We examined modulation by nitric oxide (NO) of sympathetic neurotransmitter release and vasoconstriction in the isolated pump-perfused rat kidney. Electrical renal nerve stimulation (RNS; 1 and 2 Hz) increased renal perfusion pressure and renal norepinephrine (NE) efflux. Nonselective NO synthase (NOS) inhibitors [N ω-nitro-l-arginine methyl ester (l-NAME) orN ω-nitro-l-arginine], but not a selective neuronal NO synthase inhibitor (7-nitroindazole sodium salt), suppressed the NE efflux response and enhanced the perfusion pressure response. Pretreatment with l-arginine prevented the effects of l-NAME on the RNS-induced responses. 2-(4-Carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide (carboxy-PTIO), which eliminates NO by oxidizing it to NO2, suppressed the NE efflux response, whereas the perfusion pressure response was less susceptible to carboxy-PTIO. 8-Bromoguanosine cGMP suppressed and a guanylate cyclase inhibitor [4H-8-bromo-1,2,4-oxadiazolo(3,4-d)benz(b)(1,4)oxazin-1-one] enhanced the RNS-induced perfusion pressure response, but neither of these drugs affected the NE efflux response. These results suggest that endogenous NO facilitates the NE release through cGMP-independent mechanisms, NO metabolites formed after NO2 rather than NO itself counteract the vasoconstriction, and neuronal NOS does not contribute to these modulatory mechanisms in the sympathetic nervous system of the rat kidney.
- sympathetic nerves
- nitric oxide synthase inhibitor
- guanylate cyclase inhibitor
- nitric oxide scavenger
In the cardiovascular system, NO has been demonstrated to modulate sympathetic control of vascular tone and resistance. Inhibition of NO synthase (NOS) with l-arginine derivatives facilitates vasoconstriction induced by electrical stimulation of the sympathetic nerves or by application of exogenous norepinephrine (NE) in vitro and in vivo (6, 16, 17, 20, 21, 25), indicating that endogenous NO counteracts adrenergically induced vasoconstriction by acting at least on the postsynaptic site.
NO might also modulate the sympathetic nervous system by acting on the presynaptic site, but experimental results have been inconsistent. NOS inhibition facilitates NE release induced by the nerve stimulation in the in vivo dog kidney (6) and in the isolated rat heart (19). On the other hand, the nerve stimulation-induced NE release is suppressed in the isolated rat mesenteric vasculature (24, 25) or remains unaffected in the isolated rat tail artery (21) and rabbit pulmonary artery (20) by NOS inhibition. Thus the modulation by NO of sympathetic neurotransmitter release varies among tissues, organs, or species.
In the rat kidney, there has been no direct information on the neural NE release during NOS inhibition, whereas enhancement by NOS inhibition of the adrenergically induced vasoconstriction was well demonstrated (17, 28), and contribution of the guanylate cyclase-cGMP pathway to the sympathetic nervous system has not been fully understood. In addition, it is unknown whether NO produced by neuronal NOS (nNOS), the inhibition of which is supposed to enhance neural NE release in the isolated guinea pig atria (5), modulates the sympathetic control of renal vascular resistance.
To clarify the above-mentioned issue, in the present study we examined the effects of NO- and cGMP-related drugs on increases in renal NE efflux and perfusion pressure induced by renal nerve stimulation (RNS) in the isolated pump-perfused rat kidney.
