INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

THE APPLICATION OF technology previously used in larger mammals has allowed a growing inventory of tools to become available for the phenotypic assessment of murine models of cardiovascular disease. The ability presently exists to evaluate murine myocyte contractility (23), ex vivo whole heart work performance (3), and in vivo ventricular function using two-dimensional and Doppler echocardiography (4, 21), angiocardiography (14), and, more recently, calcium transients in beating mouse hearts (5). Although these methods are useful for evaluating the heart under resting conditions, cardiovascular stress assessment is limited to pharmacological manipulations with these systems, which may not fully reproduce the maximal stress achieved with exercise in the intact, awake animal. In larger mammals with cardiac disease, the gold standard for functional assessment of the intact cardiovascular system is graded treadmill exercise (GTE) (6, 24). Recently, we have demonstrated the ability to perform cardiovascular and metabolic evaluations in awake, unrestrained mice during the physiological stress of GTE (2). In the present study, we used these techniques to evaluate cardiovascular performance in a genetically altered model, the phospholamban (PLB)-deficient mouse, with previously demonstrated elevation of basal cardiac function and attenuated inotropic and lusitropic responses to -adrenergic stimulation.

PLB regulates calcium reuptake into the cardiac sarcoplasmic reticulum and has been implicated in the contractile response to -adrenergic receptor (-AR) stimulation (9). Protein kinases, activated by -AR stimulation, phosphorylate PLB, removing tonic inhibition of the sarcoplasmic reticulum calcium ATPase, enhancing calcium reuptake into the sarcoplasmic reticulum, and resulting in enhanced myocardial contractility and relaxation. Studies using whole hearts (8, 12) or isolated myocytes (23) have shown that targeted disruption of the PLB gene (PLB/) enhances cardiac contractility and relaxation and attenuates the stimulatory effects of -AR agonists compared with wild types (PLB+/+). Similar findings were observed in vivo when ventricular function was assessed using echocardiography under anesthesia (7). These experimental systems suggest not only enhanced resting cardiovascular function but also an inability to improve cardiac performance during stress. Extrapolation of these findings to the awake animal, however, may be difficult due to the influence of anesthetic agents and to the inability to accurately simulate cardiovascular stress in such systems. To evaluate the effects of disrupting the PLB gene in the awake, active state, we first studied cardiovascular parameters in nonanesthetized, nonrestrained mice at rest and in response to -AR agonist stimulation. To determine whether loss of PLB results in decreased exercise performance due to loss of adrenergically mediated cardiac reserve, we evaluated cardiovascular performance in PLB-deficient mice during GTE.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animal subjects. PLB-deficient mice were derived from 129sv/J and CF-1 background strains using homologous recombination in murine embryonic stem cells. Disruption of the PLB gene was produced by replacing a 1.8-kb region of exon 2, which includes the entire protein coding sequence, with a neomycin resistance gene cassette. The deletion of PLB in these mice has been previously demonstrated by loss of the gene using Southern analysis and loss of the protein using Western analysis (12). Thirteen adult PLB/ mice and eleven PLB+/+ age-matched littermates between the ages of 10 and 12 wk were used for this study. There were no significant differences in sex distribution or weight between the groups (Table 1). All mice were kept in standard rodent cages with food and water ad libitum in a 12:12-h light-dark cycle.

Vessel cannulation. Mice were anesthetized with 1.5% inhaled isoflurane. Induction was performed in a closed chamber and was maintained via a nose cone throughout surgery. Carotid arterial cannulation with polyethylene (PE-10) tubing was performed as previously described (2). Intravenous ampicillin (100 mg/kg) was administered immediately after skin closure, and recovery was allowed for 24 h. All experimental protocols were approved by the Administrative Panel on Laboratory Animal Care at Stanford University and are in conformity with the "Guiding Principles for Research Involving Animals and Human Beings."

Data acquisition. After the arterial catheter was removed from a subcutaneous pocket, the catheter was connected to a Spectramed DTX Plus pressure transducer (Oxnard, CA) and analog inputs were amplified using a Gould preamplifier (model 11-1202-25; Cleveland, OH) and amplifier (model 13-4615-52). Analog signals were digitized using a Data Translation Series analog-to-digital converter (DT2801; Marlboro, MA) and were analyzed and stored using Dataflow data-acquisition software (Crystal Biotech, Hopkinton, MA). Because of the low-frequency response demonstrated previously with this fluid-filled catheter system (2), we elected to utilize only mean, rather than phasic, blood pressure (BP) recordings, although all pressure tracings clearly demonstrated a phasic waveform.

