clonidine and rilmenidine suppress hypotension-induced fos expression in the lower braistem of the...

~ Pergamon 0306-4522(94)00583-4

Neuroscience Vol. 66, No. 2, pp. 391-402, 1995 Elsevier Science Ltd

Copyright © 1995 IBRO Printed in Great Britain. All rights reserved

0306-4522/95 $9.50 + 0.00

CLONIDINE AND RILMENIDINE SUPPRESS HYPOTENSION-INDUCED FOS EXPRESSION IN THE LOWER BRAINSTEM OF THE CONSCIOUS RABBIT

Y.-W. LI* and R. A. L. D A M P N E Y t

Department of Physiology, University of Sydney, NSW 2006, Australia

Almtract---Our current knowledge of the sites of action of the centrally-acting antihypertensive drug clonidine is based almost entirely on experiments in anesthetized animals. The aim of this study was to determine, in conscious rabbits, the sites of action in the brainstem of systemically administered clonidine, as well as its oxazoline analog rilmenidine. Three groups of experiments were carried out. In the first group, hypotension was produced by continuous intravenous infusion of sodium nitroprusside, at a rate sufficient to decrease arterial pressure by 20-30 mmHg, maintained for a period of 60 min. In the second and third groups of experiments, sustained hyp0tension was also produced by nitroprusside infusion as in the first group, but this was preceded by intravenous injection of clonidine (7-30/lg/kg i.v.) or rilmenidine (150-300#g/kg i.v.), respectively. In confirmation of our previous study [Li Y.-W. and Dampney R. A. L. (1994) Neuroscience 61, 613-634], hypotension produced by nitroprusside alone induced a large increase (compared to sham control experiments) in the neuronal expression of Fos (a marker of neuronal activation) in the nucleus of the solitary tract, area postrema, the rostral, intermediate and caudal parts of the ventrolateral medulla, A5 area, locus coeruleus and subcoeruleus, and parabrachial nucleus. In comparison with this group, in rabbits pretreated with clonidine the numbers of Fos-positive cells were greatly reduced (by 76 94%) in the rostral, intermediate and caudal parts of the ventrolateral medulla, area postrema, A5 area, locus coeruleus and subcoeruleus. Clonidine pretreatment also caused a more moderate reduction (by 45%) in the number of Fos-positive cells in the nucleus of the solitary tract, but had no effect on Fos expression in the parabrachial nucleus. Double-labeling for tyrosine hydroxylase and Fos immunoreactivity showed that clonidine pretreatment greatly reduced the numbers of both catecholamine and non-catecholamine Fos-positive neurons. Rilmenidine pretreatment also greatly reduced Fos expression in the lower brainstem, with a very similar pattern to that observed after clonidine pretreatment.

The results indicate that in conscious animals both clonidine and rilmenidine cause a widespread but selective inhibition of neurons in the pons and medulla that are normally activated by a hypotensive stimulus. In contrast to previous observations in anesthetized animals, the results suggest that (i) systemic administration of both drugs inhibits non-catecholamine as well as catecholamine neurons in the ventrolateral medulla, and (ii) the regional pattern of neuronal inhibition following administration of equipotent hypotensive doses of both drugs is very similar.

Clonidine and r i lmenidine are ant ihyper tensive agents tha t lower arterial b lood pressure by an act ion in the CNS. 34'36'59 Both drugs are agonists of ~2-

adrenocep tors and putat ive imidazoline receptors, bu t differ significantly in their pharmacologica l properties. Rilmenidine, which is an oxazoline analog of clonidine, has a greater selectivity for puta t ive imidazoline receptors over ~2-adrenoceptors as compared with clonidine. ~°'2~

Our current unde r s t and ing of the central sites of act ion of bo th clonidine and r i lmenidine is based a lmost entirely on experiments in anesthet ized ani-

*Present address: Department of Pharmacology, School of Medicine, University of Virginia, Charlottesville, VA 22908, U.S.A.

tTo whom correspondence should be addressed. Abbreviations: NHS, normal horse serum; NTS, nucleus of

the solitary tract; PBS, phosphate-buffered saline; SNP, sodium nitroprusside; TH, tyrosine hydroxylase; VLM, ventrolateral medulla.

mals. In part icular , it has been shown tha t one of the ma in sites of the hypotensive act ion of systemically adminis tered clonidine and r i lmenidine is the rostral par t of the ventrolateral medulla (VLM), 7'8'18'19"21'43

which conta ins a group of bulbospina l sympathoexci- ta tory neurons. ~5'22 The effects of bo th systemically adminis tered and locally applied clonidine on the firing rate of these neurons in anesthet ized animals have been studied in detail by Guyene t and co- workers. One of their mos t interest ing observat ions is tha t clonidine inhibits only a subset of rostral V L M sympathoexci ta tory neurons, while the remainder are unaffected. 1'23'28"54 Fur the rmore , the clonidine-sensi-

tive and clonidine-insensit ive neurons differ in their electrophysiological propert ies, as well as in the pattern of their connect ions with the spinal cord and other b ra ins tem nuclei. 28 As reviewed by Guyene t e t al. , 23 a compar i son of these propert ies with those of identified ca techolamine and non-ca techolamine neurons in the rostral V L M has led to the hypothesis

391

392 Y.-W. Li and R. A. L. Dampney

that the clonidine-sensitive sympathoexcitatory neurons are catecholamine (C1) cells, while the clonidine- insensitive neurons are non-catecholamine cells.

Both clonidine and rilmenidine also affect neurons in other regions of the lower brainstem. In particular, both drugs can inhibit the activity of cells within the locus coeruleus, 39,57 although there is some evidence that when equipotent hypotensive doses of the drugs are injected, cionidine has a relatively greater effect on locus coeruleus neurons as compared to rilmenidine? 7 In addition, clonidine has also been shown to inhibit putative catecholamine cells within the caudal ventrolateral medulla (A 1 cell group) 32 and caudal ventrolateral pons (A5 cell group), ll'3°

As mentioned, the studies referred to above were carried out in anesthetized animals. There is considerable evidence, however, that anesthesia can profoundly influence the cardiovascular effects of clonidine. For example, it has long been known that intracerebroventricular administration of clonidine 5~ in anesthetized animals decreases blood pressure and heart rate at doses that are ineffective when administered intravenously. In contrast, it has been shown more recently that in conscious normotensive rats intracerebroventricular administration of clonidine results in either little change or even an increase in arterial pressure, 6'33 while in conscious spontaneously hypertensive rats it has n o effect . 49 Similar observations have also been made in the case of rilmenidine. 49 Furthermore, there is direct evidence that anesthesia can profoundly affect the response of central neurons to systemically administered clonidine, as demonstrated by a recent study in which the activity of single neurons in the locus coeruleus of rats was recorded before and after a period of halothane anesthesia) ° In the presence of halothane, the dose of clonidine required to inhibit locus coeruleus neurons was much less than that required to produce the same effect when administered after withdrawal of halothane.

