INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

THE PHYSIOLOGICAL FUNCTIONS of angiotensin II (ANG II) in the cardiovascular system are mediated mainly through the ANG II type 1 receptor (AT1-R) (7, 27), which is present in both rat and human hearts (21, 24) and shows little interspecies difference in drug affinity (14). Chronic AT1-R antagonism improves hemodynamics (28) and limits cardiac hypertrophy after myocardial infarction (23). Recently, chronic treatment with the selective AT1-R antagonist TCV-116 for 1 wk before ischemia-reperfusion (I/R) was found to improve left ventricular (LV) contractility in nonworking Langendorff hearts, suggesting that endogenous ANG II might be deleterious (30) and acute AT1-R antagonism during I/R might improve recovery of mechanical function. Because a variety of agents that decrease the rate of proton (H⁺) production from glucose metabolism (8-10) also improve recovery of mechanical function in the postischemic heart, blockade of endogenous ANG II by an AT1-R antagonist might also be expected to decrease H⁺ production. Evidence suggests that stimulation of glucose oxidation and/or inhibition of glycolysis decreases proton production (9) thereby reducing intracellular acidosis, Na⁺/H⁺ exchange, intracellular Na⁺ accumulation, Na⁺/Ca²⁺ exchange (9), intracellular Ca²⁺ overload, and I/R injury (25). Linkages among a reduction in proton production, activation of protein kinase C (PKC), and cardioprotection have been demonstrated with adenosine A₁ receptor stimulation (8, 13) and ischemic preconditioning (1, 17). ANG II, which mimics ischemic preconditioning and stimulates PKC (19), might also be expected to reduce glucose-derived proton production. However, ANG II-induced AT1-R stimulation also activates Na⁺/H⁺ exchange (16), which might exacerbate I/R injury (25).

The aim of the study was to determine whether intrinsic AT1-R stimulation plays a role in the recovery of mechanical function after I/R in isolated perfused working rat hearts. We compared the effects of acute antagonism of endogenous ANG II using the selective AT1-R antagonist losartan (14, 28) to those of the selective adenosine A₁ receptor agonist N⁶-cyclohexyladenosine (CHA), which is known to protect against I/R injury by inhibiting proton production derived from glucose metabolism (10).

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Experimental animals and isolated working heart preparations. All animals were housed and treated in accordance with guidelines of the Canadian Council on Animal Care and the American Physiological Society. Male Sprague-Dawley rats (250-350 g), which had been fed ad libitum, were anesthetized with pentobarbital sodium. Hearts were rapidly excised and placed in ice-cold Krebs-Henseleit solution. After we cannulated the aorta, a brief Langendorff perfusion was commenced using Krebs-Henseleit buffer (pH 7.4, gassed with 95% O₂-5% CO₂) at a hydrostatic pressure of 60 mmHg. Extraneous tissue was removed, and the pulmonary artery and left atria were cannulated. After an initial 10-min Langendorff perfusion was performed, we switched the hearts into the working mode by clamping the line from the reservoir and opening the left atrial line using the method of Neely et al. (20). During aerobic perfusion, atrial pacing was applied to the hearts at 300 beats/min (Grass S88 stimulator). Working hearts (5, 12) were perfused in a closed recirculating system at 37°C using an oxygenator with a large surface area in constant contact with a 95% O₂-5% CO₂ gas mixture (Fig. 1). The perfusate (100 ml) consisted of a modified Krebs-Henseleit buffer containing 2.5 mmol/l CaCl₂, 11 mmol/l glucose, 1.2 mmol/l palmitate prebound to 3% bovine serum albumin (BSA, Fraction V), and 100 mU/l insulin. Perfusions were performed at a constant hydrostatic left atrial pressure of 11.5 mmHg (index of preload) and an aortic hydrostatic pressure of 80 mmHg (index of afterload). The O₂ content of the coronary effluent was measured continuously using a probe (YSI 5331, Yellow Spring Instruments) placed in the pulmonary artery outflow line, which was connected to an O₂ meter (YSI 5300). Coronary effluent drained back into the buffer reservoir.

