INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

FAMILIAL HYPERTROPHIC CARDIOMYOPATHY (HCM) is an autosomal dominant cardiac disease associated with a high incidence of sudden death. HCM is a genetically heterogeneous disease that has been shown to be caused by mutations in the genes for a variety of cardiac myofibrillar proteins, including -myosin heavy chain (-MHC) (8), troponin T (TnT) (33), troponin I (TnI) (11), -tropomyosin (-TM) (33), myosin-binding protein C (3, 32), and myosin light chains (22).

HCM is also clinically variable, and the mutations in cardiac TnT are known to be associated with a high risk of sudden death despite an incomplete disease penetrance with moderate hypertrophy (33), in contrast to the high penetrance with more severe hypertrophy of -MHC mutations with a comparably poor prognosis (1, 6, 24, 34). Mutations in the -MHC gene in general impair the interaction between myosin and actin so that they depress cardiac performance and lead to a compensatory cardiac hypertrophy (5, 7, 13, 25). At least 13 different mutations in the TnT gene have been identified: 11 missense mutations, a mutation involving a deleted codon, and a splice donor site mutation (21). We have previously shown that two missense mutations, the substitution of Asn for Ile at position 79 (Ile79Asn) and Arg92Gln, have a Ca²⁺-sensitizing effect on the contraction of cardiac muscle without affecting the maximum Ca²⁺-activated force and the cooperativity (15), suggesting that the mutations in TnT are not a stimulus for compensatory cardiac hypertrophy and that Ca²⁺ sensitization may be responsible for the distinguishing feature of HCM caused by TnT mutations, i.e., incomplete disease penetrance and poor prognosis.

The clinical phenotype of the splice donor site mutation in intron 15 of the cardiac TnT gene has been shown to be similar to those of the missense mutations Ile79Asn and Arg92Gln (33). This mutation results in aberrant splice products encoding two truncated TnT molecules lacking COOH termini (29). One truncated mutant TnT simply lacks the COOH-terminal 14 amino acids (TnT₁₄), and the other mutant lacks 21 amino acids because of the replacement of the normal 28 terminal amino acids with 7 novel residues (TnT₂₈₍₊₇₎). It has been shown that TnT₂₈₍₊₇₎ can be incorporated into troponin complexes within a sarcomere when expressed in a cultured, transfected quail skeletal myotube and that it greatly diminishes the maximum Ca²⁺-activated force (35), suggesting that the splice donor site mutation in the TnT gene acts as a dominant-negative allele that impairs cardiac contractile performance and leads to a compensatory hypertrophy as in the case of the -MHC mutations. However, in this quail skeletal myotube expression system, it was reported to be technically impossible to control the extent of mutant TnT incorporation into the sarcomeres, and the contractile response to the submaximal Ca²⁺ concentrations (e.g., Ca²⁺ sensitivity and cooperativity) remains to be determined because of a dramatic depression of maximum force. In the present study, we attempted to determine the functional effects of this splice donor site mutation on the cardiac muscle contraction by exchanging the two truncated human cardiac TnT molecules directly into rabbit skinned cardiac muscle fibers. The extent of mutant TnTs exchanged into the skinned fibers was 50-55%, which was close to the heterozygous state expected in most patients. Both mutants increased the Ca²⁺ sensitivity, as in the case of the missense mutations Ile79Asn and Arg92Gln, and also decreased the cooperativity of the contraction of skinned cardiac muscle fibers. Whereas TnT₂₈₍₊₇₎ slightly decreased the maximum Ca²⁺-activated force, TnT₁₄ had no significant effect on the maximum force under these conditions. The results suggest that these functional changes in cardiac muscle, an increase in Ca²⁺ sensitivity and decreases in cooperativity and maximum force, might be involved in the pathogenesis of HCM associated with the cardiac TnT splice donor site mutation.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Mutagenesis of human cardiac TnT cDNA. Human cardiac TnT cDNA was amplified by RT-PCR of human heart mRNA purchased from Clontech (Palo Alto, CA). The PCR products were subcloned into the pUC119 vector for screening by restriction analysis and DNA sequencing. The obtained wild-type TnT cDNA was then constructed in the pET3-d vector, and mutagenesis was carried out by PCR. To generate TnT₂₈₍₊₇₎ cDNA, PCR was first carried out to create Spl I and EcoR I sites by using oligonucleotides 5'-GGT GGT GGA AG A AGA GG-3' (18F; Spl I site underlined) and 5'-CGG AGA ACA TTA TAT TTC-3' (794R; EcoR I site underlined). The obtained PCR product was then digested by Spl I and EcoR I and connected to oligonucleotides 5'-CAT GTC TGA CAT CGA AGA AGT GGT GGA AGA-3', to be ligated to the Nco I site of the pET-3d vector, and 5'-AAT TGC AAG ACC CGC GGG AAG GCT AG-3', to replace the COOH-terminal 28 amino acids with 7 novel amino acids. To generate TnT₁₄ cDNA, PCR was carried out to introduce a stop codon and a BamH I site at the position of Ser-275 by using oligonucleotides 5'-GGT GGT GGA AGA AGA GG-3' (18F; Spl I site underlined) and 5'-CCC GCG G TA GAC TTT CTG G-3' (837R; BamH I site underlined). The obtained PCR product, digested by Spl I and BamH I, was connected to oligonucleotide 5'-CAT GTC TGA CAT CGA AGA AGT GGT GGA AGA-3' to be ligated to the Nco I site of the pET-3d vector and thereby was joined with the vector. The complete nucleotide sequences of the mutant TnT cDNAs were confirmed by DNA sequencing.

