Nitric oxide mediates sympathetic vasoconstriction

Transcription

1 Nitric oxide mediates sympathetic vasoconstriction at supraspinal, spinal, and synaptic levels NORIKO IIDA Department of Physiology, School of Medicine, University of Hiroshima, Hiroshima , Japan Iida, Noriko. Nitric oxide mediates sympathetic vasoconstriction at supraspinal, spinal, and synaptic levels. Am. J. Physiol. 276 (Heart Circ. Physiol. 45): H918 H925, The purposes of this study were to investigate the level of the sympathetic nervous system in which nitric oxide (NO) mediates regional sympathetic vasoconstriction and to determine whether neural mechanisms are involved in vasoconstriction after NO inhibition. Ganglionic blockade (hexamethonium), 1 -receptor blockade (prazosin), and spinal section at T 1 were used to study sympathetic involvement. NO was blocked with N -nitro-l-arginine methyl ester (L-NAME). Regional blood flow in the mesenteric and renal arteries and terminal aorta was monitored by electromagnetic flowmetry in conscious rats. L-NAME (3 5 mg/kg iv) increased arterial pressure and peripheral resistance. Ganglionic blockade (25 mg/kg iv) significantly reduced the increase in resistance in the mesentery and kidney in intact and spinal-sectioned rats. Ganglionic blockade significantly decreased hindquarter resistance in intact rats but not in spinal-sectioned rats. Prazosin (200 µg/kg iv) significantly reduced the increased hindquarter resistance. We concluded that NO suppresses sympathetic vasoconstriction in the mesentery and kidney at the spinal level, whereas hindquarter tone is mediated at supraspinal and synaptic levels. nitric oxide synthase inhibitor N -nitro-l-arginine methyl ester; mesenteric, renal, and hindquarter tones; spinal transection; ganglionic blockade; 1 -receptor blockade CURRENT RESEARCH (14, 20, 23, 25) indicates that the tonic release of nitric oxide (NO) from neuronal and endothelial cells mediates vascular resistance directly via vasodilator action on smooth muscles and indirectly via modulation of sympathetic neurotransmission, but the precise regulatory mechanisms involved are not clearly understood (5). The NO-forming enzyme NO synthase (NOS) has been found not only in vascular endothelium and neurons of the sympathetic nervous system but also in nerve fibers innervating blood vessels. It has been demonstrated that NO is not only a mediator of neurotransmitter release (17) but also an inhibitory transmitter of nonadrenergic and noncholinergic (NANC) nerves (3). NO appears to have a dual function in the sympathetic nervous system in vivo: reduction of central sympathetic nerve activity (21) and inhibition of peripheral sympathetic vasoconstriction (29). Addicks et al. (1) reported that blood flow regulation via neuronal NO inhibition of norepinephrine (NE) The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. release becomes more important with the decrease in the size of blood vessels and in the expression of endothelial NOS. Neuronally produced NO has been shown to play a more important role than endothelial NO in some regional vascular beds (3, 16). However, pre- and postjunctional effects remain controversial, and the origin of adventitial NO is yet to be ascertained (6, 28). It is also unknown in which parts and at what level of the vascular system the interaction between NO and NE takes place. Recent research has shown NE to induce endothelium-dependent relaxation via stimulation of endothelial 1 - and 2 -adrenoceptors (19) and to counteract its vasoconstrictor action (14, 20, 25). The inhibitory mechanism of NO in sympathetic neurotransmission is presumably adjusted locally by a complex feedback mechanism (26) similar to the intercellular mechanism, i.e., the guanylate cyclase-cgmp system of endothelial NO in vascular smooth muscle relaxation (4). The side effects of anesthesia may lead to unique alterations in the peripheral circulation of experimental animals, as shown in the effects of anesthesia on the contributions of endothelium-derived vasodilators to circulatory control (15). For this reason, the present study examined the role of NO in blood flow regulation in the conscious resting state. Comprehensive whole body regional and systemic hemodynamic data can provide a profound understanding of flow regulation mechanisms. The aims of this study were to investigate the level of the sympathetic nervous system in which NO mediates regional sympathetic vasoconstriction and to determine whether neural mechanisms are involved in the