INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

SYMPATHOMODULATORY ACTIONS of nitric oxide (NO) have been repeatedly demonstrated in several species (10, 11, 15, 27, 36). However, the possible regulatory functions of these effects are much less clear. Recently, the modulation of excitatory inputs to the rostral ventrolateral medulla (RVLM) has been identified as a principal mechanism of action of NO within the brain stem (37). One excitatory input that may be modulated by NO is provided by peripheral chemoreceptors that are activated by systemic hypoxia (8). However, despite its sympathoexcitatory influences, hypoxia can be associated with pronounced vasodilatation in several vascular beds (23). NO may be critically involved in these specific cardiovascular responses to hypoxia at several levels. Endothelial NO has been identified as an important locally acting mediator of hypoxic vasodilation in several species (3, 4, 22, 34). Furthermore, NO is known to reduce the activity of peripheral chemoreceptors (25, 31, 32). In the present study, we examined possible regulatory effects of endogenous NO within the brain stem that could affect sympathetic activity during systemic hypoxia. Young farm pigs were studied during general anesthesia and underwent cycles of hypoxia and reoxygenation before and after inhibition of NO synthases (NOS) by N^G-nitro-L-arginine (L-NNA) in the VLM. The specificity of the observed effects was tested by the inactive enantiomer D-NNA and their reversibility by administration of an NO donor at the same sites. Furthermore, to discriminate between possible direct effects on sympathetic nerve activity induced by hypoxia and/or NO in these areas and effects mediated by chemoreceptor inputs, additional experiments were performed following complete deafferentiation by cutting carotid sinus and vagal nerves bilaterally.

	METHODS

Top Abstract Introduction Methods Results Discussion References

General procedures. Young farm pigs (n = 21, 16-20 kg body wt) were sedated by ketamine (10 mg/kg im) and anesthetized by pentobarbital sodium (12-15 mg/kg iv). Anesthesia was maintained after intubation with isoflurane (1.0-1.5%) in the inspired air consisting of 70% N₂ and 30% O₂. For infusion of drugs and for measurement of blood pressure, catheters were placed into the right brachial vein and artery, respectively. Antibiotics (100 mg/kg ampicillin) were given to prevent possible influences of infections such as the induction of expression of inducible NOS by bacterial endotoxins. The pigs were paralyzed by 0.2 mg · kg¹ · h¹ pancuronium bromide and artificially ventilated via a tracheal tube. Ventilatory parameters (airway flow and pressure, inspiratory O₂, expiratory CO₂) were continuously monitored (Dräger PM 8050). End-tidal CO₂ was kept at normal levels by adjustment of ventilatory depth and rate. Arterial blood gases were monitored with a blood gas analyzer (AVL 990, AVL List) and maintained in the normal range by administration of sodium bicarbonate solutions or adjustment of ventilation. Rectal temperature was maintained at 38.5°C by a thermostatically controlled infrared lamp. For measurements of cardiac output (CO), central venous pressure (CVP), and pulmonary arterial pressure (PAP), a biluminal (right atrium, pulmonary artery) 5F Swan-Ganz thermodilution catheter (Baxter) was inserted through a jugular vein and advanced through the right ventricle in the pulmonary artery under blood pressure control. In some experiments, CO was measured continuously with a transient time flow probe (ART, Triton) on the pulmonary artery as previously described (37). For recording of renal sympathetic nerve activity (RSNA), the left renal nerve was retroperitoneally exposed, placed on bipolar platinum electrodes, and kept in a mixture of Vaseline and paraffin oil. Neural signals were amplified (×20,000-50,000; Tektronix AM 502) and filtered (2 Hz-3 kHz). RSNA was full-wave rectified and then resistance-capacity (RC) integrated with a time constant between 7 and 10 ms.

Experimental hypoxia. Before experimental hypoxia, ventilation of the pigs was adjusted to normocapnia (arterial partial CO₂ pressure = 39-41 mmHg) and kept slightly hyperoxic (25-28% inspiratory O₂). Systemic hypoxia was initiated by replacing O₂ by N₂ in the inspired air to achieve 10% O₂ in the inspired air for 3-5 min. Hypoxic ventilation lowered arterial PO₂ to levels between 46 and 54 mmHg as detected by blood gas measurements. For reoxygenation, N₂ was withdrawn and O₂ was supplemented to reestablish normal ventilation. Ventilatory depth and rate were not changed during these cycles.

