INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

NUMEROUS STUDIES INDICATE that electrical stimulation of cutaneous and/or muscular afferents produces marked changes in arterial blood pressure, heart rate (HR), and regional blood flow in anesthetized animals. Distinct cardiovascular patterns seem to be associated with specific groups of afferents. Electrical or physiological stimulation of group III and IV cutaneous or muscular afferent fibers produces a pattern of cardiovascular adjustments characterized by hypertension, tachycardia, and increases in hindlimb blood flow (HBF) and vascular conductance (HVC; see Refs. 5, 7, 13, 20). It is conceivable that these adjustments represent an equivalent of the cardiovascular responses elicited by stimulation of hypothalamic sites involved in defense reactions (1-2, 24, 25). Results obtained from studies utilizing transection of the neuroaxis demonstrated that cardiovascular responses produced by stimulation of peripheral nerves depend on neurons located in the lower brain stem. Transection of the brain stem immediately caudal to the inferior colliculus has no effect on these adjustments, although they are blocked after spinal cord transection at cervical levels (5, 7, 13, 20).

The rostral ventrolateral medulla (RVL) contains the bulbospinal neurons responsible for the maintenance of the tonic excitation of sympathetic preganglionic neurons of the intermediolateral nucleus of the spinal cord (4, 15, 17, 16). The RVL has also been implicated as the area within the central nervous system responsible for the pressor response elicited by stimulation of somatic and visceral afferents. Chemical or electrolytic lesions of the RVL block the pressor response elicited by sciatic nerve stimulation (SNS; see Ref. 20). Blockade of RVL glutamatergic receptors abolished the pressor response to stimulation of sciatic nerve or high-threshold vagal afferents (8, 21) and the pressor response elicited by muscle contraction (3). Electrophysiological techniques demonstrated that RVL sympathoexcitatory neurons are excited by sciatic afferent stimulation (13).

Although previous studies have investigated the pressor response, whether stimulation of high-threshold afferents induces changes in blood flow remains largely unknown. In the present study, we sought to determine 1) the HBF responses to electrical stimulation of sciatic afferents; 2) the role of RVL in these responses; and 3) the possible neurotransmitters involved.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

General procedures. Studies were conducted on male Wistar rats weighing 250-350 g. Anesthesia was induced with halothane, and catheters were placed in the right carotid artery and jugular vein for measurement of arterial blood pressure and HR and administration of drugs, respectively. Once the venous catheter was in place, the animals were anesthetized with urethane (1.4 g/kg iv). The arterial catheter was connected to a Statham P23 Db strain gauge transducer to record arterial pressure. The trachea was cannulated, and the rats were mounted prone in a stereotaxic apparatus (David Kopf Instruments) with the bite bar set 11 mm below the interaural line. The rats were paralyzed with d-tubocurarine (0.5 mg/kg iv) and were artificially ventilated (7025 Rodent Ventilator-UGO BASILE) with room air. Arterial PCO₂ was maintained at 30-35 mmHg by controlling the rate of respiration. Body temperature was maintained at 37°C with a thermostatically controlled heating table. The adequacy of anesthesia without paralysis was verified by the absence of a withdrawal response to nociceptive stimulation of a hindpaw and by stable arterial pressure and HR during paralysis. A miniature Doppler flow probe was placed on the left iliac artery for measurement of the HBF and the determination of vascular conductance. HR was recorded with a SensorMedics 9857 B cardiotachometer triggered by the arterial pulse wave. Pulsatile (PAP) and mean (MAP) arterial blood pressure, HR, and HBF were recorded continuously on a six-channel Beckman polygraph.

Nerve stimulation. The left sciatic nerve was dissected in the lateral aspect of the leg. The nerve was severed distally, and the central end was placed on platinum hook electrodes for stimulation. The nerve was covered with warm mineral oil and electrically stimulated for 10 s with cathodal square wave pulses (800-1,000 µA, 1 ms, 20 Hz) obtained from a Grass S88 stimulator through a PSIU6 isolation unit. Sufficient time (10 min) was allowed between stimuli to permit the hemodynamic parameters to return to control values.

