Alterations of Tension-dependent ATP Utilization in a Transgenic Rat Model of Hypertrophic Cardiomyopathy*

Although it is established that familial hypertrophic cardiomyopathy (FHC) is caused by mutations in several sarcomeric proteins, including cardiac troponin T (TnT), its pathogenesis is still not completely understood. Previously, we established a transgenic rat model of FHC expressing a human TnT molecule with a truncation mutation (DEL-TnT). This study investigated whether contractile dysfunction and electrical vulnerability observed in DEL-TnT rats might be due to alterations of intracellular Ca2+ homeostasis, myofibrillar Ca2+ sensitivity, and/or myofibrillar ATP utilization. Simultaneous measurements of the force of contraction and intracellular Ca2+ transients were performed in right ventricular trabeculae of DEL-TnT hearts at 0.25 and 1.0 Hz. Rats expressing wild-type human TnT as well as nontransgenic rats served as controls. In addition, calcium-dependent ATPase activity and tension development were investigated in skinned cardiac muscle fibers. Force of contraction was significantly decreased in DEL-TnT compared with nontransgenic rats and TnT. Time parameters of Ca2+ transients were unchanged at 0.25 Hz but prolonged at 1.0 Hz in DEL-TnT. The amplitude of the fura-2 transient was similar in all groups investigated, whereas diastolic and systolic fura-2 ratios were found elevated in rats expressing nontruncated human troponin T. In DEL-TnT rats, myofibrillar Ca2+-dependent tension development as well as Ca2+ sensitivity of tension were significantly decreased, whereas tension-dependent ATP consumption (“tension cost”) was markedly increased. Thus, a C-terminal truncation of the cardiac TnT molecule impairs the force-generating capacity of the cycling cross-bridges resulting in increased tension-dependent ATP utilization. Taken together, our data support the hypothesis of energy compromise as a contributing factor in the pathogenesis of FHC.

Although it is established that familial hypertrophic cardiomyopathy (FHC) is caused by mutations in several sarcomeric proteins, including cardiac troponin T (TnT), its pathogenesis is still not completely understood. Previously, we established a transgenic rat model of FHC expressing a human TnT molecule with a truncation mutation (DEL-TnT). This study investigated whether contractile dysfunction and electrical vulnerability observed in DEL-TnT rats might be due to alterations of intracellular Ca 2؉ homeostasis, myofibrillar Ca 2؉ sensitivity, and/or myofibrillar ATP utilization. Simultaneous measurements of the force of contraction and intracellular Ca 2؉ transients were performed in right ventricular trabeculae of DEL-TnT hearts at 0.25 and 1.0 Hz. Rats expressing wild-type human TnT as well as nontransgenic rats served as controls. In addition, calcium-dependent ATPase activity and tension development were investigated in skinned cardiac muscle fibers. Force of contraction was significantly decreased in DEL-TnT compared with nontransgenic rats and TnT. Time parameters of Ca 2؉ transients were unchanged at 0.25 Hz but prolonged at 1.0 Hz in DEL-TnT. The amplitude of the fura-2 transient was similar in all groups investigated, whereas diastolic and systolic fura-2 ratios were found elevated in rats expressing nontruncated human troponin T. In DEL-TnT rats, myofibrillar Ca 2؉ -dependent tension development as well as Ca 2؉ sensitivity of tension were significantly decreased, whereas tension-dependent ATP consumption ("tension cost") was markedly increased. Thus, a C-terminal truncation of the cardiac TnT molecule impairs the force-generating capacity of the cycling cross-bridges resulting in increased tension-dependent ATP utilization. Taken together, our data support the hypothesis of energy compromise as a contributing factor in the pathogenesis of FHC.