vasoconstriction after NO inhibition. MATERIALS AND METHODS The present study was conducted in accordance with the guidelines of the Committee on Animal Care and Use of the University of Hiroshima School of Medicine and with the Guide for Care and Use of Laboratory Animals in the Field of Physiological Science of the Physiological Society of Japan (1988). Implantation of flow probes and catheters. Male normotensive control Wistar rats (n 93), wk of age and g body wt, were used in this study. The superior mesenteric artery, renal artery, or terminal aorta was reached retroperitoneally by a left flank incision, and an electromagnetic flow probe (type FC, Nihon Koden; 1, 1.5, or 2 mm ID) was implanted around the vessel under anesthesia with thiamylal sodium (50 mg/kg ip). A polyethylene catheter (PE-10 fused to PE-20) for arterial pressure measurement was inserted into the right femoral artery so that the tip was placed in the terminal aorta or the right common carotid artery in rats with a probe on the terminal aorta. In all rats H /99 $5.00 Copyright 1999 the American Physiological Society

2 H919 another catheter for injection of drugs was inserted into the right external jugular vein (10). Spinal cord transection. After implantation, each rat was isolated in a polyethylene cage containing wood chips. Three to four days later, blood flow and arterial pressure were measured in the conscious resting state in the home cage. To determine at what level of the sympathetic nervous system NO suppresses sympathetic vasoconstriction in regional vascular beds, 27 rats underwent spinal transection. Rats were anesthetized with ether and placed ventrally, and a midline skin incision was made in the dorsal neck. The spinous process of the vertebra prominens was cut and removed. Next the spinal cord was transected between the first and second dorsal vertebra with an ophthalmological scalpel under visual control. Immediately after spinal transection, administration of ether was stopped. Bleeding was minimal and usually ceased within s. A local anesthetic, xylocaine jelly, was applied only around the incision made for transection, and the skin was sutured (11). Arterial pressure and regional blood flows stabilized at a new plateau level several hours after transection. Throughout these procedures and thereafter during the infusion of drugs, blood flow and arterial pressure were recorded continuously. Estimation of sympathetic vasoconstriction. After stable plateau values were reached, N -nitro-l-arginine methyl ester (L-NAME, 3 5 mg/kg iv) and hexamethonium bromide (C6, 25 mg/kg iv) were infused successively at intervals of min. After arterial pressure and blood flow stabilized, the NO precursor L-arginine (L-Arg, 70 mg/kg iv) was injected by bolus. The percent decrease in vascular resistance, calculated as arterial pressure divided by regional blood flow, was used as the magnitude of sympathetic vasoconstriction in resistance vessels of the particular region. Our criterion for the level of the sympathetic nervous system in which NO mediates regional sympathetic vasoconstriction was whether there was a significant decrease in the regional peripheral resistance after ganglionic blockade with C6 in L-NAME-infused intact and spinal-sectioned rats. The presence of peripheral sympathetic vasoconstriction was also estimated from the magnitude of the drop in regional vascular resistance caused by the injection of the 1 -receptor blocking agent prazosin (200 µg/kg iv) given before or after L-Arg. Arterial pressure and blood flow plateaued at a new level within 10 min after the bolus injection of 70 mg/kg L-Arg or 200 µg/kg prazosin. Drugs. L-NAME and prazosin were purchased from Sigma Chemical (St. Louis, MO) and C6 and L-Arg from Nakarai Tesque (Kyoto, Japan). In a preliminary test we checked how long each agent was effective in the conscious rat. The maximal effects of 3 5 mg/kg L-NAME on pressure and flow lasted 90 min, and recovery toward normal levels did not occur for 2 h with 25 mg/kg C6. Statistical analysis. Hemodynamic data are expressed as means SD. Significant difference from the preceding point in the mean value after drug (L-NAME, C6, L-Arg, or prazosin) infusion was compared using two-way ANOVA for repeated measurements. Significant difference between intact and spinal-sectioned rats within each drug was analyzed by using one-way ANOVA. P 0.05 was used as the criterion for statistical significance in all experiments. RESULTS Figure 1 shows simultaneous recordings of arterial pressure