All procedures for handling animals were approved by the Animal Experimentation Committee of Tohoku University Graduate School of Pharmaceutical Sciences. The isolated pump-perfused rat kidney preparation was made according to the methods as described previously (7). Male Wistar rats, weighing 250–350 g, were anesthetized with pentobarbital sodium (50 mg/kg ip), and the kidneys were exposed via a midline incision. The aorta was ligated just above the origin of the left renal artery, and a PE-50 cannula was immediately inserted into the renal artery through the aorta. The kidney was flushed for several minutes by infusion (3 ml/min) of oxygenated Tyrode solution of the following composition (in mM): 137 NaCl, 2.7 KCl, 1.8 CaCl2, 1.1 MgCl2, 12 NaHCO2, 0.42 NaH2PO4, and 5.6d(+)-glucose. The left kidney was carefully separated from surrounding tissue, and the left renal vein and the left ureter were cut. The kidney was then removed from the rat and immediately placed in a water-jacketed chamber, the temperature of which was maintained at 37°C with thermostatically controlled water circulator (model NTT-1200, EYELA, Tokyo, Japan). The right renal artery was cannulated via the mesenteric artery, and the right kidney was also isolated in a similar manner. The right kidney was placed in another set of the perfusion chamber and treated in the same way as the left kidney. The kidney was perfused at a constant flow rate of 5 ml/min by using a peristaltic pump (model MP-3 N-H, EYELA). The perfusate was gassed with 95% O2-5% CO2 and pumped through a warming coil (37°C) fitted with a bubble trap. The perfusate passed through the kidney only once and was not recirculated. Bipolar platinum electrodes were placed around the renal artery to stimulate the renal nerves. One of the sidearms of the perfusion system was connected to a pressure transducer (model MPU-0.5, Nihon Kohden, Tokyo, Japan) for continuous monitoring of perfusion pressure with a carrier amplifier (model AP-601G, Nihon Kohden) and a linear-writing pen recorder (model SR6511, Graphtechs, Yokohama, Japan). More than 1 h was allowed for stabilization before start of experiments. The kidneys were randomly divided into nine experimental groups.
Drug solution was infused into the perfusate at 0.05 ml/min through sidearms of the perfusion system by using motor-driven infusion pumps (model 100, KD Scientific). Drug concentrations given in this paper are calculated concentrations in the perfusate.
RNS-induced increases in NE efflux and perfusion pressure were evaluated before and during treatment with a nonselective NOS inhibitor, N ω-nitro-l-arginine methyl ester (l-NAME) orN ω-nitro-l-arginine (l-NNA); a selective nNOS inhibitor, 7-nitroindazole sodium salt (7-NINA); an NO scavenger, 2-(4-carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide (carboxy-PTIO); a membrane-permeable cGMP analog, 8-bromoguanosine cGMP (8-BrcGMP); or a guanylate cyclase inhibitor, 4H-8-bromo-1,2,4-oxadiazolo(3,4-d)benz(b)(1,4)oxazin-1-one (NS-2028).
In groups 1–7, RNS was applied at increasing frequencies of 1 and 2 Hz (duration, 0.5-ms; supramaximal voltage, 30–50 V) for 2.5 min each. The perfusate sample exiting the kidney was collected before RNS was started and during the last 30 s of each RNS for determination of renal NE efflux. A series of RNS and sampling was performed three or four times at 20-min intervals: one control period followed by consecutive drug infusion periods. Our laboratory has previously confirmed that the RNS-induced changes in renal perfusion pressure and renal NE efflux were stable in four consecutive experimental periods (7). Infusion ofl-NAME (100 and 200 μM; group 1,n = 6; group 2, n = 6),l-NNA (100 and 200 μM; group 3,n = 6), 7-NINA (10 and 100 μM; group 4,n = 5), carboxy-PTIO (1, 3, and 10 μM; group 5, n = 6), 8-BrcGMP (1, 10, and 100 μM; group 6,n = 6), or NS-2028 (0.1 and 1 μM; group 7,n = 6) was started 15 min before the start of RNS in the second and third (groups 1, 3, 4, and 7) or in the second, third, and fourth (groups 5 and 6) experimental periods at increasing concentrations. In group 2, effects of l-NAME were examined in the presence ofl-arginine (1 mM). l-Arginine was infused throughout the experiments beginning at 15 min before the start of the first RNS. In groups 8 and 9, NE was infused at increasing concentrations of 40 and 60 nM (group 8) or at 60 nM (group 9). Perfusion pressure markedly increased after the start of NE infusion and then gradually decreased to a constant level. Because the initial increase was not stable in some preparations, the values were obtained more than 5 min after the start of infusion. A series of NE infusions was performed three or four times at 20-min intervals: one control period followed by two or three consecutive drug infusion periods. Infusion of l-NNA (100 and 200 μM; group 8, n = 5) or carboxy-PTIO (1, 3 and 10 μM; group 9, n = 6) was started 15 min before the start of NE infusion in the second and third or the second, third, and fourth experimental periods at increasing concentrations.
Determination of renal NE efflux.
