Altered hemodynamics in transgenic mice harboring mutant tropomyosin linked to hypertrophic cardiomyopathy

MATERIALS AND METHODS

Transgenic animals.

Transgenic mice (FVBN strain) expressing α-Tm(Asp175Asn) were produced as previously described by Muthuchamy et al. (25). Expression of the transgene was driven by the murine α-myosin heavy chain promoter, which restricts expression to the heart. All in situ and in vitro measurements were performed on a transgenic line of mice that expressed a high ratio of mutant α-Tm(Asp175Asn) to wild-type α-Tm protein. In previously published work we showed that there is 63% replacement of wild-type α-Tm with mutant α-Tm (Asp175Asn) in myofilaments from the hearts of these TG mice (25). We found no signs of gross pathology and no difference in heart weight, heart-to-body weight ratios, or longevity in the TG mice compared with NTG mice (data not shown). The mean age of NTG mice was 21.5 ± 1.3 wk, and the mean age of TG mice was 22.2 ± 1.5 wk. The mean body weight of NTG mice was 28.1 ± 0.99 g, and the mean body weight of TG mice was 30.1 ± 1.10 g.

In situ measurements.

In situ measurements were performed similarly to a method described by Lorenz and Robbins (16). Experiments were conducted on TG mice and their NTG litter mates. Male and female mice (25 and 40 g/wt) were allowed free access to food and water up to the time of the experiment. A total of 10 NTG (7 male and 3 female) and 10 TG mice (8 male and 2 female) were used. Mice were anesthetized by intraperitoneal injection of 50 μg/g body weight of ketamine and 100 μg/g body weight of thiobutabarbital sodium (Inactin, Research Biochemicals International, Natick, MA). The level of anesthesia was assessed by a toe pinch. When the mice required additional anesthesia, they were given one-fifth of the original intraperitoneal dose of both anesthetics. The mice were placed supine on a thermally controlled warming plate where their body temperature was maintained at 37°C. A tracheotomy was performed, and a short segment of polyethylene (PE)-120 tubing was inserted into the airway and secured with suture. The right carotid artery was isolated, and the distal end was tied off. The artery was canulated with a 1.4-Fr Mikro-Tip transducer (model SPR-671, Millar, Houston, TX). With the use of the continuous pressure display as a guide, the transducer was advanced retrogradely down the right carotid artery, into the aorta, through the aortic valve, and into the LV (Fig. 1 A). When a stable LV pressure waveform was noted, the transducer cable was secured in place with a suture. Before each catheterization, the transducer was calibrated in warm saline in a sealed chamber at 20 and 200 mmHg, as recommended by the manufacturer. At the end of each experiment, the catheter was immediately rezeroed in warm saline to check for drift. To gain venous access, the right femoral vein was isolated, the distal end tied off, and the proximal end catheterized with stretched PE-10 tubing. After a short length of tubing was advanced into the femoral vein and secured in place, the free end was connected to a 100-μl syringe mounted on a PHD2000 microinfusion/withdrawal pump (Harvard Apparatus, Holliston, MA). All of the surgical incisions were covered with saline-soaked gauze to minimize evaporation. The mice were allowed to stabilize after their surgery for 30 min before the experimental protocol began.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1.

Representative samples of a pressure trace during in situ measurements. A: pressure measurements made at relatively slow paper speed as the Millar catheter was advanced down the right carotid artery and into the left ventricle (LV).B–D: pressure measurements made at relatively fast paper speed during control conditions (B), after 0.08 (C), and after 0.32 ng isoproterenol (Iso) · g body wt⁻¹ · min⁻¹(D) were infused. HR, heart rate.