and superior mesenteric, renal, or hindquarter flow under conscious and resting states in an intact rat. The recordings show the changes in mean arterial pressure and regional blood flows during the successive infusions of L-NAME and C6 and the bolus injection of L-Arg. Regional blood flows decreased with elevation of arterial pressure during L-NAME infusion for min. C6 was injected min after arterial pressure and blood flow had plateaued at a new stable level. In succession, L-Arg was injected by bolus after infusion of C6 in the same way. Superior mesenteric and renal flows increased, with reduction of arterial pressure after C6 and L-Arg infusions. However, hindquarter flow decreased in parallel with the reduction of arterial pressure. Figure 2 shows the simultaneous recordings of arterial pressure and regional blood flow in an acute spinal-sectioned rat. Several hours after spinal transection, arterial pressure and regional blood flow stabilized at a new plateau level, and they were measured again in the conscious resting state. Even in spinalsectioned rats, regional blood flow decreased with elevation of arterial pressure after infusion of L-NAME. In Figs. 1 and 2, mesenteric, renal, and hindquarter hemodynamics were not measured simultaneously in the same animal but in separate animals under control conditions or after spinal cord section. The entire measurement with drug infusions in an intact or spinal-sectioned rat was completed within 90 min. The duration of the experiment was not a factor in the vascular responses. Figure 3 shows arterial pressure, heart rate, and regional blood flow responses to L-NAME, C6, and L-Arg from 36 intact and 27 spinal-sectioned rats. Mean arterial pressure significantly increased and mean superior mesenteric, renal, and hindquarter flows, as well as heart rate, significantly decreased during infusion of L-NAME in intact and spinal-sectioned rats. Further injections of C6 and L-Arg did not change blood flow in any of the beds in intact rats, although arterial pressure did fall and heart rate returned to the control level. This suggests that blood flow autoregulation was maintained even after inhibition of NO synthesis in the mesentery, kidney, and hindquarter in intact rats. On the other hand, mesenteric and renal flows in spinalsectioned rats increased with decrease of arterial pressure by infusions of C6 and L-Arg, but hindquarter flow did decrease by C6, and L-Arg had no additional effect. Heart rate still did not recover to the control level. These findings suggest that hindquarter flow autoregulation was disrupted by the depression of heart rate caused by spinal cord section and that local neurohumoral factors had a great influence on hindquarter blood flow. With spinal cord section, mesenteric resistance significantly increased [change in resistance ( R) % of control] and hindquarter resistance significantly decreased ( R % of control), whereas the change in renal resistance was not significant ( R % of control). The regional local mechanism appeared to be partially maintained by higher central nervous system control in the conscious resting state.

3 H920 Fig. 1. Simultaneous recordings of arterial pressure (AP) and superior mesenteric (MF), renal (RF), and hindquarter (HQF) flows in an intact rat. N -nitro-l-arginine methyl ester (L-NAME) and hexamethonium bromide (C6) were infused successively for periods shown as horizontal bars to total doses of 3.5, 4.3, or 5.0 mg/kg iv and 25 mg/kg iv, respectively. At arrow, L-arginine (L-Arg) was injected at 70 mg/kg iv by bolus. Figure 4 shows the percent changes in mesenteric, renal, and hindquarter resistances caused by additional infusions of L-NAME, C6, and L-Arg. The changes in regional vascular resistances are given as percentages of stable plateau values in the conscious resting state before L-NAME infusion in intact rats (n 36) and after spinal cord section in spinal-sectioned rats (n 27). Mesenteric and renal resistances were increased markedly by L-NAME infusion, and the percent increments in mesenteric and renal resistances were significantly decreased by infusions of C6 and L-Arg in intact and spinal-sectioned rats. Mesenteric resistances in both groups of rats returned to control level before L-NAME infusion. There was no significant difference between intact and spinal-sectioned rats in the percent change of mesenteric resistance after infusion of L-NAME, C6, or L-Arg. However, the percent increments in renal resistances after L-NAME, C6, and L-Arg infusions were