Familial hypertrophic cardiomyopathy (FHC) 3 is an autosomal-dominant inherited disease characterized by ventricular hypertrophy, arrhythmias, and sudden death. The phenotype of affected individuals may vary from early sudden death, marked hypertrophy with contractile dysfunction, to an asymptomatic carrier status (reviewed in Refs. [1][2][3]. In the past 10 years, several genes could be linked to FHC, almost uniformly encoding contractile proteins such as ␤-myosin heavy chain and cardiac troponin T (4). FHC has thus been labeled a "disease of the sarcomere" (5). Mutations of the cardiac troponin T (cTnT) gene are of particular interest because they are associated with a high incidence of arrhythmias and sudden death even in the absence of significant myocardial hypertrophy (6). Yet the pathogenesis of FHC is still poorly understood, and there is no clear understanding how these mutations lead to the development of FHC with diastolic dysfunction and sudden cardiac death. Biochemical and biophysical analyses have failed to identify a common mechanism underlying the alterations of cardiac contractility resulting from these mutations. For example, missense mutations in ␤-myosin heavy chain have been shown to both depress (7) as well as enhance (8) contractile function, whereas mutations in ␣-tropomyosin increase calcium sensitivity and force development of the sarcomere (9). Likewise, it is still unclear how FHC-associated cTnT mutations trigger the initiation of FHC (10). The exact molecular function of troponin T is also still a matter of debate; however, it is believed that cTnT stabilizes the troponin complex, consisting of troponin T, I, and C. In addition, cTnT may affect and potentially regulate the Ca 2ϩ sensitivity of myofibrillar ATPase activity, the level of ATPase activation, and/or force development (11)(12)(13)(14)(15).
Recently, it has been observed that FHC patients with cardiac troponin T, ␤-myosin heavy chain, and myosin-binding protein C mutations display a markedly altered phosphocreatine to ATP ratio, irrespective of the presence of myocardial hypertrophy (16). This finding led to the new hypothesis of myocardial energy depletion being a critical factor in the pathogenesis of FHC (17).
To get a closer understanding of the pathophysiology underlying FHC, we have previously (18) generated a transgenic rat model of the disease by overexpressing a C-terminal cTnT truncation (DEL-TnT), resulting from an intron 15 splice donor site mutation observed in FHC patients. Wild-type rats and rats overexpressing nonmutated human TnT served as controls. In a working heart model, DEL-TnT-transgenic rats exhibited significant systolic and diastolic dysfunction in the absence of cardiac hypertrophy. Of note, relatively small amounts of DEL-TnT (Ϸ5% of endogenous TnT) were sufficient to induce this phenotype, similar to observations in other transgenic models of hypertrophic cardiomyopathy (19,20). In contrast, transgenic rats overexpressing a nonmutated human TnT molecule (Ϸ30% of endogenous TnT) displayed improved contractile performance. Moreover, after exercise training, myocardial disarray and ventricular arrhythmias, such as ventricular tachycardia and ventricular fibrillation, were observed in DEL-TnT-transgenic rats (18). Immunofluorescence analyses of transgenic cardiomyocytes demonstrated that the mutant TnT is incorporated into the sarcomere, suggesting a dominant-negative mode of action rather than haploinsufficiency.
To further investigate the mechanisms underlying contractile dysfunction in DEL-TnT transgenic rats and specifically whether alterations of Ca 2ϩ homeostasis contribute to the phenotype, we now performed simultaneous measurements of force and intracellular Ca 2ϩ transients in isolated electrically stimulated papillary muscle strips. Furthermore, to test whether ATP utilization might be impaired in DEL-TnT transgenic rats, calcium-dependent tension development and myofibrillar ATPase activity were measured in Triton X-100 skinned cardiac fiber preparations. Rats overexpressing the normal, i.e. nontruncated, human TnT molecule as well as nontransgenic controls were studied for comparison.
The current findings confirm and extend our previous data that a C-terminal truncation of the cardiac TnT molecule impairs the force-generating capacity of the cycling crossbridges. Moreover, this effect is accompanied by increased tension-dependent ATP utilization in DEL-TnT rats, thus supporting the hypothesis of energy compromise as a contributing factor in the pathogenesis of FHC.