To analyze the myocardial response to β-adrenergic stimulation and blockade, we infused increasing concentrations of isoproterenol (Iso) (0.02, 0.04, 0.08, 0.16, and 0.32 ng Iso · g body wt⁻¹ · min⁻¹), given over 3 min at 0.1 μl/g body weight via the femoral venous catheter. The infusion vehicle consisted of 0.9% saline with 10 U/ml heparin added to prevent clotting of the venous line. Mice were allowed to recover to baseline for 10–15 min between doses. After the highest dose of Iso, the mice were again allowed to recover to baseline for 10–15 min and were then given a single bolus dose of propranolol (Pro) (100 ng/g body weight) via the femoral venous catheter. The raw data signal was amplified on the internal amplifier in a WindowGraf Chart Recorder (Gould Instrument Systems, Valley View, OH), recorded at 2,000 Hz, and analyzed using the Left Ventricular Pressure Module of the Po-ne-mah Digital Acquisition Analysis and Archive System software (Gould Instrument Systems) on a personal computer. This program performs calculation of heart rate (HR), LV end-diastolic pressure (LVEDP), LV systolic pressure (LVSP), developed pressure (DP), maximal rate of pressure increase over time (dP/dt _max), minimal rate of pressure decrease over time (dP/dt _min), and 50% time of relaxation (RT_1/2). Raw data recordings were replayed, and calculation of LV pressure parameters at baseline (control), after administration of Iso, and after administration of Pro were made at the end of the infusion period when pressure was observed to have reached a steady state by averaging 10 consecutive beats. Because the absolute change in pressure is known to influence dP/dt, we also normalized dP/dt _max and dP/dt _min by dividing by the developed pressure (DP) (dP/dt _max/DP and dP/dt _min/DP). To confirm that the dose of Pro completely inhibited the prior treatment with Iso, we infused an additional dose of Iso after the Pro infusion in a subset of the mice. In all of the cases, we found no indication of β-adrenergic stimulation. In addition, to examine the impact of infusing a volume of 0.1 μl/g body weight over 3 min on cardiac function, we infused the vehicle alone and found no effect.

SR vesicle Ca²⁺ uptake.

SR Ca²⁺ uptake was determined using a modified method from Solaro and Briggs (30) and Luo et al. (18). The hearts were dissected from mice anesthetized with 50 mg/kg body weight pentobarbital sodium and immediately placed in ice-cold saline and trimmed free of atria. The hearts were then transferred to homogenizing buffer (HB) (2 ml/100 mg wet heart), chopped into small pieces with scissors, and homogenized. The HB contained (in mmol/l) 50 KH₂PO₄, 10 NaF, 1.0 EDTA, 300 sucrose, 0.3 phenylmethylsulfonyl fluoride (PMSF), and 0.5 dithiothreitol (DTT), pH 7.0.

Ca²⁺ uptake was measured over a range of pCa values (pCa 8 to 5) and, in addition to the corresponding Ca²⁺concentration, the reaction mixture contained the following (in mmol/l): 5 MgATP²⁻, 0.5 free Mg²⁺, 40 imidazole, 10 creatine phosphate, 0.5 EGTA, 5 potassium oxalate, 5 sodium azide, 10 procaine, and 0.03 ruthenium red. The ionic strength was adjusted to 175 mM with KCl, and pH was adjusted to 7.1 with KOH. Ruthenium red and procaine were added to inhibit Ca²⁺release from the SR, whereas sodium azide was added to inhibit Ca²⁺ uptake into mitochondria (18, 30, 37). After 2 min of preincubation, the reaction was started by adding ventricular homogenate to the reaction mixture at a concentration of 0.20–0.25 mg protein/ml and proceeded at 37°C for 2 min with constant stirring. The protein concentration was determined by using the method described by Lowry et al. (17). The reaction was stopped by filtration through a 0.45-μm Millipore filter that was washed with ice-cold buffer containing 20 mM Tris and 2 mM EGTA, pH 7.0. The total SR vesicle Ca²⁺ uptake was calculated from the amount of ⁴⁵Ca bound to filters, as determined by liquid scintillation spectroscopy.

β-Adrenergic receptor density measurements.

A modification of the method designed by Hohl et al. (11) was used to measure binding of [¹²⁵I]iodocyanopindolol ([¹²⁵I]ICYP) (Du Pont-New England Nuclear), a high-affinity ligand for the β-adrenoceptors, in ventricular homogenates. Ventricular homogenates were prepared as previously described but with the use of an assay buffer (AB) containing (in mmol/l) 1.0 EDTA , 100 Tris, and 5.0 MgCl₂, pH 7.2. The homogenate was centrifuged at 3,000g for 10 min, the supernatant was discarded, and the pellet was resuspended in AB. The protein concentration was determined by using the mothod of Lowry et al. (17). Ventricular preparations (33 μg) were incubated with increasing concentrations of [¹²⁵I]ICYP (5–300 pM) at 37°C in a final volume of 300 μl for 80 min. The assay was carried out in duplicate with one set of tubes containing 5 μM Pro. At the end of 80 min, the reaction was stopped by addition of ice-cold wash buffer (WB) (10 mM Tris, 1 mM EDTA, pH 7.5) and filtration through glass fiber filters by using a Brandel cell harvester. The filters were washed twice with WB and counted in a Beckman 9000 gamma counter. To determine nonspecific binding, [¹²⁵I]ICYP binding in the presence of Pro was subtracted from total, and the dissociation constant (K _D) and maximal binding (B_max) were determined by fitting the data to a single binding site-hyperbolic curve by using GraphPad software (San Diego, CA).

