HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Tobacman, L. S. Articles by Homsher, E. Search for Related Content PubMed PubMed Citation Articles by Tobacman, L. S. Articles by Homsher, E. Social Bookmarking                      What's this?

J Biol Chem, Vol. 274, Issue 40, 28363-28370, October 1, 1999

Functional Consequences of Troponin T Mutations Found in Hypertrophic Cardiomyopathy^*

Larry S. Tobacman^§¶, David Lin^§, Carol Butters, Cheryl Landis, Nick Back, Dmitry Pavlov, and Earl Homsher

From the Departments of  Internal Medicine and ^§ Biochemistry, The University of Iowa, Iowa City, Iowa 52242 and the  Department of Physiology, University of California, Los Angeles, California 90024

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Missense mutations in the cardiac thin filament protein troponin T (TnT) are a cause of familial hypertrophic cardiomyopathy (FHC). To understand how these mutations produce dysfunction, five TnTs were produced and purified containing FHC mutations found in several regions of TnT. Functional defects were diverse. Mutations F110I, E244D, and COOH-terminal truncation weakened the affinity of troponin for the thin filament. Mutation E160 resulted in thin filaments with increased calcium affinity at the regulatory site of troponin C. Mutations R92Q and F110I resulted in impaired troponin solubility, suggesting abnormal protein folding. Depending upon the mutation, the in vitro unloaded actin-myosin sliding speed showed small increases, showed small decreases, or was unchanged. COOH-terminal truncation mutation resulted in a decreased thin filament-myosin subfragment 1 MgATPase rate. The results indicate that the mutations cause diverse immediate effects, despite similarities in disease manifestations. Separable but repeatedly observed abnormalities resulting from FHC TnT mutations include increased unloaded sliding speed, increased or decreased Ca²⁺ affinity, impairment of folding or sarcomeric integrity, and decreased force. Enhancement as well as impairment of contractile protein function is observed, suggesting that TnT, including the troponin tail region, modulates the regulation of cardiac contraction.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Familial hypertrophic cardiomyopathy (FHC)¹ can be caused by dominant mutations in genes encoding any one of several proteins of the cardiac contractile apparatus: myosin, C-protein, tropomyosin, or troponin (1-5). These molecules have central roles in contraction or its regulation, suggesting that abnormal contraction of the heart leads to the clinical, histological, and morphological manifestations of FHC. Strong support for this conclusion comes from three related types of observations: (i) purified mutant proteins retain many of their basic functions but display kinetic or other abnormalities that would be expected to alter muscle contraction (6-11); (ii) altered contraction occurs in isolated muscle fibers or cells expressing the mutant protein (12-15); and (iii) transgenic animals with FHC-linked mutations exhibit altered cardiac function (15-18).

One of the more commonly affected genes in FHC is that for the tropomyosin binding subunit of troponin, troponin T (1). Like the abnormalities in other genes that cause this syndrome, the troponin T mutations occur widely throughout the sequence, are generally missense or point deletions, and occasionally are premature truncations. Although there is less severe hypertrophy in kindreds with the troponin T mutations than is observed in other FHC patients, these kindreds show a high incidence of sudden death and have mortality rates as high as those accompanying the most severe mutations in the myosin heavy chain gene (1). This raises the possibility that the contraction abnormalities resulting from troponin T mutations may differ from those found in FHC more generally.

The troponin T mutations were first reported in 1994 and 1995 (1, 2), and since then some experimental information has emerged concerning their properties and effects. The troponin T mutation best characterized at the level of the purified protein is the missense mutant Ile⁷⁹  Asn. Troponin containing a rat recombinant form of this mutation functioned normally in many respects but had the unexpected effect of increasing the speed that thin filaments slide over a myosin fragment-coated surface (8). This was not anticipated from the known properties of troponin T and was different from the characteristic consequences of FHC-causing mutations in the cardiac myosin heavy chain gene. More recently it has been shown that this same troponin T mutation decreases the force that is exerted by myosin (13). When overexpressed in quail myotubes, two other troponin T mutations also decrease force, and a different mutation increases the unloaded shortening speed (12, 13). However, information about these and other mutants is limited, so the overall picture remains unclear. Troponin T is highly elongated and interacts with all the other polypeptides of the thin filament: tropomyosin, actin, and the other two troponin subunits, troponin I and troponin C (see Ref. 19 for review). Therefore, there are many possibilities for how the mutations could alter function.