EXPERIMENTAL PROCEDURES
Animals-Transgenic rats overexpressing the predominant adult isoform of human cTnT (288 amino acids) were generated as described previously (18). Cardiac skinned fiber preparations as well as right ventricular papillary muscles were investigated from transgenic rats overexpressing human wild-type cTnT (TnT-rats, n ϭ 10) or a truncated human cTnT molecule resulting from an intron 15 splice donor site mutation, lacking the most C-terminal 14 amino acids (DEL-TnT rats, n ϭ 10), in comparison to nontransgenic controls (NT, Sprague-Dawley rats, n ϭ 10). The age of the rats ranged from 14 to 16 months. The animals were killed by cervical dislocation, and hearts were rapidly removed. Control and transgenic animals did not differ in their body weight (NT, 414 Ϯ 42 g; DEL-TnT, 427 Ϯ 33 g; hTnT, 415 Ϯ 22 g) or their heart wet weight (NT, 1.6 Ϯ 0.1 g; DEL-TnT, 1.9 Ϯ 0.1 g; hTnT, 1.9 Ϯ 0.1 g). The investigation conformed with institutional guidelines as well as the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH Publication Number 85- 23, revised 1996).
Simultaneous Measurements of Force and Intracellular Ca 2ϩ Transients-The intracellular Ca 2ϩ transient was measured at room temperature in isolated, electrically driven right ventricular papillary muscles by the fluorescence indicator fura-2 (21). To facilitate cell loading fura-2 was used as acetoxymethyl (AM) ester. These AM esters passively cross the plasma membrane and, once inside the cell, are cleaved to cell-impermeant products by intracellular esterases. After an initial control measurement of the force of contraction (FOC), the muscle strips were incubated for 4 h in darkness to avoid photobleaching of the dye in an oxygenated (95% O 2 , 5% CO 2 ) Ringer's solution containing 5 mol/liter of fura 2-AM. The external calcium concentration was 1.8 mmol/liter. The experimental setup was purchased from Scientific Instruments, Heidelberg, Germany, and has been described previously (22). Because the right papillary muscles are more suitable for fura-2 measurements because of their smaller diameter, Ca 2ϩ transients were studied in these preparations. The cross-sectional areas of papillary muscle preparations were comparable in all experimental groups as follows: NT rats, 0.58 Ϯ 0.05 mm 2 ; TnT rats, 0.45 Ϯ 0.03 mm 2 ; DEL-TnT rats, 0.58 Ϯ 0.06 mm 2 . Left ventricular myocardium was used for skinned fiber experiments (see below).
Preparation of Triton X-100 Skinned Fibers-Skinning of the heart muscle fibers was performed as described previously (23). Briefly, the fiber bundles (diameter Ͻ0.2 mm) were dissected from left ventricular papillary muscles and permeabilized at 4°C for 20 h in a solution containing 50% (v/v) glycerol, 1% Triton X, and in mmol/liter NaN 3 10, ATP 5, MgCl 2 5, EGTA 4, 1,4-dithioerythritol 2, and imidazole 20 (pH 7.0). Subsequently, the fibers were stored in the same buffer without Triton X-100 at Ϫ20°C. The experiments were performed within 5 days.
Force and ATPase Activity Measurements-Triton X-100 skinned fiber bundles were prepared under the microscope and then mounted isometrically and connected to a force transducer (Scientific Instruments, Heidelberg, Germany). Mean sarcomere length was 1.9 Ϯ 0.03 m. ATPase activity was simultaneously measured with tension development using a linked NADH fluorescence assay (0.6 mM NADH, 140 units of lactate dehydrogenase) as described previously (23-26) (experimental setup, Scientific Instruments, Heidelberg, Germany). Relaxation solution contained 20 mM imidazole, 10 mM ATP, 5 mM NaN 3 , 5 mM EGTA, 12.5 mM MgCl 2 , and 0.2 mM Ap 5 A. The contraction solution contained calcium EGTA (5 mM) instead of EGTA. The ATP concentration was stabilized with an ATP-regenerating system, including phosphoenolpyruvate (12.5 mM) and pyruvate kinase (100 units/ ml). Both solutions were mixed with a gradient mixer so that the Ca 2ϩ concentration was incrementally increased every 15 s. Fibers were fixed in a "slack" position in relaxation solution (for composition see above), and fiber length was adjusted to an extent where resting tension was just threshold. Measurement of developed tension and myofibrillar ATPase activity started 3 s after the solution was exchanged when a stable plateau was reached. At the end of the exper-iment, the NADH decline was determined by perfusion with a defined solution not containing NADH (in mmol/liter: imidazole 20, NaN 3 5, EGTA 5, MgCl 2 12.5, phosphoenolpyruvate 5, calcium 5) as well as with the use of a the same solution plus 600 mol/liter NADH. These solutions did not contain additional calcium. Calibration of the signal was performed under conditions of a continuous perfusion of the respective calibration solutions until a stable signal had been reached. Free Ca 2ϩ concentration was determined by calculator programs designed for experiments in skinned muscle cells (27). Experiments were performed at 25°C. The ratio of ATPase activity and force was assumed as a measure for "tension cost." Table 1 summarizes the parameters measured to characterize myofibrillar function.

