Introduction

Top Abstract Introduction Methods Results Discussion References

ATP and adenosine have gained popularity for producing deliberate hypotension during surgery (Sollevi et al., 1984b; Bloor et al., 1985) and treating supraventricular tachyarrhythmia (DiMarco et al., 1983; Belhassen et al., 1983). Through various mechanisms, these substances attenuate the baroreflex-mediated increases in HR, systemic vascular resistance and sympathetic nerve activity, compared with nitroprusside (Fukunaga et al., 1982; Lagerkranser et al., 1984; Delle et al., 1988). It has been generally assumed that the actions of exogenously administered ATP are caused by adenosine, because ATP is rapidly degraded to adenosine by the ectoenzymes such as ATPase (Willamsson and Dipiertro, 1965; Sollevi et al., 1984a). However, we previously reported, using neuraxis-intact and cervically vagotomized dogs, that ATP but not adenosine has a mechanism by which it attenuates reflex increases in sympathetic nerve activity, by interacting with the vagal afferents (Taneyama et al., 1991). In that experiment, the location of the vagal afferent mechanism of ATP could not be determined, but we speculated that i.v. administered ATP, at high enough doses, may stimulate ventricular receptors from which vagal afferents arise. It has been shown that purinergic compounds, including ATP, stimulate ventricular epicardial sensory afferent endings. Activity enhancement in nodose ganglion cardiac afferent neurons occurs during brief periods of coronary occlusion and reperfusion, during which purinergic compounds are known to be released by the ischemic myocardium (Armour et al., 1994).

Vagal afferent fibers originating in the ventricles are activated by both mechanical and chemical stimulation, leading to precipitous bradycardia and hypotension (Paintal, 1955; Lagerkranser et al., 1984). This cardiac reflex attenuates baroreflex control of HR, systemic vascular resistance and sympathetic nerve activity in anesthetized and conscious animals (Chen, 1979; Trimarco et al., 1987; Zucker et al., 1989). It is interesting to note that reflex increases in blood pressure can occur when chemoreceptors associated with vagal afferents around the region of the proximal left coronary artery are activated. This is known as a cardiogenic hypertensive chemoreflex (James et al., 1975).

The main goal of this study was to elucidate whether the i.c. administration of ATP could attenuate the reflex increase in sympathetic nerve activity that occurs in response to arterial hypotension. RSNA was measured as representative efferent sympathetic nerve activity in neuraxis-intact and bilaterally cervically vagotomized dogs anesthetized with -chloralose.

	Methods

Top Abstract Introduction Methods Results Discussion References

This study was approved by the Kansas University Institutional Animal Care and Use Committee. Appropriate guidelines for the use of animals were observed during all aspects of this study.

Adult mongrel dogs (15-22 kg) were anesthetized with -chloralose (100 mg/kg i.v.). Anesthesia was maintained by continuous infusion of -chloralose (20 mg/kg/hr) during the experiment. The dogs were intubated with a cuffed endotracheal tube and ventilated with a Harvard animal ventilator (Harvard, Millis, MA), using oxygen in nitrogen (FiO₂, 0.4) at tidal volumes of 10 to 15 ml/kg and frequencies of 15 to 20 cycles/min. Arterial blood gases were measured periodically (model 175 PH/Blood Gas Analyzer, Corning, Boston, MA) and were maintained within normal limits (PaCO₂, 35-45 mm Hg; pH 7.35-7.45) by adjusting the tidal volume or frequency or by administering sodium bicarbonate. The PaO₂ exceeded 100 mm Hg in all experiments. Body temperature was maintained between 37°C and 38°C by external warming. The animals were paralyzed with pancuronium bromide (0.1 mg/kg i.v.) to avoid artifacts in sympathetic nerve activity measurements caused by muscular movement.

A thoracotomy was performed through the left fifth intercostal space. The left atrial appendage was retracted, and the LCX was dissected free near its origin. Great care was taken to avoid damage to the neural supply to the vessel, and a siliconized 27-gauge hypodermic needle connected to a fine polyethylene catheter was inserted into the LCX. The distal end of the catheter was connected to two polyethylene tubes via a Y-piece, for constant infusion of saline and administration of ATP. To prevent coagulation within the tubing, saline was infused via a Harvard infusion pump (model 2716; Harvard, Southnatic, MA), at a constant rate of 0.1 ml/min, throughout the experiments. A solid-state micromanometer-tipped catheter (TCP2 RN136 F30; Tokai Rika, Aichi, Japan) was inserted in the left ventricle via the left pulmonary vein to monitor LVEDP. Polyethylene catheters were placed in the left jugular vein for i.v. administration of agents and in the femoral artery for measurement of arterial pressure. Arterial blood pressure was monitored with a pressure transducer (DTX Spectramed, Oxnard, CA) and recorded continuously. MAP was derived by electronic integration of the pulsatile pressure signal.

