Cardiovascular responses induced by endothelin microinjection into area postrema

Regulatory Peptides, 27 (1990) 75-85 Elsevier

75

REGPEP 00857

Cardiovascular responses induced by endothelin microinjection into area postrema Alastair V. Ferguson and Pauline Smith Department of Physiology, Queen's University,Kingston, ON (Canada) (Received2 June 1989;revisedversionreceivedand accepted29 August 1989) Key words: Circumventricular organs; Nucleus tractus solitarius

Summary The recently described endothelium derived constricting factor endothelin (ET) is a 21 amino acid peptide which is the most potent endogenous vasoconstrictor yet described. Binding sites for this peptide have been demonstrated within the circumventricular structures of the brain. One of these structures, the area postrema (AP), has been implicated in central cardiovascular control mechanisms. We have therefore examined the effects of AP microinjection of ET on blood pressure in urethaneanaesthetised rats, Such treatment resulted in dose-dependent biphasic changes in arterial blood pressure (increases followed by decreases). Low doses of ET (0.2-1.0 pmol) induced significant increases ( P < 0.01), and high doses (5.0 pmol) significant decreases (P < 0.01), while at intermediate concentrations (2.0 pmol) ET caused significant increases (P < 0.05) followed by significant decreases (P < 0.01) in mean blood pressure. Other vasoconstrictors were found to be without effect following AP administration, suggesting these changes to be the result of specific action of ET. In contrast, both ET and methoxamine had similar cardiovascular actions when microinjected into regions anatomically adjacent to the AP such as the NTS, indicating that vasoconstriction in these areas induces changes in femoral arterial blood pressure. These data suggest a specific role for ET as a chemical messenger involved in central nervous system control of the cardiovascular system within AP.

Correspondence: A.V.Ferguson, Department of Physiology,Queen's University,Kingston ON, Canada K7L 3N6 0167-0115/90/$03.50 © 1990 Elsevier SciencePublishers B.V. (BiomedicalDivision)

76 Introduction

Since the discovery of endothelium-derived relaxing factors in the early 1980's, the endothelial cell has become recognised as an important functional unit in the regulation of vascular smooth muscle [1 ]. It is now well established that under the influence of a variety of circulating substances, endothelial cells have the ability to release relaxing (EDRF) or constricting (EDCF) factors, which then act as paracrine agents on local tissue. However, until recently, little attention has been paid to the potential roles of these endothelium derived factors (EDF's) in the central nervous system. The peptide endothelin (ET) is the first EDCF to be structurally characterised [2]. It is a 21 amino acid peptide containing two disulphide bonds, which appears to exist as one of three different isomers (ET-1, ET-2, ET-3) each of which has a number of minor amino acid substitutions [3]. The peptide shows considerable structural similarity to the sarafatoxins derived from asp venom [4]. It is the most potent vasoconstrictor known to date [2,3,5]. Electrophysiological recordings from cultured muscle cells have demonstrated that administration of ET results in a biphasic change in membrane potential consisting of an initial short lasting hyperpolarisation followed by a prolonged (more than 10 min) depolarisation [6]. Such effects were suggested to result from activation of a Ca z + -sensitive K + channel and from opening of a non-specific cation channel, respectively [6]. There is, at present, no direct evidence indicating functional roles for this peptide within the central nervous system. However, the recent demonstration of high densities of ET binding sites in a variety of regions of the brain including the dorsal vagal complex [7], subfornical organ (SFO) [8], and median eminence [8] suggest these to be potential CNS sites of action for this peptide. The area postrema (AP) in the rat is a midline circumventricular structure located at the dorsal surface of the medulla and thus is a part of the dorsal vagal complex. It is highly vascular, lacks the normal blood brain barrier; and has been demonstrated to contain the highest densities of binding sites for both angiotensin II (ANG) [9] and atrial natriuretic peptide (ANP) [ 10] in the mesencephalon. Anatomical tracing studies have demonstrated efferent neural connections from AP to the adjacent nucleus tractus solitarius (NTS), the parabrachial nucleus in the pons, and have suggested further projections to both spinal and hypothalamic autonomic centres [ 11,12]. A considerable body of evidence has implicated this structure as an essential chemoreceptor zone in the control of emesis [ 13,14]. Thus, circulating substances are believed to gain direct access to neural elements within this structure and so initiate the neural processes which lead to vomiting. In accordance with such a proposal, it has recently been reported that systemic ET infusion (0.72 nmol/kg per min) in dogs induces vomiting [5 ], although the specific locus at which ET acts to produce such effects has yet to be established. In addition, destruction of the AP in rats has been reported to result in hypertension [ 15], thus implicating this structure in cardiovascular control mechanisms. In accordance with such a hypothesis we have demonstrated that electrical stimulation in AP of the rat results in rapid reversible decreases in both blood pressure and heart rate [ 16], while in the rabbit lower intensity stimulation has been reported to decrease sympathetic output [ 17]. A N G has been suggested to elicit it's CNS effects on the cardiovascular system, at least in part, through actions within the AP [ 18], a suggestion which draws

