Thin Filament Interaction and Ca²⁺-Desensitization Effect of the C-Terminal End Peptide of Cardiac Troponin T and Loss of Function in Hypertrophic Cardiomyopathy Mutants

Abstract

Troponin T (TnT), the tropomyosin (Tm)-binding subunit of troponin, plays a central role in regulating striated muscle contractility. The recently identified Tm-binding site 3 in the highly conserved C-terminal end segment of TnT has a troponin I (TnI)-like inhibitory function. A restrictive proteolytic removal of the N-terminal hypervariable region (amino acids 1–71) of cardiac TnT (cTnT-ND) in vivo in adaptation to inotropy-afterload mismatch slows down ventricular end-systolic velocity to increase ejection time and stroke volume through a conformational modulation that enhances the function of Tm-binding site 3. Mutations in this segment of cTnT cause hyperactivation of cardiac muscle and hypertrophic cardiomyopathy (HCM). Here we characterized isolated C-terminal end 14 amino acids (275–288) peptide of cTnT (cTnT-C14) for interactions with Tm, F-actin, and F-actin-Tm thin filament using localized surface plasmon resonance (LSPR). Wild-type cTnT-C14 peptide showed saturable strong binding to Tm and F-actin in LSPR, which is significantly weakened by HCM mutation R278C, K280N or R286C. Contractility studies of permeabilized mouse cardiac muscle strips found that the HCM mutant peptides all lost the known Ca²⁺-desensitization function of wild-type cTnT-C14 peptide, correlating to the pathogenesis and pathophysiology of HCM. Treatment of cTnT-ND transgenic mouse cardiac muscle with the phenotype of lower than wild-type maximum force production with wild-type cTnT-C14 peptide did not produce additive effect, implicating that cTnT-C14 peptide and cTnT-ND both utilize the Tm-binding site 3 mechanism to reduce maximum activation while free cTnT-C14 peptide does not diminish the effect of cTnT-ND in troponin complex on increasing the overall sensitivity to Ca²⁺-activation. The functionality of cTnT-C14 in the form of free peptide presents a novel reagent for adjusting contractile kinetics of cardiac muscle and alleviating myofilament hyperactivation.

Graphical Abstract

LSPR analysis of the binding of cTnT-C14 peptides to tropomyosin-F-actin composite thin filament demonstrates that in comparison with the negative control peptide cTnT-E16, human cTnT-C14WT peptide had rapid association and saturable binding to F-actin-Tm thin filament followed by a rapid dissociation whereas three HCM mutant peptides showed diminished binding.

Introduction

Cardiac muscle contraction and relaxation are essential for the pumping function of the heart. The contractile apparatus of vertebrate striated muscles is regulated by Ca²⁺ via the troponin complex in the sarcomeric thin filament. The troponin complex consists of three protein subunits: The Ca²⁺-binding subunit troponin C (TnC), the inhibitory subunit troponin I (TnI), and the tropomyosin (Tm)-binding subunit troponin T (TnT)[1]. During the activation of muscle contraction, cytosolic Ca²⁺ rises to bind TnC and induces a series of conformational changes in troponin and Tm, which in turn allows myosin heads to form strong cross bridges with the actin filament to activate myosin ATPase and a conformational power stroke that forces the sliding of thin filament toward the center of sarcomere to contract the muscle[2].

TnT is at a central position in the thin filament regulation of skeletal and cardiac muscle contraction and relaxation. Extensive structure-function relationship studies have demonstrated that TnT binds TnI, TnC, Tm and F-actin[3]. Earlier research have characterized two Tm-binding sites in the middle region and the beginning of the C-terminal region of TnT[4–6]. A recent study further identified a third Tm-binding site in the C-terminal end segment of TnT encoded by the last exon of vertebrate TnT genes[7]. Supporting functional importance of the new Tm-binding site 3, earlier studies showed that deletion of the C-terminal 14 amino acids of cardiac TnT (cTnT) resulted in myofilament hyperactivation and hypertrophic cardiomyopathy (HCM) phenotypes[8–11].

The C-terminal end segment of TnT has highly conserved amino acid sequences among the cardiac, fast and slow skeletal muscle TnT isoforms and across vertebrate species[3]. A striking feature is that the 14 amino acids C-terminal end segment of cTnT (residues 275–288 of human cTnT) encoded by exon 17 of TNNT2 gene has an identical sequence in all mammalian species (from platypus to human)[7], indicating a stringently selected structure and function. Consistently, the 14 amino acids C-terminal end segment of cTnT is required for full inhibition of myofilament and substitution of the positively charged Arg and Lys residues in this segment with Ala resulted in increased activation of contraction[8,9]. Further demonstrating its functional importance, mutations within this segment, for example R278C, K280N and R286C, cause HCM[12–14].

While TnC, the Ca²⁺-receptor subunit of troponin, belongs to the calmodulin family of Ca²⁺-binding proteins[15], TnT and TnI, the two signal transmitting subunits of troponin, have evolved from the duplication of a TnI-like ancestral gene[16]. While TnT and TnI have significantly diverged in structure for specialized functions, repressed TnI-like ancestral structure is detectable in TnT. Restrictive proteolytic deletion of the evolutionarily added N-terminal variable region (residues 1–71) of TnT to remove its conformational modulation effect reconfigures the molecular conformation of TnT to bring back a TnI-like ancestral structure[16]. This reconfigured structure is recognized by a monoclonal antibody (mAb) raised against the C-terminal end segment of TnI[17]. The C-terminal end segment of cardiac TnI has an inhibitory function critical to the relaxation of cardiac muscle[18]. Like the C-terminal peptide of cardiac TnI[19,20], the TnI C-terminal segment-like ancestral conformation brought back in N-terminal truncated cTnT (cTnT-ND) conveys a structure that binds Tm[7], corresponding to a functional effect on reducing the end systolic velocity of cardiac muscle to elongate the left ventricular rapid ejection time and increase stroke volume as a compensatory adaptation to inotropy-afterload mismatch in vivo[21]. Peptide mapping studies have located this TnI-like inhibitory Tm-binding activity to the 14 amino acids C-terminal end segment of TnT[7].

Following the initial finding of the C-terminal Tm-binding site 3 of TnT[7], the present study further characterized the C-terminal 14 amino acids segment of cTnT (cTnT-C14) in the form of free peptide isolated from TnT backbone. Localized surface plasmon resonance (LSPR) spectroscopy determined the binding activity and kinetics of cTnT-C14 peptide to Tm and F-actin-Tm composite thin filaments. Its effect on Ca²⁺-activation of force production was investigated in permeabilized cardiac muscle sections and compared with that of cTnT-ND. The results provide new insights into the physiological function of the C-terminal Tm-binding site 3 of TnT and establish the functionality of cTnT-C14 free peptide as an inhibitory modulator of the contractile kinetics of cardiac muscle.

2. Materials and Methods

2.1. Amino Acid Sequence Analysis

Amino acid sequences of TnT were retrieved from NCBI GenBank database. The molecular mass and isoelectric point of peptides and their theoretical charge at pH 7.0 was calculated by using DNAStar software.

