Mutations in troponin T associated with Hypertrophic Cardiomyopathy increase Ca2+-sensitivity and suppress the modulation of Ca2+-sensitivity by troponin I phosphorylation☆

We investigated the effect of 7 Hypertrophic Cardiomyopathy (HCM)-causing mutations in troponin T (TnT) on troponin function in thin filaments reconstituted with actin and human cardiac tropomyosin. We used the quantitative in vitro motility assay to study Ca2+-regulation of unloaded movement and its modulation by troponin I phosphorylation. Troponin from a patient with the K280N TnT mutation showed no difference in Ca2+-sensitivity when compared with donor heart troponin and the Ca2+-sensitivity was also independent of the troponin I phosphorylation level (uncoupled). The recombinant K280N TnT mutation increased Ca2+-sensitivity 1.7-fold and was also uncoupled. The R92Q TnT mutation in troponin from transgenic mouse increased Ca2+-sensitivity and was also completely uncoupled. Five TnT mutations (Δ14, Δ28 + 7, ΔE160, S179F and K273E) studied in recombinant troponin increased Ca2+-sensitivity and were all fully uncoupled. Thus, for HCM-causing mutations in TnT, Ca2+-sensitisation together with uncoupling in vitro is the usual response and both factors may contribute to the HCM phenotype. We also found that Epigallocatechin-3-gallate (EGCG) can restore coupling to all uncoupled HCM-causing TnT mutations. In fact the combination of Ca2+-desensitisation and re-coupling due to EGCG completely reverses both the abnormalities found in troponin with a TnT HCM mutation suggesting it may have therapeutic potential.


Introduction
Hypertrophic cardiomyopathy (HCM) is the most common inherited cardiomyopathy and is usually associated with mutations in sarcomeric proteins. A recent study has shown that 11% of the identified mutations are in the proteins of the thin filament; actin, tropomyosin, troponin I (TnI), troponin C (TnC) and troponin T (TnT) [1]. It is generally found that HCM-causing mutations result in a 2e3 fold higher myofilament Ca 2þ -sensitivity compared to normal heart muscle and this has been proposed to be necessary to trigger the symptoms of HCM: a hyper-contractile phenotype, heart muscle hypertrophy, myocyte disarray and fibrosis [2], although the clinical manifestation of HCM is very variable, probably due to background genetic and environmental factors.
Mutations in the same sarcomeric protein genes are also associated with familial dilated cardiomyopathy (DCM), a disease characterised by a hypo-contractile phenotype with dilation of the heart chamber and thinning of cardiac muscle [3]. Whilst DCM clearly has a separate molecular mechanism from HCM, the simple hypothesis that Ca 2þ -sensitivity is reduced in familial DCM (the opposite of HCM) has proven to be incorrect since changes of Ca 2þ -sensitivity do not correlate with the phenotype [2,4,5]. Our investigations have established that a consistent feature of DCM-causing mutations in thin filament proteins is that the Ca 2þ -sensitivity is not modulated by TnI phosphorylation, a process we have called uncoupling, and we have proposed that this is a disease-mechanism [4e6].
TnI is one of several substrates of protein kinase A (PKA). In normal heart, on adrenergic stimulation PKA is activated and TnI phosphorylation at Ser22 and 23 is increased. Phosphorylation of TnI decreases myofibrillar Ca 2þ -sensitivity 2e3 fold and Table 1 Effect of the HCM related TnT K280N mutation on Ca 2þ -sensitivity and phosphorylation dependence of Ca 2þ -sensitivity using the TnT K280N patient sample (see Fig. 1). X ¼ exchanged. All data rounded to 2 significant figures. correspondingly increases the rate of Ca 2þ dissociation from TnC. This is an essential component of the lusitropic response to adrenergic stimulation, consequently DCM mutations cause a blunting of the response to b1 agonists, and this leads to a reduced cardiac reserve and predisposes the heart to failure under stress [6].
Mutations in the cardiac TnT gene (TNNT2) are the most common HCM-causing mutations after MYH7 and MYBPC3. There have been persistent reports from in vitro studies that HCM-causing mutations in the other thin filament proteins: TnI, TnC, actin and tropomyosin uncouple changes in myofilament Ca 2þ -sensitivity from TnI phosphorylation level in addition to increasing absolute Ca 2þ -sensitivity (reviewed by Messer [5] and refs therein [7e17]). However there have not been any systematic studies of mutations in TnT associated with HCM. It is possible that uncoupling is a consistent consequence of HCM-causing mutations in all thin filament proteins and if so, it could contribute to the HCM phenotype.
In this study we investigate the relationship of myofilament Ca 2þ -sensitivity to TnI phosphorylation in 7 HCM related mutations in human cardiac TnT by quantitative in vitro motility assay. Six of these are well-established mutations, whereas the seventh is a unique human heart sample from an HCM patient with a homozygous mutation, K280N. All these mutations induced uncoupling, thus extending the range of cardiac disease associated with uncoupling and raising the possibility of a physiological role in accounting for the HCM phenotype.

