INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

THERE IS CONSIDERABLE evidence that endothelin (ET) may play an important function as a neuromodulator and/or a neurotransmitter within the central nervous system (19, 24). In addition, various studies using immunohistochemistry, autoradiography, and in situ hybridization techniques have demonstrated the existence of all components of the ET system, including ET itself, its receptors, and ET-converting enzyme activity, in the brain (13, 15, 18-20, 28, 29). ET binding sites are found in the hypothalamus and the brain stem, which are known to be important sites for cardiovascular regulation (4, 16). In the brain stem, ET-1 and ET-1 binding sites are abundant in the nucleus of the solitary tract (NTS) and the ventrolateral medulla (13, 15, 18, 19). These results suggest that ET-1 within the brain stem, including the NTS, may contribute to the control of sympathetic nerve activity and arterial pressure.

The NTS is the site where afferent fibers from arterial baroreceptors, chemoreceptors, cardiopulmonary receptors, and other visceral receptors make their first synapse (1, 4, 5, 16). The NTS is thus believed to play an important role in the integration of the autonomic control of the cardiovascular system (1, 4, 5). Microinjection of ET-1 into the NTS is known to decrease both arterial pressure and heart rate in vivo (7, 24), thus suggesting that ET-1 increases neuronal activity of NTS neurons. However, it is not known whether ET-1 directly affects neuronal activity because microinjection of ET-1 into the NTS may cause medullary ischemia, which results in changes in the arterial pressure (6, 21, 22, 25). In addition, the role of endogenous ET within the NTS is yet to be elucidated. Furthermore, it has also been suggested that the injection of ET-1 into the cisterna magna sensitizes the arterial baroreflex control of heart rate (12). It is thus possible that ET-1 within the NTS influences the effect of such major neurotransmitters as L-glutamate (Glu) (32) or the responses of the NTS neurons to it.

To investigate the role of ET-1 in these pathways, we cut the semihorizontal brain slices containing the NTS of rats and recorded extracellular single-unit discharges from the medial NTS (mNTS) that responded to solitary tract (ST) stimulation, and we attempted to determine 1) whether ET-1 affects the neuronal activity of the NTS neurons, 2) whether selective ET receptor antagonists affect the neuronal activity of the NTS neurons, and 3) whether ET-1 modulates the excitability of the NTS neurons evoked by Glu.

A preliminary report of these data has been presented (26).

	METHODS

Top Abstract Introduction Methods Results Discussion References

Slice preparation. Wistar-Kyoto rats (WKY: 100-150 g, 4-6 wk old) were used in all experiments. Under ether anesthesia, the rat was euthanized by cervical dislocation, and the brain stem and cerebellum were rapidly removed to cold Krebs-Ringer solution containing (in mM) 126 NaCl, 5 KCl, 2.4 CaCl₂, 1.3 MgSO₄, 1.26 KH₂PO₄, 26 NaHCO₃, and 10 D-glucose, saturated with 95% O₂ and 5% CO₂. A semihorizontal brain stem slice (400 µm thick) containing the area postrema, the NTS, and the ST was obtained (2, 27) using a vibratome (DTK-1000; Dosaka, Kyoto, Japan) supported by 3% agar-agar. In this slice, the ST provided major afferent innervations to the neurons in the NTS and was also long enough to allow synaptic stimulation of the NTS neurons via electrical stimulation of the ST. The slice was then incubated for at least 2 h before the start of the experiment in Krebs-Ringer solution bubbled with 95% O₂ and 5% CO₂.

Recording systems. The slices were placed on a Plexiglas mesh in a submerged recording chamber and were covered with nylon mesh and a silver ring to support the tissue. The recording chamber was perfused with oxygenated Krebs-Ringer solution at 34°C at a flow rate of 3 ml/min. A concentric stimulating electrode was placed on the ST 1.0-2.0 mm distant from the recording electrode, and square pulses (100 µs and ~15 V) were applied to the ST through the stimulating electrode at 0.3 Hz. These electrodes were manipulated under direct observation with a microscope (SV-6; Zeiss, Oberkochen, Germany). Extracellular single-unit discharges were recorded from neurons that responded to electrical stimulation of the ST with a glass micropipette filled with Krebs-Ringer solution (3-5 M). This electrode was driven into the mNTS until an action potential responding to ST stimulation was encountered. The average latency of responses to ST stimulation was 7 ms. Spikes were amplified (MEZ-7200 and MEG-1200; Nihon-Kohden, Tokyo, Japan), and the output was fed into a pulse counter (MET-1100, Nihon-Kohden) to show the numbers of spikes per second. The raw neurogram and firing rate were simultaneously displayed on an oscilloscope (DS-8605; Iwatsu, Tokyo, Japan) and recorded on an optical hard-copy recorder (RAT-1200, Nihon-Kohden).

