Directory UMM :Data Elmu:jurnal:B:Brain Research:Vol879.Issue1-2.2000:

(1)

www.elsevier.com / locate / bres

Research report

Autonomic and cardiovascular reflex responses to central estrogen

injection in ovariectomized female rats

*

Monique C. Saleh, Barry J. Connell, Tarek M. Saleh

Department of Anatomy and Physiology, Atlantic Veterinary College, University of Prince Edward Island, Charlottetown, P.E.I., Canada C1A 4P3 Accepted 25 July 2000

Abstract

The role of estrogen in central autonomic nuclei was examined in ovariectomized female Sprague–Dawley rats supplemented daily for 7 days with either estrogen (5mg / kg; sc) or saline (0.9%; sc). Animals were subsequently anaesthetized with sodium thiobutabarbital (Inactin; 100 mg / kg; ip) and instrumented to record blood pressure and heart rate. Efferent vagal parasympathetic (VPNA) and renal sympathetic (RSNA) nerve activities were recorded and used to assess baseline and reflexive changes in autonomic tone. The cardiac baroreflex was evoked using a single bolus injection of phenylephrine (0.1 mg / kg) both before and following either intrathecal injection of estrogen (0.5 mM; 1 ml) or bilateral injection of estrogen (0.5 mM; 100 nl / side) into several central autonomic nuclei. In estrogen-replaced rats, both the baseline and PE-evoked values for mean arterial pressure and RSNA were significantly decreased following injection of estrogen into the nucleus tractus solitarius (NTS), rostral ventrolateral medulla (RVLM), parabrachial nucleus (PBN), central nucleus of the amygdala (CNA) and the intrathecal space. Baseline heart rate and VPNA were significantly decreased following injection of estrogen into NTS, nucleus ambiguus (Amb), PBN and the intrathecal space. PE-evoked changes in heart rate and VPNA were significantly enhanced following injection of estrogen into these same nuclei. Injection of estrogen into the insular cortex (IC) produced significant decreases in baseline and PE-evoked RSNA only. The cardiac baroreflex was significantly enhanced following injection of estrogen into all nuclei and the intrathecal space. In saline-replaced females, injection of estrogen into NTS, RVLM, Amb and the intrathecal space had similar effects on both baseline and PE-evoked parameters although of a reduced magnitude compared to estrogen-replaced rats. However, no significant changes in autonomic tone and baroreflex function were observed following the injection of estrogen into the PBN, CNA or IC of saline-replaced rats. These results demonstrate a role for estrogen in central autonomic nuclei in female rats and suggest a possible alteration of estrogen receptor distribution or efficacy within the central nervous system of estrogen-deficient female rats.  2000 Elsevier Science B.V. All rights reserved.

Theme: Endocrine and autonomic regulation

Topic: Cardiovascular regulation

Keywords: Parabrachial nucleus; Central nucleus of the amygdala; Insular cortex; Nucleus ambiguus; Nucleus tractus solitarius; Rostral ventrolateral medulla; Spinal cord

1. Introduction nucleus ambiguus and intermediolateral cell column of the

spinal cord respectively [18,19]. With the additional evi-There is increasing evidence to support a role for dence of estrogen receptor mRNA [1,22,23] as well as the estrogen as a central modulator of autonomic tone. Experi- estrogen-synthesizing enzyme, aromatase cytochrome P-ments in male and female rats have demonstrated that 450 within discrete regions of the central nervous system intravenous estrogen administration produces significant [16], the contribution of locally-produced estrogen to the changes in autonomic tone via direct effects on para- regulation of autonomic and cardiovascular reflexes be-sympathetic and be-sympathetic preganglionic neurons in the comes more likely.

Experiments in male rats have demonstrated that direct injection of estrogen into cardiovascular and autonomic

*Corresponding author. Tel.: 11-902-566-0819; fax: 1

1-902-566-nuclei in the brain stem elicits significant changes in

0832.

E-mail address: [email protected] (T.M. Saleh). autonomic tone and baroreflex function [20]. Specifically,

(2)

significant increases in parasympathetic tone and barorefl- daily subcutaneous injections of either estrogen (17b -ex sensitivity were observed following bilateral injection estradiol-3-sulfate, water soluble form; 5 mg / kg; Sigma of estrogen into the nucleus tractus solitarius and nucleus Chemical, St. Louis, MO, USA; n555) or physiological ambiguus. Conversely, sympathetic tone was significantly saline (0.9%; n553) over an additional 7 days. This decreased following injection of estrogen into the nucleus estrogen replacement regimen produces sustained serum tractus solitarius and rostral ventrolateral medulla [20]. In estrogen levels similar to that in an intact rat during general, the apparent role of estrogen in cardiovascular proestrous (40–50 pg / ml) [19].

nuclei is to produce a shift in sympatho-vagal balance On the day of experimentation, animals were anaesthet-toward the parasympathetic limb resulting in an enhance- ized with sodium thiobutabarbital (Inactin; RBI, Natick, ment of baroreflex function which has been shown to MA, USA; 100 mg / kg; ip) and instrumented to record provide antifibrillatory protection on the heart [5]. blood pressure, heart rate and efferent vagal and renal Investigation into the role of estrogen in cardiovascular nerve activities as described previously [17–19]. Blood nuclei of female rats has been limited. In estrogen-replaced pressure and heart rate data were displayed and analysed ovariectomized female rats, the cardiovascular responses to using POLYVIEW PRO/ 32 data acquisition software (Grass;

glutamate injection in the bed nucleus of the stria ter- Warwick, RI, USA). The multi-unit nerve activity was minalis were significantly enhanced when preceded by amplified by a Grass model P55 preamplifier (Grass) with injection of estrogen [6]. Similar injections of estrogen in a 100-Hz to 3-kHz bandpass and 60-Hz notch filter and non-estrogen replaced ovariectomized female rats had no then displayed using thePOLYVIEW PRO/ 32 data acquisition