View larger version (32K): [in this window] [in a new window]   Fig. 1.   Schematic of closed, recirculating working heart perfusion apparatus. Preload pressure was set at 11.5 mmHg and afterload pressure at 80 mmHg.

Heart rate and systolic and diastolic pressures (P23 Db; Gould) were recorded on a Grass model 7D polygraph. Cardiac output and aortic flow were measured using ultrasonic flow probes (Transonic T206) placed in the left atrial and aortic lines, respectively. Coronary flow was calculated as the difference between cardiac output and aortic flow. Mechanical function was measured as LV minute work, which was calculated as (systolic pressure  left atrial pressure) × cardiac output × 0.133 J. Myocardial efficiency was calculated as LV work expressed as a percentage of total potential work based on oxygen consumption (MO₂, µmol · min¹ · g dry wt¹) calculated as moles of oxygen per coronary flow. Coronary vascular conductance (CVC, ml · min¹ · mmHg¹) was calculated as the ratio of coronary flow and mean aortic pressure.

Biochemical assays. Glycolysis and glucose oxidation (6, 8) were measured simultaneously by quantitative collection of ³H₂O (liberated at the enolase step of glycolysis) and ¹⁴CO₂ (liberated at the level of pyruvate dehydrogenase complex and in the citric acid cycle) from perfusion buffer containing tracer quantities of [5-³H]glucose and [U-¹⁴C]glucose. Samples of perfusate were taken at 10-min intervals during aerobic perfusion and stored under liquid paraffin for determination of metabolic rates. To measure glycolysis, ³H₂O in perfusate samples was separated from [³H]glucose and [¹⁴C]glucose using columns containing Dowex 1-X4 anion exchange gel, as described previously (8). Glycolytic rates are expressed as micromoles of metabolized glucose per minute per gram dry weight. The closed perfusion system allowed the collection of gaseous ¹⁴CO₂ under a hyamine trap (40 ml). The hyamine reservoir was sampled at the same time as the perfusate samples. The ¹⁴CO₂ trapped as bicarbonate in the perfusate and ¹⁴CO₂ trapped in the hyamine were measured and glucose oxidation rates (µmol metabolized glucose · min¹ · g dry wt¹) were calculated as previously described (8-10). Proton production from glucose metabolism was calculated as follows. When the rates of glycolysis and glucose oxidation are identical, the net production of protons is zero. However, when the rate of glycolysis exceeds that of glucose oxidation, two protons are produced for every molecule of glucose that passes through glycolysis but is not oxidized (6). Therefore, the rate of proton production attributable to hydrolysis of ATP arising from glucose metabolism is estimated as 2 × (rate of glycolysis  rate of glucose oxidation).

Experimental protocols. Hearts were randomly assigned to control (untreated) or one of nine drug treatment groups as follows: 0.5 µmol/l CHA, 1 nmol/l angiotensin II (ANG II), 0.1 µmol/l losartan, 1 µmol/l losartan, 1 µmol/l losartan + ANG II (0.1, 1, 10, or 100 µmol/l), and 1 µmol/l losartan + 1 µmol/l PD-123,319. The concentration of CHA used was previously shown to enhance recovery of mechanical function after ischemia (10). The concentration of PD-123,319 used is 100-fold higher than its dissociation constant (K_D) for AT2 receptors as determined by binding studies (7). All hearts were subjected to an I/R protocol (Fig. 2). Hearts were prepared during an initial 10-min Langendorff perfusion and were switched to a working mode. This consisted of a 50-min period of aerobic baseline perfusion followed by 30 min of global, no-flow ischemia and aerobic reperfusion for an additional 30 min. Rates of glucose metabolism were measured in selected hearts assigned to control, CHA (0.5 µmol/l), ANG II (1 nmol/l), losartan (1 µmol/l), and losartan (1 µmol/l) + ANG II (1 nmol/l). Hearts in these groups were also subjected to an aerobic protocol (Fig. 2). After initial preparation during Langendorff perfusion, hearts were switched to a working mode for 45 min of aerobic baseline followed by a 35-min treatment period during which the effects of the various drugs and drug combinations were assessed. In both the I/R and aerobic protocols, mechanical function and glucose metabolism were measured at 10-min intervals during periods of aerobic baseline, reperfusion, and treatment periods. At the end of both perfusion protocols, the ventricles were frozen with Wollenberger clamps cooled to the temperature of liquid N₂ for the determination of total dry weight. For both protocols, glucose metabolism is reported as an average of 10-min rates during each perfusion period.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Ischemia-reperfusion (I/R; A) and aerobic perfusion (B) protocols. For both protocols, hearts were prepared (Prep) during a 10-min Langendorff perfusion. Hearts were then switched into working mode and perfused aerobically for 50 min (Baseline). Hearts assigned to I/R protocol were then subjected to 30 min of global ischemia (Ischemia) followed by 30 min of aerobic reperfusion (Reperfusion). Hearts assigned to the aerobic protocol underwent aerobic perfusion for 45 min (Baseline) followed by an additional 35 min (Treatment). Drugs were added as indicated (arrows).

