Journal of the Autonomic Nerr,ous System, 40 (1992) 91-98 © 1992 Elsevier Science Publishers B.V. All rights reserved 0165-1838/92/$05.00
Neurons in the caudal ventrolateral medulla mediate the somato-sympathetic inhibitory reflex response via G A B A receptors in the rostral ventrolateral medulla Noboru Masuda, Youichirou Ootsuka and Naohito .Terui Institute of Basic Medical Sciences, Uni~'ersityof Tsukuba, lbaraki, Japan (Received 10 April 1992) (Revision received and accepted 18 May 1992)
Key words: Rostral v e n t r o l a t e r a l m e d u l l a ; C a u d a l v e n t r o l a t e r a l m e d u l l a ; Bicuculline; K y n u r e n i c acid; G A B A ; G l u t a m a t e ; S o m a t o - s y m p a t h e t i c reflex; A r t e r i a l b a r o r e c e p t o r reflex
Abstract In urethane-anesthetized rabbits, stimulation of the sural nerve, consisting of cutaneous afferents (A-fibers), evoked reflex responses consisting of an early small excitatory component followed by a prolonged inhibitory component in renal sympathetic nerve activity. Bilateral injections of GABA antagonist, bicuculline (4 nmol/site), into the rostral ventrolateral medulla (RVLM), where sympatho-excitatory reticulospinal neurons are located, attenuated the inhibitory component in a dose-dependent manner as well as the inhibition evoked by stimulation of the aortic nerve A-fibers (baroreceptor afferents). Bilateral injections of a neurotoxic agent, kainic acid (4 nmol/site, 3 sites/side), into the caudal ventrolateral medulla (CVLM), where sympatho-inhibitory neurons with axonal projection to the RVLM are located, diminished these sympatho-inhibitory responses. Therefore it is concluded that the sympatho-inhibition evoked by activation of somatic afferents was mediated by neurons in the CVLM and by GABA receptors in the RVLM, as was the sympatho-inhibition associated with the arterial baroreceptor reflex. Bilateral injections of kynurenic acid (4 nmol/site, 3 sites/side) into the CVLM did not affect the somato-sympathetic reflex response, but diminished the sympatho-inhibition produced by activation of the baroreceptor afferents, Sympatho-inhibitory neurons in the CVLM were activated by glutamate when baroreceptor afferents were activated, but another excitatory transmitter may participate in the somato-sympathetic reflex in the CVLM.
Introduction T h e s o m a t o - s y m p a t h e t i c reflex has b e e n extensively investigated over the last 20 years , b u t its c e n t r a l o r g a n i z a t i o n , especially with respect to
Correspondence to: N. Terui, Institute of Basic Medical Sciences, University of Tsukuba, 1-1-1 Tennohdai, Tsukuba-shi, Ibaraki 305, Japan.
n e u r o t r a n s m i t t e r s , is still u n c e r t a i n . Recently, at least two kinds of cardiovascular n e u r o n s in the m e d u l l a have b e e n i d e n t i f i e d [1,2,12,20,21]. R e t i c u l o s p i n a l n e u r o n s located in the rostral ventrolateral m e d u l l a ( R V L M ) , r e f e r r e d to h e r e i n as R V L M n e u r o n s , are t h o u g h t to be excitatory i n t e r n e u r o n s that control activity of sympathetic p r e g a n g l i o n i c n e u r o n s of t h e s p i n a l c o r d [1,2,12,21]. T h e s e n e u r o n s receive various periph-" eral i n p u t s i n c l u d i n g somatic sensory o n e s
92 [19,21,22]. The other kind of cardiovascular neurons are neurons possessing axonal projection to the RVLM, which are located in the caudal ventrolateral medulla (CVLM) . These neurons, termed CVLM neurons, are activated by excitation of the arterial baroreceptor afferents, and inhibit activity of the RVLM neurons [13,20]. The CVLM neurons may also participate in the somato-sympathetic reflex. In the present study, we identified one of the neurotransmitters that mediate sympatho-inhibition evoked by stimulation of somatic afferents, and clarified the role of neurons in the caudal ventrolateral medulla in this reflex response by means of microinjections of receptor antagonists and of a neurotoxic agent into the rostral and caudal ventrolateral medulla.
