INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE NUCLEUS OF THE SOLITARY TRACT (NTS) is the initial synaptic integration site for baroreceptor and arterial chemoreceptor afferent inputs (5, 9, 39). Baroreceptor afferent activation elicits depressor responses via a sympathoinhibitory pathway from the dorsomedial NTS that involves activation of GABAergic neurons (1, 17, 18, 20, 40) in the caudal ventrolateral medulla (CVLM) (2, 12, 22) that inhibit vasoconstrictor sympathetic premotor neurons in the rostral ventrolateral medulla (RVLM). Stimulation of arterial chemoreceptors evokes pressor responses via a sympathoexcitatory pathway from the commissural NTS (commNTS) that excites vasoconstrictor sympathetic premotor neurons in the RVLM either directly (21) or through brain stem pathways that involve neurons in the A5 region (13). Although their relationship to the chemoreceptor reflex is not known, the presence of sympathoexcitatory neurons in commNTS has been suggested by: 1) the increase in arterial pressure (AP) elicited by microinjection of glutamate into commNTS (8, 26, 31) and 2) the ability of inhibition of commNTS neurons to block both afferent stimulation-evoked increases in AP (24) and those evoked by inhibition of neurons in the CVLM (30).

Although electrolytic lesions in commNTS abolish the pressor response evoked by stimulation of the arterial chemoreceptor reflex with potassium cyanide in both normotensive rats and spontaneously hypertensive rats (SHR) (6, 37), it was only in SHR that such lesions reduced the level of basal AP. These data raise the possibility that a tonically elevated level of discharge of commNTS neurons in SHR contributes to the maintenance of hypertension in this model. In the present study, we sought to provide further support for this hypothesis, by determining 1) whether the depressor effects of commNTS lesions are due specifically to inactivation of neurons rather than en passant fibers in commNTS and 2) whether a reduction in vasoconstrictor sympathetic nerve activity (SNA) accompanies the fall in AP produced by interruption of activity in commNTS neurons. A preliminary report of these results has appeared (36).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experiments were performed on adult male SHR (14-16 wk old, 300-350 g) and Wistar-Kyoto (WKY) rats (14-16 wk old, 350-400 g) obtained from Taconic Farms and on Sprague-Dawley (SD) rats (350-400 g) supplied by Charles River. The latter group of animals was selected to include a normotensive strain that was genetically unrelated to the SHR. We reasoned that differences directly related to the hypertension in the SHR should also be demonstrable between SHR and normotensive strains other than the WKY rat. Animals were initially anesthetized with 3% isoflurane in 100% O₂, the femoral artery and vein were cannulated for AP measurement and drug administration, respectively, and anesthesia was maintained with -chloralose (60 mg/kg iv) alone. Animals were unresponsive to noxious toe pinch and maintained a steady level of AP. After 3 h (before paralysis and ~1.5 h before microinjection of GABA), a supplementary dose (20 mg/kg) of -chloralose was administered. Heart rate (HR) was derived from the AP signal. The trachea was cannulated, and animals were artificially ventilated with 100% O₂. A unilateral pneumothorax was performed, and the animals were paralyzed with d-tubocurarine (0.8 mg/kg iv). End-tidal CO₂ was maintained between 3 and 4.5% with adjustments in the minute ventilation. The colonic temperature was maintained at 37°C with a thermostatically controlled heating table and heat lamp. Animals were placed in a stereotaxic apparatus (incisor bar: 11 mm below the interaural line) in prone position. A partial occipital craniotomy was performed to expose the dorsal surface of the caudal brain stem.

Microinjections (36 nl, delivered over 5 s) of GABA (50 mM) into the commNTS were performed with glass pipettes (20 µm tip diameter) coupled to a pressure injection apparatus (PicoSpritzer II). The volume of each injection was estimated from the displacement of the fluid meniscus in the pipette using a calibrated reticule. Microinjections into commNTS were made on the midline, 0.5 mm caudal to the calamus scriptorius, and 0.3 to 0.5 mm below the dorsal surface of the brain stem.

