Effects of Thyroid Hormone on Left Ventricular Performance and Regulation of Contractile and Ca2+-Cycling Proteins in the Baboon
Implications for the Force-Frequency and Relaxation-Frequency Relationships
The transcriptional, posttranscriptional, and related functional effects of thyroid hormone on primate myocardium are poorly understood. Therefore, we studied the effects of thyroid hormone on sarcoplasmic reticulum (SR) Ca2+-cycling proteins and myosin heavy chain (MHC) composition at the steady state mRNA and protein level and associated alterations of left ventricular (LV) performance in 8 chronically instrumented baboons. The force-frequency and relaxation-frequency relations were assessed as the response of LV isovolumic contraction (dP/dtmax) and relaxation (Tau), respectively, to incremental atrial pacing. Both the heart rate at which dP/dtmax was maximal and Tau was minimal (critical heart rates) in response to pacing were increased significantly after thyroid hormone. Postmortem LV tissue from 5 thyroid-treated and 4 additional control baboons was assayed for steady state mRNA levels with cDNA probes to MHC isoforms and SR Ca2+-cycling proteins. Steady state SR Ca2+-ATPase and phospholamban mRNA increased in the hyperthyroid state, and α-MHC mRNA appeared de novo, whereas β-MHC mRNA decreased. Western analysis (4 thyroid-treated and 4 control baboons) showed directionally similar changes in MHC isoforms and a slight increase in SR Ca2+-ATPase. In contrast, there was a statistically nonsignificant decrease in phospholamban protein, which resulted in a significant 40% decrease in the ratio of phospholamban to SR Ca2+-ATPase. Thus, thyroid hormone increases the transcription of Ca2+-cycling proteins and shifts MHC isoform expression in the primate LV. Our data suggest that both transcriptional and posttranslational mechanisms determine the levels of these proteins in the hyperthyroid primate heart and mediate, in part, the observed enhanced basal and frequency-dependent LV performance.
The cardiovascular manifestations of thyroid hormone excess have been studied extensively. In addition to alterations in ventricular loading conditions and interactions with the sympathetic nervous system, there is clear evidence that thyroid hormone has direct effects on cardiomyocytes that are mediated through both nuclear (transcriptional) and extranuclear (energy-regulating) mechanisms.1 2 3 4 Among the transcriptional regulatory actions of thyroid hormone are the effects on MHC isoform and SR Ca2+-cycling protein gene expression. Identification of thyroid responsive elements in the promoter region of the α-MHC and SERCA genes indicates that thyroid hormone transcriptionally regulates the expression of critical genes involved in contraction and relaxation of the cardiomyocyte.5 6
Changes in MHC isoforms are correlated with alterations in myosin ATPase activity and actinomyosin crossbridge cycle rate.7 8 9 A switch from the β (slow) to the α (fast) MHC isoform has been demonstrated in small animals (ie, rats and rabbits) that are in transition between the hypothyroid and hyperthyroid states. However, few data exist concerning the effects of thyroid hormone on α- and β-MHC gene expression in large animals and humans. Preliminary data suggest that exogenous thyroid hormone administration may increase α-MHC gene expression in the calf heart and, to a lesser extent, in the baboon heart but not in the dog heart.10 Moreover, in a single patient with dilated cardiomyopathy and severe hypothyroidism, thyroxine replacement therapy increased steady state α-MHC mRNA, accessed by sequential endomyocardial biopsies.11 Thus, the influence of thyroid hormone on MHC isoform gene expression in large mammals remains inadequately characterized.
The SR plays a pivotal role in excitation-contraction-relaxation coupling12 ; intracellular free Ca2+ concentration is regulated in part by the activity of SR proteins that control the release (ryanodine receptor), resequestration (Ca2+-ATPase and PLB), and storage of Ca2+ (calsequestrin). In particular, SERCA is known to be inhibited by PLB in its dephosphorylated state, and the stoichiometry between these critical components of SR has been demonstrated to be an important determinant of LV chamber and cardiomyocyte relaxation using targeted ablation and transgenic overexpression of this phosphoprotein in the mouse.13 14 The effects of thyroid hormone on the gene expression of these Ca2+-cycling proteins in small animals are well documented. Experimental cardiac hypertrophy produced by thyroid hormone administration results in increased rates of tension development and decline, which are associated with coordinate increases in expression of ryanodine receptor and SERCA mRNA. However, the effects of thyroid hormone on PLB are more controversial.15 16 17 18 19 Although steady state PLB mRNA expression decreases in small animals treated with thyroid hormone, PCR-quantified PLB mRNA from serial endomyocardial biopsies doubled in a patient after treatment of hypothyroidism and heart failure.11
Accordingly, the aim of the present study was to examine the effects of thyroid hormone excess on steady state mRNA and protein levels of the SR Ca2+-cycling proteins and MHC isoforms in the primate heart. In order to identify alterations in rate-dependent LV myocardial contractility and relaxation in response to thyroid hormone in a higher mammalian species, we critically examined the force-frequency and relaxation-frequency relationships, respectively. Although the physiological importance of the force-frequency relation in the conscious animal is somewhat controversial,20 21 22 study of the frequency-dependent effects of thyroid hormone on LV contractile and lusitropic properties provides insight into the role of altered contractile and Ca2+-cycling proteins on these properties in a species where baseline MHC composition is equivalent to that in humans.23
Materials and Methods
Eight sedated adult male baboons (Papio anubis) weighing 21 to 30 kg were preinstrumented for physiological monitoring using methods previously described.24 Briefly, after sedation with ketamine (10 mg/kg) and atropine (0.5 mg), the animals were intubated, and anesthesia was maintained by 1.0% to 1.5% halothane. A Konigsburg micromanometer (model P5-P7) and a polyvinyl catheter (outer diameter, 0.95 in; inner diameter, 0.06 in) were implanted in the LV apex, and miniaturized sonomicrometer pairs were placed in the endocardium across the LV anteroposterior minor axis (3 MHz, 6 mm). A second polyvinyl catheter was implanted in the right atrium for central venous access, and pacing wires were sewn to the right atrial appendage. Wires and tubes were tunneled subcutaneously into the interscapular area for later use. Postoperative pain was reduced by the use of buprenorphine HCl (0.01 mg/kg IM, every 6 hours), and postoperative antibiotics (cephonidid sodium 25 mg/kg) were administered for 5 days to reduce the risk of infection. After a minimum of 1 week for postoperative recovery, baseline hemodynamic studies were performed (see below).