Materials and Methods
Preparation of animals Experiments were performed on 18 adult albino rabbits (New Zealand White) of both sexes weighing between 2.4 and 3.8 kg. They were anesthetized with an intravenous administration of urethane (1.0-1.2 g/kg). Additional doses of anesthesia were given when necessary. After insertion of tracheal, arterial and venous cannulas, the animal was paralyzed with gallamine triethiodide (Sigma, initially 30 rag, i.v., thereafter 20 mg every 1-1.5 h), and artificially ventilated by a respiratory pump (Harvard, 661) with a gas mixture of 20% 0 2 and room air. The end-expiratory CO z concentration was constantly monitored (Beckman, LB-2) and was maintained at 3.5-4.5% by changing the volume a n d / o r the frequency of the respirator. Rectal temperature was maintained at 37-38°C by a thermostatically regulated heating pad and an infrared lamp connected to a rectal thermosensor probe. Bilateral vagal and aortic nerves were cut in the neck. Carotid sinus nerves were kept intact. After preparatory surgical procedures, the animal was fixed to a stereotaxic frame in a prone position with the head tilted downwards at an angle of about 45 degrees from the horizontal plane. The neck flexion was adjusted so that the
dorsal surface of the medulla was positioned horizontally. The atlanto-occipital membrane was exposed, cut at the midline and then retracted. To reduce respiratory movement of the brainstem, a pneumothorax was induced in all experiments. Mean arterial pressure (MAP) was maintained above 70 mmHg. When necessary, phenylephrine (50 p.g/ml in a mixture of 6c~: dextran-70 and 5% glucose solution) was injected intravenously.
Measurement of cardiovascular rariables and peripheral nerl'e activity Arterial pressure (AP), MAP and heart rate (HR) were monitored continuously in all experiments. AP was recorded from the abdominal aorta by a polyethylene catheter through the right femoral artery and connected to a pressure transducer (Nihon Kohden, MPU-0.5). H R was computed from AP pulses by a tachometer unit (Nihon Kohden, AT-600G). The left renal nerve was approached retroperitoneally through a left flank incision, and prepared for recording from near the renal artery. The central end of the cut nerve was placed on a pair of bipolar Ag electrodes connected to an amplifier, and its electrical activity was displayed on an oscilloscope (Tektronix, 7613). The lower and higher cut-off frequencies of the recording system were 100 and 5000 Hz, respectively. Multifiber renal nerve discharges (renal sympathetic nerve activity; RSNA), representing those of vasoconstrictor fibers , were converted into trains of standard pulses by a window discriminator (Nihon Kohden, EN-601J), the threshold of which was set slightly above the noise level. The discharge frequency was counted by a frequency counter with a bin width of 1 or 2 s and displayed on a polygraph (Nihon Kohden, RM6000) through a digital-to-analog converter. Electrical stimulation of peripheral nerves The right or left sural nerve, consisting of the cutaneous afferents of the hind limb, was isolated from the surrounding tissue at the surface of the gastrocnemius, cut peripherally, placed on a bipolar silver stimulating electrode, and stimulated. Rectangular pulses (0.2 ms duration, 2-3 pulses,
93 10 ms interval) were delivered from a pulse generator (Nihon Kohden, SEN 7103) through an isolation unit. At about 40 mm proximal to the stimulating site, compound action potentials were monitored continuously with Ag electrodes so that the stimulus intensity was confirmed to be sufficient to excite all of the A6 (Gnl)-fibers, as well as the A/3 (Gii)-fibers, but insufficient to excite the C (Giv)-fibers. The stimulus intensity was varied between 2.0 V and 5.0 V for activation of all A-fibers (n = 18). The central end of the left aortic nerve was prepared in the neck and mounted on stimulating electrodes. Train pulses (1.0-1.8 V intensity, 0.2 ms duration, 10 pulses, 10 ms interval) were used to evoke a reflex response. With this stimulation, only aortic nerve A-fibers, which consist exclusively of baroreceptor fibers, were activated .