The left postganglionic splanchnic sympathetic nerve was dissected using a dorsolateral approach. The nerve was cut distal to the suprarenal ganglion, and the central end was placed on a bipolar hook recording electrode, using a monopolar configuration. The SNA was filtered (bandpass: 10-300 Hz) and amplified (50K, Axon Instruments Cyber Amp 3800). Signals for AP, SNA, mixed expired CO₂, and HR were digitized at 1 kHz and recorded on the computer hard disk. The SNA signals were demeaned (mean value of all the points was calculated and subtracted from each point, producing an identical waveform, but with a mean value of zero), rectified, and integrated in 10-s epochs (DataPac 2000, Run Technology). The splanchnic SNA response (percentage of control) to inhibition of neurons in the commNTS was obtained by dividing the average integrated SNA during the 30-s period (i.e., the mean of three 10-s epochs) of maximal response to microinjections of GABA into commNTS by the average integrated SNA during the 30-s period immediately before the microinjection. Results are presented as means ± SE. Following logarithmic transformation of the average integrated values of SNA under each condition, ANOVA followed by Scheffé's post hoc test was used to assess the significance (P < 0.05) of 1) the differences among strains in the levels of SNA, mean AP (MAP), and HR that were observed both before and after application of GABA to the commNTS; 2) the changes in SNA, MAP, and HR within each strain produced by GABA application to the commNTS; and 3) the differences among strains in the amplitude of the changes in SNA, MAP, and HR produced by microinjection of GABA into commNTS.

At the end of each experiment, the site of microinjection in the commNTS was marked by an iontophoretic deposit of 2% fast green dye and the small lesion that accompanied dye deposition (see Fig. 3A). The animal was perfused transcardially with 10% formalin, and the brain stem was removed and sectioned coronally at 80 µm. Sections containing dye spots were examined, and the locations of the microinjection sites were plotted on drawings (see Fig. 3B) from a rat stereotaxic atlas (34).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Responses to microinjection of GABA into the commNTS. Figure 1 illustrates the effects of GABA microinjection into the commNTS in a SHR and a WKY rat. Within 10 s after the microinjection of GABA (1.8 nmol in 36 nl) into the commNTS in the SHR, splanchnic SNA began to decline and this was followed by decreases in AP and HR (Fig. 1A). Cardiovascular parameters returned to control levels within 2-3 min, likely due in part to local uptake or metabolism of GABA (38). In contrast, a similar microinjection of GABA in the WKY rat produced only a small and delayed reduction in SNA, no change in AP, and a modest bradycardia (Fig. 1B). In all six SHR, inhibition of neurons in the commNTS with microinjections of GABA reduced the splanchnic SNA (P < 0.001; Figs. 1 and 2), MAP (P < 0.001), and HR (P < 0.001). In six WKY rats, microinjection of GABA into commNTS produced a fall in HR (Figs. 1 and 2), but no change in SNA or MAP. Similar microinjections in four SD rats had no effect on splanchnic SNA, MAP, or HR.

View larger version (41K): [in this window] [in a new window]   Fig. 1.   Responses of a spontaneously hypertensive rat (SHR) and a Wistar-Kyoto (WKY) rat to microinjection of GABA into the commissural nucleus of the solitary tract (NTS) (GABA commNTS). A: oscillographic traces recorded from an SHR. Top, splanchnic sympathetic nerve activity (SPL SNA); bottom, integrated splanchnic nerve activity (int SPL SNA, %control), arterial pressure (AP, mmHg), and heart rate (HR, beats/min) during microinjection of GABA (arrow) into commNTS. B: same traces as in A, recorded from a WKY rat. Vertical calibration is 5 µV for the SPL SNA traces, horizontal calibration is 60 s.

View larger version (12K): [in this window] [in a new window]   Fig. 2.   Changes in cardiovascular parameters elicited by inhibition of neurons in commNTS. A: mean percent changes from control splanchnic sympathetic nerve activity (%SNA) in SHR (open bar), WKY (solid bar), and SD (shaded bar) during maximum effect of GABA microinjection into commNTS. B: mean changes in mean arterial pressure (MAP) under same conditions as in A. C: mean changes in HR (HR) under same conditions as in A. Histogram values are means ± SE. *Significantly (P < 0.01) different from WKY and SD rats; **significantly (P < 0.05) different from WKY and SD rats.

Localization of GABA microinjection sites. The sites of fast green dye deposits marking the GABA microinjection sites in the three strains of animals are illustrated in Fig. 3. In all cases, the fast green dye deposits were located in the commNTS (34). This site corresponds to that in which electrolytic lesions reduced MAP in conscious SHR (37).

View larger version (49K): [in this window] [in a new window]   Fig. 3.   Location of GABA microinjection sites. A: histological section illustrating a typical microinjection site in the commNTS. A small lesion (arrow tip) was produced by the current applied to eject the fast green dye. B: positions of fast green dye deposits (solid triangles) plotted on a drawing (bregma: 14.6 mm) from the atlas of Paxinos and Watson (34) at the level of the commNTS. Top trace, dye deposits in SHR; middle trace, dye deposits in WKY; bottom trace, dye deposits in Sprague-Dawley (SD) rats. CC, central canal; X, dorsal motor nucleus of vagus; XII, hypoglossal nucleus; sol, solitary tract; Cu, cuneate nucleus; Gr, gracile nucleus.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The major finding of this study is that inhibition of neurons in the commNTS reduced splanchnic SNA and MAP in SHR to a significantly greater degree than in normotensive WKY or SD rats. These results comprise the first evidence for a potential involvement of neurons in commNTS in the maintenance of elevated blood pressure and SNA in SHR.