Statistics. Data were analyzed using a one-way analysis of variance (ANOVA) with repeated measures followed by a Student's t-test with the Bonferroni correction for repeated comparisons. Comparisons were made between values obtained at 50 min (end of aerobic baseline period) and 110 min (end of reperfusion) for the I/R protocol. Linear-trend analysis was performed on the recovery of LV work at the end of reperfusion versus each concentration of ANG II (0.1, 1, 10, and 100 nmol/l) used in combination with losartan. Values at 45 min (end of aerobic baseline period) and 90 min (end of aerobic treatment period) were compared. Results are reported as means ± SE. Statistical significance was set at P < 0.05.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

I/R protocol. There were no significant differences in indexes of mechanical function during aerobic baseline perfusion among any of the experimental groups. After ischemia, LV work in control hearts recovered to 44 ± 9% of aerobic baseline by the end of reperfusion. Coronary flow (24 ± 1 vs. 17 ± 3 ml/min, P < 0.05) but not CVC (0.29 ± 0.01 vs. 0.23 ± 0.03 ml/min, P = 0.10) was significantly depressed by the end of reperfusion compared with baseline, with the percent values at the end of reperfusion being 78 ± 13 and 85 ± 12% of baseline, respectively (Table 1). Peak systolic pressure and cardiac output were significantly (P < 0.01) depressed at the end of reperfusion compared with aerobic baseline (67 ± 9 and 50 ± 10%, respectively). Also, both MO₂ (63 ± 14%) and myocardial efficiency (46 ± 11%) were significantly impaired at the end of reperfusion compared with aerobic baseline (P < 0.01). Rates of glycolysis exceeded those of glucose oxidation in all hearts (Fig. 3). In controls, both rates of glycolysis and proton production were elevated compared with aerobic baseline (Table 2 and Fig. 3). The rate of glucose oxidation during reperfusion was similar to that measured during the aerobic baseline period.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Glucose metabolism in control (n = 6) hearts and those treated with N 6-cyclohexyladenosine (CHA, 0.5 µmol/l, n = 5) or losartan (1 µmol/l, n = 5) subjected to 30 min of global ischemia. Rates of glucose metabolism during baseline perfusion (filled bars) compared with rates during reperfusion (open bars). A: glycolysis; B: glucose oxidation; C: proton production. * P < 0.05 vs. reperfused controls.  P < 0.05 vs. baseline.

CHA significantly enhanced the recovery of LV work (81 ± 4%) at the end of reperfusion relative to time-matched controls (44 ± 9%, Table 1 and Fig. 4A). Myocardial efficiency of oxygen consumption for contractile work at the end of reperfusion was significantly improved by CHA (98 ± 6%) compared with controls (46 ± 11%, Table 1). Rates of glycolysis and proton production were significantly depressed during reperfusion, whereas glucose oxidation rates were unaltered compared with time-matched controls (Table 2 and Fig. 3).