77 further support from the electrophysiological demonstration that AP neurons are influenced by circulating ANG [ 19-21], and also in some cases by changes in blood pressure [21 ]. In combination, this background literature suggests that ET may have specific actions within AP. The studies reported here have therefore attempted to address this issue by examining the cardiovascular effects of microinjection of the endothelial cell derived peptide ET directly into this circumventricular organ.

Materials and Methods

Surgical preparation Urethane anaesthetised (1.4 g/kg) male Sprague-Dawley rats (150-300 g) were used in all experiments. Animals were fitted with endotracheal tubes to facilitate breathing. Body temperature was maintained at 37.0 + 1.0 °C using a feedback-controlled heating blanket. A femoral arterial catheter (PE50-Intramedic) was inserted and connected to a pressure transducer, the output of which was sent to a Buxco cardiovascular analyser such that mean arterial blood pressure and heart rate could be continuously monitored and recorded on a computer-based data acquisition system (Codas). Each animal was then placed in a stereotaxic frame with it's head in the vertical position (nose down) and the dorsal medulla was surgically exposed by a midline incision such that a microinjection pipette could be stereotaxically positioned in the AP.

Chemicals All drugs administered into AP were dissolved in phosphate buffered saline (pH 7.4) which was therefore administered as a vehicle control. These included ET (PeninsulaHuman, Peptides International-ETl: 10-8-10 -5 M), and the ~-adrenergic agonist methoxamine (Vasoxyl 10-3 M, utilised as an alternative vasoconstrictor). Manipulation of the EDRF system was achieved [22] using pharmacological treatments including pretreatment with indomethacin (10 mg/kg sc. in NaHCO3, 60 min prior to ET), an inhibitor of prostacylin (an EDRF) synthesis; and methylene blue (1 ~o infused at 0.1 ml/min for 10 min, ET administered 5 min into infusion time) an inhibitor of the action of the EDRF nitrous oxide.

Experimental Protocol Microinjections were made using glass micropipettes (tip diameter < 10/am) which were attached to a Hamilton 10/al syringe, and then backfilled with drugs of interest. A small air bubble was always left in the tubing connecting the syringe to the microinjection pipette such that the movement of this bubble during injections confirmed patency of the pipette tip. This microinjection setup was then positioned using a micromanipulator over the region of the AP or the NTS, according to surface landmarks, and advanced into either of these regions according to the stereotaxic coordinates of Paxinos and Watson [23]. Following a minimum 5 min, baseline recording period microinjections were made in volumes ranging from 0.1-1.0/al, and effects on blood pressure and heart rate were

78 assessed. Using these techniques we were able to examine, in some cases in a single animal, the effects of multiple doses of a different drugs in both anatomical locations (i.e., AP and NTS) of interest in the present studies.

Histology At the completion of each experiment, rats were overdosed with pentobarbital and the animal then was perfused by administration of normal saline followed by a 10~o formalin solution through the left ventricle of the heart. The brain was then removed and stored in formalin overnight. The next day, 50 #m coronal sections were cut through the medulla using a vibratome. These sections were mounted and then stained with cresyl violet. The anatomical location of microinjection sites were then determined in this histologically prepared post-mortem tissue by an investigator unaware of the experimental data derived from the tissue being examined. Animals were then assigned to one of two experimental groups according to the location of microinjection sites in AP or NTS. Data from rats in which sites were not localised to either of these specific regions were not included for further analysis.