2.2. Synthetic Peptides

Wild-type (WT) human cTnT-C14 peptide and derivatives containing HCM-linked mutation R278C, K280N or R286C were commercially synthesized by Peptide 2.0, Inc (Chantilly, VA) at >95% purity with HPLC mass spectrometry verification of the anticipated molecular mass. High concentration stocks of the peptides were made by dissolving the dried powder in a buffer containing 0.1 M KCl, 3 mM MgCl₂, 10 mM imidazole-HCl, pH 7.0, and stored at −20°C until use in experiments. A non-Tm-binding and non-actin-binding peptide of the same length (E16) corresponding to the exon 16-encoded segment of human cardiac TnT (INVLRNRINDNQKV) was synthesized for use as a control. The endogenous myofilament origin of E16 peptide avoids potentially toxic or gain-of-function effect of randomly scrambled peptides. Experiments using E16 peptide as negative control also provide evidence to attribute the loss of function caused by the C-terminal 28 amino acids (E16+cTnT-C14)-truncated HCM mutant of cardiac TnT reported previously[10,11] to the loss of cTnT-C14 peptide.

2.3. Preparation of Tropomyosin and F-actin

Acetone powder of rabbit cardiac and back muscles was prepared from frozen tissue stocks. After minced using a stainless-steel grinder precooled at 4 °C and washed in cold water with stirring for 2–3 minutes and standing at 4°C for 20 minutes, the solubilized material was removed by squeezing out through two layers of cheesecloth. The tissue residue was sequentially washed at 4°C with 50% ethanol three times, 95% ethanol twice, and 100% acetone twice with stirring followed by filtering through cheesecloth. The resulting grayish-white acetone-washed residues were spread thinly over filter paper in a fume hood and air-dried at room temperature. The final dried acetone powder was stored at −20°C in aliquots until use.

As described previously [22], α-tropomyosin was purified from cardiac muscle acetone powder by extraction with a buffer containing 1.0 M KCl, 0.5 mM dithiothreitol (DTT), 10 mM imidazole-HCl, pH 7.0, for 16 hours at room temperature with gentle stirring. The solubilized material was collected by squeezing out through two layers of cheesecloth, and the residue was re-extracted with the same buffer for 2 hours. The extracts were combined, adjusted to pH 4.6 using HCl, and stirred at 4°C for 30 minutes. The precipitate was collected by centrifugation at 6,000 × g at 4°C for 20 minutes and dissolved in the extraction buffer with stirring for 20 min. Any insoluble materials were removed by centrifugation at 6,000 × g at 4°C for 10 minutes. The isoelectric precipitation of tropomyosin at pH 4.6 and re-dissolution at pH 7.0 were repeated two more times. The precipitate was dissolved in 0.5 mM DTT, 10 mM imidazole-HCl, pH 7.0, and precipitated on ice by slowly adding solid ammonium sulfate to 53% saturation at 0°C with stirring while maintaining the pH at 7.0 with NaOH. After standing for 30 minutes, the precipitate was removed by centrifugation at 11,000 × g at 4°C for 30 minutes. More solid ammonium sulfate was added to the supernatant to 65% saturation at 0°C while keeping the pH at 7.0, and the precipitate was collected by centrifugation at 11,000 × g at 4°C for 30 minutes. The purified tropomyosin in the final precipitate was redissolved in deionized water containing 0.5 mM DTT and dialyzed against cold water containing 2 mM β-mercaptoethanol with three changes and lyophilized.

As described previously[23], actin was extracted from rabbit back muscle acetone powder with a pre-cooled buffer containing 2 mM Tris-HCl, pH 8.0, 2 mM CaCl₂, 0.2 mM NaATP and 0.5 mM DTT with gentle stirring on ice for 30 minutes. The extraction mix was paper-filtered and centrifuged in a Beckman Coulter Optima XPN-100 Ultracentrifuge at 12,700 × g in a TYPE 50.2 Ti rotor at 4°C for 25 minutes to remove residual materials. KCl and MgCl₂ were gradually added to the supernatant to final concentrations of 50 mM and 2 mM, respectively, to initiate actin polymerization. The solution was gently stirred at room temperature for 1 hour. More KCl was added to reach a final concentration of 0.9 M followed by stirring at room temperature for 2 hours to separate tropomyosin from F-actin. The mixture was then ultracentrifuged at 244,000 × g at 4°C for 3 hours to pellet polymerized F-actin. The F-actin pellet was completely dissolved in a minimal volume of the extraction buffer and dialyzed against the same buffer at 4°C with two changes to depolymerize F-actin. The solution of depolymerized actin was ultracentrifuged at 244,000 × g at 4°C for 2 hours to remove any insoluble materials. The clarified supernatant containing purified G-actin was repolymerized by adding KCl and MgCl₂ to 50 mM and 2 mM, respectively. The final product of F-actin was pelleted by ultracentrifugation at 244,000 × g at 4°C for 3 hours, resuspended in a small volume of the polymerization buffer and stored at 4°C for use within 7–10 days.

The concentration of rabbit α-tropomyosin and F-actin stocks was determined with UV absorbance at 280 nm based on their amino acid sequences.

2.4. SDS-Polyacrylamide Gel Electrophoresis (SDS-PAGE)

To assess the purity of Tm and actin prepared for the present study, SDS-PAGE was carried out as described previously[7,23]. The resolving gel is 14% with an acrylamide/bis-acrylamide ratio of 180:1. The protein samples were prepared in an SDS-gel sample buffer containing 2% SDS and 1% β-mercaptoethanol. The samples were heated at 80°C for 5 minutes and clarified by centrifugation in a microcentrifuge at top speed (20,000 × g) at room temperature for 5 min before being loaded to the gel. Electrophoresis was run at a constant current of 22 mA per 0.75 mm thick BioRad mini-gel. The gel was then stained with Coomassie Brilliant Blue R-250, distained, and scanned at 600 dpi for image documentation.

2.5. Localized Surface Plasmon Resonance (LSPR) Assay

LSPR spectroscopy was carried out using a Nicoya Lifesciences LSPR system to study the interaction of WT or mutant cTnT-C14 peptides with Tm or F-actin-Tm composite thin filaments. Nicoya High Sensitivity Carboxyl LSPR sensor chip pre-coated with carboxyl-modified gold nanoparticles was used. The LSPR system was first flushed with 80% isopropanol at the maximum flow rate of 150 μL/min to remove any air bubbles. To activate the carboxyl groups on the gold nanoparticle surface before the coating of Tm or F-actin, 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) and N-hydroxysuccinimide (NHS) were dissolved separately in deionized water both at the concentration of 0.1 M. The EDC/NHS activation solution was made freshly by 1:1 mixing of the stocks immediately prior to use to ensure optimal activation efficiency. The EDC/NHS activation solution was injected at a flow rate of 20 μL/min for 4 minutes. The protein to be studied (Tm or F-actin) was immobilized on the chip surface by flowing the protein solution at 20 μL/min for 4 minutes for dehydration condensation of carboxyl groups on the chip with the amino groups of the protein.