Microinjections. Kainic acid (20 mM), kynurenic acid (20 mM), bicuculline methiodide (4 mM), or buffered saline (vehicle), pH 7.4, was microinjected in sites within the RVL. The site was usually 2.6 mm rostral, 2.0 mm lateral, and 2.6 mm ventral in relation to the calamus scriptorius. Microinjection pipettes were drawn from calibrated glass tubing and broken back to a tip diameter of 40 µm. Injections were made by pressure, and the volume delivered (100 nl) was controlled by monitoring the displacement of the meniscus in the pipette through a dissecting scope. Animals received intramedullary microinjections contralateral to the stimulated nerve. At the end of the experiments, the injection sites were marked with the same volume of 2% Evans blue.

Histology. The animals were perfused transcardially with 10% Formalin. Brain stems were removed, cut coronally into 40-µm-thick sections, and stained with 1% neutral red. Sections containing the dye were examined and plotted on drawings adapted from a rat stereotaxic atlas (14) with the aid of a camera lucida attachment (Nikon). A typical injection site is shown in Fig. 1. Only data for experiments in which microinjections were centered within the RVL were considered.

View larger version (56K): [in this window] [in a new window]   Fig. 1.   Histological verification of a typical microinjection into the rostral ventrolateral medulla (RVL). Right: photomicrograph of coronal section stained with 1% neutral red at the level of the center of the microinjection (dark area) in the RVL. Left: camera lucida drawing of the same coronal section. Amb, nucleus ambiguus; NTS, nucleus of the solitary tract; py, pyramidal tract; Sp5, spinal trigeminal nucleus.

Drugs. All drugs were obtained from Sigma Chemical (St. Louis, MO). Vehicle solutions consisted of PBS, pH 7.4.

Data analysis. To quantitate the cardiovascular effects of nociceptive stimulation for each parameter, the peak values during the stimulation period were considered. HVC was calculated by dividing the Doppler shift (Hz) by MAP (mmHg). Changes in HBF and HVC are expressed as a percentage of control. Results are presented as means ± SE. The effects of drugs microinjected into the RVL were assessed by a paired two-tailed t-test. A value of P < 0.05 was considered to denote a significant difference.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cardiovascular responses induced by SNS. In initial experiments, we sought to determine the effects of electrical stimulation of high-threshold fibers of the sciatic nerve on the MAP, HR, and HBF ipsilateral to the stimulated sciatic nerve. As shown in Fig. 2A, high-current stimulation of the sciatic nerve (1,000 µA, 1-ms duration, 20 Hz for 10 s) elicited a rise in MAP and HBF with a discrete tachycardia. The latency between the onset of stimulation and the onset of responses varied from 1.0 to 2.0 s. These responses continued throughout the period of stimulation and returned to resting levels immediately after discontinuation of the stimulus. The increase in HBF was associated with a simultaneous increase of the calculated vascular conductance in that territory, indicating a local active vasodilation (Table 1). In control experiments, the vasodilation was not observed systematically in all animals. In ~30% of animals, the increase in blood flow was due to an increase in blood pressure without a change in HVC. Our major interest was to understand the mechanisms underlying the vasodilation elicited by SNS. Therefore, in the present study, only the animals that exhibited vasodilation during SNS were considered.

View larger version (11K): [in this window] [in a new window]   Fig. 2.   Typical example of changes in pulsatile (PAP) and mean (MAP) arterial pressure, heart rate (HR), and hindlimb blood flow (HBF) elicited by electrical stimulation of the sciatic nerve (1,000 µA, 1 ms, 20 Hz, 10 s) before (A) and 30 min after (B) microinjection of kainic acid (2 nmol/100 nl) into the RVL contralateral to the stimulated sciatic nerve. Bars represent 10-s trains of electrical stimulation.

Effects of kainic acid into the RVL on the cardiovascular responses induced by SNS. The pain transmission pathway ascending to the superior levels of the central nervous system involves cells in the dorsal horn of the medulla with their axons crossing the midline to ascend in the contralateral anterolateral funiculus. To establish the participation of the RVL in the cardiovascular responses to SNS, we examined these responses before and after microinjection of kainic acid (2 nmol/100 nl) into the RVL contralateral to the stimulated sciatic nerve. As shown in Fig. 2B, 30 min after microinjection of kainic acid into the contralateral RVL, a fall of 10 mmHg in MAP resting level was observed. On the other hand, the increase in MAP, HR, and HBF produced by SNS was completely abolished. As shown in Table 1, after unilateral microinjections of kainic acid into the RVL, resting levels of MAP, HR, and HBF were not modified. Increases in MAP and HBF induced by SNS were reduced drastically. Although the calculated basal levels of the HVC were not modified after kainic acid, the increases in HVC evoked by SNS were abolished, and vasodilation was no longer observed. In control experiments, microinjection of saline into the contralateral RVL did not modify basal arterial blood pressure, HR, HBF, or the cardiovascular responses to SNS. Microinjections of kainic acid in the same amount and volume into the RVL ipsilateral to the stimulated sciatic nerve did not modify basal parameters or cardiovascular responses to SNS.