To characterize the functional properties of the troponin T mutations present in FHC, we now report effects of mutations introduced into bovine cardiac troponin T via mutagenesis of cDNA isolated from a bovine heart cDNA library. Four new functional abnormalities were identified using the bacterially expressed mutant proteins. One of the FHC mutants increased the calcium affinity of the troponin C binding site that regulates cardiac contraction. Another mutation decreased the thin filament-myosin subfragment 1 MgATPase rate and increased in vitro sliding speed, implying impaired myosin cross-bridge function. Several mutants caused either weakened binding of troponin to the thin filament or impaired troponin solubility, abnormalities that may affect the relative incorporation of normal and mutant troponin Ts into the thin filaments of heterozygous patients. The results are discussed for their significance regarding both FHC pathophysiology and the normal functions of troponin T in muscle contraction.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cloning of Bovine Cardiac Troponin T cDNA-- Bovine heart cDNA was isolated from a bovine heart cDNA library in the Unizap XR vector system (Stratagene). The library, created from the heart of a 2-year-old animal, was screened by colony hybridization using a primer obtained by nick translation of a rat cardiac troponin T cDNA template (20). Hybridization was done at 65 °C in the presence of 6× SSC, 20 mM NaH₂PO₄, 0.4% SDS, 10 µg/µl salmon sperm DNA, 200,000 cpm/ml probe. After primary and secondary screening the positive plaques were excised using the Stratagene Unizap excision protocol resulting in pBluescript SK vectors containing the library inserts. Clones were then screened by restriction digest. Five clones large enough to contain the bovine cardiac troponin T sequence were chosen. Sequencing was performed, with primer walking where necessary, by the dideoxy chain termination method (21). All of the isolates were found to have open reading frames encoding portions of bovine cardiac troponin T, as previously characterized by protein sequencing methods (22). Different isolates encoded regions defining the two known isoforms, which differ by the presence or absence of residues 15-19 (22). The longer isoform is predominant in the bovine heart and was used for mutagenesis. To obtain the entire coding region, fragments from two isolates were subcloned into pSP72 and combined.

The protein sequence deduced from the troponin T cDNA agreed with the published protein sequence, with one discrepancy. The codon for Glu⁴² is missing from the cDNA. Although this portion of the nucleotide sequence is based upon only one cDNA isolate, we do not believe it is an artifact. Alignment with the rat cardiac troponin T genomic sequence (23) indicates that Glu⁴² arises from a split exon at the boundary of exons 5 and 6. An alternative splice acceptor site, shifted by 3 base pairs into exon 6, would produce this result. There is precedent for this from reverse transcription-PCR analysis of human cardiac troponin T (24). Also, variable deletions of single Glu residues were found at two sites in the hypervariable region of troponin T (25) that includes exons 5 and 6, and the splice sites are functionally weak in this region (26). For the purposes of the present study, it is notable that the protein corresponding to the sequence in Fig. 1A has functional properties indistinguishable from those of troponin T isolated from the heart (see below).

Construction of a Bovine Cardiac Troponin T Expression Vector in pET3d-- Because the pET expression system (27) was successful in expressing rat cardiac troponin T (8), the same system was used for the bovine protein. In constructing the wild type vector two changes were made at the 5' end of the troponin T cDNA by PCR mutagenesis. An AflII site was created at the translation initiation site, producing an end that could be ligated into the NcoI insertion site of pET3d without changing the amino acid sequence. Also, the second codon (Ser) was changed from TCG (infrequent in Escherichia coli) to TCT. The mutagenesis primers were 5'-AAGGGCCTGGGCTTGGGC-3' (internal primer) and 5'-ATAATAATAATCGATGAGGAGGACATGTCTGACGTGGAAGAGGCC-3' (initiating Met codon is in bold), which includes a ClaI site at the 5' end for ligation into the polylinker of pSP72. PCR was performed with Vent Polymerase (New England Biolabs) for 29 cycles. DNA sequencing confirmed no incidental changes were introduced by PCR. This wild type TnT sequence was then released by digestion with AflII and HindIII (Klenow) and ligated by standard techniques into the NcoI/BamHI (Klenow) site of pET3d. DE3(BL21) were transformed and single colonies grown overnight without subculturing or isopropyl-1-thio--D-galactopyranoside induction in 500-ml volumes of LB plus ampicillin. SDS-polyacrylamide gel electrophoresis showed that the darkest band in cell homogenates was not present in untransformed cells and comigrated with bovine cardiac troponin T isolated from the heart. The recombinant protein was purified by the same method used previously for rat troponin T (8). From a liter of cell culture the procedure yields 10-20 mg of purified recombinant troponin T.

Creation of Mutant Troponin Ts Corresponding to Mutations Found in FHC-- Using PCR-based protocols similar to the one described above, the wild type troponin T in pET3d was altered to produce six different mutations. The human mutations are (with the position in the bovine protein sequence (22) in parenthesis): R92Q (89), F110I (107), E160 (157), E163K (160), E244D (240), and intron 15 G₁  A (COOH truncation of 28 residues). In humans, this last mutation results in two truncated proteins, one missing 14 COOH-terminal residues and the other in which the 28 COOH-terminal residues are replaced by seven novel residues. The bovine cardiac troponin T mutants were expressed and purified as above. To facilitate comparison with other papers, in this manuscript the bovine mutations are referred to by the corresponding position in human troponin T.

