Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Diadenosine polyphosphates (Ap_nA) are naturally occurring compounds that are present in the myocardium and platelet-dense granules (Jovanovic et al., 1998; Flores et al., 1999; Luo et al., 1999). They function as neurotransmitters and extracellular signaling molecules and as such alter platelet reactivity, vasomotor tone, and cardiac electrophysiology (Flores et al., 1999).

We have recently described the coronary vasomotor and cardiac electrophysiological effects of Ap_nA at physiologically relevant concentrations in the guinea pig (Stavrou et al., 1998, 2001). At nanomolar concentrations, they produce transient reductions in coronary perfusion pressure that become maximal and maintained with micromolar concentrations. Ventricular action potential duration and refractory period are increased with nanomolar and micromolar concentrations of Ap_nA. Mechanisms responsible for these effects involve P2 receptors for the vasomotor effects and both adenosine (P1) and P2 receptors for the electrophysiological effects (Stavrou et al., 1999a,b, 2001).

Mechanisms involving receptors to adenosine and ATP have been widely reported to account for the effects of Ap_nA on the cardiovascular system (Flores et al., 1999). Ap_nA are structurally related to ATP, but are more resistant to hydrolysis (Flores et al., 1999). As such, there are many similarities between their cardiac electrophysiological and vasomotor effects (Stavrou et al., 1999a, 2001). Despite this, the number of phosphate groups present in Ap_nA has an important influence on determining the direction and magnitude of the response of the heart and circulation to each compound (Flores et al., 1999).

Coronary vasodilatation occurs in response to ATP and is mediated by release of nitric oxide (NO) and prostanoids (Hopwood and Burnstock, 1987; Hopwood et al., 1989; Brown et al., 1992; Vials and Burnstock, 1994). In isolated rabbit hearts, release of NO has been shown to account for part of the vasodilatory response to diadenosine tetraphosphate (Ap₄A) and release of prostacyclin (PGI₂) is involved in the response to diadenosine triphosphate (Ap₃A; Pohl et al., 1991). The relative involvement of either mechanism in mediating the coronary vasomotor response to diadenosine pentaphosphate (Ap₅A) and diadenosine hexaphosphate (Ap₆A) is unknown. The involvement of NO and prostanoids in mediating the cardiac electrophysiological effects of Ap_nA has also not been investigated.

To investigate the hypothesis that NO and prostanoids are involved in mediating the coronary vasomotor and cardiac electrophysiological effects of Ap_nA, we studied the consequences of pharmacological inhibition of NO synthase and cyclooxygenase on the coronary vasomotor and cardiac electrophysiological effects of Ap₃A, Ap₄A, Ap₅A, and Ap₆A in isolated guinea pig hearts.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Perfusion Technique. Hearts were removed from male guinea pigs (Dunkin-Hartley, 400-500 g) that had been humanely killed and were mounted for Langendorff perfusion (Goulielmos et al., 1995). They were perfused at a constant flow rate (7 ml min¹) with Krebs-Henseleit buffer at 32°C. The buffer was gassed with 95% O₂/5% CO₂ to obtain pH 7.4 and contained 118.5 mM NaCl, 4.8 mM KCl, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, 24.9 mM NaHCO₃, 2.6 mM CaCl₂, 8.0 mM glucose, and 2.0 mM sodium pyruvate. Heart rate was maintained at a constant rate of 3.3 Hz (just above the spontaneous sinus rate) by right ventricular pacing at twice the diastolic pacing threshold using square wave pulses 5 ms in duration from a stimulator (ST-02; Experimetria Ltd., Budapest, Hungary.) The investigation conformed with United Kingdom legislation, the Animals (Scientific Procedures) Act of 1986 and the Guide for the Care and Use of Laboratory Animals published by the U.S. National Institutes of Health.