effect on the glutamate-evoked responses [6] suggesting system (Grass). A sampling rate of 2000 / s was used to that the effects of estrogen within the central nervous record the raw nerve signal. Non-biological noise was system may in fact be dependent on levels of circulating obtained by recording from the nerve after death and was estrogen in the peripheral vasculature. The present in- subtracted from the nerve signal recorded during the vestigation will attempt to expand our understanding of the experiment. A venous catheter (PE-10) was inserted into role of estrogen within the central nervous system as well the right femoral vein to permit administration of drugs. as the contribution of peripheral estrogen levels to the Respiration was facilitated via the insertion of an endotra-neuromodulatory function of estrogen. Specifically, the cheal tube and ventilation with room air on a Harvard effects of estrogen injection into several central autonomic rodent ventilator (65 strokes per min; 2.5-ml tidal volume). nuclei on autonomic tone and baroreflex function will be

examined in estrogen-replaced and saline-replaced ovariec- 2.2. Central injections tomized female rats. Changes in autonomic tone will be

assessed by monitoring efferent vagal and renal nerve Animals (n593) were placed in a David Kopf (Tujunga, activities. Baroreflex function will be assessed using CA, USA) stereotaxic frame and small burr holes were intravenous administration of a single dose of phenylep- drilled bilaterally through the skull to permit stereotaxic hrine hydrochloride. Finally, co-injection of a potent and insertion of a 30-gauge stainless steel, 1-ml Hamilton selective estrogen receptor antagonist, ICI 182,780 [10,25], micro-syringe. A bilateral injection of either estrogen with estrogen into brain nuclei will be used to verify the (17b-estradiol-3-sulphate; Sigma-Aldrich; St. Louis, MO, specificity and receptor-mediated action of estrogen on USA; 0.5 mM; 100 nl per side) or a combination of autonomic tone and baroreflex function. estrogen (0.5mM) and ICI 182,780 (Tocris; Bristol, UK; 1 pM; 100 nl per side) was made into each of the following nuclei according to coordinates obtained from a stereotaxic

2. Materials and methods atlas of the rat brain [15]: nucleus tractus solitarius (NTS),

nucleus ambiguus (Amb), rostral ventrolateral medulla All experiments were carried out in accordance with the (RVLM), parabrachial nucleus (PBN), central nucleus of guidelines of the Canadian Council on Animal Care and the amygdala (CNA) and insular cortex (IC). Control were approved by the University of Prince Edward Island injections of physiological saline (0.9%; 100 nl per side) Animal Care Committee. and ICI 182,780 were made into each nucleus prior to injection of estrogen and ICI 182,780 / estrogen

respective-2.1. General surgical procedures ly.

Experiments were performed on a total of 108 female 2.3. Intrathecal injections Sprague–Dawley rats (Charles Rivers; Montreal, PQ)

weighing 250–275 g. All rats were anaesthetized with A separate group of animals (n515) was instrumented sodium pentobarbitol (somnitol; 17 mg / kg) and valium for the intrathecal administration of either estrogen (0.5 (0.4 mg / kg; ip) and a bilateral ovariectomy was performed mM; 1ml) or a combination of ICI 182,780 (1 pM; 1 ml) by ligation and dissection of the ovaries. A recovery period and estrogen (0.5mM) by the insertion of a catheter (PE of 1 week was permitted prior to the commencement of 10) through a small puncture in the atlanto-occipital

(3)

membrane. The catheter was advanced caudally through Changes in nerve activity and phenylephrine-induced the subarachnoid space placing the tip of the catheter cardiovascular responses were analysed by two-way immediately rostral to the lumbar enlargement (T ) To12 . ANOVA for repeated measures followed by a Student– inject, the catheter was slowly withdrawn over a distance Newman–Keul’s post hoc analysis (Sigma Stat). In all of 5 cm while making 200-nl injections at 1-cm intervals cases, differences were considered significant if P#0.05. between the lumbar enlargement (T ) and the base of the12

cervical spinal cord (T ) Control injections of saline_{1 .} 2.6. Histology (0.9%; 1 ml) preceded intrathecal estrogen injection.

Control injections of saline / ICI 182,780 (0.9%; 1 pM; 1 At the end of each experiment animals were perfused ml) preceded intrathecal injection of estrogen / ICI 182,780. transcardially with 0.9% saline followed by 10% formalin. The brains were removed and stored in 10% formalin until 2.4. Baroreflex testing and autonomic tone the location of micro-syringe tracks could be verified

measurements histologically in thionin-stained coronal sections (6100

mm). For verification of intrathecal catheter placement, 1 In all animals, the baroreflex was evoked by the ml of blue ink was injected using the protocol described intravenous administration of phenylephrine hydrochloride above and the presence of ink in the region of the base of (PE; 0.1 mg / kg) 5 min prior to, and 5, 30, 60, 90 and 120 the cervical spinal cord (T ) to the rostral tip of the lumbar1 min following the central injection of estrogen and 5 min enlargement (T ) was confirmed.12

prior to, and 5, 30, and 60 min following the central injection of estrogen / ICI 182, 780. As well, the baroreflex

was tested 5 min prior to and 5 and 30 min following 3. Results

control injections of saline and ICI 182,780.

Sympatho-vagal balance was assessed at each time point by moni- 3.1. Effect of estrogen on baseline parameters toring changes in renal and vagal efferent nerve activities,

as well as by calculating the ratio of the bradycardic 3.1.1. Estrogen-replaced animals

response to the phenylephrine-evoked pressor response Prior to estrogen injection, mean arterial pressure (Index of Baroreflex Function5DHR /DMAP beats per (MAP) and heart rate (HR) were 100612 mmHg and

min / mmHg). 298619 beats per min respectively (n555). Baseline

values for vagal parasympathetic nerve activity (VPNA) 2.5. Data analysis and renal sympathetic nerve activity (RSNA) were 1964 mV (n555) and 1565 mV (n555) respectively. Control All data are presented as mean6standard error of the injections of saline and co-injection of estrogen / ICI mean (S.E.M.). Changes in blood pressure, heart rate and 182,780 into all nuclei and the intrathecal space had no nerve activity data were calculated at pertinent time points effect on baseline blood pressure, heart rate or nerve using the analysis mode of the POLYVIEW PRO/ 32 software activities (data not shown). Baseline mean arterial pressure

program. Changes in blood pressure and heart rate were and RSNA were significantly decreased 30 min following analysed by a one-way analysis of variance (ANOVA) for injection of estrogen into NTS (n54), RVLM (n54), PBN repeated measures followed by a Student–Newman– (n54), CNA (n54) and the intrathecal space (n53; Table Keul’s post hoc analysis (Sigma Stat, Jandel Scientific). 1). Injection of estrogen into the IC (n54) produced a

