Search

Cannabis (marijuana) is not only a frequently abused drug but also has the potential for the development of useful agents for the treatment of emesis, anorexia and multiple sclerosis. In this article, the effects of modulation of transmitter release by cannabinoids in both the CNS and the PNS of various species, including humans, will be discussed. Cannabinoids inhibit neurotransmitter release via specific presynaptic cannabinoid CB1 receptors. Studies using either the CB1 receptor antagonist and inverse agonist SR141716 or CB1-receptor-deficient mice suggest that numerous presynaptic cannabinoid receptors are tonically activated by endogenous cannabinoids and/or are constitutively active. CB1-receptor-mediated inhibition of transmitter release might explain, for example, reinforcing properties and memory impairment caused by cannabinoids.

We examined whether acetylcholine release in the hippocampus and striatum and noradrenaline release in the hippocampus is altered in CB(1) receptor-deficient mice. The electrically evoked tritium overflow from hippocampal slices preincubated with [(3)H]-choline was increased by about 100% in CB(1)(-/-) compared to CB(1)(+/+) mice whereas the electrically evoked tritium overflow from striatal slices preincubated with [(3)H]-choline and from hippocampal slices preincubated with [(3)H]-noradrenaline did not differ. The cannabinoid receptor agonist, WIN 55,212-2, inhibited, and the CB(1) receptor antagonist, SR 141716, facilitated, the evoked tritium overflow from hippocampal slices (preincubated with [(3)H]-choline) from CB(1)(+/+) as opposed to CB(1)(-/-) mice. Both drugs did not affect the evoked tritium overflow from striatal slices (preincubated with [(3)H]-choline) and from hippocampal slices (preincubated with [(3)H]-noradrenaline) from CB(1)(+/+) and CB(1)(-/-) mice. The selective increase in acetylcholine release in CB(1)(-/-) mice may indicate that the presynaptic CB(1) receptors on the cholinergic neurones of the mouse hippocampus are tonically activated and/or constitutively active in vivo.

Cannabinoid CB1 receptors occur as presynaptic receptors producing inhibition of neurotransmitter release. To elucidate their physiological role, experiments on tissues from CB1 receptor knockout mice would be helpful. We studied whether CB1 receptor-mediated inhibition of acetylcholine release is detectable in the brain of NMRI mice and of CD-1 and C57BL/6J mice (the latter two strains representing the wild-type strains of the two CB1 receptor knockout mouse models). Brain slices preincubated with [3H]choline were superfused and tritium overflow was evoked electrically (3 Hz) or by introduction of Ca2+ into Ca2+-free K+-rich medium (35 mM) containing tetrodotoxin. The eletrically evoked tritium overflow from NMRI mouse hippocampal slices was inhibited (maximally by 60%) by the cannabinoid receptor agonists CP-55,940 and WIN 55,212-2 but not affected by WIN 55,212-3 (the inactive enantiomer of WIN 55,212-2; pEC50=7.9, 7.4 and <5.5). The concentration-response curve of WIN 55,212-2 was shifted to the right by the CB1 receptor antagonist SR 141716 (apparent pA2=8.6). Compared to hippocampal slices from NMRI mice, WIN 55,212-2 1 microM inhibited the electrically evoked overflow (1) from cortical slices from NMRI mice to a lesser extent and from striatal slices not at all, (2) from hippocampal slices from CD-1 and C57BL/6J mice to an identical extent and (3) from hippocampal slices from Sprague-Dawley rats to at least the same extent. SR 141716 0.32 microM abolished the effect of WIN 55,212-2 1 microM in hippocampal slices from NMRI, CD-1 and C57BL/6J mice and in cortical slices from NMRI mice. The electrically evoked tritium overflow from NMRI mouse hippocampal slices was also inhibited by the muscarinic receptor agonist oxotremorine (maximum effect of 85%; pEC50=6.5) and this effect was antagonized by the muscarinic receptor antagonist AF-DX 384 (apparent pA2=8.3). The Ca2+-evoked tritium overflow from NMRI mouse hippocampal slices was inhibited by WIN 55,212-2 in a manner sensitive to SR 141716. In conclusion, the cholinergic axon terminals of the NMRI mouse hippocampus are endowed with presynaptic CB1 receptors. Such receptors are also detectable in the hippocampus of CD-1 and C57BL/6J mice. The maximum extent of the CB1 receptor-mediated inhibition of acetylcholine release is lower than the maximum effect mediated via the autoreceptor.

