Introduction

Top Abstract Introduction Methods Results Discussion References

The activation of the cholinergic system induces antinociception in laboratory animals (Pedigo et al., 1975; George et al., 1962; Herz, 1962; Hendershot and Forsaith, 1959; Harris et al., 1969) and humans (Hood et al., 1995). Bartolini et al. (1992) demonstrated that muscarinic analgesia in mice and rats is mediated by postsynaptic M₁ receptors. These authors reported that M₁-selective agonists McN-A-343 and AF-102B were able to produce a significant enhancement of the pain threshold, whereas the M₂-selective agonist arecaidine propargil ester (APE) was not. Moreover, Bartolini et al. (1992) have demonstrated that the M₁ antagonists dicyclomine and pirenzepine, contrary to the M₂ antagonist AF-DX 116, antagonized antinociception induced by both unselective (oxotremorine) and M₁-selective (McN-A-343, AF-102B) muscarinic agonists. It has also been reported that the antimuscarinic drug atropine, at very low doses, produces a cholinomimetic effect by inducing a central cholinergic antinociception in laboratory animals regardless of the route of administration and the noxious stimulus applied (Ghelardini et al., 1990). This paradoxical effect of atropine confirmes the previous observations made by Ferguson-Anderson (1952) that reported that the tincture of belladonna in small doses, given by mouth, had a parasympathomimetic action increasing the frequency and amplitude of gastric contractions.

The typical cholinergic symptomatology (tremors, sialorrhea, diarrhea, lacrimation, etc.) did not accompany the antinociceptive activity of atropine. The atropine-induced increase in the pain threshold was attributable to the R-(+)-enantiomer of atropine, R-(+)-hyoscyamine, because S-()-hyoscyamine was ineffective in all antinociceptive tests used (Ghelardini et al., 1992). The investigation of the antinociceptive effect of atropine demonstrated, by microdialysis techniques, that R-(+)-hyoscyamine, at effective doses, produced an increase in the ACh release from the rat cerebral cortex in vivo (Ghelardini et al., 1997). On the basis of the above-mentioned results, the racemate (Gualtieri et al., 1994) and the enantiomers (Romanelli et al., 1995) of the compound, PG-9, structurally related to atropine (fig. 1), have been synthesized to obtain a new cholinergic amplifier endowed with more intensive antinociceptive activity than atropine but as lacking in cholinergic side effects as atropine. For this purpose, (±)-PG-9 antinociceptive properties were investigated by use of the hot-plate, abdominal-constriction, paw-pressure and tail-flick tests, whereas the incidence of behavioral side effects was detected by the rota-rod test and Animex apparatus. Furthermore, the central cholinergic system has long been known to be involved in the modulation of learning and memory processes in animals and man. Drugs that affect the central cholinergic system have been found either to enhance or to hinder performance in learning and memory tests. Direct muscarinic agonists (oxotremorine, arecoline, AF-102B, RS 86, etc.), acetylcholine esterase inhibitors (physostigmine, diisopropyl fluorophosphate, eptastigmine, tacrine, etc.) and acetylcholine releasers (AFDX 116, DuP 996, etc.) potentiate test performance retention in rodents (Coyle, 1995). On the contrary, disruption of the cholinergic system impairs cognitive processes. The administration of muscarinic antagonists (scopolamine, atropine, pirenzepine and dicyclomine), inhibitors of choline uptake (hemicolinium-3) or lesions of nucleus basalis magnocellularis or injection of the cholinotoxic agent AF64A, all induce amnesia (Coyle, 1995). Considering that R-(+)-hyoscyamine was able to prevent amnesia induced by both scopolamine and dicyclomine in mice (Ghelardini et al., 1997), the potential antiamnesic activity of (±)-PG-9 was investigated with the mouse passive-avoidance test.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Chemical structure of PG-9 racemate and enantiomers.