Ca²⁺-activated MgATPase activity of myofibrils.

A modification of the method of Pagani and Solaro (27) was used to prepare cardiac myofibrils. Hearts dissected from mice anesthetized with 50 mg/kg body weight pentobarbital sodium were immediately placed in ice-cold homogenizing solution (HS), pH 7.0, of the following composition (in mmol/l): 25.0 MOPS, 60.0 KCl, and 2.5 MgCl₂. Three or four mouse hearts were weighed and pooled for each myofibrillar preparation. The hearts were trimmed free of connective tissue and then chopped into small pieces with dissecting scissors. The hearts were homogenized at a ratio of 1 ml HS to 100 mg wet heart in an Omni mixer (DuPont, Newtown, CT) fitted with a mini-micro generator attachment (IKA Homogenizers) for 4 min and then centrifuged at 3,000 g. The pellet was resuspended in HS with 1% Triton X-100 and homogenized again for 3 min. Three more cycles of homogenization and centrifugation in HS with 1% Triton X-100 were performed. After the Triton X-100 treatment, the pellet was homogenized in HS for 3 min containing 1.0 mM EGTA and centrifuged. The pellet was then resuspended in HS (without EGTA), homogenized, and centrifuged as previously described for two more cycles to remove all EGTA. Protein concentration was determined by using the method described by Lowry et al. (17). Rates of ATP hydrolysis of myofibrils were determined according to the method described by Pagani and Solaro (27), but scaled down to one-fourth of the original volume because of the small amount of protein obtained from the myofibrillar preparation. The reaction proceeded at 30°C for 10 min in a solution of the following composition (in mmol/l): 2 free Mg²⁺, 20 MOPS, 79.5 KCl, 5 MgATP²⁻, and 1.0 EGTA, and pCa values ranged from 8.0 to 4.875 at pH 7.0. Trichloroacetic acid was added to stop the reaction, and inorganic phosphate released was determined as described by Carter and Karl (6).

Ca²⁺-sensitive force measurements in skinned fiber bundles.

Measurements of the pCa-force relation, before and after PKA-dependent phosphorylation, were performed on fiber bundles as previously described by Palmiter et al. (28). Adult mice were anesthetized as described above, and hearts were quickly removed and put into cold high-relaxing solution of the following composition (in mmol/l): 53 KCl, 10 EGTA, 20 MOPS, 1 free Mg²⁺, 5 MgATP²⁻, 12 creatine phosphate, and 10 U/ml creatine phosphokinase at pH 7.0. The ionic strength of all solutions was 150 mmol/l. Papillary muscles from the LV were dissected, and small fiber bundles ∼150 μm in width and 4–5 mm long were prepared. Fiber bundles were mounted between a micromanipulator and a force transducer with cellulose-acetate glue, and the membranes were extracted in the high-relaxing solution containing 1% Triton X-100 for 30 min. A resting sarcomere length (SL) of 2.0 μm was established from laser diffraction patterns. Isometric tension was recorded on a chart recorder as fibers were first maximally contracted in solution of pCa 4.5 and then relaxed in high-relaxing solution, followed by sequential solutions of decreasing pCa values (pCa range from 8.0 to 4.5). After the first series of contractions, fiber bundles were relaxed in the solution and subjected to PKA-dependent phosphorylation by incubation in a phosphorylating buffer with the same composition as the high-relaxing solution but with 100 μM cAMP and 10 μg/ml PKA from bovine heart (Sigma Chemical, St. Louis, MO) for 30 min. In a subset of fibers, 100 ng/ml of calyculin A, a phosphatase inhibitor, was added to prevent dephosphorylation, but no greater shift in Ca²⁺-sensitivity was observed. After the incubation period, the fibers were again contracted in sequential solutions of decreasing pCa values (pCa range from 8.0 to 4.5) but containing cAMP and PKA at the same concentrations as the phosphorylating buffer. All solutions also contained the protease inhibitors pepstatin A (2.5 μg/ml), leupeptin (1 μg/ml), and PMSF (50 μM).

In a separate series of experiments, the maximum Ca²⁺-activated isometric tension (at pCa 4.5) was measured in fiber bundles at two different SL (2.0 and 2.3 μm). The SL was set at 2.0 μm as described above, and the fiber bundles were maximally contracted two to three times. SL was checked after each contraction, and, after it was determined that SL did not change, the fiber was maximally contracted again. After this last contraction, the fiber bundle was relaxed in HR and measured in two perpendicular planes (0° and 90°) and at three points (the center and two distal ends) using a mirror and a reticle fitted to the eyepiece of a movable microscope. The mean diameters from all three points were used to calculate the mean cross-sectional area. After the first measurement, SL was reset at 2.3 μm, and the fiber bundle was again contracted maximally as described above.