Purification of proteins. Expression and purification of the recombinant human cardiac TnTs were performed as described previously (15). Recombinant TnT was purified with a fast-performance liquid chromatography (FPLC) ion-exchange column, RESOURCE Q (6 ml; Pharmacia Biotech), with a linear gradient of 0-0.5 M NaCl in the presence of 20 mM Tris · HCl (pH 8.0), 6 M urea, 5 mM trans-1,2-cyclohexanediamine-N, N,N',N'-tetraacetic acid (CDTA), and 15 mM 2-mercaptoethanol; recombinant TnT was eluted at 0.3-0.4 M NaCl. Native troponin and its subunits, TnT, TnI, and troponin C (TnC), were prepared from the left ventricular myocardiums of young male albino rabbits (~3 mo old) according to the method of Tsukui and Ebashi (30), by using FPLC ion-exchange columns RESOURCE Q and RESOURCE S (6 ml) and Mono Q and Mono S HR 5/5 (Pharmacia Biotech).

Dephosphorylation and phosphorylation of TnI. Purified native cardiac TnI (0.5 mg/ml) was dephosphorylated by incubation with 50 U/ml alkaline phosphatase, type VII NL (Sigma Chemical), for 10 min at 25°C in 5 ml of a solution containing 20 mM Tris · HCl (pH 9.8), 0.25 M KCl, 1 µg/ml pepstatin A, and 0.2 mM phenylmethylsulfonyl fluoride (PMSF). The reaction was stopped by the addition of 6 M urea, and dephosphorylated TnI was purified by FPLC cation-exchange column POROS HS/M (PerSeptive Biosystems), with a linear gradient of 0-0.5 M NaCl in the presence of 20 mM Tris · HCl (pH 8.0), 6 M urea, 5 mM CDTA, and 15 mM 2-mercaptoethanol; dephosphorylated TnI was eluted at 0.3-0.4 M NaCl.

Preparation of skinned fibers and force measurements. Skinned fibers were prepared from the left ventricular trabeculae of young male albino rabbits (~3 mo old), and force measurements were performed as described previously (15). Briefly, small bundles (0.5-1 mm wide and 5-7 mm long) of trabeculae tied to glass capillary tubes were skinned with relaxing solution containing 0.5% Brij-58 for 30 min at 25°C and were stored up to 3 wk at 20°C in relaxing solution containing 50% glycerol. A small fiber (~120 µm in diameter) dissected from the stock skinned trabecula was mounted in a thermostatically controlled chamber with a capacity of 0.2 ml. The fiber length between hooks was ~1 mm, and the resting sarcomere length was set to 2.3 µm by laser diffraction. The force was measured with strain gauge UL-2GR (Minebea). The relaxing solution consisted of (in mM) 50 MOPS-KOH (pH 7.0), 100 KCl, 6 MgCl₂, 5 ATP, 4 EGTA, 0.5 DTT, and 10 creatine phosphate, as well as 35 U/ml creatine kinase. Activating solutions with desired free Ca²⁺ concentrations were prepared by adding appropriate amounts of CaCl₂, calculated as described previously (14), to the relaxing solution. All force measurements were done at 25°C.