Thyroxine tablets were crushed and concealed in fruit, and the animals took them under direct supervision. The initial dose of thyroxine was 0.25 mg/kg per day for 10 days and then 0.25 mg/kg every other day for a total of 26.8±2.7 days (range, 22 to 30 days).
Hemodynamic Data Acquisition and Analysis
The micromanometer and fluid-filled catheters were calibrated with a mercury manometer. Zero drift of the micromanometer was corrected by matching the LV end-diastolic pressure measured simultaneously through the LV catheter. The fluid-filled LV catheter was connected to a precalibrated Statham 23 dB transducer with zero pressure at the level of the mid right atrium. The transit time of ultrasound between the ultrasonic dimension crystals was measured with a multichannel sonomicrometer (Triton Technology, Inc) and converted to distance, assuming a constant velocity of sound in blood of 1.55 mm/ms.
Analog signals for high fidelity and fluid-filled LV pressures, LV short-axis dimension, LV dP/dt, and the electrocardiogram were recorded on-line on a Gould multichannel recorder at 25- and 100-mm/s paper speed, digitized through an A-D board (Dual Control Systems) interfaced to an IBM AT computer at 500 Hz, and stored on a floppy disk. Data were analyzed using algorithms and software developed in our laboratory.25 Steady state data were acquired over 5 seconds during spontaneous respiration and averaged.
Fractional shortening of the LV minor axis was calculated as (EDD−ESD)/EDD. LV volumes were derived from minor-axis diameter (D) measurements as LV volume=π/6·(D).3
The analog LV dP/dt signal was obtained on-line by electronic differentiation of the high-fidelity LV pressure signal and by digitization of the LV micromanometer pressure waveform. LV end diastole was defined as the time that LV dP/dtmax increased by at least 150 mm Hg/s for 50 milliseconds, and LV end systole was defined as the time of the maximum ratio of LV pressure to LV minor-axis dimension. LV developed pressure was calculated as the LV systolic minus end-diastolic pressure. Tau was derived from the high-fidelity LV pressure tracing using the method of Weiss et al,26 which assumes a zero asymptote and has been shown to be directionally equivalent to other mathematical approaches for quantification of isovolumic pressure decay.27 Tau is equal to the time in milliseconds for LV pressure to decay to 1/e, where e is the base of the natural logarithm; thus, decreases in the time constant reflect improved ventricular relaxation.
Analysis of the Force-Frequency Relation
In order to quantify the effects of thyroid hormone on inotropic and lusitropic LV chamber properties in vivo, we examined the response of isovolumic contraction and relaxation to incremental pacing. We defined the critical heart rate as the heart rate at which dP/dtmax and Tau reached their maxima and minima, respectively, during progressive increases in heart rate. Thus, the value beyond which dP/dtmax declined and Tau increased by 5%, respectively, or when either AV block or pulsus alternans intervened was the critical heart rate for isovolumic contraction and relaxation. This occurred in one euthyroid and two hyperthyroid animals at heart rates of 201, 340, and 330 bpm, respectively.
Hemodynamic studies were performed 1 week after instrumentation and were repeated 22 to 30 days after thyroxine administration. Animals were sedated with valium (1 mg) and ketamine (100 mg), and cholinergic blockade was achieved with atropine (0.4 to 0.8 mg IV). After baseline hemodynamic data were acquired, atrial pacing was instituted at 1.6 to 1.8 Hz and was increased at 0.2-Hz increments until there was a decrease in LV dP/dtmax or until either AV block (unresponsive to 1 mg atropine) or pulsus alternans appeared.
RNA Isolation and Northern Blot Analyses
Total cellular RNA was isolated from ≈1 g of ventricular tissue from five hyperthyroid (experimental) and four additional euthyroid (control) adult baboons. Only one of the control animals had been instrumented. The hearts were removed immediately at necropsy and quickly weighed; the LV was cross-sectioned, and transmural samples from the anterior free wall were placed in liquid nitrogen.