Electrophysiological Recordings. Action potentials, refractory periods, electrograms, and perfusion pressure were recorded and analyzed (Goulielmos et al., 1995; Stavrou et al., 2001) using a digital data acquisition and analysis system (Po-Ne-Mah; Gould Instrument Systems, Inc., Valley View, OH; 12-bit resolution, frequency response 5 kHz, sampling rates 250 Hz for perfusion pressure, 1 kHz for electrograms and action potentials). Electrograms were recorded as a volume-conducted ECG equivalent (lead II) using three silver electrodes fixed within the organ bath in a triangular arrangement. Transmembrane action potentials were recorded using the floating microelectrode technique from apical right ventricular epicardial cells (Goulielmos et al., 1995). Glass microelectrodes (tip resistances of 5-15 M) filled with 3 M KCl were mounted on flexible silver wire (diameter 125 µm), coated with Teflon except for 3 mm at the tip, which formed an Ag/AgCl junction. Microelectrodes were connected to a differential field-effect transistor input instrumentation amplifier with an input impedance of 1.5 T (built by the Department of Medical Engineering, Imperial College School of Medicine, London, UK; based on circuits described by Chapman and Fry, 1978). Action potentials were recorded as the potential between the intracellular microelectrode and a reference electrode (a KCl salt bridge connected via an Ag/AgCl junction) placed in the organ bath, as described by Penny and Sheridan (1983). All recordings were made from the apical region of the heart. Although multiple impalements were required to provide continuous electrophysiological data, all data presented are based on single, stable impalements. Using this technique, stable impalements can be achieved for at least 20 s, allowing 60 action potentials to be recorded and analyzed. Action potential duration was measured at 95% repolarization (APD₉₅). Refractory periods were determined using the extrastimulus technique. The hearts were paced at 3.3 Hz (198 beats min¹, cycle length 303 ms between beats) at twice pacing threshold, determined prior to each measurement, which did not vary and was in the range of 0.2 to 0.3 mA. The extra stimulus was introduced once after every eight regular beats at shorter coupling intervals and in decrements of 5 ms until failure to capture occurred. The effective refractory period was taken as the longest interval at which failure to capture occurred (Penny and Sheridan, 1983).

Protocol. Hearts were perfused with buffer for a control period of 20 min, and measurements were made. Perfusion was then switched to a buffer containing an inhibitor of NO synthase: [L-N^G-nitroarginine methyl ester (L-NAME), 300 µM or L-N⁵-(1-iminoethyl)ornithine (L-NIO), 30 µM]; or a cyclooxygenase inhibitor (indomethacin, 100 µM; or meclofenamate, 10 µM) that was allowed to flow for 15 min and measurements were repeated. Perfusion was then switched to a buffer containing the inhibitor and each Ap_nA (in separate experiments) at 1 nM, followed by 1 µM with 30 min allowed at each concentration. Measurements were repeated at 5-min intervals during exposure to the two concentrations and peak responses are reported. L-NIO and meclofenamate were used as alternative inhibitors of NO synthase and cyclooxygenase, respectively, to provide confirmatory data in support of observations made with L-NAME and indomethacin.

Experimental Groups. Hearts were studied in the following groups: L-NAME + Ap₃A (n = 8), L-NAME + Ap₄A (n = 8), L-NAME + Ap₅A (n = 4), L-NAME + Ap₆A (n = 4); L-NIO + Ap₃A (n = 5), L-NIO + Ap₄A (n = 5), L-NIO + Ap₅A (n = 3), L-NIO + Ap₆A (n = 3); indomethacin + Ap₃A (n = 6), indomethacin + Ap₄A (n = 5), indomethacin + Ap₅A (n = 4), indomethacin + Ap₆A (n = 4); meclofenamate + Ap₃A (n = 6), meclofenamate + Ap₄A (n = 5), meclofenamate + Ap₅A (n = 3), meclofenamate + Ap₆A (n = 3). Ap₃A, Ap₄A, Ap₅A, and Ap₆A were studied alone with five hearts in each group. Data were also obtained from hearts in which the effects of L-NAME and indomethacin in combination on the actions of Ap₃A (n = 5), Ap₄A (n = 3), Ap₅A (n = 4), and Ap₆A (n = 3) were investigated.

Statistical Analysis. Data are expressed as the mean ± standard error of the mean. Two-way repeated measures of analysis of variance, followed by Bonferroni's multiple comparison test (for changes in coronary perfusion pressure), and Dunnett's multiple comparison test (for changes in APD₉₅ and refractory period) as post-tests, were used to identify where statistically significant differences had occurred. The effects of L-NAME, L-NIO, indomethacin, and meclofenamate on coronary perfusion pressure were compared with drug-free control conditions using Student's paired t test. p  0.05 defines the probability value indicating statistical significance.

