Renal sympathetic baroreflex during normoxia and during hypoxia in conscious and in anesthetized rabbits

Sympathetic baroreceptor regulation during hypoxic hypotension in humans: new insights

BACKGROUND

Baroreceptor activation by a continuous infusion of phenylephrine selectively abolishes the muscle sympathetic nerve activity (MSNA) response to hypoxia in humans. Baroreceptor deactivation enhances the MSNA rise during hypoxia in animals. Whether this is true in humans is unknown and was tested in the present study.

METHODS

We assessed MSNA responses elicited by isocapnic hypoxia (10% O2 in N2) during baroreflex loading and unloading with phenylephrine and nitroprusside, respectively, in 19 healthy volunteers. The study was randomized and placebo-controlled.

RESULTS

Phenylephrine and nitroprusside increased and decreased, respectively, blood pressure during normoxia and hypoxia, whereas the reverse occurred for heart rate and MSNA (all P < 0.001 vs. placebo). As compared with normoxia, cardiac barosensitivity decreased during the infusion of placebo and nitroprusside in the presence of hypoxia, as well as sympathetic barosensitivity during the infusion of nitroprusside (all P < 0.05). Three patients even disclosed a reduction in arterial pressure, which became apparent at the third minute of hypoxia and worsened steadily thereafter (SBP: 91 ± 7 mmHg; DBP 47 ± 9 mmHg), in spite of a gradual rise in heart rate of 20 ± 4 bpm. Changes in baroreceptor loading conditions did not affect ventilation during normoxia and hypoxia.

CONCLUSION

Cardiac and sympathetic baroreceptor sensitivity decrease during baroreceptor unloading in the presence of peripheral chemoreceptor activation. Normal humans have limited reflex capabilities to sustain simultaneous reductions in oxygen and pressure, and may experience hemodynamic instability episodes in such condition.

Characteristics of renal sympathetic nerve single units in rabbits with angiotensin-induced hypertension

We examined the effect of chronic angiotensin (Ang II)-induced hypertension on activity of postganglionic renal sympathetic units to determine whether altered whole renal nerve activity is due to recruitment or changes in firing frequency. Rabbits were treated with a low (20 ng kg(-1) min(-1), 8 weeks) or high dose (50 ng kg(-1) min(-1), 4 weeks) of Ang II before the experiment under chloralose-urethane anaesthesia. Spontaneously active units were detected from multiunit recordings using an algorithm that separated units by action potential shape using templates that matched spikes within a prescribed standard deviation. Multiunit sympathetic nerve activity was 40% higher in rabbits treated with low-dose Ang II than in sham (P = 0.012) but not different in high-dose Ang II. Resting firing frequency was similar in sham rabbits (1.00 ± 0.09 spikes s(-1), n = 144) and in those treated with high-dose Ang II (1.10 ± 0.08 spikes s(-1), n = 112) but was lower with low-dose Ang II (0.65 ± 0.08 spikes s(-1), n = 149, P < 0.05). Unit firing rhythmicity was linked to the cardiac cycle and was similar in sham and low-dose Ang II groups but 29-32% lower in rabbits treated with high-dose Ang II (P < 0.001). Cardiac linkage followed a similar pattern during hypoxia. All units showed baroreceptor dependency. Baroreflex gain and range were reduced and curves shifted to the right in Ang II groups. Firing frequency during hypoxia increased by +39% in low-dose Ang II and +82% in shams, but the greatest increase was in the high-dose Ang II group (+103%, P(dose) = 0.001). Responses to hypercapnia were similar in all groups. Increases in sympathetic outflow in hypertension caused by low-dose chronic Ang II administration are due to recruitment of neurons, but high-dose Ang II increases firing frequency in response to chemoreceptor stimuli independently of the arterial baroreceptors.

Differential regulation of sympathetic burst frequency and amplitude following acute hypoxia in humans