greater in spinal-sectioned than in intact rats, presumably because of increased resistance produced by spinal cord section. On the other hand, the percent increment in hindquarter resistance after L-NAME infusion was not significantly different between intact and spinal-sectioned rats. Intact and spinal-sectioned rats differed in terms of hindquarter resistance after C6 infusion. The elevated hindquarter resistance after L-NAME was significantly decreased by infusion of C6 in intact but not in spinal-sectioned rats. Furthermore, hindquarter resistance (8) in intact and spinal-sectioned rats was unchanged by injection of L-Arg. The influences of L-NAME on vascular resistances were different among regional vascular beds. To elucidate whether L-NAME-induced hindquarter tone is mediated by peripheral sympathetic nerves at the synaptic level, we examined the effect of prazosin on hindquarter resistance after infusion of L-NAME in ganglion-blocked rats. Arterial pressure was increased and regional blood flows were decreased by infusion of L-NAME after ganglionic blockade. The reduced superior mesenteric and renal flows after L-NAME were partially restored by injection of L-Arg, but hindquarter flow was unchanged. Hindquarter flow was increased by supplement of prazosin given before or after L-Arg, whereas superior mesenteric and renal flows were unchanged. Figure 5 shows percent changes in mesenteric, renal, and hindquarter resistances caused by intravenous infusions of L-NAME, L-Arg, and prazosin in ganglionblocked rats with C6 (n 30). Here, the percent changes in regional vascular resistances after L-NAME, L-Arg, and prazosin infusions are given as the percentage of the stable plateau value in the conscious resting state after ganglionic blockade, but the effects of C6 on regional vascular resistances are given as the percentage of the control value before C6 infusion. Regional vascular resistances also increased signifi-

4 H921 Fig. 2. Simultaneous recordings of arterial pressure and superior mesenteric, renal, and hindquarter flows in a spinal-sectioned rat. L-NAME and C6 were infused successively for periods shown as horizontal bars to total doses of 5.0 mg/kg iv and 25 mg/kg iv, respectively. At arrow, L-Arg was injected at 70 mg/kg iv by bolus. CS, cord section. cantly by infusion of L-NAME after ganglionic blockade. NO-mediated perivascular nerves and/or endothelial NO production appeared to be involved in regional vasoconstrictor responses to L-NAME. The elevated mesenteric and renal resistances by L-NAME decreased significantly after bolus injection of L-Arg in both groups of rats, whereas the hindquarter resistance did not decrease significantly. The elevated hindquarter resistance decreased significantly in conjunction with prazosin given before or after infusion of L-Arg Fig. 3. Arterial pressure, heart rate (HR), and superior mesenteric, renal, and hindquarter flows in response to L-NAME, C6, and L-Arg from 36 intact (s) and 27 spinal-sectioned rats (r). C and CS, resting control state and cord section, respectively. Statistical comparisons were made by 2-way ANOVA for repeated measurements: *P 0.05 for significant difference from preceding point in arterial pressure, heart rate, and regional blood flow caused by infusions of L-NAME, C6, and L-Arg and by spinal cord section.

5 H922 Fig. 4. Effects of L-NAME, C6, and L-Arg infusions on mesenteric, renal, and hindquarter resistances in intact (n 36) and spinal-sectioned rats (n 27). Values are changes ( R) caused by L-NAME, C6, and L-Arg expressed as percentage of control value before L-NAME infusion in intact rats and stable plateau value after spinal cord section in spinal-sectioned rats. *P 0.05 vs. L-NAME infusion; **P 0.05 vs. L-NAME C6 infusion (2-way ANOVA for repeated measurements); # P 0.05 vs. corresponding intact rats within each drug infusion (1-way ANOVA). Statistical significance at P Fig. 5. Percent changes in superior mesenteric, renal, and hindquarter resistances caused by infusions of L-NAME (3.5 mg/kg iv), L-Arg (70 mg/kg iv), and prazosin (200 µg/kg iv) after ganglionic blockade with C6 (25 mg/kg iv). Values are expressed as percent changes from stable plateau values in conscious resting state after ganglionic blockade; percent change in regional vascular resistances by C6 is given as percentage of control value in conscious resting state before C6 infusion. *P 0.05 vs. L-NAME infusion; **P 0.05 vs. L-NAME L-Arg infusion within each vascular group (2-way ANOVA for repeated measurements).