Responses of the mNTS neurons to ET-1. After a single-unit discharge was encountered with the recording electrode in the mNTS area identified under the microscope, responses to iontophoretic application of ET-1 and Glu were elicited using a three-barrel glass electrode. This electrode was independent of the recording electrode. The most effective field was a circular area ~50 µm in diameter (11). The iontophoretic system was a Neurophore model BH-2 control unit (Medical System, New York, NY), and chemicals were prepared by the following recipes: ET-1 (10⁴ M in 1.5 mM NaOH, pH 8.5) (19) and Glu (0.5 M in distilled water, pH 7.5) (11) and NaCl (1 M) as a control current. The current used (250 ms in duration) was 8 to 20 nA for the Glu pipette and 8 to 30 nA for the ET-1 pipette, and no backing current was applied routinely. For ET-1 and Glu, the current was varied for each neuron because of variation of electrodes and ejection sites, and then currents were adjusted for ET-1 and Glu so as to get almost identical responses. In some experiments, modified Ringer solution containing low Ca²⁺ (0.1 mM) and high Mg²⁺ (4.3 mM) was perfused to eliminate neurotransmitter release from presynaptic terminals. Pharmacological examinations were performed on neurons that showed a stable firing more than 20 min. The basal firings were taken as the average spikes per second (Hz) for 5 min before drugs. The response was defined as the peak change in spikes per second.

Receptor subtype mediating by ET-1 response. To determine which receptor subtype is involved, we perfused the ET_A receptor antagonist (BQ-123) or the ET_B receptor antagonist (BQ-788) during iontophoretic ET-1 application. After the basal spontaneous neuronal activity and the stable responses to ET-1 were confirmed, Krebs-Ringer solution containing BQ-123 (10⁵-10⁴ M) or BQ-788 (10⁴ M) was then perfused to the recording chamber for 7 min at a flow rate of ~3 ml/min during the iontophoretic ET-1 application.

Effects of exogenous ET-1 or endogenous ET on the responses to iontophoretically applied Glu. To determine whether ET-1 modulates the responses to Glu, a major neurotransmitter in the NTS, we iontophoretically applied Glu before and after iontophoretic ET-1 application (n = 21). Glu was iontophoretically applied for 1 s at a fixed interval of 1 min. After the basal spontaneous neuronal activity and the stable responses to Glu were confirmed, ET-1 was iontophoretically applied for 5 s. Glu application every minute was continued.

To determine whether the responses to Glu are influenced by endogenous ET, we applied Glu iontophoretically during the perfusion with BQ-123 or BQ-788. After confirming the basal spontaneous neuronal activity and the stable responses to iontophoretically applied Glu, Krebs-Ringer solution containing BQ-123 (10⁴ M, n = 10) or BQ-788 (10⁴ M, n = 10) was perfused for 7 min to the recording chamber at a flow rate of 3 ml/min.

To investigate whether the blocking effects of BQ-123 were specific, we examined whether BQ-123 (10⁴ M) modulates the effect of ACh (0.5 M in distilled water, pH 3.5) (11, 27) on the neuronal activity of the NTS. ACh was applied iontophoretically.

Histological studies. At the end of the experiments, the slices were fixed with a 4% paraformaldehyde solution in PBS, frozen, sectioned at 100-µm thickness, and then stained with neutral red. A representative slice preparation is shown in Fig. 1.

View larger version (49K): [in this window] [in a new window]   Fig. 1.   Experimental preparation of the brain stem slice. A: photomicrograph of a 100-µm section cut with neutral red staining. NTS, nucleus of the solitary tract; AP, area postrema; DMV, dorsal motor nucleus of the vagus; ST, solitary tract; 4th, fourth ventricle. B: schematic drawing of slice preparation. Rec, recording electrode; IP, iontophoretic electrode; Stim, stimulating electrode.

Drugs. The following drugs were used: Glu (Sigma Chemical, St. Louis, MO), ET-1 (Peptide Institute, Osaka, Japan), ACh (Tocris Cookson, St. Louis, MO), ET_B receptor antagonist BQ-788 (Novabiochem, Tokyo, Japan). ET_A receptor antagonist BQ-123 was a gift from Banyu Pharmaceutical, Tokyo, Japan.