Measurement of strong cross-bridge activated force.

Strong cross-bridge force activation was measured in two ways, on the basis of a previously published method (28). Triton X-100 extracted fiber bundles were prepared and mounted on the force transducer as described above. In the first series of experiments, isometric tension was recorded as fiber bundles were contracted in solutions of sequentially increasing pMgATP (−log [MgATP]) (3.0–8.0) at pCa 9.0. Data from the pMgATP-force relations were fitted to the Hill equation for determination of the pMgATP₅₀. In the second series of experiments, fiber bundles were contracted in a solution with a pMgATP of 5.0, close to the pMgATP₅₀ determined for the NTG fibers in the previous experiments, and in a solution with a pMgATP of 8.0, which caused maximal MgATP-activated force in both NTG and TG fibers. The ratio of force at pMgATP 5.0/8.0 was determined.

Data computation and statistical analysis.

The composition of all pMgATP and pCa solutions was calculated using the Bathe computer program with the binding constants for all ionic species as reported by Godt and Lindley (8). Calculations of the pMgATP₅₀ values for the pMgATP-force relation and the pCa₅₀ for pCa-force, ATPase, and Ca²⁺uptake relations were made with the use of Prizm software (GraphPad). Force and ATPase data were fitted to the Hill equation to obtain the pCa₅₀ and pMgATP₅₀. Calculations of the pCa₅₀ and maximal velocity (V _max) for Ca²⁺ uptake measurements were made by fitting the raw data to a sigmoidal curve of variable slope. All results are presented as means ± SE. Data from in situ pressure measurements and data from force measurements (before and after PKA) were compared using a two-way repeated measure ANOVA and a Tukey post test. Comparisons of data from ATPase activity, strong cross-bridge activated force, maximum force measurements, SR Ca²⁺ uptake measurements, and β-adrenergic receptor density measurements were made using the Student's t-test. A value of P < 0.05 was the criterion for significance in all experiments.

RESULTS

To evaluate the effect of the Asp175Asn point mutation of α-Tm on cardiac function in an intact animal, we measured LV pressure in situ during control conditions (before infusion of any drugs) and after the infusion of increasing doses of Iso in NTG and TG mice. Because the control state, in this experimental situation, represents a condition where endogenous catecholamines are present, we also administered Pro, a nonspecific β-blocker, to inhibit β-adrenergic stimulation. Figure 1 A shows a pressure tracing as the catheter was advanced into the LV at a relatively slow paper speed. Figure 1,B–D, shows LV pressure, at relatively fast paper speed, in the control phase (B), at a submaximal concentration of Iso (C), and at the maximum concentration of Iso (D). Table 1 summarizes the LVEDP, DP, HR, dP/dt _max, and dP/dt _min for NTG and TG mice at the basal experimental state (control), during β-adrenergic stimulation (increasing doses of Iso), and after β-adrenergic blockade (Pro). Figure 2 shows the effect of increasing doses of Iso and a single dose of Pro on the DP and HR in both groups of mice. Although there were no significant differences in LVEDP or HR between NTG and TG mice under any conditions, DP was significantly lower in TG mice compared with NTG mice at the highest doses of Iso (0.08–0.32 ng Iso · g body wt⁻¹ · min⁻¹). The DP also increased in the NTG mice during Iso infusion (0.16 ng Iso · g body wt⁻¹ · min⁻¹) compared with the control state, but there was no corresponding increase in DP in the TG mice.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2.

Effect of increasing doses of Iso and a bolus dose of propanolol (Pro) on developed pressure (DP) (A) and HR in nontransgenic (NTG) (B) and transgenic (TG) mice. †Significant difference (P < 0.05) within the NTG group from control; ‡significant difference (P < 0.05) within the TG group from control; *significant difference (P < 0.05) between NTG and TG groups. Data are expressed as means ± SE for 10 NTG and 10 TG preparations. bw, Body weight.