These previous observations have therefore led us to re-examine the question of the central sites of action of clonidine and rilmenidine when administered systemically in conscious animals. For this purpose, we have taken advantage of our recent observation 37 that in conscious rabbits a period of sustained peripherally-induced hypotension induces the expression of the proto-oncogene c-fos in neurons within several regions in the lower brainstem, including all of those referred to above (rostral VLM, locus coeruleus, AI and A5 areas), which have been shown to be sites of action of clonidine in anesthetized animals. The expression of c-fos, which is believed to be a marker of neuronal activation, can be detected by the immunohistochemical labeling of Fos, the protein product of c-fos. 17

The general strategy in this study, then, was to determine whether pretreatment of conscious rabbits with either clonidine or rilmenidine alters the pattern of Fos expression in pontomedullary cell groups that

is normally induced by a period of peripherally- induced sustained hypotension, and if so, whether the effects of clonidine and rilmenidine differ in this respect. Secondly, since there is evidence, as described above, that clonidine selectively affects catecholamine neurons, we have combined Fos labeling with immunohistochemical labeling for the catecholamine- synthesizing enzyme tyrosine hydroxylase (TH), to determine the extent to which clonidine (and also rilmenidine) suppresses Fos expression specifically in catecholamine neurons.

EXPERIMENTAL PROCEDURES

General procedures Experiments were performed on 22 New Zealand White

rabbits University of Sydney Laboratory Animal Services, (2.5-3.5 kg) of both sexes. The general procedures have been described in detail in a previous paper) 7 Briefly, a marginal ear vein was cannulated and the rabbits were anesthetized with a mixture of alphaxalone and alphadolone (Saffan, Pitman-Moore, 10mg total steroid/kg i.v.). A central ear artery was then exposed and cannulated under local anesthesia (2% xylocaine). This procedure was completed within 15 min, after which the rabbits were allowed to recover from the anesthesia. Alphaxalone and aiphadolone are rapidly inactivated by the liver, and have a plasma half-life of approximately 7 min. 26 The rabbits were conscious within 30 min following the catheter implantation, but were then allowed to rest for 3-5 h to allow full recovery from the effects of the anesthetic. The rabbits were then moved into a padded box, and placed in a comfortable position. The ear vein catheter was connected to an infusion pump for administration of drug solution and the ear artery catheter connected to a transducer, which was linked to a polygraph chart recorder for monitoring arterial pressure and heart rate. During both the experimental and preceding resting periods, room temperature and light were maintained close to that in the animal house.

There were four experimental groups. In the first group (n = 5), the mean arterial pressure was reduced by about 25 mmHg with a continuous intravenous infusion of sodium nitroprusside (SNP; 0.2% dissolved in Ringer solution). The hypotension was maintained at a steady-state level for 60min by manually adjusting the infusion rate (range 2-4 ml/h) while observing the mean arterial pressure. In this way, the mean arterial pressure was maintained within 5 mmHg of the desired level, apart from occasional spon- taneous small changes in pressure that lasted for a few seconds. We have previously shown that the duration and magnitude of the hypotension as used in these experiments is optimal for eliciting Fos expression in pontomedullary cell groups) 7

In the second (n = 5) and third (n = 5) groups, rabbits were first pretreated with an intravenous injection of clonidine (7-30 pg/kg; Sigma) or rilmenidine (150-300 p g/kg; Servier, France). The agents were dissolved in 2 ml Ringer solution and injected slowly over 5 min. Although injection of these agents resulted in a small decrease in arterial pressure, it was necessary to infuse SNP continuously to achieve a sustained reduction in mean arterial pressure of the same magnitude and for the same period (60 min) as in the first group of experiments. The dose ranges of clonidine and rilmenidine were selected on the basis of the known dose-response relationship for these drugs in the conscious rabbit. 29,56 For both drugs, the highest dose used was that which produces a near-maximal hypotensive response, while the lowest dose produces a medium hypotensive effect.

Rabbits in the fourth group served as controls. Five

Antihypertensive drugs and Fos expression in brainstem 393

rabbits were infused with the vehicle Ringer solution (5 ml/h) for 60 min. The results of these control experiments were reported in our previous study, 37 and are presented here again for comparison with the results of the first three groups of experiments. In two other experiments, rabbits were intravenously injected with clonidine (30 ttg/kg) and subsequently infused with Ringer solution (5 ml/h) for 60 min.

In all experiments, after the period of infusion there was a waiting period of 1 h and then the rabbits were deeply anesthetized with sodium pentobarbitone (50-60 mg/kg i.v.) and perfused transcardially with 500 ml 0.01 M phosphate- buffered saline (PBS; pH 7.3), followed by 1.51 of 4% paraformaldehyde in PBS (0.1 M, pH 7.3). The brainstem was then removed, blocked and postfixed for one to two days in 20% sucrose in PBS. Five series of coronal sections (40 #m) were cut on a freezing microtome, after which they were washed in 50% ethanol for l h and stored in PBS containing 0.1% azide at 4°C.

Single staining for Fos immunoreactivity

One series of free-floating sections from each rabbit was processed for Fos immunoreactivity using an avidin biotin-peroxidase procedure. Sections from the different experimental groups were always processed according to the same protocol. After incubating in 20% normal horse serum in PBS (NHS-PBS) for 1 h, sections were incubated overnight at 4°C in polyclonal sheep anti-Fos antibody (Cambridge Research Biochemicals, OA-11-823) diluted 1:4000 in 1% NHS-PBS. After washes in 1% NHS-PBS, sections were subsequently incubated in biotinylated anti- sheep antibody (Vector Labs; 1:200 dilution in 1% NHS-PBS) for 1 h at room temperature and ExtrAvidi~ peroxidase conjugate (Sigma; 1 : 400 dilution 1% NHS-PBS) for 1 h at room temperature. After washes in PBS, sections were reacted for 20 min in a PBS solution containing 0.05% diaminobenzidine hydrochloride, 0.1% nickel ammonium sulfate and 0.01% hydrogen peroxide. Sections were then mounted onto slides subbed with gelatin-chrome, dried, dehydrated, defatted and coverslipped with DPX. As a control, some sections were incubated without Fos antibody or with normal sheep serum instead of Fos antibody. In these sections, no specific immunoreactivity was observed.

Double staining for Fos and tyrosine hydroxylase immunoreactivity

One series of sections was processed for double immunoperoxidase staining for Fos and TH immunoreactivity. First, immunoperoxidase staining for Fos immunoreactivity was performed using the same procedure as described above. The sections were then incubated in mouse anti-TH antiserum (Incstar; 1:4000 dilution) overnight at 4°C. After washes in 1% NHS-PBS, the sections were incubated in biotinylated anti-mouse antibody (Vector Labs; 1:200 dilution) for 1 h at room temperature. The sections were then rinsed in 1% NHS-PBS and incubated in ExtrAvidin- peroxidase conjugate (1:400 dilution) for l h at room

temperature. After washes in PBS, the sections were then incubated for 10-20min in a PBS solution containing 0.05% diaminobenzidine and 0.01% hydrogen peroxide in PBS. Consequently, Fos-positive cell nuclei were stained black and the cytoplasm of TH-positive cells was stained brown. Sections were mounted onto slides subbed with gelatin-chrome, dried, dehydrated, defatted and coverslipped with DPX.