Current evidence suggests that the persistent sympathetic nerve activity (SNA), commonly observed after exposure to hypoxia (HX), is mediated by chemoreceptor sensitization and or baroreflex resetting. Evidence in humans and animals suggests that these reflexes may independently regulate the frequency (gating) and amplitude (neuronal recruitment) of SNA bursts. In humans ( n = 7), we examined the regulation of SNA following acute isocapnic HX (5 min; end-tidal Po2 = 45 Torr) and euoxic hypercapnia (HC; 5 min; end-tidal Pco2 = +10 from baseline). HX increased SNA burst frequency (21 ± 7 to 28 ± 8 bursts/min, P < 0.05) and amplitude (99 ± 10 to 125 ± 19 au, P < 0.05) as did HC (14 ± 6 to 22 ± 10 bursts/min, P < 0.05 and 100 ± 12 to 133 ± 29 au, P < 0.05, respectively). Burst frequency (26 ± 7 bursts/min, P < 0.05), but not amplitude (97 ± 12 au), remained elevated 10 min post-HX. The change in burst amplitude (but not frequency) was significantly related to the measured change in ventilation ( r2 = 0.527, P < 0.001). Both frequency and amplitude decreased during recovery following HC. These data indicate the differential regulation of pattern and magnitude of sympathetic outflow in humans with sympathetic persistence following HX being specific to burst frequency and not amplitude.

Long-range correlation of renal sympathetic nerve activity in both conscious and anesthetized rats

In this study we employed both detrended fluctuation analysis (DFA) and multiscale entropy (MSE) measurements to compare the long-range temporal correlation (LRTC) of multifibre renal sympathetic nerve activity (RSNA) between conscious and anesthetized Wistar rats. It was found that both methods showed the obvious LRTC properties in conscious state. Moreover, the scaling exponent of the RSNA in conscious rats was significantly higher than that in anesthetized rats. The results of MSE analysis showed that the entropy values, derived from the conscious group, increased on small time scales and then stabilized to a relatively constant value whereas the entropy measure, derived from anesthetized animals, almost monotonically decreased. This suggests that the fractal properties of underlying dynamics of the system have been reduced by anesthesia. The results demonstrate that apparently random fluctuations in multifibre RSNA are dictated by a complex deterministic process that imparts "long-term" memory to the dynamic system. However, this memory is significantly weakened by anesthesia.

Evidence of differential control of renal and lumbar sympathetic nerve activity in conscious rabbits

We have explored the possibility that renal sympathetic nerve activity (RSNA) and vasomotor sympathetic nerve activity are differentially regulated. We measured sympathetic nerve activity (SNA) to the kidney and the hind limb vasculature in seven conscious rabbits 6-8 days after the implantation of recording electrodes. Acute infusion of N(G)-nitro-L-arginine methyl ester (L-NAME) (6 mg.kg(-1).min(-1) for 5 min) led to an increase in blood pressure (from 66 +/- 1 to 82 +/- 3 mmHg) and a decrease in heart rate (from 214 +/- 15 to 160 +/- 13 bpm). L-NAME administration caused a significantly greater decrease in RSNA than lumbar sympathetic nerve activity (LSNA) (to 68 +/- 14% vs. 84 +/- 4% of control values, respectively). Volume expansion (1.5 ml.kg(-1).min(-1)) resulted in a significant decrease in RSNA to 66 +/- 7% of control levels but no change in LSNA (127 +/- 20%). There was no difference in the gain of the baroreflex curves between the LSNA and RSNA [maximum gain of -7.6 +/- 0.4 normalized units (nu)/mmHg for LSNA vs. -7.9 +/- 0.75 nu/mmHg for RSNA]. A hypoxic stimulus (10% O2 and 3% CO2) led to identical increases in both RSNA and LSNA (195 +/- 40% and 158 +/- 21% of control values, respectively). Our results indicate tailored differential control of RSNA and LSNA in response to acute stimuli.

Glutamate receptors in RVLM modulate sympathetic baroreflex in conscious rabbits

In this study, we examined the effect of excitatory amino acid (EAA) receptor blockade in the rostral ventrolateral medulla (RVLM) on the renal sympathetic baroreflex in conscious rabbits. Rabbits were implanted with guide cannulas for bilateral microinjections into the RVLM (+2 to +3 mm from the obex, n = 8) or into the intermediate ventrolateral medulla (IVLM; 0 to +1 mm from the obex, n = 5) and with an electrode for measuring renal sympathetic nerve activity (RSNA). After 7 days of recovery, microinjection of the EAA receptor antagonist kynurenate (10 nmol) into the RVLM did not affect resting RSNA or arterial pressure. Kynurenate decreased the gain of the RSNA baroreflex by 53% but did not change the reflex range. By contrast, injection of kynurenate into the IVLM increased resting arterial pressure and RSNA by 27 mmHg and 88%, respectively, but did not alter the RSNA baroreflex gain or range. Pentobarbital sodium anesthesia attenuated the gain and range of the RSNA baroreflex by 78 and 40%, respectively. Under these conditions, microinjection of kynurenate into the RVLM did not cause any further change in the gain of this reflex. These results suggest that endogenous EAA neurotransmitters in the RVLM are important in modulating the sympathetic baroreflex in conscious rabbits. Anesthesia can mask the functional significance of EAAs in the RVLM in modulating the baroreflexes, which may explain why previous studies in anesthetized animals found no effect of blocking EAA receptors in the RVLM on sympathetic baroreflexes.