For cTnT-C14 peptide-Tm binding assays, purified Tm was dissolved at 50 μg/mL in 10 mM sodium acetate, pH 4.0. It is important to have the pH at least 0.5 units below the isoelectric point of Tm (4.6) to ensure its positive overall charge. After Tm immobilization and equilibration in LSPR assay buffer consisting of 0.1 M KCl, 3 mM MgCl₂ and 10 mM imidazole-HCl, pH 7.0, 1 M ethanolamine was injected at 20 μL/min for 4 minutes to block the unreacted carboxyl groups on the surface of gold particle on the chip. After re-equilibration with LSPR assay buffer, 1 mM 6-mercapto-1-hexanol (MCH) in 30% EtOH was injected at 20 μL/min for 12 min to block exposed gold nanoparticles and prevent nonspecific interaction with free cysteine in protein samples. cTnT-C14 or control peptide diluted at 10 μM in the LSPR assay buffer was flowed over the chip at a rate of 20 μL/min for 4 minutes to record the course of association. The flow was then switched to the LSPR assay buffer at the same rate to elute the bound peptide and the course of dissociation was recorded until the signal level was back to baseline or reaching a plateau with no further decrease. The Tm-coated chip was regenerated between assays by washing with the LSPR assay buffer or with 1 M KCl and 3 mM MgCl₂ at high flow rate for removing peptides with a strong binding to return the signal to the baseline level.

For measuring the binding of cTnT-C14 to F-actin-Tm composite thin filament, F-actin was coated at a relative low density and decorated with a high concentration of Tm aiming for a close to saturated binding. F-actin was coated to the sensor chip same as that for Tm except for at the protein concentration of 30 μg/mL. After blocking the unreacted carboxyl groups and exposed gold nanoparticles on the surface of the sensor chip as above, 10 μM Tm in the LSPR buffer was applied to the F-actin-coated chip by flowing in at the rate of 20 μL/min for 12 minutes, allowing close to saturation Tm binding to immobilized F-actin. After a 10 minutes wash with the LSPR buffer at 20 μL/min to remove free Tm, 10 μM cTnT-C14 peptide was injected at 20 μL/min flow rate for 4 minutes to measure its association to the immobilized F-actin-Tm thin filament. The flow was then switched to LSPR buffer and the dissociation of cTnT-C14 peptide was recorded. To regenerate the F-actin coated chip between assays, a 150 μL/min high-speed flow of the regeneration buffer containing 1 M KCl and 3 mM MgCl₂ was applied to remove any residual Tm and TnT-C14 peptide.

The direct binding of cTnT-C14 peptides to LSPR chip-immobilized F-actin was measured similarly to the cTnT-C14-Tm binding assay.

cTnT-C14 peptide with HCM-linked amino acid substitution R278C or R286C contains a cysteine residue which can form dimers via disulfide bonding. To prevent dimerization that could affect the binding properties for Tm or F-actin, 1 mM Tris(2-carboxyethyl)phosphine (TCEP) was added in the R278C and R286C stock solution at least 1 hour before testing to reduce disulfide bonds. After diluting the peptide to 10 μM in the LSPR working solution the final TCEP concentration was 100 μM, which did not generate any background signal while ensuring any disulfide bonds were reduced.

2.6. Measurement of Ca²⁺-Activated Force of Permeabilized Mouse Cardiac Muscle Sections

The animal procedure was carried out using a protocol approved by the Animal Care Committees of University of Illinois at Chicago and was conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Left ventricular papillary muscles from 3–4 months old male and female WT mice or transgenic mice over-expressing cTnT-ND driven by the alpha-MHC promoter to replace ~70% of the endogenic cTnT was described previously[21], both are of full C57B/L6 background, were isolated immediately after euthanasia and flash frozen at their slack length. Using the method described previously[24], longitudinally sliced 120–150 μm wide 35 μm thick cardiac muscle strips were cryosectioned. Glycerol-permeabilized cardiac muscle strips were mounted between two aluminum T-clips and transferred to a chamber on a thermo-controlled stage (802D, Aurora Scientific) at 6–8°C containing a relaxation buffer (BES 40 mM, EGTA 10 mM, MgCl₂ 6.86 mM, ATP 5.96 mM, DTT 1 mM, creatine phosphate 33 mM, creatine kinase 200U/mL, K-propionate 3.28 mM, pH 7.0, plus protease inhibitor cocktail). For contractility studies, the muscle strip was connected to a force transducer (403A, Aurora Scientific) at one end and a length controller (322C, Aurora Scientific) at the other end. The buffer was then switched to a skinning solution (relaxation buffer plus 1% Triton X-100) for 20 min. After a wash with relaxation buffer, the fully permeabilized muscle strip was placed in pCa 9.0 buffer and the sarcomere length was measured through a digital camera attached to the microscope and adjusted to 2.0 μm or 2.3 μm for mechanical force measurements. After measuring the baseline calcium activated force at pCa 6.5, 6.3, 6.0, 5.8, 5.5, 5.0, and 4.5 at 15°C, WT or HCM mutant cTnT-C14 peptides was added at 20 μM and the force-pCa measurements repeated. The force development of the muscle strip at pCa 4.5 was then measured again to verify the absence of any significant rundown effect. If the force was lower than 80% of beginning force, the data were disqualified for inclusion in the analysis. The peptides were made as 10x (200 μM) stocks in each of the different pCa buffers. The muscle strips were pre-incubated in 20 μM peptide at pCa 9.0 for at least 10 min. To maintain constant treatment of the peptide throughout the experiment, one-tenth of the buffer in the pCa testing chamber was removed before adding the same volume of 10x peptide stock in the same pCa buffer to the working concentration of 20 μM.

2.7. Data Analysis

The differences between the LSPR curves were analyzed using two-way ANOVA. The LSPR association (k_on) and dissociation (k_off) rate constants were obtained by fitting the curves using 1:1 binding model in TraceDrawer software. The binding affinity (KD) was calculated as k_off/k_on. In the fitting process using TraceDrawer, the software internally determines the apparent rate constant (k_obs) based on the association phase, which follows the relationship: k_obs = k_on· C + k_off where C is the analyte concentration (in M). The value of k_on is then derived from k_obs using the known analyte concentration and the fitted k_off. Therefore, even when the analysis is performed at a single concentration, the reported k_on retains the proper units of M⁻¹·s⁻¹, rather than s⁻¹. The LSPR curves were compared for statistical significance using two-way ANOVA followed by post hoc multiple comparisons (Tukey test). Statistical comparisons of LSPR-determined Δpm, k_on, k_off, and KD were done using unpaired Student’s t test. The force-pCa curves were plotted with the force at pCa 4.5 and 100% and fitted using a Hill exponential equation (y = START + (END-START)*x^n/(k^n+x^n)) for data analysis. Statistical comparison of maximum force and pCa₅₀ was performed using paired or unpaired Student’s t-test. All quantitative data are presented as mean ± SEM.

3. Results

To illustrate the foundation of investigating the functions of WT and HCM mutant cTnT-C14 peptides, the C-terminal end position of cTnT-C14 peptide in reference to the recently identified Tm-binding site 3, the TnI and TnC binding sties, and Tm-binding sites 1 and 2 of TnT is shown in Fig. 1A. Fig. 1B shows the amino acid sequence alignment of cTnT-C14 segment of representative mammalian species and the striking feature that this segment is 100% conserved in all mammalian hearts, reflecting a stringent evolutionary selection indicating functional importance. Physical properties of the TnT-C14 peptides were compared in Fig. 1B to demonstrate that the loss of a positive charge in the HCM mutants may be a common pathogenic factor. Supporting this notion, previous studies have shown that substituting the positively charged residues in the C-terminal end segment of TnT with neutral amino acid residue Ala resulted in impaired relaxation and overactivation of skeletal and cardiac muscles[10,11].