Determination of distribution and number of immunoreactive cell nuclei

Sections processed according to the immunoperoxidase procedure were examined with bright-field optics using a BH2 Olympus microscope at x 200. Fos immunoreactivity appeared as black staining of variable intensity within the nucleus. Fos-positive neurons were identified as those in which the nucleus was clearly stained in comparison with the unstained nucleolus, while TH-positive cells were identified as those in which the cytoplasm was clearly stained brown. The locations of Fos- and TH-positive neurons identified in this way were mapped, and the number of such cells counted, with the aid of the Magellan Image Analysis Program 27 and a Macintosh Ilcx computer.

The number of Fos- and TH-positive cells in a defined region was counted bilaterally for each series of sections. In each animal, the immunoreactive cells in all sections that contained the defined region were counted, for each side of the brain. The sections in each series were 200 #m apart, so that the number of sections that contained the defined region varied according to the rostrocaudal extent of the region examined, but was never less than five for each side. Cells in the region on each side were counted independently, so that any asymmetry in cutting the sections did not affect the accuracy of cell counts. The mean number of immunoreactive cells per section in this region per side was calculated for each animal. Mean values and standard errors for each experimental group of animals were then calculated. Anatomical correlations were made by reference to the atlas of Meesen and OlszewskiJ °

Statistical analysis

The numbers of labeled cells within each region between the different experimental groups (SNP, SNP + clonidine, SNP + rilmenidine) were compared using the non-paramet- ric Kruskal-Wallis test. Pair-wise comparisons between mean values were made using the Mann-Whitney test, with application of the Bonferroni procedure for multiple comparisons. In all cases a P-value <0.05 was taken as the level of statistical significance.

RESULTS

Control experiments

No significant changes in blood pressure and heart

rate were observed in the group of rabbits in which Ringer solution only was infused intravenously (Table 1). Immunohis tochemical staining for Fos

Table 1. Changes in mean arterial pressure and heart rate in rabbits infused intravenously with Ringer solution alone, sodium nitroprusside alone, sodium nitroprusside after pretreatment with clonidine and sodium nitroprusside after pretreatment with

rilmenidine

MAP (mmHg) HR (beats/min)

n Baseline Change Baseline Change

Ringer solution 5 87 _ 3 + 4 + 1 260 ___ 12 + 5 + 3 SNP 5 84 ___ 5 - 2 4 + 2* 285 _ 21 +66 + 5* SNP + clonidine 5 8 6 + 2 - 3 0 + 3 * 235+__11 +1 5 +1 2 SNP + rilmenidine 5 84_+2 - 2 5 + 2 * 251+3 +44+14"

Values are mean ± S.E. *P < 0.05. HR, heart rate; MAP, mean arterial pressure.


Table 2. Numbers of Fos-positive neurons per section in different brain regions in rabbits infused with Ringer solution alone, sodium nitroprusside alone, sodium nitroprusside after pretreatment with clonidine and sodium nitroprusside after

pretreatment with rilmenidine

Ringer SNP SNP SNP solution alone + clonidine + rilmenidine (n = 5) (n = 5) (n = 5) (n = 5)

NTS 7 + 1 47 _+ 5 26 _+ 4* 24 _+ 5* AP 4_+ 1 34_+7 9_+3* 13_+2" CVLM 1 _+0.4 19+_ I 2_+ 1" 2_+ 1" IVLM 3 _+ 1 30 -+ 4 6 _+ 2* 4 _+ 1 * RVLM 2 _+ 0.2 38 _+ 6 9 _+ 2* 8 _+ 2* A5 area 0.2_+0.2 17-+2 1 _+ 1" 1 _+ 1" LC&Subc 2+1 52-+7 3-+1" 3_+1" PBN I1 +2 60-+ 5 67+ 11 58_+5

Values are mean _+ S.E. *Statistically significant difference (P < 0.05) compared with the SNP alone group. AP, area postrema; CVLM, caudal ventrolateral medulla; IVLM, intermediate ventrolateral medulla; LC & Subc, locus coeruleus and subcoeruleus area; NTS, nucleus tractus solitarii; PBN, parabrachial nucleus; RVLM, rostral ventrolateral medulla.

showed that there were only a few lightly-stained Fos-positive cells in several brainstem regions, including the area postrema, the nucleus tractus solitarii (NTS), the VLM and the parabrachial nucleus (Table 2).

In the two experiments in which clonidine (30 pg/kg i.v.) was injected and followed by intravenous infusion of Ringer solution, there was a small fall in mean arterial pressure (by 5-10 mmHg) and a decrease in heart rate (by 40-60 beats/min). The depressor response began within 3-5 min following clonidine injection, and lasted for more than 1 h. In these animals, there were only a few scattered Fos- positive cells in the NTS, the VLM and the parabrachial nucleus, similar to the number and pattern observed in the control experiments in which Ringer solution only was infused.

Effects of hypotension on Fos expression

In the group, of rabbits in which mean arterial pressure was reduced by approximately 25 mmHg for a period of 60 min by intravenous infusion of SNP, there was an increase in heart rate (Table 1), presumably as a consequence of baroreceptor unloading. In the lower brainstem, there was a very marked increase in the number of Fos-positive cells in several regions (Table 2, Fig. 1), in comparison to the control group, in confirmation of our previous report. 37 In the VLM, there were increased numbers of Fos-positive cells in its rostral part (defined as the region between levels 1.8 and 3.0 mm rostral to the obex), intermediate part (between the level of the obex and 1.8 mm rostral to the obex) and caudal part (levels caudal to the obex) (Figs 1C, D, 2A). In the dorso- medial medulla, numerous Fos-positive cells were located in the NTS and the area postrema (Fig. 1D). Fos-positive cells in the NTS were distributed throughout the nucleus, although the majority were located in the region caudal or just rostral to the obex. A few scattered Fos-positive cells were also

located ventrolateral to the NTS, and in the dorsal motor nucleus of the vagus.

In the lower pons, many Fos-positive cells were located in the locus coeruleus, the subcoeruleus area and the parabrachial nucleus (Table 2, Figs 1A, 2B, C). In the parabrachial nucleus, the Fos-positive cells were typically concentrated in the subregion just lateral to the ventral part of the superior cerebellar peduncle (Figs 1A, 2C). There was also a significant number of Fos-positive cells in the ventrolateral pons, in the region just lateral to the superior olive and facial nucleus, which corresponds to the location of the A5 catecholamine cell group (Fig. 1B).

In confirmation of our previous study, 37 double labeling for Fos and TH immunoreactivity revealed that approximately two-thirds of the Fos-positive cells in the rostral VLM, where the C1 cell group is located, and in the intermediate and caudal VLM, where the A1 cell group is located, 4 were also immunoreactive for TH (Table 3). In the locus coeruleus, subcoeruleus area and A5 area in the pons, many cells were immunoreactive for both Fos and TH (Fig. 2D). Specifically, the mean number of double-labeled cells was 70-80% of the mean total number of Fos-positive cells, and also of the mean total number of TH-positive cells (Table 3).

Effects of pretreatment with clonidine or rilrnenidine on Fos expression evoked by hypotension

In the rabbits in which pretreatment with either clonidine (7-30/~g/kg) or rilmenidine (150-300 #g/kg) was followed by infusion of SNP, blood pressure was maintained at a similar level to that seen in the rabbits in which only SNP was infused (Table 1). During this period, the mean heart rate in both groups was increased compared with the baseline control level (Table 1), but the increases were variable and, in the case of the clonidine- treated group, was not statistically significant. The variable effects on heart rate presumably reflect the


competition between the bradycardic effects of clonidine and rilmenidine, 29 and the tachycardia reflexly elicited by the SNP-induced hypotension.