Measurement and recording of RSNA have been described elsewhere (Taneyama et al., 1990). Briefly, the left kidney was exposed and renal sympathetic nerves were isolated and placed on a bipolar silver electrode. Nerve impulses were amplified, rectified, integrated and continuously recorded. To quantitate RSNA, the resting spontaneous nerve discharge before i.c. infusion of saline in each experiment was defined as 100% control value. Atropine sulfate (0.1 mg/kg) and metoprolol (1 mg/kg) were administered to prevent baroreflex-mediated changes in HR from influencing RSNA. The effectiveness of the beta adrenergic and muscarinic blockade was tested with bolus injections of isoproterenol (1 µg/kg) and acetylcholine (1 µg/kg) into the LCX. Atropine and metoprolol abolished effects of these agonists on HR and blood pressure. Data were continuously measured and recorded with a digital audio tape pulse code modulation recorder (RD-110T; TEAC, Montebello, CA) and played back on a multichannel chart recorder (Omnicorder 8 M143; Sanei, Japan).

After completion of the surgical preparation, sufficient time (>1 hr) was allowed for hemodynamic stabilization before initiation of the study. The 14 dogs were divided into two groups, i.e., neuraxis-intact dogs (n = 7) and bilaterally cervically vagotomized dogs (n = 7). After base-line recordings of RSNA, MAP, HR and LVEDP, normal saline (as a control) and three different doses of ATP (100, 200 and 400 µg/kg/min i.c.) were infused in a random fashion into the LCX for 5 min at an infusion rate of 0.5 ml/min, using a Harvard syringe pump (model 2716), in both neuraxis-intact and vagotomized dogs. At least 30 min elapsed between infusions. After determination of responses, saline or ATP was infused into the left jugular vein in the same manner as i.c. infusion. In an additional six neuraxis-intact dogs, the effects of intra-LCX infusions of adenosine (100, 200 and 400 µg/kg/min) on RSNA were determined. Na₂ATP (116F70802; Sigma) and adenosine (90F0870; Sigma) were dissolved in physiological saline just before use, using a stirrer (model 152; VWR Scientific, Bronwill, CA) with water bath.

All data were expressed as mean ± S.E. Comparisons made within experimental protocols were performed using a repeated-measurement analysis of variance. Multiple comparisons between individual means were performed using Newman-Keul's method. Differences with a statistical probability of less than 0.05 were considered significant.

	Results

Top Abstract Introduction Methods Results Discussion References

Figures 1 and 2 show typical tracings of HR, MAP and RSNA during i.c. infusion of ATP in the neuraxis-intact dogs and in the cervically vagotomized dogs, respectively. ATP at 400 µg/kg/min i.c. resulted in marked reduction of RSNA in spite of arterial hypotension in intact animals (fig. 1). In the vagotomized animals, RSNA was increased when MAP was reduced after administration of 400 µg of ATP (fig. 2). The effects of i.c. ATP infusions on measured variables in the neuraxis-intact and vagotomized groups are shown in table 1. ATP at 100 µg/kg/min i.c. resulted in a significant increase in RSNA in both groups. In intact dogs, elevation of RSNA was not observed during ATP infusion of 200 µg/kg/min i.c., and ATP infusions of 400 µg/kg/min i.c. caused marked reduction of RSNA, despite a dose-dependent decrease in MAP during these infusions. Intracoronary infusion of saline as a control did not change any measured variables. In the vagotomized group, RSNA was increased dose-dependently with decreasing MAP during i.c. ATP infusions (RSNA was increased 124, 140 and 170%, with corresponding MAP of 125, 122 and 114 mm Hg, by ATP at 100, 200 and 400 µg/kg/min, respectively).

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Effect of increasing i.c. infusion doses of ATP on RSNA, HR and MAP in an intact dog. Note suppression of RSNA at the i.c. infusions of 200 and 400 µg/kg/min ATP. SBP, systemic blood pressure; mRSNA: mean (integrated) RSNA.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Effect of increasing i.c. infusion doses of ATP on RSNA, HR and MAP in a vagotomized dog. Note that vagotomy completely abolished the ATP-induced suppression of RSNA. See figure 1 for abbreviations.