RNA extraction was carried out by the acid guanidinium thiocyanate–phenol chloroform method.28 Total RNA (20 μg) was denatured at 65°C for 10 minutes, fractionated on 1% agarose gel containing 5% (vol/vol) formaldehyde, and transferred by blotting onto a charged nylon membrane (Hybond-N+, Amersham). The RNA was covalently bound to the membrane by illumination with ultraviolet light (1.2×105 μJ) in a Stratalinker (Stratagene).
Hybridization was carried out in 50% formamide at 42°C for 16 hours.2 A mixture of various random-primed as well as specifically primed cDNA probes29 labeled with 32P was used. The membranes were washed twice with 1× SSC/0.1% SDS and four times with 0.1× SSC/0.1% SDS for 15 minutes each at room temperature, air-dried, and exposed to Kodak X-omat AR film for 20 to 48 hours in a cassette containing an intensifying screen at −70°C. The autoradiographic bands were quantified using an enhanced laser densitometer (Ultrascan XL, LKB).
The following probes made from the corresponding cDNA were used for Northern blot analyses: (1) SERCA: 1.7-kb cDNA fragment of the rabbit cardiac/slow-twitch muscle Ca2+-ATPase (BamHI–3′ end),30 (2) PLB: 1.3-kb cDNA fragment of the rabbit cardiac PLB (5′-EcoRI-linker-EcoRI),31 (3) ryanodine receptor: 2.25-kb cDNA fragment corresponding to amino acids 2662 to 3413 of rabbit cardiac muscle ryanodine receptor,32 and (4) GAPDH: 1.6-kb chicken GAPDH cDNA cloned from chicken breast muscle library.33 All four of the probes were made by using [α-32P]dCTP in a random-priming reaction.
The probes for α-MHC and β-MHC were made using synthetic double-stranded deoxy oligonucleotides as follows: (5) α-MHC: 58-mer double-stranded synthetic DNA fragment corresponding to a unique 3′-untranslated region of baboon α-MHC cDNA34 end-labeled with the use of [γ-32P]ATP and T4-polynucleotide kinase and synthesized in the presence of specific forward and reverse primers in a Klenow reaction containing [α-32P]dCTP to achieve high specific activity, and (6) β-MHC: 96-mer double-stranded synthetic DNA fragment corresponding to a unique 3′-untranslated region of baboon β-MHC cDNA35 used in the presence of specific forward and reverse primers in a Klenow reaction to generate a cDNA probe.
Tissue Homogenization for MHC Proteins
α-MHC and β-MHC proteins were analyzed using Western blotting techniques on tissue extracted by methods similar to those described previously.34 Briefly, LV tissue was quickly frozen and maintained at −70°C. Tissue was powdered in liquid nitrogen, and ≈75 mg of tissue was added to cold buffer at a 1:50 (wt/vol) ratio. The homogenizing buffer (pH 8.8) consisted of 100 mmol/L Na4P2O7, 5 mmol/L EDTA, 4 mmol/L 2-mercaptoethanol, 20 μmol/L leupeptin, and 10 KIU (≈4000 KIU/mg) aprotinin. Tissue was homogenized and stirred for 2 hours at 4°C before being spun in a centrifuge at 48 000g, also at 4°C. The supernatant was saved, and glycerol was added to each sample for a final ratio of 1:1 (vol/vol).
Tissue Homogenization for PLB and SERCA
PLB and SERCA were analyzed using Western blotting techniques on tissue homogenized by methods described previously.36 Briefly, ≈100 mg of tissue powdered in liquid nitrogen was homogenized in 1 mL homogenizing buffer per 100 mg tissue. Homogenizing buffer consisted of 10 mmol/L imidazole (pH 7.0), 0.3 mol/L sucrose, 0.3 mmol/L phenylmethylsulfonyl fluoride, 10 mmol/L sodium metabisulfite, and 1 mmol/L dithiothreitol.
Western Analysis of Proteins
Protein concentrations were determined using the Bio-Rad protein assay, with bovine serum albumin serving as the protein standard. For Western blot analysis, homogenates were subjected to electrophoresis using 8% (α- and β-MHC) or 10% to 20% gradient (PLB and Ca2+-ATPase) SDS polyacrylamide gels. The amounts of protein used for each protein were in the linear range for each antibody (data not shown) and were as follows: α-MHC, 20 μg per lane; β-MHC, 0.5 μg per lane; and PLB and Ca2+-ATPase, 12.5 μg per lane. The separated proteins were then transferred to 0.45-μm nylon membranes overnight at 4°C and then washed in TBS (100 mmol/L Tris-HC1, pH 7.0, with 0.9% NaCl) three times for 5 minutes each. PLB and Ca2+-ATPase were detected on the same blot by cutting the membranes horizontally in two around the 45-kD molecular mass marker. Subsequently, the lower portion of the membrane was probed for PLB, and the upper portion was probed for Ca2+-ATPase. All membranes were blotted for 1 hour using 5% milk buffer. After another series of washes (three for 5 minutes each), blots were incubated in primary antibody for 2 hours at room temperature and then overnight at 4°C. The antibodies used were mouse monoclonal IgG1s. Anti–α-MHC and anti–β-MHC antibodies were generous gifts from Dr J.J. Leger, and each was used at a dilution of 1:500. Anti-PLB and anti–Ca2+-ATPase IgG1s were purchased from Upstate Biotechnology Inc and Affinity BioReagents, respectively, and used at a dilution of 1:1000. All antibody solutions were made up in 2% milk TBS. After allowing the primary antibody solutions to return to room temperature, blots were washed in TBS (three times for 5 minutes each) and placed into secondary antibody (goat anti-mouse–HRP conjugate) at 1:1000 for 2 hours (MHC) or 4 hours (Ca2+-ATPase and PLB) at room temperature and gently rocked. Membranes were then washed a final time in TBS (three times for 5 minutes each), and proteins were visualized with horseradish peroxidase developing agents 4CN and H2O2 (HRP detection system, BioRad). Developed membranes were washed in distilled water and allowed to dry before quantification of the scanned image using ImageQuant software, with data reported as integrated density units. In order to account for variable protein loading, values for PLB and SERCA proteins were normalized to total MHC that was quantified from the Coomassie-stained gel. Assays were performed in duplicate with tissue from 4 euthyroid and 4 hyperthyroid baboons examined simultaneously.
