INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

DESPITE THE CONSISTENT observation that chronic exercise training reduces circulating catecholamine levels during submaximal exercise (40), there is surprisingly little agreement concerning the training effects on catecholamine responsiveness of the heart (30). Cardiac contractile responsiveness to -adrenergic stimulation has been found to increase (4, 29, 42, 46), decrease (10), or remain unchanged (32, 36) with exercise training. The training effects on cardiac contractile responsiveness to -adrenergic stimulation are even more equivocal due in part to the small number of investigations in this area (25) and the disparate nature of the cardiac -adrenergic response itself (11, 14).

In general, -adrenergic stimulation of cardiac tissue is reported to induce a brief (<2 min) negative inotropy, ostensibly through action of the _1b-receptor subtype, followed by a sustained positive inotropy via pathways distal to the _1a-receptor subtype (11, 14, 18, 20, 21, 33). The mechanisms by which -adrenergic stimulation elicits a response are numerous, and perhaps as a consequence the relative importance of each mechanism remains unclear (11, 14). Specifically, -adrenergic-induced positive inotropy has been reported to be due to 1) a decrease in the transient outward K⁺ current (I_to), which elicits action potential prolongation and an increase in the slow inward Ca²⁺ current (I_Ca) (15, 16), 2) a directly stimulated increase in I_Ca (27), 3) elevated intracellular inositol 1,4,5-trisphosphate (IP₃) leading to increased sensitivity of sarcoplasmic reticular (SR) Ca²⁺ release channels (13, 21), 4) increased activity of protein kinase C leading to intracellular alkalization and increased myofilament Ca²⁺ sensitivity (7, 19, 22), 5) a pH-independent increase in myofilament Ca²⁺ sensitivity (37, 38, 47), and 6) myosin light-chain phosphorylation (9). Although the effects of -adrenergic stimulation are not typically reported to include an influence on mechanisms of cytosolic Ca²⁺ concentration ([Ca²⁺]_c) decline, namely SR Ca²⁺-ATPase, Na⁺/Ca²⁺ exchange, and sarcolemmal Ca²⁺-ATPase (11, 14), there have been reports that -adrenergic stimulation can decrease cAMP-dependent protein kinase activity via an increase in phosphodiesterase activity (1, 5, 6, 48), which could in turn suppress SR Ca²⁺-ATPase due to increased levels of unphosphorylated phospholamban (2).

Without focusing on any one mechanism by which cardiac -adrenergic responsiveness may be altered with exercise training, the present report describes characteristics of sustained positive inotropy elicited by phenylephrine (PE) exposure to isolated left ventricle (LV) cardiomyocytes and to isovolumic LV of sedentary (Sed) and run-trained (Tr) female rats. We found a diminished -adrenergic-stimulated positive inotropy after run training in measures of cardiomyocyte [Ca²⁺]_c, cardiomyocyte shortening, and isovolumic LV pressure development. These data demonstrated a training-induced suppression of the -adrenergic myocardial responsiveness of calcium regulatory mechanisms and contractile dynamics at both the cellular and whole organ levels.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal model. Female Sprague-Dawley rats were randomly assigned to a Sed group (n = 23) and an exercise Tr group (n = 22). Rats trained for a minimum of 20 wk that included a 12-wk phase during which running intensity and duration were gradually increased. By the end of the first 12 wk, rats were running 5 days/wk for 1 h/day up a 10% grade, and the daily training bout consisted of 15 min of running at 20 m/min, 30 min at 28 m/min, and 15 min at 35 m/min. At the time the rats were euthanized, Sed and Tr animals were age matched and were mature adults between 9 and 15 mo old. Immediately after each rat was killed, the adrenal glands and the spleen were dissected and weighed, tibial length was measured, and plantaris muscles were dissected, homogenized, and assayed for citrate synthase activity as previously described (43).