(pCa  pCa₅₀)n_H

TnT exchange in skinned fibers. TnT exchange in the skinned fibers was performed by the method described previously (15).

SDS-PAGE and immunoblot analyses. SDS-PAGE was carried out at 8 or 12% acrylamide concentration according to the method of Laemmli (12). The fiber samples were lysed in Laemmli's sample buffer by heating for 4 min at 95°C after having been frozen (80°C) and thawed several times. The gel was stained with silver with a staining kit (Pharmacia Biotech) or with Coomassie brilliant blue R-250. Immunoblot analysis was performed after electrophoretically transferring proteins from unstained SDS-12% polyacrylamide gels to polyvinylidene difluoride membranes (0.2 µm; Bio-Rad). After fixation with 0.2% glutaraldehyde (31), the blots were blocked with 3% nonfat dry milk and reacted with primary antibody; primary antibodies used in the present study were monoclonal anti-cardiac TnT antibody 1F2 (HyTest), monoclonal anti-cardiac TnI antibody 8E10 (HyTest), monoclonal anti-tropomyosin (sarcomeric) antibody CH1 (Sigma Chemical), and a rabbit polyclonal anti-phosphoserine antibody that should not differentiate between mono- and diphosphorylated forms of TnI (Zymed Laboratories). After being washed, the membranes were reacted with an alkaline phosphatase-conjugated second antibody, and the antigens of primary antibodies were visualized by a chemiluminescent detection reagent (CDP-Star; NEN Life Science Products). An optical densitometric scan was performed with Phoretix International gel analysis software calibrated by a photographic step tablet (21 steps; density range 0.05-3.03; Eastman Kodak). The linearity of the densitometric readings from immunoblots was confirmed by changing the amount of sample applied on the gel in steps.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Figure 1A shows the results of SDS-PAGE of the bacterially expressed wild-type and HCM-causing truncated mutants of human adult cardiac TnT. The recombinant human wild-type cardiac TnT showed an electrophoretic mobility similar to that of the native cardiac TnT purified from the hearts of adult rabbits, consistent with the study reporting that the major isoforms of adult human and rabbit cardiac TnTs show almost the same electrophoretic mobilities on a similar gel (2). Both the two truncated mutant cardiac TnTs showed slightly higher electrophoretic mobility than wild-type TnT, as expected.

View larger version (32K): [in this window] [in a new window]   Fig. 1.   A: SDS-PAGE of purified recombinant wild-type, and truncated mutants of, human cardiac troponin T (TnT). Lane 1, rabbit cardiac tissue-derived (native) TnT; lane 2, human cardiac wild-type TnT; lane 3, human cardiac truncated mutant TnT28(+7); lane 4, human cardiac truncated mutant TnT14; lane 5, human cardiac wild-type TnT plus truncated mutant TnT28(+7); lane 6, human cardiac wild-type TnT plus truncated mutant TnT14. Acrylamide concentration of separating gel was 8%, and gel was stained with Coomassie brilliant blue R-250. B: SDS-PAGE of rabbit skinned cardiac muscle fibers treated with human recombinant cardiac TnTs and reconstituted with rabbit native cardiac troponin I (TnI) and troponin C (TnC). Acrylamide concentration of separating gel was 12%, and gel was stained with a silver staining kit (Pharmacia Biotech). LC2, myosin light chain 2; TM, tropomyosin. C: immunoblot analysis of TnT-treated fibers before reconstitution with rabbit native cardiac TnI and TnC probed with primary antibodies against TnT, TM, and TnI. D: immunoblot analysis of TnT-treated fibers after reconstitution with rabbit native cardiac TnI and TnC probed with primary antibodies against TnT, TM, and TnI. In B-D, untreated (control) rabbit cardiac skinned fibers were loaded in lane 1 and fibers treated with human cardiac wild-type TnT, TnT28(+7), and TnT14 were loaded in lanes 2-4, respectively.