6 H923 (data for prazosin given before L-Arg infusion are not shown here). 1 -Receptor-mediated peripheral sympathetic vasoconstriction was indicated to be implicated in elevated hindquarter resistance by L-NAME (19, 20). Furthermore, mesenteric resistance in resting control rats was significantly increased ( R % of control) by infusion of C6, suggesting that mesenteric tone was suppressed by central sympathetic nerve activity in the conscious resting state, whereas renal and hindquarter resistances were significantly decreased ( R and % of control, respectively). In comparison with L-NAME-infused intact rats (Fig. 4), the percent decreases in mesenteric, renal, and hindquarter resistances after C6 infusion were , , and % of L-NAME, respectively. The enhanced C6 responses after L-NAME are possibly due to sympathetic nerve activity augmented by NO inhibition but not to increased resistance produced by L-NAME. DISCUSSION The aim of this study was to investigate whether spinal or supraspinal mechanisms control NO-mediated sympathetic vasoconstriction in the mesentery, kidney, and hindquarter. Our criterion for the level of the sympathetic nervous system was whether there was a significant difference in vascular response to C6 in spinal-sectioned compared with intact rats. The criterion for differentiating between supraspinal and synaptic mechanisms was also whether there was a significant difference in vascular response to 1 - receptor blockade with prazosin in ganglion-blocked rats compared with spinal-sectioned rats. The percent increments of renal resistances produced by L-NAME were significantly greater in spinalsectioned than in intact rats, presumably because of some vasoconstrictor metabolites (ANG II and vasopressin) produced by spinal cord section or release of tonic depression of the NO system in the intact state (i.e., greater contribution of NO to regional resistance when the sympathetic system is disrupted), whereas there was no significant difference between intact and spinalsectioned rats in the percent increments of mesenteric and hindquarter resistances. The elevated mesenteric and renal resistances by L-NAME were significantly reduced by infusion of C6 in intact and spinal-sectioned rats. However, C6 did not significantly decrease the hindquarter resistance in spinal-sectioned rats but significantly decreased the hindquarter resistance in intact rats. The elevated hindquarter resistances were also decreased significantly by bolus injection of the 1 -blocker prazosin after ganglionic blockade. NOmediated sympathetic vasoconstrictions in the mesentery and kidney were spinal in origin, but hindquarter tone was supraspinal in origin. Mesenteric and renal tones were potentiated more at the spinal level by spinal cord section, whereas hindquarter tone was at the synaptic level (22, 23). These neuronal mechanisms can possibly be explained by the hypothesis (30) that the supraspinal inhibitory influences on spinal sympathetic generators were released by spinal cord section and the outflow from the spinal generator to the mesentery and kidney was potentiated and also that the supraspinal excitatory input to the hindquarter was replaced by NE releases from sympathetic nerve endings by spinal cord section or ganglionic blockade. NE releases from nerve endings in the hindquarter must be modified at preand/or postsynaptic levels. The enhanced NE release from sympathetic nerve endings and/or the prejunctional depression of NO-mediated NANC nerve activity must be a likely cause of the depressed response to L-Arg in the hindquarter. In addition, hindquarter tone must be more preferentially controlled by neuronal than by endothelial NO. Alternatively, we cannot dismiss the possibility (23) that L-NAME is a muscarinic receptor antagonist and can produce abnormal tone in regional vascular beds by blocking prejunctional M 2 receptors. However, the blocking effect of L-NAME on muscarinic receptors is an unlikely cause of the depressed response to L-Arg in the hindquarter. This is because depressed responses to L-Arg in the hindquarter have also been observed by the infusion of N G -monomethyl-l-arginine (an inhibitor of NOS) (8), which does not have a blocking effect on muscarinic receptors. In the present study, higher resting resistance was observed in the kidney and hindquarter after spinal cord section and ganglionic blockade. The increased resting resistance may be partially mediated by other neurohumoral factors such as ANG II and arginine vasopressin (AVP) at regional vascular beds. This is because spinal cord section or ganglionic blockade must enhance the vasoconstrictor actions of a number