MATERIALS AND METHODS
Salts used were high analytical grade, purchased from Merck. All other chemicals were of analytical grade or the best grade commercially available.
Statistics-All values are means Ϯ S.E. unless otherwise noted. Student's t test or paired t test were used to test for significance. p values of Ͻ0.05 were accepted as significant. pCa force as well as pCa myofibrillar ATPase activity relationships were fitted by a modified Hill equation (28) where Y is the fractional force, or actomyosin-ATPase activity, pCa 50 is the Ca 2ϩ concentration giving half-maximal activation (inhibition), and H is an index of cooperativity (Hill coefficient). The pCa 50 for tension development or myofibrillar ATPase activity, all Hill coefficients, and the tension cost were analyzed by Graph Pad Prism (San Diego).

Frequency-dependent Alterations of Force and Ca 2ϩ Transients in Right Ventricular Muscle Strips of TnT Transgenic
Rats-It has been proposed that the impaired cardiac relaxation observed in working heart models of transgenic mice expressing a truncated TnT molecule might involve altered intracellular Ca 2ϩ homeostasis (20). To investigate whether dysregulation of intracellular Ca 2ϩ homeostasis is present in DEL-TnT transgenic rats, simultaneous measurements of the intracellular Ca 2ϩ transient and force of contraction were performed at 0.25 and 1.0 Hz. Table 2 and Fig. 1 summarize the results obtained for the frequency-dependent changes of the isometric force of contraction as well as intracellular Ca 2ϩ transients.
Force-At a stimulation frequency of 0.25 Hz, FOC as well as contraction (ϩdP/dt) and relaxation (ϪdP/dt) velocities were markedly decreased in rats expressing the truncated human cardiac TnT molecule (DEL-TnT) and both compared with TnT transgenic and NT rats (Fig. 1A). Conversely, FOC, ϩdP/dt and ϪdP/dt values were significantly increased in rats expressing the nonmutated human cardiac TnT molecule (TnT) compared with nontransgenic controls. The same pattern was observed at the higher stimulation frequency of 1.0 Hz. Again, DEL-TnT rats displayed impaired contractile function, whereas TnT-rats showed "supranormal" contractile behavior ( Table 2 and Fig. 1A). In all three experimental groups, a comparable frequency-dependent decline in force of contraction  Ca 2ϩ Transients-At 0.25 Hz, fura-2 amplitudes were similar in all three experimental groups. Also, diastolic and systolic fura-2 ratios were not significantly different in DEL-TnT papillary muscles compared with nontransgenic controls, whereas diastolic and systolic fura-2 ratios were found elevated in rats expressing the nonmutated human TnT (Table 2). An elevation of the stimulation frequency to 1 Hz was accompanied by a comparable increase in diastolic fura-2 ratio in all three groups ( Table 2). Systolic and diastolic fura-2 ratios were again significantly increased in TnT rats compared with NT rats. At higher stimulation frequencies, the amplitude of the intracellular Ca 2ϩ transient declined to a similar degree in all three experimental groups investigated.