Data Analysis Data from experimental animals were analysed to determine the specific (as opposed to effects also induced in other anatomical locations or effects similar to those induced by alternative vasoconstrictors) effects of administration of ET into AP. Biphasic cardiovascular responses (an increase in blood pressure followed by a later decrease) to this peptide were observed in most cases, and therefore effects of each dose of ET were measured according to the maximum change from baseline observed. As such biphasic responses were observed, data points representing both pressor and depressor effects as a result of each dose were included for analysis. In cases where only a pressor or depressor effect was observed, the absent effect was entered in data tables as a zero change value. The majority of animals to which the higher doses of ET were administered died approximately 20 min after peptide adminstration, following respiratory failure. However these rats routinely demonstrated a depressor response to ET which reached a maximum and then returned toward baseline blood pressure prior to respiratory failure occurring. In such cases, the maximum decrease in blood pressure was evaluated as the minimum value reached prior to this return toward baseline. Statistical comparisons were made utilising the Student's t-test to evaluate changes in blood pressure both within group (paired test evaluating changes in blood pressure elicited by microinjection of a specific dose of ET elicited significant changes increases or decreases in blood pressure) and between group (unpaired test comparing blood pressure effects of methoxamine to those of ET).

Results

A total of 69 animals were used in the present studies of which 27 had histologically confirmed AP microinjection sites (blood pressure 93 + 4 m m H g , heart rate 375 + 7 bpm) and 18 with NTS sites (blood pressure 86 + 5 mmHg, heart rate

79

362 + 15 bpm). In all cases the data we will present in this study are derived from examination of the cardiovascular effects of such microinjections in the 5 min period following drug administration. A typical example of an AP microinjection site is illustrated in Fig. 1A, while Fig. 1B and C show schematic medullary coronal sections summarising the locations of AP and NTS ET microinjection sites, respectively. All

A

NTS

O C

m

AP

NT$

m

•

O

• •

Fig. 1. This figure illustrates the anatomical locations of microinjection sites in both area postrema (AP) and the nucleus tractus solitarius (NTS). In A, a photomicrograph is presented showing a typical example of a microinjection site within AP (*). The schematic coronal sections shown in B and C summarise the location of all ET microinjections in AP and NTS, respectively. Scale bar = 0.1 mm.

80

remaining animals were found to have injection sites which did not specifically fit into either of these anatomical groupings. Area postrema ET was microinjected into AP in total doses ranging from 0.02 pmol (0.2 #1 10 - 7 M) to 5.0 pmol (0.5/~1 10 - 5 M). This lower dose limit for such testing in the present studies was in accordance with the observation that these doses were without effect on blood pressure (Fig. 2A). In contrast, the 5.0 pmol dose was not exceeded, as in all cases the initial cardiovascular changes (see Fig. 2D) were followed by a longer latency precipitous fall in blood pressure and heart rate, and a cessation of breathing resulting in death of the animal. As illustrated in Fig. 2B and C and summarised in Fig. 3, lower doses of ET resulted in statistically significant increases in blood pressure which were dose-

140

Blood pressure (mmHg) A 4 0.02 pmol

120

100

BO

2

4

6

8

10

12

140

40.2 pmol

B 120

100

80

2

4

6

8

10

12

2

4

6

8

10

12

6

8

10

12

140

C 120

lO0

80 140

D

42.0 pmol

120

100

80

2

4

Time (mins) Fig. 2. These polygraph record (A-D) show typical examples of the effects of differing doses of ET (total dose indicated on each record) microinjected into AP (injection time indicated by the arrow) on mean arterial blood pressure.

81

15

A

W nZ

0

o o

E E

0 --~ J O3

"

n

-lb

-50

8~

Z

!

<1

45 O.8~mol

2.0nmol

[ UETHOXAUINE]

**&

-60 O.2 (N=I I)

O.5 (N=12)

1.0 (N= I0)

2.0 (N=I~)

5.0 (N=7)