Drugs. All chemicals used for the preparation of the Krebs-Henseleit buffer were of analytic grade (Merck Ltd., Lutterworth, UK). All compounds were dissolved in Krebs-Henseleit buffer except indomethacin, which was dissolved in 0.1 M NaOH (0.2 g in 0.5 ml). Stock solutions of indomethacin were then prepared for storage by further dilution to 10² M with Krebs-Henseleit buffer and diluted to 100 µM with Krebs-Henseleit buffer for use. The final concentration of NaOH in the perfusate was sufficiently low and had no effect on any variable recorded from the hearts. All stock solutions were stored at 20°C and were thawed and diluted prior to use. Ap_nA, L-NAME, L-NIO, indomethacin, and meclofenamate were purchased from Sigma-Aldrich Co. Ltd. (Poole, UK). L-NAME was chosen because its effects have been widely characterized by us and others (Rees et al., 1990; Smith et al., 1992; Goulielmos et al., 1995) in this model. In a previous study (Goulielmos et al., 1995), we used 100 µM L-NAME to inhibit NO synthase. In pilot experiments, we found that responses to Ap_nA persisted in the presence of this concentration, so we increased it further to 300 µM. Since L-NAME at high concentrations has also been reported to antagonize muscarinic receptors (Buxton et al., 1993), we repeated the experiments using a more potent NO synthase inhibitor, L-NIO (Rees et al., 1990), which we were able to use at a lower concentration (30 µM) and obtained confirmatory data. Similarly, with indomethacin we found from pilot experiments that a concentration of 100 µM was needed, so to confirm the specificity of the effects in relation to cyclooxygenase inhibition, we used the cyclooxygenase inhibitor meclofenamate (10 µM; Vegh et al., 1990).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Vasomotor and Electrophysiological Effects of Ap_nA. Figures 1 and 2 illustrate the vasomotor and electrophysiological effects of Ap_nA at 1 nM and 1 µM, which were comparable to those that we have described previously (Stavrou et al., 1998, 2001). Transient reductions in coronary perfusion pressure were seen with 1 nM Ap_nA, as illustrated for Ap₄A in Fig. 1A. When the concentration was increased to 1 µM, larger and persistent, statistically significant reductions in perfusion pressure were observed. With Ap₃A, Ap₄A, and Ap₅A, APD₉₅ and refractory period were increased, while with Ap₆A, only refractory period was increased, as reported previously (Stavrou et al., 1998, 2001). These effects are illustrated in Fig. 2.

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Effects of diadenosine polyphosphates on coronary perfusion pressure. Panel A illustrates a representative recording from a heart receiving 1 nM Ap4A, which produced a transient reduction in perfusion pressure. Exposure to 1 µM Ap4A produced a larger reduction in perfusion pressure, which was maintained. Panel B illustrates the peak/nadir responses of hearts to Ap3A, Ap4A, Ap5A, and Ap6A at concentrations of 1 nM and 1 µM. Data were obtained from five hearts for each ApnA. *p < 0.05; **p < 0.01; ***p < 0.001 versus control, i.e., ApnA-free conditions. , control; , 1 nM; , 1 µM.

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Effects of diadenosine polyphosphates on ventricular action potential duration and refractoriness. Panel A illustrates a representative action potential recording from a heart before () and after () 1 µM Ap4A. Panels B and C illustrate the effects of Ap3A, Ap4A, Ap5A, and Ap6A at concentrations of 1 nM and 1 µM on APD95 and RP, respectively. , control; , 1 nM; , 1 µM. Data were obtained from five hearts for each ApnA. *p < 0.05; **p < 0.01 versus control, i.e., ApnA-free conditions.

Effects of NO Synthase Inhibition

Vasomotor Effects. Perfusion of hearts with either L-NAME or L-NIO increased coronary perfusion pressure (from 32.9 ± 1.4 to 46.6 ± 2.0 mm Hg, n = 24, p < 0.0001 with L-NAME and from 39.2 ± 2.1 to 55.1 ± 2.7 mm Hg, n = 16, p < 0.0001 with L-NIO), indicating that basal production of NO had been inhibited. The effects of L-NAME alone on coronary perfusion pressure in each experiment are illustrated in Fig. 3A. The vasomotor effects of Ap_nA were not abolished by inhibition of NO synthase using either L-NAME (Fig. 3A) or L-NIO (Table 1) since coronary perfusion pressure was not maintained at a constant level but instead was significantly reduced during exposure to Ap_nA. This was especially apparent with a concentration of 1 µM (Fig. 3A).

