INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

ISCHEMIA of 20 min or less results in postischemic "myocardial stunning" that persists despite the lack of irreversible damage (3). Free radicals have been implicated in stunning-induced arrhythmias, because various antioxidants attenuate these arrhythmias (4, 10) whereas free radical-generating systems increase their incidence (9). Moreover, in the dog, blocking production of · OH radicals alone attenuates reperfusion-induced arrhythmias (25, 28). To explore an antioxidant mechanism for the antiarrhythmic effect of estrogen, we measured the production of · OH radicals during reperfusion after a 20-min period of ischemia. Our aim was to determine whether administration of conjugated equine estrogen (CEE) could attenuate the burst of · OH radicals typically observed during the first 5 min of reperfusion.

The highly reactive · OH radical is difficult to measure in biological systems because it reacts with many organic molecules. It is necessary to "trap" the molecule by providing a substrate with which it will selectively react to form a measurable product. In this experiment, 200 µM sodium salicylate was included in the blood perfusate. In the presence of · OH radical, salicylate (2-hydroxybenzoic acid) is further hydroxylated to dihydroxylated benzoic acid species and thus acts as a reporter molecule (20). By measuring one product of this reaction, 2,5-dihydroxybenzoic acid (2,5-DHBA), we were able to show that the production of · OH radical was significantly attenuated during reperfusion in the presence of CEE. Although this methodology is common in our laboratory (20), others have used specific scavengers of hydroxyl radicals including N-(2-mercaptopropionyl)glycine (11, 18) and the iron chelator desferrioxamine (5).

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Experimental preparation. In compliance with methods approved by the Rutgers University Animal Care and Use Committee, 17 dogs of either gender weighing 12.9 ± 1.3 kg were used for this study. Blood samples from prior experiments revealed that endogenous estrogen in dogs of all ages of either gender (not in heat) is not detectable (<20 pg/ml, unpublished observation). In the present study dogs were randomly assigned to a CEE-treated group [10 µg/kg iv of CEE (injectable Premarin, Wyeth-Ayerst, Philadelphia, PA); n = 8] or a nontreated group that received the same volume of Premarin vehicle (0.9% saline; n = 9). Dogs were fasted for ~16 h before surgery, with water provided ad libitum. Anesthesia was induced by intravenous injection of pentobarbital sodium (30 mg/kg, Sigma, St. Louis, MO). The trachea was intubated, and artificial respiration was initiated (model 607 ventilator, Harvard Apparatus, Millis, MA). Tidal volume and respiratory rate were adjusted to maintain blood gases and pH within physiological limits (e.g., PO₂ 85-110 mmHg, PCO₂ 35-45 mmHg, and pH 7.38-7.46).

Experimental protocol. Instrumented dogs were observed for ~15 min or until recorded variables achieved a steady state. Recorded variables included heart rate (HR), MAP, CPP, LVP, ±dP/dt, CBF, ECG, the peak blood flow response on reperfusion, and the time it took to reach this peak flow response (TPFR). Heparin was supplemented every 30 min (300 U/kg iv). To assess the viability of the coronary vasculature, we observed reactive hyperemia following release of a 15-s occlusion of the cannulated LAD. This procedure was used before and after our primary protocol and is standard in our laboratory (6, 15). Regional ischemia was imposed by occluding flow in the shunt for 20 min. Twenty minutes before ischemia, we initiated a continuous, intravenous infusion of 200 µM sodium salicylate (0.2 ml/min). This low concentration of salicylate does not affect cardiac hemodynamics or postischemic recovery (14, 21). Ten minutes after the infusion of salicylate was initiated, CEE or its vehicle was administered via bolus injection into the left atrial appendage. Samples of coronary venous blood from the perfused myocardium were removed 10 min after initiation of salicylate infusion, 5 min after administration of CEE or its vehicle (just before onset of ischemia), after 15 min of ischemia, and at 2, 5, and 25 min of reperfusion. Samples were immediately centrifuged at 2,000 g for 10 min to obtain plasma. The plasma was frozen at liquid nitrogen temperatures and stored at 70°C until analyzed for 2,5-DHBA content. Absolute values for 2,5-DHBA were determined by subtracting baseline production of · OH radicals in each dog from production at 2, 5, and 25 min of reperfusion. Values represent means ± SE. We recorded all monitored variables at baseline (control), during the last 5 min of ischemia, and during the first 5 min of reperfusion.