In rabbits pretreated with cionidine (7-30/tg/kg), there was a very marked reduction in Fos expression in many lower brainstem regions, compared with the experiments in which only SNP was infused (Table 2). In every experiment, the numbers of Fos-positive cells in the rostral, intermediate and caudal parts of the VLM and in the area postrema were reduced by at least 55% compared with the mean number of Fos- positive cells in these regions in experiments in which only SNP was infused (Figs 1C, D, 2A, E). In the A5

area, locus coeruleus and subcoeruleus area, the reduction in Fos expression was even greater: in every experiment the number of Fos-positive cells in each of these areas was reduced by at least 88% compared with the experiments in which only SNP was infused (Fig. 1A, B). In the NTS, there was a more moderate reduction (mean 45%) in the number of Fos-positive cells in the clonidine-pretreated group (Table 2, Fig. 1D). In all these regions, the differences in the mean number of Fos-positive cells in the clonidine- pretreated animals compared with the animals in which SNP only was infused were statistically significant (Table 2). In contrast, clonidine pretreatment

SNP+ SNP+ SN P clonidine rilmenidine

A

B

C

D

Fig. 1. Drawings of pontomedullary sections at the level of the locus coeruleus (A), facial nucleus (B), and rostral (C) and caudal (D) VLM, showing examples of the distribution of Fos-positive neurons in three different experiments in which the rabbits were infused intravenously with SNP alone (left column), SNP following pretreatment with clonidine (10/~g/kg i.v., middle column) and SNP following pretreatment with rilmenidine (150/~ g/kg i.v., right column). Each filled square represents one Fos-positive neuron per section. Vsp, spinal trigeminal nucleus; VII, facial nucleus; XII, hypoglossal nucleus; AP, area postrema; IO, inferior olive; LC, locus coeruleus; LRN, lateral reticular nucleus; NA, nucleus ambiguus; NTS, nucleus of the solitary tract; RFN, retrofacial nucleus; SCP, superior cerebellar peduncle; SubC,

subcoeruleus.


t

Fig. 2.


Table 3. Numbers of neurons per section immunoreactive for tyrosine hydroxylase, Fos, and both tyrosine hydroxylase and Fos in different regions in the different experimental groups

SNP alone SNP + clonidine SNP + rilmenidine (n = 4) (n = 4) (n = 4)

Region TH Fos TH/Fos TH Fos TH/Fos TH Fos TH/Fos

RVLM 34+4 43+4 28+3 37+7 5+1 3-t-1" 38__+3 10+5 5+3* CVLM/IVLM 30___2 31+5 21+2 31+3 2+1 I + I * 30__+5 4+3 2 + 1 ' NTS 46+5 50-t-13 15-t-3 55+5 22+4 5__+1" 45+4 25+1 4 + 1 ' A5 area 17__+I 16+3 12+2 18__+4 0__+0 0+0" 17+2 0+ 0 0 + 0 ' LC & Subc 51-t-2 44-t-5 36___4 52+2 1+1 1__1" 44+5 3+5 1+1"

Values are mean + S.E. *Statistically significant difference (P < 0.05) compared with the SNP alone group. CVLM, caudal ventrolateral medulla; IVLM, intermediate ventrolateral medulla; LC & Subc, locus coeruleus and subcoeruleus area; NTS, nucleus tractus solitarii; RVLM, rostral ventrolateral medulla.

had no effect on the number of Fos-positive cells in the parabrachial nucleus (Table 2, Figs IA, 2C, G).

Pretreatment of rabbits with rilmenidine (150-300#g/kg) also in every case substantially attenuated Fos expression in the lower brainstem regions, with a very similar pattern to that observed following clonidine pretreatment (Table 2, Fig. 1). Compared with the experiments in which only SNP was infused, the numbers of Fos-positive cells in the VLM, area postrema, the A5 area, locus coeruleus and subcoeruleus area were greatly reduced (by 62-94%), while in the NTS there was a more moderate reduction of 49% (Table 2, Fig. 1). These differences were statistically significant (Table 2). As in the clonidine-pretreated group, rilmenidine pretreatment had no effect on Fos expression in the parabrachial nucleus (Table 2, Fig. 1A). There was no significant difference (P > 0.5 in all cases) in the numbers of Fos-positive cells in all these regions between the clonidine- and rilmenidine-treated groups.

Double-labeling for Fos and TH immunoreactivity revealed that clonidine pretreatment greatly reduced the numbers of cells in the catecholamine-synthesizing cell groups that expressed Fos following induced hypotension (Table 3). Following clonidine pretreatment, only 8% of TH-positive cells in the rostral VLM (C1 area) and 2% of TH-positive cells in the intermediate and caudal VLM (A1 area) were Fos- positive compared with 84% and 74%, respectively, in the group in which SNP only was infused. In the NTS, 5% of TH-positive cells were also Fos-positive, compared with 33% in the group in which SNP only was infused. In the A5 area and locus coeruleus/subcoeruleus, virtually no TH-positive cells were Fos-positive following clonidine pretreatment.

Rilmenidine pretreatment also greatly reduced the numbers of cells in the catecholamine-synthesizing cell groups that expressed Fos following induced

hypotension (Table 3). Moreover, the extent of the reduction was almost identical to that observed in the clonidine-pretreated group. In the rostral, intermediate and caudal VLM and NTS, between 5 and 14% of TH-positive cells were also Fos-positive, while in the A5 area and locus coeruleus/subcoeruleus virtually no TH-positive cells expressed Fos following rilmenidine pretreatment.

The effects of pretreatment of both clonidine and rilmenidine on hypotension-induced Fos expression was not confined to TH-positive cells within the regions containing catecholamine cells. Although non-catecholamine cells (i.e. cells that were not immunoreactive for TH) accounted for only a small proportion of the cells in the VLM, NTS, A5 area and locus coeruleus/subcoeruleus that expressed Fos following induced hypotension, the numbers of such cells were also significantly reduced following pretreatment with either clonidine or rilmenidine (Table 4).

DISCUSSION

In a previous study, 37 we have demonstrated that induced hypotension in conscious rabbits results in Fos expression in specific nuclei within the brain, including all of the catecholamine cell groups in the pons and medulla. As we have discussed in detail previously, 37 neurons that express Fos in response to hypotension are likely to include those that are part of central pathways that reflexly increase sympathetic vasomotor activity, as well as those that convey signals from baroreceptor and other cardiovascular receptors to higher levels of the brain. In the present study, we have confirmed these previous observations and extended them by demonstrating for the first time that in conscious animals the antihypertensive agents clonidine and rilmenidine have reproducible

Fig. 2. Photomicrographs showing Fos-positive neurons in the rostral VLM (A, E), locus coeruleus (B, F) and lateral parabrachial nucleus (C, G). The photomicrographs in D and H show neurons stained with the immunoperoxidase procedure for Fos (dark-stained nucleus) and/or TH (lighter-stained cytoplasm) immunoreactivity in the locus coeruleus. The sections on the left side (A-D) are all taken from experiments in which the animals were infused with SNP alone, while those on the right side are taken from experiments in which SNP infusion was preceded by administration of either clonidine (10/~g/kg i.v.; E, G and H)

or rilmenidine (150 pg/kg i.v.; F). Scale bar = 200 pm (A-C, E43); 50 pm (D, H).