Figure 1. The 100% conserved C-terminal end segment of mammalian cardiac TnT in comparison with human cardiomyopathy mutations and fast and slow skeletal muscle isoforms.

(A) A linear structure map of cardiac TnT with the segments encoded by each exon marked is aligned with the exon 17-encoded cTnT-C14 peptide (residues 275–288 in adult human cTnT) to show relationships to the binding sites for TnI, TnC and Tm located in the T1 and T2 chymotryptic and CB2 CNBr fragments. (B) The amino acid sequence alignment of the C-terminal end segment of TnT along with the overall isoelectric point (pI) and net charge at pH 7.0 showed that structure of the C-terminal end segment encoded by the last exon of TnT genes is highly conserved and has an identical amino acid sequence in all mammalian cardiac TnTs from platypus to human. The three cTnT-C14 peptides containing HCM mutation R278C, K280N or R286C are compared to implicate that the loss of a positive charge may be a pathogenic factor shared by these mutations. The C-terminal end segments of human fast and slow skeletal muscle TnT are aligned as examples to show the variation among muscle type TnT isoforms. The NCBI GenBank accession numbers of the TnT sequences analyzed are: Platypus cardiac, XP_028924875.1; koala cardiac, XP_020858612; rabbit cardiac, AAB51160.1; dog cardiac, AAG23715.1; bovine cardiac, NP_777196.1; dolphin cardiac, XP_059982389.1; whale cardiac, XP_057400554.1; bat cardiac, XP_066095192.1; mouse cardiac, BAB19881.1; human cardiac, AAK92231.1; human fast, NP_001354775.1; and human slow, NP_001119604.1.

The SDS-PAGE gels in Fig. 2 show that the purity of Tm and F-actin used in the studies.

Figure 2. Purified cardiac α-tropomyosin and striated muscle actin used in the present study.

The SDS-gel images show the rabbit tropomyosin and actin preparations with sufficient purities and anticipated molecular weights (32.5-kDa and 43-kDa, respectively).

3.1. Tm-Binding Kinetics of cTnT-C14 Peptide and the Loss of Function Impact of HCM Mutations

The representative LSPR traces in Fig. 3 show that WT human cTnT-C14 peptide at 2 μM concentration binds immobilized Tm with a fast association rate to reach saturation. Once the plateau of the binding signal was reached and the flow was switched to LSPR buffer, the dissociation of WT cTnT-C14 peptide occurred also with a fast rate. While LSPR analysis of the binding of cTnT-C14 peptide to chip-immobilized tropomyosin mimics that to myofilament-immobilized Tm in muscle, it is a non-equilibrium kinetics-based solid phase binding assay in which the ligand concentration for saturable binding is not always equal to K_D. Nonetheless, reproducible LSPR traces showed that cTnT-C14 peptides at 2 μM concentration can reach saturated binding, allowing the comparison of plateaued Δpm between WT and HCM mutants. The saturable Tm-binding of WT cTnT-C14 peptide with fast kinetics (Table 1) is consistent with the segment’s in situ function in cTnT under dynamic allosteric changes during cardiac muscle contraction and relaxation. The results demonstrate that the Tm-binding ability of the Tm-binding site 3 in the C-terminal end segment of TnT is retained in the form of isolated free peptide, reflecting its intrinsic conformational and functional state.

Figure 3. LSPR analysis of the binding of cTnT-C14 peptides to tropomyosin.

WT human cTnT-C14 peptide or cTnT-C14 peptides containing HCM mutation R278C, K280N or R286C at 2 μM concentration was flown over an LSPR sensor chip with saturated coating of Tm. In comparison to the negative control peptide cTnT-E16, the representative LSPR traces from three or more independent experiments demonstrate that cTnT-C14WT produced a saturable high signal level (Δpm) binding to immobilized Tm with fast association and dissociation rates. In contrast, the three HCM mutant peptides all showed significantly lower Tm-binding signal levels. The results demonstrate that the C-terminal end segment of cTnT functions as a Tm-binding structure in the form of isolated free peptide, a functionality that is significantly impaired by the HCM mutations. The LSPR curves were analyzed by comparing Δpm at the midpoint of association phase, the midpoint of plateau phase, and the midpoint of dissociation phase using two-way ANOVA followed by post hoc multiple comparisons (Tukey test) for difference between the binding of HCM mutant vs WT peptides to Tm. ***P < 0.001.

Table 1.

LSPR Maximum binding signal (Δpm), association (k_on) and dissociation (k_off) rates, and affinity (K_D) of WT and HCM mutant human cTnT-C14 peptides to immobilized tropomyosin

	Maximum Signal (Δpm)	kon (1/(M*s))	koff (1/s)	KD (M)
WT	4.20e2 ± 8.58e0	5.91e3 ± 4.81e2	3.74e-2 ± 1.31e-3	6.53e-6 ± 4.27e-7
R278C	9.54e1 ± 6.82e0 **	2.50e3 ± 1.56e2 **	2.19e-2 ± 2.09e-3 **	8.89e-6 ± 8.39e-7 *
K280N	1.08e2 ± 4.87e0 **	6.60e3 ± 5.14e2	3.51e-2 ± 5.06e-3	5.29e-6 ± 6.41e-7
R286C	1.14e2 ± 8.89e0 **	3.11e3 ± 3.81e2 **	1.66e-2 ± 1.24e-3 **	6.26e-6 ± 1.12e-6

The maximum Tm-binding signal of cTnT-C14 peptide was measured as the plateaued LSPR shift in absorption wavelength (Δpm, representative curves are shown in Fig. 3). The association (k_on) and dissociation (k_off) rate constants were obtained by fitting LSPR curves using a one-to-one binding model, and the binding affinity (K_D) was calculated as k_off/k_on. Data are presented as mean ± SEM from 5 or more independent experiments. In comparison to WT control, all three HCM mutant peptides showed significantly decreased Δpm. R278C and R286C had significantly slower k_on and k_off rates than WT control, which is also reflected by the lower affinity (K_D) of R278C.

^*P < 0.05,

^**P < 0.01 vs WT in Student’s t test.

The results in Fig. 3 and Table 1 further found that the human cTnT-C14 peptides containing any of the three HCM-linked single amino acid substitutions (R278C, K280N or R286C) all have significantly decreased maximum binding to Tm. In comparison to WT cTnT-C14 under the same concentration, time parameters and experimental conditions, the maximum Tm-binding signal (Δpm) of the mutant peptides was significantly lower. The R278C and R286C mutant peptides also showed decreased association and dissociation rates in the Tm-binding assay, of which R278C had a statistically significant decrease in affinity (Table 1).

3.2. Binding of cTnT-C14 Peptide to F-actin-Tm Composite Thin Filament and the Loss of Function Impact of HCM-linked Mutations

The LSPR results in Fig. 4 and Table 2 show that WT human cTnT-C14 peptide also strongly binds F-actin-Tm composite filaments, similar to the saturable binding to Tm alone with fast association and dissociation rates. Also similar to the binding to Tm alone, the cTnT-C14 peptide containing HCM-linked amino acid substitutions exhibited significantly weakened maximum binding signal (Δpm) to F-actin-Tm composite filament. The results support the ability of free cTnT-C14 peptide to produce functional effects on the thin filament regulation of muscle contraction and relaxation. The similar patterns of cTnT-C14 peptide interactions with F-actin-Tm filament and with Tm alone and the similarly loss of function impact of the three HCM-linked mutants suggest that the binding to Tm is likely a primary mechanism for cTnT-C14 peptide to interact with F-actin-Tm composite thin filament. On the other hand, the impact of the HCM mutations was less on the binding to Tm-F-actin filament than that to Tm alone with only R278 showing statistically significant slower dissociation rate than WT control (Table 2), implicating a change in the binding mode contributed by the presence of F-actin.

