Distribution of Cannabinoid Receptors in the Central and Peripheral Nervous System 303inhibit neurotransmitter release. The third theme was that in a few brain regions
(for example, anterior olfactory nucleus, caudate nucleus and cerebellum) CB 1
receptors are uniformly expressed at moderate levels on a single class of neurons.
1.4
Immunocytochemistry
Elucidation of the primary sequence of the CB 1 receptor allowed for production of
numerous CB 1 receptor antibodies. There have been two thorough immunocyto-
chemical mapping studies in rodent brain (Tsou et al. 1998a; Egertová and Elphick
2000) and one in spinal cord (Farquhar-Smith et al. 2000). These generally support
the results from the autoradiographic studies, with some differences in relative
intensity of staining. These variations may be due to differences in antibody access
to specific epitopes, variable post-translational modification of an epitope (e.g.,
phosphorylation), or fixation conditions. There is little evidence for alternative
splicing in the coding region of rodent CB 1 receptors (Matsuda 1997; Lutz 2002),
despite the report of alternatively splicing of the human CB 1 receptor (Shire et al.
1995; Matsuda 1997; Lutz 2002); so alternative splicing is less likely to explain the
reported differences.
The immunocytochemical studies have led to additional insights into cannabi-
noid action. The first is that rigorous electron microscopic studies in the hip-
pocampus demonstrated that in this region CB 1 is undetectable on somatic cell
membranes and dendrites, yet is very highly expressed in axon terminals and
preterminal segments (Hajos et al. 2000; Katona et al. 2000, 2001). An example of
this is shown in Fig. 2, with the labeling of four consecutive ultrathin sections of
a cortical axodendritic synapse. The second is that double-label immunostaining
experiments demonstrated that in forebrain there is a striking correlation between
cholecystokinin (CCK) and CB 1 receptor expression (Katona et al. 1999, 2001; Tsou
et al. 1999). These findings have been confirmed and extended with double-label in
situ hybridization studies (Marsicano and Lutz 1999). Thus, the cells expressing the
highest levels of CB 1 receptorsinforebrainareγ-aminobutyric acid (GABA)ergic,
CCK-positive interneurons. Although inhibition of GABA release is measured in
the in vitro electrophysiological studies, activation of CB 1 receptors in vivo will
attenuate both inhibitory transmission (generally fast, mediated by GABA A recep-
tors) (Wilson et al. 2001) as well as the slow, excitatory actions mediated by CCK
receptors (Beinfeld and Connolly 2001). Thus, the localization of CB 1 receptors
on CCK-containing neurons suggests that CB 1 receptors are well positioned to
modulate complex network behaviors (Freund 2003).
Once antibodies to the anandamide-degrading enzyme, namely fatty acid amide
hydrolase (FAAH), became available, it was apparent that in many regions FAAH
and CB 1 expression is reciprocal in nature (Egertová et al. 1998, 2003; Tsou et
al. 1998b). For example, FAAH, but not CB 1 is highly expressed in the somata
and proximal dendrites of hippocampal pyramidal cells and cerebellar Purkinje
neurons. These neurons are, in turn, densely innervated by CB 1 -positive fibers.
Thus, it has been proposed that anandamide, despite its possible presynaptic site