Effects of kynurenic acid into the RVL on the cardiovascular responses induced by SNS. Previous studies have indicated that glutamatergic receptors in the RVL are involved in the cardiovascular responses to somatic afferent stimulation. In this group of experiments, we investigated the cardiovascular responses to SNS before and after unilateral microinjection of kynurenic acid (2 nmol/100 nl) into the RVL contralateral to the stimulated sciatic nerve. As shown in Table 2, unilateral microinjection of kynurenic acid into the RVL did not alter the resting levels of MAP or HR. The HBF was reduced discretely, but this reduction was not enough to modify the resting levels of the calculated HVC. After microinjection of kynurenic acid, the tachycardia produced by SNS was reduced, and the pressor response was abolished completely. Actually, it was replaced by a discrete fall in MAP. A discrete reduction in the HBF increase in response to SNS was observed. However, in this condition, the calculated HVC indicated that the observed vasodilation was greater than that observed in the control period. Figure 3 shows a typical example of this group. As shown in Fig. 3A, the SNS (1,000 µA, 1-ms duration, 20 Hz for 10 s) elicited a rise in MAP, HR, and HBF with an associated increase in the calculated HVC, indicating a vasodilation in that territory. Ten minutes after microinjection of kynurenic acid into the RVL contralateral to the stimulated sciatic nerve (Fig. 3B), SNS performed as described above elicited a discrete fall in MAP instead of hypertension, no change in HR, and an increase in HBF similar to that observed in the control situation. The calculated HVC indicated a greater vasodilation.

View larger version (11K): [in this window] [in a new window]   Fig. 3.   Typical example of changes in PAP, MAP, HR, and HBF elicited by electrical stimulation of the sciatic nerve (1,000 µA, 1 ms, 20 Hz, 10 s) before (A) and 10 min after (B) microinjection of kynurenic acid (2 nmol/100 nl) into the RVL contralateral to the stimulated sciatic nerve. Bars represent 10-s trains of electrical stimulation.

Effects of bicuculline into the RVL on the cardiovascular responses induced by SNS. We have demonstrated that the RVL is the efferent arm of the pressor and vasodilating responses to SNS and that, by blocking glutamatergic receptors in this region, the pressor response to SNS was abolished without modifying the vasodilation in the stimulated limb. A previous study demonstrated that, in the rat, the muscle vasodilation associated with the defense reaction is due to a reduction of the sympathetic vasomotor tone. To test the hypothesis that the vasodilation induced by SNS is due to inhibition of RVL sympathoexcitatory neurons, the cardiovascular responses to SNS were analyzed before and after microinjection of the GABA_A receptor antagonist bicuculline (400 pmol/100 nl) into the RVL contralateral to the stimulated sciatic nerve. Unilateral microinjection of bicuculline into the RVL produced a discrete hypertension, without changing basal HR, HBF, or HVC levels (Table 3). The pressor response to SNS was discretely higher but without reaching statistical significance. Tachycardia was replaced by bradycardia. The increase in HBF was drastically reduced, and the calculated HVC decreased, indicating a vasoconstriction in that territory in response to SNS instead of the vasodilation observed in the control situation. As shown in Fig. 4A, the SNS (800 µA, 1-ms duration, 20 Hz for 10 s) produced an increase in MAP, HR, and HBF, with an increase in the calculated HVC. Ten minutes after bicuculline microinjection into the RVL contralateral to the stimulated sciatic nerve (Fig. 4B), the increase in HBF produced by SNS was abolished, the tachycardia was replaced by bradycardia, and the pressor response was increased. In this condition, the calculated HVC decreased, indicating a vasoconstriction in that territory.