	Methods

Top Abstract Introduction Methods Results Discussion References

Animals

Male Swiss albino mice (23-30 g) and Wistar rats (200-300 g) from Morini (San Polo d'Enza, Italy), Fisher 344 rats (200-300 g ) from Charles River (Calco, Italy) and guinea pigs (150-200 g) from Rodentia (Bergamo, Italy) breeding farms were used. Fifteen mice and four rats or guinea pigs were housed per cage. The cages were placed in the experimental room 24 h before the test for acclimatization. The animals were kept at 23 ± 1°C with a 12-h light/dark cycle, light at 7 A.M., with food and water ad libitum. All experiments were carried out according to the guidelines of the European Community Council.

Analgesic Tests

Hot-plate test. The method adopted was described by O'Callaghan and Holzman (1975). Mice were placed inside a stainless steel container, thermostatically set at 52.5 ± 0.1°C in a precision water bath from KW Mechanical Workshop, Siena, Italy. Reaction times (s), were measured with a stop-watch before and at regular intervals up to a maximum of 45 min after treatment. The endpoint used was the licking of the fore or hind paws. Those mice scoring less than 12 and more than 18 s in the pretest were rejected (30%). An arbitrary cut-off time of 45 s was adopted.

Abdominal-constriction test. Mice were injected i.p. with a 0.6% solution of acetic acid (10 ml kg¹), according to Koster et al. (1959). The number of stretching movements was counted for 10 min, starting 5 min after acetic acid injection.

Paw-pressure test. The nociceptive threshold in the rat and guinea pig was determined with an analgesimeter (Ugo Basile, Varese, Italy), according to the method described by Leighton et al. (1988). Threshold pressure was measured before treatment and 15, 30 and 45 min after treatment. Rats and guinea pigs scoring less than 30 g or more than 85 g during the test and before drug administration were rejected (25%). An arbitrary cut-off value of 250 g was adopted.

Tail-flick test. An analgesimeter from Ugo Basile (Varese, Italy) was used to perform the tail- flick test described by D'Amour and Smith (1941). The light from a project bulb, situated beneath the platform where the rat was placed, was focused through a small hole on the ventral part of the tail at a point about 4 cm from the tip. Withdrawal of the tail exposed a photocell to the light, which turned off the thermal stimulus and automatically stopped the clock. The intensity was regulated so that the reaction time varied between 2 and 4 s. The analgesia was tested before and 15, 30 and 45 min after treatment of rats. Each value was derived from the mean of three consecutive readings in which the light was focused on three adjacent points of the tail. An arbitrary cut-off value of 10 s was adopted.

Antiamnesic Test: Passive-Avoidance Test

The test was performed according to the step-through method described by Jarvik and Kopp (1967), as we modified it for testing drugs endowed with analgesic properties. The apparatus consists of a two-compartment acrylic box with a lighted compartment connected to a darkened one by a guillotine door. In the original method mice received a punishing electrical shock as soon as they entered the dark compartment, whereas in our modified method, after entry into the dark compartment, mice receive a nonpainful punishment consisting of a fall into a cold water bath (10°C). For this purpose the dark chamber was constructed with a pitfall floor. The latency times for entering the dark compartment were measured in the training test and after 24 h in the retention test. For memory disruption, mice were injected i.p. with the amnesic drugs scopolamine and dicyclomine. (±)-PG-9, physostigmine and piracetam were injected 20 min before the training session, whereas scopolamine and dicyclomine were injected immediately after termination of the training session. The maximum entry latency allowed in the retention session was 120 s. The memory degree of received punishment (fall into cold water) was expressed as the increase in seconds between training and retention latencies.

Additional Behavioral Tests

Spontaneous activity meter (Animex). Locomotor activity in mice was quantified with an Animex activity meter Type S (LKB, Farad, Sweden) set to maximum sensitivity. Mice were placed on the top of the Animex activity meter and each movement produced a signal caused by variation in inductance and capacity of the apparatus resonance circuit. These signals were automatically converted to numbers. On the day of the experiment the mice were treated and then the cage, containing five mice, was put on the measurement platform. Activity counts were made every 15 min for 45 min, starting immediately after injection of the drug. Because of the arbitrary scale adopted to quantify movements, drug-treated mice were always compared with saline-treated ones.