It has previously been demonstrated that endogenous TnT-TnI-TnC complexes in skinned muscle fibers are selectively replaced with exogenous TnT when the fibers are treated with an excess amount of TnT under acidic and high-ionic-strength conditions (9, 10, 15). To investigate the functional consequence of the splice donor site mutation in the TnT gene, an attempt was made in the present study to directly exchange the two human recombinant truncated TnTs, as well as the recombinant wild-type TnT, into adult rabbit cardiac skinned muscle fibers by this technique. SDS-PAGE analysis of the TnT-treated fibers after reconstitution with rabbit native cardiac TnI and TnC indicated that the two truncated TnTs were incorporated into the fibers without changing the stoichiometry of other myofibrillar constituents (Fig. 1B). Immunoblot analysis of the TnT-treated fibers using monoclonal antibodies against TnT, TM, and TnI directly demonstrated that the two truncated TnTs were exchanged into the skinned fibers (Fig. 1C); ratios of truncated TnT to total TnT in lanes 3 and 4 estimated by optical densitometric scans were 0.50 and 0.54, respectively. Quantification of the relative amounts of native and truncated TnT in a subgroup of fibers indicated that the extents of TnT₂₈₍₊₇₎ and TnT₁₄ exchanged were 54.5 ± 2.3 and 56.8 ± 2.8% (means ± SE of measurements on 3 fibers). The immunoblot analysis of the TnT-treated fibers also showed that the TnT treatment caused a decrease in the amount of native TnI, an inhibitory subunit of the troponin complex, in all fibers, indicating that endogenous TnT-TnI-TnC complexes were just displaced by exogenously added TnT (Fig. 1C); TnI-to-TM ratios in lanes 1, 2, 3, and 4 were 1.41, 0.65, 0.70, and 0.63, respectively. Quantification of the decreased amount of TnI in a subgroup of fibers by densitometric scans indicated that the extents of the endogenous troponin complex displaced by TnT and thus the extents of TnT exchanged into the fibers were 54.9 ± 2.6, 51.4 ± 2.1, and 55.8 ± 1.2% (means ± SE of measurements on 3 fibers) for wild-type TnT, TnT₂₈₍₊₇₎, and TnT₁₄, respectively. Consistent with this, the skinned fibers treated with these three recombinant TnTs developed a large force even in the absence of Ca²⁺ (Fig. 2A). Reconstitution of the TnT-treated skinned fibers with rabbit native cardiac TnI inhibited the Ca²⁺-insensitive force generation completely (Fig. 2B). Immunoblot analysis indicated that the reincorporation of TnI was nearly stoichiometric in all the skinned fibers treated with the three recombinant TnTs (Fig. 1D); TnI-to-TM ratios in lanes 1, 2, 3, and 4 were 1.43, 1.39, 1.47, and 1.31, respectively. Subsequent reconstitution of the skinned fibers with rabbit native cardiac TnC restored Ca²⁺-dependent force generation (Fig. 2C). The TnC-to-myosin light chain 2 ratios estimated by optical densitometric scans of the SDS-PAGE gel in B were 0.25, 0.23, 0.23, and 0.24 in lanes 1, 2, 3, and 4, respectively, indicating that the reincorporation of TnC is also stoichiometric in all cases.

View larger version (12K): [in this window] [in a new window]   Fig. 2.   Changes in force response to Ca2+ by TnT exchange into rabbit skinned cardiac muscle fibers. A: changes after treatment with recombinant human cardiac TnTs. B: changes after reconstitution with rabbit cardiac TnI. C: changes after reconstitution with rabbit cardiac TnC. Forces measured in absence of Ca2+ (pCa > 9; open bars) and in presence of a saturating amount of Ca2+ (pCa < 5.5; solid bars) were normalized to maximum force developed by same fiber before TnT treatment and are expressed as means ± SE of measurements on 8-10 fibers.

A previous study showed that the exchange of the human wild-type cardiac TnT into rabbit cardiac skinned fibers caused a decrease in the maximum force and also caused a slight rightward shift in, and a decrease in the steepness of, the normalized force-pCa relationship (15). However, the previous study also demonstrated that the exchange of tissue-derived rabbit native cardiac TnT into rabbit skinned cardiac muscle fibers had exactly the same effects on these parameters. This indicates that although the TnT exchange procedure itself has a nonspecific deteriorative effect on both contractile and regulatory systems to some extent, the displacement of the endogenous troponin complex in rabbit skinned cardiac muscle fibers with recombinant human cardiac TnT and the subsequent reconstitution with rabbit cardiac TnI and TnC constitute an effective approach for investigating the functional effects of the HCM-causing mutations in human cardiac TnT on cardiac muscle contraction. In the following experiments, the wild-type human cardiac TnT-exchanged fiber was taken as a control to detect a specific functional effect exerted on the fibers by mutant human cardiac TnT, as in the previous study (15).