[ENDOTHELINI

(p~oO Fig. 3. This histogram summarises both the pressor (open bars) and depressor (filled bars) effects of different doses of ET following administration into the AP. Each bar represents mean data + S.E.M. for grouped data, while the number of animals comprising each group is indicated on the horizontal axis. The inset shows mean responses to methoxamine microinjected into AP, and illustrates that this substance was without significant effect on blood pressure. In this and followingfigures significant changes from baseline blood pressure in response to microinjeetions are indicated by **P < 0.01, and *P < 0.05. In addition, ET responses which were significantly different to methoxamine (0.8 nmol) effects are also indicated • P < 0.05. No significant changes (P > 0.1) were observed in other groups. dependent (0.2 pmol, 5 . 6 + 1.0 m m H g , P < 0 . 0 1 ; 0.5 pmol, 1 0 . 9 + 3 . 3 m m H g , P < 0.01). Conversely, the higher doses o f this peptide administered in the present study caused statistically significant falls in arterial blood pressure (2.0 pmol, 27.8 + 7.3 m m H g , P < 0.01 ; 5.0 pmol, - 48.0 + 8.5 m m H g , P < 0.01), which were often preceeded by hypertensive effects similar to those elicited by the lower doses of the peptide as illustrated in Fig. 2C and summarised in Fig. 3. The dose-related effects observed were apparently not related to the volume of the microinjection, as similar effects of 1.0 pmol ET followed administration of this peptide in either 0.1 or 1.0/~1. We found no consistent changes in heart rate in the time period immediately following E T administration into AP. In all cases, vehicle control injections were found to be without effect following administration into the AP. In order to examine the possibility that the observed depressor effects of ET may result from a secondary release of E D R F ' s as a compensatory response to the increased local concentrations of the former peptide, we also tested the effects o f ET administration into A P in both indomethacin- and methylene blue-pretreated animals. D a t a from these experiments are presented in Table I and demonstrate that such treatments were without effect on the blood pressure changes elicited by A P ET (5.0 pmol) microinjection. -

82

TABLE I Effect of 5.0 pmol ET This table summarises the effects of microinjection of 5.0 pmol ET into the AP of indomethacin and methylene blue (Indo/mb, n = 4) pretreated animals. Statistical analyses demonstrated that these responses were not significantly different to those observed in untreated animals (Control, n = 7) in response to similar doses of ET. Indo/mb

Control

Mean

S.E.M.

Mean

S.E.M.

P

2.2 - 31.5

0.9 4.3

8.9 - 48.0

4.1 8.5

> 0.1 > 0.1

Pressor response Depressor response

In view of the potent vasoconstrictor actions of ET, we have attempted to establish whether the cardiovascular changes reported above may be secondary to the vasoconstrictor actions of this peptide within the AP. We therefore compared the effects of ET microinjection with those of a second vasoconstrictor, methoxamine (~-adrenergic agonist 0.8 and 2.0 nmol). As shown in Fig. 3 (inset), this adrenergic agonist was without significant effect (P > 0.1) on blood pressure following microinjection into AP. However, such a dose of methoxamine has a considerably greater pressor effect than 0

L~

-10

u0 of] L~ ~ , Od 0", ££ I

o 0

03 Z <3

E E

20

-50

-40

i

i

i

T

r

0.2

0.5

2.0

5,0

0.8

2,0

(N=7)

(N=6)

(N=7)

(N=6)

(N=5)

(N =,5)

[ENDOTHELIN]

(p oO

[METHOXAMINE]

( mol)

Fig. 4. The effects of ET administration into the NTS are summarised in this histogram. Each bar represents the mean + S.E.M. for grouped data in response to different doses of ET (filled bars) or methoxamine (hatched bars) microinjected into NTS. These data illustrate that both of these substances caused similar changes in blood pressure, suggesting that such effects result from their common vasoconstrictor actions. Similar annotation to Fig. 3 is used to indicate statistical significance.

83 the doses of ET we utilised in the present study, if both substances are administered systemically. Nucleus tractus solitarius

A further series of experiments were carded out in an effort to establish whether the cardiovascular actions of ET were specific to the AP microinjection sites. In these studies, we compared the effects of administration of this peptide into the immediately adjacent NTS with effects of AP administration. ET microinjection into NTS across a similar dose range to that used in the AP studies resulted in cardiovascular effects which are summarised in Fig. 4. Statistically significant decreases in blood pressure were observed in response to 0.2 pmol ( - 21.7 + 6.0 mmHg, P < 0.05), and 0.5 pmol ( - 23.3 + 5.5 mmHg, P < 0.05) of ET, while the higher doses were without significant effect (P > 0.1). We again examined whether these effects were perhaps the result of the vasoconstrictor actions of ET, by comparing the effects of this peptide to those of methoxamine administration. NTS microinjection of 0.8 and 2.0 nmol of this 0t-adrenergic agonist resulted in similar depressor responses to those observed following ET (Fig. 4), suggesting the possibility that such effects were secondary to the pressor actions of these two unrelated substances.

