Baclofen activates two distinct receptors in the rat spinal cord and guinea pig ileum

NeuropharmacologyVol.25, No.7, pp.795-798, 1986 Printed in Great Britain

0028-3908/86 $3.00 + 0.00 Pergamon Journals Ltd

PRELIMINARY

NOTES

BACLOFEN ACTIVATES TWO DISTINCT RECEPTORS IN THE RAT SPINAL CORD AND GUINEA PIG ILEUM J. Department of Halifax,

Sawynok

Pharmacology Nova Scotia

Dalhousie University, Canada, B3H 4H7

(Accepted 18 Ma&

19861

intrathecal injection, pretreatment summary - Following with both D-baclofen and 5-aminovaleric acid (5-AV) inhibited the antlnoclceptive effect of L-baclofen, but homotaurine (3-aminopropane sulphonic acid, APS) was inactive as an antagonist (rank order D-baclofen > 5-AV > APS=O). In an established GABAB system, the electrically stimulated guinea pig longitudinal muscle myenteric plexus preparation, APS and 5-AV but not D-baclofen reduced the inhibitory effect of L-baclofen (APS=5-AV > Receptors with which baclofen interacts D-baclofen=O). in the spinal cord to produce antinociception differ from GABAB receptors with respect to the rank order of potency of antagonists as well as close structural and these criteria could be used for characterianalogs, zation of such receptors. Baclofen is considered a prototype agonist for novel GABAB This classireceptors in the peripheral and central nervous systems. fication is made on the basis of (a) mimicry by GABA, (b) cross-desensitization between baclofen and GABA, (c) insensltivity to the GABAA antagonist bicuculline (Bowery et al., 1981) and (d) the results of It has not relied on the use of binding studies (Bowery et al., 1983). Recently, s-aminoantagonists because these have not been available. 1983) and homotaurine (3-aminevaleric acid (S-AV) (Muhyaddin et al., propane sulphonic acid, APS) (Giotti et al., 1983b) were reported to be while D-baclofen antagonized GABAB antagonists in peripheral tissues, the antinocfceptive effect of L-baclofen (Sawynok 6 Dickson, 1985). In the action of L-baclofen appears due to activation of the latter case, The a receptor other than the GABAB receptor (see Discussion). present study was undertaken to compare the relative potency of baclofen antagonists in spinal antinociception experiments and in an estabReceptors in this tissue lished GABAB sys tern, the guinea pig ileum. fulfill criteria (a)-(c) above (Bowery et al., 1981; Giotti et al., 1983a), and antagonism with both 5-AV and APS has been demonstrated (Giottl et al., 1983b; Kerr and Ong, 1984). METHOD Antlnociception implanted with space (Sawynok

Rats (male Sprague-Dawley 300-350 g) were experiments: chronic indwelling cannulas in the spinal subarachnoid Nociceptive thresholds were deterand Dickson, 1985).

mined using the tail flick test. Baseline latencies were 2-3 set, and a cut-off latency of 10 set was imposed. Rats were accommodated in plastic containers for intrathecal injections and tail flick determinations. All drugs were injected in 15 pl saline and flushed in with 10 til saline. See Sawynok and Dickson (1985) for futher details of these experiments. Guinea pig ileum experiments: Male guinea pigs (300-400 g) were used. The guinea pig ileum longitudinal muscle myenteric plexus preparation (Sawynok and Jhamandas, 1976) was suspended in a 2 ml organ bath and stimulated at 6/min at 80% of the maximal twitch height. Under these conditions, a 40% inhibition of the twitch response could be obtained. Drugs were added in a fixed order (see Fig. 2) and a 30 minute rest with washes permitted between drug additions to minimize desensitiaation (Giotti et al., 1983b), but some reduction in control responses did occur. The effect of L-baclofen was calculated as a percentage of the twitch height in the presence of the antagonist. Cumulative volumes of drug added to the bath were less than 3% of the total volume. 795

Preliminary Notes

796

120

100

20

0

Figure

1.