Analysis of production of · OH radicals. With the use of sodium salicylate as a probe for · OH, the production of 2,5-DHBA (an index of · OH radical production) was quantified by coulemetric detection after HPLC separation. Aliquots (670 µl) of plasma samples were deproteinized with 330 µl of 30% trichloroacetic acid and then centrifuged at 2,000 g to sediment the denatured protein. The supernatant was filtered through a 0.2-µm filter, and 2,5-DHBA was detected according to the method of Powell and Hall (21). Briefly, 60 µl of the filtered supernatant were injected (20-µl loop) into an HPLC (Perkin-Elmer High Sensitivity LC, series 410) equipped with a coulemetric detector (Coulechem model 5100, ESA). Detection parameters were as follows: electrode 1: oxidation potential, +0.4 V; electrode 2: reduction potential, 0.25 V; guard potential, 0.030 V. We quantified 2,5-DHBA by using a standard software package (Omega 2, Perkin-Elmer) for integration of peaks.

Drug preparation. Premarin was purchased in 25-mg secules (iv preparation). The contents were diluted in 0.9% saline to a final concentration of 100 µg/ml.

Data analysis. Student's t-test for paired data was used to compare measurements between the CEE-treated group and the saline-treated (control) group. For multiple comparisons within groups, we used ANOVA for repeated measures and Fisher's least significant difference for comparison of means. All data are presented as means ± SE, with statistically significant differences routinely established at P < 0.05.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Throughout the control and ischemic periods there were no significant differences between CEE-treated and nontreated dogs relative to the measured concentrations of 2,5-DHBA. During the first few minutes of reperfusion, the typical burst in production of · OH radicals (21, 25, 28) was observed in nontreated dogs. The peak of this burst occurred at 5 min of reperfusion and was markedly reduced or abolished in CEE-treated dogs (Figs. 1A and 2). Individually, the reduction in the production of · OH radicals was seen in each dog that received CEE. With respect to gender, however, the effects of CEE were not as prominent in males as in females (Fig. 1B).

View larger version (25K): [in this window] [in a new window]   Fig. 1.   A: absolute changes in measured 2,5-dihydroxybenzoic acid (2,5-DHBA) levels are shown in estrogen-treated [+ conjugated equine estrogen (+CEE)] and nontreated (CEE) dogs. Normal burst of · OH radical production is seen in first 5 min of reperfusion in nontreated dogs. Production of · OH radical was attenuated in +CEE dogs and was significantly less at peak production period (5 min) of reperfusion. B: absolute changes in venous effluent 2,5-DHBA levels are shown by gender for 2- (R2V), 5- (R5V), and 25-min (R25V) periods of reperfusion. * Significant compared with nontreated dogs (P < 0.05).

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Representative chromatograms of 2,5-DHBA at 2-min measurement for a CEE male (A), a CEE female (B), a +CEE male (C), and a +CEE female (D). Dashed lines emphasize differences in peaks between A and B vs. C and D.

In dogs treated with CEE, reperfusion hyperemia was attenuated as previously reported (15, 16). This pattern was not seen in nontreated dogs. This resulted in a significant prolongation in TPFR in CEE-treated dogs (Table 1). In these studies endogenous levels of estrogen in male and female (not in heat) dogs were not detectable (<20 pg/ml, unpublished observation). Treatment with a dosage of 10 µg/kg CEE resulted in a circulating plasma concentration of 146 ± 22 pg/ml, which is physiological for females and supraphysiological for males. Therefore, it is likely that the antioxidant properties of estrogens and its metabolites are evident during the normal estrous cycle and could account for gender differences in vulnerability to cardiac disease.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