Central angiotensin and baroreceptor control of circulation

Angiotensin (Ang) receptors are located in many important central nuclei involved in the regulation of the cardiovascular system. While most interest has focused on forebrain circumventricular actions, areas of the brainstem such as the nucleus of the solitary tract and the ventrolateral medulla contain high concentrations of AT1 receptors. The present review encompasses the physiological role of Ang II in the hindbrain, particularly in relation to its influence on baroreflex control mechanisms. In rabbits there are sympatho-excitatory AT1 receptors in the rostral ventrolateral medulla (RVLM), accessible to Ang II from the cerebrospinal fluid. Activation of these receptors acutely increases renal sympathetic nerve activity (RSNA) and RSNA baroreflex responses. However, blockade of endogenous Ang receptors in the brainstem also shows sympathoexcitation, suggesting there is greater endogenous activity of a sympathoinhibitory Ang II action. Microinjections of angiotensin antagonists into the RVLM showed relatively little tonic activity of endogenous Ang II influencing sympathetic activity in conscious rabbits. However, Ang II receptors in the RVLM mediate sympathetic responses to airjet stress in conscious rabbits. Similarly with respect to heart rate baroreflexes, there appears to be little tonic effect of angiotensin in the brainstem in normal conscious animals. Chronic infusion of Ang II for two weeks into the fourth ventricle of conscious rabbits inhibits the cardiac baroreflex while infusion of losartan increases the gain of the reflex. These actions suggest that Ang II in the brainstem modulates sympathetic responses depending on specific afferent and synaptic inputs in both the short term but importantly also in the long term, thus forming an important mechanism for increasing the range of adaptive response patterns.

Influence of pontine A5 region on renal sympathetic nerve activity in conscious rabbits

The effects of inhibiting the neural activity in the pontine A5 region on renal sympathetic responses to baroreflex and/or chemoreflex activation were examined in conscious rabbits. Eight rabbits were chronically instrumented with guide cannulas for bilateral microinjections into the A5 area and an electrode for measuring renal sympathetic nerve activity (RSNA). Baroreflex curves were obtained under conditions of normoxia and hypoxia (10% O(2) + 3% CO(2)) after injections into the A5 region of the GABA receptor agonist muscimol or vehicle solution. Under normoxia, injections of muscimol did not affect resting RSNA or blood pressure but increased the range of the RSNA baroreflex by 24 and 33% at doses of 175 or 875 pmol, respectively, without affecting the reflex gain. Hypoxia alone increased resting RSNA by 63%, as well as the range and gain of the RSNA baroreflex by 53 and 89%, respectively, without affecting blood pressure. However, under hypoxia, muscimol increased resting RSNA by 37 and 47% but decreased the gain of the RSNA baroreflex by 19 and 34% at doses of 175 or 875 pmol, respectively, without affecting the reflex range. The effects of muscimol on RSNA were mediated via changes in the amplitude of the sympathetic bursts, whereas burst frequency remained unaffected. These data suggest that the A5 region has a little tonic influence on RSNA in conscious rabbits but serves to limit the renal sympathetic responses to baroreceptor unloading or chemoreceptor stimulation. The different changes in the baroreflex range and gain evoked by muscimol under normoxia and hypoxia indicate that the A5 modulatory action may depend on the activity of the afferent inputs to this region.

Hypoxia inhibits myogenic reactivity of renal afferent arterioles by activating ATP-sensitive K+ channels