I

a

I

I

1

a

40 20 30 DOSE ANTAGONIST

IO

1

(

60 nmole

I

1

200

100

)

Comparative activity of antagonists in inhibiting the antinociceptive effect of L-baclofen injected intrathecally to (0) D-Baclofen (data recalculated from Sawynok and rats. 1985 and is included for the purpose of comparing Dickson, two supplies of D-baclofen), (0) D-Baclofen (R), (A) S-aminovaleric acid, (0) homotaurine (3-aminopropane sulphonic acid), (V) GABA. Data is expressed as a percentage of the control Antinociceptive Index (sum of tail flick readings after L-baclofen injection minus value prior to injection), which in all experiments ranged from 8-11 sec. Values are means f 8.e.m. for n = 4-9 (n=3 for GABA). Dashed line indifor control values converted to a cates lower range of 8.e.m. **p< 0.005 compared to control percentage score. * p< 0.05, values in individual experiments (Analysis of variance and Student Newman Keuls test).

\ 30

/1.: 20

10

, i

,

4

CUMULATIVE L BACLOFEN

14

DOSE ( ,tM )

44

0

-liooo

-loo DOSE ANTAGOINST

-loo (,,M

)

evoked twitch response of the Inhibition of elect ri ,cally guinea pig longitudinal muscle-myenteric plexus preparation by L-baclofen added cumulatively (A) and by taclofen 10 PM in the absence and presence of (B) S-aminovaleric acid (C) homotaurine (3-aminopropane sulphonic acid) and (D) D-bacloValues are means l s.e.m. for n=6-8 except for A fen (RI. Shaded areas indicate range of 8.e.m. of control where n=21. ** p < 0.001 compared to * p< 0.01, responses to L-baclofen. control values (ANOVA and Student Newman Keuls test). D-baclofen and L-baclofen were supplied by Ciba Geigy, Basel, Drugs: Switzerland, and 3-APS and S-AV by the Aldrich Chemical Company. Previously, D-baclofen which was approximately 99% pure was used (Sawynok & Dickson, 1985), but a repurified form is now available and this supply (D-baclofen(R) for repurified) was used in the present study .

Figure

2.

Preliminary

Notes

RESULTS Pretreatment with D-baclofen inhibited the antinociceptive effect of L-baclofen In the tail flick test when both agents were injected However, D-baclofen intrathecally (Sawynok 6 Dickson, 1985). produced and the effect of L-baclofen was assessed in the intrinsic activity, presence of such activity. In the present study, a repurified (R) supply of D-baclofen has been used. D-Baclofen(R), up to 10 pg, lacked intrinsic activity but blocked the effect of a subsequent dose of L-baclofen in a dose related manner (Fig. 1). Pretreatment with 5-AV also inhibited L-baclofen in these experiments but it was less potent Pretreatment with APS and GABA were without than D-baclofen (Fig. 1). antagonist effects (Fig. 1). S-AV, APS and GABA did not exhibit intrinsic activity in the tail flick test. The cumulative addition of L-baclofen to the guinea pig longitudinal muscle-myenteric plexus preparation produced a dose related inhibition of the electrically stimulated twitch response (Fig.ZA). 5-AV 300 PM and 1000 uM reduced the inhibitory effect of L-baclofen (Fig.ZB). These doses produced a transient contracture which was completed well before L-baclofen was added, and a slight inhibition (< 10%) of twitch height. APS 300 uM and 1000 PM also inhibited the effect of L-baclofen (Fig. 2C), produced a transient contracture and inhibited the twitch by < 10%. (In both cases, contractures are due to activation of GABAA receptors (Bowery and Brown, 1974; Giotti et al., 1983a)). D-Baclofen(R) lacked antagonist activity at 300 PM, while at 1000 fl only 3 of the 7 preparations were reduced (to 53, 39, 37% of control). The net effect at this dose, however, was not significant (Fig. 2D). These doses of D-baclofen(R) produced 9.6-12.9% inhibition of the twitch hefght, and higher doses were not tested because of marked intrinsic inhibitory effects. DISCUSSION The present study confirms that pretreatment with D-baclofen inhibits the antinociceptive effect of L-baclofen when injected intrathetally (Sawynok 6 Dickson, 1985). The repurified supply of D-baclofen used in this study lacked intrinsic activity suggesting that intrinsic activity observed previously was due to contamination with the L-isomer. Of other putative antagonists evaluated, only 5-AV showed antagonist actlvi ty . The rank order of potency was D-baclofen> 5-AV > APS=O. In the guinea pig ileum, inhibitory effects of L-baclofen are mediated by GABAB receptors (see Introduction). We confirm antagonism of L-baclofen by 5-AV and APS in this preparation (Glotti et al., 1983b; Kerr and Ong, 1984), with approximate equivalent potency as antagonists, in agreement with similar pA2 values reported for them (4.1 for 5-AV, Muhyaddln et al., 1982; 4.2 for APS, Giotti et al., 1983b). However, D-baclofen(R) was not a good antagonist in the guinea pig ileum. The rank order of potency of antagonists in the ileum was APS = 5-AV > D-baclofen = 0. It has been suggested that the spinal baclofen receptor which mediates the antinociceptive effect of baclofen is not a GABAB receptor (Sawynok & Dickson, 1985). This was based on observations that GABA did not mimic the effect of baclofen following intrathecal injection, and the rank order of potency of analogs of baclofen (p-chlorophenyl GABA), in particular m-chlorophenyl GABA (MCPG) and o-chlorophenyl GABA (OCPG) which differed markedly in the spinal antinoclceptlon system (MCPG > OCPG) and at GABAB receptors (OCPG > MCPG). The results of the present study, i.e. the rank order of potency of antagonists differs for the spinal cord and guinea pig ileum, can be considered a third major difference between the two systems and supports the notion that baclofen is activating distinct receptor populations in these preparations. It has already been argued that simple mimicry of an effect of baclofen by GABA and bicuculllne insensitivity are insufficient criteria to implicate GABAB receptors in a pharmacological It is proposed that the rank action of baclofen (Fung et al., 1985). order of potency of the ortho and meta chloro substituted analogs of baclofen and of the antagonists APS, 5-AV and D-baclofen be evaluated prior to implicating baclofen or GABAB receptors in such actions.