Protein Purification-- All experiments were performed with proteins purified to homogeneity. Cardiac tropomyosin, whole troponin, and troponin subunits were obtained from bovine hearts (28, 29). Troponin was also obtained by reconstitution from subunits, i.e. by serial dialysis of 1:1:1 mixtures of troponin I, troponin C, and the various forms of troponin T (29). Labeling of various thin filament components was performed as described previously: tropomyosin with ³H on Cys¹⁹⁰ (30), troponin T with ³H on Cys³⁹ (31), or TnC with IAANS on Cys⁸⁴ and Cys³⁵ (32, 33). Rabbit skeletal muscle actin (34) and myosin subfragment 1 (35) and heavy meromyosin (36) were obtained by standard techniques. Skeletal muscle myosin was used instead of bovine cardiac myosin because of its greater stability, higher ATPase rate, and many similar interactions with the regulated thin filament: (i) effects of Ca²⁺ on ATPase V_max, ATPase actin K_app, and true myosin S1-ATP affinity for the thin filament (28, 37, 38); (ii) ATPase rate linearity with myosin S1 concentration (37, 39, 40); (iii) cooperative ATPase activation by the free Ca²⁺ concentration (32, 37, 39); and (iv) cooperative ATPase activation by bound Ca²⁺ (40).

Assays-- Published methods were used to measure the thin filament-myosin S1 MgATPase rate (41), the binding of troponin-tropomyosin to actin (30), and fluorescent titration of calcium binding to the IAANS-labeled TnC regulatory site (site II) with excitation at 332 nm and emission at 510 nm (32). The speed of thin filament movement over a heavy meromyosin-coated surface was measured as previously (36), with the following modifications. To diminish troponin-tropomyosin-induced bundling of filaments, rhodamine-phalloidin actin was allowed to bind to the HMM-coated surface in the absence of ATP, washed twice with buffer, and then incubated in the presence of 0.5 µM troponin-tropomyosin for 10 min at room temperature. This solution was then replaced with a solution containing 0.1 µM troponin-tropomyosin, ATP, oxygen-scavenging enzymes, and either pCa5 or pCa9.

The effects of the mutations on the affinity of troponin for actin-tropomyosin were determined by modification of a competitive binding assay (42). F-actin samples contained identical, saturating amounts of unlabeled tropomyosin and control ³H-labeled cardiac troponin and variable amounts of competing mutant troponin. After centrifugation, displacement of the ³H-labeled troponin from the thin filament was measured by liquid scintillation counting of the supernatants. Control troponin binding is tight (K > 10⁸ M¹) under these conditions (43), which makes this competitive approach necessary. The ratio (K_R) of the affinities of competitor troponin:radiolabeled control troponin for actin-tropomyosin was measured by fitting the data to a simple competition model as in Ref. 44.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Complete Coding Sequence of Bovine Cardiac Troponin T-- The bovine heart is a convenient source for cardiac troponin, because this tissue can be used to purify large amounts of troponin and/or troponin subunits. To procure mutant troponin T subunits that could be combined with TnI and TnC to create nonchimeric troponin complexes, the bovine cardiac troponin T cDNA was isolated. Fig. 1A shows 993 base pairs of bovine cardiac troponin T cDNA, including the entire coding region and partial 5' and 3' noncoding region sequences of 51 and 87 base pairs, respectively. The encoded amino acid sequence was consistent with the predominant adult bovine cardiac troponin T isoform, as previously determined by protein sequencing (22).

View larger version (46K): [in this window] [in a new window]   Fig. 1.   Mapping of human cardiomyopathy-causing mutations to sequence position and functional domains in bovine cardiac troponin T. A, troponin T was cloned from a bovine heart cDNA library, as described under "Materials and Methods." Both the cDNA and the encoded amino acid sequences are shown. The GenBankTM accession number is AF175558. The data are consistent with prior, protein sequencing data (22). B, location of identified FHC-causing troponin T mutations. The arrows indicate the positions of the mutations introduced into the bovine troponin T sequence, with the corresponding human mutations in parentheses. Many but not all of the mutations are in the portion of troponin T that forms the extended tail domain of troponin (19, 44). Serial NH2-terminal truncations of this region result in troponin complexes that are progressively shorter, with the Stokes radius decreasing linearly with the number of residues deleted (44). Within the tail region, the hypervariable portion contains no known mutations, and the mutation in the functionally critical region 95-119 (44) is a conservative Phe to Ile change. Similarly, the region that cross-links to the other troponin subunits is spared (as far as is known) except for a conservative Glu to Asp mutation. Two other FHC-implicated sites (human R278T (1) and A104V (58) mutations) have not been examined for functional defects and are not shown in the figure.