Thyroid Function Tests
Thyroid function tests were performed before the baseline experiment with the animals in the euthyroid state and before the terminal experiment with the animals in the hyperthyroid state. The tests were performed at Cincinnati (Ohio) Veterinary Laboratory, Inc. T3 RIA, T4, and free T4 levels were measured at each state.
In order to test the global hypothesis of equality of within-animal means, repeated measures ANOVA was performed across the sequence of euthyroid unpaced, euthyroid paced, and hyperthyroid states. Greenhouse-Geisser corrections were used to correct for multisample asphericity. When significant differences were found, group means were compared with prespecified contrasts, and a Bonferroni correction was used to control for multiple comparisons. Paired mean data were compared with Student's t tests. Western blot data were compared using unpaired t tests on nonnormalized data averaged from duplicate samples. Normalized data failed an equal variance test and were therefore compared with Mann-Whitney rank sum tests. A value of P<.05 was considered significant. Data are expressed as mean±SD.
Thyroid hormone supplementation produced a state characterized by clinical and laboratory evidence of thyrotoxicosis. In the hyperthyroid state, the thyroid function tests were reported to be in excess of 500 ng/dL for T3 RIA, 20 μg/dL for T4, and 4.2 ng/dL for free T4. The normal values for baboons in our laboratory (n=11) are 105±49 ng/dL, 5.2±1.4 μg/dL, and 1.4±0.5 ng/dL, for T3 RIA, T4, and free T4, respectively. Body weight was significantly reduced after supplementation with thyroxine (26.4±2.9 versus 23.5±2.3 kg, P<.05).
The hemodynamic effects of chronic thyroid administration in the conscious baboon are shown in Fig 1⇓ and Table 1⇓. Thyroid hormone significantly increased the baseline heart rate, LV systolic and developed pressures, and the peak rate of isovolumic LV pressure development (dP/dtmax). Concomitantly, thyroid hormone decreased Tau and the shortening fraction, whereas LV end-diastolic pressure and velocity of circumferential shortening were unchanged. By contrast, at matched heart rates, thyroid hormone increased LV velocity of circumferential shortening and end-diastolic pressure, whereas the augmentations of isovolumic contraction (dP/dtmax) and especially relaxation (Tau) were less prominent. These data suggest that the positive inotropic and lusitropic effects of thyroid hormone are partially mediated by increased contraction frequency.
The LV end-diastolic and end-systolic dimensions were similar in the hyperthyroid and euthyroid state at baseline but were significantly larger in the hyperthyroid than euthyroid state when compared at matched heart rates.
Effects of Incremental Atrial Pacing on dP/dtmax
The effects of incremental pacing on LV dP/dtmax (force-frequency response) in the euthyroid and hyperthyroid states are shown in Fig 2⇓. The force-frequency relations were biphasic in each animal, with an initial positive ascending limb and a subsequent negative descending limb. Incremental atrial pacing in both the euthyroid and hyperthyroid states resulted in significant increases in dP/dtmax from the initial paced heart rate to the critical heart rate (1956±502 to 2399±428 mm Hg/s and 3275±1161 to 3516±1159 mm Hg/s, respectively; both P<.05).
In order to account for the preload dependence of dP/dtmax, we examined the relation between heart rate and dP/dtmax corrected for the instantaneous EDD, assessed as (dP/dtmax)/EDD. An identical relationship was seen (data not shown), suggesting that the biphasic nature of this relationship was not the result of heart rate–mediated decreases in LV end-diastolic volume (preload). In addition, only small changes in LV end-diastolic dimension occurred at paced rates near the critical heart rate.
Thyroid hormone shifted the force-frequency curves upward and to the right and delayed the development of the descending limb of the force-frequency relation (Fig 2⇑); thus, the critical heart rate for dP/dtmax was significantly higher in the hyperthyroid than euthyroid state (284±46 versus 194±29 bpm, P<.001).