View larger version (38K): [in this window] [in a new window]   Fig. 3.   Effects of inhibition of NO synthase using L-NAME (panel A; ) and inhibition of cyclooxygenase using indomethacin (panel B; ) on the actions of diadenosine polyphosphates at 1 nM () and 1 µM () on coronary perfusion pressure. Peak/nadir responses are indicated. Data were obtained from hearts perfused with L-NAME + Ap3A (n = 8), L-NAME + Ap4A (n = 8), L-NAME + Ap5A (n = 4), L-NAME + Ap6A (n = 4), indomethacin + Ap3A (n = 6), indomethacin + Ap4A (n = 5), indomethacin + Ap5A (n = 4), and indomethacin + Ap6A (n = 4). , control. L-NAME had no effect on the vasomotor response to each ApnA, while indomethacin prevented the vasomotor effects of Ap4A and Ap5A. *p < 0.05; **p < 0.01; ***p < 0.001 versus ApnA-free conditions, i.e., L-NAME or indomethacin.

Electrophysiological Effects. Perfusion of hearts with L-NAME or L-NIO had no effect on APD₉₅ or RP (Table 2 and Fig. 4). Inhibition of NO synthase using either compound prevented the cardiac cellular electrophysiological effects of each Ap_nA, so that APD₉₅ and RP remained constant in the presence of 1 nM and 1 µM Ap_nA, compared with control values (Fig. 4, Table 2).

View larger version (38K): [in this window] [in a new window]   Fig. 4.   Effects of inhibition of NO synthase using L-NAME on the actions of diadenosine polyphosphates at 1 nM and 1 µM on APD95 (panel A) and RP (panel B). Data were obtained from hearts perfused with L-NAME + Ap3A (n = 8), L-NAME + Ap4A (n = 8), L-NAME + Ap5A (n = 4), and L-NAME + Ap6A (n = 4). L-NAME prevented the effects of ApnA on action potential duration and refractory period. , control; , L-NAME; , 1 nM; , 1 µM.

Effects of Cyclooxygenase Inhibition

Vasomotor Effects. Perfusion of hearts with indomethacin and meclofenamate tended to reduce coronary perfusion pressure (from 40.9 ± 2.5 to 29.0 ± 2.0 mm Hg, n = 19, with indomethacin and from 44.0 ± 4.2 to 35.2 ± 3.6 mm Hg, n = 17, with meclofenamate) as has been reported previously (Grover et al., 1994; Véronneau et al., 1997). Inhibition of cyclooxygenase attenuated the vasomotor effects of Ap₃Awith 1 µM Ap₃A in the presence of indomethacin, the reduction in perfusion pressure just failed to reach statistical significance (p = 0.058). Inhibition of cyclooxygenase prevented the vasomotor effects of Ap₄A and Ap₅A, but did not abolish the vasomotor effects of Ap₆A at 1 µM (Fig. 3B, Table 1).

Electrophysiological Effects. Perfusion of hearts with indomethacin or meclofenamate had no effect on APD₉₅ or RP (Fig. 5, Table 2). Inhibition of cyclooxygenase using either compound prevented the cardiac cellular electrophysiological effects of each Ap_nA, so that APD₉₅ and RP remained constant in the presence of 1 nM and 1 µM Ap_nA, compared with control values (Fig. 5, Table 2).

View larger version (34K): [in this window] [in a new window]   Fig. 5.   Effects of inhibition of cyclooxygenase using indomethacin on the actions of diadenosine polyphosphates at 1 nM and 1 µM on APD95 (panel A) and RP (panel B). Data were obtained from hearts perfused with indomethacin + Ap3A (n = 6), indomethacin + Ap4A (n = 5), indomethacin + Ap5A (n = 4) and indomethacin + Ap6A (n = 4). Indomethacin prevented the effects of ApnA on action potential duration and refractory period. , control; , indomethacin; , 1 nM; , 1 µM.

Effects of Combined NO Synthase and Cyclooxygenase Inhibition

Vasomotor Effects. Combined inhibition of NO synthase and cyclooxygenase (using L-NAME and indomethacin) produced effects similar to those seen with indomethacin alone with regard to the effects of 1 µM Ap_nA since the vasomotor responses were blocked or attenuated (Fig. 6A). A different pattern was apparent when the effects of 1 nM Ap_nA were studied. With Ap₃A a slight but nonsignificant reduction in perfusion pressure was observed, but with Ap₄A, Ap₅A, and Ap₆A perfusion pressure increased (Fig. 6A).