The sites of FHC mutations investigated in the present work are indicated in Fig. 1B. They are distributed widely within the amino acid sequence and in every case are in residues conserved between human and bovine sequences. Between positions 67 and 284 of the bovine sequence, which includes all the identified FHC sites, there are seven relatively conservative differences from the human sequence and 97% identity. Note that many of the troponin T sites implicated in FHC, including the Arg⁹² mutational "hot spot" (45) are within conserved portions of the troponin tail. The troponin tail plays a key role in anchoring troponin-tropomyosin onto actin, both in the Ca²⁺ conformation of the thin filament and in the inhibitory conformation found in the absence of Ca²⁺ (19, 30, 44, 46). Residues 1-184 correspond to rabbit fast skeletal muscle troponin T fragment T1, which binds tightly to tropomyosin (via residues 98-184) and part of which (residues 1-97) spans the tropomyosin-tropomyosin overlap joint (46-50).

Properties of Troponin(ex15-16)-- Troponin T interacts with tropomyosin, TnI, TnC, and probably actin and has a critical role in anchoring the regulatory complex onto the thin filament (19). Therefore, alterations in the binding of troponin to the thin filament are potential consequences of the troponin T mutations found in FHC kindreds. To test this, we assembled thin filaments containing actin, tropomyosin, and radiolabeled control troponin and then displaced the troponin by addition of wild type or mutant troponin complexes. Fig. 2A shows a representative experiment for troponin(ex15-16). Troponin(ex15-16) lacks troponin T COOH-terminal residues 257-284 but nevertheless is able to displace troponin from the thin filament. The mutant troponin is not as good a competitor as control unlabeled troponin, whether Ca²⁺ is present (Fig. 2A,  versus ) or Ca²⁺ is absent (Fig. 2A,  versus ). Interestingly, troponin(ex15-16) is a better competitor (more displacement) in the presence () than in the absence () of Ca²⁺. (This aspect is also shown in Fig. 4). A critical feature in interpreting this is that Ca²⁺ weakens the binding of the normal, labeled troponin to the thin filament approximately 2-fold (31, 42, 43, 46). The data in Fig. 2A therefore indicate that Ca²⁺ increases the effectiveness of troponin(ex15-16) as a binding competitor, because Ca²⁺ has less of an effect on the mutant troponin than on the radiolabeled (normal) troponin. More precisely, K_R (the ratio of the affinities of the two troponins) is 2-fold greater in the presence of Ca²⁺, increasing from 0.22 to 0.43. This 2-fold effect is the same magnitude as the previously established weakening effect of Ca²⁺ on the binding of the wild type labeled troponin, so the data imply that Ca²⁺ has no effect on the binding of troponin(ex15-16) to the thin filament.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Altered troponin and troponin-tropomyosin binding to actin using troponin(ex15-16). A, competitive binding experiment in which wild type troponin or troponin(ex15-16) displaces radiolabeled troponin from actin-tropomyosin. Open symbols, displacement measured in the presence of EGTA. Filled symbols, displacement measured in the presence of 104 M Ca2+. Squares, control troponin; triangles, troponin(ex15-16). Solid lines are best fit curves relating the data to Equation 1 in Ref. 44 and indicate KR, the affinity of the competing troponin relative to the affinity of the troponin being displaced. For control troponin KR = 1.12 ± 0.2 and 1.03 ± 0.11 in the absence and presence of Ca2+, respectively. For troponin(ex15-16) the corresponding KR values are 0.22 ± 0.03 and 0.43 ± 0.02. The conditions were as follows: 25 °C, 6 µM F-actin, 3 µM unlabeled tropomyosin, 1 µM bovine cardiac troponin that had been 3H-labeled on TnT Cys39, 10 mM Tris-HCl (pH 7.5), 300 mM KCl, 3 mM MgCl2, 0.2 mM dithiothreitol, 0.3 mg/ml bovine serum albumin, and either 0.5 mM EGTA or 0.1 mM CaCl2. B, troponin-induced binding of tropomyosin to actin. One function of troponin is to increase the affinity of tropomyosin for actin. This effect is easily observable in the presence of high ionic strength (300 mM KCl), under which condition tropomyosin binds very poorly to actin unless troponin is present. Troponin(ex15-16) () did not promote this process as well as did control reconstituted troponin () or as well as native bovine troponin (). The conditions were as follows: 25 °C, 10 µM rabbit skeletal muscle actin, 0.5 µM 3H-bovine cardiac tropomyosin, 10 mM Tris-HCl (pH 7.5), 300 mM KCl, 3 mM MgCl2, 0.2 mM dithiothreitol, 0.3 mg/ml bovine serum albumin, and 0.5 mM EGTA.