Table 1

Effect of estrogen injection in central autonomic nuclei on baseline parameters

Injection site DMAP (mmHg) DHR (bpm) DVPNA (%) DRSNA (%)

E2 Saline E2 Saline E2 Saline E2 Saline

a a a a a a a a

NTS 22564 22065 23568 22562 4069 3064 23066 22567

a a a a

Amb 2462 2563 24569 23062 55611 4069 2562 2665

a a a a

RVLM 23065 22567 2564 2562 865 562 23564 23065

a a a a

PBN 22065 2965 22565 2764 3065 963 22563 2966

a a

CNA 22565 2865 2663 2865 564 563 22264 2864

IC 2563 2564 2864 2766 562 562 22065 2865

a a a a a a a a

Intrathecal 23066 22565 23065 22565 3065 2065 23567 22566 a

Peak changes (mean6S.E.) in baseline heart rate (HR) and vagal parasympathetic nerve activity (VPNA) occurred 5 min following estrogen injection. Peak changes in mean arterial pressure (MAP) and renal sympathetic nerve activity (RSNA) occurred 30 min following the bilateral microinjection of estrogen in estrogen-replaced (E ) and saline-replaced (Saline) ovariectomized female rats. NTS, nucleus tractus solitarius; Amb, nucleus ambiguus;2 RVLM, rostral ventrolateral medulla; PBN, parabrachial nucleus; CNA, central nucleus of the amygdala; IC, insular cortex. Asterisk indicates significantly different from pre-injection value (ANOVA; P,0.05).

(4)

significant decrease in RSNA only (Table 1). Baseline heart rate was significantly decreased and VPNA was heart rate was significantly decreased and VPNA was significantly increased at 5 min post-estrogen injection into significantly increased at 5 min post-estrogen injection into NTS, Amb (n54) and the intrathecal space (Table 1). All NTS, Amb (n54), PBN and the intrathecal space (Table baseline parameters returned to pre-injection values 60 min 1). All baseline parameters returned to pre-injection values post-estrogen injection. Injection of estrogen into PBN 60 min post-estrogen injection. (n54), CNA (n54) and IC (n54) had no significant effect

on MAP, HR, VPNA or RSNA (Table 1). 3.1.2. Saline-replaced animals

Prior to estrogen injection mean arterial pressure (MAP) 3.2. Effect of estrogen on baroreflex function and heart rate (HR) were 111615 mmHg and 329622

beats per min respectively (n553) which were signifi- 3.2.1. Estrogen-replaced animals

cantly elevated compared to estrogen-replaced animals Testing of the baroreflex with phenylephrine (PE) prior (P,0.05). Baseline values for vagal parasympathetic nerve to central injection of estrogen evoked an increase in MAP activity (VPNA) and renal sympathetic nerve activity (1762 mmHg; n555) accompanied by a reflexive de-(RSNA) were 1663 mV (n553) and 1865 mV (n553) crease in HR (21162 beats per min; n555; Figs. 1A and respectively. Control injections of saline and co-injection 2A). As well, during baroreflex testing VPNA was in-of estrogen / ICI 182,780 into all nuclei and the intrathecal creased 45610% (n555) and RSNA was decreased space had no effect on baseline blood pressure, heart rate 2668% (n555) relative to baseline levels (Fig. 1A). The or nerve activities (data not shown). Baseline mean arterial index of baroreflex function at this time point was pressure and RSNA were significantly decreased 30 min 0.760.05 beats per min / mmHg (n555; Fig. 4A). RSNA following injection of estrogen into NTS (n54), RVLM and the pressor response to PE were significantly at-(n54) and the intrathecal space (n54; Table 1). Baseline tenuated 30 min following injection of estrogen into NTS

Fig. 1. Cardiovascular and autonomic responses to phenylephrine injection (PE; 0.1 mg / kg;↑) before (baseline) and 5 min following the bilateral (100 nl per side) micro-injection of either saline (0.9%; control) or estrogen (0.5mM; estrogen) into the NTS in estrogen-replaced (A) and saline-replaced (B) ovariectomized female rats. Alterations in blood pressure, heart rate and nerve activities measured during baroreflex testing originate at the arrow (↑) and were measured for 1 min following PE injection (time scale bar51 min).

(5)

Fig. 2. Mean changes from baseline in mean arterial pressure (MAP) and heart rate (HR) in response to phenylephrine injection (0.1 mg / kg; i.v.) before and following central injection of saline (0.9%, control) or estrogen (0.5mM; 100 nl per side; estrogen) in estrogen-replaced (A) or saline-replaced (B) ovariectomized female rats. Asterisks indicate significance (P,0.05; ANOVA) from pre-injection value (before). NTS, nucleus tractus solitarius; Amb, nucleus ambiguus; RVLM, rostral ventrolateral medulla; PBN, parabrachial nucleus; CNA, central nucleus of the amygdala; IC, insular cortex; i.t., intrathecal space of the spinal cord.