Figure 4. LSPR analysis of the binding of cTnT-C14 peptides to tropomyosin-F-actin composite thin filament.

The LSPR sensor chip was coated with a low density of F-actin (at the flow concentration of 30 μg/mL). After blocking and a saturating binding of Tm to immobilized F-actin (at the flow concentration of 10 μM) followed by a wash to remove free Tm, WT or HCM mutant human cTnT-C14 peptide was injected at 2 μM concentration to measure the association to and dissociation from the reconstituted F-actin-Tm thin filaments. The representative LSPR traces from at least three independent experiments show an initial peak representing the binding of Tm to F-actin followed by the binding peak of cTnT-C14 peptides. The results demonstrate that in comparison to the negative control peptide cTnT-E16, human cTnT-C14WT peptide had rapid association and saturable high signal level (Δpm) binding to F-actin-Tm thin filament followed by a rapid dissociation, similar to its interaction to Tm alone (Fig. 3). Also similar to the weakened binding to Tm alone, the three HCM mutant peptides showed significantly diminished binding signal to the F-actin-Tm thin filament. The results confirm that the C-terminal end peptide of cTnT interacts with muscle thin filament, which is weakened by the HCM mutations. The LSPR curves were analyzed by comparing Δpm at the midpoint of association phase, the midpoint of plateau phase, and the midpoint of dissociation phase using two-way ANOVA followed by post hoc multiple comparisons (Tukey test) for difference between the binding of HCM mutant vs WT peptides to F-actin-Tm. ***P < 0.001.

Table 2.

LSPR Maximum binding signal (Δpm), association (k_on) and dissociation (k_off) rates, and affinity (K_D) of WT and HCM mutant human cTnT-C14 peptides to immobilized F-actin–Tm composite filament

	Maximum Signal (Δpm)	kon (1/(M*s))	koff (1/s)	KD (M)
WT	4.34e2 ± 8.94e0	5.27e3 ± 5.32e2	3.62e-2 ± 3.43e-3	7.13e-6 ± 1.04e-6
R278C	1.34e2 ± 1.26e1 **	3.93e3 ± 2.49e2	2.00e-2 ± 1.35e-3 *	5.10e-6 ± 1.48e-7
K280N	1.09e2 ± 5.10e0 **	6.45e3 ± 5.77e2	4.10e-2 ± 8.86e-3	6.50e-6 ± 1.30e-6
R286C	1.26e2 ± 1.09e1 **	3.85e3 ± 6.99e2	2.56e-2 ± 5.22e-3	7.56e-6 ± 1.58e-6

The maximum F-actin-Tm binding signal was measured as the plateaued LSPR shift in absorption wavelength (Δpm, representative curves are shown in Fig. 4). The association (k_on) and dissociation (k_off) rate constants were obtained by fitting LSPR curves using a one-to-one binding model, and the binding affinity (K_D) was calculated as k_off/k_on. Data are presented as mean ± SEM from 5 or more independent experiments. WT cTnT-C14 binds to F-actin-Tm composite thin filament similarly to that to Tm alone. All three HCM mutant peptides showed significantly decreased Δpm.

^*P < 0.05,

^**P < 0.01 vs WT in Student’s t test.

3.3. cTnT-C14 Peptide Directly Binds F-actin

The C-terminal domain of TnT was previously reported with an activity in binding F-actin[25]. Knowing the cTnT-C14 peptides bind F-actin-Tm composite filament similarly to but with detectable difference from their binding patterns for Tm alone (Tables 1 and 2), we further investigated their direct binding to F-actin in LSPR under the same conditions. The results in Fig. 5 and Table 3 show that WT cTnT-C14 peptide strongly binds immobilized F-actin as shown by the maximum binding signal (Δpm). The F-actin-binding of cTnT-C14 peptide was significantly decreased by the HCM mutations of which R278C and R286C showed significantly slower association and dissociation rates and overall affinity whereas K280N showed decreased dissociation rate (Table 3). The results demonstrate that the C-terminal end segment of TnT is involved in the previously detected direct binding of TnT to F-actin[25], an activity potentially contributing to the thin filament regulation of striated muscle contraction and relaxation.

Figure 5. LSPR analysis of the direct binding of cTnT-C14 peptides to F-actin.

LSPR sensor chip was coated with F-actin at flow concentration of 30 μg/mL to test the interactions of WT and HCM mutant human cTnT-C14 peptides with F-actin. The representative LSPR traces from at least three independent experiments show that cTnT-C14WT produced clear saturable binding signal (Δpm) in contrast to the significantly weakened bindings of the HCM mutants. The LSPR curves were analyzed by comparing Δpm at the midpoint of association phase, the midpoint of plateau phase, and the midpoint of dissociation phase using two-way ANOVA followed by post hoc multiple comparisons (Tukey test) for difference between the binding of HCM mutant vs WT peptides to F-actin. ***P < 0.001.

Table 3.

LSPR Maximum binding signal (Δpm), association (k_on) and dissociation (k_off) rates, and affinity (K_D) of WT and HCM mutant human cTnT-C14 peptides to immobilized F-actin

	Maximum Signal (Δpm)	kon (1/(M*s))	koff (1/s)	KD (M)
WT	3.97e2 ± 8.63e0	4.15e3 ± 1.47e2	2.78e-2 ± 1.30e-3	6.71e-6 ± 2.72e-7
R278C	1.44e2 ± 7.94e0 **	1.45e3 ± 1.60e2 **	1.69e-2 ± 8.28e-4 **	1.22e-5 ± 1.46e-6 *
K280N	1.12e2 ± 7.32e0 **	3.81e3 ± 4.53e2	4.15e-2 ± 5.84e-3	1.24e-5 ± 2.80e-6
R286C	1.12e2 ± 7.69e0 **	1.57e3 ± 1.86e2 **	1.72e-2 ± 1.08e-3 **	1.14e-5 ± 9.87e-7 **

The maximum F-actin-binding signal of cTnT-C14 peptide was measured as the plateaued LSPR shift in absorption wavelength (Δpm, representative curves are shown in Fig. 5). The association (k_on) and dissociation (k_off) rate constants were obtained by fitting LSPR curves using a one-to-one binding model, and the binding affinity (K_D) was calculated as k_off/k_on. Data are presented as mean ± SEM from 5 or more independent experiments. In comparison to WT control, all three HCM mutant peptides showed significantly decreased Δpm. R278C and R286C had significantly slower k_on and k_off rates than WT control, which is also reflected by the lower affinity (K_D) of R278C.

^*P < 0.05,

^**P < 0.01 vs WT in Student’s t test.

3.4. HCM mutations in cTnT-C14 Peptides Abolish Its Ca²⁺-Desensitization Activity

In contrast to the Ca²⁺-desensitization effect of WT cTnT-C14 peptide[7], addition of cTnT-C14 peptides containing HCM-linked amino acid substitution mutation R278C, K280N or R286C to permeabilized WT mouse left ventricular papillary muscle sections did not produce a detectable functional effect (Fig. 6). The results demonstrate that their diminished Tm- and thin filament-binding ability is an underlying mechanism for these cTnT point mutations to produce pathological effects and cause myocardial hyperactivation in HCM.