Application of drugs. SNAP was from Alexis Chemicals. All other drugs were from Sigma. The substances were dissolved in distilled water. For preparation of the final concentrations, all drugs were further diluted in Ringer solution shortly before administration.

For application of substances, the dorsal surface of the medulla oblongata was exposed at the region from 2-3 mm caudal to 5-6 mm rostral to the obex extending 4-5 mm laterally on both sides. For administrations to the VLM, a catheter (diameter 0.3 mm) was connected to a pulled glass pipette (tip diameter 0.05-0.1 mm) that was stereotaxically positioned near the midline ~5-6 mm rostral from the obex. The pipette was then advanced through the brain stem to the ventral medullary surface with a micromanipulator. The position of the catheter tip was functionally verified by instantaneous excitatory sympathetic responses to injections (0.3 ml) of glutamate (0.5 mol/l) and by ex vivo examination of the spread of a test injection of 0.5 ml Alcian blue at the end of the experiments. The catheter position and the spread of injected dye as well as a representative glutamate response are shown in Fig. 1. Central NOS inhibition at the VLM was carried out by short-term infusion (within 5 min, 1 ml/min) of L-NNA (0.3 mmol/l). The specificity of L-NNA was verified with the inactive enantiomer of L-NNA, i.e., D-NNA administered similarly at the same concentration. Effects of exogenous NO were tested by short-term infusion of SNAP (0.1 mmol/l) following NOS inhibition.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   A: camera lucida drawing showing injection site on the ventral surface of the medulla. Spread of a test injection of Alcian blue (0.5 ml) is indicated as gray area on the drawing. B: effects of glutamate injection (0.3 ml, 0.5 mol/l) at the same site in this animal on blood pressure (BP), renal sympathetic nerve activity (RSNA), and heart rate (HR). XII, nervus hypoglossus; IX, nervus glossopharyngeus; X, nervus vagus; V, nervus trigeminus; VII, nervus fascialis; VI, nervus abducens; VLM, ventrolateral medulla.

Data analysis. Heart rate (HR) was derived from the blood pressure signal. Total peripheral resistance (TPR) was calculated as [mean arterial pressure (MAP)  CVP]/CO. RSNA was RC-integrated and measured in arbitrary units. For presentation in summarizing figures and for statistical evaluation, RSNA data were normalized to individual control activities at the beginning of the experiments and expressed in percent (control = 100%). Responses to hypoxia were determined by calculating individual differences from control values. Peak responses were routinely used for calculation of the responses to hypoxia. The similarity of measurements of peak values and corresponding values of areas under the response curves had been checked in preliminary experiments. All direct measurements were stored on a linear recorder (Gould) and on DAT-tape for further computing. All data were analyzed by ANOVA (for repeated measurements where appropriate). Comparison of means was carried out with Tukey's Studentized range test. Differences of P < 0.05 were considered to be significant. Values are reported as means ± SE.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Effects of central NO on baseline sympathetic activity and hemodynamic parameters. Baseline RSNA and hemodynamic values and the effects of inhibition of NO-synthesis and exogenous NO in form of S-nitroso-N-acetyl-penicillamine (SNAP) within the VLM on these parameters are listed in Table 1. Effects of L-NNA and SNAP on baseline sympathetic activity were overall small and caused no relevant hemodynamic effects. Nevertheless, NOS inhibition by L-NNA tended to increase baseline RSNA on administration to the RVLM while SNAP significantly reduced baseline RSNA. The inactive analog D-NNA had no measurable effects on baseline measurements.