Time to peak Ca 2ϩ transient and time to half-peak Ca 2ϩ transient decays were not significantly different in DEL-TnT, TnT, and NT rats at 0.25 Hz. However, at a higher stimulation frequency (1.0 Hz), both time to peak Ca 2ϩ transients as well as time to half-peak Ca 2ϩ transient decays were prolonged in the DEL-TnT group compared with TnT rats (Fig. 1).
Tension Development and Myofibrillar ATPase Activity in DEL-TnT Transgenic Rat Hearts-To examine the possible influence of the cardiac troponin T truncation mutation (DEL-TnT) on tension-dependent ATP utilization in the transgenic model, simultaneous measurements of tension and myofibrillar ATPase activity were performed in skinned fibers at increasing extracellular Ca 2ϩ concentrations. Fig. 2 shows representative original tracings of the experiments. Fig. 3 and Table 3 summarize the results.
In DEL-TnT rats, maximal Ca 2ϩ -activated tension and Ca 2ϩ sensitivity of tension development were significantly decreased compared with rats expressing the nontruncated human TnT molecule ( Table 3). The steepness (n Hill ) of the Ca 2ϩ tension relationship was increased to a similar degree in DEL-TnT and TnT compared with NT fibers ( Table 3). The expression of a truncated TnT molecule in transgenic rat hearts did not significantly alter basal and maximal myofilament ATPase activity nor the Ca 2ϩ -dependent increase in actomyosin-ATPase activity compared with nontransgenic controls. However, Ca 2ϩ sensitivity of the myofibrillar ATPase activity was significantly shifted to the right in DEL-TnT compared with NT rats (p Ͻ 0.05; Table 3).
Tension Development and Myofibrillar ATPase Activity in TnT Transgenic Rat Hearts-To investigate whether myofibrillar function might also be altered by expressing the nontruncated human TnT molecule in rat myocardium, simultaneous measurements of tension and myofibrillar ATPase activity were performed in skinned fibers from these animals. Fig. 2 shows representative original tracings of the experiments. Fig. 3 and Table 3 summarize the results. In TnT rats, maximal Ca 2ϩactivated tension was significantly increased compared with both DEL-TnT and NT fibers, whereas Ca 2ϩ sensitivity of tension development was not different (Table 3). Moreover, basal and maximal ATPase activity as well as the Hill coefficient of the Ca 2ϩ /tension relation, which characterizes the cooperativity of the myofilaments, were significantly increased compared with NT rats (Fig. 3 and Table 3). These results imply that the

. Frequency-dependent alterations in isometric force of contraction (A) and the intracellular Ca 2؉ transient (B-D) measured in right ventricular papillary muscle strips of TnT transgenic rats.
A, isometric force of contraction. In all three groups a frequency-dependent decline in force was observed. At low (0.25 Hz) and elevated stimulation frequency (1 Hz), contractility was decreased in rats expressing the human TnT molecule (DEL-TnT, n ϭ 7) as compared with rats expressing the wild-type human TnT molecule (TnT, n ϭ 9) and nontransgenic animals (NT, n ϭ 10). B, amplitude of the fura-2 transient. The amplitude of the Ca 2ϩ transient was similar in all three groups at low and elevated stimulation frequencies. At increased stimulation frequencies the fura-2 amplitude declined. C, time to peak Ca 2ϩ transient. At elevated stimulation frequencies, time to peak Ca 2ϩ transient was significantly prolonged in transgenic rats expressing the truncated human TnT molecule (DEL-TnT) versus rats expressing nontruncated human TnT. D, time to half-peak Ca 2ϩ decay. At elevated stimulation frequencies, time to half-peak Ca 2ϩ decay was significantly prolonged in DEL-TnT transgenic rats compared with rats expressing nontruncated human TnT. ϩ, p Ͻ 0.05 versus 0.25 Hz; *, p Ͻ 0.05 versus NT; #, p Ͻ 0.05 versus TnT.
incorporation of a full-length human TnT molecule into the myofilaments may increase the actin-myosin interaction and thus increase ATP turnover. In contrast, myofibrillar Ca 2ϩ sensitivity of tension was not significantly influenced by the expression of a human TnT molecule in rat cardiac tissue.