Figure 6. HCM mutations in cTnT-C14 peptide cause a loss of the function in modulating myofilament Ca²⁺-sensitivity.

Consistent with their diminished Tm- and thin filament-binding ability, the force-pCa curves, inset tables and bar graphs show that the treatment of permeabilized sections of WT mouse left ventricular papillary muscle sections with 20 μM cTnT-C14 peptides containing HCM mutation R278C (A), K280N (B) or R286C (C) did not produce any detectable effect on the Ca²⁺-activation of contraction as compared with the force-pCa curve of WT mouse cardiac muscle without peptide treatment. Data are presented as mean ± SEM. In contrast to the Ca²⁺-desensitization effect of WT cTnT-C14 peptide on WT mouse cardiac muscle sections shown previously[7], the results demonstrate a loss of function caused by the HCM mutations, corresponding to hyperactivation phenotypes. N = 4 strips for R278C, N = 3 strips for K280N and N = 5 strips for R286C studies.

3.5. Diminished Effects of Free cTnT-C14 Peptide on the Contractility of Permeabilized cTnT-ND vs WT Mouse Cardiac Muscles

Our previous study have shown that free cTnT-C14 peptide interacts with WT mouse myofilaments to produce a Ca²⁺-desensitization effect on the contraction of permeabilized cardiac muscle sections[7]. Our earlier studies have demonstrated that the molecular conformation of the C-terminal end segment of cTnT is modulated by the N-terminal variable region and a restrictive proteolytic deletion of amino acids 1–71 restores a TnI like inhibitory structure[16,19] which is located to the cTnT-C14 segment[7]. To investigate the effect of cTnT-ND on adjusting ventricular contractile kinetics in compensatory adaptation to inotropy-afterload mismatch in vivo[21] via conformational modulating the inhibitory function of the cTnT-C14 segment, contractility studies showed that permeabilized cTnT-ND mouse cardiac papillary muscle strips had an increased overall Ca²⁺-sensitivity (higher pCa₅₀) (Fig. 7A) but decreased maximum force development (Fig. 7B) in comparison to WT control, consistent with its effect on reducing the post-activation inotropy to elongate ventricular ejection time and increase stroke volume shown in ex vivo working heart studies[21]. The positive length-tension (Fig. 7C) and length-Ca²⁺-sensitivity (Fig. 7A) responses are preserved in cTnT-ND cardiac muscle similar to that of WT control. The results suggest that the decrease in maximum force development by cTnT-ND is likely by a post-activation myofilament Ca²⁺-desensitization via the N-terminal deletion-enhanced TnI-like inhibitory activity of the C-terminal end Tm-binding site 3[7,19].

Figure 7. cTnT-ND reduces the maximum force development of cardiac muscle while increases pCa₅₀.

While cTnT-ND restores the TnI-like inhibitory function in the C-terminal end C14 segment, contractility of permeabilized left ventricular papillary muscle sections from cTnT-ND transgenic mice[21] demonstrated a higher pCa₅₀ reflecting higher overall Ca²⁺-sensitivity than WT control (A). On the other hand, cTnT-ND transgenic mouse cardiac muscle showed a clear trend of lower maximum force production (Fmax dev) at pCa 4.5 than WT mouse cardiac muscle control (B) while preserving the positive length-tension (C) and length-Ca²⁺-sensitivity responses (A) similar to that of WT control. Data are presented as mean ± SEM. *P<0.05 and **P<0.01 vs sarcomere length (SL) 2.0 μm in Student t test. ^#P<0.05 vs WT in Student t test. N = 4 fibers each in WT and cTnT-ND for (A) and (C). N = 8 in WT and 4 in cTnT-ND for (B).

To further understand the mechanism underlying the modulation of cardiac muscle contractility by free cTnT-C14 peptide, the force-pCa response of permeabilized cTnT-ND mouse left ventricular papillary muscle sections showed that in contrast to its Ca²⁺-desensitization effect on WT mouse cardiac muscle shown previously[7] the treatment of permeabilized cTnT-ND mouse muscle strips with 20 μM WT cTnT-C14 peptide did not have detectable effect on Ca²⁺-activation of contraction at both short (2.0 μm) and long (2.3 μm) sarcomere lengths (Fig. 8A). This finding suggests that the mechanism of post-activation myofilament Ca²⁺-desensitization enhanced in situ by cTnT-ND also underlies the effect of added free cTnT-C14 peptide on WT cardiac muscle. Since this capacity has been utilized, the addition of free cTnT-C14 peptide to cTnT-ND cardiac muscle will not produce further additive effect through the same mechanism. Different from the effect of cTnT-ND (Fig. 7), treatment with free cTnT-C14 peptide did not decrease the maximum force development of WT (Fig. 8B) or cTnT-ND ((Fig. 8C) mouse cardiac muscle strips. An interesting finding is that while cTnT-C14 reduces Ca²⁺-sensitivity of WT cardiac muscle[7], it does not reverse the effect of cTnT-ND on increasing Ca²⁺-sensitivity (Fig. 8A).

Figure 8. cTnT-C14 peptide does not produce additive effect on Ca²⁺-sensitivity of permeabilized cTnT-ND mouse cardiac muscle sections.

(A) In contrast to its significant Ca²⁺-desensitization effect on WT mouse left ventricular papillary muscle sections reported previously[7], the force-pCa curves and inset table show that the addition of WT cTnT-C14 peptide to permeabilized sections of cTnT-ND mouse left ventricular papillary muscle did not produce any detectable effect on pCa₅₀ and cooperativity (n) at sarcomere lengths (SL) of 2.0 μm or 2.3 μm. Data are presented as mean ± SEM calculated from measurements of multiple strips. The treatment with cTnT-C14 peptide did not decrease the maximum force development (Fmax dev) of cTnT-ND (B) and WT (C) cardiac muscle sections. N = 5 fibers in each group. *P < 0.05 vs SL 2.0 μm in Student’s t test.

4. Discussions

Troponin-tropomyosin interaction is at the center of the regulation of striated muscle contraction and relaxations with physiology and pathophysiology importance. Recent progress in molecular dynamics simulation and cryo-electron microscopy studies have produced valuable models of troponin-tropomyosin and troponin-thin filament interactions[26–30]. The current structural models of troponin complex and striated muscle thin filament derived from static crystallography and 3D reconstruction of cryo-electron microscopy images, however, have not identified a definitive location for the C-terminal end segment of TnT, likely reflecting its allosteric structure and function. Experimental determination of troponin and thin filament protein structure-function relationship and interactions remain essential for structural biology and computational simulation to build informative higher order models of muscle thin filament structure and regulation. Following the recent identification and localization of the third Tm-binding site of TnT in the C-terminal end segment [7], our present study further characterized the intrinsic functionality of free cTnT-C14 peptide in isolation from the TnT backbone and investigated its mechanistic basis in comparison with the C-terminal end segment of endogenous cTnT in situ under Ca²⁺ regulation and conformational modulation. The data provide several new insights into the function of the Tm-binding site 3 in the context of structure-function relationship of TnT.