Table 4. Numbers of non-catecholamine Fos-positive neurons per section in different regions in the different experimental groups

Region

SNP SNP SNP alone + clonidine + rilmenidine

(n =4) (n=4) (n=4)

RVLM 15 + 4 2 __+ 1" 5 __+ 3* CVLM/IVLM 10 + 3 1 + I* 2 + 2* NTS 35+12 17-t-3" 15+4" A5area 4+2 0+0" 0+0" LC & SubC 8 + 2 1 __+ I* 2 ___ 1"

Values are mean ___ S.E. *Statistically significant difference (P < 0.05) compared with the SNP alone group. CVLM, caudal ventrolateral medulla; IVLM, intermediate ventrolateral medulla; LC & Subc, locus coeruleus and subcoeruleus area; NTS, nucleus tractus solitarii; RVLM, rostral ventrolateral medulla.

but non-uniform suppressive effects on Fos expression in pontomedullary cell groups. In particular, both clonidine and rilmenidine produced a very marked suppression of Fos expression in both catecholamine and non-catechotamine cells in the VLM, NTS, A5 area, locus coeruleus and subcoeruleus, but had no effect in the parabrachial nucleus. Fur- thermore, both drugs produced a virtually identical pattern of suppression of Fos expression.

Methodological considerations

A major advantage of using a change in Fos expression as an indicator of the effect of clonidine and rilmenidine in different neuronal cell groups is that the experiments could be carried out entirely in conscious animals. This is a particularly important consideration since, as described in the Introduction, the effects of clonidine and rilmenidine are profoundly influenced by anesthesia, and our current understanding of the central antihypertensive mechanisms of these compounds is based almost entirely on studies performed in anesthetized animals.

Previous studies using the method of in vivo voltammetry have shown that the metabolic activities of catecholamine neurons in the rostral VLM and locus coeruleus are reduced by clonidine or rilmenidine, 41,45'57'58 but such studies did not provide any information about the effects of these drugs on individual neurons in these regions and could not detect effects on non-catecholamine neurons. A second advantage of the Fos method, therefore, is that it has a cellular resolution and can be combined with immunohistochemical procedures so that the trans- mitter properties of individual neurons in these and other areas can be detected.

At the same time, the Fos method also has certain limitations which need to be taken into account when interpreting the results. In the present case, it is clear that effects can only be observed in cell groups that express Fos in response to the hypotensive challenge, so that any inhibitory effect of clonidine and/or rilmenidine outside these regions could not be detected. The hypotensive challenge does, however, induce Fos expression in all the pontomedullary cell groups that have been shown in electrophysiological

or electrochemical studies to contain neurons that are inhibited by systemic administration of clonidine or rilmenidine. Therefore, this method as used in the present study is capable of providing a picture of the sites of action of these drugs that is at least as complete as that provided by previous electrophysiological or electrochemical studies, as well as having the advantages referred to above.

As mentioned in the Experimental Procedures, the dose ranges of clonidine and rilmenidine used in the present study were based on the determination by Head and Burke 29 and Szabo et al. 56 of the dose-response relationship for these drugs in the conscious rabbit. For both drugs, the highest dose used was that which produces a near-maximal hypotensive response, while the lowest dose produces a medium hypotensive effect. Although we did not attempt to determine systematically the relationship between dose and degree of Fos expression, it was clear that with both drugs a similar degree of suppression of Fos expression occurred with all doses that were used.

Effects on Fos expression in different cell groups

As we have previously discussed in detail, 37 the hypotension-induced Fos expression in cells within the rostral VLM and in the catecholamine cells in the A1 and A5 groups can be explained as a consequence of activation of these cells in response to baroreceptor unloading. It is therefore possible that suppression of Fos expression in these regions by clonidine and rilmenidine is simply due to inhibition of transmission of baroreceptor signals through the NTS, so that baroreflex disinhibition of neurons within the rostral VLM and A1 and A5 cell groups does not occur. This seems highly unlikely, however, since systemically administered clonidine or rilmenidine does not reduce the gain of the baroreceptor reflex in either animals or humans. 3'42'53'6° Further, our results showed that neither clonidine nor rilmenidine significantly affected the degree of Fos expression in the lateral parabrachial nucleus, which receives baroreceptor signals transmitted via the NTS. 31

It therefore seems more likely that the reduction in Fos expression in the rostral VLM, A1 and A5 cell


groups following clonidine or rilmenidine pretreatment is due primarily to a direct effect of these drugs on neurons in these regions rather than to inhibition of transmission of baroreceptor signals through the NTS. In support of this, microinjection of clonidine or rilmenidine directly into the rostral VLM has been shown to produce the same effects on arterial pressure as intravenous injections of the drugs. 9,21,43 Simi- larly, both AI and A5 neurons are inhibited by direct application of clonidine. 3°'32 Nevertheless, our results do not rule out the possibility that the inhibitory effects of systemically administered clonidine or rilmenidine are due partly to an action of these drugs on neurons, located in other regions, that in turn may influence the activity of neurons in the rostral VLM, A1 and A5 regions.

Previous studies in anesthetized animals have used the method of single-unit recording or in vivo voltammetry to study the effects of clonidine and rilmenidine on the firing pattern of neurons in different pontomedullary cell groups. In the following sections our results will be discussed in relation to these previous findings, with particular attention to the rostral VLM and locus coeruleus, since most previous studies of the central sites of action of clonidine and rilmenidine have focused on these two regions.

Rostral ventrolateral medulla. In rabbits pretreated with either clonidine or rilmenidine, the total number of rostral VLM neurons that expressed Fos following SNP-induced hypotension was reduced by approximately 80%, compared with experiments in which there was no pretreatment. In particular, about 80% of the catecholamine cells in the rostral VLM normally express Fos following a period of hypotension, but this proportion was greatly reduced by both drugs. This finding is consistent with in vivo voltammetry studies, which have shown that the metabolic activity of catecholamine neurons in the rostral VLM is greatly reduced by clonidine and rilmenidine. 41'57'58 Similarly, single-unit recording studies have shown that the firing rate of a subset of rostral VLM neurons, that have electrophysiological and anatomical properties indicative of C 1 catecholamine cells, is also inhibited by both intravenous and iontophoretically applied clonidine. 1'28'54

In confirmation of our previous study, 37 approximately one-third of the cells in the rostral VLM that expressed Fos following SNP-induced hypotension were not immunoreactive for TH. The number of such cells that expressed Fos following induced hypotension was decreased in both clonidine- and rilmenidine-pretreated animals, indicating that these drugs also affect non-catecholamine cells in the rostral VLM. At first sight, this finding does not appear to be consistent with those of previous electrophysiological studies in anesthetized animals by Guyenet and co-workers. These workers found that barosensitive rostral VLM cells that were insensitive to intravenous clonidine 28'54 had electrophysiological and other properties indicative of fast-conducting "pace-

maker" neurons in the rostral VLM that are not immunoreactive for TH (for review see Guyenet 22). A more recent study, however, has shown that there is considerable overlap in the electrophysiological properties of clonidine-sensitive and -insensitive neurons in the rostral VLM, 1 making it difficult to distinguish between the two types of neurons according to electrophysiological criteria alone. Secondly, it is possible that clonidine has a more selective effect on catecholamine neurons in the presence of anesthesia. Consist- ent with this, halothane anesthesia has been shown to potentiate the response of locus coeruleus neurons to intravenous clonidine? °

