Any drug that reduces excitatory drive would be expected to impair high-frequency stimulus-induced LTP by limiting the level of postsynaptic depolarisation achieved during the induction train. This would in turn reduce the relief of the voltage-dependent block of N-methyl-D-aspartate (NMDA) receptor-gated channels by Mg2+ and hence reduce the postsynaptic Ca2+ entry that is required to trigger the processes that lead to synaptic potentiation. The question of whether cannabi-noids inhibit baseline excitatory transmission is therefore an important one. The suppression of inhibitory transmission by cannabinoids in the hippocampus has been well documented, but whether cannabinoids also suppress excitatory glu-tamatergic transmission (as they do in the cerebellum, see the chapter by Szabo and Schlicker, this volume) is less clear cut. Thus, there are reports stating explicitly that WIN55,212-2 either inhibits (Misner and Sullivan 1999; Al-Hayani and Davies 2000; Ameri and Simmet 2000; Hajos et al. 2001), or does not inhibit (Terranova et al. 1995; Paton et al. 1998; Al-Hayani and Davies 2000), excitatory synaptic transmission in the CA1 region. This apparent discrepancy has now been resolved by the demonstration that the most commonly used CB1 receptor agonist, WIN55,212-2, at tenfold higher concentrations, also activates a TRPV1-like receptor, which is also sensitive to rimonabant (Hajos et al. 2001; Hajos and Freund 2002). Thus, in slices prepared from CB1+/+ mice, perfusion of WIN55,212-2 inhibited pharmacologically isolated excitatory postsynaptic currents (EPSCs) with an EC50 of 2.01 ^M, and pharmacologically isolated inhibitory postsynaptic currents (IPSCs) with an EC50 of 0.24 ^M (Hajos et al. 2001). In slices prepared from CBr/-mice, WIN55,212-2 no longer inhibited evoked IPSCs, but still inhibited evoked EPSCs. This inhibition of excitatory transmission was mimicked by the TRPV1 agonist capsaicin (10 ^M), and was blocked by the TRPV1 antagonist, capsazepine (10 ^M). The fact that the suppression of EPSCs (as well as IPSCs) by WIN55,212-2 is blocked by rimonabant is significant, since this is the criterion by which an effect would previously have been judged to be CB1 receptor mediated. The tenfold concentration difference in the EC50 of WIN55,212-2 in blocking inhibitory, as opposed to excitatory, transmission also explains why the drug has been reported to selectively block paired-pulse depression of population spikes (an effect dependent on feedback inhibitory transmission) but not baseline synaptic transmission (Paton et al. 1998).
Whether this central TRPV1-like receptor has the same properties as the better characterised peripheral TRPV1 receptors (Szallasi and Di Marzo 2001), and whether it represents the same non-CB1 non-CB2 receptor characterised by Breivo-
gel et al. (2001; see chapter by Pertwee, this volume) remains to be determined. Equally, it is hard to resolve whether the effects of other cannabinoids are likely to be mediated by the TRPV1-like receptor. Available evidence suggests that the re-ceptoris also activated by CP55,940, but that it is not blocked by AM251 (Hajos and Freund 2002). The answer is therefore that some cannabinoids do indeed inhibit excitatory transmission in the hippocampus, but via a TRPV1-like receptor. This in turn raises the question: Is inhibition of the induction of LTP by cannabinoids secondary to suppression of baseline excitatory transmission?
Several of the studies on LTP reported above have commented in passing on any associated effects of cannabinoids on baseline field EPSP or population spike responses, but answers have been contradictory. Thus, some have found no change (Terranova et al. 1995; Stella et al. 1997; Paton et al. 1988; Schweitzer et al. 1999), whereas others have found an inhibition of baseline excitatory responses (Nowicky et al. 1987; Misner and Sullivan 1999). The question was addressed directly by Misner and Sullivan (1999) who found that the EPSC was reduced by about 50% in slices prepared from neonatal rats. Perfusion of 5 ^M WIN55,212-2 blocked the induction of LTP by high-frequency stimulation (100 Hz for 200 ms), but this could be overcome by manipulations designed to overcome the Mg2+-block of NMDA receptor-gated channels, i.e. reducing the concentration of Mg2+ in the perfusion medium, or by slightly depolarising the recorded cell during the high-frequency train. They therefore concluded that the effect of WIN55,212-2 was to reduce the excitatory drive and therefore the extent of activation of NMDA receptors.
Note that lack of an effect of cannabinoids on the low-frequency-evoked synap-tic responses in some of the experiments described above does not exclude the possibility that the drugs may have an effect on the high-frequency response required to induce LTP. It would therefore be important to establish the effects of cannabinoid receptor ligands on the response to repetitive high-frequency stimulation, as well as on the response to low-frequency stimulation. Though some of the experiments described above suggest that cannabinoids might inhibit induction of LTP by suppressing excitatory drive, none of them distinguishes the possibility that cannabinoids block LTP via an action on the TRPV1-like receptor rather than the CB1 receptor. WIN55,212-2 is an agonist at both, and rimonabant inhibits both. Resolution of this question must await the development of ligands that will reliably differentiate the two receptors, and/or investigation of the effects of cannabinoids on the induction of LTP in CB1"/" animals.
Was this article helpful?