Perfusate samples were transferred into chilled tubes. As an internal standard, 250 pg of 3,4-dihydroxybenzylamine were added to each tube, and catecholamines were extracted from perfusate by means of alumina adsorption. NE concentration was determined by high-performance liquid chromatography (pump, model LC-100; autoinjector, model CMA-200; column, Biophase ODS-IV 4.0-mm ID × 110-mm length, Bioanalytical Systems, West Lafayette, IN) with electrochemical detection (model LC-4C, Bioanalytical Systems), as recently described (8). The NE efflux was calculated by multiplying the perfusate NE concentration by the perfusion rate.
l-NAME, l-NNA, l-arginine HCl, and NE [(−)-arterenol bitartrate] were obtained from Sigma Chemical (St. Louis, MO). Carboxy-PTIO, 7-NINA · H2O, and NS-2028 were obtained from Dojindo (Kumamoto, Japan), TOCRIS (Bristol, UK), and Alexis Biochemicals (San Diego, CA), respectively. NS-2028 was dissolved in ethanol and diluted with the perfusate (final concentration of ethanol in the perfusate was 0.01%). Other drugs were dissolved in distilled water and diluted with the perfusate.
All values are expressed as means ± SE. RNS- or NE-induced responses in three or four experimental periods were compared by using single-factor analysis of variance for repeated measures, and Dunnett's test was applied to analyze statistical differences between the values obtained in the first experimental period and those in other experimental periods. Differences at P < 0.05 were considered to be statistically significant.
Ranges of basal perfusion pressure and NE efflux (before RNS) were 50–60 mmHg and 0.1–0.3 ng/min, respectively, in all groups. Treatment with l-NAME (group 1) andl-NNA (group 2) increased basal perfusion pressure from 54 ± 3 mmHg (control) to 56 ± 4 mmHg (100 μM) and 59 ± 4 mmHg (200 μM) (P < 0.01), and from 54 ± 3 mmHg (control) to 57 ± 4 mmHg (100 μM) and 64 ± 5 mmHg (200 μM) (P < 0.01), respectively. NS-2028 (group 7) also increased the pressure from 60 ± 2 mmHg (control) to 62 ± 2 mmHg (0.1 μM) and 65 ± 2 mmHg (1 μM) (P < 0.01). Basal NE efflux was unaffected. Other drugs had no effects on basal perfusion pressure and NE efflux.
RNS elevated NE efflux and perfusion pressure in a frequency-dependent manner in the first experimental period of groups 1–7. Figure 1 is a trace of original record of renal perfusion pressure from group 3. The RNS-induced changes in NE efflux and perfusion pressure ingroups 1 and 3 are shown in Fig.2; the RNS-induced increases in NE efflux were attenuated and the increases in perfusion pressure were enhanced during treatment with l-NAME (group 1) andl-NNA (group 3) at 100 and 200 μM. Figure 1demonstrates an example of the enhancement by l-NNA treatment of the perfusion pressure response. l-NNA also enhanced the increases in perfusion pressure induced by NE infusion (group 8, Fig. 2).
The effects of NOS inhibitors and other drugs were evaluated by calculating the RNS- or NE infusion-induced responses in the drug treatment periods as percentage of the control responses (the changes obtained in the first experimental periods). The results are shown in Figs. 3-6. l-NAME (Fig.3) and l-NNA (Fig.4) at 200 μM reduced the 2-Hz RNS-induced NE efflux response by ∼50% of the control response and enhanced the perfusion pressure response by ∼200–250% (2- to 2.5-fold) of the control response. In the presence of 1 mMl-arginine (group 2), l-NAME failed to affect the RNS-induced responses except that l-NAME at 200 μM enhanced the perfusion pressure response to 2-Hz RNS by ∼30% (Fig. 3). 7-NINA (group 4) at 10 and 100 μM did not affect the RNS-induced NE efflux and perfusion pressure responses (Fig. 3).
Carboxy-PTIO at 1, 3, and 10 μM (group 5) attenuated the RNS-induced NE efflux response in a concentration-dependent manner (Fig. 5) with slight enhancement of the 1-Hz RNS-induced perfusion pressure response at 3 and 10 μM (Fig. 5). Carboxy-PTIO enhanced the NE-induced perfusion pressure response (group 9, Fig. 5), but the enhancement (by ∼40% of the control response at 10 μM) was smaller than the enhancement observed with l-NNA (group 8, Fig. 4).