In this study, we observed that inhibition of commNTS neurons had no effect on the splanchnic SNA or MAP in WKY and SD rats. The similarity between these results and those in our previous studies, in which electrolytic lesions of the commNTS produced no change in MAP in conscious, normotensive SD, or Wistar rats (6, 37) suggests that anesthesia is unlikely to account for the absence of an effect of GABA microinjection in the commNTS of normotensive animals in the present study. Similarly, excitation of commNTS neurons has been shown to increase MAP in both unanesthetized and urethane-anesthetized rats (8). As with all such microinjection studies, we cannot completely discount the possibility that GABA may have influenced neurons in the vicinity of commNTS to produce the effects observed in this study. From a previous study (11), we expect that an effective concentration of GABA would not have spread more than 400 µm from the injection site, allowing us to conclude that the effects noted in earlier lesion studies are, in fact, due to reduced activity of local neurons in commNTS.

The dramatic contrast between the amplitudes of the depressor responses of SHRs and normotensive animals to inhibition of commNTS neurons supports the hypothesis that sympathoexcitatory neurons in commNTS have an increased level of tonic activity in SHR, but little or no basal activity in the normotensive strains and that such NTS neurons may contribute to the elevated level of vasoconstrictor SNA seen in this and previous comparisons of SHR and WKY (19, 25, 28, 29). Although the greater depressor response in the SHR could be accounted for by a markedly reduced sensitivity to the inhibitory effects of GABA in the commNTS neurons in the normotensive strains, the absence of a fall in MAP in the normotensive strains after electrolytic lesions of commNTS (6, 37) would make this possibility seem less likely. Whether the commNTS neurons contributing to the support of elevated SNA and MAP in SHR are those that mediate the sympathetic activation during peripheral chemoreceptor activation (4, 13, 21, 35) remains to be determined. It is of interest, however, that altered synchronization between respiratory and sympathetic outflows during chemoreceptor reflexes have been described in the SHR (7, 10), although this alteration does not appear to be expressed as a difference in the amplitudes of the chemoreceptor reflex-evoked pressor responses between SHR and WKY rats (14, 37).

Ito et al. (15) have demonstrated an exaggerated, kynurenic acid-sensitive excitatory input to sympathetic, vasomotor neurons in the RVLM of SHR, although the source of this input has not been identified. Similar to our microinjections of GABA into commNTS in the current study, microinjection of kynurenic acid into the RVLM of normotensive rats had little effect on MAP. These results led Ito et al. (15) to propose that an imbalance of inhibitory and glutamate-mediated, excitatory influences on RVLM vasomotor neurons contributes to the elevated MAP in SHR. Neuroanatomical (32, 33, 42) and electrophysiological (21) evidence supports a direct projection from the commNTS to the RVLM. The commNTS is the primary termination site of peripheral chemoreceptor afferents (5, 9), and commNTS neurons are necessary for the sympathoexcitation and pressor responses evoked by peripheral hypoxia (13, 35). The latter responses, in turn, are mediated by activation of excitatory amino acid receptors in the RVLM (3, 22, 27). An increased sensitivity of arterial chemoreceptors to hypoxia in SHR (10), an altered sympathetic response to hypoxia in SHR (7), and enhanced depressor responses to hyperoxia in hypertensive subjects (16, 41) have suggested an involvement of the arterial chemoreceptor reflex in hypertension. In contrast, activation of peripheral chemoreceptors elicits similar pressor responses in SHR and WKY rats (14, 37). Taken together, these findings are consistent with a model in which neurons in the commNTS contribute, through a glutamate-mediated input, to an enhanced excitation of RVLM vasomotor neurons in SHR. The potential interaction of the peripheral chemoreceptor reflex with such NTS sympathoexcitatory neurons remains to be determined.

In the present study, HR also decreased after microinjection of GABA into the commNTS in SHR and WKY, but not in SD rats. If this result is indicative of a reduction in cardiac SNA, it would suggest that the commNTS neurons in the SHR and WKY play a role in maintaining cardiac sympathetic outflow as well as that to vasoconstrictor targets (i.e., that represented by the splanchnic SNA).

In summary, our present data show that tonically active neurons in commNTS contribute to the maintenance of elevated SNA and high blood pressure in SHR and that if such neurons are present in normotensive WKY and SD strains, their level of tonic discharge is likely to be markedly lower than in SHR. These results implicate a population of sympathoexcitatory neurons in commNTS in the hypertension in the SHR.