Another aspect of this process, not tested in Fig. 2A, is whether the mutation alters the binding of the troponin-tropomyosin complex to actin. In Fig. 2B, this process is examined by addition of troponin to samples containing actin and tropomyosin, under conditions in which the tropomyosin does not bind to actin unless troponin is added. The figure shows that the troponin-tropomyosin complex binds more weakly to actin when the troponin(ex15-16) is used () than when control reconstituted troponin is used () or when native bovine troponin is used (). The figure is representative of three experiments, in each of which the mutant troponin produced at least 20% less binding of tropomyosin to actin.

Thin filaments containing troponin(ex15-16) exhibited abnormal Ca²⁺-sensitive regulation of myosin function, in selected respects. In vitro motility was examined, in which heavy meromyosin was made to adhere to a microscope slide and the unloaded sliding of rhodamine-phalloidin containing thin filaments was assessed by epifluorescence microscopy (36, 51). As was also true for control thin filaments, sliding of thin filaments containing troponin(ex15-16) was Ca²⁺-dependent; with virtually no movement in the absence of Ca²⁺ and >85% of filaments moving consistently in the presence of Ca²⁺. The speed of this unloaded sliding has a Gaussian distribution, with width approximately 20% of the mean speed (36). When using the same preparation of actin and myosin, measurement of many filaments allows the mean speed to be determined with high precision, with 2% or better standard error calculable from repeated mean speed determinations (36). Table I presents the mean speed of filaments containing various troponin T mutations, in each case relative to paired mean speed determinations using control troponin. Unloaded sliding of thin filaments containing troponin(ex15-16) was 7% faster than sliding of control thin filaments in the presence of 50 mM ionic strength and 16% faster than control in the presence of 100 mM ionic strength (this last value is not shown in Table I). For comparison, in vitro sliding speed of thin filaments containing the human Ile⁷⁹  Asn mutation is as much as 50% faster than control filaments in the presence of 100 mM ionic strength (8). (In recent experiments this increase is approximately 20%).²

A different aspect of Ca²⁺-sensitive regulation of myosin activity is examined in solution MgATPase assays. The MgATPase rate and the sliding rate are limited by different parts of the cross-bridge cycle and often are affected differently by changes in contractile protein structure, including troponin T mutations (reviewed in Ref. 8). When control troponin was added to solutions of myosin S1, actin, and tropomyosin (Fig. 3), the MgATPase rate increased when Ca²⁺ was present () and decreased when Ca²⁺ was absent (). However, when troponin(ex15-16) was added there was significantly impaired regulation. When Ca²⁺ was present () the MgATPase rate was only slightly higher than when it was absent (), resulting in a level of activity that was only one-third the level for control thin filaments activated by Ca²⁺ (). This indicates that the troponin T mutation alters myosin function. The figure shows one of three experiments, which used slightly different troponin concentrations but gave similar results; in each case the MgATPase rate fell when troponin(ex15-16) was added in the presence of Ca²⁺.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Troponin(ex15-16) causes impaired Ca2+-sensitive regulation of the myosin S1-thin filament MgATPase rate. When troponin is added to actin-tropomyosin thin filaments, the myosin S1 MgATPase rate increases when Ca2+ is present () and decreases when Ca2+ is absent (). However, when troponin(ex15-16) is added the MgATPase rate is only slightly higher when Ca2+ is present () than when absent (). The conditions were as follows: 25 °C, 7 µM F-actin, 2 µM tropomyosin, 0.3 µM myosin S1, 1 mM ATP, 20 mM imidazole (pH 7.1), 3.5 mM MgCl2, 7 mM KCl, 1 mM dithiothreitol, 0.5 mM dibromoBAPTA. Filled symbols indicate measurements with 0.6 mM CaCl2 also present.

Properties of Troponin Mutants Containing Troponin T Mutations R92Q or F110I-- As indicated under "Materials and Methods," each of the mutant troponin Ts was mixed with TnC and TnI to form a troponin complex. After a series of dialysis steps that gradually remove denaturant and 1 M KCl, complex formation was confirmed by a final step of gel filtration chromatography (29, 44). For the R92Q and F110I mutants, this procedure failed on repeated occasions because the proteins precipitated during the dialysis protocol. More than 50% of the protein became insoluble, including TnI and TnC as well as the mutant TnT. This suggests that the TnT mutants did not fold normally under these in vitro conditions. For troponin(R92Q), a soluble troponin complex at 1 mg/ml concentration finally was obtained by leaving 1 M KCl in the dialysis buffer and also in the gel filtration buffer. For troponin(F110I) 1 M KCl did not prevent precipitation of a 1 mg/ml protein solution, and it was necessary also to include 20% glycerol in the dialysis and gel filtration buffers. The weak effect of troponin(F110I) in competition experiments (Fig. 4), therefore may merely reflect aggregation under the conditions of the experiment (0.3 M KCl and no glycerol). It is unclear whether similar behavior would occur for the corresponding mutation within the human cardiocyte, where the abnormality might be mitigated by cellular conditions and/or by human versus bovine subtle differences in protein sequence.