View larger version (26K): [in this window] [in a new window]   Fig. 6.   Effects of combined inhibition of NO synthase using L-NAME and of cyclooxygenase using indomethacin on the vasomotor (panel A) and electrophysiological (panels B and C) effects of Ap3A (n = 5), Ap4A (n = 3), Ap5A (n = 4) and Ap6A (n = 3). Peak responses are indicated. Data are presented before and after addition of ApnA at concentrations of 1 nM and 1 µM to hearts perfused with L-NAME and indomethacin. *p < 0.05 versus ApnA-free conditions as indicated. Combined inhibition of NO synthase and cyclooxygenase prevented the electrophysiological effects of ApnA and attenuated the vasomotor effects of ApnA at 1 µM. With 1 nM Ap4A, Ap5A, and Ap6A, the direction of the vasomotor responses was reversed so that perfusion pressure increased. , L-NAME/indomethacin; , 1 nM; , 1 µM.

Electrophysiological Effects. Combined inhibition of NO synthase and indomethacin produced effects similar to those seen with either L-NAME or indomethacin alone. APD₉₅ and RP were unchanged by the diadenosine polyphosphates (Fig. 6, B and C).

Stability of the Preparation

We have previously shown that under our experimental conditions hearts are stable for several hours. We confirmed this using a series of hearts (n = 5) as time controls without application of inhibitors or diadenosine polyphosphates. APD₉₅, RP, and perfusion pressure remained stable for 2 h (APD₉₅: 170.3 ± 5.8 versus 171.0 ± 7.5 ms; RP: 147.5 ± 7.5 versus 153.3 ± 1.7 ms; perfusion pressure: 46.2 ± 2.3 versus 45.2 ± 3.6 mm Hg).

In pilot experiments, we examined the reproducibility of the effects of the diadenosine polyphosphates on the heart. We obtained reproducible responses in both naïve preparations and after washout (for Ap₄A and Ap₅A), which were of comparable magnitude to those seen in experiments in which diadenosine polyphosphates were applied cumulatively.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

This study has demonstrated that inhibition of cyclooxygenase and NO synthase prevents the cardiac electrophysiological effects of Ap_nA, while cyclooxygenase inhibition modulates the vasomotor effects of Ap_nA in a manner that is related to the structure of the Ap_nA compound. Inhibition of NO synthase had no effect on the vasomotor effects of Ap_nA. Novel observations from this study are 1) the involvement of cyclooxygenase- and NO synthase-dependent mechanisms in the cardiac electrophysiological effects of Ap_nA, and 2) the involvement of cyclooxygenase-dependent mechanisms in the coronary vasomotor effects of Ap₅A but not Ap₆A.

Ap_nA are structurally related to ATP and adenosine. Adenosine produces its coronary vasodilator effects via adenosine A₂ receptors, ATP-dependent K⁺ (K_ATP) channels, and mechanisms which involve release of NO and endothelium-derived hyperpolarizing factor (EDHF; Olsson and Pearson, 1990; Hein and Kuo, 1999; Rubio et al., 1999). ATP, on the other hand, produces its coronary vasodilator effects via P2 receptors and mechanisms that involve release of both NO and prostanoids (Hopwood and Burnstock, 1987; Hopwood et al., 1989; Olsson and Pearson, 1990; Brown et al., 1992; Vials and Burnstock, 1994). Our data indicate the cyclooxygenase-dependent nature of the vasomotor responses of the coronary circulation to Ap_nA and their independence of NO synthase, which suggests the involvement of prostanoids.

Adenosine produces its cardiac electrophysiological effects via adenosine A₁ receptors and mechanisms that involve release of NO (Olsson and Pearson, 1990; Rubio et al., 1999). ATP, on the other hand, produces its cardiac electrophysiological effects via both P2 receptors and adenosine receptors that are activated after its subsequent metabolism to adenosine (Belardinelli et al., 1995; Matsuura et al., 1996). The cardiac electrophysiological effects of Ap_nA were dependent on both cyclooxygenase and NO synthase, suggesting the involvement of both prostanoids and NO.

The cellular mechanisms coupling stimulation of adenosine receptors with cardiac electrophysiological and coronary vasomotor effects have been widely investigated (Olsson and Pearson, 1990). Similarly, the mechanisms coupling stimulation of P2 receptors by ATP with coronary vasomotor effects are well known (Olsson and Pearson, 1990), but the potential link between NO production and cardiac electrophysiological effects is less apparent. Our results demonstrate that inhibition of NO synthase prevented the increase in ventricular action potential duration and refractory period mediated by Ap_nA. Possible mechanisms that would account for this are the observations that P2 receptor-mediated NO production can delay conduction and reduce heart rate via cholinergic effects in sinoatrial and atrioventricular nodal cells (Han et al., 1994; reviewed by Prendergast and Shah, 1999), which would lead to increased refractoriness and increased action potential duration in the intact heart. NO has been shown to activate cyclooxygenase and stimulate release of prostanoids (Salvemini et al., 1993), and such a link between enzymatic systems would account for our observations.