Recent findings implicate K+ channels as important modulators of myogenic tone and possible mediators of the vasodilatory effects of hypoxia. In the present report, we examined the effects of hypoxia on myogenic vasoconstriction of renal afferent arterioles. Using the in vitro perfused hydronephrotic rat kidney model, we observed precisely graded decreases in arteriolar diameter when renal perfusion pressure was increased. Normal myogenic reactivity was observed over PO2 levels of 150 to 80 mm Hg. Reducing PO2 to 60, 40, and 30 mm Hg resulted in a significant progressive inhibition of myogenic reactivity. At approximately 20 mm Hg, myogenic vasoconstriction was essentially abolished, whereas the vasoconstriction induced by 30 mmol/L KCl was unaffected. The addition of 1.0 mumol/L glibenclamide completely restored myogenic vasoconstriction during hypoxia. In contrast, 1.0 mmol/L tetraethylammonium did not alter the effects of hypoxia. To investigate the relation between hypoxia-induced vasodilation and smooth muscle oxidative phosphorylation, we monitored changes in arteriolar levels of reduced NADH during exposure to hypoxia. Arterioles preconstricted by elevated pressure were optically isolated for simultaneous monitoring of vessel diameter and NADH fluorescence (360-nm excitation, 450-nm emission). Reducing perfusate PO2 from 150 to 20 mm Hg resulted in progressive loss of myogenic tone with no change in arteriolar NADH. These findings indicate that lowering PO2 within a physiological range attenuates myogenic reactivity of the renal afferent arteriole by causing the activation of ATP-sensitive K+ channels.(ABSTRACT TRUNCATED AT 250 WORDS)

Interactions between the circulatory effects of central hypovolaemia and arterial hypoxia in conscious rabbits

1. Eight conscious rabbits were repeatedly subjected to progressive reduction in central blood volume by gradually inflating a thoracic inferior vena caval-cuff so cardiac index (CI) fell at a constant 8.5% of baseline/min. 2. Caval-cuff inflations were performed after 10 min exposure to 100, 21, 12-14 and 8-10% O2, with and without the addition of 3-4% CO2, in randomized order. 3. The haemodynamic response to progressive reduction in central blood volume was biphasic. In Phase I, systemic vascular conductance index (SVCI) fell linearly, supporting mean arterial pressure (MAP). When CI had fallen to a critical level, Phase II occurred in which SVCI rose abruptly, MAP plummeted and respiratory drive progressively increased. 4. During Phase I, there were independent linear relationships between PaCO2 (but not PaO2) and the rates at which SVCI and MAP changed during the progressive fall of CI. The higher the level of PaCO2, the greater was the rate of fall of SVCI and the less the rate of fall of MAP. 5. There was an inverted U-shaped effect of PaO2 on the level of CI at which Phase II occurred: (a) during hyperoxia (100% O2), Phase II occurred later than during normoxia (21% O2); and (b) across the normoxic and hypoxic gas mixtures (21-8% O2, with and without added CO2), there was an independent linear relationship between PaO2 (but not PaCO2 or PaO2 x PaCO2) and the level of CI at which Phase II occurred. That is, the lower the level of PaO2, the later was the onset of Phase II. This interaction is best explained by an increased level of central sympathetic vasoconstrictor drive during hypoxia.

Effect of pentobarbital anesthesia on renal sympathetic nerve activity in the rabbit

The effects of pentobarbital (PB) anesthesia on arterial pressure, heart rate, and renal nerve activity (RNA) were studied in chronically instrumented intact rabbits and rabbits with sinoaortic baroreceptor denervation (SAD). In intact rabbits, PB caused an initial decrease in mean arterial pressure from 82 +/- 2 to 42 +/- 5 mm Hg, which then returned to the control level within 5 min. RNA increased by 144 +/- 19% in response to hypotension induced by PB then also returned to the control level within 5 min. Heart rate increased by 56 +/- 10 from 244 +/- 11 beats/min and remained elevated for 60 min. Sensitivities of baroreflex control of heart rate and RNA were markedly impaired by PB anesthesia. In SAD rabbits, PB caused hypotension (-46 +/- 5 mm Hg), which lasted for 15 min. RNA decreased initially by 80 +/- 5%, then gradually returned toward the control level. However, at 45 min after injection of PB, RNA was still depressed significantly. Tachycardia induced by PB was abolished by SAD, i.e. heart rate actually decreased by 53 +/- 10 beats/min. These results suggest that in intact rabbits steady-state effects of PB were: no change in mean arterial pressure or RNA, and tachycardia, all due to baroreflex compensation. However, in the absence of baroreflex compensation, PB elicited hypotension, bradycardia, and a decrease in RNA.