8-BrcGMP (group 6) attenuated and NS-2028 (group 7) enhanced the RNS-induced perfusion pressure response in a concentration-dependent manner (Fig. 6). However, neither 8-BrcGMP nor NS-2028 affected the RNS-induced NE efflux response (Fig. 6).
The present study aimed to clarify whether endogenous NO modulates the renal sympathetic nervous system at the pre- and postsynaptic sites. Effects of NOS inhibitors and cGMP-related drugs on increases in renal NE efflux (an index for neural NE release) and perfusion pressure (an index for vasoconstriction) induced by RNS (activation of the renal sympathetic nerves) were examined in the isolated pump-perfused rat kidney. We obtained new findings suggesting that in the rat kidney1) endogenous NO, produced by constitutive endothelial NOS (eNOS) but not by nNOS, acts on the presynaptic site to facilitate neural NE release during activation of the renal sympathetic nervous system; 2) NO metabolites act on the postsynaptic site to counteract the neurally induced vasoconstriction more potently than NO itself; and 3) the presynaptic action of NO is independent of the guanylate cyclase-cGMP pathway.
The NOS inhibitors l-NAME and l-NNA suppressed the RNS-induced increase in NE efflux. Although we did not measure renal NE clearance, the renal spillover of NE determined as NE efflux (total of NE output from the renal vein and the ureter) may be able to reflect changes in the neural NE release. Despite the reduced NE efflux response, the NOS inhibitors markedly enhanced the RNS-induced increase in perfusion pressure. Pretreatment with excessive amount of the NOS substrate l-arginine almost abolished the effects ofl-NAME on the RNS-induced responses. These results demonstrate that NOS inhibition reduces neural NE release during sympathetic activation in the rat kidney, as has been observed in the isolated rat mesenteric vasculature (24, 25).l-NNA also enhanced the exogenous NE (NE infusion)-induced increases in perfusion pressure. Taken together, the data suggest that NO may play a facilitatory role in the neurotransmitter release and an inhibitory role in the vasoconstriction during activation of the sympathetic nervous system of the rat renal vasculature. Our present study is the first that revealed the pre- and postsynaptic modulation by endogenous NO in the rat kidney.
NOS has three isozymes [eNOS, nNOS, and inducible NOS (iNOS)], each of which is inhibited by l-NAME andl-NNA. 7-NINA is one of the nNOS selective inhibitors (2). In the isolated guinea pig atria, 7-NINA enhances positive chronotropic and inotropic responses induced by cardiac nerve stimulation but not the responses induced by exogenous NE (5), suggesting that NO produced by nNOS suppresses the neurotransmitter release. Recent work in our laboratory suggested that nNOS is responsible for endothelin-B-receptor-mediated suppression of the cholinergic adrenal catecholamine secretion in dogs (10). nNOS is also found in the rat kidney (22) and is considered to participate in the control of renal hemodynamics (11) and renin secretion (18). In the present study, however, 7-NINA failed to affect the RNS-induced responses. Although the concentration range of 7-NINA (10–100 μM) was lower than those of l-NAME and l-NNA (100–200 μM) because of its low solubility, 7-NINA at 100 μM could substantially inhibit nNOS-mediated responses, as demonstrated in previous reports (2,5). Therefore, NO production by nNOS does not seem to participate in the modulation by endogenous NO of the renal sympathetic nervous system.
Because iNOS is not expressed in normal condition, NO production in our experiments can be exclusively related to eNOS activity. Adrenergic stimulation may activate eNOS in vascular endothelial cells, and NO released from the endothelial cells may act on both the pre- and postsynaptic sites of the renal vascular sympathetic nervous system. It is also possible that NO is released from the renal sympathetic nerve endings, but there has been no information on the distribution of eNOS in sympathetic nerves of the rat kidney.
The released NO is immediately changed to NO2 and then metabolized to nitrosothiols or other substances, and these metabolites also have biological actions (26). For more appropriate evaluation of the role of NO, therefore, it may be necessary to examine effects of drugs that selectively eliminate NO. In this regard, we examined the effect of an NO scavenger, carboxy-PTIO, that eliminates NO by oxidizing NO to NO2 without affecting NOS (1). Carboxy-PTIO, like the nonselective NOS inhibitors, suppressed the RNS-induced NE efflux response. This result confirms the facilitatory role of NO in renal neurotransmitter release.
