C

Original article 1001

Differential muscarinic recepto

r gene expression levelsin the ventral medulla of spontaneously hypertensive andWistar–Kyoto rats: role in sympathetic baroreflex functionNatasha N. Kumara, Jorgen Fergusonb, James R. Padleya, Paul M. Pilowskya

and Ann K. Goodchilda

Objective We demonstrated previously that central

muscarinic cholinergic receptor (mAChR) activation

increased splanchnic sympathetic nerve activity and

sympathetic baroreflex function via activation of mAChR in

the rostral ventrolateral medulla (RVLM), and we found that

some RVLM bulbospinal neurons contain muscarinic M2R

mRNA. Here, we examined the gene expression, cellular

distribution and functional role of muscarinic receptors in

the RVLM in spontaneously hypertensive rats (SHR)

compared with Wistar–Kyoto (WKY) rats.

Method and results Using the sensitive technique of

quantitative real time reverse transcriptase-PCR, M2R

mRNA level was elevated two-fold (P < 0.05) and M4R mRNA

was downregulated two-fold (P < 0.001), with all other

receptors expressed at similar levels, in the rostral ventral

medulla of SHR compared with WKY. Bulbospinal, but not

catecholaminergic neurons, in the RVLM expressed M2R

mRNA (M2RR), and similar numbers were found in the

RVLM of SHR and WKY. Could elevated M2R within

individual neurons or enhanced presynaptic activity reflects

enhanced cholinergic effects in the RVLM? Activation of

central mAChR using oxotremorine evoked a larger

increase in mean arterial pressure in SHR compared with

WKY (P < 0.01); however, oxotremorine-induced increases

in splanchnic sympathetic nerve activity, and sympathetic

baroreflex function were similar in SHR and WKY.

Conclusion These data indicate that enhanced pressor

responses in SHR, following centrally mediated mAChR

opyright © Lippincott Williams & Wilkins. Unauth

0263-6352 � 2009 Wolters Kluwer Health | Lippincott Williams & Wilkins

activation, are not associated with RVLM-mediated

constriction of the splanchnic circulation or effects on the

sympathetic baroreflex, but could reflect modified mAChR

gene expression elsewhere. RVLM-dependent splanchnic

sympathetic nerve activity effects, evoked by mAChR

activation, are not mediated by the differential M2/M4

receptor mRNA levels identified in SHR compared with

WKY. J Hypertens 27:1001–1008 Q 2009 Wolters Kluwer

Health | Lippincott Williams & Wilkins.

Journal of Hypertension 2009, 27:1001–1008

Keywords: baroreceptor reflex, blood pressure, cholinergic, hypertension,rostral ventrolateral medulla, sympathetic nerve activity

Abbreviations: ACh, acetylcholine; C1, adrenergic catecholaminergic cellgroup; Ct, cycle threshold; CTB, cholera toxin B; DIG, digoxigenin; -ir,immunoreactive; ISH, in situ hybridization; MA, methylatropine; MAP, meanarterial pressure; OXO, oxotremorine; RVLM, rostral ventrolateral medulla;RVM, rostral ventral medulla; SBP, systolic blood pressure; SCOP,scopolamine; SHR, spontaneously hypertensive rats; SNA, sympatheticnerve activity; WKY, Wistar–Kyoto rat

aAustralian School of Advanced Medicine, Macquarie University and bFaculty ofMedicine, Royal North Shore Hospital, The University of Sydney, New SouthWales, Australia

Correspondence to Associate Professor Ann K. Goodchild, Australian School ofAdvanced Medicine, Level 1 Dow Corning Building, Macquarie University, NSW2109, AustraliaTel: +61 2 9850 4016; fax: +61 2 9850 4010;e-mail: ann.goodchild@vc.mq.edu.au

Received 6 June 2008 Revised 1 November 2008Accepted 23 December 2008

IntroductionCholinergic cell groups in the midbrain, hypothalamus

and forebrain provide extensive projections throughout

the brain [1–3]. Acetylcholine (ACh) plays a role in the

central control of cardiovascular regulation [3–7], particu-

larly within the rostral ventrolateral medulla (RVLM), a

site crucial to the tonic and reflex control of arterial

pressure [8]. Spinally projecting (bulbospinal) RVLM

neurons are the main origin of spinal sympathetic outflow

and are divided into those that contain, or lack catechol-

amine biosynthetic enzymes, the C1 or non-C1 neurons,

respectively [9,10]. ACh is synthesized, stored and

released within the RVLM, and muscarinic receptors

are present there [3,11–13]. Activation of muscarinic

receptors in the RVLM elicits pressor and sympathoex-

citatory responses [14–17]. Almost, all centrally evoked

ACh-mediated sympathetic and cardiovascular responses

can be blocked by RVLM muscarinic receptor blockade

[3].

ACh in the RVLM may play a role in the pathogenesis of

hypertension [12,18–20]. ACh content [21], release of

ACh [19] and choline acetyltransferase activity [22] in the

RVLM are all increased in spontaneously hypertensive

rats (SHR) compared with normotensive rats. Further-

more, inhibition of acetylcholinesterase or muscarinic

receptors in the RVLM of SHR evokes larger hyperten-

sion or hypotension, respectively, [19] than in normoten-

sive animals. These data together suggest that cholin-

ergic transmission is enhanced within the RVLM of SHR.

orized reproduction of this article is prohibited.