In our earlier study (15), we looked for a functional correlate, a change in hemodynamics that could explain the decreased arrhythmias during reperfusion. Hyperemia on reperfusion was diminished, and therefore reflow was more sluggish in the presence of CEE (15, 16). A slower reflow is known to decrease reperfusion arrhythmogenesis (26). We cannot be certain that hemodynamic effects did not play a role; however, we have observed in a Langendorff-perfused rat heart that estrogen at concentrations of up to 1,000 pg/ml had no effects on coronary flow but did produce concentration-dependent improvement in postischemic myocardial function, indicating a direct effect of estrogen (unpublished observation). Whether diminished hyperemia was the result of antioxidant scavenging properties of estrogen or caused by some other effect of estrogen was not clear. Recent studies have shown that vascular reactivity can be altered by application of oxidants and that antioxidants reverse this process (8, 31). Thus our studies suggest that it might be the antioxidant properties of CEE that account for the diminished hyperemia. Estrogens also appear to stabilize membranes, thus decreasing membrane fluidity (32). This action might make the endothelium or underlying vascular smooth muscle less responsive to endogenously produced vasodilators. This could contribute to the diminished hyperemia we observed on reperfusion.

The manner in which estrogen exerts its antioxidant actions has not been fully elucidated and might involve multiple mechanisms. Different forms of estrogen have different antioxidant potencies. In terms of their abilities to inhibit lipid peroxidation, the catechol estrogens 17-dihydroequilin and 4-hydroxyestrone, metabolites of equilin and 17-estradiol, respectively, have been found to be the most potent. A general order of potency would be as follows: estrone < 17-estradiol < equilin < 17-dihydroequilin < 2- or 4-hydroxyestrone (27, 29, 30). A similar relationship has been found with regard to scavenging of the · OH radical (23). With respect to · OH radical, scavenging was apparent at low micromolar concentrations. Because the plasma concentration of estrogens in our dogs was in the nanomolar range we do not believe that scavenging was the major mechanism at play in our study. Rather, we believe that the estrogens were interfering with the catalytic activity of redox-active transition metals. Numerous studies in vitro have shown that estrogens have the potential to affect the redox chemistry of transition metals and, in particular, iron and copper (12, 24, 30). It is by this mechanism that estrogens are thought to inhibit peroxidation of lipids and lipoproteins (23, 24, 27). Such a mechanism might also account for the "antiatherogenic" effect of estrogen in premenopausal women (1, 19). The involvement of redox-active transition metals in postischemic reperfusion injury has been a subject of intense study. Many investigators have demonstrated a role for iron and, to a lesser extent, copper in this type of oxidative injury (13, 22). In general, agents that impair the redox activity of transition metals, either through chelation or through hindered changes in oxidation states, exert a protective effect. Certain estrogens and estrogen metabolites reduce Fe³⁺ and at the same time retard oxidation of Fe²⁺, thus interfering with the redox activity of this metal (24). Conceptually, estrogens may be viewed as being most similar to a class of nitroxide compounds that rapidly oxidize Fe²⁺ and prevent reduction of Fe³⁺, thereby interfering in the redox activity of the metal (17). We (7) have shown that some nitroxides can reduce the occurrence of reperfusion arrhythmias. Thus it seems reasonable to suggest that estrogens might reduce reperfusion arrhythmias through a similar mechanism. In addition, estrogen has been shown to possess · OH radical-scavenging properties (23). Agents with this property are known to be cardioprotective and to prevent electrophysiological changes that foster arrhythmogenesis (2). For example, studies have shown the cardioprotective effect of the free radical scavenger N-(2-mercaptopropionyl)glycine and its effects on hydroxyl radical production (11, 18).

In conclusion, our study demonstrates that CEE attenuates the postischemic production of · OH radicals. Production of · OH radicals has been associated with reperfusion-induced arrhythmias. We suggest that the antioxidant properties of CEE may be involved in attenuation of cardiac reperfusion injury as evidenced by decreased hyperemia and reperfusion arrhythmias.