Cardiomyocyte isolation. Cardiomyocytes were obtained from the left ventricular free wall and septum from 17 Sed and 16 Tr rat hearts. All chemicals and reagents were obtained from Sigma (St. Louis, MO) except where noted. Rats were heparinized (250 U ip) and then anesthetized with pentobarbital sodium (35 mg/kg body wt ip; Abbott, North Chicago, IL). Hearts were rapidly excised and placed in ice-cold saline solution. The aorta was then cannulated, and the heart was retrogradely perfused using a modified Langendorff perfusion apparatus that could deliver three different solutions maintained at pH 7.4 and 37°C and bubbled with 95% O₂-5% CO₂ gas. The first solution was a bicarbonate-based modified Krebs-Henseleit buffer, the second solution was a nominally Ca²⁺-free Krebs-Henseleit buffer, and the third solution contained an additional 375 U/ml collagenase (Worthington, Freehold, NJ) and 420 U/ml hyaluronidase. Left ventricular and septal myocardium was minced, placed in a collagenase and hyaluronidase solution, and mechanically agitated. Isolated cardiomyocytes were suspended in bicarbonate-based medium 199, plated on laminin-coated glass coverslips, and incubated for 2-8 h at 37°C in a humidified 5% CO₂-balance room air atmosphere. One coverslip from each preparation was placed under a microscope, and images of all cardiomyocytes were recorded on video tape. These video images were examined for visual length and width using National Institutes of Health Image 1.41 video frame grabbing software.

Cardiomyocyte experimental protocol. Coverslips were incubated for 5 min in the presence of 0.05% vol DMSO-2 µM fura 2-AM (Molecular Probes, Eugene, OR). Each coverslip was removed from the fura 2 loading medium and used to form the bottom plate of a custom-built flow-through chamber (44). The chamber was placed on the stage of an inverted microscope (Nikon Diaphot) fitted with a 40× oil-immersion objective. Coverslips were superfused with a normal Tyrode solution (in mM, 140 NaCl, 6 KCl, 1 MgCl₂, 2 CaCl₂, 10 glucose, 2 pyruvate, 5 HEPES, pH 7.4) including 0.1 µM propranolol and maintained at 29 ± 0.2°C by conductive heating of superfusate as it passed through glass chambers heated by circulating water (Brinkmann Instruments, Westbury, NY). Cardiomyocytes were electrically paced via field stimulation using platinum electrodes with stimulus duration of 0.5 ms, voltage of 1.5 times stimulation threshold, and at a pacing frequency of 0.5 Hz (Grass Instruments, Quincy, MA).

Measurements of [Ca²⁺]_c dynamics. Fura 2 fluorescence was induced with a fluorescence microscopy system (IonOptix, Milton, MA) fitted with optical filters of 400 and 360 nm. This choice of filters takes advantage of a linear relationship between [Ca²⁺]_c and the fluorescence ratio (R) when an excitation wavelength over 390 nm is used (45). Fluorescence intensities were recorded as photon-counting rates using a personal computer with a sampling frequency of 200 Hz. The value for cardiomyocyte fluorescence background was determined for each cell by superfusion of Ca²⁺-free Tyrode and 1 µM digitonin for 4 min, which released cytosolic fura 2 and the subsequent measure of fluorescence with Ca²⁺-free Tyrode as superfusate.

(1)

Measurement of cardiomyocyte shortening dynamics. The positions of cardiomyocyte edges were determined using a video edge detection device (Crescent Electronics, Sandy, UT) and recorded using an A to D converter of the same personal computer that recorded fluorescence. Custom-made software was used to analyze the recorded cardiomyocyte shortening transients to determine the following characteristics: peak shortening expressed as a percentage of resting length, time to peak shortening, maximal shortening velocity, maximal shortening rate defined as maximal velocity/peak shortening, maximal relaxation velocity, maximal relaxation rate, and times to 25, 50, 75, and 90% relaxation.