(n54), RVLM (n54), PBN (n54), CNA (n54) and the (2162 mmHg; n553) accompanied by a reflexive de-intrathecal space (n53; Figs. 2A and 3A). Injection of crease in HR (2962 beats / min; n553; Figs. 1B and 2B). estrogen into the IC (n54) produced a significant decrease As well, during baroreflex testing VPNA was increased in RSNA only (Fig. 3A). VPNA and the reflexive bradycar- 35611% (n553) and RSNA was decreased 2067% (n5 dia to PE injection were significantly increased 30 min 53) relative to baseline levels (Fig. 1B). The index of post-estrogen injection into NTS, Amb (n54), PBN and baroreflex function at this time point was 0.4560.05 beats the intrathecal space (Figs. 2A and 3A). Baroreflex func- per min / mmHg (n553 Fig. 4B). RSNA and the pressor tion was significantly enhanced following injection of response to PE were significantly attenuated 30 min estrogen into all autonomic nuclei and the intrathecal space following injection of estrogen into NTS (n54), RVLM (Fig. 4A). PE-evoked changes in MAP, HR, VPNA and (n54) and the intrathecal space (n54; Figs. 2B and 3B). RSNA returned to pre-injection values 90 min post-es- VPNA and the reflexive bradycardia to PE injection were trogen injection. significantly increased 30 min post-estrogen injection into NTS, Amb (n54) and the intrathecal space (Figs. 2B and 3.2.2. Saline-replaced animals 3B). Baroreflex function was significantly enhanced fol-Testing of the baroreflex with phenylephrine (PE) prior lowing injection of estrogen into NTS, Amb, RVLM and to central injection of estrogen evoked an increase in MAP the intrathecal space (Fig. 4B). PE-evoked changes in

(6)

Fig. 3. Mean changes in vagal (VPNA) and renal (RSNA) efferent nerve activities given as a per cent change from baseline (0%) in response to phenylephrine injection (0.1 mg / kg; i.v.) following central injection of saline (0.9%, control) or estrogen (0.5mM; estrogen) in estrogen-replaced (A) or saline-replaced (B) ovariectomized female rats. Asterisks indicate significance (P,0.05; ANOVA) from pre-injection value (before). NTS, nucleus tractus solitarius; Amb, nucleus ambiguus; RVLM, rostral ventrolateral medulla; PBN, parabrachial nucleus; CNA, central nucleus of the amygdala; IC, insular cortex; i.t., intrathecal space of the spinal cord.

MAP, HR, VPNA and RSNA returned to pre-injection unilateral or outside the intended region produced no values 90 min post-estrogen injection. Injection of estrogen significant effects on baseline parameters nor on PE-into PBN (n54), CNA (n54) and IC (n54) had no evoked changes in the same parameters (data not shown). significant effect on the PE-evoked changes in MAP, HR, As well, the sites of injection for saline and ICI 182,780 VPNA or RSNA (Figs. 2B and 3B). controls have been omitted from Figs. 5 and 6 for clarity. 3.3. Histological verification of cannulae placement

4. Discussion

Figs. 5 and 6 are composite diagrams indicating the

bilateral placement of microinjection cannulae in NTS, In the present study, direct injection of estrogen into RVLM and Amb (Fig. 5A and B) and PBN, CNA and IC central autonomic nuclei produced significant changes in (Fig. 6A,B and C). Data from animals in which both both baseline parameters and PE-evoked changes in the cannulae were not localized to the appropriate nucleus same parameters thereby providing additional evidence to have been omitted. Injections of estrogen which were support a role for estrogen as a central modulator of

(7)

Fig. 4. Effect of central estrogen (0.5 mM; 100 nl / side; estrogen) injection on baroreflex function in estrogen-replaced (A) and saline-replaced (B) ovariectomized female rats. Index of baroreflex function given as a ratio of the peak change in heart rate to blood pressure in response to a single dose of phenylephrine (0.1 mg / kg; i.v.). Control injections of saline (0.9%; control) had no effect on baroreflex function. Asterisks indicate significance (P,0.05; ANOVA) from pre-injection value (before). NTS, nucleus tractus solitarius; Amb, nucleus ambiguus; RVLM, rostral ventrolateral medulla; PBN, parabrachial nucleus; CNA, central nucleus of the amygdala; IC, insular cortex; i.t., intrathecal space of the spinal cord.

cardiovascular autonomic reflexes. These findings are 4.1. Estrogen receptors in the CNS consistent with similar experiments in male rats [20] with

the exception that the cardiovascular and autonomic re- Several studies have demonstrated the presence of sponses to estrogen were increased in magnitude in the diverse estrogen receptor populations within discrete re-estrogen-replaced female rats suggesting that circulating gions of the central nervous system. Autoradiographic estrogen may in fact influence the effects of estrogen studies have localized both estrogen receptor subtypes, within the central nervous system. Consistent with this ERaand ERb, in significant concentrations on cell bodies, finding is the apparent inefficacy of estrogen within axons and terminals of autonomic regulatory nuclei forebrain nuclei of saline-replaced female rats. Among the throughout the neuroaxis from the spinal cord to the scenarios that must be considered are an alteration in insular cortex in several species including the rat and estrogen receptor density or function resulting from re- human [1,11,22]. The presence of estrogen receptors on the duced serum estrogen levels as are observed in ovariec- membrane of cells in these nuclei suggests that estrogen tomized female rats. may be able to produce immediate modulatory effects on

(8)

(i.e. Amb.) or indirectly by an attenuation of sympathetic output at the level of the brain stem (RVLM) or spinal cord [18]. In addition, injection of estrogen into forebrain and midbrain nuclei suggests a possible role for estrogen as a modulator of ascending and / or descending car-diovascular and autonomic information.

4.2. Estrogen vs. saline-replaced female rats

Co-injection of estrogen with the estrogen receptor antagonist ICI 182,780 provides confirmation that the changes in autonomic tone and cardiovascular reflex function observed following estrogen injection are in fact mediated by the activation of estrogen receptors. With this assumption, it is tempting to speculate that the apparent ineffectiveness of estrogen in forebrain nuclei of saline-replaced rats was due to an absence of estrogen receptors. In situ hybridization studies have demonstrated a correla-tion between estrogen receptor mRNA in the hypothalamus and serum estrogen levels in ovariectomized dogs and rats [21,24]. Specifically, estrogen receptor mRNA was found to be very low in the ovariectomized dog or rat and significantly increased following exogenous estrogen ad-ministration [21,24]. The functional significance of these findings remains unclear and warrants further investiga-tion. However, if estrogen receptor populations are in fact down-regulated in response to decreased serum estrogen levels, results in this study would suggest a progressive loss of receptors originating in the forebrain and proceed-ing towards the brain stem.