Tension Cost in TnT and TnT-DEL Rats-Tension-dependent ATP utilization, a parameter reflecting myofibrillar economy, was analyzed by the ratio of myofibrillar ATPase activity over tension (tension cost). Although TnT transgenic rats showed a tension cost ratio (0.71) comparable with nontransgenic rats (0.74), the slope of tensiondependent ATP utilization was significantly decreased in the DEL-TnT group (0.31; p Ͻ 0.05 (Table 3 and Fig. 4)). Myofibrillar ATP consumption was thus markedly increased in rats expressing a truncated TnT molecule compared with controls.

DISCUSSION
Although multiple mutations in sarcomeric genes have been found in patients with FHC, the pathogenesis of the disease remains poorly understood. We have previously established a transgenic rat model for FHC that displays typical FHC features, such as diastolic dysfunction and a propensity toward arrhythmias (18). To gain additional insight into the pathophysiology of this animal model, this present study investigated myofibrillar tension development, intracellular Ca 2ϩ homeostasis, and actomyosin ATPase activity in hearts of DEL-TnT transgenic rats. Our findings demonstrate that a C-terminal truncation of the cardiac TnT molecule markedly impairs the forcegenerating capacity of the cycling cross-bridges. In addition, calcium transients were altered in transgenic rats expressing mutant TnT but not in animals carrying a nonmutated human TnT transgene. Finally, sig-

FIGURE 2. Original recordings of the simultaneous measurements of myofibrillar ATPase activity (top) and developed tension (bottom) of ventricular skinned fiber preparations of nontransgenic (NT) rats, rats expressing a human TnT molecule with a truncation mutation (DEL-TnT), and rats expressing the nonmutated human TnT molecule (TnT).
Ca 2ϩ -dependent tension and myofibrillar ATPase activity were found to be significantly impaired in DEL-TnT rats.

TABLE 3 Tension development, myofibrillar ATPase activity, and tension cost in TnT transgenic rats
The abbreviations used are as follows: NT, nontransgenic rats; DEL-TnT, rats overexpressing the truncated TnT molecule; TnT, rats overexpressing nonmutated human TnT; ␣, slope of the myofibrillar ATPase/tension relationship; pCa 50 , Ca 2ϩ concentration at which a 50% increase of tension or myofibrillar ATPase activity was achieved. *, p Ͻ 0.05 versus NT; #, p Ͻ 0.05 versus TnT.

Parameter
NT (n ‫؍‬ 13) DEL-TnT (n ‫؍‬ 9) TnT (n ‫؍‬ 6) nificantly increased tension-dependent ATP utilization was observed in DEL-TnT rats, suggesting that compromised cardiac energy homeostasis contributes to the pathogenesis of FHC.

Reduced Myofibrillar Tension Development because of a Truncation of the Human TnT
Molecule-In isolated working heart preparations from transgenic rats, we have shown previously that DEL-TnT hearts display contractile dysfunction, whereas hearts expressing nonmutated human TnT actually reveal enhanced contractility compared with wild-type controls. We could now corroborate and extend these findings by measuring FOC in isolated papillary muscle strips as well as tension development in skinned fiber preparations. Again, DEL-TnT resulted in significant depression of FOC, whereas a nonmutated human TnT improved FOC and tension development, respectively. Although it is now widely accepted that a dominant-negative mechanism accounts for the deleterious effects of sarcomeric protein mutations in FHC (3,4), it is still unclear precisely how these mutations modulate myofilament function and eventually lead to the clinical phenotype of hypertrophy, contractile dysfunction, and lethal arrhythmias.