Isovolumic heart experimental protocol. Isovolumic left ventricular pressure (LVP) was recorded from 6 Sed and 6 Tr rat hearts using methods previously reported by our laboratory (25). Rats were heparinized (250 U ip) and then anesthetized with pentobarbital sodium (35 mg/kg body wt ip; Abbott). Hearts were rapidly excised and placed in ice-cold saline solution. The aorta was cannulated, and the heart was retrogradely perfused at 85 mmHg with a Krebs-Henseleit-bicarbonate solution containing (in mM) 1.75 CaCl₂, 117.4 NaCl, 4.7 KCl, 1.2 MgSO₄, 1.2 KH₂PO₄, 24.7 NaHCO₃, 11.0 glucose, 5.0 pyruvate, 0.5 EDTA, and 0.1 µM propranolol and maintained at 37°C and pH 7.4.

Analysis. All statistical analyses were performed using SPSS v.6.1 (SPSS, Chicago, IL). Contrasts between characteristics of Sed and Tr groups were determined by unpaired, two-tailed t-tests. To test the relative sensitivity of cardiomyocytes of the training groups to -adrenergic stimulation, a 2 (Sed, Tr) × 2 (C, PE) × 4 (baseline, 1 min, 3 min, 5 min) repeated-measures ANOVA was performed on all characteristics of fluorescence ratio and shortening dynamics. From these analyses, a significant "PE × duration" interaction was taken to indicate a PE effect in a group-independent manner, and a significant "Tr × PE × duration" interaction indicated a differential response of the experimental groups (Sed vs. Tr) to PE. To test the relative sensitivity of isovolumic LVs of the experimental groups to -adrenergic stimulation, a 2 (Sed, Tr) × 4 (baseline, 1 min, 3 min, 5 min) repeated-measures ANOVA was performed on all LV pressure characteristics, and a significant "Tr × duration" interaction was used to indicate a differential response of the experimental groups to PE. All data are presented as means ± SE. To reduce the possibility of committing a type II interpretive error, i.e., a false negative, significance was considered at both the P  0.05 and P  0.10 levels per the principles described by Williams et al. (49).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal model. Training did not significantly affect body weight, tibial lengths, adrenal weights, or spleen weight in this study (Table 1), which is consistent with results of previous training studies using female Sprague-Dawley rats (28, 31, 34, 35). Cardiomyocyte length but not width was increased, and citrate synthase activity of the plantaris muscle homogenates was increased by run training (Table 1). Treadmill training also induced left ventricular hypertrophy, as indicated by the higher absolute and relative LV weights for the Tr group (Table 1) in the rats that were used in the isovolumic LV experiments. The combined results of LV hypertrophy, cardiomyocyte lengthening, and increased citrate synthase activity provide central and peripheral verification that our treadmill training protocol was effective in producing a trained state in this animal model as has been described previously (26, 28, 31, 34, 35).

Differentiating -adrenergic response from effects of experiment duration. Figure 1 presents representative fluorescence ratio and cardiomyocyte shortening transients for Sed cardiomyocytes at baseline and after 5 min under the control (C) and 5 µM PE conditions. As depicted in Fig. 1, peak [Ca²⁺]_c, indicated by R_peak and R_diff, increased with experiment duration under both the C and PE conditions. In addition, cardiomyocyte shortening and maximal shortening velocity increased with experiment duration under both the C and PE conditions. The inotropic effect of experiment duration on [Ca²⁺]_c and shortening dynamics under the C condition may reflect a gradually increasing intracellular calcium load during the course of these experiments, as reported previously (34). This observation underscores the importance of the C cardiomyocytes in differentiating the effects of PE from the inotropic effects of experiment duration alone.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Representative fluorescence ratio and shortening transients recorded before baseline and after 5 min during control (C) conditions and during exposure to 5 µM phenylephrine (PE). A: inotropic effects due to experiment duration alone were observed in C cardiomyocytes. B: inotropic effects were also observed after exposure to PE. Inotropic effects of PE on calcium regulation and contractile function were differentiated from inotropic effects of experiment duration alone by comparison between C and PE cardiomyocytes.