Sources of contractile proteins
Troponin was isolated from human or mouse heart muscle using an anti-cardiac troponin I (TnI) monoclonal antibody affinity column as described [18]. We used heart muscle from TnT R92Q transgenic mice and non-transgenic littermates [19] and explanted heart from a patient with a homozygous TnT K280N mutation with donor heart muscle as control [20]. Donor hearts had no history of cardiac disease and normal ECG and ventricular function and were obtained when no suitable transplant recipient was found. Approval was granted by the Human Research Ethics Committees of both the University of Sydney (Protocol No. 2814) and St. Vincent's Hospital (Protocol No. H91/048/1a) for collection and distribution of the human heart samples and by the NHS National Research Ethics Service, South West London REC3 (10/H0803/147) for study of the samples. Patients gave written consent with PIS approved by the relevant ethical committee. All samples are anonymised. The investigations conform to the principles of the Declaration of Helsinki. Recombinant human cardiac TnT K280N (T3 isoform) was introduced into donor heart troponin by exchange as described [18]. Recombinant whole troponin incorporating HCM-causing mutations in TnT (T3 isoform) was prepared as described [21,22].

Manipulation and measurement of TnI phosphorylation level
Troponin isolated from human heart samples has a high level of phosphorylation, which was reduced by treatment with shrimp alkaline phosphatase (Sigma P9088). Recombinant TnI was phosphorylated by treatment with protein kinase A (PKA) catalytic subunit (Sigma, P2645-400) as previously described [4,23]. TnI phosphorylation levels in isolated troponin were measured by phosphate affinity SDS-PAGE as described by Messer et al. [24].

Quantitative in vitro motility assay (IVMA)
Thin filaments were reconstituted with 10 nM rabbit skeletal or mouse cardiac muscle a-actin (labelled with TRITC phalloidin), tropomyosin (40e60 nM) and troponin (60 nM) to study Ca 2þregulation of filament motility by the quantitative in vitro motility assay (IVMA) [4,18,25]. Thin filament movement over a bed of immobilised rabbit fast skeletal muscle heavy meromyosin (100 mg/ ml) was compared in two channel motility cells in which troponin varied by a single factor (mutation, phosphorylation state or treatment with drug). The temperature was set to 29 C. Filament movement was recorded and analysed as previously described [26], yielding two parameters, the fraction of filaments moving and the speed of moving filaments.
The fraction motile and the sliding speeds were measured over a range of Ca 2þ concentrations to generate Ca 2þ -activation curves as shown previously [4,17]. The data were fitted to the 4-variable Hill equation to yield a value for EC 50 and n H . EC 50 values from replicate experiments were analysed by paired t-test since the distribution of EC 50 has been shown to be normal.

Human heart sample with TnT K280N mutation
We initially studied a TnT mutation in a unique cardiac muscle sample from a patient with HCM due to a homozygous mutation in TnT, K280N. The patient had a septal myectomy but subsequently required a heart transplant at age 26. The mutation was identified and contractility of skinned myocytes was investigated in a previous study [27].
We isolated troponin from the sample and confirmed that asparagine had exclusively replaced a lysine at amino acid 280 (K280N) by mass spectrometry (Supplemental data). The TnI phosphorylation, measured by phosphate affinity SDS-PAGE, was in the range 1.4e1.6 mols Pi/mol TnI. This is not significantly different from donor heart troponin (Table 1) and substantially greater than heart samples from septal myectomy operations, indicating that the myectomy operation had restored phosphorylation levels to normal although this did not prevent subsequent heart failure requiring heart transplant [27].
We investigated the effect of the TnT K280N mutation on troponin function in thin filaments reconstituted with actin and human cardiac tropomyosin. We used IVMA to study thin filament Ca 2þ -regulation of unloaded movement over immobilised heavy meromyosin. When we compared thin filaments containing donor heart and TnT K280N patient heart troponin there was no difference in the Ca 2þ -sensitivity or maximum sliding speed (Fig. 1, Table 1). We previously found that this pattern of results was typical of muscle from the interventricular septum of hypertrophic obstructive cardiomyopathy (HOCM) patients that generally also had an uncoupled relationship between Ca 2þ -sensitivity and TnI phosphorylation accompanied by very low levels of TnI phosphorylation [23]. We therefore determined the effect of dephosphorylating the TnT K280N patient troponin on Ca 2þ -regulation. In contrast to the 3.1-fold increase in Ca 2þ -sensitivity of donor heart troponin when it was dephosphorylated, the Ca 2þ -sensitivity of thin filaments containing TnT K280N patient troponin was independent of the level of phosphorylation (Fig. 2, Table 1). We then replaced the TnT in the TnT K280N patient sample with recombinant wild-type TnT by an exchange reaction (exchange was 89% [23]) and observed that the dependence of Ca 2þ -sensitivity on TnI phosphorylation was restored.
In a previous study of troponin from patients with obstructive HCM, we found that the myofilament Ca 2þ -sensitivity was similar to donor heart troponin, as observed here, and did not change with TnI phosphorylation level independently of the gene that was mutated [23]. In order to determine the direct effects of the TnT K280N mutation, we investigated the effect of recombinant human cardiac TnT K280N on thin filament regulation. Comparison of donor heart troponin and donor heart troponin with the TnT component replaced by TnT K280N showed that the mutation increased Ca 2þ -sensitivity 1.7-fold; a similar shift in Ca 2þ -sensitivity was obtained with 90% and 45% exchanged (Fig. 2, Table 2). However, the Ca 2þ -sensitivity of troponin containing recombinant TnT K280N did not depend on TnI phosphorylation. The increased Ca 2þ -sensitivity due to recombinant TnT K280N is typical of HCMcausing mutations in vitro and indicates that the absence of Ca 2þsensitivity shift in the patient sample was probably due to secondary abnormalities similar to troponin from the IVS of HOCM patients.