Statistical analysis. The data are expressed as means ± SE. Statistical significance was evaluated by an analysis of variance with repeated measures or Student's paired t-test as appropriate. Differences were considered significant at a value of P < 0.05.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Response of the mNTS neurons to ET-1. Fifty-five neurons that responded to the ST stimulation were examined, and all of those neurons responded to iontophoretically applied Glu and ET-1. ET-1 increased the neuronal activity from 2.1 ± 0.2 to 5.4 ± 0.4 spikes/s (P < 0.01). These effects were transient and current dependent (Fig. 2) and were observed even during perfusion with a solution containing low Ca²⁺ and high Mg²⁺, which eliminated synaptic transmission (data not shown). Figure 2 shows that spikes decrease or cease after intense excitation by Glu and the other excitants. This may come from inactivation of sodium channels owing to strong depolarization or from activation of an electrogenic sodium pump, because these phenomena were even observed under low-Ca²⁺ and high-Mg²⁺ solution. The current itself (through one of the multibarrel pipettes filled with NaCl solution) did not affect the neuronal activity.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Example of increased neuronal activity evoked by iontophoretically applied endothelin (ET)-1 (104 M; 10, 20, and 40 nA) on a single neuron of the medial NTS (mNTS) in a rat brain stem slice. Raw neurograms are shown at top, and firing rate is shown at bottom. Iontophoretically applied ET-1 increased neuronal activity as a function of applied current.

The increased neuronal activity evoked by iontophoretically applied ET-1 was attenuated by BQ-123 but not by BQ-788. The magnitude of increase in neuronal activity evoked by iontophoretically applied ET-1 was attenuated from 3.2 ± 0.6 to 2.2 ± 0.4 spikes/s during perfusion with BQ-123 (10⁵ M) (P < 0.05, n = 10). The magnitude of increase in neuronal activity evoked by iontophoretically applied ET-1 was attenuated from 2.3 ± 0.2 to 0.9 ± 0.2 spikes/s during perfusion with BQ-123 (10⁴ M) (P < 0.01, n = 10, Fig. 3, A and C). The magnitude of increase in neuronal activity evoked by iontophoretically applied ET-1 did not differ before or during perfusion with a high concentration of BQ-788 (10⁴ M) (2.9 ± 1.0 vs. 2.6 ± 0.7 spikes/s, NS, n = 10, Fig. 3, B and C).

View larger version (25K): [in this window] [in a new window]   Fig. 3.   Effects of ET receptor antagonists on ET-1-induced increases in neuronal activity of mNTS. A and B: raw neurograms (top) and firing rate (bottom) evoked by iontophoretically applied ET-1 before and during perfusion with BQ-123 (104 M) (A) or BQ-788 (104 M) (B) are shown. C and D: bar graphs showing change in the neuronal activity evoked by iontophoretically applied ET-1 before and during perfusion with BQ-123 (104 M) (C) or BQ-788 (104 M) (D). ** P < 0.01 vs. values before ET-1.

To examine whether the blocking effect of BQ-123 on the excitatory response evoked by ET-1 was specific for ET-1, we applied ACh iontophoretically to the neurons in the mNTS during perfusion with BQ-123 (10⁴ M). The magnitude of increase in neuronal activity evoked by the iontophoretically applied ACh did not change before or during perfusion with BQ-123 (2.6 ± 0.9 vs. 3.4 ± 0.7 spikes/s, NS, n = 3).

Effects of ET receptor antagonists on the spontaneous activity of the mNTS neurons. Perfusion with BQ-123 (10⁴ M) decreased the control spontaneous neuronal activity from 2.2 ± 0.2 to 1.8 ± 0.2 spikes/s (P < 0.01, n = 23), but that with BQ-788 (10⁴ M) did not alter the basal spontaneous neuronal activity (from 3.0 ± 0.5 to 3.0 ± 0.5 spikes/s, NS, n = 13).

Effects of ET-1 on the response to iontophoretically applied Glu. Iontophoretically applied Glu to the same site as ET-1 increased the neuronal activity of the mNTS neurons as shown in Fig. 4A. After ET-1 application, the magnitude of increase in neuronal activity evoked by iontophoretically applied Glu was augmented from 4.6 ± 0.5 to 7.4 ± 0.8 spikes/s (P < 0.01, n = 21, Fig. 4B). This augmentation peaked at 1-5 min after ET-1 application and then gradually returned to the control response by ~20 min (Fig. 4A).