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Left ventricular (LV) work during baseline, ischemia, and reperfusion. A: control hearts (no treatment, , n = 14); CHA (0.5 µmol/l, , n = 11); losartan 1 µmol/l (, n = 10), and 0.1 µmol/l (, n = 6). * P < 0.05 vs. controls. B: angiotensin II (ANG II) hearts (1 nmol/l, , n = 7); ANG II (1 nmol/l) plus losartan (1 µmol/l, , n = 7). * P < 0.05 vs. controls.

Losartan (1 µmol/l) prevented the recovery of LV work during reperfusion (0.0 ± 0.0% of aerobic baseline). Compared with controls, losartan at 1 µmol/l, but not at 0.1 µmol/l, depressed myocardial efficiency, peak systolic pressure, cardiac output, and coronary flow during reperfusion and had no significant effect on CVC (Table 1 and Fig. 4A). Losartan did not alter rates of glucose oxidation during reperfusion compared with controls (Table 2 and Fig. 3). However, compared with controls, rates of glycolysis and proton production were significantly depressed in losartan-treated hearts.

ANG II alone had no significant effect on the recovery, during reperfusion, of mechanical function, MO₂, or efficiency compared with controls. Concentrations of ANG II (0.1, 1, 10, and 100 µmol/l) and losartan (1 µmol/l) reversed the inhibition of function observed with the use of losartan alone in a concentration-dependent manner (Table 1 and Fig. 5). ANG II at 1 nmol/l, but not 0.1 nmol/l, used in combination with losartan significantly reversed the inhibition of LV work during reperfusion compared with losartan alone. Efficiency, MO₂, peak systolic pressure, cardiac output, coronary flow, and CVC were not significantly altered in hearts treated with a combination of losartan or ANG II (0.1 or 1 nmol/l). Higher concentrations of ANG II (10 and 100 nmol/l) used in combination with losartan did not significantly alter the recovery of LV work, peak systolic pressure, cardiac output, coronary flow, CVC, MO₂, and efficiency compared with hearts treated with 1 nmol/l ANG II in combination with losartan (Table 1 and Fig. 5). Although there appeared to be a trend of reduction in recovery of LV work during reperfusion with increasing concentrations of ANG II (1, 10, and 100 nmol/l), it was not statistically significant. The combination of PD-123,319 with losartan had no significant effect on recovery of mechanical function, MO₂, or efficiency compared with time-matched controls (Table 1).

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Recovery of left ventricular (LV) work during reperfusion expressed as a percentage of baseline for control (C, n = 14) and those treated with CHA (0.5 µmol/l, n = 11), losartan alone (L, 1 µmol/l, n = 10), or in combination with 0.1 (n = 6), 1 (n = 7), 10 (n = 6), and 100 (n = 6) nmol/l ANG II. * P < 0.05 vs. Control.

Aerobic protocol. Mechanical function, MO₂, and efficiency in all hearts subjected to the aerobic protocol were stable during the aerobic baseline period (Table 3). After 80 min of aerobic perfusion in control hearts, LV work was slightly depressed (by 9 ± 3%) relative to baseline values. None of the drug treatments had any significant effect on mechanical function, MO₂, or efficiency compared with controls.

In control hearts, rates of glycolysis and proton production during the treatment period were similar to baseline, but the rate of glucose oxidation increased gradually throughout aerobic perfusion and was higher during the treatment period compared with aerobic baseline (Table 4 and Fig. 6). CHA reduced rates of glycolysis and proton production compared with aerobic time-matched controls but had no significant effect on glucose oxidation (Table 4 and Fig. 6). Losartan (Table 3), ANG II, and ANG II + losartan (Table 5) did not alter mechanical function.