Another scenario which may be entertained is that of an accelerated clearance or inactivation of estrogen in fore-brain nuclei of saline-replaced female rats. There is evidence to suggest that steroid hormones are rendered

Fig. 5. Serial schematic diagram of coronal sections (6100mm) of the

inactive when sulfated and returned to a bioactive form in

rat brain depicting the approximate placement of microinjection cannulae

the presence of a sulphatase enzyme [9]. Estrone sulfatase

within brain stem nuclei. (A) Rostral ventrolateral medulla (RVLM) and

nucleus ambiguus (Amb). (B) Nucleus tractus solitarius (NTS). Numbers _{activity has been demonstrated in several tissues including} on the right indicate distance from bregma in mm [15]. Injection sites are _{brain [9], however investigation into the correlation} be-depicted unilaterally for purposes of clarity. Open circles represent sites

tween estrone sulfatase activity and serum estrogen levels

of estrogen injection which produced significant changes in either

has been limited to studies involving tissue homogenates

cardiovascular parameters or autonomic tone during baroreflex testing.

from brain and pituitary [2]. The authors found estrone

Closed circles represent sites of estrogen / ICI 182,780 injection in which

phenylephrine-induced changes in cardiovascular parameters and au- _{sulphatase activity to be two-fold higher in the anterior} tonomic tone were not significantly different from pre-injection values. _{pituitary of intact females compared to ovariectomized} Both saline and ICI 182,780 control injection sites have been omitted

female or male rats suggesting that normal female rats are

from the figure for clarity. 12, hypoglossal nucleus; Amb, nucleus

exposed to a larger pool of estrogen in its bioactive form

ambiguus; AP, area postrema; cc, corpus callosum; LPGi, lateral

while ovariectomized females and males possess greater

paragigantocellular nucleus; NTS, nucleus tractus solitarius; py,

pyrami-dal tract; ROb, raphe obscurus nucleus; RVLM, rostral ventrolateral _{quantities of estrogen in an inactive form. No such} nucleus; Sp5C, spinal trigeminal nucleus, caudal part; Sp5I, spinal _{correlation was observed between serum estrogen levels} trigeminal nucleus, interpolar part.

and estrone sulfatase activity in tissue homogenates of the preoptic area, medial basal hypothalamus or cerebral neurotransmission and ultimately influence the integrated cortex [2]. Further investigation into enzyme activity response from cells in these nuclei. Evidence to date would within more circumscribed brain nuclei may provide a suggest that the overall effect of estrogen within autonomic more accurate picture of the differential clearance of nuclei is to enhance parasympathetic tone perhaps by estrogen in autonomic nuclei in response to alterations in direct activation of parasympathetic preganglionic neurons serum estrogen.

(9)

Fig. 6. Serial schematic diagram of coronal sections (6100mm) of the rat brain depicting the approximate placement of microinjection cannulae within the (A) insular cortex (IC), (B) central nucleus of the amygdala (CNA) and (C) parabrachial nucleus (PBN). Numbers on the right indicate distance from bregma in mm [15]. Injection sites are depicted unilaterally for purposes of clarity. Open circles represent sites of estrogen injection which produced significant changes in either cardiovascular parameters or autonomic tone during baroreflex testing. Closed circles represent sites of estrogen / ICI 182,780 injection in which phenylephrine-induced changes in cardiovascular parameters and autonomic tone were not significantly different from pre-injection values. Both saline and ICI 182,780 control injection sites have been omitted from the figure for clarity. 4V, fourth ventricle; BL, basolateral amygdaloid nucleus; CA3, field CA3 of Ammon’s horn; cc, corpus callosum; CNA, central amygdaloid nucleus; CPu, caudate putamen; IC, insular cortex; Ic, internal capsule; LH, lateral hypothalamic area; PBN, parabrachial nucleus; py, pyramidal tract; Sp5O, spinal trigeminal nerve, oral part.

4.3. Central estrogen effects on baroreflex function injection of estrogen on baroreflex function. The baroreflex is the result of a complex yet highly refined communica-Several studies have examined the influence of circulat- tion between several central autonomic nuclei [13]. With ing estrogen on baroreflex function. The sensitivity of the the demonstration of estrogen receptors as well as es-cardiac baroreflex has been shown to change with fluctua- trogenic projections between these nuclei [3], it is surmis-tions in serum estrogen levels such as might occur during able that estrogen plays a role in modulating this car-varying stages of the menstrual cycle [14], following diovascular reflex. Indeed, evidence from this study would ovariectomy [7,8] or during early or late pregnancy [12]. suggest that estrogen may act at several levels to influence Experiments in conscious ovariectomized female rats have baroreflex function. The fact that injections of ICI 182,780 demonstrated an enhanced baroreflex sensitivity following had no effect on baseline or PE-evoked parameters sug-peripheral estrogen administration [4,7,8]. To date, few gests estrogen does not likely exert a tonic influence on studies have investigated the effect of direct central baroreflex function or autonomic tone within the nuclei

(10)

sion and neuropeptidergic characterization of estrogen receptors

included in this study. Further investigation into the

(ERa and ERb) throughout the rat brain: anatomical evidence of

conditions under which estrogen is released into central

distinct roles of each subtype, J. Neurobiol. 36 (1998) 357–378.

nuclei will provide valuable insight into the role of this _{[12] L. Leduc, N. Wasserstrum, T. Spillman, D.B. Cotton, Baroreflex} hormone in central cardiovascular regulation. function in normal pregnancy, Am. J. Obstet. Gynecol. 165 (1991)

886–890.

[13] A.D. Loewy, K.M. Spyer, Central autonomic pathways, in: A.D. Loewy, K.M. Spyer (Eds.), Central Regulation of Autonomic

Acknowledgements

Function, Oxford University Press, New York, 1990, pp. 88–103. [14] C.T. Minson, J.R. Halliwill, T.M. Young, M.J. Joyner, Influence of

This works was funded by a grant ([_{615122) from the} _{the menstrual cycle on sympathetic activity, baroreflex sensitivity,}

Heart and Stroke Foundation of Prince Edward Island. and vascular transduction in young women, Circulation 101 (2000) 862–868.