Rota-rod test. The apparatus consisted of a base platform and a rotating rod of 3-cm diameter with a nonslippery surface. The rod was placed at a height of 15 cm from the base. The rod, 30 cm in length, was divided into five equal sections by six disks. Thus, up to five mice were tested simultaneously on the apparatus, with a rod-rotating speed of 16 r.p.m. The integrity of motor coordination was assessed based on the number of falls from the rod in 30 s according to Vaught et al. (1985). The performance time was measured before and 15, 30 and 45 min after treatment.

In Vitro Functional Studies

Isolated rabbit vas deferens (M₁). Experiments on isolated rabbit vas deferens were performed according to the method described by Eltze (1988) and modified by Dei et al. (1995). The preparations were maintained at 32°C and tissues were stimulated through platinum electrodes by square-wave pulses (2 ms, 0.1 Hz, 10-30 V). Contractions were measured isometrically after tissues had been equilibrated for 1 h, then a cumulative dose-response curve for the inhibitory effect of McN-A-343 was plotted.

Isolated guinea-pig left atria (M₂). Isolated left atria were prepared according to the method described by Eltze et al. (1985) and modified by Dei et al. (1995). Bath fluid temperature was maintained at 30°C. Atria were stimulated electrically (1 Hz, 1 ms, 4-10 V) by means of two platinum electrodes. Carbachol negative inotropic effects on isometric atria contractions were recorded before and 1 h after perfusion with antagonists.

Isolated guinea pig ileum (M₃). Isolated ileum fragments were prepared according to Eltze and Figala (1988). Bath fluid temperature was maintained at 37°C. Isotonic ileum contractions induced by ACh were recorded before and 1 h after perfusion with antagonists.

Guinea pig isolated uterus (M₄). Experiments on isolated immature guinea pig uterus were performed according to Dörje et al. (1990). The preparations were maintained at 30°C and after 1 h equilibration period isotonic contractions to carbachol were recorded. Initially the tissues were exposed to a single concentration of carbachol (3 nmol l¹) to check the responsiveness to the agonist, and a dose-response curve for carbachol was obtained.

Determination of antagonist affinities. After a stabilization for 30 to 60 min, agonist concentration-response curves were plotted before and after equilibration with antagonists. In separate control experiments no significant changes in tissue sensitivity to the agonist were observed during the period required for the determination of two concentration-response curves. The antagonists were allowed to equilibrate for 60 min. No more than two concentrations of antagonist were tested in the same preparation. Agonist EC₅₀ values in the absence and presence of antagonists were determined graphically for the calculation of dose ratios.

Acetylcholinesterase activity. Acetylcholinesterase activity was assayed according to Ellman et al. (1961), with 0.5 mM acetylthiocholine iodide as substrate. The (±)-PG-9 inhibitory effect was tested at various concentrations on a purified preparation of acetylcholinesterase from the electric eel.

Drugs

The following drugs were used: PG-9 racemate was prepared according to Gualtieri et al. (1994); R-(+)-PG-9 and S-()-PG-9 were prepared according to Romanelli et al. (1995); R-(+)-hyoscyamine was prepared according to Gualtieri et al. (1991); SDZ 205557 was prepared in the Department of Pharmaceutical Sciences of the University of Florence, Italy, according to the method described by Romanelli et al. (1993); atropine sulfate, carbamylcholine chloride, physostigmine hemisulfate and yohimbine hydrochloride (Sigma, Milan, Italy), HC-3, pirenzepine dihydrochloride, naloxone hydrochloride, quinpirole hydrochloride, (R)--methylhistamine dihydrochloride, NAN 190, McN-A-343 (R.B.I., Milan, Italy); acetylcholine chloride (Merck, Rome, Italy); morphine hydrochloride (U.S.L. 10/D, Florence, Italy), diphenhydramine hydrochloride and AFDX-116 (De Angeli, Milan, Italy); clomipramine hydrochloride (anafranil), CGP 35348 and reserpine (Ciba Geigy, Basel, Switzerland); dicyclomine hydrochloride (Le Petit, Milan, Italy). Other chemicals were of the highest quality commercially available. All drugs were dissolved in isotonic (0.9% NaCl) saline solution or dispersed in 1% sodium carboxymethylcellulose immediately before use, except reserpine which was dissolved in a 20% solution of ascorbic acid and R-(+)-hyoscyamine that was dissolved in 0.1 M HCl and then diluted with saline (1:10). Drug concentrations were prepared so that the necessary dose could be administered in a volume of 10 ml kg¹ by s.c., i.p. and p.o. routes or 5 ml kg¹ by the i.v. route. Intracerebroventricular administration was performed under ether anesthesia with isotonic saline as solvent, according to the method described by Haley and McCormick (1957) for mice and which we adapted for rats. During anesthesia, mice and rats were grasped firmly by the loose skin behind the head. A 0.4-mm external diameter hypodermic needle attached to a 10-µl syringe was inserted perpendicularly through the skull at a depth of no more than 2 mm into the brain of the mouse and 4 mm into the brain of the rat, where 5 µl (mice) or 10 µl (rats) were then administered. The injection site was 1.5 mm (mice) or 2.5 mm (rats) from either side of the middle of a line drawn through to the anterior base of the ears. To ascertain that the drugs were administered exactly into the cerebral ventricle, some mice and rats were injected i.c.v. with 5 to 10 µl of diluted 1:10 Indian ink and their brains examined macroscopically after sectioning. The accuracy of the injection technique in both mice and rats was evaluated, and 95% of the injections were correct.