View larger version (11K): [in this window] [in a new window]   Fig. 4.   Typical example of changes in PAP, MAP, HR, and HBF elicited by electrical stimulation of the sciatic nerve (800 µA, 1 ms, 20 Hz, 10 s) before (A) and 10 min after (B) microinjection of bicuculline (400 pmol/100 nl) into the RVL contralateral to the stimulated sciatic nerve. Bars represent 10-s trains of electrical stimulation.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present study demonstrated that electrical stimulation of sciatic nerve afferents of Urethane-anesthetized, paralyzed, and artificially ventilated rats produced stimulus-locked increases in blood pressure, HR, and blood flow of the hindlimb ipsilateral to the stimulated sciatic nerve. Furthermore, SNS was associated with increased calculated HVC, indicating vasodilation in the stimulated hindlimb. Unilateral microinjections of the neuroexcitatory/neurotoxic agent kainic acid into the contralateral RVL abolished pressor, HR, and blood flow responses to SNS. Similarly, microinjections of the glutamatergic antagonist kynurenic acid into the contralateral RVL selectively blocked the pressor responses without changing the increase in HBF and vascular conductance. In contrast, microinjections of the GABA_A receptor antagonist bicuculline selectively blocked the vasodilation of the stimulated limb, whereas pressor responses were unaffected. These data suggest that cardiovascular responses to SNS are mediated by neurons within the RVL. Glutamatergic synapses within this region appear to mediate pressor responses while GABAergic synapses may mediate vasodilatory responses.

Sciatic afferent fibers innervate cutaneous and muscle thermo-, mechano-, and nociceptors (6). Previous studies have indicated that low-frequency stimulation with intensities sufficient to stimulate group II and III afferent fibers produced falls in blood pressure, whereas increasing the frequency of stimulation or the stimulus strength to include group IV (C) afferents always produced pressor responses (9, 19). In a previous study, Morrison and Reis (13) demonstrated that the sympathoexcitation evoked by single stimuli applied to the sciatic nerve of Urethane-anesthetized rats is mainly due to stimulation of group III (A) afferents, although during high-frequency stimulation temporal facilitation may contribute to increasing the role of group IV afferents in the somatosympathetic reflexes observed. The parameters used in the present study are well within the range that stimulates group III and IV afferents. During preliminary experiments to determine the best parameters of stimulation, we observed that the thresholds for pressor and blood flow responses were very similar. Also, in accordance with previous studies (20), we never observed falls in blood pressure in response to stimulation of the sciatic nerve. Therefore, it is conceivable that the increases in arterial blood pressure and HBF induced by the SNS in the present study arise from the stimulation of the same set of afferents fibers, probably A (group III) and C (group IV) cutaneous and muscle afferents.

A new result obtained in the present study was the demonstration that in rats, beside producing tachycardic and pressor responses, SNS is also associated with vasodilation in the territory of the stimulated limb. In a previous study (5), we reported that high-intensity cutaneous stimulation produced a similar pattern of pressor and vasodilatory responses in the stimulated limb of anesthetized or decerebrate-unanesthetized cats. This vasodilatory response was not modified by cholinergic blockade with atropine, indicating that it did not involve activation of sympathetic cholinergic vasodilatory fibers. Yardley and Hilton (25) obtained significant muscle vasodilation in rats by stimulating hypothalamic sites associated with the defense reaction. According to these authors, the vasodilation was due to a combined reduction of the sympathetic vasoconstrictor tone in the initial stage and activation of ₂-adrenergic receptors by catecholamines released in the later stage. In preliminary experiments (data not shown), we found that the vasodilation induced by SNS was not affected by systemic administration of propranolol up to 4 mg/kg, suggesting that activation of ₂-receptors is not essential for the vasodilatory response. Hence, it is conceivable that the increase in the HVC evoked by SNS observed here is largely due to withdrawal of sympathetic vasoconstrictor tone.

Unilateral microinjection of the neurotoxic agent kainic acid into the RVL contralateral to the stimulation site abolished the cardiovascular responses to SNS. These results are in accordance with previous studies indicating that the RVL represents a critical site for pressor responses induced by stimulation of cutaneous and visceral afferents (8, 13, 20, 21). Our results further extend those previous studies demonstrating that, besides integrating pressor responses, the RVL also mediated the vasodilation induced by SNS. Important to note is the fact that, although blocking responses to SNS, kainic acid unilaterally applied to the RVL did not change basal levels of arterial pressure or the calculated HVC. Therefore, the elimination of vasodilatory responses may not be attributed to a loss of vasomotor tone of the stimulated limb.