[15] G. Paxinos, C. Watson, The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 1986.

References _{[16] C.E. Roselli, S.E. Abdelgadir, O.K. Ronnekleiv, S.A. Klosterman,}

Anatomic distribution and regulation of aromatase gene expression [1] A. Amundesson, O. Hermanson, A. Blomqvist, Estrogen receptor- in the rat brain, Biol. Reprod. 58 (1998) 79–87.

like immunoreactivity in the medullary and spinal dorsal horn of the [17] T.M. Saleh, B.J. Connell, Role of 17b-estradiol in the modulation of female rat, Neurosci. Lett. 196 (1996) 25–28. baroreflex sensitivity in male rats, Am. J. Physiol. 275 (1998) [2] P.B. Connolly, J.A. Resko, Estrone sulphatase activity in rat brain R770–R778.

and pituitary: effects of gonadectomy and the estrous cycle, J. [18] T.M. Saleh, B.J. Connell, Centrally mediated effect of 17b-estradiol Steroid Biochem. 33 (1989) 1013–1018. on parasympathetic tone in male rats, Am. J. Physiol. 276 (1999) [3] K.P. Corodimas, J.I. Morrell, Estradiol-concentrating forebrain and R474–R481.

midbrain neurons project directly to the medulla, J. Comp. Neurol. [19] T.M. Saleh, B.J. Connell, 17b-estradiol modulates baroreflex sen-291 (1990) 609–620. sitivity and autonomic tone of female rats, J. Autonom. Nerv. Syst. [4] M. El-Mas, A.A. Abdel-Rahman, Estrogen enhances baroreflex 80 (2000) 148–161.

control of heart rate in conscious ovariectomized rats, Can. J. [20] M.C. Saleh, T.M. Saleh, B.J. Connell, 17b-estradiol modulates Physiol. Pharmacol. 76 (1998) 381–386. baroreflex sensitivity and autonomic tone in the brainstem and spinal [5] M. Esler, The autonomic nervous system and cardiac arrhythmias, cord of male rats, Brain Res. 867 (2000) 200–209.

Clin. Auton. Res. 2 (1992) 33–135. [21] P.J. Shughrue, C.D. Bushnell, D.M. Dorsa, Estrogen receptor mRNA [6] L.N. Gibbons, J. Ciriello, Effect of oestrogen on the cardiovascular in female rat brain during the estrous cycle: a comparison with responses to glutamate stimulation of bed nucleus of the stria ovariectomized females and intact males, Endocrinology 131 (1992) terminalis, Fundam. Clin. Pharmacol. 11 (1997) 105s–108s. 381–388.

[7] X. He, W. Wang, J.T. Crofton, L. Share, Effects of 17b-estradiol on [22] P.J. Shughrue, L.V. Lane, I. Merchenthaler, Comparative distribution sympathetic activity and pressor response to phenylephrine in of estrogen receptor-a and -b mRNA in the rat central nervous ovariectomized rats, Am. J. Physiol. 275 (1998) R1202–R1208. system, J. Comp. Neurol. 388 (1997) 507–525.

[8] X. He, W. Wang, J.T. Crofton, L. Share, Effects of 17b-estradiol on [23] R.B. Simerly, C. Chang, M. Muramatsu, L.W. Swanson, Distribution the baroreflex control of sympathetic activity in conscious ovariec- of androgen and estrogen receptor mRNA-containing cells in the rat tomized rats, Am. J. Physiol. 277 (1999) R493–R498. brain: an in situ hybridization study, J. Comp. Neurol. 294 (1990) [9] R. Hobrick, Steroid sulphotransferases and steroid sulphate sulfat- 76–95.

ases: characteristics and biological roles, Can. J. Biochem. Cell Biol. [24] H. Tani, T. Inaba, S. Matsuyama, Y. Takamori, T. Sawada, Enhance-63 (1985) 1127–1144. ment of estrogen receptor gene expression in the mediobasal [10] S.M. Hyder, C. Chiappetta, L. Murthy, G.M. Stancel, Selective hypothalamus during anestrus in the beagle bitch, Neurosci. Lett.

inhibition of estrogen-regulated gene expression in vivo by the pure 227 (1997) 149–152.

anti-estrogen ICI 182, 780, Cancer Res. 57 (1997) 2547–2549. [25] A.E. Wakeling, M. Dukes, J. Bowler, A potent specific pure anti-[11] N. Laflamme, R.E. Nappi, G. Drolet, C. Labrie, S. Rivest, Expres- estrogen with clinical potential, Cancer Res. 51 (1991) 3867–3873.

(1)

Fig. 2. Mean changes from baseline in mean arterial pressure (MAP) and heart rate (HR) in response to phenylephrine injection (0.1 mg / kg; i.v.) before and following central injection of saline (0.9%, control) or estrogen (0.5mM; 100 nl per side; estrogen) in estrogen-replaced (A) or saline-replaced (B) ovariectomized female rats. Asterisks indicate significance (P,0.05; ANOVA) from pre-injection value (before). NTS, nucleus tractus solitarius; Amb, nucleus ambiguus; RVLM, rostral ventrolateral medulla; PBN, parabrachial nucleus; CNA, central nucleus of the amygdala; IC, insular cortex; i.t., intrathecal space of the spinal cord.

(n

5 4), RVLM (n

5 4), PBN (n

5 4), CNA (n

5 4) and the

(21

6 2 mmHg; n

5 53) accompanied by a reflexive

de-intrathecal space (n

5 3; Figs. 2A and 3A). Injection of

crease in HR (

2

9

6 2 beats / min; n

5 53; Figs. 1B and 2B).

estrogen into the IC (n

5 4) produced a significant decrease

As well, during baroreflex testing VPNA was increased

in RSNA only (Fig. 3A). VPNA and the reflexive bradycar-

35

6 11% (n

5 53) and RSNA was decreased 20

6 7% (n

5 dia to PE injection were significantly increased 30 min

53) relative to baseline levels (Fig. 1B). The index of

post-estrogen injection into NTS, Amb (n

5 4), PBN and

baroreflex function at this time point was 0.45

6 0.05 beats

the intrathecal space (Figs. 2A and 3A). Baroreflex func-

per min / mmHg (n

5 53 Fig. 4B). RSNA and the pressor

tion was significantly enhanced following injection of

response to PE were significantly attenuated 30 min

estrogen into all autonomic nuclei and the intrathecal space

following injection of estrogen into NTS (n

5 4), RVLM

(Fig. 4A). PE-evoked changes in MAP, HR, VPNA and

(n

5 4) and the intrathecal space (n

5 4; Figs. 2B and 3B).