Figure 3 compares the force-pCa relationships of the fibers into which the two truncated mutants of human cardiac TnT were exchanged with the force-pCa relationship of the fibers into which the human wild-type cardiac TnT was exchanged. Table 1 summarizes the Ca²⁺ sensitivity (pCa₅₀) and cooperativity (n_H) in the force-pCa relationships, as well as the normalized maximum force. Although the maximum force generated by the TnT₁₄-exchanged fibers was not statistically significantly different from that generated by the wild-type TnT-exchanged fibers, the TnT₂₈₍₊₇₎-exchanged fibers developed a slightly but statistically significantly lower level of maximum force than the wild-type TnT-exchanged fibers (Fig. 3 and Table 1). Both these two truncated TnTs conferred a higher Ca²⁺ sensitivity and a lower cooperativity than did wild-type TnT, as demonstrated by a significant increase in pCa₅₀ and a significant decrease in n_H (Table 1). Before TnT exchange, there were no statistically significant differences in Ca²⁺ sensitivity and cooperativity between these three fiber groups (Fig. 3, inset).

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Effect of exchanging truncated mutant TnT into skinned cardiac muscle fibers on force-pCa relationships. Force-pCa relationships determined after exchange of recombinant human cardiac wild-type TnT (), TnT28(+7) (), and TnT14 () were compared; forces were normalized to averaged maximum force in fibers with wild-type TnT exchanged and are means ± SE of measurements on 8-10 fibers. Inset, force-pCa relationships determined in respective fibers before TnT exchange; forces were normalized to maximum force developed by each fiber.

It is known that phosphorylation of TnI by PKA results in a decrease in the Ca²⁺ sensitivity of cardiac muscle contraction, which may play an important role in the positive inotropic effect of -adrenergic stimulation. To see whether the truncated mutant TnT affects this Ca²⁺-desensitizing effect of TnI phosphorylation by PKA, the wild-type TnT- and the two truncated TnT-treated fibers were reconstituted with dephosphorylated or phosphorylated TnI and TnC. Figure 4A shows an immunoblot analysis of the dephosphorylated or phosphorylated TnI with antibodies against TnI and phosphoserine, which confirms the phosphorylation and dephosphorylation of TnI used for reconstitution. Immunoblot analyses of the fibers reconstituted with dephosphorylated or phosphorylated TnI with antibodies against TnI and TM also confirmed that both the phosphorylated and dephosphorylated TnIs were incorporated stoichiometrically into the fibers (Fig. 4B). In both the wild-type and truncated TnT-exchanged fibers, force-pCa relationships conferred by the reconstitution with phosphorylated TnI were significantly shifted rightward compared with those conferred by the reconstitution with dephosphorylated TnI (Fig. 4B), indicating that the Ca²⁺ sensitivity is decreased by TnI phosphorylation in the presence of truncated TnT as well as in the presence of wild-type TnT; it should be noted that the pCa₅₀ values for the fibers reconstituted with untreated TnI in Fig. 3 (Table 1) are intermediate between the values for the fibers reconstituted with phosphorylated and dephosphorylated TnI (Table 2), in agreement with the fact that the untreated TnI has already been phosphorylated considerably in vivo (Fig. 4A). The truncated TnT-exchanged fibers always showed significantly higher Ca²⁺ sensitivity and lower cooperativity than did the wild-type TnT-exchanged fibers irrespective of whether they were reconstituted with phosphorylated or dephosphorylated TnI (Fig. 4B and Table 2). Before TnT exchange, there were no statistically significant differences in Ca²⁺ sensitivity and cooperativity between these four fiber groups (data not shown). These results indicate that the two truncated mutant TnTs increase the Ca²⁺ sensitivity and decrease the cooperativity independently of the phosphorylation of TnI by PKA. Although the two truncated mutant TnTs had no significantly different effect on these two parameters, TnT₂₈₍₊₇₎ conferred a significantly lower maximum force than wild-type TnT and TnT₁₄ when reconstituted with phosphorylated TnI.