HCM-causing TnT mutations studied in transgenic mice
We next examined the HCM-related TnT mutation R92Q in the tropomyosin binding domain of TnT expressed in mouse models [19]. The level of mutation expression in these lines is 67%. Compared to troponin from non-transgenic mouse hearts, the TnT R92Q mutation increased Ca 2þ -sensitivity and uncoupled the relationship between Ca 2þ -sensitivity and TnI phosphorylation (Figs. 3 and 4, Table 3) [17].

HCM-causing TnT mutations studied using recombinant troponin
A further five HCM-causing mutations in TnT were examined using recombinant troponin expressed as the whole troponin complex in Escherichia coli. DE160, S179F and K273E are produced by missense mutations in the TNNT2 gene   Table 3. Error bars are ± SEM. Blue, WT thin filaments; Blue hatched, unphosphorylated WT thin filaments; Purple, HCM-causing TnT mutant thin filaments; Purple hatched, unphosphorylated mutant thin filaments. The previously studied ACTC E99K and TPM1 E180G were included as a comparison. A. All the HCM-causing TnT mutants show increased Ca 2þ -sensitivity (decreased EC 50 ) compared to WT thin filaments. B. All the HCM-causing TnT mutants are uncoupled in comparison to WT thin filaments.  (Fig. 4A). All of these mutations fully uncoupled the relationship between Ca 2þ -sensitivity and TnI phosphorylation, Figs. 3 and 4B, Table 3.

EGCG can recouple the uncoupled HCM related mutations in TnT
In a previous study we found that the uncoupling effect of DCMcausing mutations could be reversed by Epigallocatechin-3-gallate (EGCG) and related compounds in vitro and in myofibrils [17]. EGCG could also restore coupling to the HCM mutations, TPM1 E180G and ACTC E99K. We therefore investigated whether EGCG could restore coupling to HCM related mutations in TnT. Fig. 5A shows that modulation of Ca 2þ -sensitivity by TnI phosphorylation was fully restored to thin filaments containing the TnT R92Q mutation. This is shown by a shift of both curves to the right on addition of 100 mM EGCG with the phosphorylated curve shifting more to bring back the difference in Ca 2þ -sensitivity due to phosphorylation (as shown in Fig. 1C). This can also be shown by plotting the difference in fraction motile at a fixed Ca 2þ concentration (Fig. 5B). The other mutations were tested at a fixed concentration of 0.037 mM Ca 2þ and likewise showed restoration of modulation of Ca 2þ -sensitivity by TnI phosphorylation (Fig. 5C). Thus EGCG is capable of restoring coupling to every mutant troponin that is uncoupled both in DCM and HCM.