The dose-response effects of i.v. infusion of ATP in intact and vagotomized groups are summarized in table 2. RSNA was increased and MAP was decreased in a dose-dependent fashion during i.v. ATP infusions in both groups of animals. The effects of i.c. infusions of adenosine on measured variables in intact dogs (table 3) were essentially the same as the effects of i.c. ATP infusions in vagotomized dogs (table 1). HR and LVEDP remained unchanged in all groups during ATP and adenosine infusions.

Maximum changes in RSNA (RSNA) and MAP (MAP) during i.c. and i.v. infusions of ATP in the intact and vagotomized groups are shown in figures 3 and 4. The RSNA was increased dose-dependently with decreasing MAP during i.v. ATP infusion in intact dogs. On the other hand, RSNA was attenuated by 23% with 400 µg/kg/min i.c. ATP infusion, despite the fact that MAP was decreased dose-dependently and to a greater extent than during i.v. ATP infusions (fig. 3). This suppression of RSNA by the i.c. ATP infusion was completely abolished by bilateral cervical vagotomy, and the MAP became identical between the i.c. and i.v. ATP infusions (fig. 4).

View larger version (24K): [in this window] [in a new window]   Fig. 3.   Maximum changes in RSNA (RSNA) and MAP (MAP) produced by i.c. and i.v. infusions of ATP in intact animals. Note that RSNA was attenuated by i.c. infusion of 200 and 400 µg/kg/min ATP, despite the fact that MAP was decreased dose-dependently and more profoundly than with i.v. ATP infusion. IC, i.c. infusion of ATP; IV, i.v. infusion of ATP. * P < .05 compared with i.v. infusion of ATP.

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Maximum changes in RSNA (RSNA) and MAP (MAP) during i.c. and i.v. infusions of ATP in vagotomized animals. The levels of RSNA and MAP were identical for i.c. and i.v. infusions. See figure 3 for abbreviations.

	Discussion

Top Abstract Introduction Methods Results Discussion References

Our data suggest that ATP stimulates left ventricular receptors with vagal afferents, as has been indicated previously (Armour et al., 1994), and that this is one mechanism by which ATP attenuates reflex increases in sympathetic nerve activity in response to arterial hypotension. On the other hand, adenosine does not affect the ventricular receptors in our experiment. Adenosine, as well as ATP, can activate nodose ganglion cardiac afferent neurons associated with ventricular epicardial sensory endings in dogs (Armour et al., 1994). It was demonstrated that infusion of adenosine into the left anterior descending coronary artery produced a dose-dependent reflex increase in blood pressure in humans (Cox et al., 1989). The circumflex coronary artery, into which ATP and adenosine were infused in our study, supplies primarily the left ventricular posterior wall, where cardiac receptors with vagal afferent endings are abundant (Frink and James, 1971; Thoren, 1979). The ventricular receptors consist of barosensitive receptors (Downing, 1979) and chemosensitive receptors (Coleridge and Coleridge, 1979). When ventricular contractility and ventricular filling pressure are increased (for example, by inotropic agents), these barosensitive receptors can be stimulated mechanically, leading to a reflex decrease in efferent sympathetic nerve activity mediated by vagal afferent pathways (Paintal, 1955; Holmberg and Zucker, 1986). In our preliminary study, a large i.v. bolus dose of ATP (4 mg/kg) administered to intact dogs decreased both left ventricular contractility (maximal dP/dt was decreased from 2142 ± 10 to 943 ± 6 mm Hg/sec) and LVEDP (6.9 ± 0.6 to 1.7 ± 0.7 mm Hg), but RSNA was suppressed (100 ± 0 to 9 ± 2%) rather than increased (n = 7). In the present study, RSNA was decreased, even though LVEDP remained unchanged, during i.c. infusion of ATP. Therefore, it is unlikely that ATP infused into the coronary artery stimulated barosensitive ventricular receptors. It is more likely that ATP stimulated chemosensitive ventricular receptors in a manner similar to the Bezold-Jarisch reflex produced by agents such as veratrum alkaloids (Jarisch, 1941; Frink and James, 1971; Chen, 1979). Interestingly, veratrum alkaloids, unlike ATP, may increase ventricular contractility (Goodman and Gilman, 1955). Therefore, they can stimulate both ventricular baroreceptors and chemosensitive receptors, leading to additive reflex reduction of efferent sympathetic nerve activity.