View larger version (30K): [in this window] [in a new window]   Fig. 4.   A: immunoblot analysis of dephosphorylated and phosphorylated TnI. Purified TnI was dephosphorylated by treatment with alkaline phosphatase or phosphorylated by cAMP-dependent protein kinase (PKA) after dephosphorylation as described in MATERIALS AND METHODS, and subjected to immunoblotting with anti-TnI antibody and anti-phosphoserine antibody. Lane 1, untreated TnI; lane 2, phosphorylated TnI after dephosphorylation; lane 3, dephosphorylated TnI. B, top: immunoblot analysis of TnT-treated fibers after reconstitution with dephosphorylated or phosphorylated TnI and TnC probed with primary antibodies against TM and TnI. Lanes 1 and 4, untreated (control) rabbit cardiac skinned fibers; lane 2, wild-type TnT-exchanged fiber reconstituted with phosphorylated TnI; lane 3, wild-type TnT-exchanged fiber reconstituted with dephosphorylated TnI; lane 5, TnT28(+7)-exchanged fiber reconstituted with phosphorylated TnI; lane 6, TnT28(+7)-exchanged fiber reconstituted with dephosphorylated TnI. TnI-to-TM ratios in lanes 1-6, estimated by optical densitometric scans, were 0.90, 0.88, 0.98, 0.95, 0.87, and 0.88, respectively. Data are representative of those from 3 other experiments. Bottom: effect of TnI phosphorylation on force-pCa relationships in wild-type and truncated mutant TnT-exchanged skinned fibers. Wild-type TnT- (, ), TnT28(+7)- (, ), and TnT14 (, )-treated fibers were reconstituted with TnI that had been dephosphorylated by phosphatase (, , ) or phosphorylated by PKA (, , ) and were further reconstituted with TnC. Forces were normalized to maximum force developed by each fiber and are means ± SE of measurements on 3-5 fibers.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The present study demonstrates that both the two truncated cardiac TnTs produced by the HCM-causing splice donor site mutation increase the Ca²⁺ sensitivity and decrease the cooperativity of the Ca²⁺-activated force generation in cardiac muscle and that the short truncated mutant, TnT₂₈₍₊₇₎, furthermore decreases the maximum force. By the same method used in the present study, we have previously shown that the HCM-causing missense mutations Ile79Asn and Arg92Gln in the TnT gene also increase the Ca²⁺ sensitivity but do not alter the cooperativity and maximum force (15). Because the increase in Ca²⁺ sensitivity is expected to increase the force generated at submaximal intracellular Ca²⁺ concentrations, these HCM-causing TnT mutations may have a common functional effect of directly enhancing the cardiac performance, in contrast to the missense mutations in the -MHC gene, which are considered to cause a depression of cardiac performance that leads to a compensatory hypertrophy.

Because the increased Ca²⁺ sensitivity would also involve an increase in the myofibrillar ATPase activity (i.e., an increase in the utilization of ATP by actomyosin ATPase) at submaximal intracellular Ca²⁺ concentrations, these mutations in TnT may facilitate the rapid exhaustion of the intracellular ATP under severe stress, leading to an imbalance in energy supply and demand in the heart that might be responsible for sudden death (23). The increased Ca²⁺ sensitivity may also cause a diastolic dysfunction by impairing relaxation, as has recently been shown in hearts from transgenic mice expressing the human cardiac truncated mutant TnT₂₈₍₊₇₎ (28). The -adrenergic stimulation activates PKA and causes a decrease in Ca²⁺ sensitivity through the phosphorylation of TnI, which plays a cardioprotective role in preventing the overstimulation of myofilaments by a dramatic increase in the intracellular Ca²⁺ concentration (23). The present study shows that the truncated TnT has a Ca²⁺-sensitizing effect independent of the phosphorylation of TnI by PKA, suggesting that these unfavorable effects of the increased Ca²⁺ sensitivity may be augmented by the positive inotropic effect of -adrenergic stimulation. On the other hand, the decrease in cooperativity caused by the truncated TnT is expected to make the force response of cardiac muscle to changes in intracellular Ca²⁺ concentration less steep, but it is difficult to predict its impact on cardiac performance.