Response to exercise. After acclimation for 1 h and after stable baseline HR, BP, O₂, and CO₂ measurements were recorded, 2.5 m/min incremental increases in treadmill belt speed and 2° increments in angle of inclination were made every 3 min until the mouse exhibited signs of exhaustion. Exhaustion was defined as the mouse spending >50% of the time or >15 consecutive seconds on the shock grid. Exercise capacity was defined as time to exhaustion during GTE. To assess exercise differences between the genotypes, mean values for each parameter were calculated at baseline, at each exercise workload, and at peak exercise. The percent changes between resting and peak exercise values were also calculated. To indirectly assess ventricular stroke volumes during exercise, we calculated mean oxygen pulse values (O₂/HR) at each workload and performed linear regression analysis of oxygen pulse versus workload.

Response to -adrenergic agonists. After stable baseline HR and BP measurements were recorded, 1 µg/kg intra-arterial boluses of isoproterenol were administered in volumes of 0.03 ml of normal saline. This dose has been previously shown to produce maximal HR and BP responses in mice (2). Continuous recordings were performed for 30 min after the bolus or until baseline values were reestablished.

Statistical analyses. Resting and maximal exercise parameters were compared between the two genotypes by using Student's t-test. The percent change in a given parameter from rest to maximal exercise was also compared between genotypes by using a Student's t-test. Values of a parameter during GTE were compared between genotypes by using both Student's t-test at each workload and a two-way ANOVA for repeated measures for the entire GTE protocol. To determine whether significant changes occurred during exercise for a specified parameter, values for that parameter at each workload were compared with the resting value by using a one-way ANOVA with a post hoc Dunnett's test. To compare the oxygen pulse curves during exercise, average slopes and intercepts derived from the linear regression analyses were compared between genotypes using a Student's t-test. Data from -AR agonist studies were analyzed to establish maximal HR elevation and maximal decrease in BP. Percent changes from resting values were compared between groups using Student's t-test. All values are reported as means ± SD. Statistical significance was defined as a P value < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Surgical results. Overall survival through surgery and the 24-h recovery period was 92%. There was a 100% catheter patency rate among surviving animals after recovery, leaving 9 PLB+/+ mice and 13 PLB/ mice for resting studies. Two knockout animals had late catheter failures and were unable to be evaluated during GTE, leaving 9 PLB+/+ mice and 11 PLB/ mice for the exercise and -AR agonist studies. The overall success rate for completion of the study was 83%.

Resting parameters. Resting, nonanesthetized, nonrestrained cardiovascular and metabolic parameters for both groups are shown in Table 1. PLB/ mice were hypertensive compared with wild-type littermates (125 ± 19 vs. 109 ± 16 mmHg, P < 0.05). No differences were found in resting HR, O₂, CO_2, or RER between the groups.

Graded treadmill exercise. At maximal exercise, values for all parameters were nearly identical between PLB-deficient and wild-type mice (Table 1). Exercise capacity was also similar between the two groups, as measured by O₂ at peak exercise and the duration of exercise protocol performed (Fig. 1). With the exception of BP, both groups exhibited typical murine cardiovascular and metabolic responses to incremental treadmill exercise (2), demonstrating linear increases in HR, O₂, CO₂, and RER with increasing workload (Fig. 1). PLB+/+ mice maintained a steady mean BP during exercise, whereas PLB/ mice were hypertensive at the onset of exercise, with a subsequent fall in mean BP at maximal exercise (P < 0.05 by Dunnett's test between resting value and value at 27.5 m/min). Thus PLB/ mice had a significantly different percent change in mean BP with exercise between genotypes (P < 0.05 by Student's t-test) (Table 1). All measured parameters returned to baseline values within 15 min of cessation of exercise.

View larger version (30K): [in this window] [in a new window]   Fig. 1.   Cardiovascular (A and B) and metabolic (D-F) responses to graded treadmill exercise (GTE) from baseline to 27.5 m/min belt speed and 22° slope. Note linear increases in heart rate (HR), oxygen consumption (O2), and carbon dioxide production (CO2) with exercise for PLB+/+ () and PLB/ () mice and a drop in mean blood pressure (BP) with exercise in PLB/ mice. Duration of exercise (C) is not different between groups. Values represent means ± SD. RER, respiratory equivalent ratio.

