- Email: [email protected]
Neuronal localization of cannabinoid receptors and second messengers in mutant mouse cerebellum
301
Brain Research, 552 (1991) 301-310 Elsevier ADONIS 000689939116730T BRES 16730
Neuronal localization of cannabinoid receptors and second messengers in mutant mouse cerebellum M. Herkenham 1, B.G.S. Groen 1, A.B. Lynn 1, B.R.
D e C o s t a 2 a n d E . K . Richfield 1'*
1Section on Functional Neuroanatomy, National Institute of Mental Health, and 2Laboratory of Medicinal Chemistry, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD (U.S.A.) (Accepted 22 January 1991) Key words: Tetrahydrocannabinol; [3H]CP 55,940; Forskolin; Phorbol ester; Adenylate cyclase; Phosphoinositide
Four lines of mutant mice were used to investigate (1) the neuronal localization of cannabinoid receptors in the cerebellar molecular layer and (2) the anatomical association of these receptors with elements of the two second messenger systems in the brain. Two of the mutant lines Purkinje cell degeneration and nervous - - are selectively deficient in Purkinje cells; the other two - - weaver and reeler - - are deficient in granule cells. In the heterozygous mice, [3H]CP 55,940 binding to cannabinoid receptors was discretely and densely localized to the molecular layer, as was [3H]forskolin binding to adenylate cyclase and [3H]phorbol 12,13-dibutyrate binding to protein kinase C, a component of the phosphoinositide cycle. [3H]CP 55,940 and [3H]forskolin binding was selectively reduced in weaver and reeler homozygous mice but unchanged in Purkinje cell deficient and nervous homozygotes. No decreases in [3H]phorpbol 12,13-dibutyrate binding were found in any of the homozygous mutants relative to the heterozygous littermates. The results suggest that cannabinoid receptors and adenylate cyclase are localized to granule cell axons in the molecular layer, whereas protein kinase C is equally distributed in parallel fibers and Purkinje cell dendrites.
-
-
INTRODUCTION The actions of d9-tetrahydrocannabinol (A9-THC),the major psychoactive ingredient of marijuana (Cannabis sativa) 22, appear to be mediated by the recently characterized cannabinoid receptor 3'1°. A9-THC and other natural and synthetic cannabinoids produce characteristic m o t o r effects in animals. Structure-activity relationships and localization patterns 1° suggest that cannabinoid receptors in the basal ganglia and cerebellum mediate the catalepsy and ataxia elicited by systemic administration of cannabinoids in mice and dogs 17A9"2°. Recently, we have shown that the receptors in the basal ganglia are localized tO striatal neurons, very densely to their axons and terminals in the globus pallidus, entopeduncular nucleus, and substantia nigra pars reticulata 7'8. The other area in the brain containing very dense cannabinoid receptor binding is the cerebellum, where receptors are discretely localized to the molecular layer. We wondered whether this too might represent axonal localization. This possibility was reinforced by observations of axonal binding of ligands marking second messengers, in patterns that match the cannabinoid binding patterns in both the basal ganglia and cerebellum 5'29'3°. Cannabinoids are known to modulate the adenylate cyclase second messenger system via the
cannabinoid receptor 11-13, but evidence for modulation of the phosphoinositide (PI) cycle is scarce 2'23. Specific questions arise regarding the anatomical correlations. First, are cannabinoid receptors in the cerebellum neutonally localized, and if so, to which cerebellar elements are they confined? Second, do they reside on the same elements that are enriched in either of the second messengers? These questions have been addressed using section binding assays and autoradiography performed on lines of mutant mice showing selective losses of either Purkinje cells or granule cells. The neuronal localization patterns are compared with patterns of [3H]forskolin and [3H]phorbol 12,13-dibutyrate ([3H]PDBu) binding to the adenylate cyclase and PI systems, respectively. MATERIALS AND METHODS Animals Litters of mice were obtained from Jackson Laboratories (Bar Harbour, ME). Four lines of cerebeUar mutants were obtained (Table I). Upon arrival, each litter, including the mother, was housed in one cage. The litters contained animals homozygous and heterozygous for the mutant gene. Homozygotes showed slow weight gain and ataxia, whereas heterozygotes showed normal weight gains and normal motor behavior. Animals were given tap water and dry food ad libitum, and were housed at approximately 24 °C with lights on from 06.00 h to 18.00 h. After the appropriate survival time, to ensure maximal expression of the phenotype (Table I), the mice were decapitated, and the
*Present address: University of Rochester, Monroe Community Hospital, Rochester, NY 14620, U.S.A. Correspondence: M. Herkenham, Section on Functional Neuroanatomy, NIMH, Bldg. 36, Rm. 2D-15, Bethesda, MD 20892, U.S.A.