Figure 3 shows dP/dt _max (A), dP/dt _max/DP (B), dP/dt _min (C), and dP/dt _min/DP (D) in NTG and TG mice in the control state, during Iso infusion, and after Pro infusion. There were no significant differences between the groups in dP/dt _max or dP/dt _max/DP in either the control state or after infusion of Pro. However, during infusion of Iso there was a concentration-dependent increase observed in contractility in both groups, but neither dP/dt _max nor dP/dt _max/DP increased to the same extent in TG versus NTG mice (Fig. 3,A and B). Compared with NTG mice, cardiac relaxation, as measured by dP/dt _min and dP/dt _min/DP, was significantly lower in the control state at all doses of Iso and during β-adrenergic blockade in TG mice (Fig. 3, C and D). Moreover, there were significant increases in dP/dt _min and dP/dt _min/DP in the NTG mice during infusion of Iso (0.08 ng Iso · g body wt⁻¹ · min⁻¹) compared with levels in the control state but no corresponding increase in either relaxation parameter in the TG mice. The RT_1/2 is shown in Fig.4. Although the RT_1/2 in TG mice was not significantly different from NTG mice in the control state or during Pro infusion, there was a significant difference at the first two doses of Iso (0.02 and 0.04 ng Iso · g body wt⁻¹ · min⁻¹).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3.

The effect of increasing doses of Iso and a bolus dose of Pro on derived pressure parameters in NTG and TG mice. A: maximal rate of pressure rise (dP/dt _max). B: DP-normalized dP/dt _max(dP/dt _max/DP). C: maximal rate of pressure decline (dP/dt _min). D: DP-normalized dP/dt _min(dP/dt _min/DP). †Significant difference (P < 0.05) within the NTG group from control; ‡significant difference (P < 0.05) within the TG group from control; *significant difference (P < 0.05) between NTG and TG groups. Data are expressed as means ± SE for 10 NTG and 10 TG preparations.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 4.

The effect of increasing doses of Iso and a bolus dose of Pro on 50% time of maximal relaxation (RT_1/2) in NTG (■) and TG (○) mice. †Significant difference (P < 0.05) within the NTG group from control. ‡Significant difference (P < 0.05) within the TG group from control. *Significant difference (P< 0.05) between NTG and TG groups. Data are expressed as means ± SE for 10 NTG and 10 TG preparations.

The results shown in Figs. 3 and 4 demonstrate a reduced relaxation rate of TG hearts versus NTG hearts that is exacerbated by β-adrenergic stimulation; therefore, we sought to determine whether changes in SR Ca²⁺ uptake rate or an alteration in the β-adrenergic receptor density could, in part, be responsible for the altered relaxation in TG mice. Figure 5shows the result of oxalate-supported, Ca²⁺ uptake into SR vesicles prepared from NTG and TG mouse hearts. We could detect no difference in either the pCa₅₀ or theV _max for SR Ca²⁺ uptake in TG mice compared with NTG mice. Figure 6 shows the result of [¹²⁵I]ICYP binding in ventricular homogenates prepared from NTG and TG mouse hearts. Although there was no difference in β-adrenergic receptor density, a small but significant difference was found in the K _D in TG compared with NTG mouse hearts [K _D (NTG) = 65.7 ± 7.1 fmol/mg, (n = 6) andK _D (TG) = 42.3 ± 5.4 fmol/mg,n = 5]. This difference in receptor affinity, however, cannot explain the blunted response to β-adrenergic stimulation found in situ in TG mice, since a decrease in the K _Dwould be expected to enhance the response to Iso.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 5.

Ca²⁺ uptake into sarcoplasmic reticulum (SR) vesicles prepared from NTG and TG hearts measured as a function of pCa and corresponding maximal velocity (V _max) (inset). Data are expressed as means ± SE. pCa₅₀ (NTG) = 6.49 ± 0.04 (n = 7); pCa₅₀ (TG) = 6.54 ± 0.04 (n= 7). V _max (NTG) = 440.6 ± 16.6 nmol Ca²⁺ · mg⁻¹ · min⁻¹ (n = 7);V _max (TG) = 454.3 ± 18.9 nmol Ca²⁺ · mg⁻¹ · min⁻¹ (n = 7).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 6.

β-Adrenergic receptor binding of [¹²⁵I]iodopindolol ([¹²⁵I]ICYP) in ventricular homogenates from NTG and TG mouse hearts. Saturation isotherm data are expressed as the means ± SE. Dissociation constant (K _D) (NTG) = 65.7 ± 7.1 fmol/mg (n = 6), K _D (TG) = 42.3 ± 5.4* fmol/mg (n = 5). Maximal binding (B_max) (NTG) = 105.3 ± 4.41 fmol/mg (n = 6), and B _max (TG) = 102.7 ± 4.33 fmol/mg (n = 5). *Significant difference between NTG and TG groups (P < 0.05).