It has been reported that chemical activation of sensory endings with vagal afferents in the inferoposterior left ventricle of dogs (an area supplied primarily by the LCX) results in greater vasodepressor responses than those resulting from chemical activation of receptors in the anterior left ventricle, which is supplied primarily by the left anterior descending artery (Walker et al., 1978). Based on these findings, we chose to infuse agents into the circumflex coronary artery in this study.

As we suspected from our previous study (Taneyama et al., 1991), the results of i.c. infusions of ATP in vagotomized animals (table 1) were essentially the same as the results of i.c. infusions of equivalent doses of adenosine in neuraxis-intact animals (table 3). RSNA was dose-dependently increased in both of these groups as MAP decreased. These results confirm the previously reported findings (Taneyama et al., 1991) that ATP attenuates reflex increases in efferent sympathetic nerve activity via the vagal afferents and that adenosine lacks such a vagal afferent mechanism.

In neuraxis-intact animals, i.v. infusions of ATP decreased MAP dose-dependently, and a dose-dependent increase in RSNA ensued (fig. 3). This is likely because ATP infused i.v. was metabolized to adenosine by the time it reached the left ventricle and the ventricular receptors could not be stimulated to suppress RSNA (fig. 3, top).

Suppression of efferent sympathetic outflow during i.c. infusions of ATP contributed to decreased MAP, along with the direct vasodilatory effects of ATP. Therefore, reduction of MAP was greater with i.c. ATP than with i.v. ATP (fig. 3, bottom). This conclusion is supported by the fact that reduction of MAP became identical with i.v. and i.c. infusions of ATP in vagotomized animals (fig. 4, bottom). ATP is a far more potent vasodilator than adenosine (Burnstock, 1980a,b). However, the reduction of MAP was the same during i.v. ATP infusion in intact dogs, i.c. ATP infusion in vagotomized dogs and i.c. adenosine infusion in intact dogs. Again, this is likely the result of ATP being metabolized to adenosine by the time it reaches peripheral arterial circulation.

Both ATP and adenosine produce dose-dependent decreases in HR (Fukunaga et al., 1982; Taneyama et al., 1991). In this study, HR remained unchanged, even though MAP was decreased. Doses of ATP and adenosine might not have been large enough to affect the cardiac conduction system in fully atropinized dogs. In the previous study, although MAP was decreased, i.v. bolus injections of ATP and adenosine did not produce bradycardia in fully atropinized dogs until doses were increased to 1 mg/kg (Taneyama et al., 1991).

It is well known that, in addition to cardiac vagal afferents, afferent cardiac sympathetic nerve fibers also are involved in cardiac reflexes (Brown, 1979; Casati et al., 1979). We cannot speculate on whether ATP interacted with the cardiac sympathetic afferents in our experiment, because cardiac sympathetic denervation was not performed. If sympathetic afferents had been stimulated by i.c. infusion of ATP, then the vagal afferent mechanism to suppress the efferent sympathetic nerve activity would have been partially inhibited.

ATP is a primary storage form for adenyl compounds. It is released into the circulation by crossing cell membranes via special transport mechanisms (Burnstock, 1980a,b). As an endogenous vasoactive substance, ATP produces vasodilation, as well as depression of myocardial contractility and cardiac conduction (Bloor et al., 1985). These actions counteract the inotropic, chronotropic and vasoconstrictive effects of catecholamines and can prevent sympathetic overstimulation of the cardiovascular system. ATP is found in the coronary circulation during vasodilatory responses to hypoxia (Paddle and Burnstock, 1974; Forrester and Williams, 1977). Plasma levels of ATP may increase to >50 times normal during arterial occlusion (Forrester, 1972). Our data suggest that ATP contributes to hemodynamic stability by stimulating ventricular chemoreceptors, with resulting attenuation of reflex increases in sympathetic activity, in addition to its direct inhibitory effects on the cardiovascular system.

In conclusion, our results suggest that ATP attenuates reflex increases in efferent sympathetic nerve activity possibly by stimulating left ventricular chemoreceptors associated with vagal afferent pathways. This vagal afferent mechanism is one of the mechanisms by which ATP attenuates baroreflex-mediated hemodynamic changes and may contribute to cardiovascular stability in cases of sympathetic overactivity.