Modulation of the sympathetic responses to hypoxia by NO. A representative original tracing of the effects of NO on sympathetic and hemodynamic responses to hypoxia is shown in Fig. 2. A summary of correspondent mean data (measured at the end of the hypoxic period) of the responses of renal sympathetic activity (RSNA) and hemodynamics (TPR and MAP) to hypoxia is given in Fig. 3. Inhibition of NO synthesis in the VLM by L-NNA significantly enhanced the sympathoexcitatory response to systemic hypoxia whereas with D-NNA the responses remained almost unchanged from control. Hypoxia also caused moderate increases in HR (Tables 1 and 2). However, the modulation of these responses by the pharmacological treatments was not statistically significant. The effects of the drugs applied on responses to hypoxia in the pulmonary vascular bed are shown in Fig. 4. Hypoxic vasoconstriction, indicated by pronounced increases in PAP, was not affected by the experimentally induced changes in sympathetic activity in the present experiments.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Modulation of sympathetic responses to hypoxia by nitric oxide (NO). Representative tracings showing effects of inhibition of NO synthesis and exogenous NO in the VLM on BP, RSNA, mean pulmonary arterial pressure (PAP), cardiac output (CO), and HR during cycles of hypoxia and reoxygenation in a pig. Control (Ringer) solution (A), NG-nitro-L-arginine (L-NNA, 0.3 mmol/l; B), and S-nitroso-N-acetyl-penicillamine (SNAP, 0.1 mmol/l; C) were administered to the ventral surface of the medulla as shown in Fig. 1.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Graph summarizing effects of inhibition of NO synthesis and exogenous NO in the VLM on sympathetic responses to hypoxia and the correspondent changes in mean arterial blood pressure (MAP) and total peripheral resistance (TPR) in pigs. Control (Ringer) solution (n = 14), NG-nitro-D-arginine (D-NNA; 0,3 mmol/l, n = 4), L-NNA (0.3 mmol/l, n = 10), and SNAP (0.1 mmol/l, n = 10) were administered to the ventral surface of the medulla. Measurements were made at the end of the hypoxic period. * P < 0.05, ** P < 0.01 vs. control. # P < 0.05 vs. L-NNA.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Central NO has no effect on hypoxic vasoconstriction in the lungs. Graph summarizes effect of hypoxia on mean pulmonary arterial pressure (mPAP) before and after D-NNA, L-NNA, or SNAP to the VLM. Drugs and concentrations were as described in Fig. 3; n = 10, D-NNA: n = 4.

Role of carotid sinus and aortic chemoreceptors and baroreceptors. The dependency on intact chemoreceptor afferents of the observed responses to hypoxia was demonstrated in seven pigs that were completely deafferentiated by cutting of carotid sinus and vagal nerves bilaterally. After denervation, these pigs had almost normal baseline RSNA (107 ± 8%) and hemodynamics (CO: 3.25 ± 0.33 l/min; MAP: 93.5 ± 5.3 mmHg; PAP: 19.5 ± 1.2 mmHg; HR 134 ± 3 beats/min; for control values see Table 1). A representative original tracing of the responses to hypoxia in one of the denervated animals is shown in Fig. 5. Hypoxia caused no significant changes in sympathetic activity in these animals anymore, and the corresponding hemodynamic responses markedly differed from those in intact control animals. A summary of these data and a statistical comparison with intact animals are given in Table 2. Neither NOS inhibition nor NO donor administration in the VLM had any significant effects on RSNA or on hemodynamic responses to hypoxia in animals with disrupted chemoreceptor and baroreceptor inputs.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Effect of deafferentiation of carotid sinus and aortic chemoreceptors on sympathetic and hemodynamic responses to hypoxia. Representative tracings showing course of BP, RSNA, mean pulmonary arterial pressure (PAP), CO, and HR during a cycle of hypoxia and reoxygenation in a denervated pig (carotid sinus and vagal nerves cut bilaterally). aU, arbitrary units.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The results of this study reveal a new regulatory function of NO. Activation of chemoreceptor afferents may induce NO release in brain stem regions that regulate sympathetic functions, and the RVLM is probably an important site of its action. Thereby NO may attenuate excess activation of presympathetic vasomotor neurons during hypoxia and subsequently cause sympathoinhibition during reoxygenation. These centrally mediated effects of hypoxia could be part of a separate regulation of organ blood flow during hypoxia/reoxygenation that is independent of the local metabolic state.