Previous studies have demonstrated that kynurenic acid, a broad-spectrum glutamate receptor antagonist, when administered intracisternally blocks the pressor response and the increase in sympathetic nerve activity produced by SNS (22). When microinjected into the RVL, kynurenic acid blocked the pressor response to static muscular contraction (3) and the sympathoexcitation produced by stimulation of high-threshold vagal afferents (21). Similar results were obtained in the present study. Unilateral microinjection of kynurenic acid into the RVL contralateral to the stimulated sciatic nerve blocked the pressor and HR responses to SNS. Although the increase in HBF during SNS diminished after kynurenic acid microinjection, the increase in HVC was greater than in the control situation. It is interesting to note that, after glutamatergic blockade of the RVL, SNS was often associated with depressor responses, i.e., falls in blood pressure. One possible explanation for these depressor responses is that once the pressor response is blocked, the vasodilation of the stimulated hindlimb decreases systemic vascular resistance, reducing MAP. A reversal of the pressor response to SNS to a depressor response was also observed after bilateral blockade of non-N-methyl-D-aspartate glutamatergic receptors within the RVL (8).

The existence of distinct mechanisms for the pressure and blood flow responses was also suggested in some experiments in which, despite marked increases in blood pressure and HBF produced by SNS, calculated vascular conductance did not rise, and vasodilation of the stimulated limb could not be observed. In agreement with our previous observation, even in these animals, microinjecting kynurenic acid into the contralateral RVL selectively blocked the pressor response. The increases in HBF remained unchanged, and in this condition the calculated HVC showed a significant increase, indicating a vasodilatory response to SNS. Accordingly, it is reasonable to believe that, in these animals, a vasodilatory mechanism was activated during SNS, although it was masked by the pressor response.

Taken together, the results obtained with microinjection of kainic or kynurenic acid indicate that pressor and blood flow responses to SNS are integrated within the RVL contralateral to the stimulated nerve and that glutamatergic synapses mediate pressor but not blood flow responses. These data raise the possibility that the vasodilation produced by SNS may be due to a selective inhibition of the RVL sympathoexcitatory neurons, resulting in a reduction of the sympathetic vasoconstrictor activity in the stimulated limb. An alternative explanation would assume the existence of vasodilator neurons in the RVL activated by nonglutamatergic synapses during SNS. However, the existence of such neurons in the rat is still lacking.

Results obtained after microinjection of the GABA_A antagonist bicuculline support the former hypothesis. Bicuculline microinjection into the RVL selectively blocked the blood flow responses to SNS, whereas pressor responses were not affected. Although unilateral bicuculline microinjection produced a significant increase in systemic arterial blood pressure, an equivalent increase in HBF was observed, and the calculated vascular conductance levels after bicuculline were comparable to those obtained in the control situation. These results are compatible with the hypothesis that vasodilatory responses to SNS were due to GABAergic inhibition of RVL sympathoexcitatory neurons.

A previous study demonstrated that electrical stimulation of the sural nerve in the anesthetized rabbit produced a biphasic response characterized by an early peak of excitation followed by a late inhibitory trough of the renal nerve sympathetic activity (10). A similar biphasic excitatory-inhibitory response was observed in the activity of RVL sympathoexcitatory neurons during electrical stimulation of cutaneous afferents (23). Both studies demonstrated that the inhibitory component of these responses was abolished after microinjection of bicuculline into the RVL.

Neuroanatomical data indicate that GABAergic neurons are distributed widely within the medulla oblongata reticular formation, including the RVL, the caudal ventrolateral medulla (CVL), the nucleus tractus solitarius, and raphe (18, 11, 12). Within the RVL, putative adrenergic and nonadrenergic bulbospinal neurons receive GABAergic synaptic profiles (12). A possible mechanism for the vasodilatory response would include the activation of these GABAergic cells by spinoreticular afferents, inducing inhibition of RVL sympathoexcitatory neurons. Because in our study the vasodilatory responses were not abolished after kynurenic acid microinjection into the RVL, it is conceivable that these GABAergic neurons are located outside the RVL. Alternatively, if these cells are within the RVL, they are excited by a nonglutamatergic mechanism. Results obtained by Masuda et al. (10) favored the hypothesis that the CVL represents the source of the sympathoinhibition induced by cutaneous afferent stimulation. These authors demonstrated that, after chemical inactivation of the CVL with kainic acid microinjections, the late inhibitory component of the somatosympathetic reflex was reduced severely.

In concert with previous studies, the present results suggest that the RVL integrated the cardiovascular responses to stimulation of somatic afferents. Within the RVL, specific groups of sympathoexcitatory neurons may be excited or inhibited according to their function, leading to the observed integrated cardiovascular response that included hypertension and selective redistribution of blood flow.

Perspectives