4.1. The Isolated cTnT-C14 Free Peptide Retains Intrinsic Function of the Tm-binding Site 3 of TnT

To further establish the intrinsic function of the C-terminal end 14 amino acids segment of cardiac TnT, the present study focused on its activity in the form of free peptide in isolation from the TnT backbone. Rationale of this research design is based on that a) this segment is encoded by a single exon as a likely genetic and functional unit (Fig. 1A), b) it is a likely allosteric structure extending from the core of troponin complex as indicated by the crystallography data[31,32], and c) it is a uniquely conserved structure with a 100% identical amino acid sequence in all mammalian cardiac TnT implicating a stringently selected functionality (Fig. 1B). Supporting the hypothesis that the C-terminal end 14 amino acids segment remains an intrinsic functionality when isolated from the TnT backbone and demonstrating the difference between the functions of added free cTnT-C14 peptide and the endogenous C-terminal end segment of TnT in situ, the new data show that the C-terminal end 14 amino acid free peptide of human cTnT retains the abilities to bind Tm and actin thin filament (Figs. 3 and 4), similar to that of a mini-TnT construct containing the exons 15–17-encoded segment[7]. The micromolar affinity with fast association and dissociation kinetics of the interaction of cTnT-C14 with Tm and F-actin-Tm thin filament supports the Tm-binding site 3 to function in modulating contractility of cardiac muscle during the dynamic cycle of contraction and relaxation.

It is an important finding that the cTnT-C14 peptide also directly binds F-actin (Fig. 5). This activity is consistent with the early report that the T2 fragment of TnT (Fig. 1A) bound F-actin[25]. Our result now localizes the actin-binding activity of TnT to the C-terminal end segment, which is a function retained by cTnT-C14 peptide in isolation from the troponin backbone and structurally integrative with the Tm-binding site 3. This finding lays a new platform for the structure-function model of troponin regulation of muscle contraction and relaxation. The binding of cTnT-C14 to Tm and F-actin may both contribute to the binding to composite F-actin-Tm thin filament. Therefore, we compared the kinetic parameters of the binary or tertiary bindings. Summarized in Table 4, WT cTnT-C14 peptide has higher maximum binding (Δpm) to F-action-Tm than to F-actin alone. The R278C mutant decreases Tm-binding more than F-actin-binding but has less decrease in F-actin-Tm-binding than Tm-binding. The association and dissociation rates of WT cTnT-C14 peptide are faster in Tm-binding than that in F-actin-binding. This pattern suggests that Tm-binding may be a primary function of Tm-binding site 3 in native thin filaments. The observation that the three HCM mutations had detectable differences in the impacts on the binding kinetics for Tm, F-actin-Tm and F-actin (Table 4) may implicate specific functional change in the cTnT-C14 peptide and is worth further investigation.

Table 4.

Qualitative comparison of the binding of WT and HCM mutant cTnT-C14 peptides to Tm, F-actin and F-actin-Tm composite thin filament

	Peptide
Comparison		WT	R278C	K280N	R286C
Tm vs F-Actin	Δpm	-	↓ **	-	-
	kon	↑ **	↑ **	↑ **	↑ **
	koff	↑ **	-	-	-
	KD	-	-	↑ *	↑ **
F-Actin-Tm vs Tm alone	Δpm	-	↑ *	-	-
	kon	-	↑ **	-	-
	koff	-	-	-	-
	KD	-	↑ **	-	-
F-Actin-Tm vs F-Actin alone	Δpm	↑ *	-	-	-
	kon	-	↑ **	↑ *	↑ *
	koff	-	-	-	-
	KD	-	↓ **	-	-

The kinetic data with statistical significance (*p < 0.05; **P < 0.01 in Student’s t test) shown in Tables 1–3 are summarized in paired comparisons of the binding of WT and HCM mutant cTnT-C14 peptide to Tm, F-actin and F-actin-Tm composite thin filament. The arrows indicate the direction of statistically significant differences in the maximum binding signal (Δpm), association rate (k_on) and dissociation (k_off) rates and binding affinity (K_D) of the indicated comparisons between the binding to the first partner and the binding to the second partner listed in the first column (↑ denotes increase, ↓ denotes decrease and - indicates no statistically significant difference).

4.2. Loss of Tm- and Thin Filament-binding Function Underlies the Pathological Impact of HCM Mutations in cTnT-C14

The C-terminal end segment of cardiac TnT encoded by exon 17 of TNNT2 gene is highly conserved in vertebrate species and has identical amino acid sequence in all mammalian cardiac TnT in the current database (Fig. 1B). This unique structural conservation reflects a stringent structure-function relationship and a high selection value during the evolution of mammalian hearts. Using LSPR that is a sensitive kinetic approach for measuring protein-protein and protein-ligand interactions[33] to compare the three HCM mutant human cTnT-C14 peptides for interactions with Tm, F-actin-Tm thin filament and F-actin alone with WT control clearly showed their loss of function. The significantly weakened maximum binding (Δpm) of cTnT-C14 peptide for Tm, F-actin and F-actin-Tm thin filament caused by each of the HCM-associated point mutations (Figs. 3, 4 and 5 and Tables 1, 2 and 3) confirms the stringent structural requirement. Consistently, cardiac muscle contractility studies showed that the HCM mutant cTnT-C14 peptides had lost the Ca²⁺-desensitization effect, showing a pathogenic and pathophysiologic mechanism by which impairment of Tm-binding site 3 causes myocardial hypercontractility. The detectable differences in the maximum binding and kinetic parameters of the three HCM mutant cTnT-C14 peptides for Tm and F-actin (Table 4) suggest functional impacts involving different residues in the C-terminal segment of cTnT.

Supporting our finding, previous studies have shown that the R278C mutant of cTnT increases Ca²⁺-sensitivity with reduced thin filament binding affinity[12,34]. The loss of Tm-binding site 3 function in the inhibitory regulation of cardiac muscle is likely to hinder the coordinated transition from contraction to relaxation to cause impaired relaxation and hypercontractile phenotype as shown by the study of Ala substitution for the positively charged Arg and Lys residues in the C-terminal end segment of TnT[8,9], implicating a common pathogenic mechanism for mutations in the C-terminal end Tm-binding site 3 to cause HCM. The findings that deletion of the N-terminal variable region of cTnT enhances (Fig. 7) and the HCM mutations weakens (Fig. 6) the inhibitory function of the C-terminal Tm-binding site 3 connect the in vitro function of the cTnT C14 peptide to the organ level function in transgenic mouse hearts and the in vivo HCM phenotypes.

4.3. Conformationally Modulated Functionality of the Tm-binding Site 3 of cTnT

The initial lead in our finding of the Tm-binding site 3 in the C-terminal end 14 amino acid segment of cTnT and its regulatory function[7] was an observation that the restrictive deletion of the evolutionarily added N-terminal variable region of cTnT in adaptation to inotropy-afterload mismatch in vivo[35] restores an evolutionarily repressed TnI C-terminus-like[16] Tm-binding inhibitory structure[19,20]. This physiological adaptation by modulating the molecular conformation and function of TnT[36,37] selectively tunes down ventricular end-systolic velocity to elongate ejection time and sustain stroke volume against high afterload[21], indicating a physiological function of the C-terminal end Tm-binding site 3 of TnT in regulating contractile kinetics.