1

1

1

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Oxygen pulse curves during GTE. Although there is a trend toward a steeper slope in PLB/ mice (), there is no significant difference compared with PLB+/+ mice (). Values represent means ± SD.

Isoproterenol response. Administration of the -AR agonist isoproterenol resulted in expected tachycardic and hypotensive responses within 1 min of injection in both PLB/ and wild-type mice (Fig. 3). No differences were found in maximally stimulated HR (691 ± 15 vs. 684 ± 17 beats/min, respectively) or mean BP (90.7 ± 8.1 vs. 83.1 ± 4.5 mmHg, respectively) between the two groups. Percent changes from basal values in response to isoproterenol were also not different between PLB/ and PLB+/+ mice (21.4 ± 2.0 and 26.2 ± 2.4% for HR and 21.7 ± 3.1 and 22.1 ± 1.5 for BP, respectively) (Fig. 4).

View larger version (30K): [in this window] [in a new window]   Fig. 3.   Representative HR (diamonds) and mean BP (squares) responses to a single intra-arterial bolus of 1 µg/kg isoproterenol for PLB/ (A) and PLB+/+ (B) mice. Note rapid responses of both parameters within 1 min after administration and return to baseline before 10 min.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Isoproterenol-stimulated changes in HR and mean BP for PLB+/+ and PLB/ mice. Columns and error bars represent means ± SD.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

This study provides the first evaluation of the role of PLB in the cardiovascular and metabolic responses of the intact, awake animal at rest and during exercise. Ablation of PLB has been shown to improve cardiac contractility and relaxation in vitro (8, 12, 23) and in vivo (7). Although basal cardiac function is enhanced, cardiac reserve appears to be diminished as evidenced by an attenuated response to -adrenergic stimulation. These studies, however, utilize techniques that do not necessarily reproduce the physiological responses of the intact, nonanesthetized organism and cannot assess the cardiovascular response to severe physiological stress such as that provided by exercise. Our primary aim, therefore, was to determine whether the in vitro enhancements of basal cardiac function demonstrated in PLB-deficient mice would confer a cardiovascular advantage in vivo or whether decreased adrenergic reserve would lead to impaired function during cardiovascular stress.

GTE provides one of the most severe physiological stresses to the cardiovascular system and has been extensively used to quantitatively evaluate cardiovascular fitness and disease (13, 19, 20, 22). We have previously shown that treadmill exercise in mice produces quantifiable and reproducible changes in cardiovascular and metabolic indexes similar to those in humans and other species (2). With the use of our established exercise protocol, PLB-deficient mice achieve normal peak and submaximal exercise values for all measured parameters, indicating that, during severe exercise stress, their overall cardiovascular and metabolic physiology is not significantly different from that of controls. The fact that they reach the same peak exercise workload and peak O₂ indicates that their exercise capacities are also normal. These results are somewhat surprising, because PLB-deficient mice show attenuated cardiac contractile responses to -AR stimulation. Thus one might expect that this reduction in adrenergic reserve would lead to a limited cardiovascular response to physiological stress such as exercise, during which adrenergic stimulation plays a major role. However, as we have recently demonstrated in ₁-adrenergic receptor knockout mice, a significant attenuation of the chronotropic response to exercise does not reduce total exercise capacity (15). Our findings suggest, similarly, that the majority of the PLB deficiency-induced alterations in myocardial contractility and reserve are adequately counterbalanced by other regulatory pathways, most likely those involving regulation of ventricular preload and afterload and oxygen extraction.

In our discussion of peak O₂, it is important to emphasize its distinction from maximum O₂ (O_{2 max}). In a review of the literature, we found that O_{2 max} is very difficult to establish reproducibly in mice using treadmill exercise (1). Thus we make no assertion of establishing O_{2 max} in this study, and we cannot deny the possibility that differences in O_{2 max} values may exist for PLB-deficient mice that our study would not have been able to detect. We have demonstrated, however, that the exercise workloads used in our protocol induce a quantifiable and reproducible level of cardiovascular stress (2) and that peak O₂ and peak exercise values established using our protocol are useful for evaluating cardiac reserve in genetically altered mice (15).