797

798

Preliminary Notes

ACKNOWLEDGEMENTS I am particularly grateful to Drs. Gagneaux and Bencze at Ciba and the Geigy, Base1 for their continued supplies of L- and D-baclofen I wish to thank C. provision of a repurified form of D-baclofen. Dickson and A. Reid for technical assistance in the intrathecal experiThis work was supported by the MRC of Canada. ments. REFERENCES aowery, N.G. and Brown, D.A. (1974) Depolarization action of GABA and related compounds on rat superior cervical ganglion in vitro. -Br. J. Pharmac. 50:205-218. Bowery, N.G., Doble, A., Hill, D.R., Hudson A.L., Shaw, J.S., Turnbull, M.J. and Warrington, R. (1981) Bicuculline-insensitive GABA Eur. J. receptors on peripheral autonomic nerve terminals. Pharmacol. 71: 53-70 Hill, D.R. and Hudson, A.L. (1983) Characteristics of Bowery, N.G., GABAB receptor binding sites on rat whole brain synaptic membrances. Br. J. Pharmac. 78:191-206. Swarbrick, M.J. and Fillenz, M. (1985) Effect of baclofen Fung, S.C., on -in vitro noradrenaline release from rat hippocampus and cereNeurochem. Int. bellum: an action at an a2-adrenoceptor. 7:155-163. Glotti, A., Luzzi, S., Spagnesi, S. and Zllletti, L. (1983a) GABAA and GABAB receptor-mediated effects in guinea pig ileum. Br. J. Pharmac. 78:469-478. S. and Zilletti, L. (1983b) HomotauGiotti, A., Luzzl, S., Spagnesl, rine: a GABAB antagonist in guinea-pig ileum. Br. J. Pharmac. 79:855-862. Evidence that ethylenediamine acts In Kerr, D.I.B., and Ong, J. (1984) the isolated ileum of the guinea-pig by releasing endogenous GABA. Br. J. Pharmac. 83:169-177. Muhyaddin, M., Roberts, P.J. and Woodruff, G.N. (1982) Presynaptic y-aminobutyric acid receptors in the rat anococcygeus muscle and their antagonism by 5-aminovaleric acid. Br. J. Pharmac. 77~163-168. Sawynok, J. and Dickson, C. (1985) D-Baclofen is an antagonist at baclofen receptors mediating antinociception in the spinal cord. Pharmacology. 31:248-259. K.H. (1976) Inhibition of acetylcholine Sawynok, J. and Jhamandas, release from cholfnergic nerves by adenosine, adenine nucleotides and morphine: antagonism by theophylline. J. Pharmacol. Exp. Ther. 197: 179-190