RSNA returned to pre-injection values 90 min post-es-

VPNA and the reflexive bradycardia to PE injection were

trogen injection.

significantly increased 30 min post-estrogen injection into

NTS, Amb (n

5 4) and the intrathecal space (Figs. 2B and

3.2.2. Saline-replaced animals

3B). Baroreflex function was significantly enhanced

fol-Testing of the baroreflex with phenylephrine (PE) prior

lowing injection of estrogen into NTS, Amb, RVLM and

to central injection of estrogen evoked an increase in MAP

the intrathecal space (Fig. 4B). PE-evoked changes in

(2)

110 M.C. Saleh et al. / Brain Research 879 (2000) 105 –114

Fig. 3. Mean changes in vagal (VPNA) and renal (RSNA) efferent nerve activities given as a per cent change from baseline (0%) in response to phenylephrine injection (0.1 mg / kg; i.v.) following central injection of saline (0.9%, control) or estrogen (0.5mM; estrogen) in estrogen-replaced (A) or saline-replaced (B) ovariectomized female rats. Asterisks indicate significance (P,0.05; ANOVA) from pre-injection value (before). NTS, nucleus tractus solitarius; Amb, nucleus ambiguus; RVLM, rostral ventrolateral medulla; PBN, parabrachial nucleus; CNA, central nucleus of the amygdala; IC, insular cortex; i.t., intrathecal space of the spinal cord.

MAP, HR, VPNA and RSNA returned to pre-injection

unilateral or outside the intended region produced no

values 90 min post-estrogen injection. Injection of estrogen

significant effects on baseline parameters nor on

PE-into PBN (n

5 4), CNA (n

5 4) and IC (n

5 4) had no

evoked changes in the same parameters (data not shown).

significant effect on the PE-evoked changes in MAP, HR,

As well, the sites of injection for saline and ICI 182,780

VPNA or RSNA (Figs. 2B and 3B).

controls have been omitted from Figs. 5 and 6 for clarity.

3.3. Histological verification of cannulae placement

4. Discussion

Figs. 5 and 6 are composite diagrams indicating the

bilateral placement of microinjection cannulae in NTS,

In the present study, direct injection of estrogen into

RVLM and Amb (Fig. 5A and B) and PBN, CNA and IC

central autonomic nuclei produced significant changes in

(Fig. 6A,B and C). Data from animals in which both

both baseline parameters and PE-evoked changes in the

cannulae were not localized to the appropriate nucleus

same parameters thereby providing additional evidence to

have been omitted. Injections of estrogen which were

support a role for estrogen as a central modulator of

(3)

Fig. 4. Effect of central estrogen (0.5 mM; 100 nl / side; estrogen) injection on baroreflex function in estrogen-replaced (A) and saline-replaced (B) ovariectomized female rats. Index of baroreflex function given as a ratio of the peak change in heart rate to blood pressure in response to a single dose of phenylephrine (0.1 mg / kg; i.v.). Control injections of saline (0.9%; control) had no effect on baroreflex function. Asterisks indicate significance (P,0.05; ANOVA) from pre-injection value (before). NTS, nucleus tractus solitarius; Amb, nucleus ambiguus; RVLM, rostral ventrolateral medulla; PBN, parabrachial nucleus; CNA, central nucleus of the amygdala; IC, insular cortex; i.t., intrathecal space of the spinal cord.

cardiovascular autonomic reflexes. These findings are

4.1. Estrogen receptors in the CNS

consistent with similar experiments in male rats [20] with

the exception that the cardiovascular and autonomic re-

Several studies have demonstrated the presence of

sponses to estrogen were increased in magnitude in the

diverse estrogen receptor populations within discrete

re-estrogen-replaced female rats suggesting that circulating

gions of the central nervous system. Autoradiographic

estrogen may in fact influence the effects of estrogen

studies have localized both estrogen receptor subtypes,

within the central nervous system. Consistent with this

ER

a

and ER

b

, in significant concentrations on cell bodies,

finding is the apparent inefficacy of estrogen within

axons and terminals of autonomic regulatory nuclei

forebrain nuclei of saline-replaced female rats. Among the

throughout the neuroaxis from the spinal cord to the

scenarios that must be considered are an alteration in

insular cortex in several species including the rat and

estrogen receptor density or function resulting from re-

human [1,11,22]. The presence of estrogen receptors on the

duced serum estrogen levels as are observed in ovariec-

membrane of cells in these nuclei suggests that estrogen

tomized female rats.

may be able to produce immediate modulatory effects on

(4)

112 M.C. Saleh et al. / Brain Research 879 (2000) 105 –114

(i.e. Amb.) or indirectly by an attenuation of sympathetic

output at the level of the brain stem (RVLM) or spinal

cord [18]. In addition, injection of estrogen into forebrain

and midbrain nuclei suggests a possible role for estrogen

as a modulator of ascending and / or descending

car-diovascular and autonomic information.

4.2. Estrogen vs. saline-replaced female rats

Co-injection of estrogen with the estrogen receptor

antagonist ICI 182,780 provides confirmation that the

changes in autonomic tone and cardiovascular reflex

function observed following estrogen injection are in fact

mediated by the activation of estrogen receptors. With this

assumption, it is tempting to speculate that the apparent

ineffectiveness of estrogen in forebrain nuclei of

saline-replaced rats was due to an absence of estrogen receptors.