Effects of Incremental Atrial Pacing on Isovolumic Relaxation
The effects of incremental pacing and thyroid hormone on Tau (relaxation-frequency) are shown in Fig 2⇑. The relaxation-frequency relations were biphasic in each animal, with an initial negative descending limb demonstrating accelerated relaxation and a subsequent positive ascending limb demonstrating impaired and incomplete isovolumic relaxation. Incremental atrial pacing in both the euthyroid and hyperthyroid states resulted in significant decreases in Tau from the initial paced heart rate to the critical heart rate (34.3±7.8 to 25.3±5.9 milliseconds and 22.5±4.8 to 18.5±5.0 milliseconds, respectively; both P<.05).
Thyroid hormone shifted the relaxation-frequency curves downward and to the right and delayed the onset of “impaired and incomplete” isovolumic relaxation (ascending limb). There were no significant differences between the absolute values of the critical heart rates for dP/dtmax and Tau in either the euthyroid or hyperthyroid state.
Steady State Expression of MHC Isoforms and SR Proteins
Steady state SERCA and PLB mRNA levels (normalized to GAPDH and expressed in arbitrary units) were significantly higher in hyperthyroid than control hearts (1.69±0.12 versus 0.94±0.24 AU and 1.5±0.68 versus 0.37±0.11 AU, respectively, both P<.001) (Fig 3⇓). Steady state levels of ryanodine receptor mRNA increased in the hyperthyroid hearts; however, because of the large number of base pairs (16 kb), the increase could not be quantified.
Steady state β-MHC mRNA levels were significantly lower in the hyperthyroid than control animals (0.91±0.18 versus 1.60±0.10, P<.0001). Control baboons did not express α-MHC mRNA; by contrast, this myosin isoform was expressed in response to thyroid hormone (Fig 4⇓).
Expression of MHC Isoforms and SR Proteins
Western blot analyses of LV tissue demonstrated a 33% decrease in the β isoform of MHC protein and the de novo appearance of α-MHC in hyperthyroid compared with euthyroid baboons (Fig 5⇓). These findings are consistent with those observed for steady state mRNA expression for these proteins in the same tissue. Furthermore, similar to the change in mRNA expression for Ca2+-ATPase, Ca2+-ATPase protein increased by 29% in hyperthyroid compared with euthyroid tissue. Although large increases in steady state expression of PLB mRNA were found in the LV myocardium from hyperthyroid baboons, PLB protein levels decreased significantly (23%). However, when data were normalized to MHC from the same gel, the differences between euthyroid and hyperthyroid ventricles for PLB and SERCA protein levels were not statistically significant (Table 2⇓). Nevertheless, in both cases (normalized and nonnormalized data), the ratio of PLB to Ca2+-ATPase, an index of SR function, was significantly less in hyperthyroid than euthyroid baboons.
The principal findings of the present study in the chronically instrumented nonhuman primate are that (1) hyperthyroidism of short duration (3 to 4 weeks) results in augmented indices of isovolumic LV contraction (dP/dtmax) and relaxation (Tau) and enhanced frequency-dependent rates of isovolumic LV pressure development and decay, and (2) these changes are associated with altered regulation of cardiomyocyte MHC composition and Ca2+-cycling proteins.
In small mammals, the myocardial functional effects of thyroid hormone are known to be mediated in part through transcriptionally regulated alterations in the Ca2+-cycling proteins and MHC isoforms; specifically, thyroid hormone upregulates the expression of genes encoding for the SERCA, ryanodine receptor, and α-MHC isoform. Data regarding the effects of thyroid hormone on steady state PLB mRNA are conflicting, with reports of an increase, decrease, or no change in gene expression.1 2 11 15 16 17 18 19 37
In order to explore the molecular basis for the effect of thyroid hormone on large mammals with a close phylogenetic relationship to humans, gene expression for contractile and Ca2+-cycling proteins was examined in LV myocardium from baboons treated with thyroid hormone and from untreated control baboons. We found a 79% increase in SERCA mRNA expression that was associated with a greater than fourfold increase in mRNA coding for PLB. In addition, there was a 43% decrease in β-MHC mRNA that was associated with the de novo appearance of α-MHC mRNA in the hyperthyroid myocardium. It should be emphasized that the extent to which these changes reflect a difference in the rate of transcription and/or message stability cannot be determined from the present studies.
Changes in steady state mRNA levels were associated with comparable changes in protein levels for β-MHC (−33%) and the de novo appearance of α-MHC. Protein levels of the SR-ATPase increased while PLB decreased in the thyroid-treated baboons. However, these changes did not reach statistical significance after the levels were normalized to the level of total MHC (Table 2⇑). It is likely that differences in statistical significance between normalized and nonnormalized data were due to the variance associated with MHC and the obligatory small sample size. Taken together, these data suggest that T4-mediated posttranscriptional and/or posttranslational events are important in the regulation of these proteins. The direction of changes produced by thyroid hormone in the Ca2+-cycling proteins and in the magnitude of the significant change in stoichiometry between PLB and the SERCA are remarkably similar to recent studies by Kiss et al18 performed in the rat. It should be recognized that the control animals were not all sham-operated. However, data from the instrumented control did not differ from the other data. Moreover, there is no evidence that chronic catheter implantation produces significant changes in message or protein levels of contractile or Ca2+-cycling proteins. In recent studies in the guinea pig, we have found a small (≈10%) α- to β-MHC shift (unlike the β- to α-MHC shift we observed in the present study) 1 week after surgery (authors' unpublished data, 1994). Thus, it is unlikely that the differences in protein and mRNA that we observed in thyroid-treated animals can be explained by the lack of a sham-operated control group.