The poor solubility of these mutant troponins precluded their examination in motility or myosin S1 MgATPase assays. However, high ionic strength improved the solubilty, and troponin binding to the thin filament is best measured under high salt conditions (30). Therefore, troponin(R92Q) and troponin(F110I) were studied as in Fig. 2A to determine the effects of these mutations on the affinity of troponin for the thin filament. The results are shown in Fig. 4, which also includes data for several other troponin T mutations. Troponin(R92Q) had the same affinity for actin-tropomyosin as did wild type tropomyosin (i.e. K_R = 1) both in the presence and in the absence of Ca²⁺. Despite the presumed alteration in folding, once the troponin was folded it was fully functional in this respect. In contrast, troponin(F110I) bound very weakly to the thin filament, regardless of the Ca²⁺ concentration. It is unclear whether this merely indicates poor binding because of aggregation or is because of an authentically low affinity of monodisperse troponin(F110I) for actin-tropomyosin.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Effect of troponin T mutations on the affinity of troponin for actin-tropomyosin. Data were obtained from experiments conducted as described in the legend to Fig. 2A and are expressed relative to the affinity of wild type labeled troponin for actin-tropomyosin measured under the same conditions, either the presence (filled bars) or the absence (open bars) of calcium. The affinities of troponins containing the F110I, E244D, and ex15-16 mutations are significantly different from control (p < 0.01).

Properties of Troponin(E160)-- Troponin(E160) bound normally to the thin filament in competitive binding assays (Fig. 4), and the troponin(E160)-tropomyosin complex also bound normally (data not shown). Like troponin(I79N) (8), troponin(E160) produced normal thin filament assembly yet altered myosin function. However, the pattern of this alteration differed from that of either troponin(I79N) or troponin(ex15-16). Unlike either of the other mutations, E160 altered the Ca²⁺ concentration dependence of the thin filament-myosin S1 MgATPase rate (Fig. 5A). Interestingly, activation required slightly lower concentrations of Ca²⁺, implying higher apparent Ca²⁺ affinity for the regulatory site of TnC when this troponin T point deletion was present () than when it was absent (). The effect was small but statistically significant based upon paired and repeated titrations: K_app = 6.2 ± 0.6 × 10⁵ M¹ for the mutant troponin and K_app = 4.0 ± 0.5 × 10⁵ M¹ for wild type troponin. Like the I79N mutation but unlike the ex15,16 mutation, the E160 mutation had no effect on the myosin S1 MgATPase rate in the presence of saturating Ca²⁺ or in the absence of Ca²⁺ (data not shown). Unlike the other two mutants, troponin(E160) decreased rather than increased the in vitro unloaded sliding speed. Measurements of in vitro motility showed a slightly slower speed in the presence of this mutation, an 8% decrease (Table I). Unloaded muscle shortening reportedly is unchanged by this mutation (13), but with error estimates too large to view this as a conflict with the present in vitro motility data.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Thin filaments containing troponin(E160) have increased Ca2+ affinity. A, representative experiment showing that lower Ca2+ concentrations are required to activate the myosin S1-thin filament MgATPase rate when the troponin contains the TnT E160 mutation () than when it contains wild type TnT (). The maximum activated rate in the presence of saturating Ca2+ concentrations was not significantly affected by the mutation. (Similar results were obtained on this point by varying the troponin concentration as described in the legend to Fig. 3; data not shown.) Conditions are as described in the legend to Fig. 3, except the troponin or troponin(E160) concentration was constant at 1 µM, and the concentration of CaCl2 was varied to control the free Ca2+ concentration (32). B, representative experiment showing that thin filaments containing troponin(E160) have an increased affinity for Ca2+ at the TnC regulatory site II. Ca2+ binding was measured by labeling the TnC with IAANS prior to reconstitution with TnI and the mutant or wild type TnT. Normalized data are shown. Addition of Ca2+ caused a 60% increase in fluorescence intensity for control thin filaments, and similarly, a 56% increase in intensity when the mutant troponin was present. The conditions were as follows: 25 °C, 5 µM F-actin, 2 µM tropomyosin, 0.6 µM troponin or troponin(E160), 5 mM imidazole (pH 7.1), 3.5 mM MgCl2, 7.5 mM KCl, 1 mM dithiothreitol, 0.5 mM dibromoBAPTA.

To determine the mechanism for the shift in Ca²⁺ K_app seen in the MgATPase assays, troponin was reconstituted with IAANS-modified TnC. This fluorophore permits spectroscopic monitoring of Ca²⁺ binding to the TnC regulatory site on reconstituted thin filaments (32). Fig. 5B shows a representative result. There is an increased Ca²⁺ affinity on thin filaments containing troponin(E160) (). The Ca²⁺ affinity for troponin(E160) thin filaments was 1.4 ± 0.1 × 10⁵ M¹, compared with a value of 0.99 ± 0.09 × 10⁵ M¹ for control thin filaments. Both of these values are tighter than in the MgATPase assays (Fig. 4A), because the IAANS labeling of TnC causes a 2-3-fold increase in regulatory site Ca²⁺ affinity (32). However, there is concurrence between the two experiments on the effect of the troponin T E160 mutation, a 40- 50% increase in Ca²⁺ affinity.