(0.0, 0.5, and 1.0 mm caudal to the obex, 2.75 mm lateral to the midline and 3.5 mm from the dorsal surface of the medulla) were also determined sterotaxically. As shown in a previous paper , in order to cover the rostro-caudal extent of the CVLM, the multiple injections were needed to affect the arterial baroreceptor reflex. In order to evaluate the effect of chemicals, it is necessary to inject bilaterally. However, in some experiments the medulla was split at the midline and the rostral end of the first segment of the right cervical spinal cord was transected. In such preparations, only neural connections between the left medulla and the left spinal cord were kept intact and the effect of chemicals could be estimated by unilateral injection. In the following text, this preparation is termed the 'hemi-medulla' preparation.
Injection of chemicals into the ventrolateral medulla The methods of microinjection of chemicals into the medulla are described in a previous paper . T o summarize briefly, solutions were delivered through a single or 3-barrel glass capillary connected to a pressure source (1-5 kg/cm 2) via a solenoid valve. The injection volume was controlled by the pressure and opening time of the solenoid valve, and was monitored directly by the movement of the meniscus of each solution in the capillary. The following solutions were injected (volume 10-200 nl); bicuculline methiodide (Sigma; 20 raM) which was dissolved in artificial cerebrospinal fluid (ACSF; 124 mM NaC1, 2 mM KCI, 2 mM MgCI2, 2 mM CaCI2, 1.25 mM KHzPO4, 26 mM NaHCO3, 11 mM glucose, pH 7.4-7.6; ), kynurenic acid (Wako; 40 raM, dissolved in ACSF, pH adjusted to 7.4-7.6 with a few drops of 1 N NaOH), 2% Pontamine sky blue (Tokyo Kasei) dissolved in ACSF as a marker for the injection site, and kainic acid (Wako; 20 mM plus 2% Pontamine sky blue dissolved in ACSF) for chemical lesion. The injection sites of the rostral part of medulla were 2.5-3.0 mm rostral to the obex, 2.5-3.0 mm lateral to the midline and 4.0-4.5 mm ventral from the dorsal surface of the medulla where most RVLM neurons are located [15~21,22]. Three injection sites of the caudal part of the medulla
Data analysis Reflex responses in RSNA were analyzed by constructing peri-stimulus time histograms (PSTH) with the aid of a computer (ATAC 450, Nihon Kohden). Response magnitude of an inhibitory component evoked by stimulation of the sural nerve was calculated as follows: (mean spike frequency of pre-stimulus p e r i o d - mean spike frequency of response period) / (mean spike frequency of pre-stimulus period) ×100%. The pre-stimulus period was 100 ms before the onset of stimulation. The response period to the sural nerve stimulation was defined as the first second after the RSNA was lowered to the pre-stimulus level. Response magnitude of inhibition evoked by stimulation of the aortic nerve was calculated in the same way, but the response period was 200 ms and started 150 ms after the onset of stimulation. The amplitude of excitation was calculated as follows: (spike number at the bin of the peak of excitation) / (mean spike number per bin of pre-stimulus period) × 100%. The effect of chemicals on the reflex response was measured at 10-30 min after the injection (or after the last injection when multiple injections were made), since the maximal effect was obtained within 30 min and after 2 h the reflex responses recovered except in the case of the kainic acid injection.
94 Numerical values in the text were expressed as m e a n _+ 1 SD.
Histological examination At the e n d of each experiment, the a n i m a l was perfused with saline followed by 10% formalin. The brain a n d spinal cord were post-fixed, frozen and sectioned at 50 g i n . Injection sites were d e t e r m i n e d by the deposits of P o n t a m i n sky blue.