Analysis of a quail skeletal myotube expression system has shown that the truncated mutant TnT₂₈₍₊₇₎ is incorporated into the sarcomeres of myotubes in vivo and greatly inhibits (by ~80%) the maximum force generation (35). This indicates that the splice donor site mutation in the TnT gene acts as a dominant-negative allele but not as a null allele. The present study indicates that this truncated TnT can also be incorporated into the sarcomeres of cardiac muscle in vitro as effectively as wild-type TnT and inhibits the maximum force generation. Of interest is the fact that the fibers with this mutant TnT exchanged produce significantly less force than the fibers with wild-type TnT or TnT₁₄ exchanged in the absence of Ca²⁺ even before the reconstitution with TnI and TnC (Fig. 2A). Because the extent of TnT₂₈₍₊₇₎ exchanged was no different from the extents of wild-type TnT and TnT₁₄ exchanged, this may indicate that TnT₂₈₍₊₇₎ itself has a direct inhibitory effect on the force-generating interaction of the thin filament with myosin cross bridges. Another interesting finding is that the inhibition of the maximum force by TnT₂₈₍₊₇₎ has only been observed in the fibers reconstituted with phosphorylated TnI (Table 2). This suggests that the maximum force-generating capability of cardiac muscle in patients might be impaired by -adrenergic stimulation. However, the inhibitory effect of TnT₂₈₍₊₇₎ on the force generation in skinned rabbit cardiac muscle observed in the present study was quite small (~20%) compared with that observed in the quail skeletal myotube expression system. This discrepancy is probably due to a difference in the amount of the truncated TnT incorporated into the sarcomere; replacement of endogenous TnT with truncated TnT was nearly complete in the quail skeletal myotube expression system, whereas it was 50-55% in the skinned cardiac muscle system in the present study. Because HCM is an autosomal dominant disorder, most patients with the splice donor site mutation in the TnT gene are expected to be heterozygous, with the normal and the two truncated mutant proteins being expressed and probably incorporated into the sarcomere in the ratio of 2:1:1 (29). The present study showed that TnT₁₄ has no effect on maximum force-generating capability so that the splice donor site mutation in the TnT gene could not have a large effect on the maximum force-generating capability of cardiac muscle in patients. Because intact cardiac muscle is generally thought to be never activated beyond about half-maximal level (23) and because both truncated mutant TnTs are expected to just increase the force at the submaximal Ca²⁺ concentration (Fig. 3), the splice donor site mutation in the TnT gene would potentially have an enhancing effect on the force generation in the living heart.

The contraction of vertebrate striated muscle is regulated by a specific regulatory protein complex, troponin, located at regular intervals in the thin filament. TnT has an important structural role in anchoring the other constituents of the troponin complex, TnI and TnC, to TM (17). Previous studies of the rabbit fast-twitch skeletal TnT molecule have revealed that a highly -helical region of 81 amino acid residues at the NH₂-terminal side of the molecule constitutes a primary, strong TM-binding site and that a COOH-terminal region of the molecule constitutes a second, weak TM-binding site (20, 27). Chymotryptic digestion is known to split the fast-twitch skeletal TnT into two subfragments designated TnT₁ and TnT₂ (or TnT₂); TnT₁ is the NH₂-terminal three-fifths of TnT containing the primary TM-binding region, and TnT₂ is the remaining COOH-terminal two-fifths of TnT constituting the second TM-binding region and retaining a partial regulatory function (16, 26). TnT₂ is further digested into a smaller subfragment, TnT₂, which lacks the COOH-terminal 17 residues (26, 27). It has been shown that the regulatory function retained in TnT₂ (or TnT₂) is lost in TnT₂ such that the troponin complex cannot inhibit the contractile response in the absence of Ca²⁺ (18, 19), indicating that the COOH-terminal 17 residues of rabbit fast-twitch skeletal TnT play an important role in allowing the troponin complex to inhibit muscle contraction in the absence of Ca²⁺. It should be noted that these COOH-terminal 17 amino acid residues of rabbit fast-twitch skeletal TnT are just homologous to the COOH-terminal 14 amino acid residues of human cardiac TnT, which are lost in the HCM-causing truncated mutants, TnT₁₄ and TnT₂₈₍₊₇₎. Thus the splice donor site mutation in cardiac TnT associated with HCM may impair the inhibitory action of the troponin complex, leading to an increase in the Ca²⁺ sensitivity. The present study also showed that both the truncated mutant TnTs in HCM conferred a lower cooperativity than wild-type TnT, providing the first evidence that this COOH-terminal region of cardiac TnT is involved in the cooperative Ca²⁺ regulation of contraction.

Recently, missense mutation Asp175Asn in the -TM gene has been shown to increase the Ca²⁺ sensitivity without changing the cooperativity and maximum force-generating capability in skeletal muscle fibers of HCM patients (4), suggesting that the increased Ca²⁺ sensitivity may be a common functional effect of the HCM-causing mutations in the two distinct thin-filament regulatory proteins, cardiac TnT and -TM. Further studies on the function of other mutations in the TnT gene would clarify this possibility.