of neurohumoral systems, as well as inhibition of NO synthesis (16). NO is not only a negative modulator of sympathetic nerve activity but also an important modulator of the vasoconstrictor influence of ANG II and AVP in regional blood circulation. The regulation of regional blood flow is intimately associated with the local interaction between the vasodilator NO and several vasoconstrictor systems such as ANG II, AVP, and the sympathetics. In addition, the diverse regional effects in NO-mediated peripheral and organ perfusions may reflect variation in the sensitivity of different vascular beds to these neurohumoral factors (24). We did not measure plasma levels of ANG II and AVP before and/or after spinal cord section and ganglionic blockade. However, the balance of the vasodilator NO and the vasoconstrictor systems at renal and hindquarter beds will most likely shift toward the vasoconstrictor state when the plasma levels of ANG II and AVP and/or the sensitivity of vascular beds are elevated by spinal cord section or ganglionic blockade. It is conceivable that higher resting resistances in the kidney and hindquarter are due to increased vascular sensitivity to circulating vasoconstrictor factors potentiated by spinal cord section or ganglionic blockade. Zanzinger et al. (29) reported that intravenous administration of the -adrenergic receptor antagonist prazosin almost completely reversed hypertension caused by

7 H924 L-NAME and that inhibition of peripheral sympathetic vasoconstriction is an important mechanism of in vivo NO vasodilation. In contrast, Huang et al. (9) reported that the vasoconstrictor and pressor responses to L-NAME were not attenuated by pretreatment with the ganglion blocker C6, suggesting that a neurogenic mechanism is not involved in L-NAME-induced abnormal tone and resultant hypertension. Our experimental results with conscious intact rats demonstrated that increased mesenteric, renal, and hindquarter resistances by L-NAME were significantly decreased by the ganglion blocker C6 and that the percent decreases by C6 were greater in L-NAMEinfused intact rats than in resting control rats. A neurogenic mechanism was involved in regional vasoconstrictor responses to L-NAME. On the other hand, the vasoconstrictor responses to L-NAME observed in ganglion-blocked vascular beds might be explained by NO-mediated perivascular nerve activity (27) and endothelial NO production potentiated by ganglionic blockade or spinal cord section, in addition to the vasoconstrictor actions of the neurohumoral system (16). The difference between the findings of Zanzinger et al. (29) and Huang et al. (9) is most likely due to the different species studied (e.g., cat and rat) and different anesthesia (e.g., -chloralose and thiobutabarbital sodium salt) rather than some universal mechanism applicable to all species. Neuronal NO appears to regulate regional blood flows interdependently at the supraspinal, spinal, and synaptic levels. For a profound understanding of the NO-mediated blood flow regulatory mechanism in the regional vascular beds, further exploration of the cellular mechanisms (4, 7, 12), the membrane K channel (2, 13, 18), and the neural mechanism of NANC nerve fibers (27) is needed. In conclusion, the results of the present study show that 1) NO suppresses sympathetic vasoconstriction in the mesentery and kidney at the spinal level and 2) in the hindquarter, sympathetic vasoconstriction is suppressed by NO at the supraspinal and synaptic levels. The author thanks Drs. Juro Iriuchijima and Nobukuni Ogata for helpful support. Address for reprint requests: N. Iida, Dept. of Physiology, School of Medicine, University of Hiroshima, Kasumi, Minami-ku, Hiroshima , Japan ( iida@mcai.med.hiroshima-u.ac.jp). Received 14 April 1998; accepted in final form 21 October REFERENCES 1. Addicks, K., W. Bloch, and M. Feelisch. Nitric oxide modulates sympathetic neurotransmission at the prejunctional level. Microsc. Res. Tech. 29: , Bolotina, V. M., S. Najibi, J. J. Palacino, P. J. Pagano, and R. A. Cohen. Nitric oxide directly activates calcium-dependent potassium channels in vascular smooth muscle. Nature 368: , Bredt, D. S., P. M. Hwang, and S. H. Snyder. Localization of nitric oxide synthase indicating a neural role for nitric oxide. Nature 347: , Bryan, R. M., M. L. Steenberg, Jr., M. Y. Eichler, T. D. Johnson, M. W. G. Swafford, and M. S. Suresh. Permissive roleofnoin 2 -adrenoceptor-mediated dilations in rat cerebral arteries. Am. J. Physiol. 269 (Heart Circ. Physiol. 