Statistical Analysis

Results are given as the mean ± S.E.M.; analysis of variance, followed by Fisher's Protected Least Significant Difference procedure for post hoc comparison, was used to verify the significance between two means. P values of less than .05 were considered significant. Data were analyzed with the StatView for the Macintosh computer program (1992).

	Results

Top Abstract Introduction Methods Results Discussion References

Antinociceptive activity of PG-9. (±)-PG-9, as shown in figure 2, produced a dose-dependent increase in the pain threshold in the mouse hot-plate test after s.c. (10-40 mg kg¹; fig. 2A), i.c.v. (10-30 µg/mouse; fig. 2B), p.o. (30-60 mg kg¹; fig. 2C) and i.v. (10-30 mg kg¹; fig. 2D) administration. The antinociceptive effect of (±)-PG-9, regardless of the route of administration, peaked 15 min after injection and then slowly diminished. Figure 3, A and B, illustrates the analgesic effect of (±)-PG-9 in the mouse acetic acid abdominal constriction test. (±)-PG-9 induced an increase in the pain threshold in a dose-dependent manner starting from the dose of 10 mg kg¹ s.c. (fig. 3A). (±)-PG-9 showed antinociceptive properties also after the injection of 10 to 30 µg/mouse i.c.v. (fig. 3B).

View larger version (35K): [in this window] [in a new window]   Fig. 2.   Dose-response curves of (±)-PG-9 administered s.c. (A), i.c.v. (B), p.o. (C) and i.v. (D) in the mouse hot-plate test. The doses are expressed as milligrams per kilogram s.c., p.o. and i.v. and as micrograms per mouse i.c.v. Vertical lines show S.E.M.  P < .05; *P < .01 in comparison with saline controls. Each point represents the mean of at least 10 mice.

View larger version (39K): [in this window] [in a new window]   Fig. 3.   (A) Dose-response curve of (±)-PG-9 administered s.c. and effect of atropine (5 mg kg-1 i.p.), dicyclomine (10 mg kg-1 i.p.) and reserpine (2 mg kg-1 i.p.) pretreatments on antinociception induced by (±)-PG-9 (20 mg kg-1 s.c.) in the mouse abdominal constriction test induced by 0.6% acetic acid. Atropine, dicyclomine and reserpine were injected, respectively, 15 min, 15 min and twice 48 and 24 h before (±)-PG-9 administration. The nociceptive responses were recorded 15 min after (±)-PG-9 administration. (b) Dose-response curve of (±)-PG-9 administered i.c.v. Vertical lines show S.E.M  P < .05; *P < .01 in comparison with saline controls. °P < .01 in comparison with (±)-PG-9 (20 mg kg-1 s.c.). Numbers inside the columns indicate the number of mice.