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Rates of glucose metabolism measured under aerobic conditions for control hearts (n = 6) and those treated with CHA (0.5 µmol/l, n = 5) or losartan (1 µmol/l, n = 5). Rates of glucose metabolism during baseline (filled bars) compared with treatment (open bars). A: glycolysis; B: glucose oxidation; C: proton production. * P < 0.05 vs. control treatment.  P < 0.05 vs. baseline.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

There are two new findings in this study. First, acute treatment with losartan, a selective AT1-R antagonist, prevented recovery of mechanical function in postischemic rat hearts compared with untreated hearts in a concentration-dependent manner. Second, this response was reversed by concomitant treatment with ANG II in a concentration-dependent manner. Concentrations of ANG II above those required to reverse the response to losartan did not further improve recovery of postischemic mechanical recovery over that observed for time-matched controls. These findings suggest that the effect of acute treatment with losartan on I/R injury is mediated by AT1-R antagonism. Furthermore, the inhibition of recovery of LV mechanical function by AT1-R antagonism in the absence of exogenous ANG II is evidence that intrinsic stimulation of the AT1-R contributes to the functional recovery of postischemic myocardium.

Mechanisms. In agreement with previous reports using the same experimental model (10, 15), the adenosine A₁ agonist CHA enhanced recovery of mechanical work. This cardioprotective effect was associated with improved recovery of myocardial efficiency during reperfusion, supporting the concept that CHA improves the coupling of oxidative metabolism to contractile work in postischemic myocardium. This beneficial effect appears to be mediated by limitation of the production of protons from glucose metabolism (25), thereby reducing Ca²⁺ overload and improving efficiency by reversing the algorithm: increased intracellular protons  increased clearance of H⁺ via the H⁺/Na⁺ exchanger  increased intracellular Na⁺  increased exchange of intracellular Na⁺ for extracellular Ca²⁺  increased cytosolic Ca²⁺ (8). The final clearance of the high cytosolic Ca⁺ involves an energy-dependent mechanism that reduces ATP available for contraction, thereby reducing myocardial efficiency (18).

In contrast to CHA-induced cardioprotection, selective AT1-R antagonism with losartan completely inhibited recovery of LV work and myocardial efficiency of postischemic hearts in a concentration-dependent manner. Compared with control hearts, losartan lowered the rate of glycolysis but not that of proton production. This losartan-induced reduction of glycolysis is most likely the consequence of a much lower workload in these hearts compared with controls. Because losartan did not significantly influence rates of glycolysis and glucose oxidation in aerobic hearts, the depression of LV work and myocardial efficiency in the postischemic hearts cannot be attributed to an increase in proton production and intracellular acidosis. In addition, the fact that neither losartan nor ANG II influenced the rates of glucose metabolism during aerobic perfusion suggests that AT1-R-mediated stimulation of Na⁺/H⁺ exchange (16) is not due to alterations in proton production.

Reversal of the losartan-mediated inhibition of LV work and myocardial efficiency by concomitant administration of increasing concentrations of ANG II has several implications. First, it suggests that the presence of endogenous ANG II played a role in the recovery under control conditions, although we did not measure the endogenous ANG II concentration. Because neither losartan nor ANG II had a significant effect on mechanical function or efficiency in nonischemic hearts, endogenous ANG II did not seem to be involved in the maintenance of inotropy.