To determine the effect of the Asp175Asn mutation of α-Tm on myofilament MgATPase rate, we measured Ca²⁺-sensitive Mg-ATPase activity in myofilaments from NTG and TG mouse hearts. Figure7 shows the pCa-ATPase relationship and the corresponding pCa₅₀ values. There was a significant increase in the Ca²⁺ sensitivity of ATPase activity in myofilaments from TG hearts compared with NTG; pCa₅₀ (NTG) = 5.89 ± 0.06 (n = 10, from 5 groups, 3–4 hearts/group); pCa₅₀ (TG) = 6.11 ± 0.04 (n = 12, from 8 groups, 3–4 hearts/group). However, basal and maximal Ca²⁺-activated myofibrillar ATPase activities were similar, supporting previous findings that showed no shift in myosin heavy chain isoforms in the TG mouse hearts (25). In addition, a comparison of the Hill coefficients (n _H) of the pCa-ATPase relations between NTG and TG myofilaments showed no difference in cooperativity [n (NTG) = 2.33 ± 0.42;n (TG) = 2.10 ± 0.44].

• Download figure
• Open in new tab
• Download powerpoint

Fig. 7.

Ca²⁺-dependent MgATPase activity and corresponding pCa₅₀ values (inset) in myofibrils purified from NTG and TG mouse hearts. Data are expressed as means ± SE. pCa₅₀ (NTG) = 5.89 ± 0.06 (n = 10, from 5 groups, 3–4 hearts/group); pCa₅₀ (TG) = 6.10 ± 0.04 (n = 12, from 8 groups, 3–4 hearts/group). Hill coefficient (n _H) (NTG) = 2.33 ± 0.42; Hill coefficient n _H (TG) = 2.10 ± 0.44.

To analyze whether the altered physiological response to β-adrenergic stimulation observed in the TG mice in situ was due to an alteration at the myofilament level, we simulated the in vivo effect of β-adrenergic stimulation on the myofilaments by PKA-dependent phosphorylation. Ca²⁺ sensitivity of force was measured before and after treatment with PKA. Figure8 shows the pCa-force relation in skinned fiber bundles prepared from NTG and TG mouse hearts before and after treatment with PKA and the corresponding pCa₅₀ values. The myofilaments from TG hearts demonstrated a significant leftward shift in the pCa-force relationship compared with myofilaments from NTG hearts, as determined by comparing the pCa₅₀ values. When treated with PKA, both NTG and TG pCa-force relationships shifted rightward, indicating a decrease in Ca²⁺ sensitivity, but PKA-treated TG fibers were still more sensitive to Ca²⁺compared with the PKA-treated NTG fibers. The ΔpCa₅₀between NTG and TG fibers was 0.08 ± 0.01 before PKA treatment and 0.07 ± 0.02 after PKA treatment. NTG and TG myofilament pCa-force relations had similar n _H values both before and after PKA treatment (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 8.

Normalized pCa-force relations and corresponding pCa₅₀ values (inset) of detergent extracted fiber bundles prepared from mouse heart papillary muscles, at sarcomere length 2.0 μm. Force was normalized to the corresponding maximum force at pCa 4.5. Data are expressed as means ± SE. pCa₅₀: NTG = 5.77 ± 0.02 (n = 5, from 4 hearts); NTG-PKA = 5.71 ±0.01‡ (n = 5, from 4 hearts); TG = 5.84 ± 0.01* (n = 6, from 5 hearts); TG-PKA = 5.77 ± 0.02†‡ (n= 6, from 5 hearts). *Significant difference between NTG and TG groups (P < 0.05); †significant difference between NTG-PKA and TG-PKA; ‡significant difference within groups.

Our findings of increased sensitivity to Ca²⁺ of force and ATPase rate in TG compared with NTG myofilaments could arise from an altered activation by Ca²⁺, by strong cross-bridges, or both (4, 33). We therefore also measured strong cross-bridge activated force in NTG and TG myofilament bundles by varying MgATP concentration at pCa 9.0. At relatively high pMgATP values, rigor cross-bridges can activate force even under relaxing conditions (28). Figure9 shows the pMgATP-force relation and the corresponding pMgATP₅₀ values for both groups. Sensitivity to strong cross-bridge activation was similar between the groups [pMgATP₅₀ (NTG) = 4.94 ± 0.06,n = 7; (TG) = 5.04 ± 0.07, n= 7]. Comparison of the n _H values showed no significant difference in cooperativity between groups (data not shown). Prolonged treatment in rigor conditions may damage myofilaments; therefore, we performed a separate series of experiments in which myofilaments were exposed to rigor conditions for only brief periods. We compared the ratios of force at pMgATP 5.0, close to the pMgATP₅₀, and at 8.0, the pMgATP that caused maximal contraction, in NTG and TG myofilaments. This ratio was not different between the groups (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 9.