Discussion

Our data demonstrate that microinjection of the endothelium-derived peptide ET into the AP results in changes in blood pressure which are dose-dependent. This 21 amino acid peptide appears to be particularly potent in eliciting these changes in that pmol concentrations are effective. We have also demonstrated that the observed cardiovascular effects are not reproduced by similar microinjection of a second vasoconstrictor, methoxamine, into the AP in doses which elicit significantly larger increases in blood pressure than 5 pmol ET given systemically. These findings argue strongly against the possibility that the observed effects of ET occur indirectly as a result of the previously demonstrated vasoconstrictor actions of this peptide [2,3,5]. Microinjections of ET into the NTS were also undertaken to examine whether the effects observed following AP administration were specific to an action of the peptide in this anatomical location. Data from these experiments show that, although ET microinjection into the NTS results in decreases in blood pressure, methoxamine had similar effects, suggesting that in this region vasoconstriction may be the primary factor inducing these changes in blood pressure. However an alternative hypothesis which cannot be ruled out at this time is that both these substances have similar direct effects on neuronal activity and thus blood pressure following administration into this region. Additional information indicating that ET effects are specific to the AP microinjection sites is derived from the observations (1) that the largest responses to ET occur following microinjection into AP rather than NTS and (2) that methoxamine has no effect after administration into AP (despite it's observed NTS effects) indicating that this substance does not diffuse into NTS following microinjection in similar volumes. We have recently reported that electrical stimulation in AP decreases arterial blood

84

pressure [ 16], suggesting that depolarisation of AP neurons elicits depressor effects. In accordance with such data we would hypothesise that the pressor effects observed in response to ET may result from hyperpolarisation of AP neurons, while the sustained depressor action following higher doses of this peptide may result from a long-lasting depolarisation of neurons within this circumventricular structure. There is at present no direct evidence for such effects of ET on neuronal excitability. However, Van Renterghem et al. [6], have recently reported that ET has an initial hyperpolarising followed by sustained depolarising effect on membrane potential in cultured muscle cells. Such reports of long-lasting effects of ET are in accordance with receptor binding studies demonstrating that specific binding of this peptide to vascular smooth muscle cells is highly resistant to dissociation [24]. Initial examination of our data suggests that there may in fact be two differentially sensitive mechanisms through which ET elicits these changes in blood pressure, perhaps through actions on separate groups of neurons within the AP. Although we have no definitive evidence to either support or refute such a hypothesis, future studies utilising electrophysiological single unit recording techniques to examine the effects of local administration of ET on the activity of AP neurons could address such a possibility. An alternative explanation is that the depressor effects of ET observed in the present study may result from a compensatory local release action of endothelium-derived releasing factors (EDRF), such has been suggested to take place in the isolated perfused mesenteric artery [25]. However, our observation that neither methylene blue nor indomethacin pretreatment significantly influenced the observed responses to ET in AP argues against such an explanation. In conclusion, these studies have demonstrated potent and anatomically specific cardiovascular actions ofET microinjected into the AP. These findings suggest that this peptide, which is produced and released from endothelial cells, may have specific cardiovascular actions following the attachment to binding sites within this circumventricular structure which are not directly associated with its vasoconstrictor actions. Whether such actions result from direct effect on AP neuronal activity will require further electrophysiological analysis. The physiological significance of ET actions within the AP is unclear at the present time and future studies will be required to determine whether this peptide acts as a true circulating hormone (being released at distant sites and carried to the AP within the circulation), or perhaps as a local paracrine agent being released from AP endothelial cells and acting at local sites within this structure.

Acknowledgements This work was supported by a grant from the Heart and Stroke Foundation of Ontario. Thanks to Dr. J.L. Wallace for the generous gift of ET and for his input to the development of these studies.

References 1 Greenberg, S. and Diecke, F.J.P., Endothelium-derived relaxing and contracting factors: New concepts and new findings, Drug Development Research, 12 (1988) 131-149.