Recently, Guyenet and co-workers 25 reported that, amongst the population of presympathetic neurons in the rostral VLM, ~2A-adrenoceptor-like immunoreactivity was found in the vast majority (>95%) of catecholamine neurons, but in only very few (< 5%) of the non-catecholamine neurons. Our observation that clonidine and rilmenidine suppressed Fos expression in both catecholamine and non-catecholamine putative sympathoexcitatory neurons in the rostral VLM therefore suggests that these drugs affect neurons in this region that lack ~:A-adrenoceptors. In particular, it is possible that these drugs may act on putative imidazoline receptors in the rostral VLM. 2°

Locus coeruleus. Hypotension in rabbits infused with SNP alone evoked Fos expression in about 70% of the TH-positive cells in both the locus coeruleus and subcoeruleus. Following clonidine pretreatment, the number of Fos-positive cells was reduced by about 95%, and virtually none of the remaining Fos-positive cells were catecholamine cells. This observation supports previous electrophysiologica139 and electrochemical 44,58 studies which showed that clonidine has a powerful inhibitory action on catecholamine cells in the locus coeruleus. The locus coeruleus receives afferent inputs from neurons in the ventral medulla, some of which are CI cells, and it may be suggested that the reduced Fos expression in the locus coeruleus is a secondary consequence of inhibition of C1 cells by clonidine. This is most unlikely, however, since the input to the locus coeruleus from C1 cells is inhibitory rather than excitatory. 2 Thus, it is more likely that intravenous clonidine directly inhibits the activity of locus coeruleus neurons in conscious animals. In agreement with this, iontophoretically applied clonidine has been shown to inhibit locus coeruleus neurons in anesthetized animals. 39

A previous study in anesthetized rats has shown that intravenous rilmenidine only results in a decrease in the metabolic activity of locus coeruleus neurons (as measured with in vivo voltammetry) when very high doses of the drug (15 mg/kg), much greater than those required to elicit a maximal hypotensive effect, are administered. 57 In contrast, our observations demonstrated that rilmenidine pretreatment greatly reduces Fos expression in catecholamine neurons within the locus coeruleus and subcoeruleus, even


when injected in doses (150-300#g/kg) that cause only a moderate degree of hypotension. Thus, they do not support the view that hypotensive doses of rilmenidine selectively inhibit rostral VLM neurons but not locus coeruleus neurons, at least in conscious rabbits.

Nucleus o f the solitary tract and area postrema. As in the rostral VLM, both clonidine and rilmenidine pretreatment decreased the number of both catecholamine and non-catecholamine cells in the NTS and area postrema that expressed Fos following hypotension, although the effect on catecholamine cells was somewhat greater. Both of these regions have a very high density of ~2A-adrenoceptors, which do not appear to be restricted to catecholamine neurons. 48 Activation of these receptors by clonidine and other ~2-adrenoceptor agonists has been shown to facilitate the baroreceptor reflex. 35'55 Our observation that rilmenidine has a similar effect as clonidine on Fos expression in the NTS and area postrema suggests that it may have a similar effect on the baroreceptor reflex, as a consequence of its affinity to ~2-adrenoceptors in the NTS.

Other catecholamine cell groups. The A1 cell group is believed to be an important component of the central pathways mediating the baroreceptor reflex regulation of the release of vasopress~n and adrenocorticotropic hormone from the hypothalamus. 5'~2'~6'38 Previous experiments in anesthetized animals using the in vivo voltammetry technique have shown that clonidine suppresses the increase in metabolic activity of catecholamine neurons in the A1 cell group that is normally evoked by hemorrhagic hypotension. 45 Furthermore, an electrophysiological study in anesthetized rats has shown that systemically administered clonidine inhibits the activity of A1 cells that project to the paraventricular nucleus in the hypothalamus. 32 In agreement with this, we found that Fos expression in AI cells was prevented by clonidine pretreatment.

It has also been demonstrated in anesthetized animals that intravenous injection of clonidine suppresses the secretion of vasopressin and adrenocorticotropic hormone. 46 Therefore, our finding that clonidine virtually eliminates the expression of Fos in A1 cells suggests that, in conscious animals, these cells may be one of the central sites at which this drug acts to reduce the release of vasopressin and adrenocorticotropic hormone. Further, our results suggest that rilmenidine has a similar effect.

Hypotension-induced Fos expression in the A5 area was also completely abolished by clonidine and rilmenidine pretreatment. This observation supports and extends previous studies in anesthetized rats that

have shown that the activity of single putative A5 cells is inhibited by both local and systemic application of clonidine. 1~'3° The A5 cells, like C1 cells in the rostral VLM, project directly to and excite sympathetic preganglionic vasomotor nuclei. 3° Thus, inhibition of A5 cells may also be a major factor responsible for the hypotensive actions of clonidine and rilmenidine.

Antihypertensive actions o f clonidine and rilmenidine

An important question is whether the sites of action of clonidine and rilmenidine within the brainstem of normotensive animals are the same sites at which these drugs act to lower arterial pressure in hypertensive animals. Although there is no evidence that the anatomical organization of central cardiovascular pathways in hypertensive animals is different to that in normotensive animals, a number of studies have identified differences in the functional properties of central cardiovascular neurons in spontaneously hypertensive as compared to normotensive rats (e.g. Refs 14 and 52). Thus, it is possible that clonidine and rilmenidine may have different effects, or even different sites of action, in the brainstem of hypertensive animals as compared to normotensive animals. This question would need to be addressed by future studies.

Pharmacological properties o f clonidine and rilmenidine

As mentioned in the Introduction, it is well known that both clonidine and rilmenidine are agonists of ~2-adrenoceptors as well as putative imidazoline receptors, although rilmenidine has a higher selectivity for imidazoline receptors. There are differing views, however, as to which of these receptor types mediates the hypotensive actions of the drugs. 19"24'47 The present observations do not directly address this question, but do show that both drugs produced virtually identical patterns of suppression of Fos expression, despite differences in their pharmacological properties and differences in the regional distribution in the brainstem of ~t2-adrenoceptors and imidazoline receptors. 13'2° Therefore, if there are regional differences in the relative effects of the two drugs on brainstem neurons in conscious animals, these are not reflected by differences in the pattern of Fos expression in these regions.

Acknowledgements--This work was supported by the National Health and Medical Research Council of Australia and the National Heart Foundation of Australia. We are grateful to the Institut de Recerches Internationales Servier, France, for their gift of rilmenidine. We also thank Ms Li-Ping Liu for technical assistance, Mr Jaimie Poison for photography, and Assoc. Prof. Richard Bandler for his comments on the manuscript.

REFERENCES

1. Allen A. M. and Guyenet P. G. (1993) ct2-Adrenoceptor-mediated inhibition of bulbospinal barosensitive cells of rat rostral medulla. Am. J. Physiol. 265, R1065 R1075.