Training effect on -adrenergic response of cardiomyocyte [Ca²⁺]_c dynamics. Several variables describing [Ca²⁺]_c dynamics, specifically R_rest, R_peak, R_diff, time to R_peak, and k_rise, were not different between the Sed and Tr groups and demonstrated experiment duration effects under the C condition that were not different from those under the PE condition. This result was expected, as there have been reports that -adrenergic stimulation may affect cardiomyocyte [Ca²⁺]_c only subtly, occasionally, or after PE exposure greater than 5 min (8, 18, 19, 22, 47).

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Characteristics of cardiomyocyte cytosolic Ca2+ concentration ([Ca2+]c) in response to 5 µM PE. A: integral of fluorescence ratio transient (Rint) tended to rise with experiment duration, as illustrated with C cardiomyocytes. During PE exposure, Rint increased dramatically in sedentary (Sed) group, but was relatively unaffected in trained (Tr) group. These results for Rint, which represented effective [Ca2+]c available to the myofilaments during contraction, indicated that -adrenergic stimulation normally increases activation of myofilaments by [Ca2+]c, but not in trained state. B: rate of [Ca2+]c decline (kfall) was differentially affected by PE in Sed and Tr groups. Sed group clearly experienced a decrease in kfall with PE exposure, whereas kfall of Tr group was either unaffected or increased with PE. * P < 0.05 compared with Sed.

Training effect on -adrenergic response of cardiomyocyte shortening dynamics. A strong positive inotropy was elicited by PE in these isolated cardiomyocytes as demonstrated by increased peak shortening. After 5-min exposure to 5 µM PE, peak shortening increased by ~65% in the Sed group and by ~50% in the Tr group (Fig. 3A). The significant PE × duration interaction indicated that -adrenergic stimulation of the PE cardiomyocytes induced a greater increase in peak shortening than experiment duration alone in the C cardiomyocytes. This increase in cardiomyocyte shortening due to -adrenergic stimulation has been similarly observed by others (8, 15, 18, 19, 47). Moreover, in the present study there was a Tr × PE × duration interaction (P = 0.098), indicating that the Sed group was more sensitive to positive inotropy elicited by PE than the Tr group. These interactions can be visualized in Fig. 3A, where the increase in peak shortening between baseline and 5 min was greater under the PE condition than under the C condition, but was also more dramatic in the Sed compared with Tr group. These results are the first of their kind to suggest that positive inotropy due to PE is reduced by run training in the rat LV cardiomyocyte.

View larger version (33K): [in this window] [in a new window]   Fig. 3.   Characteristics of cardiomyocyte shortening in response to 5 µM PE. A: peak shortening increased with experiment duration in both Sed and Tr groups, as seen for C cardiomyocytes. Peak shortening was strongly influenced by PE exposure in both experimental groups and more so in Sed than Tr. B: time to peak shortening was significantly reduced by exposure to PE but was not differentially reduced between Sed and Tr groups. C: maximal rate of shortening, which normalizes the maximum shortening velocity relative to peak shortening, significantly increased in Sed group after PE exposure and was less affected in Tr group. D: maximal relaxation rate, as well as times to 25, 50, 75, and 90% recovery, were unaffected by PE. These latter results are consistent with current understanding that -adrenergic stimulation does not significantly influence myocardial relaxation.