NOS are detectable within the brain stem of all species so far studied (38). Therefore, although there may be species differences, modulation of the responses to hypoxia by NO could be a general principle. The present experiments were carried out with mechanical ventilation under normocapnic conditions to keep sympathomodulatory influences of respiratory neurons (16), pulmonary stretch (5, 26), and central chemosensitivity (28) constant. Nevertheless, the sympathoexcitatory input to the RVLM provided by these factors may be similarly influenced by NO in spontaneously breathing animals. Likewise, probably due to the very low vagal activity subsequent to the long-lasting anesthesia (>8 h), we could not observe any bradycardic effects of hypoxia. However, the sympathetic responses to hypoxia observed during control conditions in the present study were qualitatively similar to those previously reported from neonatal pigs (29) and from other species such as dogs (12), cats (16), and rabbits (13).

An important question is to what extent the observed effects may be influenced by actions of NO that are independent of the effects on afferent chemoreceptor inputs. One possible factor could be the sympathoinhibitory input provided by baroreceptors. Baroreceptor unloading could contribute to some extent to sympathoexcitation during hypoxic hypotension and rapid blood pressure increases during reoxygenation may be accompanied by baroreflex-mediated sympathoinhibition as well. However, in a number of species including pigs it has been shown that baroreflex functions may not be significantly altered by NO (14, 21, 35-37). Therefore, the modulation by NO of the responses to hypoxia may be primarily mediated by interactions with the chemoreceptor input (8). A cellular basis of these effects could be an active release of NO on chemoreceptor stimulation. Effects of chemoreceptor afferents are probably mediated by N-methyl-D-aspartate (NMDA) receptors in the RVLM (1, 7, 18), which may cause NOS activation by increasing intracellular calcium as shown in studies on other brain regions (2, 17). Recent studies on rats (19) suggest that aspartate rather than glutamate may be the endogenous agonist at these NMDA receptors. The amount of NO released may be determined by the chemoreceptor input depending on the intensity of the hypoxic stimulus. The therefore probably pulsatile and spatially restricted pattern of NO release, however, can be at most partially mimicked by exogenous administration of NO donors to the ventral surface of the medulla. This may explain why after NOS inhibition, control responses could not be completely restored by SNAP in our experiments. Another general methodological problem is the specificity of NOS-inhibiting drugs in vivo. It is likely that the effects of L-NNA observed in this study were NOS specific. First, the inactive enantiomer D-NNA was without effects in our present experiments, and second, in a previous study on pigs using identical methods (37) we have shown that the effects of NOS inhibition induced by L-NNA can be similarly produced by selective antagonists of the neuronal isoform of NOS. Thus the now observed regulation of sympathetic responses to hypoxia by NO may represent a physiological example of endogenously activated neuronal NO release causing sympathoinhibition by attenuation of excitatory amino acid effects in RVLM neurons.

The results of this study were apparently not significantly influenced by direct actions of NO on baseline sympathetic activity (Table 1). This seems to be in contrast to a previous study on cats where direct injections of L-NNA into the RVLM significantly increased sympathetic tone (36). However, in the present experiments, all drugs were administered to the ventral surface of the VLM. This implies that nearby structures including the caudal VLM (CVLM), which provides an important inhibitory input to the RVLM (6, 7, 30), were also affected by the treatments. Attenuation of glutamate effects by NO in the CVLM as well could therefore counteract its effects on baseline activity within the RVLM. Recent studies in conscious rabbits (21) and men (9) that suggest that systemic NOS inhibition may not significantly alter baseline sympathetic tone support this hypothesis. However, during hypoxia such effects may be of minor importance, since the CVLM is probably no relay in the transmission of the sympathetic chemoreflex within the brain stem (8).

The quantitatively different effects of NO on RSNA and HR during hypoxia/reoxygenation could indicate that NO differentially regulates sympathetic outflow to different organs despite its relatively high diffusibility in neuronal structures (20, 33). However, sympathetic outflow to different organs is rather heterogeneous during hypoxia under normal conditions as well (13, 16). Therefore, the differences of the effects of NO could be secondary to the differential integration of the chemoreceptor input in the medulla. These aspects have to be addressed in further studies.

Perspectives