DOI:10.1097/HJH.0b013e3283282e5c

C

1002 Journal of Hypertension 2009, Vol 27 No 5

Of the five known subtypes of muscarinic (M1–M5)

receptors [23,24], early pharmacological studies impli-

cated the M2 receptor in mediating the pressor effects

evoked in the RVLM, as AFDX-116, a muscarinic recep-

tor agonist, reversed the excitatory effects of physostig-

mine [15,25]. In SHR, there is more M2 receptor mRNA

in a large brainstem area that included the RVLM com-

pared with normotensive controls [26]. We have recently

demonstrated that a subpopulation of bulbospinal, non-

C1 neurons in the RVLM of normotensive animals con-

tains M2 receptor mRNA [3]. Activation of muscarinic

receptors in the RVLM in normotensive rats evokes

substantial facilitation of the sympathetic baroreceptor

reflex [3]. Furthermore, baroreceptor activation induces

an increase in the release of ACh in the RVLM [27].

Thus, we tested the hypothesis that M2 receptor gene

expression levels are increased in the RVLM of SHR and

are associated with an increased number of bulbospinal

cells containing M2 receptor mRNA and increased oxo-

tremorine (OXO)-induced cardiovascular function. We

first determined the relative amounts of the five types

of muscarinic receptors in the rostral ventral medulla

(RVM) in SHR and WKY using quantitative real time

reverse transcriptase-PCR (RT-PCR). Secondly, we

compared the number of bulbospinal (C1 and non-C1)

neurons in the RVLM that contain M2R mRNA in SHR

and WKY, using combined in-situ hybridization (ISH)

and immunohistochemistry. Thirdly, we examined the

cardiovascular effects of central muscarinic receptor acti-

vation and determined sympathetic baroreflex function

in SHR compared with WKY. Finally, we asked whether

or not constitutive activation of muscarinic receptors

contributes to baseline baroreflex control of blood pres-

sure in these rat lines.

Materials and methodAnimalsAnimals were bred and housed in a ratio of 12 : 12 h light–

dark environment, with access to food and water adlibitum. Experiments were conducted in accordance with

the Code of Practice for the Care and Use of Animals for

Scientific Purposes and approved by the Animal Care and

Ethics Committee of Royal North Shore Hospital and the

University of Technology, Sydney, Australia.

Phenotype validation: spontaneously hypertensive andWistar–Kyoto ratsSystolic blood pressure (SBP) was determined by tail

cuff plethysmography under light (1–2%) halothane

anaesthesia (in O2, Fluothane; Zeneca Pharmaceuticals,

Macclesfield, UK). The SBP between male SHR

(163� 5 mmHg) and WKY (108� 4 mmHg) was statisti-

cally different (P< 0.0001, Student’s t-test) in each group.

Three groups of animals (20–22 weeks old) were used to

determine relative M1–M5 receptor mRNA levels in the

RVM (SHR, n¼ 6; WKY, n¼ 6), the relative numbers of

opyright © Lippincott Williams & Wilkins. Unautho

M2 receptor mRNA containing bulbospinal and catechol-

aminergic neurons (SHR, n¼ 6; WKY, n¼ 5) and tonic and

reflex cardiovascular effects of activation of central muscar-

inic receptors by oxotremorine (SHR, n¼ 5; WKY, n¼ 4).

Molecular biological experimentsTissue extraction and RNA isolation for PCR

Rats were anaesthetized with sodium pentobarbital

(70 mg/kg, i.p.), then transcardially perfused with ice-cold

sterile 0.9% NaCl. The RVM (Bregma �11.8 to �12.5;

located ventral to the nucleus ambiguus, immediately

caudal to the facial nucleus) was isolated from two

consecutive 350 mm coronal brainstem sections cut frozen

on a vibrating microtome (Leica VT1000S; Gladesville,

New South Wales, Australia) and cerebellum (posterior

lobe), were immediately dissected into RNAlater cryo-

protectant reagent (Ambion Inc., Austin, Texas, USA)

and stored at �708C until RNA isolation.

RNA was extracted using the SV total RNA isolation

system (Promega, Madison, Wisconsin, USA), and a

DNase procedure was included, as described previously

[28]. The integrity and concentration of total RNA was

determined using a spectrophotometer (Beckman-Coul-

ter DU-800; Fullerton, California, USA) and considered

suitable if the A260/A280 nm reading was more than 1.8.

Total RNA (150 ng) was reverse transcribed in 40 ml

reactions, using ImProm-II Reverse Transcription Sys-

tem, oligo dT primers and RNasin ribonuclease inhibitor

(Promega).

Real time quantitative PCR

Real time quantitative PCR experiments were run using

previously described M1-M5 receptor and glyceralde-

hyde-3-phosphate dehydrogenase (GAPDH) specific pri-

mers and PCR conditions [3]. Forty cycles of real time

PCR were carried out with an initial denaturing step at

958C for 10 min, followed by 958C with a 20 s hold,

60–638C with a 20 s hold (annealing) and then 728C with

a 20 s hold. Fluorescence data were acquired at the end of

the extension step at 858C for 10 s, to eliminate primer

dimer artifact.

All samples were analyzed in duplicate, and RNA

samples that had not been reverse transcribed were

included as negative controls to confirm the absence of

contaminating genomic DNA. Amplicon specificity and

reproducibility was confirmed by ethidium bromide agar-

ose (2%) gel electrophoresis and melt curve analysis [29].

Each muscarinic receptor mRNA level was expressed

relative to the level of the internal control reference gene

GAPDH, using relative standard curve method. A stan-

dard curve was derived for each gene product within the

cycle threshold range for cDNA samples as follows: a

stock of double purified PCR product (Qiagen QiaQuick

PCR purification kit; Doncaster, Victoria, Australia) for

rized reproduction of this article is prohibited.