302 brains were removed, frozen in isopentane at -30 °C, and stored at -40 °C until use. Whole brains were mounted onto cryostat pedestals with embedding matrix (M-l, Lipshaw). Serial 20 pm thick sections were cut in a cryostat at -18 °C through the whole brain in one and through the whole cerebellum in the remaining 4 brains in every group. At levels interspaced by 100/~m, 7 serial sections were saved. The sections were thaw-mounted onto separate gelatincoated slides, desiccated overnight at 0 °C, and stored at -40 °C until use. One set of sections from each series was fixed in formalin, defaned, and stained in Cresyl violet for examination of cytoarchitecture; the remaining sets were saved for binding assays. Binding assays [3H]CP 55,940 is one of a series of synthetic cannabinoids whose structure and biological activity have been documented 3A°'14.Assay conditions were described previously1°. Slide-mounted sections taken from 5 animals in each group were brought to room temperature and incubated at 37 °C for 2.5 h in 50 mM Tris-HC1 buffer, pH 7.4, with 5% bovine serum albumin (BSA) and 10 nM [3H]CP 55,940 (sp. act. 48 Ci/mmol; custom radiolabeled at New England Nuclear). Sections were washed at 0 °C for 2 x 2 h in the same buffer with 1% BSA and dried. Nonspecific binding was determined in adjacent sections by the addition of 10 pM CP 55,244 and represented about 20% of the total. Slides were exposed to 3H-sensitive film (LKB or Amersham) for 2 (nervous, reeler) or 3 (pcd, weaver) weeks and developed (D19, Kodak) at 20 °C for 4 min. [3H]Forskolin binding to adenylate cyclase associated with guanine nucleotide binding proteins was carried out as described previously29. Sections were incubated in cytomailers containing 50 mM Tris-HCl, pH 7.7, 100 mM NaCl, 5 mM MgCI2, and 10 nM [3H]forskolin (sp. act. 38.6 Ci/mmol; New England Nuclear) for 10 min at 24 °C. Slides were washed 2 x 2 rain in same buffer at 4 °C, briefly dipped in 4 °C water, and dried. Nonspecific binding was assessed in adjacent sections by the addition of 10 pM forskolin (Sigma) and represented < 5% of the total. Slides were exposed to film for 12 weeks before developing. [3H]PDBu binding to protein kinase C was carried out as described previously3°. Sections were incubated in cytomailers containing 50 mM Tris-HC1, pH 7.7, 100 mM NaCl, 1 mM CaCl2, and 2.5 nM [3H]PDBu (sp. act. 19.1 Ci/mmol; New England Nuclear) for 60 min at 33 °C. Following incubation, slides were washed 2 x 1 min in ice cold buffer and dried. Nonspecific binding was assessed by the addition of 1 /~M PDBu (Sigma) and represented < 5% of the total. Slides were exposed to film for 2 weeks before developing. Data analysis Sections from heterozygotes and homozygotes from each mutant line were co-exposed together along with tritium standards on the same films. Developed films were digitized with a solid state video camera (Sierra Scientific) and Macintosh II computer-based system for quantitative densitometry using IMAGE software (Wayne Rasband, Research Services Branch, NIMH). Transmittance levels were converted to fmol/mg tissue using the tritium standards (Amersham 3H Microscales). Light transmittance was measured in sections at levels containing the deep cerebellar nuclei. A threshold function was applied to each section to select and measure transmittance confined to the molecular layer. For each animal, structure, and ligand, measures from 2-4 sections containing the structure were averaged. The process was performed on images from adjacent sections incubated for total and for nonspecific binding, and the difference was computed to determine specific binding for the molecular layer. RESULTS
TABLE I Strains of mice used and times of sacrifice Mutant type
Strain
Deficiency
Time of onset of maximal defect (d) °
Age at sacrifice (d)
Purkinje cell degeneration ( p c d ) Nervous Weaver Reeler
C57BL C3HeB B6CBA B6C3
Purkinje cell Purkinje cell Granule cell Granule cell
>28 23-50 20-22 37-56
31 74-77 59-68 21-26
a Literature sources: pcd21; nervous16'25'27; weaver26"27;reeler 6'24'27.