Reflex responses in RSNA Reflex response in R S N A evoked by stimulation of the A/3- a n d A 6 - f i b e r s of the sural nerve, the hind limb c u t a n e o u s afferent, consisted of a
small early excitation (excitatory c o m p o n e n t ) l~)llowed by p r o l o n g e d inhibition (inhibitory compon e n t ) (Fig. IB). In seven animals out of 18 CNS intact animals, the excitatory c o m p o n e n t was not a p p a r e n t , but after blockade of the inhibitory c o m p o n e n t (see below) or in 'hemi-medulla" p r e p a r a t i o n s (see Materials a n d Methods) this excitatory c o m p o n e n t b e c a m e clearer. T h e onset latency of the excitatory c o m p o n e n t was 110 _+ 24 ms and the a m p l i t u d e of this c o m p o n e n t was 191 _+48% (n = 10) in CNS intact animals. In ' h e m i - m e d u l l a ' p r e p a r a t i o n s , these values were 126_+ 15 ms a n d 356_+ 175% ( n = 7 ) , respectively. T h e onset latency of the inhibitory compon e n t was 184 + 26 ms and the response magnitude of this c o m p o n e n t was 93_+ 11% (n = 18). C o m p l e t e inhibition of R S N A was of a d u r a t i o n
pulses/bin B SUstlm
After Injection of Bicuculline into RVLM
Fig. 1. The effect of bilateral injections of bicuculline into the RVLM on aortic nerve- and somatic nerve-sympathetic reflexes. Reflex responses in RSNA evoked by stimulation (horizontal bar) of the aortic nerve (A and C) and by stimulation (arrows) of the sural nerve (B and D). In this and following figures, each peri-stimulus histogram (5 ms bin) was constructed from 64 successive trials. Before (A and B) and after (C and D) bicuculline (40 raM) injection (100 nl/site, a single site/side). E° Distribution of Pontamine sky blue that was the same as the volume of the bicuculline solution was indicated by solid areas in the frontal section at 2.5 mm rostral to the obex. Abbreviations in this and in the following figures are: ION, inferior olivary nucleus; LRN, lateral reticular nucleus; NA, nucleus ambiguus; NTS, nucleus tractus solitarius; NV, nucleus of trigeminal nerve; NXI1, nucleus of hypoglossal nerve; Ph, prepositus hypoglossal nucleus; Pyr, pyramidal tract; RFN, retrofacial nucleus: Tr sp V, spinal tract of trigeminal nerve.
equal to 795 + 256 ms (n = 18) and inhibition persisted for more than 1.3 s in CNS intact animals. In 'hemi-medulla' preparations, onset latency of the inhibitory component slightly increased to 224 _+ 28 ms and duration of the inhibition reduced greatly to about 500 ms (n = 7). These reflex components were mediated by supraspinal structures because neither an excitatory nor an inhibitory response with the same latency was evoked in RSNA after complete spinal transection at the C1 level (n = 2). The onset latency of inhibition in RSNA produced by stimulation of the left aortic nerve Afiber was 146 + 17 ms. Inhibition persisted for about 300 ms and the response magnitude of the
B Inhibition by SU stim
A Inhibition by AN stim
'°t =~_~~L40t ~o
Dose I nmol )
1 ( nmol )
C Excitation by SU stim g 600 ]
Dose } nmol )
Fig. 2. The effect of bilateral injections of bicuculline into the R V L M on the inhibition evoked by stimulation of the aortic nerve (A) and the inhibitory component (B) and the excitatory component (C) evoked by stimulation of the sura] nerve (n = 3). In A and B, the control response magnitude of the inhibition evoked by stimulation of the aortic nerve and control response magnitude of the inhibitory component evoked by stimulation of the sural nerve were expressed as ]00% (ordinate), respectively. In C, mean spike frequency of the pre-stimulus period of PSTH of control animals was expressed as 100% (ordinate) and each amplitude of excitation was calculated. Mean amplitude of excitation of control animals is 280:1:68% (n = 3). In each panel, the dose (abscissa, nmol) is converted from the volume of bicuculline (20 mM) that was injected cumulatively. Error bars indicate +1 SD.
inhibition was 71 + 16% (n = 18) (Fig. 1A). In 'hemi-medulla' preparations, these values were not changed significantly.