Training effect on -adrenergic response of isovolumic heart pressure dynamics. The LV pressure tracings recorded during the isovolumic LV experiments demonstrated significant positive inotropy induced by PE (Fig. 4) as reported by others (17, 25). As illustrated in Fig. 4 and summarized in Table 2, devLVP generally increased and the time to peak pressure decreased with PE exposure in both the Sed and Tr groups. Although there was no observed differential increase in devLVP between the Sed and Tr groups in response to PE, there was a significant Tr × duration interaction for time to peak pressure (Table 2), providing a strong indication that the Sed LV was more sensitive to PE than Tr.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Representative left ventricular pressure (LVP) tracings from isovolumic left ventricle in response to 5 µM PE. A: LVP tracings from Sed rat heart demonstrated significant positive inotropy induced by PE. B: LVP tracings from Tr rat heart also demonstrated significant positive inotropy induced by PE. Comparisons of time to peak pressure indicated that hearts of Tr group were less sensitive to inotropic effects of PE than Sed.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present study demonstrated that chronic run training of the female rat suppressed the sustained positive inotropic response to -adrenergic stimulation in isolated left ventricular cardiomyocytes and isovolumic LV. We specifically observed the following positive inotropic effects elicited by PE on Sed rat left ventricular cardiomyocytes: 1) increased effective [Ca²⁺]_c available during contraction, i.e., R_int, 2) decreased rate of [Ca²⁺]_c decline, i.e., k_fall, 3) increased peak shortening, and 4) increased maximal rate of shortening. In contrast, PE exposure of trained rat cardiomyocytes to PE did not induce an increase in the effective [Ca²⁺]_c available for contraction, R_int, nor a decrease in the rate of [Ca²⁺]_c decline, k_fall. Furthermore, PE elicited a diminished positive inotropy in the trained rat myocardium, as demonstrated by peak shortening and maximal shortening rate of cardiomyocytes and by time to peak pressure and (+dP/dt)_max of isovolumic LVs.

The acute effects of -adrenergic stimulation reportedly include a very brief (<1 min) positive inotropy followed by an otherwise brief (<2 min) negative inotropy (11, 14, 18, 33). We did not observe these acute effects due to -adrenergic stimulation, possibly because our temporal resolution (1 observation per 2 min) was not designed to detect such brief and subtle occurrences. Therefore, the present study provided no insights into the possible effects of run training on the acute inotropic effects of -adrenergic stimulation on the myocardium.

Our results do suggest that the sustained -adrenergic effects on calcium regulation and contractile function of left ventricular myocardium are suppressed by chronic run training. The putative sustained effects of -adrenergic stimulation on calcium regulation may include, but are not necessarily limited to, 1) reduction in I_to leading to increased I_Ca (15, 16), 2) direct increase in I_Ca (27), 3) increased intracellular content of IP₃ (3, 13, 21, 23), and 4) a downregulation of cAMP-dependent processes (1, 5, 6, 48). We do not report here nor know of any direct evidence suggesting that training alters the otherwise normal effects of -adrenergic stimulation on I_to, I_Ca, or IP₃. Therefore, inferring from our data that -adrenergic responses of these specific mechanisms are suppressed by training would be purely speculative at this time.

However, we do have direct evidence that the rate of [Ca²⁺]_c decline, which is dominated by SR Ca²⁺ uptake, is normally reduced in LV cardiomyocytes exposed to PE (Fig. 2B). Because SR Ca²⁺-ATPase activity is known to be suppressed by unphosphorylated phospholamban (2), and -adrenergic stimulation has been shown to increase phosphodiesterase activity and reduce cellular cAMP (1, 5, 6, 48), it would be reasonable to infer that we observed a decreased rate of SR Ca²⁺ uptake induced by -adrenergic stimulation of phosphodiesterase activity in our cardiomyocyte model. The absence of this effect in cardiomyocytes of run-trained rats implies that chronic exercise training may lead to a decrease in the ability of the LV cardiomyocyte to increase phosphodiesterase activity after -adrenergic stimulation and/or that phosphorylated phospholamban content may be intrinsically reduced with chronic exercise training and to such a degree as to be relatively unaffected by an increase in phosphodiesterase activity. It is also possible that an increase in myofilament sensitivity to Ca²⁺ (11, 14) may have reduced the rate of [Ca²⁺]_c decline in the Sed state, but not in the Tr state. We conclude that the training-induced suppression of the -adrenergic response in k_fall, like other variables examined in this study, must be due to an intrinsic downregulation of the -adrenergic signaling pathway in the present training model, as has been reported for swim-trained rats (50).