Properties of Troponin(E163I) and Troponin(E244D)-- The troponin T mutation in troponin(E244D) is a conservative change, was identified in just one FHC patient, and was not confirmed by genomic analysis of that kindred (1). Nevertheless this mutation is located in a region of troponin T that is highly conserved and that has been implicated in interactions with the other troponin subunits. In support of the conclusion that this mutation caused the cardiomyopathy, Fig. 4 demonstrates that it caused abnormal troponin function, i.e. impaired troponin binding to the thin filament. In MgATPase assays, it resulted in normal regulation, with Ca²⁺ K_app = 5 ± 1 × 10⁵ M¹. In motility assays it caused a small, 5% increase in sliding speed (Table I).

FHC troponin T mutation E163K (1) occurs at the same position as mutation E160 from the protein perspective; residues 160-163 are all glutamates so point deletions at any of these glutamates are equivalent. We have not examined the charge reversal E to K mutation, but we did examine a neutralization of this charge, creating troponin E163I. MgATPase titrations employing thin filaments with troponin(E163I) were similar to those employing troponin(E160), with K_app = 6.0 ± 1.5 × 10⁵ M¹. The increased K_app for troponin(E163I) further supports the validity of the small change in Ca²⁺ affinity that was described above for troponin(E160). Motility assays using troponin(E163I) showed no effect on the unloaded sliding speed (Table I).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The shared clinical features of FHC, whether caused by mutations in troponin or in other components of the sarcomere (myosin, C-protein, or tropomyosin), indicate at least partial convergence toward common mechanisms. However, there are obvious functional differences among the several mutated molecules, somewhat different disease patterns are distinguishable depending upon the affected gene (1, 52, 53), and even within one gene such as troponin T the mutations occur in more than one functional region. This suggests that the pathophysiological mechanisms must be understood for several of the mutations and that significant variation may be present. The present report provides experimental support for this variation. For example, Fig. 4 summarizes the effects of various troponin T mutations on troponin binding to the thin filament. It can be seen that the affinities of the mutants vary, and some are decreased (K_R < 1). However, almost all of the mutant troponins bind tightly to the thin filament, differing no more than severalfold from the control troponin (i.e. K_R  0.2). Note that these are affinities relative to that of control troponin, shown by previous work to bind to actin-tropomyosin with an affinity that is very tight: 1-5 × 10⁸ M¹ (43, 46, 54). This suggests that the mutant troponins could be incorporated into the thin filaments of the cell, with specific binding despite mutations in the anchoring subunit, troponin T. Two of the mutants (troponin(R92Q) and troponin(E160)) have no detectable change in thin filament affinity (K_R = 1), similar to previous findings for troponin(I79N) (8). In contrast, troponin(E244D), troponin(ex15-16), and possibly (see above) troponin(F110I)) have weaker than normal binding (K_R < 1). This may be relevant for patients heterozygous for these mutations (and also for transgenic models), providing a mechanism tending to promote greater thin filament incorporation of the normal allele than the mutant allele.

Two of the mutants in the present study (R92Q and F110I) exhibited poor solubility implying abnormal protein folding. It is interesting that these two sites are in or immediately proximate to a region of TnT proposed to form a critical portion of the troponin tail (Fig. 1B) (44). The insolubility also may be related to the findings of Sweeney, Watkins, and co-workers (1, 13), who found that overexpression of mutant but not wild type human cardiac troponin T in quail myotubes caused abnormal sarcomere assembly with disruptions or aggregates within the sarcomeric structure. To varying degrees depending upon the mutation, they reported these abnormalities for each of examined mutations: I79N, R92Q, 160E, and the truncation of exons15 and 16. In transgenic mice expressing low levels of R92Q, 15% of the myocardium exhibited disarray (16), which is similar to the quail in vitro result but could be a consequence of abnormal contraction rather than of abnormal folding. Abnormal folding of the troponin T mutants found in FHC may be a general feature contributing in some way to the pathophysiological pattern. However, both the prior and present work suggest that folding is not so disrupted as to prevent substantial incorporation into the sarcomere. The troponin binding data (Fig. 4) show that the mutant troponins bind tightly enough to actin-tropomyosin to displace control troponin, when added in modest excess.