2. Aston-Jones G., Astier B. and Ennis M. (1992) Inhibition of noradrenergic locus coeruleus neurons by C1 adrenergic cells in the rostral ventral medulla. Neuroscience 48, 371-381.


3. Badoer E., Head G. A. and Korner P. I. (1983) Effects ofintracisternal and intravenous alpha-methyldopa and clonidine on haemodynamics and baroreceptor-heart rate reflex properties in conscious rabbits. J. cardiovasc. Pharmac. 5, 760-767.

4. Blessing W. W., Howe P. R. C., Joh T. H., Oliver J. R. and Willoughby J. O. (1986) Distribution of tyrosine hydroxylase and neuropeptide Y-like immunoreactive neurons in rabbit medulla oblongata, with attention to colocalization studies, presumptive adrenaline-synthesizing perikarya, and vagal preganglionic cells. J. comp. Neurol. 248, 285 300.

5. Blessing W. W. and Willoughby J. O. (1985) Inhibiting the rabbit caudal ventrolateral medulla prevents baroreceptor- initiated secretion of vasopressin. J. Physiol., Lond. 367, 253-265.

6. Bonham A. C., Trapani A. J., Portis L. R. and Brody M. J. (1984) Studies on the mechanism of the central antihypertensive effect of guanabenz and clonidine. J. Hypertension 4, $543-$546.

7. Bousquet P. and Guertzenstein P. G. (1973) Localization of the central cardiovascular action of clonidine. Br. J. Pharmac. 49, 573-579.

8. Bousquet P. and Schwartz J. (1983) ct-Adrenergic drugs: pharmacological tools for the study of the central vasomotor control. Biochem. Pharmac. 32, 1459-1465.

9. Bousquet P., Feldman J. and Schwartz J. (1984) Central cardiovascular effects of alpha adrenergic drugs: differences between catecholamines and imidazolines. J. Pharmac. exp. Ther. 230, 232-236.

10. Bricca G., Dontenwill M., Molines A., Feldman J., Tibirica E., Belcourt A. and Bousquet P. (1989) Rilmenidine selectivity for imidazoline receptors in human brain. Eur. J. Pharmac. 163, 373-377.

11. Byrum C. E., Stornetta R. and Guyenet P. G. (1984) Electrophysiological properties of spinally-projecting A5 noradrenergic neurons. Brain Res. 303, 15-29.

12. Carlson D. E. and Gann D. S. (1987) Responses of adrenocorticotropin and vasopressin to hemorrhage after lesions of the caudal ventrolateral medulla in rats. Brain Res. 406, 385 390.

13. Cedarbaum J. M. and Aghajanian G. K. (1977) Catecholamine receptors on locus coeruleus neurons: pharmacological characterization. Eur. J. Pharmac. 44, 375-385.

14. Chan R. K. W., Chan Y. S. and Wong T. M. (1991) Electrophysiological properties of neurons in the rostral ventrolateral medulla of normotensive and spontaneously hypertensive rats. Brain Res. 549, 118-126.

15. Dampney R. A. L. (1994) Functional organization of central pathways regulating the cardiovascular system. Physiol. Rev. 74, 323-364.

16. Day T. A. (1989) Control of neurosecretory vasopressin cells by noradrenergic projections of the caudal ventrolateral medulla. In Central Neural Organization of Cardiovascular Control. Progress in Brain Research (eds Ciriello J., Caverson M. M. and Polosa C.), Vol. 81, pp. 303-317. Elsevier, Amsterdam.

17, Dragunow M. and Faull R. (1989) The use of c-fos as a metabolic marker in neuronal pathway tracing. J. Neurosci. Meth. 29, 261-265.

18, Ernsberger P., Giuliano R., Willette R. N., Granata A. R. and Reis D. J. (1988) Hypotensive actions of clonidine analogues correlates with binding affinity at imidazole and not alpha-2-adrenergic receptors in the rostral ventrolateral medulla. J. Hypertension 6, $554-$557.

19. Ernsberger P., Giuliano R., Willette R. N. and Reis D. J. (1990) Role of imidazole receptors in the vasodepressor response to clonidine analogs in the rostral ventrolateral medulla. J. Pharmac. exp. Ther. 253, 408 418.

20. Ernsberger P., Meeley M. P., Mann J. J. and Reis D. J. (1987) Clonidine binds to imidazole binding sites as well as to %-adrenoceptors in the ventrolateral medulla. Eur. J. Pharmac. 134, 1-13.

21. Gomez R. E., Ernsberger P., Feinland G. and Reis D. J. (199.1) Rilmenidine lowers arterial pressure via imidazole receptors in brainstem C1 area. Eur. J. Pharmac. 195, 181-191.

22. Guyenet P. G. (1990) Role of the ventral medulla oblongata in blood pressure regulation. In Central Regulation of Autonomic Functions (eds Loewy A. D. and Spyer K. M.), pp. 145-167. Oxford University Press, New York.

23. Guyenet P. G., Haselton J. R. and Sun M. K. (1989) Sympathoexcitatory neurons of the rostroventrolateral medulla and the origin of the sympathetic vasomotor tone. In Central Neural Organization of Cardiovascular Control. Progress in Brain Research (eds Ciriello J., Caverson M. M. and Polosa C.), Vol. 81, pp. 105 116. Elsevier, Amsterdam.

24. Guyenet P. G., Lynch K. R., Allen A. M., Rosin D. L. and Stornetta R. L. (1995) ct2-Adrenergic receptors rather than "imidazoline" binding sites mediate the sympatholytic effect of clonidine in the rostral ventrolateral medulla: a new look at the evidence. In Ventral Brainstem Mechanisms and Control Functions (eds Trough O., Millis R., Kiwull-Schone H. and Schl~fke M. E.). Marcel Dekker, New York (in press).

25. Guyenet P. G., Stornetta R. L., Riley T., Norton F. R., Rosin D. L. and Lynch K. R. (1994) Alpha2A-adrenergic receptors are present in lower brainstem catecholaminergic and serotonergic neurons innervating spinal cord. Brain Res. 638, 285-294.

26. Gyermek L. and Soyka L. F. (1975)Steroid anesthetics. Anesthesiology 42, 331-344. 27. Halasz P. and Martin P. (1985) Magellan, A Programme for the Quantitative Analysis of Histological Sections. Halasz

& Martin, Sydney. 28. Haselton J. R. and Guyenet P. G. (1989) Electrophysiological characterization of putative C1 adrenergic neurons in

the rat. Neuroscience 30, 199-214. 29. Head G. A. and Burke S. (1991) Importance of central noradrenergic and serotonergic pathways in the cardiovascular

actions of rilmenidine and clonidine. J. cardiovasc. Pharmac. 18, 819-826. 30. Huangfu D., Koshiya N. and Guyenet P. G. (1991) A5 noradrenergic unit activity and sympathetic nerve discharge

in rats. Am. J. Physiol. 261, R393-R402. 31. Jhamandas J. H. and Harris K. H. (1992) Influence of nucleus tractus solitarius stimulation and baroreceptor activation

on rat parabrachial neurons. Brain Res. Bull. 28, 565-571. 32. Kannan H., Osaka T., Kasai M., Okuya S. and Yamashita H. (1986) Electrophysiological properties of neurons in

the caudal ventrolateral medulla projecting to the paraventricular nucleus of the hypothalamus in rats. Brain Res. 376, 342-350.