Effect of injection of chemicals on inhibitory responses Injections of bicuculline methiodide (20 mM of 200 nl/site), an antagonist of y-aminobutyric acid (GABA), into the bilateral RVLM increased the tonic activity of the renal nerve to 280% (mean, n = 4) of control level and also increased in mean arterial pressure (MAP) to 170 mmHg. These increases in RSNA and MAP lasted about 1 h and then gradually returned to the control level. The inhibitory component of the somato-sympathetic reflex and inhibition of RSNA produced by the aortic nerve stimulation completely disappeared or greatly reduced their response magnitude (Fig. 1C, D). Reduction of the inhibitory component by injection of bicuculline into the RVLM was dose-dependent and very similar to that of the inhibition evoked by aortic nerve stimulation (Fig. 2A, B). Injections of kainic acid (20 mM, 100 nl/site, 3 sites/side) into the bilateral CVLM in CNS intact animals (n = 2) or unilateral CVLM in a 'hemi-medulla' preparation (n = 1) decreased transiently the RSNA and MAP. However, 10-20 min after injections the RSNA and MAP increased over the control level (mean; +30%, + 30 mmHg, respectively, n = 3). These increases in RSNA and MAP lasted more than 1 h and then gradually returned to the control level. The inhibitory component of the somato-sympathetic reflex and the inhibition of RSNA produced by the aortic nerve stimulation completely disappeared (Fig. 3) or greatly reduced in the response magnitude (less than 20% of control response, n = 3). Multiple injections of kynurenic acid (40 mM, 200 nl/site, 3 sites/side) into the CVLM in the CNS intact animals (n = 2) or 'hemi-medulla' preparations (n = 2) increased in RSNA and MAP (mean; + 5 2 % and +20 mmHg, respectively) for more than 1 h. This manipulation did not affect the somato-sympathetic reflex, but the inhibitory response evoked by stimulation of the aortic nerve attenuated (Fig. 4) or almost corn-
pletely disappeared (7% of control response n = 4).
Barosensitive reticulospinal neurons in the RVLM respond to stimulation of the cutaneous or muscle afferents [8,19,22]. The temporal response pattern in these neurons is the same or very similar to the one observed in activity of the peripheral vasoconstrictor nerve when somatic afferents were activated, but were preceded by about 100-140 ms [8,22]. Electrolytic or chemical lesion of the RVLM diminishes pressor response and the reflex responses in the renal nerve evoked by stimulation of the hind limb somatic afferents [11,17]. These previous studies indicate that the RVLM neurons are responsible for the medullamediated component  of the somato-sympathetic reflex.
The effect of injection of chemicals on excitatory responses Injections of bicuculline into the RVLM either did not change the amplitude of the early excitatory component evoked by stimulation of the sural nerve (n = 2), or they increased it ( n - - 2 ) . But this increase was not dose-dependent (Figs. I and 2C). Chemical lesions by injections of kainic acid into the CVLM either did not affect (Fig. 3), or slightly increased, the amplitude of the early excitatory component. Multiple injections of kynurenic acid into the CVLM produced almost no effect on the early excitatory component of the somato-sympathetic reflex (Fig. 4).
B SU sUm
After Injection of Kainic Acid into CVLM
pulses/bin D ,
_ . . . . . . . .
ou,ses/0,o . _ . _ _ _ . _ 1
rostrocaudal 0 mm
Fig. 3. The effect of bilateral injections of kainic acid into the CVLM on aortic nerve- and somatic nerve-sympathetic reflexes. Reflex responses in RSNA evoked by stimulation (horizontal bar) of the aortic nerve (A and C) and by stimulation (arrows) of the sural nerve (B and D). Before (A and B) and after (C and D) kainic acid (20 mM) injection (100 nl/site, 3 sites/side). E. Distributions of Pontamine sky blue that was dissolved in kainic acid solution were indicated by solid areas in the frontal section at the obex level. Abbreviations as in legend to Fig. 1.