Normalized pMgATP-force relations and corresponding pMgATP₅₀ values (inset) in detergent-extracted fiber bundles prepared from NTG and TG mouse heart papillary muscles, at a shortening length (SL) of 2.0 μm. Data are expressed as means ± SE. Force was normalized to the corresponding maximum force. pMgATP₅₀ (NTG) = 5.04 ± 0.07 (n = 7, from 4 hearts), and pCa₅₀ (TG) = 4.94 ± 0.06 (n = 7, from 4 hearts).

To investigate whether maximal Ca²⁺-activated force was altered in myofilaments from TG mice, we measured maximal Ca²⁺-activated tension at both a short (2.0 μm) and long (2.3 μm) SL to simulate tension at different ventricular volumes. Figure 10 shows the tension generated by both groups at different SL. At SL 2.0 μm, mean tension in NTG fibers was 31.3 ± 2.0 mN/mm² and 25.4 ± 2.8 mN/mm² in TG. At SL 2.3 μm, mean tension in NTG fibers was 37.0 ± 1.7 mN/mm² and 33.4 ± 2.4 mN/mm² in TG. Compared with NTG fibers, maximum tension of TG fibers was not significantly different at either sarcomere length.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 10.

Tension in skinned fiber bundles from NTG and TG mouse papillary muscles at SL = 2.0 μm and 2.3 μm. Data are expressed as means ± SE. Tension at SL 2.0 μm: NTG = 31.3 ± 2.0 mN/mm² (n = 5, from 3 hearts) and TG = 25.4 ± 2.8 mN/mm²(n = 6, from 3 hearts). Tension at SL 2.3 μm: NTG = 37.0 ± 1.7 mN/mm² (n = 12, from 6 hearts) and TG = 33.4 ± 2.4 mN/mm²(n = 11, from 4 hearts).

DISCUSSION

Experiments reported here are the first to compare the effects of β-adrenergic stimulation in an in situ preparation with the effects of PKA-dependent phosphorylation on Ca²⁺-activation of myofilaments from NTG and TG-α-Tm(Asp175Asn) mouse hearts. During β-adrenergic stimulation, PKA is activated and phosphorylates, among other targets, TnI. Phosphorylation of TnI is believed to be partially responsible for the increased relaxation rate observed during β-adrenergic stimulation through an allosteric effect, decreasing Ca²⁺ binding to TnC and/or increasing cross-bridge turnover rate (4, 40). The results of our in situ experiments indicate that in the absence of β-adrenergic stimulation, the α-Tm(Asp175Asn) mutation in TG mice manifests itself primarily as a defect in cardiac relaxation. However, during β-adrenergic stimulation, impaired relaxation is exacerbated, and both contractility and pressure development are diminished. Results of Ca²⁺-sensitive force measurements, before and after PKA-dependent phosphorylation, suggest that the impaired relaxation detected in situ is a result of increased Ca²⁺ sensitivity. Moreover, this altered relaxation in the TG hearts cannot be explained by changes in SR Ca²⁺ uptake rate, because SR vesicles prepared from NTG and TG mouse hearts had a similar maximal rate of Ca²⁺ uptake and pCa₅₀. Furthermore, this diminished response to β-adrenergic stimulation cannot be attributed to a decrease in β-adrenergic receptor density or affinity. At high HR, as occurs during β-adrenergic stimulation, altered sensitivity to Ca²⁺ in the hearts of TG mice may not permit full cardiac relaxation, thereby compromising LV filling and altering both contractility and LV pressure development. The hemodynamic changes produced by the mutation during a stress, when adrenergic activity is high, may be the stimulus for the hypertrophic response.

The in situ findings of altered hemodynamics are consistent with our previous studies with this same line of TG mice (25). In the working heart preparations perfused with crystalloid buffers, these TG mouse hearts had slowed relaxation rates and decreased contractility in the basal state compared with NTG hearts. After perfusion with Iso, relaxation rates and contractility increased to a lesser extent in the TG hearts compared with NTG. There was also an increase in LVEDP in TG working hearts, but we did not find this in in situ LV pressure measurements. This difference with regard to LVEDP between the present study and previous results may be due to compensatory mechanisms operating when the heart is beating in the closed chest of an intact animal compared with an isolated preparation that is not blood perfused. In confirmation of the in situ results presented here, echocardiography measurements in the previous study on lightly anesthetized TG mice showed no difference in LV end-diastolic dimension compared with controls. In the previous study, the TG mouse hearts also showed myocyte hypertrophy, myocyte disarray and fibrosis but showed no changes in MHC, actin, Tn, or Tm isoform expression.