View larger version (39K): [in this window] [in a new window]   Fig. 4.   Effects of ET-1 on L-glutamate (Glu)-induced increases in neuronal activity of mNTS. A: example of augmentation of responses to iontophoretically applied Glu (0.5 M, 10 nA) after iontophoretically applied ET-1 (104 M, 28 nA). Raw neurogram is shown at top, and firing rate is shown at bottom. B: bar graph showing magnitude of increases in neuronal activity evoked by iontophoretically applied Glu before and after iontophoretically applied ET-1. ** P < 0.01 vs. before ET-1 application.

Effects of ET receptor antagonists on the responses to iontophoretically applied Glu. The magnitude of increase in neuronal activity evoked by iontophoretically applied Glu was attenuated from 4.5 ± 0.3 spikes/s to 2.9 ± 0.6 spikes/s during perfusion with BQ-123 (10⁴ M) (P < 0.05, n = 10, Fig. 5, A and C). The magnitude of increase in neuronal activity evoked by iontophoretically applied Glu did not differ before or during perfusion with BQ-788 (4.8 ± 1.4 vs. 5.0 ± 1.1 spikes/s, NS, n = 3, Fig. 5, B and D).

View larger version (27K): [in this window] [in a new window]   Fig. 5.   Effects of ET receptor antagonists on Glu-induced increases in neuronal activity of mNTS. A and B: raw neurogram is shown at top, and firing rate is shown at bottom. BQ-123 but not BQ-788 attenuated Glu-induced increases in neuronal activity of mNTS. C and D: bar graphs showing magnitude of increases in neuronal activity evoked by iontophoretically applied Glu before and during perfusion with BQ-123 (104 M) (C) or BQ-788 (104 M) (D). * P < 0.05 vs. values before perfusion with BQ-123 or BQ-788.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Our results indicate that ET-1 increases the neuronal activity of mNTS neurons via ET_A receptors. Second, these findings suggest that endogenous ET contributes to the control of the basal spontaneous neuronal activity of the mNTS neurons. Finally, ET-1 augmented the responses to Glu, which is considered to be the major neurotransmitter from baroreceptors to the NTS (32), via ET_A receptors, thus suggesting that ET-1 may facilitate synaptic transmission in the NTS.

Effect of exogenously applied ET-1 on the neuronal activity of the mNTS neurons. We demonstrated that the iontophoretically applied ET-1 increased the neuronal activity of mNTS neurons. This finding also supports the results reported by other investigators (7, 24), which indicated that the microinjection of ET-1 into the NTS decreases both arterial pressure and heart rate in anesthetized rats in vivo. Furthermore, in the present study, iontophoretically applied ET-1 increased the neuronal activity of mNTS neurons during perfusion with modified Ringer solution containing low Ca²⁺ and high Mg²⁺, thus suggesting that ET-1 mainly acts on the postsynaptic membrane of neurons of the mNTS because the magnitude of increase in neuronal activity evoked by ET-1 after blocking synaptic transmission did not differ from that during perfusion with normal Ringer solution.

Because ET-1 is a potent vasoconstrictor agent, it is difficult to confirm in vivo whether the cardiovascular response of ET-1 results from neuronal activity or from the ischemia produced by intense vasoconstriction. It has been reported that ET-1 applied to the middle cerebral artery of the rat reduces the local cerebral blood flow (25). It has also been shown that the intracisternal administration of 5-500 pmol ET induces basilar artery contraction in vivo (22). Furthermore, it has been demonstrated that the intracisternal administration of 30 pmol ET-1 in WKY markedly increases the blood pressure with concomitant reductions in the blood flow through the caudal medulla (21). Our present studies using medulla slice preparations have clearly demonstrated that ET-1 affects neuronal activity of the mNTS directly. The possible increase of neuronal activity via ischemic insult by ET-1 injection in vivo systems can be excluded because the brain slice preparation is independent from the vascular system.

The increased neuronal activity evoked by ET-1 is mediated by ET_A receptors because it was attenuated by BQ-123 but not by BQ-788. The effect is specific for ET-1 because the responses to ACh did not change after BQ-123.