the somato-sympathetic reflex was attenuated dose-dependently in the same manner as was the attenuation of the inhibition associated with the arterial baroreceptor reflex (Figs. 2, 3). Barosensitive neurons in the CVLM that project to the RVLM are thought to participate in the arterial baroreceptor reflex by inhibiting activity of the RVLM neurons . These neurons in the CVLM were also activated by somatic afferent stimulation . Therefore, the same neurons in the CVLM may mediate the arterial baroreceptor reflex and inhibitory component of the somatosympathetic reflex via G A B A receptors of neurons in the RVLM. If so, at least two kinds of receptors must exist for excitatory transmission at CVLM neurons. One must be the glutamate receptor because injections of kynurenic acid, an glutamate antagonist, into the CVLM abolished sympatho-inhibition accompanied with the arterial baroreceptor reflex [4,5]. The second may be other than the glutamate receptor because the
Injection of kainic acid, a neurotoxic agent, into the CVLM diminished the inhibitory component of the somato-sympathetic reflex as well as the arterial baroreceptor reflex (Fig. 3). Iwamura et al.  reported that the 'silent period' produced by stimulation of the sciatic nerve was eliminated after electrolytic lesions were made in the ventral medulla of the cat. This observation and our present results may suggest that neurons in the CVLM are responsible for so the called 'silent period' or 'post-excitatory depression' that was commonly observed in the sympathetic nerve activity . Injection of bicuculline, a GABA antagonist, into the RVLM also diminished the inhibitory component of the somato-sympathetic reflex as well as the arterial baroreceptor reflex (Fig. 1). It has already been reported that the arterial baroreceptor reflex is mediated by G A B A receptors of neurons in the RVLM [5,15,18,23]. In the present experiments, the inhibitory component of
RSNA~'~ After Injection of Kynurenic Acid into CVLM
RSN~AJ 6 4 T ~ 0 100 ms
Fig. 4. The effect of bilateral injections of kynurenic acid into the C V L M on aortic nerve- and somatic nerve-sympathetic reflexes. Reflex responses in R S N A evoked by stimulation (horizontal bar) of the aortic nerve (A and C) and by stimulation (arrows) of the sural nerve (B and D). Before (A and B) and after (C and D) kynurenic acid (40 raM) injection (100 nl/site, 3 sites/side). E. Distributions of Pontamine sky blue that was the same volume of kynurenic acid solution was indicated by solid areas in the frontal section at 0.5 m m caudal to the obex. Abbreviations as in legend to Fig. 1.
inhibitory component of the somato-sympathetic reflex was not affected by the same procedure (Fig. 4). The excitatory component of the somato-sympathetic reflex was not ascribable to a disinhibition of the CVLM neurons since this component was still observed after G A B A receptors at the RVLM had been blocked. The excitatory neurotransmitter at the RVLM for the sympatho-excitation associated with activation of the somatic afferents, as well as the excitatory neurotransmitter at the CVLM for the inhibitory component, must be determined.
Acknowledgements We thank Miss Kazumi Kamemoto for technical assistance. This study was supported by Grants-in-Aid 03670049 for Scientific Research from the Ministry of Education, Science and Culture of Japan.
References 1 Barman, S.M. and Gebber, G.L., Axonal projection patterns of ventrolateral medullospinal sympathoexcitatory neurons, J. Neurophysiol., 53 (1985) 1551-1566. 2 Brown, D.L. and Guyenet, P.G., Electrophysiological study of cardiovascular neurons in the rostral ventrolateral medulla in rats, Circ. Res., 56 (1985) 359-369. 3 Dorward, P.K., Burke, S.L., Jiinig, W. and Cassel, J., Reflex responses to baroreceptor, chemoreceptor and nociceptor inputs in single sympathetic neurones in the rabbit and the effect of anaesthesia on them, J. Auton. Nerv. Syst., 18 (1987) 39-54. 4 Gordon, F.K., Aortic baroreceptor reflexes are mediated by N M D A receptors in caudal ventrolateral medulla, Am. J. Physiol., 252 (1987) R628-R633. 5 Guyenet, P.G., Fritz, T.M. and Donaldson, S.R., Role of excitatory amino acids in rat vagal and sympathetic baroreflexes, Brain Res., 407 (1987) 272-284. 6 lwamura, Y., Uchino, Y., Ozawa, S. and Kudo, N., Excitatory and inhibitory components of somato-sympathetic reflex, Brain Res., 16 (1969) 351-358. 7 Koizumi, K. and Brooks, C.M., The integration of autonomic system reactions: a discussion of autonomic reflexes, their control and their association with somatic reactions, Biochem. Exp. Pharmacol., 67 (1972) 1-68. 8 Kumada, M., Terui, N. and Kuwaki, T., Arterial baroreceptor reflex: its central and peripheral neural mechanisms, Progr. Neurobiol., 35 (1990) 331-361. q Kumada, M., Terui, N., Saeki, Y. and Kuwaki, T..