Regulatory properties of the homozygous TnT K280N mutant patient tissue sample
This sample enabled us to study troponin with an HCM related mutation in TnT in the context of the patient's heart. Contractility in the same sample has also been studied in skinned muscle fibres and in single myofibrils [33]. Both the isolated troponin and the skinned muscle fibres indicated that myofilament Ca 2þ -sensitivity was not very different from donor heart controls: the two were indistinguishable in IVMA whilst TnT K280N patient skinned muscle had a 20% higher Ca 2þ -sensitivity in skinned fibres. The studies in myofibrils indicated that the kinetics of contraction at saturating Ca 2þ were altered by the mutation, with k ACT and slow k REL both being increased compared to donor heart myofibrils, similar to a previous study with an HCM mutation in myosin, however, Ca 2þ dependence was not measured [34].
Exchange experiments permit the TnT K280N mutation to also be studied in troponin with a recombinant mutation introduced into 'normal' donor heart troponin. The regulatory properties are different, indicating that secondary changes have occurred to the troponin in the HCM patient's heart as previously found by Bayliss et al. [23]. Recombinant human cardiac TnT K280N increased myofilament Ca 2þ -sensitivity 1.7-fold in the IVMA, as is common with HCM-causing mutations (Fig. 2, Table 2). It is interesting that the effect of 95% mutation, a similar level to that found in the patient, and 45% mutation produced a similar Ca 2þ -sensitivity shift, suggesting that this mutation was dominant negative in common with most HCM-causing mutations. However, there is no data available on individuals with a heterozygous TnT K280N mutation that would allow us to confirm this conclusion.
In contrast to measurements of Ca 2þ -sensitivity, the effect of the mutation on the kinetics of myofibrillar force production and relaxation at saturating Ca 2þ concentrations showed that exchange of recombinant human cardiac TnT K280N into donor and also of wild-type troponin into TnT K280N patient myofibrils gave fully reversible effects [33]. Since IVMA measures unloaded movement at 29 C, whilst studies on myofibrils or permeabilised cells measure isometric force only at saturating Ca 2þ at 15 C, the studies are not directly comparable as we discussed previously [17] and therefore not contradictory.
The patient sample is unusual since the HCM phenotype was found with a high level of TnI phosphorylation, equivalent to donor heart muscle. Previously HCM was studied in tissue from septal myectomies that always had a very low level of phosphorylation of both TnI and MyBP-C [24,35,36]. Since the patient had a myectomy prior to the transplant it seems likely that the operation was beneficial to the septum, as determined by the TnI phosphorylation level, but the disease-causing mutation remained and it was still uncoupled (as is the case with DCM). The myectomy did not prevent subsequent heart failure, perhaps due to diastolic dysfunction.
When we tested the Ca 2þ sensitivity at high and low levels of phosphorylation there was no difference, i.e. the troponin was uncoupled. This is similar to findings in myectomy samples where uncoupling was independent of the HCM-causing mutations [23]. When we repeated the test with donor troponin exchanged with recombinant human cardiac TnT K280N, Ca 2þ -sensitivity was still uncoupled, so in this case a single point mutation in TnT is sufficient to uncouple Ca 2þ -sensitivity from TnI phosphorylation.

Uncoupling is a common feature of HCM related mutations in TnT
An important question addressed in this manuscript is whether all mutations in TnT that are associated with HCM are uncoupled. We therefore examined six additional well-characterised HCMlinked mutations in TnT and found that they all increased Ca 2þsensitivity as previously reported [22,37e40], and all exhibited uncoupling of Ca 2þ -sensitivity from TnI phosphorylation levels. The mutant TnT R92Q troponin extracted from transgenic mice, expressed at 65% [19], and the five TnT mutations studied using recombinant troponin were all fully uncoupled. Therefore, mutations in TnT join mutations in TnI (R145G and R21C, measured by ATPase, IVMA and in myofibrils and skinned fibres and K206Q, measured by ATPase and IVMA), TnC (L29Q, measured by IVMA and ATPase and Y5H, I148V and M103I, measured in skinned papillary muscle), actin (E99K, measured by IVMA and in myofibrils) and tropomyosin (E180G, measured by IVMA) as mutations demonstrated to cause uncoupling in vitro [7e11, 13,14,17,32,41]. Thus, for HCM-causing mutations in genes coding for sarcomeric thin filament proteins, Ca 2þ -sensitisation together with uncoupling in vitro is likely to be a common consequence and they may contribute to the HCM phenotype.
We recently demonstrated that EGCG and related compounds could restore coupling to DCM-causing mutations in thin filament proteins [17]. We now show that EGCG can also restore coupling to all the uncoupled HCM-causing TnT mutations. Re-coupling by EGCG was also observed for ACTC E99K and TPM1 E180G suggesting a common mechanism [17]. In fact the combination of Ca 2þdesensitisation and re-coupling reverses both the abnormalities found in troponin with an HCM mutation. This is illustrated for the TnT R92Q mutation where the EGCG treatment fully restores motility to the levels of wild-type Ca 2þ -sensitivity and phosphorylation dependence (Fig. 6). This suggests a potential use for recoupling compounds based on EGCG as treatment for HCM. Error bars represent SEM of four measurements of motility in the same motility chamber. The TnT R92Q mutation causes the thin filaments to be uncoupled and EGCG re-couples the system and brings back the difference in Ca 2þ -sensitivity seen with wild-type (or donor) thin filaments.

Funding
This work was supported by grants from the British Heart Foundation (RG/11/20/29266 and FS/12/24/29568) and the Seventh Framework Program of the European Union 'BIG-HEART' (grant agreement 241577).