The ratio of O₂ to HR, or oxygen pulse, can be used as a measure of HR-independent cardiovascular performance during exercise (22). The normal oxygen pulse values of PLB-deficient mice indicate a normal increase in O₂ for a given increase in HR during exercise. Because PLB-deficient mice are unable to increase cardiac contractility both in vitro and in vivo, there must be alternative compensatory mechanisms to allow normal cardiac performance. A rearrangement of the Fick equation directly relates the oxygen pulse to left ventricular stroke volume (SV) and oxygen extraction (Ca_O2  Cv_O2), suggesting two possible mechanisms for maintaining cardiac performance during exercise: increasing stroke volume and/or increasing oxygen extraction. We have previously demonstrated that oxygen extraction plays a role in augmenting murine cardiac output by measuring an ~30% increase in oxygen extraction during exercise in wild-type mice (15). This is unlikely to be the sole mechanism, however, because in these same mice the oxygen pulse increased by ~55%. Thus increases in stroke volume must also occur during exercise. Therefore, it is possible that PLB-deficient mice compensate for an inability to increase contractility by enhancing oxygen extraction and/or by increasing stroke volume.

By what mechanism is the latter achieved? In mammalian hearts, stroke volume can be enhanced by 1) improving ventricular contractility, 2) reducing afterload, such as by lowering systemic vascular resistance, or 3) enhancing preload, such as by improving ventricular filling and end-diastolic pressure (19). It is unlikely that PLB-deficient mice significantly increase intrinsic ventricular contractility during exercise, as suggested by previous in vitro and in vivo studies (7, 12). A reduction in ventricular afterload, however, is suggested by the drop in mean BP that PLB-deficient mice demonstrate during exercise. Furthermore, echocardiographic studies have shown enhanced ventricular filling velocities in PLB-deficient mice, suggesting improved ventricular preload (7). Both of the latter mechanisms may allow PLB-deficient mice to increase their stroke volumes during exercise and compensate for their inability to enhance myocardial contractility. Thus we have evidence suggesting that compensatory changes in stroke volume do occur in PLB-deficient mice during exercise; however, we cannot rule out the possibility that increases in both stroke volume and oxygen extraction are involved in maintaining normal exercise capacities.

PLB-deficient mice have higher resting BPs than their wild-type littermate controls. Although basal hypertension has not been previously reported in awake mice, it has been observed in anesthetized animals (11). An elevation of resting mean BP may be the result of enhanced basal myocardial contractility, previously demonstrated in PLB-deficient hearts, or of alterations in basal peripheral vascular tone.

The significant drop in BP seen in PLB-deficient mice in response to exercise may simply reflect their elevated resting BPs. Other factors, however, may also play a role. Exercise in larger mammals results in a redistribution of blood flow away from vasoconstricting splanchnic beds and toward vasodilated skin and active muscle beds (16-18). Lack of PLB may preferentially attenuate the response to sympathetically mediated vasoconstriction during exercise while the locally mediated vasodilatory responses to acidosis, hypoxia, and hyperkalemia in working muscle and epidermal beds remain intact, resulting in a reduction in total vascular resistance and a drop in mean BP during prolonged exercise. Indeed, PLB-deficient aortic rings show a rightward shift in the force-concentration curves for the -adrenergic agonist phenylephrine, confirming a decreased sensitivity to vasoconstrictive stimuli (10).

A normal chronotropic response to -AR agonists has been previously demonstrated in PLB-deficient mice in vitro and in vivo under anesthesia (7, 12). We confirm the normal HR response in the absence of anesthesia and show a normal BP response as well. This suggests that the effects of pharmacological -adrenergic stimulation on cardiac chronotropy and mean systemic arterial pressure remain intact in PLB-deficient mice and must involve effector systems other than those mediated by PLB. This does not contradict our findings of altered resting and exercising BPs in PLB-deficient mice, however, because BP homeostasis involves mechanisms more complex than purely -adrenergic stimulation.

In summary, we find that despite subtle alterations in cardiovascular performance during exercise, the exercise capacities of PLB-deficient mice are not different from wild-type mice, suggesting that attenuated adrenergic reserve does not compromise overall cardiovascular performance. Alternative regulatory pathways appear to adequately compensate for any deficiency that a lack of PLB imposes on the normal cardiovascular system during exercise. This does not imply that PLB inhibition could not be advantageous in the presence of cardiac pathology. PLB inhibition, much like -adrenergic stimulation, may have a profound beneficial effect on exercise capacity in animals with failing hearts and diminished cardiovascular reserve that would not be apparent in animals with normal hearts.

This study also demonstrates the importance of examining cardiovascular regulation in an intact animal both at rest and during physiological stress such as that provided by treadmill exercise. Such methodology is critical for evaluating cardiovascular phenotypes resulting from manipulations of the murine genome and, more importantly, for determining their relevance to whole body function.