1

1

1

View larger version (30K): [in this window] [in a new window]   Fig. 4.   Dose-response curve of i.p. (±)-PG-9 in the rat (A) and guinea-pig (B) paw pressure test and antagonism exerted by atropine (5 mg kg-1 i.p.; C) and HC-3 (1 µg/mouse i.c.v.; D) in the rat paw pressure test. Atropine and HC-3 were injected respectively 15 min and 5 h before (±)-PG-9 administration. Vertical lines show S.E.M. *P < .01 in comparison with saline controls; °P < .01 in comparison with (±)-PG-9 (30 mg kg-1 i.p.). Each point represents the mean of at least 5 rats and 4 guinea pigs. The doses of (±)-PG-9 are expressed as milligrams per kilogram i.p.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Dose-response curves of (±)-PG-9 administered i.p. in the Wistar rats (A) and in the Fisher 344 rats (B) in the tail-flick test. The doses are expressed as milligrams per kilogram i.p. Vertical lines show S.E.M. *P < .01 in comparison with saline controls. Each point represents the mean of at least six rats.

View larger version (42K): [in this window] [in a new window]   Fig. 6.   Dose-response curves of R-(+)-PG-9 (A, C) and S-()-PG-9 (B, D) in the mouse hot-plate test and abdominal-constriction test. The doses are expressed as milligrams per kilogram s.c. Vertical lines show S.E.M.  P < .05; *P < .01 in comparison with saline controls. The number of mice is shown in columns.

1

1

1

1

1

View larger version (51K): [in this window] [in a new window]   Fig. 7.   Area under the curve (AUC) of the analgesic effect of (±)-PG-9 in comparison with R-(+)-hyoscyamine, morphine, diphenhydramine and clomipramine in the mouse hot-plate test. AUC represents the sum of four measurements. Antinociceptive responses were evaluated every 15 min and exactly 15, 30, 45 and 60 min after administration for R-(+)-hyoscyamine, diphenhydramine and clomipramine and 30, 45, 60 and 75 min after morphine injection. Numbers inside the columns indicate the number of mice.

Antagonism of the (±)-PG-9 induced antinociception. In the mouse hot-plate test, the antinociceptive effect of (±)-PG-9 (30 mg kg¹ s.c.) was not antagonized by naloxone (1 mg kg¹ i.p.; fig. 8D), CGP-35348 (2.5 µg/mouse i.c.v.), (R)--methylhistamine (10 mg kg¹ i.p.), quinpirole (0.1 mg kg¹ i.p.), SDZ-205557 (10 mg kg¹ i.p.), NAN 190 (0.5 µg/mouse i.c.v.) (data not shown) and, in the abdominal-constriction test, by reserpine (2 mg kg¹ i.p.) (fig. 3). Conversely, atropine (5 mg kg¹ i.p.), pirenzepine (0.1 µg/mouse i.c.v.) and hemicolinium-3 (1 µg/mouse or rat i.c.v.) were able to completely prevent (±)-PG-9 antinociception in the mouse hot-plate (fig. 8, A-C), abdominal-constriction (fig. 3A) and rat paw-pressure tests (fig. 4, C and D). All antagonists were injected 15 min before (±)-PG-9, with the exception of reserpine (injected twice 48 and 24 h before the test), HC-3 (injected 5 h before the test) and CGP 35348 (injected 5 min before (±)-PG-9).

View larger version (32K): [in this window] [in a new window]   Fig. 8.   Effect of atropine (5 mg kg-1 i.p.; A), pirenzepine (0.1 µg/mouse i.c.v.; B), HC-3 (1 µg/mouse i.c.v.; C) and naloxone (1 mg kg-1 i.p.; D) on antinociception induced by (±)-PG-9 in the mouse hot-plate test. Atropine, pirenzepine, HC-3 and naloxone were injected, respectively, 15 min, 5 min, 5 h and 15 min before (±)-PG-9 administration. Vertical lines show S.E.M. *P < .01 in comparison with saline controls; °P < .01 in comparison with (±)-PG-9 (30 mg kg-1 s.c.). Each point represents the mean of at least 10 mice. The doses of (±)-PG-9 are expressed as milligrams per kilogram s.c.