We studied whether serotonin release in the CNS is inhibited via cannabinoid receptors. In mouse brain cortex slices preincubated with [3H]serotonin and superfused with medium containing indalpine and metitepine, tritium overflow was evoked either electrically (3 Hz) or by introduction of Ca2+ (1.3 mM) into Ca2+-free K+-rich (25 mM) medium containing tetrodotoxin. The effects of cannabinoid receptor ligands on the electrically evoked tritium overflow from mouse brain cortex slices preincubated with [3H]choline and on the binding of [3H]WIN 55,212-2 and [35S]GTPgammaS to mouse brain cortex membranes were examined as well. In superfused mouse cortex membranes preincubated with [3H]serotonin, the electrically evoked tritium overflow was inhibited by the cannabinoid receptor agonist WIN 55,212-2 (maximum effect of 20%, obtained at 1 microM; pEC50=7.11) and this effect was counteracted by the CB1 receptor antagonist SR 141716 (apparent pA2=8.02), which did not affect the evoked tritium overflow by itself. The effect of WIN 55,212-2 was not shared by its enantiomer WIN 55,212-3 but was mimicked by another cannabinoid receptor agonist, CP-55,940. WIN 55,212-2 also inhibited the Ca2+-evoked tritium overflow and this effect was antagonized by SR 141716. Concentrations of histamine, prostaglandin E2 and neuropeptide Y, causing the maximum effect at their respective receptors, inhibited the electrically evoked tritium overflow by 33, 69 and 73%, respectively. WIN 55,212-2 (1 microM) inhibited the electrically evoked tritium overflow from mouse brain cortex slices preincubated with [3H]choline by 49%. [3H]WIN 55,212-2 binding to mouse cortex membranes was inhibited by CP-55,940, SR 141716 and WIN 55,212-2 (pKi=9.30, 8.70 and 8.19, respectively) but not by the auxiliary drugs indalpine, metitepine and tetrodotoxin (pKi<4.5). [35S]GTPgammaS binding was increased by WIN 55,212-2 (maximum effect of 80%, pEC50=6.94) but not affected by WIN 55,212-3. In conclusion, serotonin release in the mouse brain cortex is inhibited via CB1 receptors, which may be located presynaptically and are not activated by endogenous cannabinoids. The extent of inhibition is smaller than that obtained (1) via another three presynaptic receptors on serotoninergic neurones and (2) via CB1 receptors on cholinergic neurones in the same tissue.

We have studied (i) the effect of antisense oligodeoxynucleotides complementary to CB1 mRNA on the CB1 receptor binding in hippocampus, striatum and cerebral cortex of the rat; (ii) the possible mechanism of action of one of the antisense oligodeoxynucleotides; and (iii) its effect on two functional CB1 receptor-mediated effects. Synthetic oligodeoxynucleotides or saline were administered to male Wistar rats by the intracerebroventricular (i.c.v.) route twice daily for 3 days. Antisense oligodeoxynucleotides corresponding to the nucleotides 4 to 21 (AS1; GCCATCTAGGATCGACTT) and -8 to 12 (AS2; GATCGACTTCATAACCTCAG) and a mismatch oligodeoxynucleotide differing from AS1 in 6 positions (MM; TCCAGCTACTATGGACTG) were used. The dissociation constant (K(D)) of rat CB1 cannabinoid receptors, labelled by the radioligand [3H]-SR141716, did not differ in membranes from rats treated with saline, AS , AS2 or MM. The density of receptor binding (Bmax) was reduced by the antisense oligodeoxynucleotides, 10nmol, in the hippocampus (AS 1, -40%; AS2, -20%) and striatum (AS1, -29%; AS2 -6%), but not in the cerebral cortex. When the dose of AS1 was raised to 30 nmol, the reduction of Bmax in the hippocampus and striatum was only marginally increased; a dose of 3nmol of AS1 reduced Bmax in both brain regions by somewhat more than the half-maximum effect. The mismatch oligodeoxynucleotide MM (3-30nmol) did not affect Bmax. In the second part of the study, RNA obtained from the three brain regions of rats pretreated with AS1 10 nmol, MM 10 nmol or saline was analyzed using reverse transcription-polymerase chain reaction of CB1 receptor mRNA and of beta-actin mRNA levels (used as reference value). The ratio of CB1 receptor mRNA over beta-actin mRNA after treatment with AS1 did not differ from the ratios following treatment with saline or MM in the hippocampus, striatum and cerebral cortex. Finally, pretreatment with antisense oligodeoxynucleotide AS1 30nmol attenuated two functional effects via CB1 receptors, i.e., the facilitatory effect of WIN 55,212-2 on [35S]-GTPgammaS binding in rat hippocampus membranes and the inhibitory effect of WIN 55,212-2 on acetylcholine release in rat hippocampus slices. In conclusion, (i) two antisense oligodeoxynucleotides reduce the density of CB1 receptors in the rat hippocampus and striatum after i.c.v. administration. (ii) The effect of the antisense oligodeoxynucleotide AS1 does not appear to be related to breakdown of CB1 receptor mRNA. (iii) Pretreatment with AS1 attenuated the CB1 receptor-mediated effect in two functional models.