Previous investigations suggest that augmented rate-dependent indices of contractility and relaxation in hyperthyroid animals result from an abbreviated time course of Ca2+ transients and alterations in the kinetics of actinomyosin crossbridge cycling.38 39 40 Thus, the increased velocity of circumferential shortening and rates of pressure development and decay observed here may be explained in part by the de novo expression of the α-MHC isoform, which is associated with an increased catalytic activity of myosin ATPase relative to β-MHC and by increased SERCA gene expression, which is associated with increased uptake and release of Ca2+ from the SR. In addition, the posttranscriptionally mediated reduction in protein expression of PLB substantially diminished the ratio of this phosphoprotein to SERCA (Table 2⇑). This significant alteration in the stoichiometry between the SR pump and its major inhibitor would be expected to further enhance resequestration and release of Ca2+ from the SR and resultant LV isovolumic relaxation and contraction.13 14 It should be noted that we cannot exclude thyroid hormone–mediated alterations in basal phosphorylation status of PLB from these studies; however, in another preparation, we failed to observe differences in this potential determinant of SR function.41
The Force-Frequency Relationship
Heart rate is an important determinant of myocardial performance, but chronotropic effects on myocardial contractility in the conscious animal are controversial. Some studies suggest that the influence of heart rate on contractility is minimal, whereas others report effects only during either adrenergic stimulation or exercise.20 21 22 In contrast, the present data indicate that heart rate has a significant positive effect on myocardial contractility in the conscious sedated baboon. With the baboons in the euthyroid state, incremental pacing produced a significant 26% increase in dP/dtmax. Thus, the present results with baboons in the euthyroid state are strikingly similar to the 28% increase in dP/dtmax with incremental pacing in dogs reported by Freeman et al.20 Since dP/dtmax is preload dependent and since increased heart rates resulted in variably reduced LV end-diastolic volumes, it is likely that the magnitude of the force-frequency effect was underestimated in the present study.
We have also shown that the force-frequency relationship in the sedated baboon is biphasic, with an initial positive slope (ascending limb) and a subsequent negative slope (descending limb). Although a bell-shaped relationship was described earlier in isolated muscle strips,42 43 a descending limb of the force-frequency relationship has never to our knowledge been described in the intact normal animal.20 44 45 It should be recognized that the stimulation frequencies necessary to produce the descending limb of this relationship were greater than stimulation frequencies used previously.
Thyroid hormone caused a significant increase in the critical heart rate for dP/dtmax; this rightward shift in the onset of the descending limb of the force-frequency response suggests that the time course of the Ca2+ transient (and hence, excitation-contraction coupling) is abbreviated by thyroid hormone. The alterations in contractile and Ca2+-cycling proteins that we observed in response to thyroid hormone may explain the delay in the onset of the descending limb. It is unlikely that the descending limb is due to ischemia, because thyroid hormone–induced increases in myocardial oxygen consumption should have the opposite effect on the critical heart rate (ie, an earlier onset of the descending limb). We postulate that in the euthyroid state, an increase in the stimulation frequency augments the velocity of contraction to a maximum level determined by the maximal capacity for SR Ca2+ release and sequestration; in the hyperthyroid state, the increased number of SERCA pumps and the decrease in the relative abundance of PLB to the SERCA allow more rapid Ca2+ cycling. The coordinate increased rate of actinomyosin crossbridge formation mediated by increased α-MHC and resultant increased myosin ATPase shifts the critical heart rate for isovolumic contraction to the right. Other actions of thyroid hormone vis-à-vis effects on L-type Ca2+ channels46 and Na+,K+-ATPase pumps47 or alterations in the blood volume48 may contribute to an enhancement of systolic LV performance.
The Relaxation-Frequency Relationship
A phenomenon similar to the force-frequency relationship was observed when we examined the effects of heart rate on the time constant of isovolumic relaxation; specifically, there was a 26% shortening of Tau with increasing heart rate. Thus, similar to the effect on contraction, relaxation is augmented at higher rates of stimulation. Although the latter has been documented by others,20 49 the present study is the first to demonstrate a nonlinear relationship between the heart rate and Tau. The time constant decreased monotonically with incremental pacing until a critical heart rate was achieved; after which, the relationship reversed, and Tau increased. These data suggest that after exceeding a critical heart rate, the ability of the SR to resequester Ca2+ becomes limited, and this results in incomplete relaxation.
Thyroid hormone altered the relaxation-frequency relationship. At baseline heart rates, Tau was significantly shorter and the critical heart rate of Tau was significantly higher in the hyperthyroid than euthyroid state. These data are consistent with a thyroid hormone–induced shortening of the time course for Ca2+ reuptake.38 40 46 Thus, with the thyroid hormone–induced increase in abundance of SERCA coupled with the decrease in protein expression of the SERCA inhibitor, PLB, the ability to resequester Ca2+ into the SR becomes enhanced even at very high rates of stimulation.