The TnT molecule is critical for binding of the regulatory proteins troponin C and troponin I to tropomyosin on the thin filament (29 -31). The association of TnT and tropomyosin facilitates tropomyosin assembly on the actin filament (32) and thus the cooperative activation of muscle contraction (33,34). Given that the TnT C terminus also binds troponin C and I, it seems possible that alterations of this region because of a mutation could directly influence myofibrillar cooperativity. Moreover, it has recently been shown that PKC-dependent phosphorylation of TnT at Thr 195 , Thr 204 , Thr 285 (35), and Thr 206 (36), all located at the C terminus of cardiac TnT, results in a decrease in maximum tension in murine fiber preparations. It has been suggested that the degree of TnT phosphorylation affects myofilament activation by modulating the interaction with troponin I (15), which may account for the observed functional impairment. This study provides evidence that a truncation of the TnTmoleculemodulatesbothCa 2ϩdependent tension and myofibrillar ATPase activity, consistent with the notion that conformational changes of the C terminus can affect TnT function. These alterations could contribute to the contractile dysfunction described previously in isolated working heart preparations of DEL-TnT rats (18). The critical role of the TnT C terminus in this regard is further supported by a recent study demonstrating that not only PKC-but also Rho-dependent cTnT phosphorylation at Ser 278 and Thr 287 are able to regulate tension and ATPase activity (37).
In contrast, transgenic rats overexpressing the wild-type nontruncated human TnT molecule display unchanged Ca 2ϩ sensitivity of myofibrillar tension development yet a significant increase in maximal Ca 2ϩ -activated tension compared with controls. These findings imply that functional differences must exist between rat and human cTnT. Interestingly, the same principal finding, i.e. improved performance of human wild-type TnT, has also been observed in a transgenic mouse model of FHC (38), further suggesting distinct properties of human versus rodent TnT. Although human and rat TnT display 89% of amino acid identity, the N termini show significant sequence variations. Of note, this region is also critical in the regulation of cross-bridge kinetics and calcium sensitivity (39), possibly accounting for the observed species-specific functions. In line with this, our current results indicate that expression of a wild-type human TnT molecule in rat myocardium may alter the co-operativity between thick and thin filaments resulting in improved Ca 2ϩ -dependent tension development without altering the tension-dependent energy demand (see below). However, we cannot exclude that other molecular adaptations such as myosin isoform switching (40) or altered expression of calcium handling proteins (see below) contribute to the differential effects of the transgenes on force of contraction as well.
Intracellular Ca 2ϩ Homeostasis in TnT Transgenic Rats-From functional studies in murine models of TnT-associated FHC, it has been proposed that diastolic dysfunction observed in these animals might involve a dysregulation of intracellular Ca 2ϩ homeostasis (29,41). This study provides evidence that at least at higher stimulation frequencies the regulation of intracellular Ca 2ϩ transients is in fact altered in rats expressing the truncated human TnT molecule. At a stimulation frequency of 1 Hz, both time to peak Ca 2ϩ transient and time to peak Ca 2ϩ decay were found to be prolonged (Fig. 1). Intracellular Ca 2ϩ transients are subject to tight control and depend in part on the activity of SERCA2a as well as the Na ϩ /Ca 2ϩ exchanger (42,43). In this regard, it has been shown recently that isolated In DEL-TnT rats tension-dependent ATP utilization was significantly decreased (i.e. tension cost was increased), whereas TnT transgenic rat hearts are comparable with nontransgenic rats. cardiomyocytes from transgenic mice expressing an I79N TnT mutation also display a slowed Ca 2ϩ transient decay. This was associated with differential activation of the Ca 2ϩ -dependent Na ϩ /Ca 2ϩ exchanger and the increased occurrence of ventricular arrhythmias (41). In line with this, ventricular arrhythmias have also been observed in our DEL-TnT rat model (18), raising the possibility that Na ϩ /Ca 2ϩ exchanger activity might also be altered in transgenic rats expressing a truncated human TnT molecule.
In contrast, in rats expressing a nontruncated human TnT molecule, we did not observe a prolongation of calcium transients, yet diastolic and systolic fura-2 ratios were found to be significantly elevated. This finding may be due to a complex adaptation to altered myofilament calcium sensitivity in TnT transgenic animals.
Alterations in calcium cycling proteins have been observed in other transgenic models of hypertrophic cardiomyopathy. Semsarian et al. (44) found that mice carrying an ␣-myosin heavy chain mutation (R403Q) display a marked down-regulation of the ryanodine receptor, calsequestrin, as well as triadin, whereas SERCA2a levels were unchanged. Remarkably, these changes could be reversed upon treatment with the calcium channel inhibitor diltiazem. It thus remains to be seen if dysregulated expression of calcium handling proteins also contributes to the phenotype of DEL-TnT rats.
