between cardiac troponin I and the N-terminal domain of cardiac troponin C. Biochemistry. 1997;36:13305–13311.[Medline] [Order article via Infotrieve]
31. Zhang R, Zhao J, Potter JD. Phosphorylation of both serines in cardiac troponin I is required to decrease the Ca²⁺-affinity of cardiac troponin C. J Biol Chem. 1995;270:30773–30780.[Abstract/Free Full Text]
32. Guo X, Wattanapermpool J, Palmiter KA, Murphy AM, Solaro RJ. Mutagenesis of cardiac troponin I: role of the unique NH₂-terminal peptide in myofilament activation. J Biol Chem. 1994;269:15210–15216.[Abstract/Free Full Text]
33. Krudy G, Kleerekoper Q, Guo X, Howarth JW, Solaro RJ, Rosevear PR. NMR studies delineating spatial relationships within the cardiac troponin I-troponin C complex. J Biol Chem. 1994;269:23731–23735.[Abstract/Free Full Text]
34. Potter JD, Sheng Z, Pan B-S, Zhao J. A direct regulatory role for troponin T and a dual role for troponin C in the Ca²⁺ regulation of muscle contraction. J Biol Chem. 1995;270:2557–2562.[Abstract/Free Full Text]
35. Rarick HM, Tang HP, Guo X, Tu X-H, Martin AF, Solaro RJ. Interactions of the N-terminal region of cardiac TnI with the structural cTnC site and cTnT are involved in the Ca²⁺-dependent activation of MgATPase activity [abstract]. Biophys J. 1998;74:A52.
36. Noland TA, Guo X, Raynor RL, Averyhart-Fullard V, Jideama NM, Solaro RJ, Kuo JF. Cardiac troponin I mutants: phosphorylation by protein kinases C and A and regulation of Ca²⁺-stimulated MgATPase of reconstituted actomyosin S-1. J Biol Chem. 1995;43:25445–25454.
37. Van Eyk JE, Kay CM, Hodges RS. A comparative study of the interactions of synthetic peptides of the skeletal and cardiac troponin I inhibitory region with skeletal and cardiac troponin C. Biochemistry. 1991;30:9974–9981.[Medline] [Order article via Infotrieve]
38. Malhotra A, Nakouzi A, Bowman J, Puttrick P. Expression and regulation of mutant forms of cardiac TnI in a reconstituted actomyosin system: role of kinase dependent phosphorylation. Mol Cell Biochem. 1997;170:99–107.[Medline] [Order article via Infotrieve]
39. Rarick HM, Tu X, Solaro RJ, Martin AM. The C-terminus of cardiac troponin I is essential for full inhibitory activity and Ca²⁺-sensitivity of rat myofibrils. J Biol Chem. 1997;272:26887–26892.[Abstract/Free Full Text]
40. Dong W-J, Chandra M, Xing J, Solaro RJ, Cheung HC. Conformation of the N-terminal segment of a monocysteine mutant of troponin I from cardiac muscle. Biochemistry. 1997;36:6745–6753.[Medline] [Order article via Infotrieve]
41. Dong W-J, Chandra M, Xing J, She M, Solaro RJ, Cheung HC. Phosphorylation-induced distance change in a cardiac muscle troponin I mutant. Biochemistry. 1997;36:6754–6761.[Medline] [Order article via Infotrieve]
42. Anderson PAW, Malouf NN, Oakeley A, Pagani ED, Allen PD. Troponin T isoform expression in humans: a comparison among normal and failing adult heart, fetal heart, and adult and fetal skeletal muscle. Circ Res. 1991;60:1226–1233.
43. Pearlstone JR, Smillie LB. Troponin T fragments: physical properties and binding to troponin C. Can J Biochem. 1978;56:521–527.[Medline] [Order article via Infotrieve]
44. Chong PCS, Hodges RS. Photochemical cross-linking between rabbit skeletal troponin subunits: troponin I-troponin T interactions. J Biol Chem. 1982;257:2549–2555.[Free Full Text]
45. Schaertl S, Lehrer SS, Geeves MA. Separation and characterization of the two functional regions of troponin T involved in muscle thin filament regulation. Biochemistry. 1995;34:15890–15894.[Medline] [Order article via Infotrieve]
46. Noland TA Jr, Raynor RL, Kuo JF. Identification of sites phosphorylated in bovine cardiac troponin I and troponin T by protein kinase C and comparative substrate activity of synthetic peptides containing the phosphorylation sites. J Biol Chem. 1989;264:20778–20785.[Abstract/Free Full Text]
47. Brenner B. Changes in calcium sensitivity at the cross-bridge level. In: Allen DAP, Lee JL, eds. The Modulation of Cardiac Calcium Sensitivity. Oxford, UK: Oxford University Press; 1993:197–214.
48. Lehrer S. The regulatory switch of the muscle thin filament: Ca²⁺ or myosin heads? J Muscle Res Cell Motil. 1994;15:232–236.[Medline] [Order article via Infotrieve]
49. Campbell K. Rate constant of muscle force redevelopment reflects cooperative activation as well as cross-bridge kinetics. Biophys J. 1997;72:254–262.[Abstract/Free Full Text]
50. Bremel RD, Murray JM, Weber A. Manifestations of cooperative behaviour in the regulated actin filament during actin-activated ATP hydrolysis in the presence of calcium. Cold Spring Harb Symp Quant Biol. 1972;37:267–275.
51. Swartz DR, Moss RL. Influence of a strong binding myosin analog on calcium sensitive mechanical properties of skinned skeletal muscle fibers. J Biol Chem. 1992;267:20497–20506.[Abstract/Free Full Text]
52. Muthuchamy M, Grupp I, Grupp G, O'Toole B, Kier A, Boivin G, Neumann J, Wieczorek D. Molecular and physiological effects of over expressing striated muscle ß-Tropomyosin in the adult murine heart. J Biol Chem. 1995;270:30593–30603.[Abstract/Free Full Text]
53. Regnier M, Martyn DA, Chase PB. Calimidazolium alters Ca²⁺ regulation of tension redevelopment rate in skinned skeletal muscle. Biophys J. 1996;71:2786–2794.[Abstract/Free Full Text]
54. El-Saleh S, Solaro RJ. Calmidazolium, a calmodulin antagonist, stimulates Ca-calmodulin and Ca-troponin C dependent activation of striated myofilaments. J Biol Chem. 1987;262:17240–17246.[Abstract/Free Full Text]
55. Wattanapermpool J, Guo X, Solaro RJ. The unique amino-terminal peptide of cardiac troponin I regulates myofibrillar ATPase activity only when it is phosphorylated. J Mol Cell Cardiol. 1995;27:1383–1391.