The role of endogenous ET-1 on the neuronal activity in the mNTS. Both ET_A and ET_B receptors are widely distributed in a variety of tissues, including the central nervous system. In the brain stem, there are abundant ET receptor binding sites. There are ET_A receptor binding sites in the intermediate and the caudal NTS (15, 18, 29). The ET_A receptor antagonist inhibited not only the responses to ET-1 but also the control spontaneous neuronal activity. These results thus indicate that responses to ET-1 are mediated by the ET_A receptors located on the postsynaptic membrane and that endogenous ET helps control the spontaneous neuronal activity of the mNTS. It has been reported that microinjection of BQ-123 into the NTS evokes a biphasic arterial pressure response in vivo characterized by an initial rapid hypotension followed by a sustained pressor effect (24). It is therefore possible that endogenous ET in the NTS may increase the spontaneous neuronal activity of NTS neurons, which thus influences basal arterial pressure and sympathetic nerve activity. The biphasic depressor/pressor changes in arterial pressure after microinjection of BQ-123 (24) probably indicate that this compound is a partial agonist and has a brief excitatory effect on the cell before blocking the receptors.

Augmented responses to Glu evoked by ET-1. Our findings do not clarify the precise mechanism(s) by which ET-1 augments the responses to Glu. In vascular smooth muscle, ET-1 elicits a biphasic increase in intracellular calcium concentration ([Ca²⁺]_i) (10, 31). It has been shown that ET-1 induces the depolarization of the ventral root in the newborn rat spinal cord through the L-type calcium channel (33). It has also been shown that ET-1 modulates [Ca²⁺]_i of neurons in primary cultures of rat cerebellum or slices of the hippocampus (14, 30). We thus speculate that ET-1 may increase [Ca²⁺]_i by augmenting the release from the intracellular stores or by facilitating the influx through voltage-dependent or receptor-operated calcium channels in the neurons of the mNTS, which may augment the response to Glu. Furthermore, D'Amico et al. (3) have shown that pretreatment with MK-801, an N-methyl-D-aspartate (NMDA) receptor antagonist, but not 6-cyano-7-nitroquinoxaline-2,3-dione, a non-NMDA receptor antagonist, blocks the cardiovascular responses caused by ET-1 in the periaqueductal gray area. Hashim et al. (8) have also shown that the hypotensive effect of intracerebroventricular ET-1 is attenuated by excitatory amino acid receptor antagonists. As a result, there may be some linkage between ET receptor activation and glutamate receptor activation, which may be related to an increase in [Ca²⁺]_i.

We demonstrated that during perfusion with BQ-123, the neuronal responses to Glu were attenuated, thus suggesting that endogenous ET-1 may also influence synaptic transmission in the NTS. The effects of BQ-123 are specific because 1) the equimolar concentrations of ET_B receptor antagonist did not affect the increased neuronal activity evoked by ET-1 in the mNTS and 2) BQ-123 did not affect the responses to iontophoretically applied ACh.

The results of the present study suggest that ET-1 may augment synaptic transmission and may thus facilitate the arterial baroreflex. In agreement with our results, Itoh and Buuse (12) showed that the intracisternal ET-1 administration facilitates arterial baroreflex control of the heart rate. On the other hand, a preliminary report has also shown that microinjection of BQ-123 into the NTS increases the arterial baroreflex sensitivity, thus suggesting that the activation of ET_A receptors inhibits the arterial baroreflex (23). Kuwaki et al. (19) have shown that intracisternal ET-1 administration inhibits the arterial baroreflex. In that study, however, ET-1 caused marked hypotension and bradycardia. It is thus difficult to compare the reflex response because of the marked baseline differences in both the arterial pressure and heart rate before and after ET-1. In addition, intracisternal ET-1 also affects other important ET-1-sensitive areas, such as the ventrolateral medulla, more than the NTS (19). Furthermore, we could not determine from our study whether the neurons whose activity was augmented by Glu received inputs only from the arterial baroreceptors. It is possible that they received inputs from chemoreceptors and/or other visceral receptors (29). Further studies are needed to clarify the role of ET-1 on the neurons of the NTS that synapse with afferent fibers from arterial baroreceptors.

In summary, our results suggest that ET-1 increases the neuronal activity of the mNTS via ET_A receptors on the postsynaptic membrane while endogenous ET helps control the basal spontaneous neuronal activity of the mNTS. In addition, ET-1 also facilitates synaptic (glutamatergic) transmission in the NTS. Together with the results of other studies in vivo, our results therefore suggest that ET-1 within the NTS may play an important role in the autonomic control of the cardiovascular system.

Perspectives