Myofilament ATP Utilization in TnT Transgenic Rats-FHC is considered to be a disease of the sarcomere (5), as multiple different mutant alleles have been described in at least nine genes encoding cardiac contractile proteins (for review see Ref. 4). Yet it has been difficult to devise a "unifying hypothesis" of its pathogenesis, because FHC-associated mutations have vastly variable (or even opposing) effects, i.e. on sarcomeric calcium sensitivity, force development, and ATPase activity. Therefore, it has been proposed recently that a common feature shared by many (if not all) different classes of FHC-causing mutations might be inefficient sarcomeric ATP utilization (17,39,45). The high risk of sudden death in FHC patients with TnT mutations could thus be attributed to alterations in cardiac energy consumption resulting from an increase in the energy cost of force production (39). This concept is also supported by the fact that inherited disorders of mitochondrial function, i.e. CD36 deficiency (46) or Friedreich ataxia (47), result in a phenocopy of hypertrophic cardiomyopathy. Moreover, carriers of FHC-associated mutations that do not (yet) display cardiac hypertrophy already reveal clearly abnormal phosphocreatine/ ATP ratios (16), suggesting that altered energy metabolism is not merely secondary to hypertrophic growth of the myocardium but rather is an early feature in disease progression. This study provides additional evidence that dysregulation of myofibrillar energy utilization could contribute to the functional alterations associated with FHC. The tension cost, reflecting the ratio of myofibrillar ATPase activity and tension was markedly altered in DEL-TnT transgenic rats, such that myofibrillar energy consumption for a given tension was significantly increased (Fig. 4). Although enhanced and inefficient ATP consumption in DEL-TnT mutant hearts might be negligible under basal conditions, it may become limiting in situations of increased cardiac work load, such as physical exercise, and car-diac arrhythmias or even sudden cardiac death might ensue. Findings in other transgenic models of FHC further support this notion. Javadpour et al. (48) reported that in (R92Q)-TnT transgenic mouse cardiac ATP utilization is increased as well and resulted in the impaired ability of the heart to recruit its contractile reserve. Similarly, Montgomery et al. (49) reported an increase in tension cost (and unchanged Ca 2ϩ -activated maximal ATPase activity) in two independent transgenic mouse models expressing TnT mutations. Moreover, Stelzer et al. (50) investigated a mouse model with overexpression of C-terminally truncated TnT and observed enhanced thin filament activation at submaximal calcium concentrations. However, in contrast to our work, an increase in calcium-activated force was found, which may be due to the different truncation mutation analyzed in this study.
Finally, in transgenic rat hearts expressing nonmutated human TnT, we observed a significantly increased force production without an increase in tension cost, suggesting a supranormal or even potentially beneficial effect on contractility. However, we believe that these data need to be interpreted with caution, because it is not known what the effects of human TnT were at the physiological rat heart rate of 500 -600/min. Yet it is conceivable that subtle differences in TnT function must exist between humans and rodents in order to adapt the sarcomere to an about 10-fold difference in heart rate.
Limitations of the Study-Although our findings support previous in vivo data of the DEL-TnT transgenic rat model (18), the current measurements have been performed in vitro and thus cannot necessarily be generalized. In addition, calcium transients and force measurements in right ventricular papillary muscle were performed under conditions not identical to the physiological situation (stimulation frequency of 0.25-1 Hz, temperature of 25°C) in order to obtain stable and reproducible signals. However, it appears unlikely that these limitations introduced systematic errors, because the conclusion of the study largely relies on the relative differences between transgenic rats and controls.

CONCLUSION
In conclusion, this study provides evidence that overexpression of a cardiac TnT molecule lacking the 14 C-terminal amino acids impairs myofibrillar tension development and increases tension-dependent myofibrillar ATPase activity. Taken together, our data support the hypothesis of energy compromise as a contributing factor in the pathogenesis of FHC.