C

Muscarinic receptor function in hypertensive rats Kumar et al. 1003

each gene was expressed in arbitrary units, serially

diluted and logarithms (base 10) of concentrations were

plotted against cycle threshold for accurate determination

of PCR efficiency and relative mRNA copy number

between unknown samples.

Statistical analysis for real time PCR

Mean cycle threshold (Ct, calculated by the second

derivative maximum method) values for SHR and

WKY samples in PCR replicas were converted to arbitrary

copy number using the standard curve equation calcu-

lated by linear regression analysis (Corbett rotogene soft-

ware; Corbett Life Science, New South Wales, Australia).

For each cDNA sample, copy number for each muscarinic

receptor was normalized to copy number for GAPDH.

PCR products for each muscarinic receptor were recov-

ered enzymatically from agarose gels using QiaQuick gel

extraction kit (Qiagen, Doncaster, Victoria, Australia),

and the cDNA was sequenced using the ABI Prism

3730 platform (SUPAMAC, Camperdown, New South

Wales, Australia).

Retrograde labeling of spinally projecting rostral

ventrolateral medulla neurons

SHR (n¼ 6) and WKY (n¼ 5) rats were anaesthetized with

sodium pentobarbital (70 mg/kg i.p.), and cholera toxin B

(CTB, 1% in 200 nl; Sapphire Biosciences, New South

Wales, Australia) was injected bilaterally into the upper

thoracic spinal cord (T1-T2), as described previously [30].

Three days later, rats were re-anaesthetized and transcar-

dially perfused with 0.9% NaCl, followed by freshly pre-

pared 4% paraformaldehyde/0.1 mol/l phosphate buffer.

Combined in-situ hybridization and immunohistochemistry

Tissues were processed under RNase-free conditions,

using digoxigenin (DIG)-labelled sense and antisense

riboprobes targeting the rat M2 receptor [3], as described

previously [31]. Spinal cord injection sites were later

verified by immunohistochemistry for CTB using avidin

horseradish peroxidase and a nickel-diaminobenzidine

reaction [32].

Multiple labeling was carried out on brainstem sections in

order to identify CTB immunoreactive (-ir), tyrosine

hydroxylase-ir and M2R mRNA containing (M2Rþ)

neurons in the VLM as previously described [31]. Brain-

stem sections were incubated with primary antibodies

(48C for 48 h): alkaline phosphatase-conjugated sheep

anti-DIG (1 : 1000, Roche Inc., Nutley, New Jersey,

USA), mouse antityrosine hydroxylase (1 : 2000; Sigma-

Aldrich, Castle Hill, New South Wales, Australia) and

rabbit anti-CTB (1 : 5000; Virostat Inc., Westbrook,

Maine, USA), followed by incubation with fluorophore-

conjugated secondary antisera in 2% normal horse serum

(48C for 24 h): 7-amino-4-methylcoumarin-3-acetic acid-

NHS ester (AMCA) -conjugated donkey antimouse

IgG (1 : 500, Jackson Immunoresearch, West Grove,

opyright © Lippincott Williams & Wilkins. Unauth

Pennsylvania, USA) and fluorescein-5-isothiocyanate

(FITC)-conjugated donkey antisheep IgG (1 : 500;

Jackson Immunoresearch). The detection steps for DIG-

labelled ISH neurons were as previously described [31].

Cell counts and data analysis

Sections were mounted on glass slides and cover-slipped

with Vectashield (Vector Laboratories, Burlingame,

California, USA). Anatomical markers, including the

poles of the inferior olive and facial nucleus, were used

to indicate Bregma level according to the rat brain atlas

[33]. Images were acquired using a fluorescence micro-

scope (Zeiss Z1, Gottingen, Germany), processed with

Zeiss Axiovision software (Zeiss Z1) and assembled with

CorelDraw (Version 12). Cell counts were taken from

sections at 200 mm intervals through the VLM. Data are

expressed as mean�SEM.

Physiological experimentsGeneral procedure

The experiments described here are similar to those

carried out in Sprague–Dawley rats described previously

[34]. Briefly, rats were anaesthetized (urethane 1.3 g/kg),

paralyzed (pancuronium dibromide 0.8 mg induction and

0.4 mg/h maintenance, i.v.) and ventilated (Ugo Basile

rodent ventilator, Comerio, Italy) with 100% O2 mixed

with room air. The carotid artery and jugular vein were

catheterized for arterial pressure measurement and drug

administration, respectively. The left greater splanchnic

preganglionic nerve was isolated, cut and recorded.

Rectal temperature was monitored and maintained

between 36.5 and 37.58C, with a thermoregulated blanket

and infrared lamp.

Sympathetic baroreceptor reflex function

To determine the effects of central activation of muscar-

inic receptors on the baroreceptor reflex, all peripheral

muscarinic receptors were blocked with methylatropine

(2 mg/kg, i.v.), injected 5 min prior to the administration

of the broad-spectrum muscarinic receptor agonist OXO

(0.2 mg/kg, i.v.), as described previously [3,34]. Barore-

ceptors were activated (phenylephrine – 10 mg/kg, i.v.)

and unloaded (sodium nitroprusside – 10 mg/kg, i.v.) at

5 and 30 min post-OXO injection. To determine whether

or not muscarinic receptor activation played a role in

setting baseline baroreceptor function, scopolamine

(SCOP, 1 mg/kg, i.v.) was injected intravenously, and

baroreceptors were activated and unloaded. These

responses were compared with those evoked at com-

mencement of the experiment.

Data analysisChanges in mean arterial pressure (MAP), sympathetic

nerve activity (SNA) and heart rate (HR) were expressed as

mean�SEM. Sympathetic baroreceptor function curves

were determined, before and after OXO administration.