33. Kawasaki H. and Takasaki K. (1986) Central alpha-2 adrenoceptor-mediated hypertensive response to clonidine in conscious, normotensive rats. J. Pharmac. exp. Ther. 236, 810-818.

34. Kobinger W. (1978) Central a-adrenergic systems as targets for hypotensive drugs. Rev. Physiol. Biochem. Pharmac. 81, 40 100.


35. Kubo T., Goshima Y., Hata H. and Misu Y. (1990) Evidence that endogenous catecholamines are involved in a2-adrenoceptor-mediated modulation of the aortic baroreceptor reflex in the nucleus tractus solitarii of the rat. Brain Res. 526, 313-317.

36. Laubie M., Poignant J.-C., Scuvee-Moreau J., Dabire H., Dresse A. and Schmitt H. (1985) Pharmacological properties [n(dicyclopropyl-methyl)amino-2-oxazoline], (S 334 I), an alpha2-adrenoceptor agonist. J. Pharmac., Paris 16, 259-278.

37. Li Y.-W. and Dampney R. A. L. (1994) Expression of Fos-like protein in brain following sustained hypertension and hypotension in conscious rabbits. Neuroscience 61, 613-634.

38. Li Y.-W., Gieroba Z. J. and Blessing W. W. (1992) Chemoreceptor and baroreceptor responses of A1 area neurons projecting to supraoptic nucleus. Am. J. Physiol. 263, R310 R317.

39. Marwaha J. and Aghajanian G. K. (1982) Relative potencies of alpha-1 and alpha-2 antagonists in the locus coeruleus, dorsal raphe and dorsal lateral geniculate nuclei: an electrophysiological study. J. Pharmac. exp. Ther. 222, 287-293.

40. Meesen H. and Olszewski J. (1949) A Cytoarchitectonic Atlas of the Rhombencephalon of the Rabbit. S. Karger, Basel. 41. Mermet C. and Quintin L. (1991) Effect of clonidine on catechol metabolism in the rostral ventrolateral medulla: an

in vivo electrochemical study. Eur. J. Pharmac. 204, 105-107. 42. Muzi M., Goff D. R., Kampine J. P., Roerig D. L. and Ebert T. J. (1992) Clonidine reduces sympathetic activity but

maintains baroreflex responses in normotensive humans. Anesthesiology 77, 864 871. 43. Punnen S., Urbanski R., Krieger A. J. and Sapru H. N. (1987) Ventrolateral medullary pressor area: site of hypotensive

action of clonidine. Brain Res. 422, 336 346. 44. Quintin L., Buda M., Hilaire G., Bardelay C., Ghignone M. and Pujol J.-F. (1986) Catecholamine metabolism in the

rat locus coeruleus as studied by in vivo differential pulse voltammetry. III. Evidence for the existence of an ~2-adrenergic tonic inhibition in behaving rats. Brain Res. 375, 235-245.

45. Quintin L., Ghignone M. and Pujol J.-F. (1987) The ability of the ~2-adrenergic agonist clonidine to suppress central noradrenergic hyperactivity secondary to hemodynamic or environmental stimuli. J. cardiovasc. Pharmac. 10, S128-S134.

46. Reid I. A., Ahn J. N., Trinh T., Shackelford R., Weintraub M. and Keil L. C. (1984) Mechanism of suppression of vasopressin and adrenocorticotropic hormone secretion by clonidine in anesthetized dogs. J. Pharmac. exp. Ther. 229, 1-8.

47. Reis D. J., Regunathan S., Wang H., Feinstein D. L. and Meeley M. P. (1992) Imidazoline receptors in the nervous system. Fundam. clin. Pharmac. 6, Suppl. 1, 23s-29s.

48. Rosin D. L., Zeng D., Stornetta R. L., Norton F. R., Riley T., Okusa M. D., Guyenet P. G. and Lynch K. R. (1993) Immunohistochemical localization of ~2A-adrenergic receptors in catecholaminergic and other brainstem neurons in the rat. Neuroscience 56, 139-155.

49. Sannajust F., Cerutti C., Koenig-B~rard E. and Sassard J. (1992) Influence of anaesthesia on the cardiovascular effects of rilmenidine and clonidine in spontaneously hypertensive rats. Br. J. Pharmac. 11t5, 542-548.

50. Saunier C. F., Akaoka H., de la Chapelle B., Charl&y P. J., Chergui K., Chouvet G., Buda M. and Quintin L. (1993) Activation of brain noradrenergic neurons during recovery from halothane anesthesia. Anesthesiology 79, 1072-1082.

51. Schmitt H. (1977) The pharmacology of clonidine and related products. In Handbook of Experimental Pharmacology (ed. Gross F.), Vol. 39, pp. 299 396. Springer, Berlin.

52. Smith J. K. and Barron K. W. (1990) Cardiovascular effects of L-glutamate and tetrodotoxin microinjected into the rostral and caudal ventrolateral medulla in normotensive and spontaneously hypertensive rats. Brain Res. 506, 1-8.

53. Spiers J. P., Harron D. W. G. and Wilson R. (1990) Acute and chronic effects of rilmenidine on baroreflex function in conscious dogs. Eur. J. Pharmac. 181, 235-240.

54. Sun M.-K. and Guyenet P. G. (1986) Effect of clonidine and y-aminobutyric acid on the discharges of medullo-spinal sympathoexcitatory neurons in the rat. Brain Res. 368, 1-17.

55. Sved A. F., Tsukamoto K. and Schreihofer A. M. (1992) Stimulation of ~2-adrenergic receptors in nucleus tractus solitarius is required for the baroreceptor reflex. Brain Res. 576, 297-303.

56. Szabo B., Urban R. and Starke K. (1993) Sympathoinhibition by rilmenidine in conscious rabbits: involvement of c~2-adrenoceptors. Naunyn-Schmiederberg's Arch. Pharmac. 348, 593-600.

57. Tibiriqa E., Feldman J., Mermet C., Monassier L., Gonon F. and Bousquet P. (1991) Selectivity of rilmenidine for the nucleus reticularis lateralis, a ventrolateral medullary structure containing imidazoline-preferring receptors. Eur. J. Pharmac. 209, 213-221.

58. Tibiri9a E., Mermet C., Feldman J., Gonon F. and Bousquet P. (1989) Correlation between the inhibitory effect on catecholaminergic ventrolateral medullary neurons and the hypotension evoked by clonidine: a voltammetric approach. J. Pharmac. exp. Ther. 250, 642 647.

59. Van Zwieten P. A., Thoolen M. J. M. C., Jonkman F. A. M., Willfert B., De Jong A. and Timmermans P. B. M. W. M. (1986) Central and peripheral effects of $3341 [(N-dicyclopropylmethyl)-amino-2-oxazoline] in animal models. A rchs int. Pharmacodyn. 279, 130-149.

60. Watkins L. and Maixner W. (1993) Hypotensive effect of clonidine is not mediated by enhanced baroreflex gain in rats. J. cardiovasc. Pharmac. 21, 264 271.

(Accepted 4 November 1994)

clonidine and rilmenidine suppress hypotension-induced fos expression in the lower braistem of the...

Documents