Barosensitive neurons in the rostral and caudal ventrolav eral medulla and their responses to activation of somatic afferent fibers. In M. Yoshikawa, M. Uono, II. Tanabc and S. lshikawa (Eds.), New Trends in Autonomic Nervous System Research, Elsevier Science Publishers B.V.. Amsterdam, 1991, pp. 393-396. 10 Lipski, J., Bellingham, M.C., West, M.J. and Pilowsky, P., Limitations of the technique of pressure microinjection of excitatory amino acids for evoking responses from localized regions of the CNS, J. Neurosci. Methods, 26 (1988) 169-179. 11 McAllen, R.M., Mediation of the fastigial pressor response and a somatosympathetic reflex by ventral medullary neurons in the cat, J. Physiol. (Lond.). 368 (1985) 423-433. 12 McAllen, R.M., Identification and properties of snb-retrofacial bulbospinal neurones: a descending cardiovascular pathway in the cat, J. Auton. Nerv. Syst., 17 (1986) 151 - 164. 13 Masuda, N., Terui. N.. Koshiya, N. and Kumada, M., Neurons in the caudal ventrolateral medulla mediate the arterial baroreceptor reflex by inhibiting barosensitivc reticulospinal neurons in the rostral ventrolateral medulla in rabbits, J. Auton. Nerv. Syst., 34 (1991) 103-118. 14 Numao, Y., Saito, M., Terui, N. and Kumada, M , Physiological and pharmacological properties of the three subsystems constituting the aortic nerve-renal sympathetic reflex in rabbits, J. Auton. Nerv. Syst., 9 (1983) 36t-380. 15 Saeki, Y., Terui, N. and Kumada, M., Participation of ventrolateral medullary neurons in the renal-sympathetic reflex in rabbits, Jpn. J. Physiol., 38 (1988) 267-281. 16 Sato, A. and Schmidt, R.F., Somatosympathetic reflexes: afferent fibers, central pathways, discharge characteristics, Physiol. Rev., 53 (1973) 916-947. 17 Stornetta, R.L., Morrison, S.F., Ruggiero, D.A. and Reis, D.J., Neurons of rostral ventrolateral medulla mediate somatic pressor reflex, Am. J. Physiol., 256 (1989) R448 R462. 18 Sun, M.-K. and Guyenet, P.G., GABA-mediated baroreceptor inhibition of reticulospinal neurons, Am. J. Physiol., 249 (1985) R672-R680. 19 Sun, M.-K. and Spyer, K.M., Nociceptive inputs into rostral ventrolateral medulla-spinal vasomotor neurones in rats, J. Physiol. (Lond.), 436 (1991) 685-700. 20 Terui, N., Masuda, N., Saeki, Y. and Kumada, M., Activity of barosensitive neurons in the caudal ventrolateral medulla that send axonal projections to the rostral ventrolateral medulla in rabbits, Neurosci. Lett., 118 (199(I) 211-214. 21 Terui, N., Saeki, Y. and Kumada, M., Barosensory neurons in the ventrolateral medulla in rabbits and their responses to various afferent inputs from peripheral and central sources, Jpn. J. Physiol., 36 (1986) 1141-1164. 22 Terui, N., Saeki, Y. and Kumada, M., Confluence of barosensory and nonbarosensory inputs at neurons in the ventrolateral medulla in rabbits, Can. J. Physiol. Pharmacol., 65 (1987) 1584-1590. 23 Urbanski, R.W, and Sapru, H.N.. Putative neurotransmitters involved in medullary cardiovascular regulation, J. Auton. Nerv. Syst., 25 (1988) 181-193.