Reflected by the dramatic change of the affinity of the anti-TnI C-terminal mAb TnI-1 from no significant binding to intact cTnT to the near anti-TnI level strong binding to cTnT-ND[7], the conformation of the Tm-binding site 3 of TnT can be modulated from a largely repressed state in intact cTnT to the restored state in cTnT-ND. Similar to the isolated C-terminal end 27 amino acids Tm-binding inhibitory segment of cardiac TnI which contains the epitope recognized by mAb TnI-1[19,20], free cTnT-C14 peptide retains the TnI-like conformational state in solution with Tm- and F-actin-binding activities (Figs. 3 and 5) and the function in modulating muscle contractility[7]. The functionality of isolated cTnT-C14 free peptide indicates an intrinsic activity of the C-terminal end Tm-binding site 3 in an unregulated active state when the conformational and anchoring restraints are removed. The physiological level decrease of maximum force development in cTnT-ND transgenic mouse cardiac muscle at pCa 4.5 (Fig. 7B) with increased pCa₅₀ reflecting increased overall Ca²⁺-sensitivity (Fig. 7A) implicates a post-activation Ca²⁺-desensitization of cardiac myofilament.

Different from the effect of cTnT-ND (Fig. 7), treatment with cTnT-C14 peptide does not decrease the maximum force development of WT (Fig. 8B) or cTnT-ND (Fig. 8C) mouse cardiac muscle at pCa 4.5. The lack of additive effects of exogenous cTnT-C14 peptide and endogenous cTnT-ND (Fig. 8A) suggests that free cTnT-C14 peptide acts via the same mechanism as that of the Tm-binding site 3 enhanced in cTnT-ND in situ to physiologically adjust cardiac muscle contractile kinetics. While retaining the post-activation Ca²⁺-desensitization effect on WT cardiac muscle[7], cTnT-C14 does not reverse the effect of cTnT-ND on increasing the overall Ca²⁺-sensitivity (pCa₅₀). To explain this difference, it is worth noting that Tm-binding of the endogenous C-terminal segment of cTnT is under Ca²⁺-regulation. A previous study of our group showed that reconstituted troponin complex using mini-TnT that retains only the TnI and TnC binding sites and Tm-binding site 3[7] and C-terminal truncated cardiac TnI showed a saturable binding to Tm at pCa 9 but not pCa 4[38] (figure 5 in the previous work published before the finding of Tm-binding site 3 in TnT). This observation is supported by the effect of Ca²⁺-TnC on reducing the binding of the T2 fragment of TnT to F-actin[25]. These data together let us propose that the functionality of the C-terminal end segment of TnT in muscle is dynamically changing between allosteric states of troponin complex during Ca²⁺-regulated contraction and relaxation. In contrast, the free cTnTC14 peptide acts as a Ca²⁺-independent regulator to tune muscle contractile kinetics in coordination with the Ca²⁺-regulated Tm-binding site 3 of endogenous cTnT in the troponin complex with a physiologically relevant pharmacological effect.

4.4. Potential Use of cTnT-C14 Peptide as A Therapeutic Reagent

Our previous study demonstrated that the cTnT-C14 peptide can act at micromolar concentration to adjust the Ca²⁺-sensitivity of cardiac muscle[7]. The saturable binding to Tm and F-actin-Tm thin filament at 2 μM (Figs. 3 and 4) and its potent Ca²⁺-desensitization activity present a potential value in correcting pathological hypercontractility. The fast kinetics of cTnT-C14 peptide interactions with Tm and F-actin-Tm thin filament are compatible with the rhythmic contraction and relaxation of cardiac muscle. The result that adding of cTnT-C14 peptide to cTnT-ND cardiac muscle strips does not produce additive effect on Ca²⁺-desensitization of cTnT-ND cardiac muscle (Fig. 8A) is plausible by not only showing that the application of free cTnT-C14 peptide safely targets an endogenous physiological mechanism but also demonstrating its effect on cardiac muscle contractility is physiologically restrictive avoiding the negative impact of over-desensitization. Another plausible feature of free cTnT-C14 peptide is its addition to permeabilized cardiac muscle strips does not decrease maximum force production (Fig. 8B). Based on its constitutive Tm/thin filament binding activity in coordination with the Ca²⁺-regulated Tm-binding site 3 of endogenous cTnT, the characterization of cTnT-C14 peptide presents a novel reagent to modulate cardiac muscle contractile kinetics for the development of therapeutic applications.

4.5. Limitations of the Present Study and Future Directions

Pioneer work in the 1970–80’s by the Smillie group and others mapped the linear structure-biochemical function relationship of TnT using enzymatic and chemically cleaved fragments. The studies detected binding affinities for TnC, TnI, Tm and actin in the chymotryptic fragment T2 containing the C-terminal ~100 amino acids[5] and another Tm-binding site in the middle region cyanogen bromide fragment CB2[4] (Fig. 1A). The precise locations of the TnC and TnI binding sites of TnT in the T2 region have been confirmed by high resolution crystallography studies of partial cardiac and skeletal muscle troponin complexes[32,33]. The crystallography structures of troponin complex did not resolve the middle region, the N-terminal variable region and the C-terminal end segment of TnT, likely reflecting their nature as flexible structural domains. Our follow up protein binding studies using molecular engineered TnT segments further mapped the Tm-binding site 1 that expands a large portion of the conserved middle region and localized the Tm-binding site 2 to the beginning of the T2 fragment of TnT[6]. Revising the longstanding striated muscle thin filament model based on two Tm-binding sites in TnT, we recently reported the identification and localization of a novel and allosterically regulated Tm-binding site 3 in the C-terminal end 14 amino acids segment of TnT[7] (Fig. 1A).

Our present study focused on the kinetics of Tm- and F-actin-binding of isolated cTnT-C14 free peptide and its effect on the activation of permeabilized cardiac muscle in coordination with the Tm-binding site 3 of endogenous cTnT. Our LSPR study is limited by using simplified system without Ca²⁺-regulated troponin complex in the absence of myosin and other activities of the C-terminal end segment of TnT may also have functional effects. Previous studies have reported crosslinking of residue K282 in the cTnT-C14 segment[39] and a FRET change of adjacent residue V274[40] for potential interactions with TnC. The current knowledge collectively indicates novel functions of the C-terminal end segment of TnT in regulating striated muscle contractility, meriting further studies to explore the molecular mechanisms. Further characterization of the pharmacological functionality of cTnT-C14 peptide in a dynamic physiological system such as beating cardiomyocytes will help to develop it into a therapeutic modulator of cardiac muscle contractility and kinetics. In addition to adding new insights into the mechanisms of troponin-Tm regulation of striated muscle contraction and underscoring the pathological basis of HCM mutations in Tm-binding site 3 of cTnT, the new findings of the present study lays a foundation for future research and development.

Highlights.

• Isolated C-terminal 14 amino acids peptide of cardiac troponin T (cTnT-C14) binds tropomyosin and F-actin.

• cTnT-C14 peptide produces a Ca²⁺-desensitization effect on permeabilized cardiac muscle, non-additive to the effect of enhanced Tm-binding site 3 in N-terminal truncated cTnT.

• cTnT-C14 peptides containing hypertrophic cardiomyopathy mutations show diminished binding to tropomyosin and F-actin and a loss of Ca²⁺-desensitization function.

• cTnT-C14 peptide presents a novel reagent for correcting myocardial hypercontractility.