In situ hybridization studies have demonstrated a

correla-tion between estrogen receptor mRNA in the hypothalamus

and serum estrogen levels in ovariectomized dogs and rats

[21,24]. Specifically, estrogen receptor mRNA was found

to be very low in the ovariectomized dog or rat and

significantly increased following exogenous estrogen

ad-ministration [21,24]. The functional significance of these

findings remains unclear and warrants further

investiga-tion. However, if estrogen receptor populations are in fact

down-regulated in response to decreased serum estrogen

levels, results in this study would suggest a progressive

loss of receptors originating in the forebrain and

proceed-ing towards the brain stem.

Another scenario which may be entertained is that of an

accelerated clearance or inactivation of estrogen in

fore-brain nuclei of saline-replaced female rats. There is

evidence to suggest that steroid hormones are rendered

Fig. 5. Serial schematic diagram of coronal sections (6100mm) of the

inactive when sulfated and returned to a bioactive form in

rat brain depicting the approximate placement of microinjection cannulae

the presence of a sulphatase enzyme [9]. Estrone sulfatase

within brain stem nuclei. (A) Rostral ventrolateral medulla (RVLM) and

nucleus ambiguus (Amb). (B) Nucleus tractus solitarius (NTS). Numbers

_{activity has been demonstrated in several tissues including}

on the right indicate distance from bregma in mm [15]. Injection sites are

_{brain [9], however investigation into the correlation}

be-depicted unilaterally for purposes of clarity. Open circles represent sites

tween estrone sulfatase activity and serum estrogen levels

of estrogen injection which produced significant changes in either

has been limited to studies involving tissue homogenates

cardiovascular parameters or autonomic tone during baroreflex testing.

from brain and pituitary [2]. The authors found estrone

Closed circles represent sites of estrogen / ICI 182,780 injection in which

phenylephrine-induced changes in cardiovascular parameters and au-

_{sulphatase activity to be two-fold higher in the anterior}

tonomic tone were not significantly different from pre-injection values.

_{pituitary of intact females compared to ovariectomized}

Both saline and ICI 182,780 control injection sites have been omitted

female or male rats suggesting that normal female rats are

from the figure for clarity. 12, hypoglossal nucleus; Amb, nucleus

exposed to a larger pool of estrogen in its bioactive form

ambiguus; AP, area postrema; cc, corpus callosum; LPGi, lateral

while ovariectomized females and males possess greater

paragigantocellular nucleus; NTS, nucleus tractus solitarius; py,

pyrami-dal tract; ROb, raphe obscurus nucleus; RVLM, rostral ventrolateral

_{quantities of estrogen in an inactive form. No such}

nucleus; Sp5C, spinal trigeminal nucleus, caudal part; Sp5I, spinal

_{correlation was observed between serum estrogen levels}

trigeminal nucleus, interpolar part.

and estrone sulfatase activity in tissue homogenates of the

preoptic area, medial basal hypothalamus or cerebral

neurotransmission and ultimately influence the integrated

cortex [2]. Further investigation into enzyme activity

response from cells in these nuclei. Evidence to date would

within more circumscribed brain nuclei may provide a

suggest that the overall effect of estrogen within autonomic

more accurate picture of the differential clearance of

nuclei is to enhance parasympathetic tone perhaps by

estrogen in autonomic nuclei in response to alterations in

direct activation of parasympathetic preganglionic neurons

serum estrogen.

(5)

4.3. Central estrogen effects on baroreflex function

injection of estrogen on baroreflex function. The baroreflex

is the result of a complex yet highly refined

communica-Several studies have examined the influence of circulat-

tion between several central autonomic nuclei [13]. With

ing estrogen on baroreflex function. The sensitivity of the

the demonstration of estrogen receptors as well as

es-cardiac baroreflex has been shown to change with fluctua-

trogenic projections between these nuclei [3], it is

surmis-tions in serum estrogen levels such as might occur during

able that estrogen plays a role in modulating this

car-varying stages of the menstrual cycle [14], following

diovascular reflex. Indeed, evidence from this study would

ovariectomy [7,8] or during early or late pregnancy [12].

suggest that estrogen may act at several levels to influence

Experiments in conscious ovariectomized female rats have

baroreflex function. The fact that injections of ICI 182,780

demonstrated an enhanced baroreflex sensitivity following

had no effect on baseline or PE-evoked parameters

sug-peripheral estrogen administration [4,7,8]. To date, few

gests estrogen does not likely exert a tonic influence on

studies have investigated the effect of direct central

baroreflex function or autonomic tone within the nuclei

(6)

114 M.C. Saleh et al. / Brain Research 879 (2000) 105 –114

sion and neuropeptidergic characterization of estrogen receptors

included in this study. Further investigation into the

(ERa and ERb) throughout the rat brain: anatomical evidence of

conditions under which estrogen is released into central

distinct roles of each subtype, J. Neurobiol. 36 (1998) 357–378.

nuclei will provide valuable insight into the role of this

_{[12] L. Leduc, N. Wasserstrum, T. Spillman, D.B. Cotton, Baroreflex}

hormone in central cardiovascular regulation.

function in normal pregnancy, Am. J. Obstet. Gynecol. 165 (1991)

886–890.

[13] A.D. Loewy, K.M. Spyer, Central autonomic pathways, in: A.D. Loewy, K.M. Spyer (Eds.), Central Regulation of Autonomic

Acknowledgements

Function, Oxford University Press, New York, 1990, pp. 88–103. [14] C.T. Minson, J.R. Halliwill, T.M. Young, M.J. Joyner, Influence of

This works was funded by a grant (

[

_{615122) from the}

_{the menstrual cycle on sympathetic activity, baroreflex sensitivity,}

Heart and Stroke Foundation of Prince Edward Island.

and vascular transduction in young women, Circulation 101 (2000)

862–868.

[15] G. Paxinos, C. Watson, The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 1986.

References

_{[16] C.E. Roselli, S.E. Abdelgadir, O.K. Ronnekleiv, S.A. Klosterman,} Anatomic distribution and regulation of aromatase gene expression [1] A. Amundesson, O. Hermanson, A. Blomqvist, Estrogen receptor- in the rat brain, Biol. Reprod. 58 (1998) 79–87.

inhibition of estrogen-regulated gene expression in vivo by the pure 227 (1997) 149–152.