On the other hand, carboxy-PTIO only slightly enhanced the RNS-induced perfusion pressure response. This might be due to the reduced neural NE release, because carboxy-PTIO apparently enhanced the exogenous NE-induced perfusion pressure response. However, l-NAME andl-NNA suppressed the RNS-induced NE efflux response almost to the same extent as did carboxy-PTIO, and the enhancement by carboxy-PTIO of the exogenous NE-induced perfusion pressure response was smaller than that observed with l-NNA. It was reported that an inhibitory effect of carboxy-PTIO on NO-dependent relaxation was smaller than the effect of l-NNA in the isolated rabbit urethra (27). Carboxy-PTIO does not interrupt the metabolic pathway after NO2, whereas NOS inhibitors reduce NO and its metabolites. It is therefore possible that NO metabolites, rather than NO itself, predominantly play an inhibitory role in the adrenergically induced vasoconstriction in the rat kidney.
We also examined whether the pre- and postsynaptic modulation by NO could be related to activation of guanylate cyclase by applying a selective guanylate cyclase inhibitor, NS-2028 (14), and a membrane-permeable cGMP analog, 8-BrcGMP. We had observed that NS-2028 at 1 μM abolished suppression by an NO donor (sodium nitroprusside) of the RNS-induced perfusion pressure response; 2-Hz RNS-induced changes in perfusion pressure (n = 3) were 11 ± 3 mmHg (control), 3 ± 2 mmHg (sodium nitroprusside 30 nM), and 11 ± 3 mmHg (sodium nitroprusside 30 nM and NS-2028 1 μM). 8-BrcGMP at 30 μM was reported to reverse NE-induced afferent arteriolar vasoconstriction in the perfused hydronephrotic rat kidney (9). In the present study, NS-2028 (0.1 and 1 μM), like l-NAME and l-NNA, enhanced the RNS-induced perfusion pressure response, and 8-BrcGMP (10 and 100 μM) suppressed the RNS-induced perfusion pressure response. These results indicate participation of the guanylate cyclase-cGMP pathway in the neural control of renal vascular resistance. However, neither NS-2028 nor 8-BrcGMP, even at their highest concentrations, affected the RNS-induced NE efflux response. Changes in cGMP level, therefore, may not modulate the sympathetic neurotransmitter release in the rat kidney. The failure of the guanylate cyclase inhibitor to mimic the presynaptic effects of the NOS inhibitors or NO scavenger suggests that mechanisms other than the guanylate cyclase-cGMP pathway mediate the facilitation by NO of renal neurotransmitter release.
Our present study demonstrates that NO plays a facilitatory role in the neurotransmitter release in the rat renal sympathetic nervous system, the mechanisms of which are independent of the guanylate cyclase-cGMP pathway. It is well known that NO can open Ca2+-activated K+ channels without activating guanylate cyclase in vascular smooth muscle cells (3). However, the opening of Ca2+-activated K+ channels at the presynaptic site would reduce the NE release by interfering with membrane depolarization. Other cGMP-independent actions of NO have been reported. NO elevates cellular cAMP level by directly activating adenylate cyclase in cardiac myocytes (23), and NO causes gonadotropin secretion from pituitary cells through extracellular Ca2+-dependent mechanisms (15). Because elevation of cAMP level and influx of extracellular Ca2+can be causal factors for facilitation of sympathetic neurotransmitter release, postulation of these mechanisms may explain the results obtained from the present study. However, it is unknown whether these mechanisms exist in the presynaptic site of the sympathetic nervous system of the rat kidney. In addition, the findings of our experiments do not exclude the possibility that the NO- and cGMP-related drugs might affect synthesis, reuptake, or renal clearance of NE. It should also be noted that NO reacts with superoxide to form peroxynitrite. NOS inhibitors or carboxy-PTIO could prevent production of peroxynitrite, which has been suggested to stimulate neurotransmitter release in the central nervous system through cGMP-independent mechanisms (13). Further elucidation will be required to clarify the mechanism underlying the presynaptic modulation by NO.
Address for reprint requests and other correspondence: H. Hisa, Laboratory of Pharmacology, Graduate School of Pharmaceutical Sciences, Tohoku Univ., Aobayama, Sendai 980-8578, Japan (E-mail:).
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.
First published January 17, 2002;10.1152/ajpregu.00697.2001
- Copyright © 2002 the American Physiological Society