NO has also been reported to inhibit the production and action of EDHF in the coronary circulation (Nishikawa et al., 2000), and the combination of inhibition of both NO synthase and cyclooxygenase has been used as a means of determining the relative involvement of EDHF in vasomotor responses (Nishikawa et al., 1999). As expected, with regard to the electrophysiological effects of Ap_nA, combined inhibition of NO synthase and indomethacin produced effects similar to those seen with either L-NAME or indomethacin alone. With regard to the vasomotor effects, when the hearts were challenged with the higher (1 µM) dose of Ap_nA in the combined presence of L-NAME and indomethacin, similar responses to those seen with indomethacin alone were observed. The increases in perfusion pressure seen with 1 nM Ap₄A, Ap₅A, and Ap₆A were not expected, and further work is required to determine the underlying mechanisms that may involve the emergence of a dominant effect involving P2X receptors, as has been reported in the mesenteric arterial bed (Busse et al., 1988; Ralevic et al., 1995).

Previous studies in rabbit isolated hearts (Pohl et al., 1991) have reported that release of NO accounts for part of the coronary vasomotor response to Ap₄A and that release of PGI₂ is involved in the response to Ap₃A. We are not aware of any reports describing comparable studies with Ap₅A and Ap₆A. Our results in the guinea pig are consistent with those seen in rabbits with Ap₃A but vary slightly from those seen with Ap₄A. This may be related to species differences comparable with those that have been reported previously by Froldi and Belardinelli (1990) and Song and Belardinelli (1996). To our knowledge, the involvement of NO and PGI₂ in the cardiac electrophysiological effects of Ap_nA has not been investigated.

Nakae et al. (1996) proposed the involvement of K_ATP channels in mediating the vasodilatory effects of Ap₄A in anesthetized pigs. Evidence is available that PGI₂ release may activate K_ATP channels producing vasodilatation (Jackson et al., 1993), but the effects of K_ATP channel inhibitors were beyond the scope of this study. Interestingly, Grover et al. (1994) have proposed that the cyclooxygenase inhibitor meclofenamate may, in addition to being a cyclooxygenase inhibitor, also act as a cardiac K_ATP channel inhibitor. The similarity of effects between indomethacin and meclofenamate indicates that the primary mechanism involves cyclooxygenase products, but whether these indirectly activate K_ATP channels as suggested by Jackson et al., (1993) remains to be investigated. The relative involvement of K_ATP channels in mediating the cardiac electrophysiological effects of Ap_nA also remains to be investigated and presents a potential paradox. While K_ATP channels may be involved in the coronary vasomotor effects of Ap_nA, a potential role in the electrophysiological effects of Ap_nA is difficult to envisage since activation of myocardial K_ATP channels reduces action potential duration (Nichols et al., 1991). Studies undertaken by Jovanovic et al. (1997) do, however, indicate that Ap_nA inhibit K_ATP channel activity when applied intracellularly to isolated ventricular myocytes and this would prevent action potential duration shortening. For this to occur, Ap_nA would need to be transported intracellularly, as has been proposed by Brandts et al. (1998).

In conclusion, our data demonstrate that the cardiac electrophysiological and coronary vasomotor effects of Ap_nA are mediated by NO synthase- and cyclooxygenase-dependent mechanisms and the structure of the Ap_nA has an important bearing on the dependence of these mechanisms with regard to their vasomotor effects. The principal involvement of Ap_nA in the heart is probably in acute coronary syndromes associated with platelet activation, such as myocardial ischemia, when Ap_nA are released from platelet-dense granules. Since their effects are mediated via P1 and P2 receptors, Ap_nA may have protective or deleterious effects that, as this study demonstrates, are mediated via release of NO and cyclooxygenase products. If the endothelium were to be damaged or dysfunctional, then it is likely that these effects would be prevented. Ap_nA are also hydrolyzed by ectoenzymes present on the vascular endothelium, and this also indicates the potentially important involvement of the endothelium in modulating their effects.