38): H1171 H1174, Cohen, R. A., B. Tesfamariam, and R. M. Weisbrod. Endothelium inhibits adrenergic neurotransmission. In: Endothelium- Derived Relaxing Factors, edited by G. M. Rubanyi and P. M. Vanhoutte. Basel: Karger, 1990, p Faber-Zuschratter, H., and G. Wolf. Ultrastructural distribution of NADPH-diaphorase in cortical synapses. Neuroreport 5: , Falcone, J. C., L. Kuo, and G. A. Meininger. Endothelial cell calcium increases during flow-induced dilation in isolated arterioles. Am. J. Physiol. 264 (Heart Circ. Physiol. 33): H653 H659, Gardiner, S. M., A. M. Compton, T. Bennett, R. M. J. Palmer, and S. Moncada. Control of regional blood flow by endothelium-derived nitric oxide. Hypertension 15: , Huang, M., M. L. Leblanc, and R. L. Hester. Systemic and regional hemodynamics after nitric oxide synthase inhibition: role of a neurogenic mechanism. Am. J. Physiol. 267 (Regulatory Integrative Comp. Physiol. 36): R84 R88, Iida, N. Different flow regulation mechanisms between celiac and mesenteric vascular beds in conscious rats. Hypertension 25: , Iriuchijima, J. Regional distribution of vasoconstrictor tone in acute spinal rats. Jpn. J. Physiol. 40: , Jameson, M., F. X. Dai, T. Luscher, J. Skopec, A. Diederich, and D. Diederich. Endothelium-derived contracting factors in resistance arteries of young spontaneously hypertensive rats before development of overt hypertension. Hypertension 21: , Khan, S. A., W. R. Mathews, and K. D. Meisheri. Role of calcium-activated K channels in vasodilation induced by nitroglycerine, acetylcholine and nitric oxide. J. Pharmacol. Exp. Ther. 267: , Lacolley, P. J., S. J. Lewis, and M. J. Brody. Role of sympathetic nerve activity in the generation of vascular nitric oxide in urethane-anesthetized rats. Hypertension 17: , Loeb, A. L., I. Godeny, and D. E. Longnecker. Anesthetics alter relative contributions of NO and EDHF in rat cremaster muscle microcirculation. Am. J. Physiol. 273 (Heart Circ. Physiol. 42): H618 H621, Manning, R. V., Jr., L. Hu, and T. D. Williamson. Mechanisms involved in the cardiovascular-renal actions of nitric oxide inhibition. Hypertension 23: , Modin, A., E. Weitzberg, T. Hokfelt, and J. M. Lundberg. Nitric oxide synthase in the pig autonomic nervous system in relation to the influence of N G -nitro-l-arginine on sympathetic and parasympathetic vascular control in vivo. Neuroscience 62: , Najibi, S., C. L. Cowan, J. J. Palacino, and R. A. Cohen. Enhanced role of potassium channels in relaxations to acetylcholine in hypercholesterolemic rabbit carotid artery. Am. J. Physiol. 266 (Heart Circ. Physiol. 35): H2061 H2067, Nase, G. P., and M. A. Boegehold. Postjunctional 2 -adrenoceptors are not present in proximal arterioles of rat intestine. Am. J. Physiol. 274 (Heart Circ. Physiol. 43): H202 H208, Pegoraro, A. A., O. A. Carretero, D. H. Sigmon, and W. H. Beierwaltes. Sympathetic modulation of endothelium-derived relaxing factor. Hypertension 19: , Sakuma, I., H. Togashi, M. Yoshioka, H. Saito, M. Yanagida, M. Tamura, T. Kobayashi, H. Yasuda, S. S. Gross, and R. Levi. N G -methyl-l-arginine, an inhibitor of L-arginine-derived nitric oxide synthesis, stimulates renal sympathetic nerve activity in vivo. A role for nitric oxide in the central regulation of sympathetic tone? Circ. Res. 70: , Sander, M., P. G. Hansen, and R. G. Victor. Sympathetically mediated hypertension caused by chronic inhibition of nitric oxide. Hypertension 26: , Schwarz, P., R. Diem, N. J. Dun, and U. Forstermann. Endogenous and exogenous nitric oxide inhibits norepinephrine release from rat heart sympathetic nerves. Circ. Res. 77: , 1995.

8 H Sigmon, D. H., and W. H. Beierwaltes. Angiotensin II: nitric oxide interaction and the distribution of blood flow. Am. J. Physiol. 265 (Regulatory Integrative Comp. Physiol. 34): R1276 R1283, Tesfamariam, B., R. M. Weisbrod, and R. A. Cohen. Endothelium inhibits responses of rabbit carotid artery to adrenergic nerve stimulation. Am. J. Physiol. 253 (Heart Circ. Physiol. 22): H792 H798, Thorin, E., and J. Atkinson. Modulation by the endothelium of sympathetic vasoconstriction in an in vitro preparation of the rat tail artery. Br. J. Pharmacol. 111: , Toda, N., and T. Okamura. Regulation by nitroxidergic nerve of arterial tone. News Physiol. Sci. 7: , Wang, T., Z. Xie, and B. Lu. Nitric oxide mediates activitydependent synaptic suppression at developing neuromuscular synapses. Nature 374: , Zanzinger, J., J. Czachurski, and H. Seller. Inhibition of sympathetic vasoconstriction is a major principle of vasodilation by nitric oxide in vivo. Circ. Res. 75: , Zong, S., M. J. Kenney, and G. L. Gebber. High power, low frequency components of cardiac, renal, splenic and vertebral sympathetic nerve activities are uniformly reduced by spinal cord transection. Brain Res. 556: , 1991.