Guinea-pig hippocampal slices preincubated with [3H]noradrenaline were superfused with medium containing desipramine and rauwolscine and rat striatal slices preincubated with [3H]dopamine were superfused with medium containing nomifensine; the effect of cannabinoid receptor ligands on tritium overflow stimulated by NMDA or kainate was examined. Furthermore, the affinity of the drugs for cannabinoid CB1 receptors was determined in rat brain cortex membranes using [3H]SR 141716. In guinea-pig hippocampal slices preincubated with [3H]noradrenaline, tritium overflow stimulated by NMDA 100 microM and 1000 microM and by kainate 1000 microM was inhibited by the cannabinoid receptor agonists CP-55,940 and/or WIN 55,212-2. The CB1 receptor antagonist SR 141716 increased the NMDA (1000 microM)-stimulated tritium overflow but did not affect tritium overflow stimulated by NMDA 100 microM or kainate 1000 microM. The inhibitory effect of WIN 55,212-2 on the NMDA (100 microM)- and kainate (1000 microM)-evoked tritium overflow was antagonized by SR 141716. In rat striatal slices preincubated with [3H]dopamine, WIN 55,212-2 inhibited the NMDA (1000 microM)-stimulated tritium overflow. SR 141716, which, by itself, did not affect tritium overflow, counteracted the inhibitory effect of WIN 55,212-2. [3H]SR 141716 binding to rat cortical membranes was inhibited by SR 141716, CP-55,940 and WIN 55,212-2 (pKi 8.53, 7.34 and 5.93, respectively) but not affected by desipramine, rauwolscine and nomifensine (pKi < 5). In conclusion, activation of CB1 receptors inhibits the NMDA- and kainate-stimulated noradrenaline release in guinea-pig hippocampus and the NMDA-stimulated dopamine release in rat striatum. The explanation for the facilitatory effect of SR 141716 might be that it acts as an inverse agonist at CB1 receptors or that these receptors are activated by endogenous cannabinoids.

The present study was carried out to clarify whether the imidazolines clonidine, moxonidine and cirazoline as well as the guanidine aganodine inhibit noradrenaline release in the rat and rabbit brain via imidazoline receptors, alpha 2-adrenoceptors and/or histamine H3 receptors. Slices or synaptosomes from the rat or the rabbit brain were incubated with 3H-noradrenaline and exposed to phenoxybenzamine, which irreversibly blocks presynaptic alpha 2-adrenoceptors and, at considerably lower potency, imidazoline receptors. Tritium overflow in the superfused preparations was evoked electrically (3 Hz; slices) or by K+ 15 mmol/l (synaptosomes). Noradrenaline and rauwolscine, which possess low affinity, if any, for imidazoline receptors, were used as reference drugs. The evoked overflow in rat brain cortex slices and synaptosomes and in rat medulla oblongata slices, not exposed to phenoxybenzamine, was inhibited by clonidine, moxonidine and noradrenaline. Phenoxybenzamine markedly attenuated the effect of each drug to about the same extent. In rabbit brain cortex slices, not exposed to phenoxybenzamine, the evoked overflow was inhibited by clonidine, moxonidine, aganodine and noradrenaline, facilitated by BDF 6143 (4-chloro-2-(2-imidazoline-2-yl-amino)-isoindoline), idazoxan and rauwolscine and not affected by cirazoline. In slices exposed to phenoxybenzamine, the inhibitory effects of the imidazolines, of aganodine and of noradrenaline were again attenuated by about the same high degree, the facilitatory effects of BDF 6143, idazoxan and rauwolscine were abolished and cirazoline produced a slight inhibition of the evoked overflow. The latter effect was not affected by high concentrations of rauwolscine and idazoxan (at which these drugs act antagonistic at imidazoline receptors in other models). The specific binding of 3H-N alpha-methylhistamine to H3 receptors in rat brain cortex membranes was displaced only by high concentrations of moxonidine (pKi = 6.16) and at even lower affinity by aganodine, BDF 6143, cirazoline, clonidine and idazoxan (pKi < 5). Histamine, which was used as a reference drug, proved to be very potent (pKi = 8.20). In conclusion, imidazolines affect noradrenaline release in the rat and rabbit brain cortex and medulla oblongata via alpha 2-adrenoceptors but not via imidazoline receptors resembling the presynaptic imidazoline receptors previously identified in peripheral tissues of the rabbit. In addition, the involvement of I1- or I2-imidazoline binding sites or of H3 receptors is very improbable in view of the low affinity of aganodine, moxonidine and/or clonidine for these recognition sites and/or incompatibility of the rank order of their affinities with the potencies of the drugs in inhibiting noradrenaline release.