Reflex responses to baroreceptor, chemoreceptor and nociceptor inputs in single renal sympathetic neurones in the rabbit and the effects of anaesthesia on them

Reflex responses of renal postganglionic neurones to stimulation of arterial baroreceptors, arterial and central chemoreceptors and cutaneous nociceptors, and the rhythmicity of their resting activity were studied in paralyzed, artificially ventilated rabbits, anaesthetized with either alfathesin or chloralose-urethane. A 'vasoconstrictor' response pattern was seen in all units. Perivascular balloon-induced falls in blood pressure increased firing while pressure rises silenced 90% of units and reduced firing in the rest. Resting activity was linked to pressure changes within the cardiac cycle and to the artificial respiratory cycle. The largest excitation occurred during hypoxia and injections of CO2 saturated solutions into the carotid artery while hypercapnia and stimulation of cutaneous nociceptors only slightly increased firing. Parameters characterizing rhythmicities and reflex responses were unimodally distributed with no apparent subgrouping of units on quantitative grounds. Unit response patterns were similar to those recorded in the whole renal nerve. With one exception, no silent units were found which responded to the afferent inputs studied. Nor was there a small-spike fibre group which was excited by angiotensin. However, reflex responses were significantly influenced by the anaesthetic regime selected for use. Under alfathesin, baroreceptor and chemoreceptor reflexes were double those found with chloralose-urethane. Under chloralose-urethane, hypoxia increased both rhythmicities, while under alfathesin, cardiac rhythmicity was decreased and respiratory rhythmicity was variably affected. We concluded that renal sympathetic neurones are a functionally uniform population which behave like vasoconstrictors.

The renal sympathetic baroreflex in the rabbit. Arterial and cardiac baroreceptor influences, resetting, and effect of anesthesia

Curves relating renal sympathetic nerve activity and mean arterial pressure were derived in conscious rabbits during ramp changes in mean arterial pressure, elicited by perivascular balloon inflation. The renal sympathetic nerve activity-mean arterial pressure relationship consisted of a high-gain sigmoidal region about resting, where renal sympathetic nerve activity rose or fell in response to moderate falls and rises of mean arterial pressure. With larger pressure rises, renal sympathetic nerve activity first fell to a lower plateau and then reversed at even higher mean arterial pressure. When mean arterial pressure was lowered below resting, renal sympathetic nerve activity rose to an upper plateau and then reversed abruptly toward resting at low mean arterial pressure. Both arterial and cardiac baroreceptors exerted substantial inhibitory influences on renal sympathetic nerve activity at all pressure levels. These effects appeared additive over the central high gain region of the curve, but beyond this region there were non-additive interactions. The latter were affected considerably by alfathesin anesthesia. In other experiments, we studied the effects of sustained alterations in resting mean arterial pressure induced by infusing nitroprusside and phenylephrine, which produced rapid resetting of the renal baroreflex. The latter could be accounted for, in part, by resetting of the threshold of the arterial baroreceptors and in part by contributions from other afferents, probably the cardiac receptors. During resetting associated with nitroprusside-induced falls in resting blood pressure, high-gain reflex adjustments in renal sympathetic nerve activity to moderate changes in mean arterial pressure were preserved, but during resetting associated with phenylephrine-induced rises in resting mean mean arterial pressure, the resting renal sympathetic nerve activity lay on the lower curve plateau, resulting in reduction in the apparent gain of the reflex renal sympathetic nerve activity response to moderate changes in mean arterial pressure.

The aortic nerve-sympathetic reflex in the rat

The effects of stimulation of aortic nerve A- and C-fibers on the renal and cardiac sympathetic nerve activities in anesthetized and immobilized Sprague-Dawley rats were investigated. A separate aortic nerve was found in 46 rats (90%) out of 51. Activation of A- and C-fiber groups, alone or in combination, resulted in an inhibition of renal and cardiac nerve activities. However, an excitatory component preceding the inhibitory component, representing the reflex response to stimulation of non-barosensory afferent fibers contained in the carotid sinus or aortic nerve, was never observed. This result provides electrophysiological evidence supporting the view that the rat's aortic nerve does not contain a significant amount of functionally active non-barosensory afferents. As with the aortic nerve reflex in the rabbit and cat, the sympatho-inhibitory action of C-fibers was more powerful and longer-lasting than that of A-fibers. Furthermore, the C-fiber reflex was elicited at stimulus frequencies as low as 2 Hz. No significant difference was found between the reflex response of cardiac and renal nerves. On the other hand, stimulation of the superior laryngeal nerve, which constitutes an important pathway carrying arterial baroreceptor fibers, caused a reflex sympathetic response typically consisting of excitatory and inhibitory components. Thus, the rat's aortic nerve provides a useful experimental means to activate selectively central neural structures associated with barosensory afferents and to elicit the reflex response homologous to that in the arterial baroreceptor reflex in rabbits and cats.