Distinct cannabinoid sensitive receptors regulate hippocampal excitation and inhibition

Abstract

One of the well-known effects of cannabinoids is the impairment of cognitive processes, including short-term memory formation, by altering hippocampal and neocortical functions reflected in network activity. Acting on presynaptically located G protein-coupled receptors in the hippocampus, cannabinoids modulate the release of neurotransmitter molecules. CB1 cannabinoid receptors, so far the only cloned cannabinoid receptor type in the CNS, are selectively expressed on the axon terminals of a subset of GABAergic inhibitory interneurons containing the neuropeptide cholecystokinin. Activation of CB1 receptors reduces GABA release from presynaptic terminals, thereby increasing the excitability of principal cells. Novel, non-CB1 cannabinoid sensitive receptors are present on the hippocampal excitatory axon terminals, which suppress glutamate release. These cannabinoid receptors have distinct pharmacological features compared to CB1, i.e. WIN 55,212-2 is an order of magnitude less potent in reducing glutamatergic transmission than in inhibiting GABAergic postsynaptic currents, and the novel receptor binds vanilloid receptor ligands. Thus, at least two different cannabinoid sensitive presynaptic receptors regulate network activity in the hippocampus, CB1 via the GABAergic interneurons, and a new receptor via a direct action on pyramidal cell axon terminals.

Introduction

Several lines of evidence indicate that, in addition to the cloned G protein-coupled CB1 and CB2, other cannabinoid sensitive receptors are present both in the peripheral and central nervous systems, which may participate in well-described effects of cannabinoids, such as analgesia, hypotension, catalepsy, and memory impairments (for review see Ameri, 1999). The exclusive role of CB1 and CB2 receptors in mediating cannabinoid actions has been first challenged by the study of Calignano et al. (1998) showing that an N-acylethanolamine derivative, palmitoylethanolamide (PEA), that lacks significant affinity to either cloned receptors, plays a substantial role in pain sensation, and its effect can be attenuated by the CB2-preferring antagonist SR144528. However, the analgesic effect of PEA has never been documented in CB2 knockout mice. Another recent study pointed out that the mesenteric vasodilatation induced by cannabinoids is mediated by a non-CB1, non-CB2, non-vanilloid receptor (Jarai et al., 1999).

In the CNS, the first results arguing against the exclusive role of CB1 receptors in mediating cannabinoid actions emerged from the comparison of receptor localization data with electrophysiological and pharmacological measurements. The former showed a localization of CB1 on GABAergic, but not on glutamatergic terminals (Katona et al., 1999), whereas the latter demonstrated a CB1 receptor-mediated presynaptic effect on both glutamatergic (Misner and Sullivan, 1999, Shen et al., 1996, Sullivan, 1999) and GABAergic transmission (Hájos et al., 2000, Hoffman and Lupica, 2000, Irving et al., 2000). The evidence for cannabinoid sensitive G protein-coupled receptors different from CB1, CB2 in the brain derived from studies using CB1 receptor knockout mice (Breivogel et al., 2001, Di Marzo et al., 2000, Hájos et al., 2001). The present review will correlate anatomical, physiological and pharmacological data from the hippocampus of wild type and CB1 receptor knockout animals with the aim to confirm that indeed, a new CB receptor exists, which has a characteristic location, pharmacology and function.

These studies were largely focussed on the hippocampus, since the administration of cannabinoids alters memory functions that are linked to neuronal networks of the cerebral cortex, including in particular the hippocampus. The hippocampal formation, as all other cortical areas, consists of two major neuron types: excitatory principal cells and inhibitory interneurons. The former include hippocampal pyramidal cells and dentate granule cells accounting for 85–90% of the entire neuronal population (Woodson et al., 1989). These principal cells have a rather uniform appearance, whereas GABAergic interneurons form heterogeneous populations with distinct anatomical and functional features (for review see Freund and Buzsaki, 1996). To understand the mechanism of cannabinoid actions and the possible functions of endocannabinoids in cortical neuronal circuits, the localization as well as the physiological and pharmacological properties of cannabinoid receptors has to be determined differentially for the two major cell classes.