Antiamnesic activity of (±)-PG-9. Pretreatment with (±)-PG-9 (10-30 mg kg¹ i.p.), injected 20 min before the training session, prevented the amnesia induced by scopolamine (1 mg kg¹ i.p.) and dicyclomine (2 mg kg¹ i.p.) in the mouse passive-avoidance test. (±)-PG-9 enhanced the entrance latency up to a value similar to that produced by control animals (fig. 9). (±)-PG-9, at 1 mg kg¹ i.p., was completely ineffective (fig. 9). The antiamnesic effect of (±)-PG-9 was equal to that produced by the cholinesterase inhibitor physostigmine (0.2 mg kg¹ i.p.) and the nootropic drug piracetam (30 mg kg¹ i.p.).

View larger version (56K): [in this window] [in a new window]   Fig. 9.   Effect of i.p. (±)-PG-9 in comparison with piracetam and physostigmine on scopolamine and dicyclomine induced amnesia in mouse passive-avoidance test. Punishment consists of a fall into cold water. (±)-PG-9, piracetam and physostigmine were injected 20 min before the training session. Scopolamine and dicyclomine were injected immediately after the training session.  P < .05; *P < .01 in comparison with saline controls; °P < .01 and **P < .01 in comparison with dicyclomine- and scopolamine-treated mice, respectively. Each column represents the mean of at least 29 mice.

Evaluation of the PG-9 effect on spontaneous activity and motor coordination. The motor coordination of mice treated with (±)-PG-9, R-(+)-PG-9 and S-(-)-PG-9 was evaluated by use of the rota-rod test (table 1), whereas their spontaneous activity was investigated by use of the Animex apparatus. The rota-rod performance of mice treated with (±)-PG-9 at the dose of 40 mg kg¹ s.c., 30 µg/mouse i.c.v., 60 mg kg¹ p.o. or 30 mg kg¹ i.v., and both enantiomers at the dose of 30 mg kg¹ s.c. was not impaired compared with controls (table 1). On the contrary, (±)-PG-9 administered at higher doses (50 and 60 mg kg¹ s.c., 40 µg/mouse i.c.v., 80 mg kg¹ p.o. or 50 mg kg¹ i.v.) as well as R-(+)-PG-9 (40 mg kg¹ s.c.) and S-()-PG-9 (40 mg kg¹ s.c.) significantly impaired the rota-rod performance (table 1). The number of falls by control animals progressively decreased at every measurement because the mice learned how to balance on the rotating rod. The spontaneous motility of mice was not modified by treatment with (±)-PG-9 (30 and 40 mg kg¹ s.c.) as revealed by the Animex apparatus (data not shown).

View this table: [in this window] [in a new window]   TABLE 1 Effect of (±)-PG-9, R-(+)-PG-9 and S-()-PG-9 in the mouse rota-rod test

In vitro functional studies. (±)-PG-9 blocked the McN-A-343-induced inhibition of twitch contractions of the rabbit vas deferens (pK_B = 6.71 ± 0.05), antagonized the negative inotropic carbachol-induced effect in the guinea pig left atrium (pK_B = 6.85 ± 0.10), the contractile responses to acetylcholine in guinea pig ileum (pK_B = 6.84 ± 0.08) and to carbachol in immature guinea pig uterus (pK_B = 7.72 ± 0.05) as shown in table 2. Increasing concentrations of (±)-PG-9 produced parallel shifts of the agonist concentration-response curves progressively to the right and no appreciable change in basal tension or maximum agonist response was observed (data not shown). pA₂ values of R-(+)-hyoscyamine and AFDX-116, used as reference drugs, are shown in table 2. The selectivity ratios for (±)-PG-9, R-(+)-hyoscyamine and AFDX-116, obtained as differences between respectively pK_B or pA₂ values, are reported in table 2.

8

4

	Discussion

Top Abstract Introduction Methods Results Discussion References

(±)-PG-9 was able to induce antinociception in mice, rats and guinea pigs and to prevent impairment of the acquisition of a passive-avoidance response induced by antimuscarinic drugs. Antinociception was elicited regardless of which noxious stimulus was used: thermal (hot-plate and tail-flick tests), chemical (abdominal-constriction test) and mechanical (paw-pressure test). (±)-PG-9 antinociception was obtained without visibly modifying animal gross behavior. Moreover, (±)-PG-9-treated mice showed a complete integrity of motor coordination in the rota-rod test, as well as normal spontaneous motility as revealed by the Animex test.