The frequency-dependent potentiation of both inotropy (dP/dtmax) and lusitropy (Tau) tended to be blunted in the hyperthyroid compared with the euthyroid state. These attenuated responses suggest that the subcellular processes responsible for LV isovolumic contraction and relaxation are near maximal capacity and that attempts to further augment chamber function are limited by the biophysical properties of the apparatus responsible for excitation-contraction-relaxation (ie, the number of actinomyosin crossbridges, SR Ca2+ pumps, and/or channels available). These findings are in agreement with those studies in which the β-adrenergic effects on the relaxation rate and intracellular Ca2+ concentration were decreased in isolated muscle preparations from hyperthyroid rats.38 39 50
In summary, we have critically examined the effects of thyroid hormone on cardiac mechanics, force-frequency and relaxation-frequency relations, and steady state protein and mRNA levels for the Ca2+-cycling proteins and MHC isoforms in the primate. Heart rate–independent effects on isovolumic contraction and relaxation (dP/dtmax and Tau, respectively) and augmented force- and relaxation-frequency relations were associated with de novo gene and protein expression of the α-MHC isoform and decreases in steady state β-MHC mRNA and protein; regulation of Ca2+-cycling proteins is complex, with increases in steady state mRNA and a substantial increase in the relative abundance of the SR-ATPase to PLB. The ability of thyroid hormone to favorably regulate cardiac MHC isoforms and Ca2+-cycling proteins with resultant improvement in contraction and relaxation of the primate myocardium could be exploited for therapy of heart failure using congeners that are devoid of the extranuclear effects of thyroid hormone.51
Selected Abbreviations and Acronyms
|EDD||=||LV end-diastolic dimension|
|ESD||=||LV end-systolic dimension|
|KIU||=||kallikrein inhibitor units|
|LV||=||left ventricle (ventricular)|
|MHC||=||myosin heavy chain|
|Tau||=||time constant of LV isovolumic relaxation|
This study was supported in part by National Institutes of Health grant HL-33579, SCOR in Heart Failure P-50-HL52318, and American Heart Association Fellowship Grant SW-94-34-F. We gratefully acknowledge the technical assistance of Gary Flesher and Tom Friede and the secretarial support of Norma Burns.
Reprint requests to Brian D. Hoit, MD, Division of Cardiology, University of Cincinnati Medical Center, PO Box 670542, Cincinnati, OH 45267-0542.
- Received February 9, 1996.
- Accepted July 24, 1996.
Polikar R, Burger AG, Scherrer U, Nicod P. The thyroid and the heart. Circulation. 1993;87:1435-1441.
Davis PJ, Davis FB. Acute cellular actions of thyroid hormone and myocardial function. Ann Thorac Surg. 1993;56:S16-S23.
Hartong R, Wang N, Kuiokawa R, Lazar M, Glass CK, Apriletti JW, Dillman WH. Delineation of three different thyroid hormone-response elements in promoter of rat sarcoplasmic reticulum Ca2+ ATPase gene. J Biol Chem. 1994;269:13021-13029.
Korecky B, Zak R, Schwartz K, Aschenbrenner V. Role of thyroid hormone in regulation of myosin composition, contractility and size of heterotopically isotransplanted rat heart. Circ Res. 1987;60:824-830.
Hoh JF, McGarth PA, Hale PT. Electrophoretic analysis of multiple forms of rat cardiac myosin: effects of hypophysectomy and thyroxine replacement. J Mol Cell Cardiol. 1977;10:1053-1076.
Morkin E. Regulation of myosin heavy chain genes in the heart. Circulation. 1993;87:1451-1460.
Landenson PW, Sherman SI, Baughman KL, Ray PE, Feldman AM. Reversible alterations in myocardial gene expression in a young man with dilated cardiomyopathy and hypothyroidism. Proc Natl Acad Sci U S A. 1992;89:5251-5255.
Arai M, Matsui H, Periasamy M. Sarcoplasmic reticulum gene expression in cardiac hypertrophy and heart failure. Circ Res. 1994;74:555-564.
Hoit BD, Khoury SF, Kranias EG, Ball N, Walsh RA. In vivo echocardiographic detection of enhanced left ventricular function in gene-targeted mice with phospholamban deficiency. Circ Res. 1995;77:632-637.
Arai M, Otsu K, MacLennan DH, Alpert NR, Periasamy M. Effects of thyroid hormone on the expression of mRNA encoding sarcoplasmic reticulum proteins. Circ Res. 1991;69:266-276.
Nagai R, Zarain-Herzberg Z, Brandl CJ, Fujii J, Tada M, MacLennan DH, Alpert NR, Periasamy M. Regulation of myocardial Ca2+ ATPase and phospholamban mRNA expression in response to pressure overload and thyroid hormone. Proc Natl Acad Sci U S A. 1989;86:2966-2970.
Rohrer D, Dillman WH. Thyroid hormone markedly increases the mRNA coding for sarcoplasmic reticulum Ca+2 ATPase in the rat heart. J Biol Chem. 1988;263:6941-6944.
Kiss E, Jakab G, Kranias EG, Edes I. Thyroid hormone–induced alterations in phospholamban protein expression: regulatory effects on sarcoplasmic reticulum Ca2+ transport and myocardial relaxation. Circ Res. 1994;75:245-251.