Data were normalized, curve fitted with nonlinear

orized reproduction of this article is prohibited.

C

1004 Journal of Hypertension 2009, Vol 27 No 5

Fig. 1

mRNA levels of muscarinic receptors 1–5 (M1R–M5R) inspontaneously hypertensive and Wistar–Kyoto rats. (a) M1–M5receptors (lanes 1–5, respectively) and GAPDH (lane 6) PCR productswere amplified from RVM cDNA and visualized on an ethidium bromidestained 2% agarose gel. pGEM DNA marker is shown in lane 7, and thenumbers on the right indicate the base pair size of DNA markerfragments. (b) Relative mRNA levels of the M1–M5R in RVM andcerebellum in SHR (black columns) compared with age-matched WKY(grey columns) rats. M2R mRNA is elevated, and M4R mRNA isreduced in the RVM of SHR compared with WKY. Similar levels ofM1R, M3R and M5R were seen in the tissues investigated. Values aremean�SEM, n¼6 per group. GAPDH, glyceraldehyde-3-phosphatedehydrogenase; RVM, rostral ventral medulla; SHR, spontaneouslyhypertensive rats; WKY, Wistar–Kyoto rats.

regression in the form of a sigmoidal dose–response curve

and theoretical baroreflex curves were then constructed.

Baroreflex gain was calculated from curves of the first

derivative of the baroreceptor function curve. Data were

analysed using the Student’s t-test, with a significance

level achieved when P< 0.05.

ResultsQualitative M1-M5 muscarinic receptor mRNAexpression within the rostral ventral medulla andcerebellumAll five muscarinic receptor subtypes were detected in

the RVM and cerebellum of both SHR and WKY rats. To

confirm specificity, melt curves were analyzed, and all

PCR amplicons were resolved on a 2% agarose gel

(Fig. 1a). cDNA sequencing confirmed that PCR ampli-

cons conformed to their United States National Center

for Biology Information GenBank accession numbers, as

described previously [3].

M2R mRNA is elevated and M4R mRNA is decreased inthe rostral ventral medulla of spontaneouslyhypertensive compared with Wistar–Kyoto ratsIn the RVM of SHR, there was a two-fold increase in

M2R mRNA level (P< 0.05) and a two-fold decrease in

M4R mRNA level (P< 0.001) when compared with WKY

(Fig. 1b). No significant differences were detected in

mRNA levels of M1R, M3R or M5R in the RVM or of any

muscarinic receptor subtypes in the cerebellum of SHR

compared with WKY (Fig. 1b).

M2R mRNA is expressed in bulbospinal non-C1 neuronsin the rostral ventrolateral medulla of spontaneouslyhypertensive and Wistar–Kyoto ratsFigure 2a shows an example of retrogradely labelled

(CTB-ir), M2R mRNA containing (M2Rþ) neurons

amongst tyrosine hydroxylase-ir neurons in the RVLM.

A similar number of C1 (tyrosine hydroxylase-ir, Fig. 2b),

bulbospinal (CTB-ir, SHR 232� 14 cells, n¼ 6; WKY

239� 21 cells, n¼ 5, Fig. 2c) and C1 bulbospinal neurons

(SHR 117� 14 cells, 49% of total CTB-ir, n¼ 6; WKY

117� 16 cells, 49% of total CTB-ir, n¼ 5) were quantified

in the RVLM in SHR and WKY. No C1 neurons and,

therefore, no bulbospinal C1 neurons were M2Rþ. In

contrast, 13% of all bulbospinal neurons in the RVLM

were M2Rþ in SHR and WKY (SHR 171/1391 CTB-ir

cells, n¼ 6; WKY 161/1196 CTB-ir cells, n¼ 5). Approxi-

mately, 25% of bulbospinal, non-C1 neurons expressed

M2R mRNA in both SHR (115� 15 cells, 25%) and WKY

(112� 10 cells, 27%), and this was not significantly

different between strains.

Tonic and reflex cardiovascular effects of activation ofcentral muscarinic receptors in spontaneouslyhypertensive and Wistar–Kyoto ratsFigure 3a shows in a single study that following barore-

ceptor activation and unloading (performed twice, aster-

opyright © Lippincott Williams & Wilkins. Unautho

isks), methylatropine and then OXO were administered.

OXO evoked a large increase in SNA and arterial pressure.

The sympathetic effects of baroreceptor loading and

unloading were greatly enhanced 5 min following OXO,

but this returned to control levels 30 min after OXO. SCOP

did not affect the control responses of baroreceptor

loading and unloading. Grouped data (Figs 3b and 4)

show that OXO evoked a rapid increase in MAP that

was greater in SHR (86.2� 3.2 mmHg) compared with

WKY (52.3� 6.4 mmHg) [P¼ 0.007, Fig. 3b(i)]. The peak

rise in MAP occurred within 1 min of injection in both

strains, and MAP returned to preinjection levels after 25–

30 min. In contrast, OXO evoked a similar level of splanch-

nic sympathoexcitation in both SHR (114.5� 16.1%)

and WKY (100.0� 14.4%) [Fig. 3b(ii)]. OXO evoked a

rized reproduction of this article is prohibited.

C

Muscarinic receptor function in hypertensive rats Kumar et al. 1005

Fig. 2

Bulbospinal, catecholaminergic and M2R mRNA containing (M2Rpositive) neurons in the rostral ventrolateral medulla in spontaneouslyhypertensive and Wistar–Kyoto rats. (a) (i): Inset in low-power image (i)is magnified to show M2R positive (ii, black), TH-ir (iii, red) andbulbospinal (CTB-ir, green) neurons in the same field. Dual labeledneurons are indicated (white arrows). Scales bars¼500 mm (i) and20 mm (ii,iii and iv). (b) The number and distribution of C1 neurons(TH-ir) in the RVLM was similar in SHR and WKY. M2R mRNA wasabsent in C1 neurons in the RVLM of SHR and WKY. (c) The numberand distribution of bulbospinal (CTB-ir) neurons was similar in theRVLM of SHR and WKY. The number of M2R-positive, bulbospinalneurons in the RVLM was not significantly different in SHR and WKY.M2R-positive, bulbospinal neurons constituted almost 30% of non-C1bulbospinal neurons in the RVLM, in both SHR and WKY. CTB, choleratoxin B; CVLM, caudal ventrolateral medulla; -ir, immunoreactive; RVLM,rostral ventrolateral medulla; SHR, spontaneously hypertensive rats;TH, tyrosine hydroxylase; WKY, Wistar–Kyoto rats.

Fig. 3

Cardiovascular and sympathetic effects of central muscarinic receptoractivation and inhibition in spontaneously hypertensive and Wistar–Kyoto rats. (a) Original trace of a physiological experiment in a WKY ratshowing HR, Av SNA and AP. Intravenous injections of the peripherallyacting muscarinic antagonist MA, muscarinic agonist OXO andmuscarinic antagonist SCOP are indicated by arrows. Asterisks (�)show times at which baroreceptors were loaded (phenylephrine, i.v.)and unloaded (sodium nitroprusside, i.v.). OXO increased HR, SNA andAP. The baroreflex effects on SNA were enhanced 5 min after OXOadministration compared with before or 30 min following OXO. SCOPhad no effect on sympathetic baroreflex function. (b) Grouped datashowing that OXO evoked greater pressor responses in SHR (black)than in WKY (grey, P<0.01), but had similar effects on SNA and HR.AP, arterial pressure; AU, astronomical unit; Av SNA, averagedsympathetic nerve activity; bpm, beats per minute; HR, heart rate; MA,methylatropine; MAP, mean arterial pressure; OXO, oxotremorine;SCOP, scopolamine; SHR, spontaneously hypertensive rats; sSNA,splanchnic sympathetic nerve activity; WKY, Wistar–Kyoto rats.

sympathetically mediated increase in HR that was not

significantly different between SHR [66.8� 6.7 beats/min

(n¼ 6)] and WKY [43.3� 13.9 beats/min (n¼ 4)] [P¼ 0.19,

Fig. 3b(iii)].

The gain of the sympathetic baroreflex for SHR and

WKY were similar at baseline (Fig. 4a,b), although shifted

to a higher level of MAP (see below). OXO significantly

increased the range and gain of the baroreflex in both

strains and significantly shifted the curve to a higher level

of MAP in each strain (Fig. 4a curves control versus OXO,

opyright © Lippincott Williams & Wilkins. Unauth

WKY P< 0.001; SHR P< 0.001), but there was no sig-

nificant difference in the degree of shift between the

strains. SCOP failed to evoke any change in the range or

gain of the sympathetic baroreflex in either SHR or WKY

compared with the control curve evoked at the beginning

of the experiment (Fig. 4c).

DiscussionThe main findings of this study are first that whilst all five

muscarinic receptor subtypes are expressed in the RVM,

orized reproduction of this article is prohibited.

Copyright © Lippincott Williams & Wilkins. Unautho

1006 Journal of Hypertension 2009, Vol 27 No 5

Fig. 4

Effect of central muscarinic receptor activation on sympatheticbaroreflex function in spontaneously hypertensive (black lines) andWistar–Kyoto rats (grey lines). (a) Sympathetic baroreflex functioncurves in SHR and WKY before (basal; solid lines) and 5 min after(dashed lines) central OXO administration. OXO evoked a largerpressor response in SHR compared with WKY. OXO facilitatedsympathetic baroreflex function in both SHR and WKY. (b) Graphsshowing the first derivative of the curves shown in (a). Sympatheticbaroreflex gain was significantly enhanced, following the centraladministration of OXO in SHR and WKY. However, there are nosignificant differences in gain between SHR and WKY either before orafter OXO administration. Values are mean�SEM. (c) Derivedbaroreflex function curves in SHR and WKY before any drugs [solidlines as shown in (b)] and following SCOP administration (hatchedlines). Under these conditions, SCOP has no effect on baroreflexfunction in SHR or WKY. OXO, oxotremorine; SCOP, scopolamine;SHR, spontaneously hypertensive rats; SNA, sympathetic nerve activity;WKY, Wistar–Kyoto rats.

M2 receptor mRNA levels are elevated and M4 receptor

mRNA levels are reduced in the RVM of SHR compared

with WKY. A large (27%), but similar proportion, of

presympathetic, non-C1 RVLM neurons expresses

M2R mRNA in SHR and WKY. The enhanced pressor

response evoked by central muscarinic receptor acti-

vation in SHR is not associated with increased constric-

tion of the splanchnic circulation, tachycardia or effects

on baroreflex control of sympathetic activity. Finally,

central muscarinic receptor blockade had no effect on

the baroreflex in either strain. These findings indicate

that alteration of M2 and M4 receptor mRNA expression

in the RVLM of SHR may contribute to hypertension via

an alteration in muscarinic receptor number or sensitivity

or both. The mechanism is not reflected by changes in the

number of RVLM sympathoexcitatory neurons that

express M2R and does not involve enhanced visceral

vasoconstriction, sympathetic drive to the heart or

impaired sympathetic baroreflex function.

Receptor autoradiography clearly shows that M2 receptor

mRNA is highly abundant in rat RVLM [35]. We have

confirmed the findings of Gattu et al. [26] that all five

muscarinic receptors are present in the ventral medulla of

both SHR and WKY. Using real time RT-PCR overcame

the limitations of end-point detection in RT-PCR used

previously. Confirming previous data [26], we have

demonstrated that M2 receptor mRNA expression is

almost two-fold greater in the RVM of SHR compared

with WKY. In contrast [26], we describe for the first time

that the M4 receptor transcription is reduced by 50% in

the RVM of SHR compared with WKY, but found no

differences in M3 receptor mRNA expression. No studies

have measured muscarinic receptor gene expression in

the RVLM alone. Gattu et al. [26] used a 2 mm rostral

caudal length that included the RVLM, whereas we have

used a 700 mm rostral caudal length ventral slice that

included the RVLM. Differences found may be related

to the different regions included in these slices; however,

it is noteworthy that M4 receptor expression is also

reduced in the hypothalamus of SHR [36]. An 18%

increase in receptor binding in the SHR RVLM using

tritiated AFDX-384 has been demonstrated [26], but

could include postsynaptic as well as presynaptic receptor

that was not examined in this study.

Early physiological data indicated that activation of M2

receptors mediated the sympathoexcitation evoked in

Sprague–Dawley rats [15,25,35]. These data were

acquired prior to the cloning of receptor subtypes M3–

M5 [23,24] and used pharmacological ligands that show a

less than five-fold difference in selectivity for one sub-

type over the other [18,37]. Indeed, muscarinic receptor

specific ligands are, as yet, unavailable [38,39].

As sympathoexcitatory neurons in the RVLM project

to the intermediolateral column of the spinal cord,

rized reproduction of this article is prohibited.

C

Muscarinic receptor function in hypertensive rats Kumar et al. 1007

retrograde tracing studies were combined with ISH and

immunohistochemistry in order to characterize the

neurochemistry of bulbospinal neurons in the RVLM.

Interestingly, M2R mRNA was not present in bul-

bospinal C1 neurons in the RVLM of SHR or WKY.

This is similar to our finding in Sprague–Dawley rats [3].

The number of tyrosine hydroxylase-ir neurons in

the RVLM are similar in adult SHR compared with

WKY, despite an increase in the phenylethanolamine-

N-methyl transferase-ir neurons in the medulla in

4-week SHR [40], so their number at least probably

do not contribute to the pathogenesis of hypertension.

We have demonstrated for the first time the presence of

M2R mRNA in 27% of non-C1 bulbospinal neurons of

the RVLM in SHR and WKY. These sympathoexcitatory

neurons potentially have an enkephalinergic phenotype

[41–43]. Although numbers of M2R-positive neurons are

similar, the mRNA content of the M2 receptor in each

cell is different between SHR and WKY, potentially

increasing the sensitivity of the RVLM to muscarinic

receptor activation.

In support of this idea, we have shown that OXO evoked

a larger increase in MAP in SHR compared with WKY.

This pressor effect has been described previously using

other cholinergic agents [19]. However, our novel find-

ings that OXO evoked similar levels of splanchnic sym-

pathoexcitation, tachycardia and enhanced sympathetic

baroreflex function in both SHR and WKY argues against

our hypothesis that increased mRNA levels in the RVM

are associated with increased cardiovascular function. We

have demonstrated previously that although OXO-

evoked increases in SNA and HR are entirely abolished

by muscarinic blockade in the RVLM in Sprague–Daw-

ley rats, the pressor response was attenuated, but not

abolished, and increases in cutaneous blood flow were

mediated independently of the RVLM [3]. OXO-

mediated RVLM effects (SNA and HR) are not different

in SHR or WKY, but effects at other central sites could

be, as the pressor effects evoked by OXO are greater in

SHR than WKY. These may result from the total of

haemodynamic effects elicited at multiple brain sites.

Our data show that the range and sensitivity of the

sympathetic baroreflex under control conditions is not

different between SHR and WKY, as described pre-

viously [44]. The newly described OXO-induced signifi-

cant increase in the range and gain of the sympathetic

baroreflex is similar in both SHR and WKY and mimics

that which we have described previously in Sprague–

Dawley rats [3]. This enhancement is entirely dependent

upon RVLM muscarinic receptor activation [3]. As no

differences in sympathetic baroreflex function evoked by

OXO were seen between SHR and WKY, this indicates

that baroreflex function is not associated with the

increased M2 or decreased M4 mRNA levels seen

in SHR.

opyright © Lippincott Williams & Wilkins. Unauth

Thus, it appears that tonic and reflex, sympathetically

mediated, RVLM-dependent effects evoked by muscar-

inic receptor activation are not mediated by the elevated

levels of M2 receptor mRNA or the reduced level of M4

receptor mRNA identified in SHR. The question as to

whether mRNA and protein levels are correlated remains

to be determined. It is also clear from the current study

that constitutive levels of activation of muscarinic recep-

tors in SHR, as reported previously [18,26,45], do not

contribute to normal operation of the sympathetic bar-

oreflex as SCOP failed to alter baroreflex function. These

experiments should be conducted in alternate hyperten-

sive rat strains to validate that modified muscarinic

receptor expression and OXO-evoked pressor effects

are linked.

Our findings show that the pressor response evoked by

stimulation of central muscarinic receptors is exaggerated

in SHR, although mediated neither by excessive

increases in splanchnic sympathoexcitation nor attenu-

ated baroreflex function. Similarly, the range and gain of

the sympathetic baroreflex were significantly enhanced

in SHR, but to a similar degree to that seen in WKY and

the Sprague–Dawley rats [3]. The exaggerated pressor

responses in SHR are associated with increased M2

receptor or decreased M4 receptor mRNA expression

or both, within the RVM that neither alter normal bar-

oreflex function nor affect its facilitation after muscarinic

receptor activation in the RVLM.

AcknowledgementThis work was supported by grants from the National

Health and Medical Research Council of Australia

(211023, 211196, 457068 and 457069) Garnett Passe

and Rodney Williams Memorial Foundation, North

Shore Heart Research Foundation (6-05/06) and North-

ern Sydney Central Coast Area Health (2006 : 03). J.R.P.

and N.N.K. are supported by Australian Postgraduate

Awards.

There are no conflicts of interest.

References1 Kubo T, Hagiwara Y, Sekiya D, Fukumori R. Midbrain central gray is involved

in mediation of cholinergic inputs to the rostral ventrolateral medulla of therat. Brain Res Bull 1999; 50:41–46.

2 Henderson Z, Igielman F, Sherriff FE. Organisation of the visceral solitarytract nucleus in the ferret as defined by the distribution of cholineacetyltransferase and nerve growth factor receptor immunoreactivity. BrainRes 1991; 568:35–44.

3 Padley JR, Kumar NN, Li Q, Nguyen TBV, Pilowsky PM, Goodchild AK.Central command regulation of circulatory function mediated bydescending pontine cholinergic inputs to sympathoexcitatory rostralventrolateral medulla neurons. Circ Res 2007; 100:284–291.

4 Kubo T. Cholinergic mechanism and blood pressure regulation in thecentral nervous system. Brain Res Bull 1998; 46:475–481.

5 Krstic MK, Djurkovic D. Cardiovascular response to intracerebro-ventricularadministration of acetylcholine in rats. Neuropharmacology 1978;17:341–347.

6 Krstic MK. A further study of the cardiovascular responses to centraladministration of acetylcholine in rats. Neuropharmacology 1982;21:1151–1162.

orized reproduction of this article is prohibited.

C

1008 Journal of Hypertension 2009, Vol 27 No 5

7 Feldman DS, Buccafusco JJ. Localization of cholinergic neurons involved inthe cardiovascular response to intrathecal injection of carbachol. Eur JPharmacol 1993; 250:483–487.

8 Pilowsky PM, Goodchild AK. Baroreceptor reflex pathways andneurotransmitters: 10 years on. J Hypertens 2002; 20:1675–1688.

9 Phillips JK, Goodchild AK, Dubey R, Sesiashvili E, Takeda M, Chalmers J,et al. Differential expression of catecholamine biosynthetic enzymes in therat ventrolateral medulla. J Comp Neurol 2001; 432:20–34.

10 Schreihofer AM, Guyenet PG. Identification of C1 presympathetic neuronsin rat rostral ventrolateral medulla by juxtacellular labeling in vivo. J CompNeurol 1997; 387:524–536.

11 Wei J, Walton EA, Milici A, Buccafusco JJ. M1-M5 muscarinic receptordistribution in rat CNS by RT-PCR and HPLC. J Neurochem 1994;63:815–821.

12 Kubo T, Ishizuka T, Asari T, Fukumori R. Acetylcholine release in the rostralventrolateral medulla of spontaneously hypertensive rats. Clin ExpPharmacol Physiol Suppl 1995; 22:S40–S42.

13 Kubo T, Fukumori R, Kobayashi M, Yamaguchi H. Altered cholinergicmechanisms and blood pressure regulation in the rostral ventrolateralmedulla of DOCA-salt hypertensive rats. Brain Res Bull 1998; 45:327–332.

14 Arneric SP, Giuliano R, Ernsberger P, Underwood MD, Reis DJ. Synthesis,release and receptor binding of acetylcholine in the C1 area of the rostralventrolateral medulla: contributions in regulating arterial pressure. BrainRes 1990; 511:98–112.

15 Giuliano R, Ruggiero DA, Morrison S, Ernsberger P, Reis DJ. Cholinergicregulation of arterial pressure by the C1 area of the rostral ventrolateralmedulla. J Neurosci 1989; 9:923–942.

16 Sundaram K, Sapru H. Cholinergic nerve terminals in the ventrolateralmedullary pressor area: pharmacological evidence. J Auton Nerv Syst1988; 22:221–228.

17 Kubo T, Taguchi K, Sawai N, Ozaki S, Hagiwara Y. Cholinergic mechanismsresponsible for blood pressure regulation on sympathoexcitatory neurons inthe rostral ventrolateral medulla of the rat. Brain Res Bull 1997; 42:199–204.

18 Lazartigues E, Brefel-Courbon C, Tran MA, Montastruc JL, Rascol O.Spontaneously hypertensive rats cholinergic hyper-responsiveness: centraland peripheral pharmacological mechanisms. Br J Pharmacol 1999;127:1657–1665.

19 Kubo T, Ishizuka T, Fukumori R, Asari T, Hagiwara Y. Enhanced release ofacetylcholine in the rostral ventrolateral medulla of spontaneouslyhypertensive rats. Brain Res 1995; 686:1–9.

20 Kubo T, Hagiwara Y, Endo S, Fukumori R. Activation of hypothalamicangiotensin receptors produces pressor responses via cholinergic inputsto the rostral ventrolateral medulla in normotensive and hypertensive rats.Brain Res 2002; 953:232–245.

21 Li P, Zhu DN, Kao KM, Lin Q, Sun SY. Role of acetylcholine, corticoids andopioids in the rostral ventrolateral medulla in stress-induced hypertensiverats. Biol Signals 1995; 4:124–132.

22 Helke CJ, Muth EA, Jacobowitz DM. Changes in central cholinergicneurons in the spontaneously hypertensive rat. Brain Res 1980; 188:425–436.

23 Bonner TI, Buckley NJ, Young AC, Brann MR. Identification of afamily of muscarinic acetylcholine receptor genes. Science 1987;237:527–532.

24 Bonner TI, Young AC, Brann MR, Buckley NJ. Cloning and expression ofthe human and rat m5 muscarinic acetylcholine receptor genes. Neuron1988; 1:403–410.

25 Sundaram K, Krieger AJ, Sapru H. M2 muscarinic receptors mediatepressor responses to cholinergic agonists in the ventrolateral medullarypressor area. Brain Res 1988; 449:141–149.

26 Gattu M, Wei J, Pauly JR, Urbanawiz S, Buccafusco JJ. Increasedexpression of M2 muscarinic receptor mRNA and binding sites in the rostralventrolateral medulla of spontaneously hypertensive rats. Brain Res 1997;756:125–132.

27 Kubo T, Asari T, Yamaguchi H, Fukumori R. Baroreceptor activation causesrelease of acetylcholine in the rostral ventrolateral medulla of the rat. ClinExp Hypertens 1998; 20:245–257.

28 Nuyts S, Van Mellaert L, Lambin P, Anne J. Efficient isolation of total RNAfrom Clostridium without DNA contamination. J Microbiol Methods 2001;44:235–238.

29 Ririe KM, Rasmussen RP, Wittwer CT. Product differentiation by analysis ofDNA melting curves during the polymerase chain reaction. Anal Biochem1997; 245:154–160.

30 Goodchild AK, Llewellyn-Smith IJ, Sun QJ, Chalmers J, Cunningham AM,Pilowsky PM. Calbindin-immunoreactive neurons in the reticular formationof the rat brainstem: catecholamine content and spinal projections. J CompNeurol 2000; 424:547–562.

opyright © Lippincott Williams & Wilkins. Unautho

31 Li Q, Goodchild AK, Seyedabadi M, Pilowsky PM. Preprotachykinin AmRNA is colocalized with tyrosine hydroxylase-immunoreactivity inbulbospinal neurons. Neuroscience 2005; 136:205–216.

32 Llewellyn-Smith IJ, Pilowsky P, Minson JB. The tungstate-stabilizedtetramethylbenzidine reaction for light and electron microscopicimmunocytochemistry and for revealing biocytin-filled neurons. J NeurosciMethods 1993; 46:27–40.

33 Paxinos G, Watson C. The rat brain in stereotaxic coordinates (5th ed.).San Diego, CA: Academic Press Inc.; 2005.

34 Padley JR, Overstreet DH, Pilowsky PM, Goodchild AK. Impaired cardiacand sympathetic autonomic control in rats differing in acetylcholinereceptor sensitivity. Am J Physiol Heart Circ Physiol 2005; 289:H1985–H1992.

35 Ernsberger P, Arango V, Reis DJ. A high density of muscarinic receptors inthe rostral ventrolateral medulla of the rat is revealed by correction forautoradiographic efficiency. Neurosci Lett 1988; 85:179–186.

36 Wei J, Milici A, Buccafusco JJ. Alterations in the expression of the genesencoding specific muscarinic receptor subtypes in the hypothalamus ofspontaneously hypertensive rats. Circ Res 1995; 76:142–147.

37 Li M, Yasuda RP, Wall SJ, Wellstein A, Wolfe BB. Distribution of m2muscarinic receptors in rat brain using antisera selective for m2 receptors.Mol Pharmacol 1991; 40:28–35.

38 Wess J, Eglen RM, Gautam D. Muscarinic acetylcholine receptors: mutantmice provide new insights for drug development. Nat Rev Drug Discov2007; 6:721–733.

39 Caulfield MP, Birdsall NJ. International Union of Pharmacology. XVII.Classification of muscarinic acetylcholine receptors. Pharmacol Rev 1998;50:279–290.

40 Howe PRC, Lovenberg W, Chalmers JP. Increased number ofPNMT-immunofluorescent nerve cell bodies in the medulla oblongata ofstroke-prone hypertensive rats. Brain Res 1981; 205:123–130.

41 Stornetta RL, Schreihofer AM, Pelaez NM, Sevigny CP, Guyenet PG.Preproenkephalin mRNA is expressed by C1 and non-C1 barosensitivebulbospinal neurons in the rostral ventrolateral medulla of the rat. J CompNeurol 2001; 435:111–126.

42 Guyenet PG, Schreihofer AM, Stornetta RL. Regulation of sympathetictone and arterial pressure by the rostral ventrolateral medulla after depletionof C1 cells in rats. Ann N Y Acad Sci 2001; 940:259–269.

43 Schreihofer AM, Guyenet PG. Sympathetic reflexes after depletion ofbulbospinal catecholaminergic neurons with anti-DbH-saporin. Am JPhysiol Regul Integr Comp Physiol 2000; 279:R729–R742.

44 Ono A, Kuwaki T, Kumada M, Fujita T. Differential central modulation of thebaroreflex by salt loading in normotensive and spontaneously hypertensiverats. Hypertension 1997; 29:808–814.

45 Makari NF, Trimarchi GR, Buccafusco JJ. Contribution of pre andpostsynaptic components to heightened central cholinergic activity inspontaneously hypertensive rats. Neuropharmacology 1989; 28:379–386.

rized reproduction of this article is prohibited.