(±)-PG-9 antinociception was prevented by the nonselective muscarinic antagonist atropine, the M₁-antagonists pirenzepine and dicyclomine and the ACh depletor HC-3, demonstrating, like R-(+)-hyoscyamine (see introduction), antinociceptive properties underlying a cholinergic mechanism. Both enantiomers of PG-9, R-(+) and S-(), contrary to atropine in which the analgesic activity resides only in the R-(+)-isomer (Ghelardini et al., 1992), showed very similar antinociceptive properties in the presence of either a thermal or chemical stimulus. However, in both analgesic tests used, R-(+)-PG-9 was more effective than S-()-PG-9 even if the statistical significance was not reached. Furthermore, (±)-PG-9 showed greater efficacy than that exerted by R-(+)-hyoscyamine. The analgesic effect of (±)-PG-9 was also compared with the analgesia induced by some analgesic drugs such as morphine, diphenhydramine and clomipramine at the highest doses that did not impair the rota-rod performances. By comparing the areas under the curve, the antinociceptive efficacy of (±)-PG-9 (30 mg kg¹ s.c.) was almost equal to that exerted by morphine (8 mg kg¹ s.c.), but was greater than that induced by diphenhydramine (20 mg kg¹ s.c.) and clomipramine (20 mg kg¹ s.c.).

Other neurotransmitter systems, such as opioid, GABAergic, catecholaminergic, serotoninergic and histaminergic, are not involved in (±)-PG-9 antinociception because the opioid antagonist naloxone, the GABA_B antagonist CGP-35348 and the polyamine depletor reserpine, were all unable to prevent the effect of (±)-PG-9. The doses and administration schedules of the above-mentioned drugs were ideal for preventing antinociception induced by morphine (Ghelardini et al., 1992), the GABA_B agonist baclofen (Malcangio et al., 1991) and the antidepressant drugs clomipramine and amitriptyline (Galeotti et al., 1995), respectively.

(±)-PG-9 exerted its antinociceptive effect by acting centrally. In fact, it was possible to reach the same intensity of analgesia by injecting directly into the cerebral ventricles doses (10-30 µg/mouse) of (±)-PG-9 which were 50 times lower than those needed parenterally. Dependence of the antinociception on a retrodiffusion of the drug from the cerebral ventricles to the periphery can thus be ruled out.

The prevention by the i.c.v. injection of the M₁-antagonist pirenzepine and the ACh depletor HC-3 further supports the hypothesis of a central cholinergic mechanism for (±)-PG-9 antinociception and indicates a presynaptic facilitation of cholinergic transmission by (±)-PG-9. A postsynaptic mechanism of action can be ruled out because HC-3 can not antagonize antinociception induced by agonists of postsynaptic muscarinic receptors such as oxotremorine, McN-A-343 and AF-102B (Bartolini et al., 1987, 1992).

The hypothesis of a presynaptic cholinergic mechanism for (±)-PG-9 agrees with previous results that demonstrate, by microdialysis studies, an increase in ACh release from rat cerebral cortex induced by both R-(+)-PG-9 and S-()-PG-9 administration (Romanelli et al., 1995). This effect occurred in the same range of doses in which the above-mentioned compound exerted its antinociceptive activity. Because ACh release can be increased by blocking M₂/M₄ muscarinic autoreceptors (Lapchak et al., 1989; Töröcsik and Vizi, 1991; McKinney et al., 1993; Stillman et al., 1993) and because R-(+)-hyoscyamine not only increased ACh release (Ghelardini et al., 1997) but also showed a very high affinity for the prepuberal guinea pig uterus putative M₄ receptors (Ghelardini et al., 1993), the (±)-PG-9 affinity profile toward muscarinic receptor subtypes was investigated in vitro. The affinity profile of (±)-PG-9 versus M₁ (rabbit vas deferens), M₂ (guinea pig atrium), M₃ (guinea pig ileum) and putative M₄ receptors (prepuberal guinea pig uterus) was evaluated by in vitro functional studies. Because R-(+) and S-()-PG-9 were not endowed with a different analgesic profile, their in vitro selectivity toward the muscarinic receptor subtypes was not considered worth investigating. The M₄ muscarinic receptor subtype has been defined as putative because it has not been confirmed that the mRNA codifying M₄ is expressed in prepuberal uterine tissue. However, pharmacological and biochemical studies show that the M₄ putative receptor of prepuberal guinea pig uterus has a pharmacological and biochemical profile identical with that of the muscarinic M₄ receptor subtype expressed in the rat striatum (McKinney et al., 1991; Waelbroeck et al., 1992) and in NG 108-15 cells (Leiber et al., 1984; Marc et al., 1986). (±)-PG-9 showed, like R-(+)-hyoscyamine, a M₄/M₁ muscarinic receptor subtype selectivity ratio (10.2 times) higher than the M₂/M₁ selectivity ratio (1.4 times).

The antinociception induced by (±)-PG-9 may be caused by the antagonism of the M₄ muscarinic autoreceptor. The selectivity on blocking the M₂/M₄ toward M₁ was evaluated because Bartolini et al. (1992) demonstrated that the muscarinic postsynaptic receptor responsible for central cholinergic antinociception belongs to the M₁ subtype.

The antinociceptive efficacy of (±)-PG-9 was greater than that of R-(+)-hyoscyamine. (±)-PG-9 is also endowed with very low anticholinesterase activity as demonstrated by the in vitro evaluation of its IC₅₀ value (IC₅₀ = 1.5·10⁴ M). It is possible that (±)-PG-9 is able to amplify cholinergic neurotransmission through the antagonism of the muscarinic autoreceptor and that this effect, in turn, is potentiated by its low cholinesterase inhibitory activity. However, we cannot exclude that other mechanisms able to potentiate the endogenous cholinergic system may be involved in the antinociception induced by (±)-PG-9.

D₂ dopaminergic (Gorell and Czarnecki, 1986; Wedzony et al., 1988; Scatton, 1992; Imperato et al., 1993), H₃ histaminergic (Clapham and Kilpatrick, 1992), 5-HT₄ serotoninergic heteroreceptors (Consolo et al., 1994), all located on central cholinergic neurons, increase ACh release. Therefore, the involvement of the above-mentioned heteroreceptors was investigated. Quinpirole (D₂ agonist), R-()-methylhistamine (H₃ agonist) and SDZ-205557 (5-HT₄ antagonist), at doses able to prevent antinociception induced respectively by haloperidol (Ghelardini et al., 1992), thioperamide (Malmberg-Aiello et al., 1994), BIMU 1 and BIMU 8 (Ghelardini et al., 1996), failed to prevent (±)-PG-9 antinociception. It has also been observed that the activation of the serotoninergic autoreceptor 5-HT_1A enhances ACh release from the guinea pig cortex (Bianchi et al., 1990). Pretreatment with the 5-HT_1A selective antagonist NAN 190 at doses which block the antinociception induced by 5-HT_1A agonists (Ghelardini et al., 1994), did not prevent the enhancement of the pain threshold produced by (±)-PG-9 administration. The present data suggest that the above-mentioned receptors, even though they are able to increase ACh release, are not involved in (±)-PG-9 mechanism of analgesic action.

(±)-PG-9 was able to prevent impairment of the acquisition of a passive-avoidance response induced by the antimuscarinic drugs scopolamine and dicyclomine in mice. Because stimulation of the cholinergic system improves cognitive processes (Coyle, 1995), it is reasonable to suppose that the antiamnesic effect induced by (±)-PG-9 could be related to its ability to activate the cholinergic system. In our experimental conditions, (±)-PG-9 was administered before mice received the aversive stimulus in correspondence to the maximum analgesic effect. The ability of (±)-PG-9 to enhance the pain threshold by abolishing the perception of the punishing stimulus may have influenced the results obtained. An electric shock, reported as the punishing stimulus in the original method (Jarvik and Kopp, 1967), was thus substituted by a nonpainful stimulus consisting of a fall into cold water.

In summary, our results have shown that (±)-PG-9 is able to produce dose-dependent antinociception in rodents and guinea pigs as well as antiamnesic activity in mice, without impairing motor coordination, by potentiating endogenous cholinergic activity.