Freeman GL, Little WC, O'Rourke RA. Influence of heart rate on left ventricular performance in conscious dogs. Circ Res. 1987;61:455-464.
Kambayashi M, Miura T, Oh BH, Rockman HA, Murata K, Ross J Jr. Enhancement of the force-frequency effect on myocardial contractility by adrenergic stimulation in conscious dogs. Circulation. 1992;86:572-580.
Miura T, Miyazaki S, Guth BD, Kambayashi M, Ross J Jr. Influence of the force-frequency relation on left ventricular function during exercise in conscious dogs. Circulation. 1992;86:563-571.
Henkel RD, VandeBerg JL, Shade RE, Leger JJ, Walsh RA. Cardiac beta myosin heavy chain diversity in normal and chronically hypertensive baboons. J Clin Invest. 1989;83:1487-1493.
Hoit BD, Shao Y, Kinoshito A, Gabel M, Husain A, Walsh RA. Effects of angiotensin II generated by an angiotensin converting enzyme-independent pathway on left ventricular performance in the conscious baboon. J Clin Invest. 1995;95:1519-1527.
Walsh RA, O'Rourke RA. Direct and indirect effects of calcium entry blocking agents on isovolumic left ventricular relaxation in conscious dogs. J Clin Invest. 1985;75:1426-1434.
Weiss JL, Frederickson JW, Weisfeldt ML. Hemodynamic determinants of the time course of fall in canine left ventricular pressure. J Clin Invest. 1976;58:751-760.
Starling MR, Montgomery DG, Mancini GBJ, Walsh RA. Load independence at the rate of isovolumic relaxation in man. Circulation. 1987;76:1274-1281.
Kingston RE. Preparation and analysis of RNA. In: Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K, eds. Short Protocols in Molecular Biology. 2nd ed. New York, NY: Greene Publishing Associates/John Wiley & Sons Inc; 1992:4-23.
Otsu K, Willard HF, Khanna VK, Zomzato F, Green NM, MacLennan DH. Cloning of cDNA encoding the Ca2+ release channel (ryanodine receptor) of rabbit cardiac muscle sarcoplasmic reticulum. J Biol Chem. 1990;265:13472-13483.
Wei JY, Spurgeon HA, Lakatta EG. Electromechanical responsiveness of hyperthyroid cardiac muscle to β-adrenergic stimulation. Am J Physiol. 1982;243:E114-E122.
Bing OHL, Hague NL, Perreault CL, Conrad CH, Brooks WW, Sen S, Morgan JP. Thyroid hormone effects on the intracellular calcium and inotropic responses of rat ventricular myocardium. Am J Physiol. 1994;267:H1112-H1121.
Kiss E, Ball NA, Kranias EG, Walsh RA. Differential changes in cardiac phospholamban and sarcoplasmic reticular Ca2+-ATPase protein levels: effects on Ca2+ transport and mechanics in compensated pressure-overload hypertrophy and congestive heart failure. Circ Res. 1995;77:759-764.
Hasenfuss G, Reinecke H, Studer R, Meyer M, Pieske B, Holtz J, Holubarsch C, Posival H, Just H, Drexler H. Relation between myocardial function and expression of sarcoplasmic reticulum Ca+2 ATPase in failing and nonfailing human myocardium. Circ Res. 1994;75:434-442.
Mulieri LA, Hasenfuss G, Leavitt B, Allen PD, Alpert NR. Altered myocardial force-frequency relation in human heart failure. Circulation. 1992;85:1743-1750.
Anderson PAW, Manring A, Serwer GA, Benson DW, Edwards SB, Armstrong BE, Sterba RJ, Floyd RD IV. The force-interval relationship of the left ventricle. Circulation. 1979;60:334-348.
Johnson EA. Force frequency relationship of cardiac muscle. Am J Physiol. 1979;13:475-496.
Kim B, Smith TW, Marsh JD. Effect of thyroid hormone on slow calcium channel function in cultured chick ventricular cells. J Clin Invest. 1987;80:88-94.
Orlowski J, Lingrel JB. Thyroid and glucocorticoid hormones regulate the expression of multiple Na,K-ATPase genes in cultured neonatal rat myocytes. J Biol Chem. 1990;265:3462-3470.
Gay R, Lee RW, Appleton C, Olajos M, Martin GV, Morkin E, Goldman S. Control of cardiac function and venous return in thyrotoxic calves. Am J Physiol. 1987;252:H467-H473.
Miura T, Miyazaki S, Guth B, Indolfi C, Ross J Jr. Heart rate and force frequency effects on diastolic function of the left ventricle in exercising dogs. Circulation. 1994;89:2361-2368.
Beekman RE, van Hardeveld C, Simonides WS. On the mechanism of the reduction by thyroid hormone of β-adrenergic relaxation rate stimulation in rat heart. Biochem J. 1989;259:229-236.
Pennock GD, Raya TE, Bahl JJ, Goldman S, Morkin E. Combination treatment with captopril and the thyroid hormone analogue 3,5-diiodothyropropionic acid: a new approach to improving left ventricular performance in heart failure. Circulation. 1993;88:1289-1298.