Second, the results also suggest that the losartan-induced inhibition of mechanical recovery was mediated directly via myocardial AT1-R antagonism and is reversed by displacement of losartan from its receptor by ANG II in a concentration-dependent manner. If reversal of the losartan effect by ANG II had occurred via a non-AT1-R-related mechanism, then increased cardioprotection would be expected in the presence of the higher ANG II concentrations. However, this was not the case, suggesting that reversal of the losartan response by ANG II is mediated by competition at a specific receptor site rather than via a non-AT1-R-mediated mechanism. Because the selectivity of losartan for the AT1-R is dependent on its concentration, the low concentration of free losartan in our study would ensure even greater AT1-R selectivity (7). The 3% BSA (30 mg/ml) in our perfusate is equivalent to concentrations of albumin in human plasma (4) and binds as much as 99% of losartan, thus reducing the concentration of free losartan available to compete for AT1-R binding sites (3), resulting in a 100-fold lower free concentration of ~10 nmol/l (7). At this concentration, which is two times the K_D (concentration for 50% receptor occupancy) for losartan at the AT1-R (7), 67% of AT1-R would be occupied by losartan {[agonist]/K_D + [agonist], or 10/(2 + 10) × 100 = 67%}. Because the K_D value for ANG II binding to AT1-R is 1 nmol/l (14), 50% of AT1-R [1/(1 + 1) × 100] would be occupied by 1 nmol/l ANG II. In the absence of an available selective AT1-R agonist, the low free concentration of losartan and reversal of its response by increasing ANG II concentrations represents the best available functional evidence that the responses observed are mediated by the AT1-R. Our recent demonstration that selective ANG II type 2 receptor (AT2-R) antagonism with PD-123,319 can improve the recovery of mechanical function after I/R injury (11) indicates that AT2-R stimulation would be expected to have a deleterious effect on functional recovery. Thus our finding that the combination of the AT1-R and the AT2-R antagonists did not influence recovery relative to untreated hearts does not necessarily indicate a role of the AT2-R in the ANG II-induced reversal of the deleterious effects of losartan. These data do not exclude the possibility of functional antagonism between the beneficial effects of PD-123,319 and the deleterious effects of losartan. Therefore, AT2-R stimulation cannot account for the reversal of the losartan response by ANG II.

Third, despite the ability of ANG II to overcome the depressant action of losartan, ANG II alone was not cardioprotective in our study. Therefore, it would appear that any potential protection afforded by AT1-R stimulation and PKC activation is counterbalanced by effects that exacerbate I/R injury. Because ANG II is not selective for AT1-R, it is possible that a deleterious response mediated by AT2-R stimulation accounts for this phenomenon.

The deleterious response to losartan suggests that selective AT1-R stimulation may be cardioprotective. This seems unlikely at first because of the known deleterious effects of ANG II on I/R injury (29). However, these deleterious effects (29) were obtained using a Langendorff model that differs from the working model used in our study in at least two ways. First, hearts perfused by the Langendorff method do not perform external work, whereas working hearts eject perfusate against a fixed afterload. A major consequence of this is that the energy demand of the Langendorff-perfused heart is less than that of the working heart (5, 12). For this reason, measurements of MO₂ and myocardial efficiency in the Langendorff model are not physiologically relevant and were indeed not done in the study of Yoshiyama et al. (29). Second, the Langendorff heart is not responsible for its own coronary circulation, whereas in the working heart coronary circulation is under autoregulatory control (5, 12). In that study (29), an increase in coronary vasoconstriction could have masked any direct cardioprotective action of the ANG II. Indeed, cardioprotection has been associated with AT1-R stimulation in rabbit hearts where it mimicked ischemic preconditioning and limited infarct size (19).

None of the ANG II concentrations used had any significant effect on the degree of recovery of postischemic mechanical function. Even at the highest concentration of ANG II, there was no significant vasoconstriction during reperfusion. Therefore, any potentially beneficial coronary vasodilation due to AT1-R antagonism is not observed in this model. Our recent finding that AT2-R stimulation might be detrimental to recovery of postischemic function (11) leads us to speculate that, in the absence of changes in coronary vascular tone, the potential beneficial effects of AT1-R stimulation by ANG II are counterbalanced by deleterious AT2-R stimulation. The recent evidence of interaction between AT1-R and AT2-R in ANG II-mediated cardiomyocyte hypertrophy (2) supports this view.

In conclusion, the findings in this study demonstrate that acute AT1-R antagonism exacerbated I/R injury, and this effect was reversed by a range of ANG II concentrations. This indicates that intrinsic AT1-R stimulation modulates recovery of mechanical function in the postischemic heart. In addition, the enhanced I/R damage by AT1-R blockade did not involve increased proton production from glucose metabolism. The exacerbation of I/R injury by acute AT1-R antagonism is in contrast to the beneficial effects on myocardial pathology observed during chronic administration. Importantly, the results suggest that endogenous AT1-R activation contributes to the recovery of mechanical function in the postischemic myocardium and that selective AT1-R stimulation or agonism might provide a novel approach to the management of I/R injury.