Once the abnormal troponin Ts are incorporated into the thin filaments, they cause several functional abnormalities, with details variable depending upon the mutation. Table II presents a summary of several reports. The current study indicates that mutation 160E causes enhanced apparent Ca²⁺ affinity in MgATPase assays, which is shown to be attributable to an increase in Ca²⁺ affinity at the TnC regulatory site. This is the first demonstration that an FHC-causing troponin T mutation alters Ca²⁺ affinity per se, a basic mechanistic insight. In the sarcomere, the apparent Ca²⁺ affinity also depends upon thin filament-thick filament interactions. As assessed in tension versus pCa experiments the 160E mutation slightly increases the apparent Ca²⁺ affinity (13). This was stated to be a small effect and was not quantified but nevertheless parallels our data with the purified troponin. Notably, some FHC-causing mutations in tropomyosin (14) and C-protein (15) also enhance Ca²⁺ sensitivity, and a preliminary report describes the same phenomenon for FHC troponin I mutations (55). This therefore appears to be a recurring phenomenon observed in a significant subset of FHC patients. However, it is not always present. For example, troponin T mutants I79N and R92Q weaken Ca²⁺ sensitivity in tension versus pCa data after overexpression in quail myotubes (13) or in rat cardiocytes (56), and the I79N mutation has no effect on Ca²⁺ affinity in solution experiments where cooperative effects of cross-bridges are minimal (8). A report from Morimoto and co-workers (9) that conflicts with these data indicates that exchange of the I79N and R92Q mutants into cardiac fibers causes increased Ca²⁺ sensitivity. If true, this suggests that increased Ca²⁺ sensitivity is typical for most of the troponin T mutations occurring in FHC. However, a qualification is that the cardiac fiber exchange procedure itself decreased force by 40%. A preliminary report describing mutant TnT exchange into porcine cardiac fibers supports the conclusion that several of the TnT mutants cause an increased Ca²⁺ sensitivity (57).

View this table: [in this window] [in a new window]   Table II Functional defects due to FHC-causing mutations in troponin T Footnotes a-d and g refer to observations using homogeneous mutant TnT, and footnotes e, f, and h-j refer to observations using mixtures of wild type and mutant TnT. ND, not determined. , increased; , decreased.

Several of the troponin T mutations cause decreased force in the presence of saturating Ca²⁺ concentrations (13). The I79N mutation exhibits decreased force as well as increased speed, suggesting that the mutation accelerates cross-bridge detachment from actin (8). The ex15-16 has also been shown to decrease force, and we now demonstrate that it has the same additional feature of increased sliding speed. However, this is not a consistent pattern for all of the mutations, because R92Q does not decrease force but does increase shortening speed, and E160 decreases force and either decreases speed or has no effect (Table II). In solution MgATPase assays, which are not limited by the same kinetic step as either sliding speed or maximum force (see Ref. 8), the ex15-16 mutation causes a decreased rate, and no other mutation has a detectable effect.

Table II shows that effects of FHC troponin T mutations include several aspects of increased function (higher sliding or shortening speed; higher Ca²⁺ affinity). On the other hand, decreased force results from three of the four troponin T mutations tested for this parameter. Decreased force, increased unloaded shortening or sliding, altered Ca²⁺ sensitivity, and impairment of folding/thin filament binding/sarcomere structure are each observed for several mutations. The relative contributions of these various factors for each mutation or for the troponin T mutations generally may become clearer with further investigation. However, a consensus from Table II is that many investigative groups find results that differ depending upon the specific mutation. The table also suggests a problem: that understanding of the mutations is complicated by disparity in experimental approaches and apparently conflicting results. A partial explanation for these disparities may be that the mutations alter sarcomeric integrity and troponin incorporation into the sarcomere, as well as having more direct effects. Biochemical data using purified proteins may be particularly helpful in this respect, because thin filament assembly can be measured and controlled with high sensitivity and because the use of purified proteins excludes considerations of intracellular and extracellular disarray. Regardless of future insights, the already evident effects of the FHC-causing mutations indicate that troponin T has a complex role in the modulation of cross-bridge function and in the Ca²⁺-sensitive control of contraction.

	ACKNOWLEDGEMENT

We thank Dr. J. Lin of the University of Iowa for the gift of rat cardiac troponin T cDNA.

	FOOTNOTES

^* This work was supported by American Heart Association Grant-in-Aid 9550128N and National Institutes of Health Grant NHLBI-38834 (to L. T.) and National Institutes of Health Grant AR-30988 (to E. H.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed: Dept. of Internal Medicine, University of Iowa, 200 Hawkins Dr., SE610-GH, Iowa City, IA 52242. Tel.: 319-356-3703; Fax: 319-356-3086; E-mail: larry- tobacman{at}uiowa.edu.

² D. Lee and E. Homsher, unpublished data.

	ABBREVIATIONS

The abbreviations used are: FHC, familial hypertrophic cardiomyopathy; PCR, polymerase chain reaction; TnT, troponin T; IAANS, 2-(4'-iodoacetamidoanilino)naphthalene-6-sulfonic acid.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES