European Journal
of Pharmacology,
195 (1991)
233-240
233
Elsevier Science Publishers B.V. 0014.2999/91/$03.50 ADONIS 001429999100243M 0 1991
EJP 51786
Central effects of mus a
l
gonists and antagonists on hi
Julie C. Barnes and Fiona F. Roberts Departmen? oj Neuropharmacologv,
Glaxo Group Research Ltd., Park Road, Ware. Herqorrlshire
SGI.? ODP, U.K.
Received 27 September 1990, revised MS received 21 December 1990. accepted 8 January 1991
The in vivo central effects of a range of full and partial muscarinic receptor agonists have been investigated on hippocampai theta rhythm and blood pressure. In the isoflurane-anaesthetised rat, pretreated with N-methylscopolamine. i.v. administration of arecoline. oxotremorine, arecaidine propargyl ester, aceclidine and pilocarpine produced dose-dependent increases in the frequency of hippocampal the&a rhythm and blood pressure, with an order of potency of arecoline = oxotremorine = arecaidine propargyl ester > ace&dine > pilocarpine. To increase theta wave frequency, pilocarpine showed a low maximum response and possessed antagonist activity against arecoline, indicating that pilocarpine was acting as a partial agonist. AF102B failed to alter blood pressure or theta rhythm. Intraventricular injections of scopolamine and the M, receptor-selective antagonist, pirenzepine, produced dose-dependent antagonism of the enhanced theta wave frequency and hypertensive response produced by arecoline. The differences in antagonist potency for the two responses was less than 6-fold, which indicated that both the increase in hippocampal theta wave activity and increase in blood pressure may have been mediated through muscarinic receptors of the M, subtype. Further studies using a wider range of antagonists will be required to confirm this conclusion.
Theta rhythm; Blood pressure; Muscarinic receptor agonists: Muscarinic receptor antagonists: Muscarinic receptor subtypes; (Rat)
1. Introduction
Over the past few years, evidence for the existence of multiple subtypes of muscarinic receptors has greatly expanded and molecular cloning (Bonner et al., 1987; 1988) and pharmacological studies (Buckley et al., 1989) have indicated that at least five different subtypes now exist. The pharmacological evidence to date is based on the relative potencies of a range of muscarinic antagonists. The prototype selective antagonist, pirenzepine (Hammer et al., 1980) possesses higher affinity for the M, subtype than non-M, subtypes, whereas the antagonists, atropine and scopolamine are non-selective (Freedman et al., 1988a; Buckley et al., 1989; Lazareno et al., 1990). In addition, antagonists such as himbacine, 4-diphenylacetoxy-N-methyl-piperidine methiodide (4DAMP), AF-DX 116, methoctramine and hexahydrosiladifenidol (Anwar-ul et al., 1986; Barlow et al., 1976; Giachetti et al., 1986; Giraldo et al., 1988; Mutschler and Lambrecht, 1984), also show varying degrees of selectivity for one or more of the M, to M,
Correspondence to: J.C. Barnes, Department of Neuropharmacologv, Glaxo Group Research Ltd., Park Road. Ware, Hertfordshire SG12 ODP, U.K.
subtypes (Lazareno et al., 1990) or the corresponding m2 to m5 cloned receptor subtypes (Buckley et al., 1989). Within the CNS, muscarinic receptors play an important role in mediating the actions of acetylcholine and a number of studies have been performed in an attempt to identify the different muscarinic receptor subtypes mediatir,g some of these effects, produced by either the natural transmitter itself (Hagan et al., 1987a.c; Hunter and Roberts, 1988) or by the administration of muscarinic agonists (Gene Erwin et al., 1988; Hagan et al., 1987b; Stewart et al., 1989). In the present studies, we have attempted to identify the muscarinic receptor subtypes within the brain which mediate agonist-induced changes in hippocampal theta rhythm and agonist-induced hypertension. It is well established that two types of hippocampal theta rhythm exist. An atropine-insensitive form (type 1) appears only when an animal performs such motor patterns as head movements or walking, while an atropine-sensitive form (type 2) is evident in the absence of any movement (Robinson, 1980). Type 2 theta rhythm can be measured during urethane (Olpe et al.. 1987; Kramis et al., 1975) or halothane (Bevan, 1984) anaesthesia and Olpe er al. (1987) have shown that the frequency of type 2 theta wave activity can be enhanced
v mu.scsrinic agonists. No further study to date has tnvcsttgated the effects of antagonists other than the non-seltxtive agents. scopolamine and atropine on the nist-induced changes in the frequency of theta Stimulation of central muscarinic receptors by muscarinic agonists increases blood pressure and this effect can be shown either by administering the agonists tly into the cerebral ventricles (Lang and Rush, : Day and Roach, 1977) or following their systemic administration when peripheral muscat-uric receptors are blocked by N-methylscopolamine (Pazos et al., 1986). This effect is mediated through an increase in sympathetic outflow (Wu and Wei. 1984) but there is still some controversy about which muscarinic receptor subtype within the brain is involved (Scheucher et al., 1987; Pazos et al.. 1986). The aims of the present studies were 2-fold. Firstly. we have investigated the effects of a range of muscarinic agonists on hippocampal theta wave activity and on blood pressure. conducting all experiments in the presence of peripheral muscarinic blockade. Secondly, in an attempt to identify the central muscarinic receptor subtypes mediating the agonist-induced responses, we have investigated the antagonist potencies of pirenzepine and scopolamine following their administration directly into the cerebral ventricles. It has been demonstrated that pirenzepine can identify M,-mediated behaviours (Hagan et al.. 1987a) and can differentiate between central M, and M, mediated effects in vivo (Caulfield et al., 1983).
2. Materials and methods 2. I. Animals
All experiments used male Lister-hooded rats (Glaxo) weighing between 300-350 g. The animals were maintained on a 12 h light/dark cycle and housed in groups of four per cage with food and water available ad libitum. 2.2. Surgical procedure
At the start of each experiment, anaesthesia was induced with isoflurane (Forane, ICI), delivered through a face mask using an oxygen/nitrous oxide carrier. Anaesthesia was subsequently maintained at a concentration of l-2% isoflurane through a tracheal cannula. Body temperature was regulated by a thermostatically controlled heating blanket and was maintained constant at 38°C for the duration of each experiment. For the electrophysiological studies, rats were placed in a stereotaxic frame with the incisor bar positioned 5 mm above the ear bars. Hippocampal electrical activity
was recorded from a bipolar electrode positioned in the CA1 region of the hippocampus at the following coordinates: anterior 4.1, lateral 1.8, vertical 2.4-2.6, (according to Paxinos and Watson, 1982). The correct position of the electrode was confirmed histologically at the end of each experiment. The shaft of the electrode was 0.25 mm in diameter and was insulated to within 1 mm from the tip. The non-insulated poles were concentric to each other and were separated by an insulated part, 0.5 mm long. The two poles of the electrode were connected to a 3-point input plug leading to a Grass 7P511 differential amplifier, the third point being connected to an indifferent electrode consisting‘of a screw located in the skull anterior to bregma. The amplified electrical activity was recorded on paper but was also digitised through an ASD (analogue-to-digital converter) and stored on IBM-PC, allowing subsequent off-line spectral analysis by fast Fourier transform (BIODATA). High speed sampling was performed in sections of 60 s, at a rate of 128 samples/s, and during epochs of 2 s with an interepoch gap of 3 s. A 256-point FFT analysis was subsequently performed on each 60 s section to yield estimates of EEG power (pV’/Hz) in 0.5 Hz bands, up to 64 Hz. In the cardiovascular studies, arterial blood pressure was measured from the cannulated right common carotid artery by means of a Gould transducer connected to a Grass 7P122 DC amplifier. All changes in blood pressure were measured in mm Hg. In all experiments, the left femoral vein was cannulated for the i.v. administration of drugs and in such cases, drugs were administered in a volume of 0.1 ml followed by 0.1 ml flush-in volume of saline (0.9%). For the i.c.v. administration of drugs, a stainless steel guide fitted with an injection unit was positioned unilaterally at the level of the lateral ventricle using the following coordinates: anterior 6.0, lateral 1.5, vertical -3.5 (Paxinos and Watson, 1982). The position of the injection unit was confirmed histologically by the injection of dye at the end of each experiment. All drugs were freshly prepared in either physiological saline, for i.v. administrations, or in artificial CSF (aCSF), for i.c.v. administrations. The composition of aCSF was (mM): NaCl 124; KC1 2; KH,PO, 1.25; MgSO, 2; NaHCO, 25; glucose 11; CaCl, 2. In both the electrophysiological and cardiovascular experiments, animals were pretreated with N-methylscopolamine (0.5 mg/kg i.v.), in order to block peripheral muscarinic receptors. In addition, in view of our preliminary observations that changes in blood pressure produced marked changes in the EEG activity, EEG measurements were made following pretreatment with the ganglion blocking agent, chlorisondamine (0.1 mg/kg i.v.) to prevent the increases in blood pressure produced by the agonists. Only one agonist was examined in each experiment. The agonist was administered in increasing
235
doses at intervals of between 10 and 25 min, depending on the duration of each agonist response. Antagonist effects of pirenzepine and scopolamine were measured against a fixed dose of arecoline (0.3 mg/kg). Arecoline was used for these experiments because its short duration of action allowed reproducible agonist effects to be established. Antagonists were injected in 1 ~1 aCSF cumulatively, 10 min prior to arecoline and no more than two antagonist doses were administered in each experiment. Drugs for these studies were obtained from the following sources: arecoline, oxotremorine, pilocarpine, pirenzepine, scopolamine and N-methylscopolamine (Sigma); aceclidine (Cilag-Chemie); chlorisondamine (Ciba-Geigy); arecaidine propargyl ester and AFlO2B (synthesised in our laboratories). 2.3. Statistical analysis Analysis of variance was used for all statistical comparisons, followed by Dunnett’s test where appropriate,
3. Results
3.1. Agonist and antagonist effects on hippocampal theta wave activity The electrical activity recorded from the CA1 region of the hippocampus in the isoflurane anaesthetised rat was character&d by three different wave forms; rhythmical theta wave activity of frequencies of 3-5 Hz, fast activity which occurred alone or which was superimposed on theta waves, and slow waves with frequencies of OS-2 Hz. Under deep anaesthesia (2% isoflurane), the predominant waveform was either fast or slow activity, with no apparent theta rhythm, but at 1% isoflurane, theta rhythm appeared during approximately 90% of the recording. The amplitude of the theta waveform ranged between 300-600 pV. Under 1% isoflurane anaesthesia, arecoline, oxotremorine, are&dine propargyl ester, aceclidine and pilocarpine produced dose-dependent increases in the frequency of theta wave activity. Example trace responses to arecoline taken from a single experiment, with the corresponding superimposed power spectrum obtained by FFT analysis are shown in fig. 1. Agonist effects on theta wave frequency were also accompanied by changes in the power of the signal, and although there was a tendency for agonist doses in the lower range to increase power and higher doses to produce decreases, these effects were inconsistent and therefore have not been analysed in detail. Figure 2 shows the mean increases in theta wave frequency produced by all the agonists. The most potent agonists were arecoline, oxotremorine and arecaidine propargyl ester, all pro-
Frequency (Hz! 1s Fig. 1. Example responsesIOarecoline on hippocampaltheta rhythm, recorded from a single experiment. (A-D) Original trace recordings following treatment with saline, arecoline 0.03, 0.1.0.3 mg/kg respectively. (E) Represents the superimposed power spectra by FIT anafysis and shows the rightward shift of the peaks with increasing doses of arecoline (same as shown in A-D).
ducing significant increases in frequency at doses greater than 0.03 mg/kg (P < 0.05). Acechdine and pilocarpine were much weaker and produced significant effects only at doses greater than 1 mg/kg (P K 0.05). AF102B produced a vety small increase in the frequency but compared with the baseline frequency measured after saline treatment, this response failed to reach statistical significance at any dose (P > 0.05). The agonist-induced increases in frequency (above that measured after saline treatment) ranged from 2.0 & 0.3 Hz (for pilocarpine) to 4.3 f 0.2 Hz (for oxotremorine). With the exception of pilocarpine, maximum responses to the agonists were not reached at the doses used. In view of the small 9
8
1
PL
’
Saline
1
001
01
1C
100
Fig. 2. Effect of muscarinic agonists on the frequency of bippocampal APE 9. theta rhythm. For all agonists (arecoline 0, oxotremorine aceclidine A, pilocarpine V, AF102B 0). the frequency was measured front the peak of the power spectrum created by FFT analysis. Each point represents the mean f S.E.M. of three to six experiments. Statistically significant differences are indicated as * P < 0.05 (versus saline).
80
-
70
-
60
-
I” 50
-
0’ g) 40
-
z :
C 1 .o
20
-
of
pilocdrpine against responses to arecoline
(0.3 mgjkg) on hippocampd theta rhythm. The columns represent the mean peak frequency ( + S.E.M.) measured from the FFT power trace B). arecoline spectrum following saline (0. trace A). arecolin lus pilocarpine 10 plus 3 mg/kg (I trace C). arecol D). Statistically significant differences are indicated m&J as * Pi 0.05 (versus saline). + P i 0.05 (versus arecoline alone).
maximum increase in frequency produced by pilocarpine. we investigated the possibility that pilocarpine was acting as a partial agonist and therefore wou!d antagonise the increase in frequency produced by the more efficacious agonist. arecoline (fig. 3). In these experiments, the increase in frequency of theta wave activity produced by arecoline administered 10 min after 10 mg/kg pilocarpine was significantly less than the response to arecoline alone (P < 0.05). No antagonism of the response to arecoline was produced after 3 mg/kg pilocarpine. CSF
1
P
A100
1000 AntagonIst dose (nmol I.Cv
D
3. Antagonist activity
A. Amfictal
-
00 t
0.0 Pi.
30
B. Plrenzeplne
I
Fig. 5. Comparison of antagonist potencies. Each point represents the mean ( + S.E.M.) frequency measured after arecoline (0.3 mg/kg). The squares represent responses before (0) and following (m) scopolamine and the circles represent responses before (0) and following (0) pirenzepine.
In the antagonist studies, control responses to arecoline (0.3 mg/kg i.v.) were established prior to the administration of either scopolamine, pirenzepine or aCSF. The increase in frequency produced by arecoline in all these experiments was 3.1 f 0.3 Hz (significantly different from saline, P < 0.05) and there was no significant difference between the control arecoline response prior to aCSF, scopolamine or pirenzepine. Artificial CSF failed to alter this agonist response (fig. 4A). However, although pirenzepine and scopolamine failed to significantly modify baseline theta wave frequency, both antagonists dose dependently reduced the increase in frequency produced by arecoline (fig. 4B and C). In C. Scopolamine
6.0 5.0 4.0 3.0 2.0
0.0 0 160 240
Fig 4. Effect of i.c.v. injection Of artificial CSF (A), pirenzepine (B) or scopolamine (C) on EEG responses induced by arecoline (0.3 mg/kg). The bars represent the mean ( +S.E.M.) frequency produced by saline (0. trace I). arecoline (ttt, trace 2) or arecoline following the i.c.v. administration of two doses or volumes of antagonist or vehicle (c:. trace 3 or trace 4 respectively) Antagonist doses Oqg/gI) are indicated under each bar. Statistically significant differences are indicated as * P < 0.05 (versus saline) or + P e 0.05 (versus arecoline control).
237
3.2. Agonist and antagonist effects on blood pressure
1.0
0.1 Dose
10.0
(mg/kgl
Fig. 6. Effect of muscarinic agonists on blood pressure measured after pretreatment with N-methylscopolamine. Each point represents the mean (*S.E.M.) increase in blood pressure produced by arecoline (0). oxotremorine #I). are&dine propargyl ester (e), aceclidine (A) and pilocarpine (v).
In these experiments, mean resting biood pressure ranged between 75 to 120 mm Hg. All agonists, with the exception of AF102B produced dose-related increases in blood pressure. The hypertensive effects of armline, arecaidine propargyl ester, oxotremorine, ace&dine and pilocarpine are shown in fig. 6. As in the EEG studies, arecoline, oxotremorine and are&dine propargyl ester were the most potent, with pilocarpine and aceclidine showing approximately NO-fold lower potency. Maximum responses to each agonist were not reached. AF102B failed to increase blood pressure at doses up to 20 mg/kg (data not shown). In the antagonist studies, scopolamine and pirenzepine, but not aCSF, dose dependently reduced the hypertensive response to arecoline (fig. 7). Relative antagonist potencies were estimated by plotting increases in blood pressure against nmol dose of antagonist (fig. 8). In these experiments, scopolamine was approximately 6-fold more potent than pirenzepine. 3.3. Comparison blood pressure
order to directly compare the potencies of the two antagonists, the frequency of theta wave activity measured after arecoline was plotted against antagonist dose, expressed in nmol (fig. 5). An estimation of the relative potency of scopolamine and pirenzepine indicated that scopolamine was approximately 3 times more potent than pirenzepine.
A
An~f~c~alCSF
C
70 ?
of antagonist
effects on theta rhythm and
In order to identify the selectivity of pirenzepine to antagonise either the changes in EEG or blood pressure produced by arecoline, we compared the absolute doses required to antagonise both effects. Although higher doses of pirenzepine were required to antagonise the increase in theta wave frequency, the doses used were only 2- to 3-fold higher than those required to antagonise the hypertensive effect. Similar small differences in
Scopolamine
70 F
60
.
60
60 1
P 50
40
30
20
10
0
9
Fig. 7. Effect of i.c.v. injection of artificial CSF (A), pirenzepine (B) or scopolamine (C) on hypertensive responses produced by arecoline (0.3 mg/kg). The bars represent the mean ( f S.E.M.) response produced by arecoline. alone (0) or following the i.c.v. administration of two cumulative doses or volumes of antagonist or vehicle (Coor respectively). Antagonist doses (ug/gl) are indicated under each bar. Statistically significant differences are indicated as * P -z 0.05 (versus arecoline control).
10
1
100
AntagonIstdose
1000 hmol
1C.-J1
Fig. 8. Comparison of antagonist potencies. Each point represents the mean ( f S.E.M.) increase in blood pressure measured after arecoline (0.3 mg/kg). The squares represent responses before (0) and follow) scopolamine, and the circles before (0) and following (e) pirenzepine.
entified with scopolamine (compare
the present study. we have shown that in the rat. muscarinic agonistS inr3n+maesthetised iso CW~ the ~~~e~c~ of hippocampal theta rhythm and prgs~ure by activating muscarinic recepin the brain. The evidence for the effects being ated centrally are drawn from the observations that th responses were measured in the presence of the ~~ate~ary muscarinic antagonist, N-methyllamme. which does not penetrate into the CNS and secondly. the effects were antagouised by the i.c.v. ad~~stration of ~o~la~ne. Previous studies. reporting muscarinic agonist-induced changes in type 2 theta rhythm and increases in pressure (see Introduction), have used only a range of agonists. We have extended this work by investigating the effects of a range of full and partial ‘sts with varying affinities for muscarinic receptors. most potent agonists were oxotremorine, arecoline and are&dine propargyl ester. whereas pilocarpine and ace&dine were much weaker and AFlO2B was inactive. In addition. differences in the magnitude of the responses were also identified and were particularly evident in the EEG experiments. For example, the theta rhythm recorded after the administration of oxotremarine and arecaidine propargyl ester achieved much higher fr~uencies than that measured during treatment with pi&at-pine, which actually possessed antagonist activity at the higher doses. Many factors can account for such differences. It is well recognised that the ability of au agonist to elicit a response depends not only on the affinity of the agonist for the receptor but also its efficacy. as well as other factors such as receptor density and the efficiency of coupling of the receptor to the s~mul~s-r~~nse mechanism (Kenakin, 1986). Thus, compounds with low efficacy may act as full agonists in tissues which possess an efficiently coupled stimulus-response mechanism, but in other tissues, in which the coupling to response is less efficient, they will behave as partial agonists (Ringdahl, 1987; Ringdahl et al., 1987; Freedman et al., 1988b; Fisher et al., 1983). In the present EEG studies, the obse~ation that pilocarpine behaved as a weak agonist with a low maximum respan% while oxotremorine and arecoline were more efficacious correlates well with their relative intrinsic activities in CNS tissue (Freedman et al., 1988b). Pilocarpine was shown to antagonise the EEG response to am&ne which is analogous to that reported by Fisher et al. 0983) showing antagonist properties of pilocarPine against more efficacious agonists. The inactivity of AFLO2B to increase theta wave frequency may also be a
reflection of its low intrinsic activity as reported by Nakahara et 31. (1990). In a number of in vitro preparations, we have also identified partial agonist properties of AFIOZB. and show it to be less efficacious than even pilocarpine (unpublished observations). In the present antagonist studies, we have used the M,-selective antagonist, pirenzepine to antagonise responses induced by arecoline and have measured a potency approximately 3- to &fold less than that of the non-selective antagonist, scopolamine. A number of authors have reported affinity estimates for these two antagonists in radioligand binding studies (Freedman et al., 19883; Watson et al., 1986; Lazareno et al., 1990; Hagan et al., 1987b) and have shown that although scopolamine has the highest affinity (approximate PKi = 9.2). pirenzepine is M,-selective (pK, = 8.1 on M, versus 6.4 on M,, 7.1 on M,) and therefore the difference in affinity between the two antagonists is much less on the M, subtype than on non-M, receptors (12 on M,, 630 on M,, 126 on M3). On this basis, the relatively small potency difference measured between scopolamine and pirenzepine in the present experiments supports the hypothesis that the agonist-induced increases in theta wave frequency and blood pressure are mediated through M, receptors rather than M, or M, receptors. Furthermore, in the recent report by Lazareno et al. (1990), the pharmacology of a mammalian M, receptor in rabbit lung has been detailed and on this recep tar, pirenzepine shows an affinity estimate of 7.5, intermediate to that on the M, and M, subtypes. The difference in affinity between scopolamine and pirenzepine on this subtype is still greater than that measured on the M, subtype, and this would further tend to support the involvement of receptors of the M, subtype than the M, subtype as mediators of the in vivo effects reported here. While our antagonist data suggest the involvement of M, receptors for both the EEG and cardiovascular effects, accurate comparisons of antagonist potencies are often confounded by different penetration into brain tissue from the ventricles, particularly when comparing compounds with differing lipophilicity. In this particular study, if the antagonist distribution was very different due to the greater polarity of pirenzepine than scopolamine, one would expect the relative difference in potency to be, if anything, over estimated. This would clearly not influence the inte~retation of our findings. However in the present experiments, we have identified small differences in the absolute potencies of the antagonists between the EEG and cardiovascular effects, which may be a reflection of differential penetration throughout brain tissue. Higher doses of pirenzepine were required to antagonise the EEG response than tile hypertension. Similar differences were also measured with scopolamine and therefore, although these differences are unlikely to be related to receptor
239
seiectivity, they could reflect possible differences in penetration to ~ffe~nt sites within the brain. There is indirect evidence to suggest that muscarinic agonists stimulate theta rhythm by an action in the medial septum (see Bland, 1986; Olpe et al., 1987; Stewart and Fox, 1989) while it would appear more likely for the hypertensive effects of muscarinic agonists to be mediated through the lateral septum (Scheucher et al., 1987). Thus, if muscarinic agonists were acting in the medial septum to increase theta wave frequency but in the lateral septum to increase blood pressure, one might expect the more medial areas to be less accessible for ~tago~sts after i.c.v. ad~~stration, and hence the higher doses required to antagonise the EEG response. However, a full explanation for such differences in the absolute antagonist potencies cannot be provided until brain penetra~on from the ventricular spaces and distribution in brain tissue is measured using radiolabeled antagonists. In summary, we have shown that brain-penetrating, high efficacy musearinic agonists produce increases in the frequency of hippocampal theta rhythm and increases in blood pressure. Partial agonist properties of pilocarpine have also been identified. Antagonists studies indicated that both agonist-induced responses may be mediated through muscarinic receptors of the M, subtype, although this conclusion requires support from further studies, comparing relative potencies of a wider range of selective muscarinic antagonists.
We would like to thank E. Rogers for her technical assistance, Ciba-Geigy for the generous gift of chlorisondamine and S. Guntrip for the synthesis of arecaidine pmpargyl ester and AFlO2B.
References Anwar-ul, S., H. Gilani and L.B. Cobbin, 1986, The cardioselectivity profile of himbacine: a muscarinic receptor antagonist, NaunynScluniedeb. Arch. Pharmacol. 332 16. Barlow, R-B.. K.J. Berry, P.A.M. Glenton, N.M. Nickolaou and K.S. Soh, 1976. A compa~son of affinity constants for muscarinic-sensitive acetylcholine receptors in guinea-pig atrial pacemaker cells at 29°C and in ileum at 29’C and 37°C. Br. J. Pharmacol. 58,613. Bevan, P., 1984. Effect of muscarinic ligands on the electrical activity recorded from the ~pp~ampus: a quantitative approach, Br- J. Pharmacol. 82,431. Bland, B.H.. 1986, The physiology and pharmacology of hippocampal formation theta rhythms, Prog Neurobiol. 26.1. Banner, T.L, N.J. Buckley. AC. Young and M.R. Braun, 1987, Identification of a famify of muscarinic receptor genes, Science 237, 527. Bonner, T-1.. A.C. Young, M.R. Braun and N.J. Buckley, 1988, Cloning and expression of the human and rat m5 muscarinic receptor genes, Neuron 1.403. Buckley, N.J.. T.1. Banner, CM. Buckley and M.R. Braun. 1989,
Antagonist binding properties of Bve cloned muscari& receptors expressed in CHO-Kl cells, Mol. Pharmacol. 35,469. Caulfield, M.P.. G.A. Higgins and D.W. Straughan, 1983, Central administration of the muscatinic receptor subtype-selective antagonist pircnxepine selectively impairs passive avoidance learn ing in the mouse, J. Pharm. Pharmacol. 35, 131. Day. M.D. and A.G. Roach,1977, Cardiovasctdar effects of carbachol and other cholinomitnetics administered into the cm&r& vattick of conscious cats, Clin. Exp. Pharmacol. Physiol. 4,431. Fisher,S.K., P.D. Klingerand B.W. Agranoff, 1983, Muscat&& agonistbinding and pho~o~pid turnover in brain, J. Biol. Chem. 258.7358. Freedman. S.B., H.S. Beer and EA. Harley, 1988a. M uscar& Ml, M2 receptor binding. Relationship with functional efficacy, European J. Phannacol. 156,133. Freedman, S.B., EA. Harley and L.L. iversen, 1988b. Relative affmities of drugs acting at chotinoceptors in displacing agonist and antagonist radioligands: the NMS/OXO-M ratio as an index of efficacy at cortical muscarinic receptors, Br. 1. Pharmacol. 93,437. Gene Erwin. V., A. Korte and B.C. Jones, 1988, Central muscarnnc chohnergic influences on ethanol sensitivity in long-sleep and short-sleep mice, J. Pharmacol. Exp. Ther. 247, 857. Giachetti. A., R. Micheletti and E. Montagna, 1986, Cardioselective profile of AF-DX 116, a muscarinic M2 receptor antagonist, Life Sci. 38, 1663. Giraido, E., R. Micheletti. E. Montagna, A. Giachetti. M.k Vigano, H. Ladinsky and C. Melchoirre. 1988, Binding and functional characterisation of the cardioselective muscarinic antagonist, meth~tr~ne, J. Pharmacol. Exp. Ther. 244.1016. Hagan. J.J., J.H.M. Jansen and C.L.E. Broekkamp, 1987a, Blockade of spatial learning by the Ml muscarinic antagonist pirenxepine, Psychnpharmaco!ogy 93,470. Hagan, J.J., J.A.D.M. Tonnaer, H. Rijk and C.L.E. Broekkamp, 1987b. Cholinergic stimulation of drinking from the lateral hypothalamus indications for M2 muscarinic receptor mediation. Pharmacol. B&hem. Behav. 26,771. Hagan, J.J.. J.A.D.M. Tonnaer, H. Rijk, CL-E. Broekkamp and A.M.L. Van Delft, 1987c. Facilition of ampheta~~~du~ rotation by muscarinic antagonists is correlated with M2 receptor affinity, Brain Res. 410. 69. Hammer, R.. C.P. Berrie, N.J.M. Birdsall, A.S.V. Burgen and E.C. Hulme, 1980, Pirenxepine distinguishes between different subclasses of musc~nic receptor, Nature 283.90. Hunter, A.J. and F.F. Roberts, 1988, The effect of pirenxepine on spatial learning in the Morris water maxe, Pharmacol. Biochem. Behav. 30,519. Kenakin, T.P., 1986, False premiss. Receptor resewe is a tissue misnomer, Trends Pharmacol. Sci. 7.93. Kramis, R., C.H. Vandenvolf and B.H. Bland, 1975, Two types of hippocampal rhythmical slow activity in both the rabbit and the rat: Refations to behaviour and effects of atropine. diethyl ether. urethane and pentobarbitoi. Exp. Neural. So, 58. Lang, W.J. and M.L. Rush, 1973, Cardiovascular responses to injections of cholinomimetic drugs into the cerebral ventricles of unanaesthetised dogs, Br. J. Pharrnacol. 47, 196. Lazareno, S., N.J. Buckley and F.F. Roberts, 1990, Character&ion of muscarinic M4 binding sites in rabbit lung, chick heart and NG108-15 cells, Mol. Pharmacol. 38, 805. Mutschler. E. and G. Lambrecht, 1984, Selective mUsCariniC agonists and antagonists in functional tests, Trends Pharmacol. sci. 5. Suppl. 39. Nakahara, N., N. Fujise, G. Kawanishi and F. Mizobe.19!%Central muxar$c activitiesof an MI-selective agonist: preferential effect on reversal of amnesia, Brain Res. 507. 172. olpe, HR., K. Klebs, E. Rung, P. Campiche. A. Giatt.R. Grtmann. F. DAmato,M.F. Poz~a and C. Mondadori, 1987. Chohnomimet-
rk, A.. K.-H.
&fwts
Wiedrrhold and J.M. Palacios 1956. Central pressor mdwxd h> muscarinic rwvptor agonists. evidence for a
pr&xninant role of the M2 receptor subtype. European J. Pharmacol. 125. 63. ~i~~abl. B.. 1987. Scleclivity of parti agonists related IO axotremorine based on differences in muscarinic receptor reserve hctween the guinea-pig ileum and urinary bladder, Mol. Pharmaeel. 31. 351. Rin8dahl. B.. H. Roth and D.J. Jenden. 1987. Regional differences in nx~plor rtxwvr for analogs of ox.otremorine in viva: Implicaiions for development of selective muscarinic agonists. J. Pharmacol. Exp. Ther. 242.464. Robinson. T.E. 1980. Hippocampal rhythmic slow activity (RSA; thets): a critical analysis of srlccted studies and discussion of w-ible species differences. Brain Res. Rev. 2, 69. Scheucber, A.. C.J. Pirola. MS. Bslda. S.M. Dabsys. A.L. Alvarez. S.
Finkielman and V.E. Nahmod. 1987. Musfdrinic Ml receptors In the lateral septal area mediate cardiovascular responses to cholinergic agouists and bradykinin: supersensitivty induced by chronic treatment with atropine. Neuropharmacology 26. 181. Stewart. B.R.. P. Jenner and C.D. Marsden, 1989. Assessment of the muscarinic receptor subtype involved in the mediation of pilocarpine-induced purposeless chewing behaviour. Psychopharmacology 97. 228. Stewart. M. and S.E. Fox. 1989. Two populations of rhythmically bursting neurons in rat medial septum are revealed by atropine. J. Neurophysiol. 61. 982. Watson, M.. W.R. Roeske and H.I. Yamamura. 1986, [‘HJPirenzepine and [ “Hlquinuclidinyl benzilate binding to rat cerebral cortical and cardiac muscarinic cholinergic sites. II. Characterisation and regulation of antagonist binding to putative muscarinic subtypes. J. Pharmacol. Exp. Ther. 237. 419. Wu. Y.Y. and E.T. Wei, 1983. Mechanisms underlying the pressor responses to acute and chronic inlraventricular administration of carbachol in the rat, J. Pharmacol. Exp. Ther. 228, 354.
of Pharmacology,
195 (1991)
233-240
233
Elsevier Science Publishers B.V. 0014.2999/91/$03.50 ADONIS 001429999100243M 0 1991
EJP 51786
Central effects of mus a
l
gonists and antagonists on hi
Julie C. Barnes and Fiona F. Roberts Departmen? oj Neuropharmacologv,
Glaxo Group Research Ltd., Park Road, Ware. Herqorrlshire
SGI.? ODP, U.K.
Received 27 September 1990, revised MS received 21 December 1990. accepted 8 January 1991
The in vivo central effects of a range of full and partial muscarinic receptor agonists have been investigated on hippocampai theta rhythm and blood pressure. In the isoflurane-anaesthetised rat, pretreated with N-methylscopolamine. i.v. administration of arecoline. oxotremorine, arecaidine propargyl ester, aceclidine and pilocarpine produced dose-dependent increases in the frequency of hippocampal the&a rhythm and blood pressure, with an order of potency of arecoline = oxotremorine = arecaidine propargyl ester > ace&dine > pilocarpine. To increase theta wave frequency, pilocarpine showed a low maximum response and possessed antagonist activity against arecoline, indicating that pilocarpine was acting as a partial agonist. AF102B failed to alter blood pressure or theta rhythm. Intraventricular injections of scopolamine and the M, receptor-selective antagonist, pirenzepine, produced dose-dependent antagonism of the enhanced theta wave frequency and hypertensive response produced by arecoline. The differences in antagonist potency for the two responses was less than 6-fold, which indicated that both the increase in hippocampal theta wave activity and increase in blood pressure may have been mediated through muscarinic receptors of the M, subtype. Further studies using a wider range of antagonists will be required to confirm this conclusion.
Theta rhythm; Blood pressure; Muscarinic receptor agonists: Muscarinic receptor antagonists: Muscarinic receptor subtypes; (Rat)
1. Introduction
Over the past few years, evidence for the existence of multiple subtypes of muscarinic receptors has greatly expanded and molecular cloning (Bonner et al., 1987; 1988) and pharmacological studies (Buckley et al., 1989) have indicated that at least five different subtypes now exist. The pharmacological evidence to date is based on the relative potencies of a range of muscarinic antagonists. The prototype selective antagonist, pirenzepine (Hammer et al., 1980) possesses higher affinity for the M, subtype than non-M, subtypes, whereas the antagonists, atropine and scopolamine are non-selective (Freedman et al., 1988a; Buckley et al., 1989; Lazareno et al., 1990). In addition, antagonists such as himbacine, 4-diphenylacetoxy-N-methyl-piperidine methiodide (4DAMP), AF-DX 116, methoctramine and hexahydrosiladifenidol (Anwar-ul et al., 1986; Barlow et al., 1976; Giachetti et al., 1986; Giraldo et al., 1988; Mutschler and Lambrecht, 1984), also show varying degrees of selectivity for one or more of the M, to M,
Correspondence to: J.C. Barnes, Department of Neuropharmacologv, Glaxo Group Research Ltd., Park Road. Ware, Hertfordshire SG12 ODP, U.K.
subtypes (Lazareno et al., 1990) or the corresponding m2 to m5 cloned receptor subtypes (Buckley et al., 1989). Within the CNS, muscarinic receptors play an important role in mediating the actions of acetylcholine and a number of studies have been performed in an attempt to identify the different muscarinic receptor subtypes mediatir,g some of these effects, produced by either the natural transmitter itself (Hagan et al., 1987a.c; Hunter and Roberts, 1988) or by the administration of muscarinic agonists (Gene Erwin et al., 1988; Hagan et al., 1987b; Stewart et al., 1989). In the present studies, we have attempted to identify the muscarinic receptor subtypes within the brain which mediate agonist-induced changes in hippocampal theta rhythm and agonist-induced hypertension. It is well established that two types of hippocampal theta rhythm exist. An atropine-insensitive form (type 1) appears only when an animal performs such motor patterns as head movements or walking, while an atropine-sensitive form (type 2) is evident in the absence of any movement (Robinson, 1980). Type 2 theta rhythm can be measured during urethane (Olpe et al.. 1987; Kramis et al., 1975) or halothane (Bevan, 1984) anaesthesia and Olpe er al. (1987) have shown that the frequency of type 2 theta wave activity can be enhanced
v mu.scsrinic agonists. No further study to date has tnvcsttgated the effects of antagonists other than the non-seltxtive agents. scopolamine and atropine on the nist-induced changes in the frequency of theta Stimulation of central muscarinic receptors by muscarinic agonists increases blood pressure and this effect can be shown either by administering the agonists tly into the cerebral ventricles (Lang and Rush, : Day and Roach, 1977) or following their systemic administration when peripheral muscat-uric receptors are blocked by N-methylscopolamine (Pazos et al., 1986). This effect is mediated through an increase in sympathetic outflow (Wu and Wei. 1984) but there is still some controversy about which muscarinic receptor subtype within the brain is involved (Scheucher et al., 1987; Pazos et al.. 1986). The aims of the present studies were 2-fold. Firstly. we have investigated the effects of a range of muscarinic agonists on hippocampal theta wave activity and on blood pressure. conducting all experiments in the presence of peripheral muscarinic blockade. Secondly, in an attempt to identify the central muscarinic receptor subtypes mediating the agonist-induced responses, we have investigated the antagonist potencies of pirenzepine and scopolamine following their administration directly into the cerebral ventricles. It has been demonstrated that pirenzepine can identify M,-mediated behaviours (Hagan et al.. 1987a) and can differentiate between central M, and M, mediated effects in vivo (Caulfield et al., 1983).
2. Materials and methods 2. I. Animals
All experiments used male Lister-hooded rats (Glaxo) weighing between 300-350 g. The animals were maintained on a 12 h light/dark cycle and housed in groups of four per cage with food and water available ad libitum. 2.2. Surgical procedure
At the start of each experiment, anaesthesia was induced with isoflurane (Forane, ICI), delivered through a face mask using an oxygen/nitrous oxide carrier. Anaesthesia was subsequently maintained at a concentration of l-2% isoflurane through a tracheal cannula. Body temperature was regulated by a thermostatically controlled heating blanket and was maintained constant at 38°C for the duration of each experiment. For the electrophysiological studies, rats were placed in a stereotaxic frame with the incisor bar positioned 5 mm above the ear bars. Hippocampal electrical activity
was recorded from a bipolar electrode positioned in the CA1 region of the hippocampus at the following coordinates: anterior 4.1, lateral 1.8, vertical 2.4-2.6, (according to Paxinos and Watson, 1982). The correct position of the electrode was confirmed histologically at the end of each experiment. The shaft of the electrode was 0.25 mm in diameter and was insulated to within 1 mm from the tip. The non-insulated poles were concentric to each other and were separated by an insulated part, 0.5 mm long. The two poles of the electrode were connected to a 3-point input plug leading to a Grass 7P511 differential amplifier, the third point being connected to an indifferent electrode consisting‘of a screw located in the skull anterior to bregma. The amplified electrical activity was recorded on paper but was also digitised through an ASD (analogue-to-digital converter) and stored on IBM-PC, allowing subsequent off-line spectral analysis by fast Fourier transform (BIODATA). High speed sampling was performed in sections of 60 s, at a rate of 128 samples/s, and during epochs of 2 s with an interepoch gap of 3 s. A 256-point FFT analysis was subsequently performed on each 60 s section to yield estimates of EEG power (pV’/Hz) in 0.5 Hz bands, up to 64 Hz. In the cardiovascular studies, arterial blood pressure was measured from the cannulated right common carotid artery by means of a Gould transducer connected to a Grass 7P122 DC amplifier. All changes in blood pressure were measured in mm Hg. In all experiments, the left femoral vein was cannulated for the i.v. administration of drugs and in such cases, drugs were administered in a volume of 0.1 ml followed by 0.1 ml flush-in volume of saline (0.9%). For the i.c.v. administration of drugs, a stainless steel guide fitted with an injection unit was positioned unilaterally at the level of the lateral ventricle using the following coordinates: anterior 6.0, lateral 1.5, vertical -3.5 (Paxinos and Watson, 1982). The position of the injection unit was confirmed histologically by the injection of dye at the end of each experiment. All drugs were freshly prepared in either physiological saline, for i.v. administrations, or in artificial CSF (aCSF), for i.c.v. administrations. The composition of aCSF was (mM): NaCl 124; KC1 2; KH,PO, 1.25; MgSO, 2; NaHCO, 25; glucose 11; CaCl, 2. In both the electrophysiological and cardiovascular experiments, animals were pretreated with N-methylscopolamine (0.5 mg/kg i.v.), in order to block peripheral muscarinic receptors. In addition, in view of our preliminary observations that changes in blood pressure produced marked changes in the EEG activity, EEG measurements were made following pretreatment with the ganglion blocking agent, chlorisondamine (0.1 mg/kg i.v.) to prevent the increases in blood pressure produced by the agonists. Only one agonist was examined in each experiment. The agonist was administered in increasing
235
doses at intervals of between 10 and 25 min, depending on the duration of each agonist response. Antagonist effects of pirenzepine and scopolamine were measured against a fixed dose of arecoline (0.3 mg/kg). Arecoline was used for these experiments because its short duration of action allowed reproducible agonist effects to be established. Antagonists were injected in 1 ~1 aCSF cumulatively, 10 min prior to arecoline and no more than two antagonist doses were administered in each experiment. Drugs for these studies were obtained from the following sources: arecoline, oxotremorine, pilocarpine, pirenzepine, scopolamine and N-methylscopolamine (Sigma); aceclidine (Cilag-Chemie); chlorisondamine (Ciba-Geigy); arecaidine propargyl ester and AFlO2B (synthesised in our laboratories). 2.3. Statistical analysis Analysis of variance was used for all statistical comparisons, followed by Dunnett’s test where appropriate,
3. Results
3.1. Agonist and antagonist effects on hippocampal theta wave activity The electrical activity recorded from the CA1 region of the hippocampus in the isoflurane anaesthetised rat was character&d by three different wave forms; rhythmical theta wave activity of frequencies of 3-5 Hz, fast activity which occurred alone or which was superimposed on theta waves, and slow waves with frequencies of OS-2 Hz. Under deep anaesthesia (2% isoflurane), the predominant waveform was either fast or slow activity, with no apparent theta rhythm, but at 1% isoflurane, theta rhythm appeared during approximately 90% of the recording. The amplitude of the theta waveform ranged between 300-600 pV. Under 1% isoflurane anaesthesia, arecoline, oxotremorine, are&dine propargyl ester, aceclidine and pilocarpine produced dose-dependent increases in the frequency of theta wave activity. Example trace responses to arecoline taken from a single experiment, with the corresponding superimposed power spectrum obtained by FFT analysis are shown in fig. 1. Agonist effects on theta wave frequency were also accompanied by changes in the power of the signal, and although there was a tendency for agonist doses in the lower range to increase power and higher doses to produce decreases, these effects were inconsistent and therefore have not been analysed in detail. Figure 2 shows the mean increases in theta wave frequency produced by all the agonists. The most potent agonists were arecoline, oxotremorine and arecaidine propargyl ester, all pro-
Frequency (Hz! 1s Fig. 1. Example responsesIOarecoline on hippocampaltheta rhythm, recorded from a single experiment. (A-D) Original trace recordings following treatment with saline, arecoline 0.03, 0.1.0.3 mg/kg respectively. (E) Represents the superimposed power spectra by FIT anafysis and shows the rightward shift of the peaks with increasing doses of arecoline (same as shown in A-D).
ducing significant increases in frequency at doses greater than 0.03 mg/kg (P < 0.05). Acechdine and pilocarpine were much weaker and produced significant effects only at doses greater than 1 mg/kg (P K 0.05). AF102B produced a vety small increase in the frequency but compared with the baseline frequency measured after saline treatment, this response failed to reach statistical significance at any dose (P > 0.05). The agonist-induced increases in frequency (above that measured after saline treatment) ranged from 2.0 & 0.3 Hz (for pilocarpine) to 4.3 f 0.2 Hz (for oxotremorine). With the exception of pilocarpine, maximum responses to the agonists were not reached at the doses used. In view of the small 9
8
1
PL
’
Saline
1
001
01
1C
100
Fig. 2. Effect of muscarinic agonists on the frequency of bippocampal APE 9. theta rhythm. For all agonists (arecoline 0, oxotremorine aceclidine A, pilocarpine V, AF102B 0). the frequency was measured front the peak of the power spectrum created by FFT analysis. Each point represents the mean f S.E.M. of three to six experiments. Statistically significant differences are indicated as * P < 0.05 (versus saline).
80
-
70
-
60
-
I” 50
-
0’ g) 40
-
z :
C 1 .o
20
-
of
pilocdrpine against responses to arecoline
(0.3 mgjkg) on hippocampd theta rhythm. The columns represent the mean peak frequency ( + S.E.M.) measured from the FFT power trace B). arecoline spectrum following saline (0. trace A). arecolin lus pilocarpine 10 plus 3 mg/kg (I trace C). arecol D). Statistically significant differences are indicated m&J as * Pi 0.05 (versus saline). + P i 0.05 (versus arecoline alone).
maximum increase in frequency produced by pilocarpine. we investigated the possibility that pilocarpine was acting as a partial agonist and therefore wou!d antagonise the increase in frequency produced by the more efficacious agonist. arecoline (fig. 3). In these experiments, the increase in frequency of theta wave activity produced by arecoline administered 10 min after 10 mg/kg pilocarpine was significantly less than the response to arecoline alone (P < 0.05). No antagonism of the response to arecoline was produced after 3 mg/kg pilocarpine. CSF
1
P
A100
1000 AntagonIst dose (nmol I.Cv
D
3. Antagonist activity
A. Amfictal
-
00 t
0.0 Pi.
30
B. Plrenzeplne
I
Fig. 5. Comparison of antagonist potencies. Each point represents the mean ( + S.E.M.) frequency measured after arecoline (0.3 mg/kg). The squares represent responses before (0) and following (m) scopolamine and the circles represent responses before (0) and following (0) pirenzepine.
In the antagonist studies, control responses to arecoline (0.3 mg/kg i.v.) were established prior to the administration of either scopolamine, pirenzepine or aCSF. The increase in frequency produced by arecoline in all these experiments was 3.1 f 0.3 Hz (significantly different from saline, P < 0.05) and there was no significant difference between the control arecoline response prior to aCSF, scopolamine or pirenzepine. Artificial CSF failed to alter this agonist response (fig. 4A). However, although pirenzepine and scopolamine failed to significantly modify baseline theta wave frequency, both antagonists dose dependently reduced the increase in frequency produced by arecoline (fig. 4B and C). In C. Scopolamine
6.0 5.0 4.0 3.0 2.0
0.0 0 160 240
Fig 4. Effect of i.c.v. injection Of artificial CSF (A), pirenzepine (B) or scopolamine (C) on EEG responses induced by arecoline (0.3 mg/kg). The bars represent the mean ( +S.E.M.) frequency produced by saline (0. trace I). arecoline (ttt, trace 2) or arecoline following the i.c.v. administration of two doses or volumes of antagonist or vehicle (c:. trace 3 or trace 4 respectively) Antagonist doses Oqg/gI) are indicated under each bar. Statistically significant differences are indicated as * P < 0.05 (versus saline) or + P e 0.05 (versus arecoline control).
237
3.2. Agonist and antagonist effects on blood pressure
1.0
0.1 Dose
10.0
(mg/kgl
Fig. 6. Effect of muscarinic agonists on blood pressure measured after pretreatment with N-methylscopolamine. Each point represents the mean (*S.E.M.) increase in blood pressure produced by arecoline (0). oxotremorine #I). are&dine propargyl ester (e), aceclidine (A) and pilocarpine (v).
In these experiments, mean resting biood pressure ranged between 75 to 120 mm Hg. All agonists, with the exception of AF102B produced dose-related increases in blood pressure. The hypertensive effects of armline, arecaidine propargyl ester, oxotremorine, ace&dine and pilocarpine are shown in fig. 6. As in the EEG studies, arecoline, oxotremorine and are&dine propargyl ester were the most potent, with pilocarpine and aceclidine showing approximately NO-fold lower potency. Maximum responses to each agonist were not reached. AF102B failed to increase blood pressure at doses up to 20 mg/kg (data not shown). In the antagonist studies, scopolamine and pirenzepine, but not aCSF, dose dependently reduced the hypertensive response to arecoline (fig. 7). Relative antagonist potencies were estimated by plotting increases in blood pressure against nmol dose of antagonist (fig. 8). In these experiments, scopolamine was approximately 6-fold more potent than pirenzepine. 3.3. Comparison blood pressure
order to directly compare the potencies of the two antagonists, the frequency of theta wave activity measured after arecoline was plotted against antagonist dose, expressed in nmol (fig. 5). An estimation of the relative potency of scopolamine and pirenzepine indicated that scopolamine was approximately 3 times more potent than pirenzepine.
A
An~f~c~alCSF
C
70 ?
of antagonist
effects on theta rhythm and
In order to identify the selectivity of pirenzepine to antagonise either the changes in EEG or blood pressure produced by arecoline, we compared the absolute doses required to antagonise both effects. Although higher doses of pirenzepine were required to antagonise the increase in theta wave frequency, the doses used were only 2- to 3-fold higher than those required to antagonise the hypertensive effect. Similar small differences in
Scopolamine
70 F
60
.
60
60 1
P 50
40
30
20
10
0
9
Fig. 7. Effect of i.c.v. injection of artificial CSF (A), pirenzepine (B) or scopolamine (C) on hypertensive responses produced by arecoline (0.3 mg/kg). The bars represent the mean ( f S.E.M.) response produced by arecoline. alone (0) or following the i.c.v. administration of two cumulative doses or volumes of antagonist or vehicle (Coor respectively). Antagonist doses (ug/gl) are indicated under each bar. Statistically significant differences are indicated as * P -z 0.05 (versus arecoline control).
10
1
100
AntagonIstdose
1000 hmol
1C.-J1
Fig. 8. Comparison of antagonist potencies. Each point represents the mean ( f S.E.M.) increase in blood pressure measured after arecoline (0.3 mg/kg). The squares represent responses before (0) and follow) scopolamine, and the circles before (0) and following (e) pirenzepine.
entified with scopolamine (compare
the present study. we have shown that in the rat. muscarinic agonistS inr3n+maesthetised iso CW~ the ~~~e~c~ of hippocampal theta rhythm and prgs~ure by activating muscarinic recepin the brain. The evidence for the effects being ated centrally are drawn from the observations that th responses were measured in the presence of the ~~ate~ary muscarinic antagonist, N-methyllamme. which does not penetrate into the CNS and secondly. the effects were antagouised by the i.c.v. ad~~stration of ~o~la~ne. Previous studies. reporting muscarinic agonist-induced changes in type 2 theta rhythm and increases in pressure (see Introduction), have used only a range of agonists. We have extended this work by investigating the effects of a range of full and partial ‘sts with varying affinities for muscarinic receptors. most potent agonists were oxotremorine, arecoline and are&dine propargyl ester. whereas pilocarpine and ace&dine were much weaker and AFlO2B was inactive. In addition. differences in the magnitude of the responses were also identified and were particularly evident in the EEG experiments. For example, the theta rhythm recorded after the administration of oxotremarine and arecaidine propargyl ester achieved much higher fr~uencies than that measured during treatment with pi&at-pine, which actually possessed antagonist activity at the higher doses. Many factors can account for such differences. It is well recognised that the ability of au agonist to elicit a response depends not only on the affinity of the agonist for the receptor but also its efficacy. as well as other factors such as receptor density and the efficiency of coupling of the receptor to the s~mul~s-r~~nse mechanism (Kenakin, 1986). Thus, compounds with low efficacy may act as full agonists in tissues which possess an efficiently coupled stimulus-response mechanism, but in other tissues, in which the coupling to response is less efficient, they will behave as partial agonists (Ringdahl, 1987; Ringdahl et al., 1987; Freedman et al., 1988b; Fisher et al., 1983). In the present EEG studies, the obse~ation that pilocarpine behaved as a weak agonist with a low maximum respan% while oxotremorine and arecoline were more efficacious correlates well with their relative intrinsic activities in CNS tissue (Freedman et al., 1988b). Pilocarpine was shown to antagonise the EEG response to am&ne which is analogous to that reported by Fisher et al. 0983) showing antagonist properties of pilocarPine against more efficacious agonists. The inactivity of AFLO2B to increase theta wave frequency may also be a
reflection of its low intrinsic activity as reported by Nakahara et 31. (1990). In a number of in vitro preparations, we have also identified partial agonist properties of AFIOZB. and show it to be less efficacious than even pilocarpine (unpublished observations). In the present antagonist studies, we have used the M,-selective antagonist, pirenzepine to antagonise responses induced by arecoline and have measured a potency approximately 3- to &fold less than that of the non-selective antagonist, scopolamine. A number of authors have reported affinity estimates for these two antagonists in radioligand binding studies (Freedman et al., 19883; Watson et al., 1986; Lazareno et al., 1990; Hagan et al., 1987b) and have shown that although scopolamine has the highest affinity (approximate PKi = 9.2). pirenzepine is M,-selective (pK, = 8.1 on M, versus 6.4 on M,, 7.1 on M,) and therefore the difference in affinity between the two antagonists is much less on the M, subtype than on non-M, receptors (12 on M,, 630 on M,, 126 on M3). On this basis, the relatively small potency difference measured between scopolamine and pirenzepine in the present experiments supports the hypothesis that the agonist-induced increases in theta wave frequency and blood pressure are mediated through M, receptors rather than M, or M, receptors. Furthermore, in the recent report by Lazareno et al. (1990), the pharmacology of a mammalian M, receptor in rabbit lung has been detailed and on this recep tar, pirenzepine shows an affinity estimate of 7.5, intermediate to that on the M, and M, subtypes. The difference in affinity between scopolamine and pirenzepine on this subtype is still greater than that measured on the M, subtype, and this would further tend to support the involvement of receptors of the M, subtype than the M, subtype as mediators of the in vivo effects reported here. While our antagonist data suggest the involvement of M, receptors for both the EEG and cardiovascular effects, accurate comparisons of antagonist potencies are often confounded by different penetration into brain tissue from the ventricles, particularly when comparing compounds with differing lipophilicity. In this particular study, if the antagonist distribution was very different due to the greater polarity of pirenzepine than scopolamine, one would expect the relative difference in potency to be, if anything, over estimated. This would clearly not influence the inte~retation of our findings. However in the present experiments, we have identified small differences in the absolute potencies of the antagonists between the EEG and cardiovascular effects, which may be a reflection of differential penetration throughout brain tissue. Higher doses of pirenzepine were required to antagonise the EEG response than tile hypertension. Similar differences were also measured with scopolamine and therefore, although these differences are unlikely to be related to receptor
239
seiectivity, they could reflect possible differences in penetration to ~ffe~nt sites within the brain. There is indirect evidence to suggest that muscarinic agonists stimulate theta rhythm by an action in the medial septum (see Bland, 1986; Olpe et al., 1987; Stewart and Fox, 1989) while it would appear more likely for the hypertensive effects of muscarinic agonists to be mediated through the lateral septum (Scheucher et al., 1987). Thus, if muscarinic agonists were acting in the medial septum to increase theta wave frequency but in the lateral septum to increase blood pressure, one might expect the more medial areas to be less accessible for ~tago~sts after i.c.v. ad~~stration, and hence the higher doses required to antagonise the EEG response. However, a full explanation for such differences in the absolute antagonist potencies cannot be provided until brain penetra~on from the ventricular spaces and distribution in brain tissue is measured using radiolabeled antagonists. In summary, we have shown that brain-penetrating, high efficacy musearinic agonists produce increases in the frequency of hippocampal theta rhythm and increases in blood pressure. Partial agonist properties of pilocarpine have also been identified. Antagonists studies indicated that both agonist-induced responses may be mediated through muscarinic receptors of the M, subtype, although this conclusion requires support from further studies, comparing relative potencies of a wider range of selective muscarinic antagonists.
We would like to thank E. Rogers for her technical assistance, Ciba-Geigy for the generous gift of chlorisondamine and S. Guntrip for the synthesis of arecaidine pmpargyl ester and AFlO2B.
References Anwar-ul, S., H. Gilani and L.B. Cobbin, 1986, The cardioselectivity profile of himbacine: a muscarinic receptor antagonist, NaunynScluniedeb. Arch. Pharmacol. 332 16. Barlow, R-B.. K.J. Berry, P.A.M. Glenton, N.M. Nickolaou and K.S. Soh, 1976. A compa~son of affinity constants for muscarinic-sensitive acetylcholine receptors in guinea-pig atrial pacemaker cells at 29°C and in ileum at 29’C and 37°C. Br. J. Pharmacol. 58,613. Bevan, P., 1984. Effect of muscarinic ligands on the electrical activity recorded from the ~pp~ampus: a quantitative approach, Br- J. Pharmacol. 82,431. Bland, B.H.. 1986, The physiology and pharmacology of hippocampal formation theta rhythms, Prog Neurobiol. 26.1. Banner, T.L, N.J. Buckley. AC. Young and M.R. Braun, 1987, Identification of a famify of muscarinic receptor genes, Science 237, 527. Bonner, T-1.. A.C. Young, M.R. Braun and N.J. Buckley, 1988, Cloning and expression of the human and rat m5 muscarinic receptor genes, Neuron 1.403. Buckley, N.J.. T.1. Banner, CM. Buckley and M.R. Braun. 1989,
Antagonist binding properties of Bve cloned muscari& receptors expressed in CHO-Kl cells, Mol. Pharmacol. 35,469. Caulfield, M.P.. G.A. Higgins and D.W. Straughan, 1983, Central administration of the muscatinic receptor subtype-selective antagonist pircnxepine selectively impairs passive avoidance learn ing in the mouse, J. Pharm. Pharmacol. 35, 131. Day. M.D. and A.G. Roach,1977, Cardiovasctdar effects of carbachol and other cholinomitnetics administered into the cm&r& vattick of conscious cats, Clin. Exp. Pharmacol. Physiol. 4,431. Fisher,S.K., P.D. Klingerand B.W. Agranoff, 1983, Muscat&& agonistbinding and pho~o~pid turnover in brain, J. Biol. Chem. 258.7358. Freedman. S.B., H.S. Beer and EA. Harley, 1988a. M uscar& Ml, M2 receptor binding. Relationship with functional efficacy, European J. Phannacol. 156,133. Freedman, S.B., EA. Harley and L.L. iversen, 1988b. Relative affmities of drugs acting at chotinoceptors in displacing agonist and antagonist radioligands: the NMS/OXO-M ratio as an index of efficacy at cortical muscarinic receptors, Br. 1. Pharmacol. 93,437. Gene Erwin. V., A. Korte and B.C. Jones, 1988, Central muscarnnc chohnergic influences on ethanol sensitivity in long-sleep and short-sleep mice, J. Pharmacol. Exp. Ther. 247, 857. Giachetti. A., R. Micheletti and E. Montagna, 1986, Cardioselective profile of AF-DX 116, a muscarinic M2 receptor antagonist, Life Sci. 38, 1663. Giraido, E., R. Micheletti. E. Montagna, A. Giachetti. M.k Vigano, H. Ladinsky and C. Melchoirre. 1988, Binding and functional characterisation of the cardioselective muscarinic antagonist, meth~tr~ne, J. Pharmacol. Exp. Ther. 244.1016. Hagan. J.J., J.H.M. Jansen and C.L.E. Broekkamp, 1987a, Blockade of spatial learning by the Ml muscarinic antagonist pirenxepine, Psychnpharmaco!ogy 93,470. Hagan, J.J., J.A.D.M. Tonnaer, H. Rijk and C.L.E. Broekkamp, 1987b. Cholinergic stimulation of drinking from the lateral hypothalamus indications for M2 muscarinic receptor mediation. Pharmacol. B&hem. Behav. 26,771. Hagan, J.J.. J.A.D.M. Tonnaer, H. Rijk, CL-E. Broekkamp and A.M.L. Van Delft, 1987c. Facilition of ampheta~~~du~ rotation by muscarinic antagonists is correlated with M2 receptor affinity, Brain Res. 410. 69. Hammer, R.. C.P. Berrie, N.J.M. Birdsall, A.S.V. Burgen and E.C. Hulme, 1980, Pirenxepine distinguishes between different subclasses of musc~nic receptor, Nature 283.90. Hunter, A.J. and F.F. Roberts, 1988, The effect of pirenxepine on spatial learning in the Morris water maxe, Pharmacol. Biochem. Behav. 30,519. Kenakin, T.P., 1986, False premiss. Receptor resewe is a tissue misnomer, Trends Pharmacol. Sci. 7.93. Kramis, R., C.H. Vandenvolf and B.H. Bland, 1975, Two types of hippocampal rhythmical slow activity in both the rabbit and the rat: Refations to behaviour and effects of atropine. diethyl ether. urethane and pentobarbitoi. Exp. Neural. So, 58. Lang, W.J. and M.L. Rush, 1973, Cardiovascular responses to injections of cholinomimetic drugs into the cerebral ventricles of unanaesthetised dogs, Br. J. Pharrnacol. 47, 196. Lazareno, S., N.J. Buckley and F.F. Roberts, 1990, Character&ion of muscarinic M4 binding sites in rabbit lung, chick heart and NG108-15 cells, Mol. Pharmacol. 38, 805. Mutschler. E. and G. Lambrecht, 1984, Selective mUsCariniC agonists and antagonists in functional tests, Trends Pharmacol. sci. 5. Suppl. 39. Nakahara, N., N. Fujise, G. Kawanishi and F. Mizobe.19!%Central muxar$c activitiesof an MI-selective agonist: preferential effect on reversal of amnesia, Brain Res. 507. 172. olpe, HR., K. Klebs, E. Rung, P. Campiche. A. Giatt.R. Grtmann. F. DAmato,M.F. Poz~a and C. Mondadori, 1987. Chohnomimet-
rk, A.. K.-H.
&fwts
Wiedrrhold and J.M. Palacios 1956. Central pressor mdwxd h> muscarinic rwvptor agonists. evidence for a
pr&xninant role of the M2 receptor subtype. European J. Pharmacol. 125. 63. ~i~~abl. B.. 1987. Scleclivity of parti agonists related IO axotremorine based on differences in muscarinic receptor reserve hctween the guinea-pig ileum and urinary bladder, Mol. Pharmaeel. 31. 351. Rin8dahl. B.. H. Roth and D.J. Jenden. 1987. Regional differences in nx~plor rtxwvr for analogs of ox.otremorine in viva: Implicaiions for development of selective muscarinic agonists. J. Pharmacol. Exp. Ther. 242.464. Robinson. T.E. 1980. Hippocampal rhythmic slow activity (RSA; thets): a critical analysis of srlccted studies and discussion of w-ible species differences. Brain Res. Rev. 2, 69. Scheucber, A.. C.J. Pirola. MS. Bslda. S.M. Dabsys. A.L. Alvarez. S.
Finkielman and V.E. Nahmod. 1987. Musfdrinic Ml receptors In the lateral septal area mediate cardiovascular responses to cholinergic agouists and bradykinin: supersensitivty induced by chronic treatment with atropine. Neuropharmacology 26. 181. Stewart. B.R.. P. Jenner and C.D. Marsden, 1989. Assessment of the muscarinic receptor subtype involved in the mediation of pilocarpine-induced purposeless chewing behaviour. Psychopharmacology 97. 228. Stewart. M. and S.E. Fox. 1989. Two populations of rhythmically bursting neurons in rat medial septum are revealed by atropine. J. Neurophysiol. 61. 982. Watson, M.. W.R. Roeske and H.I. Yamamura. 1986, [‘HJPirenzepine and [ “Hlquinuclidinyl benzilate binding to rat cerebral cortical and cardiac muscarinic cholinergic sites. II. Characterisation and regulation of antagonist binding to putative muscarinic subtypes. J. Pharmacol. Exp. Ther. 237. 419. Wu. Y.Y. and E.T. Wei, 1983. Mechanisms underlying the pressor responses to acute and chronic inlraventricular administration of carbachol in the rat, J. Pharmacol. Exp. Ther. 228, 354.