We observed substantial positive inotropy due to -adrenergic stimulation in our isolated cardiomyocytes and a relatively diminished response due to training. Although this phenomenon may have been due in part to changes in calcium regulation, it has been proposed that myofilament Ca²⁺ sensitivity increases with -adrenergic stimulation (11, 14), although the manner by which this effect is elicited may be due to intracellular alkalization (7, 19, 22), myosin light-chain phosphorylation (9), and/or by other undetermined means (37, 38, 47). Whereas the present study provides no direct evidence to suggest the manner by which an increase in myofilament Ca²⁺ sensitivity occurred, cardiomyocytes of the run-trained rats did respond to PE exposure with increased contractile function without significant changes in calcium regulation. We therefore conclude that an increase in myofilament Ca²⁺ sensitivity was most likely responsible for the positive inotropy induced in the cardiomyocytes of the trained rats, and we assume that a similar increase in myofilament Ca²⁺ sensitivity also occurred in the cardiomyocytes of the Sed rats.

We report here a relatively diminished -adrenergic responsiveness of isovolumic LV pressure characteristics after run training, most notably in measures of time to peak LV pressure. These results of isovolumic LV contractile function in response to PE mimic those of the LV cardiomyocytes and therefore we must conclude that run training suppressed cardiac -adrenergic responsiveness at the cellular and whole organ level in this rat model of exercise. The results, however, contradict our earlier finding that run training enhanced -adrenergic responsiveness of isovolumic LV (25). One plausible explanation for the discrepancy between the two studies is the present use of a female Sprague-Dawley rat and the previous use of a male Fischer 344 (25). It has been reported previously that run training of male rats will induce an increase in cardiac function not observed in female rats (39), and cardiac adaptations to exercise training have been suggested to be gender specific (30). Instead of trying to reconcile the present isovolumic LV results for female Sprague-Dawley rats with those of male Fischer 344 rats (25), we will confine our interpretation of the present isovolumic LV data within the context of our cardiomyocyte data and conclude only that the suppressed -adrenergic response at the cardiomyocyte level was reflected at the whole organ level in the present model of run training.

It is interesting to note that after -adrenergic stimulation in the Sed group, we observed a decreased rate of [Ca²⁺]_c decline in cardiomyocytes and yet observed a possible increased rate of isovolumic LV pressure decline and vice versa in the Tr group. These seemingly paradoxical results may not be suitable for direct comparison. Although the prolongation in the [Ca²⁺]_c transient due to -adrenergic stimulation in the Sed group reflects a slowed myofilament deactivation, this prolongation was relatively small (<10%) and may not be reflected directly in mechanical relaxation. This was the case for the unloaded cardiomyocytes (Fig. 3D), whose relaxation was dependent on myofilament deactivation as well as on the restoring forces of intrinsic elastic elements. In the case of the isovolumic LV (Table 2), the rate of pressure decline was additionally dependent on elastic and viscous elements of the LV that were not influenced by [Ca²⁺]_c at all. It is likely that run training changed the elastic and viscous characteristics of the cardiomyocytes and/or other elements of the LV (51) and thereby influenced relaxation in a manner not predictable by [Ca²⁺]_c dynamics alone.

In conclusion, our animal model of run training clearly acquired a suppressed cardiac -adrenergic responsiveness that was pervasive to the effects of -adrenergic stimulation on calcium regulatory mechanisms in the cardiomyocyte. Although laboratory treadmill training has been implicated as a stressful training protocol, which could induce regular elevated plasma catecholamine levels that may subsequently desensitize the heart to adrenergic stimulation (24, 41), we found no change in adrenal or spleen weights due to treadmill training. Therefore the stress associated with treadmill running was not morphologically significant in our animal model, and we presume that the decreased cardiac -adrenergic responsiveness was a characteristic of the trained state. The current study does not explain the specific -adrenergic pathways or mechanisms that may have been modulated by run training. Although we have inferred from our data that run training induced a downregulation of the normal -adrenergic increase in phosphodiesterase activity, we have presented no direct evidence for this effect and there are several other pathways and mechanisms that may have been altered by training. It would therefore be valuable to focus future studies on investigating specific -adrenergic pathways and affected mechanisms that may become downregulated with run training. In addition, future studies will have to pay close attention to the role of gender and/or strain of animal models while investigating the effects of run training on cardiac functions, including catecholamine responsiveness.