85 2 Yanagisawa, M., Kurihara, H., Kimura, S., Tomobe, Y., Kdbayashi, M., Mitsui, Y., Yazaki, Y., Goto, K. and Masaka, T., A novel potent vasoconstrictor peptide produced by vascular endothelial cells, Nature, 332 (1988) 411-415. 3 Masaki, T., The discovery, the present state, and the future prospects of endothelin, J. Cardiovasc. Pharmacol., 13 (1989) SI-S4. 4 Takasaki, C., Tamiya, N., Bdolah, A., Wollberg, Z. and Kochva, E., Sarafotoxins $6: several isotoxins from Atractaspis engadensis (burrowing asp) venom that affect the heart, Toxicon, 26 (1988) 543-548. 5 Goetz, K.L., Wang, B.C., Madwed, J.B., Zhu, J.L. and Leadley R.J., Jr., Cardiovascular, renal and endocrine responses to intravenous endothelin in conscious dogs, Am. J. Physiol., 255 (1988) RI064-R1068. 6 Van Renterghem, C., Vigne, P., Berhanin, J., Schmid-Alliana, A., Frelin, C. and Lazdunski, M., Molecular mechanism of action of the vasoconstrictor peptide endothelin, Biochem. Biophys. Res. Commun., 157 (1988) 977-985. 7 Jones, C. R., Hiley, C. R., Pelton, J.T. and Mohr, M., Autoradiographic visualization of the binding sites for [~25I] endothelin in rat and human brain, Neurosc. Lett., 97 (1989) 276-279. 8 Koseki, C., Imai, M., Hirata, Y., Yanagisawa, M. and Masaki, T., Binding sites for endothelin-1 in rat tissues: An autoradiographic study, J. Cardiovasc. Pharmaeol., 13 (1989) S153-S154. 9 Mendelsohn, F.A.O., Quirion, R. Saavedra, J.M. and Anguilea, G., Autoradiographic localization of angiotensin II receptors in rat brain, Proc. Natl. Acad. Sci. USA, 8"1 (1984) 1575-1579. 10 Bianchi, C., Gutkowska, J., Ballak, M., Thibault, G., Garcia, R., Genest, J. and Cantin, M., Radioautographic localization of ~25I-artial natriuretic factor bindillg sites in the brain, Neuroendocrinology, 44 (1986) 365-372. 11 Shapiro, R.E. and Miselis, R.R., The Central Connections of the Area Postrema of the Rat, J. Comp. Neurol., 234 (1985) 344-364. 12 Van der Kooy, D. and Koda, L.Y., Organization of the Projections of a Circumventricular Organ: The Area Postrema in the Rat, J. Cornp. Neurol., 219 (1983) 328-338. 13 Borison, H.L., Borison, R. and McCarthy, L.E. Role of the area postrema in vomiting and related functions, Fed. Proc., 43 (1984) 2955-2958. 14 Borison, H.L.,Areapostrema: chemoreceptor trigger zone for vomiting-is that all?, Life Sci., 14(1974) 1807-1817. 15 Ylitalo, P., Karppanen, H. and Paasonen, M. K., Is the area postrema a control centre ofblood pressure?, Nature, 274 (1974) 58-59. 16 Ferguson, A.V. and Marcus, P., Area postrema stimulation induced cardiovascular changes in the rat., Am. J. Physiol., 255 (1988) R855-R860. 17 Hasser, E. M., Nelson, D.O., Haywood, J. R. and Bishop, V. S., Inhibition of renal sympathetic nervous activity by area postrema stimulation in rabbits, Am. J. Physiol., 253 (1987) H91-H99. 18 Casto, R. and Phillips, I., Cardiovascular actions of microinjections to angiotensin II in the brain stem of rats, Am. J. Physiol., 246 (1984) R811-R816. 19 Carpenter, D.O., Briggs, D.B., Knox, A.P. and Strominger, N., Excitation of area postrema neurons by transmitters, peptides and cyclic nucleotides, J. Neurophysiol., 59 (1988) 358-369. 20 Carpenter, D.O., Briggs, D.B. and Strominger, N., Responses of neurons of canine area postrema to neurotransmitters and peptides, Cellular Molec. Neurobiol., 3 (1983) 113-126. 21 Papas, S., Smith, P. and Ferguson, A.V., Electrophysiological evidence that systemic angiotensin influences rat area postrema neurons. Am. J. Physiol., in press. 22 Wallace, J. L., Cirino, G., De Nucci, G., McKnight, W. and McNaughton, W. K., Endothelin has potent ulcergenic and vasocostrictor actions in the stomach, Am. J. Physiol., 256 (1989) G661-G666. 23 Paxinos, G. and Watson, C. The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 1982. 24 Hirata, Y., Yoshimi, H., Takaichi, S., Yanagisawa, M. and Masaki, T., Binding and receptor down regulation of a novel vasoconstrictor endothelin in cultured rat muscle cells, FEBS. Lett., 239 (1988) 13-17. 25 De Nucci, G., Thomas, R., D'Orleans-Juste, P., Antunes, E., Walder, C., Warner, T.D. and Vane, J.R., Pressor effects of circulating endothelin are limited by its removal in the pulmonary circulation and by the release of prostacyclin and endothelium-derived relaxing factor, Proc. Natl. Acad. Sei. USA, 85 (1988) 9797-9800.