Our results complement and extend findings on transgenic mice overexpressing β-Tm (TG-α-Tm) in the hearts (23, 24, 28,38). Myofilaments from both TG-α-Tm(Asp175Asn) and TG-β-Tm hearts demonstrated increased Ca²⁺ sensitivity of force and ATPase activity. When subjected to PKA-dependent phosphorylation, TG-β-Tm myofilaments had virtually no shift in Ca²⁺sensitivity of force, whereas TG-α-Tm(Asp175Asn) myofilaments were less sensitive to Ca²⁺. However, myofilaments from both of these transgenic models were significantly more sensitive to Ca²⁺ after PKA treatment compared with the respective PKA-treated NTG myofilaments. Strong cross-bridge activation, however, had different effects in these two models. TG-β-Tm myofilaments were more sensitive to strong cross-bridge activation, whereas TG-α-Tm(Asp175Asn) myofilaments were not. Isolated working heart preparations performed on transgenic β-Tm hearts demonstrated that relaxation rate is slowed in a manner consistent with our findings in TG-α-Tm(Asp175Asn) mice. Moreover, TG-β-Tm mice that express a very high ratio of β-Tm to α-Tm in the heart were found to develop hypertrophy and increased mortality.

Differences between TG-α-Tm(Asp175Asn) and TG-β-Tm models may be explained by considering the amino acid differences with regard to α-Tm. Compared with α-Tm, α-Tm(Asp175Asn) has one less negative charge in the Ca²⁺-dependent TnT binding region (12,22). Switching from α-Tm to α-Tm(Asp175Asn) would be expected to predominantly affect Ca²⁺-dependent binding to TnT and to exert its effect on thin filament activation only in the presence of Ca²⁺. On the other hand, β-Tm has 39 amino acid differences and two additional negative charges in the COOH-terminal region (Ser229Glu and His276Asn) compared with α-Tm. The amino acid differences in β-Tm, compared with α-Tm, are primarily in the Ca²⁺-independent Tm-TnT binding region (12, 22). Thus, when switching from α-Tm to β-Tm, it is not surprising that sensitivity to both Ca²⁺ and strong cross-bridge activation was altered.

How does the Asp175Asn mutation in Tm alter the activation of the thin filament? The three-state model of thin filament activation describes “open,” “closed,” and “blocked” states of the Tn-Tm complex with actin, corresponding to the state of cross-bridge attachment (14, 33). In terms of this model, α-Tm(Asp175Asn) may increase the likelihood of Tm movement from the closed to the open state in the presence of Ca²⁺. This increases the probability of cross-bridge cycling, easing activation of force development and ATPase activity, but only under activating conditions. In the absence of Ca²⁺ the mutation showed no effect on strong cross-bridge activation, suggesting that increased Ca²⁺ sensitivity is a result of altered interaction at the Ca²⁺-dependent Tm-TnT binding site rather than altered Tm-actin interaction. Furthermore, cooperativity and maximum Ca²⁺-activated tension and ATPase activity were unchanged in myofilaments from TG hearts compared with NTG, suggesting that the mutation does not affect the force/cross bridge or the number of strong cross bridges bound during activation. Overall, our conclusions are consistent with in vitro studies by other investigators examining the Asp175Asn mutation of Tm. Bing et al. (2) found that α-Tm(Asp175Asn) in thin filaments had no effect on activation in the absence of Tn, but when titrated with Tn in the presence of Ca²⁺, there was an increase in sliding velocity compared with wild-type Tm in an in vitro motility assay. Golitsina et al. (9) reported that α-Tm(Asp175Asn) labeled with pyrene iodoacetamide at cysteine-190 had normal binding to actin in both the absence and presence of Tn; however, this mutant Tm had an altered conformation on actin only in the presence of Tn, myosin S1, and Ca²⁺.

The hemodynamic alterations we have uncovered with the Asp175Asn mutation of α-Tm may be the stimulus for the hypertrophic response, but these defects may also directly contribute to a lethal event during exercise when the β-adrenergic system is activated, given that exercise is associated with increased incidence of death by FHC (21). The mutation could cause death indirectly, by activating a hypertrophic response leading to progressive pathological changes, directly, by compromising cardiac function during physiological stress, or through a combination of direct and indirect actions. We have provided evidence of a blunted relaxation-response to β-adrenergic stimulation in TG mice with the Asp175Asn mutation. Moreover, this altered relaxation compromises cardiac contractility and LV pressure development during β-adrenergic stimulation. Our finding of decreased LVDP during β-adrenergic stimulation may be particularly important because clinical data on people with hypertrophic cardiomyopathy show that hypotension during a graded exercise test is the best prognosticator of risk of sudden death (19).