of the cerebellum of each of the homozygous mutants, with a normal appearance of the brains of the heterozygous littermates. In every case, reduced cerebellar size was seen only in those animals that had showed reduced body weight and difficulties in m o v e m e n t and posture. The gross as well as the microscopic appearance of the cerebella noted in these brains conformed to previous detailed descriptions of the m u t a n t s 6A6'21'24-27. The normal cerebellar molecular layer consists of Purkinje and Golgi cell dendrites, granule cell axons (parallel fibers), scattered stellate and basket cells, and extrinsic afferent axons (climbing fibers) from the inferior olive. In each of the 4 homozygous m u t a n t types, the depth of the molecular layer a n d the arrangements of neurons in the Purkinje and granule cell layers is characteristically altered (Figs. 1 and 2). Two of the m u t a n t lines show selective losses of Purkinje cells. Mice heterozygous for the pcd gene show a n o r m a l appearing cerebellum (Fig. l a , c). Mice homozygous for the pcd gene show a dramatic loss of virtually all of their Purkinje neurons (Fig. ld) and a c o m m e n s u r a t e reduction in the width of the molecular layer and volume of the deep cerebellar nuclei (Fig. l b , d ) , reflecting the loss of Purkinje cell dendrites and axons in these locations, respectively. The nervous gene m u t a t i o n produces a phenotypic defect similar to that of the pcd, although surviving Purkinje cells are evident (Fig. lh). The two other m u t a n t lines examined in this study are characterized by relatively greater losses of granule than Purkinje cells; in these, the cerebellar deformation is much more dramatic (Fig. 2). T h e m a j o r defects in the mice homozygous for the weaver and reeler genes are a greatly s h r u n k e n molecular layer and an intermingling of Purkinje cells in the granule layer, which is reduced in size (Fig. 2). In both lines, however, the size of the deep cerebellar nuclei is normal in the homozygotes (Fig. 2b,
f). Gross appearances of the brains at the time of brain removal showed dramatic differences in the appearance
Figure 3 shows the pattern of [3H]CP55,940 binding in the cerebellum of each of the m u t a n t lines. As in all other
303
i .....
.
.
.
.
.
.
.
.
lib
•
I
Illm •
II
I
'~ '1)
, , • .-,
_ • -...
~ ,-
..
::,, ,,,
,,..
• ~ ,~.....
.. .i, ~
,
~%.,
"*, .~,
*
8
~e v
o
t O 'P ~11
etb s* a *
.¢
,-~
~p
0 . io " , D
~ t
~,
lie
Fig. 1. Photographs of Nisslostained sections of heterozygous and homozygous gene mutations leading to losses of Purkinje cells in the cerebellum. The heterozygous Purkinje cell-deficient (+/pc,d) mouse has a normal appearance of lamination and organization (a and c), but the homozygous (ped/pcd) mouse shows nearly complete loss of Purkinje cells (b and d). Similarly, the normal appearance of the heterozygous nervous (+/nr) mouse (e and g) is contrasted with the major depletion of Purkinje neurons in the homozygous (nr/nr) mouse (f and h). Large arrow in the low-power photographs point to the deep cerebellar nuclei, which are reduced in size in the homozygotes. Small arrows in the high-magnification photomicrographs point to Purkinje cell bodies. Scale bar in b = 1 mm; in c = 100/tm.
mammalian species examined 1°, cannabinoid receptors in the heterozygous control mice are discretely and densely
localized to the cerebellar molecular layer, with very low binding in the granule cell layer, the cerebellar white
304
b ,el
G ~d
wvlwv
~
9
•
Q
"
I , , , ,
¸ ,
4
• I,
Fig. 2. Photographs of Nissl-stained sections of heterozygous and homozygous gene mutations leading to losses of granule cells in the cerebellum. The heterozygous weaver (+/wv) mouse has a normal appearance of lamination and organization (a and c), but the homozygote (wv/wv) shows greatly reduced molecular and granule layers and intermingling of Purkinje cells with granule cells (b and d). The normal appearance of the heterozygous reeler (+/rl) mouse (e and g) contrasts with tremendously reduced cortical size in the homozygous (rl/rl) mouse (f). High magnification reveals near absence of a molecular layer and Purkinje cells intermingled in a greatly reduced granule layer (h). Large arrows in the low-power photographs point to the deep cerebellar nuclei, which retain their comparative size in the homozygotes. Curved arrow in f points to a visible portion of the molecular layer, where densitometric measurements were taken. Small arrows in the high-magnification photomicrographs point to Purkinje cell bodies. Same scale as in Fig. 1.
305
Fig. 3. Photographs of film images of [3H]CP55,940 binding to cannabinoid receptors in hetereozygous and homozygous mutants of each of the 4 lines. Heterozygous brains are shown in the top panel of each pair; homozygotes at the bottom. Abbreviations: pcd, Purkinje cell degeneration; nr, nervous; wv, weaver; rl, reeler. Same scale as in Fig. lb.
matter, and the deep cerebeilar nuclei. This pattern was also seen in the two lines of homozygous mutants with selective losses of Purkinje cells (pcd, Fig. 3b; nervous,
Fig. 3d). In contrast, the homozygous mutants with reductions of granule cells showed markedly reduced [3H]CP55,940 binding (weaver, Fig. 3f, reeler, Fig. 3h).
306
Fig. 4. Photographs of film images of [3H]forskolin binding to adenylate cyclase in heterozygous and homozygousmutants of each of the 4 lines. Heterozygous brains are shown in the top panel of each pair; homozygotesat the bottom. Abbreviations as in Fig. 3.
A nearly identical normal appearance and mutant pattern were seen with [3H]forskolin binding. In all of the heterozygotes, binding was selectively dense in the cerebellar molecular layer and very sparse in the granule
cell layer, white matter, and deep cerebellar nuclei (Fig. 4a,c,e,g). Among the homozygotes, only the weaver and reeler mutants showed reductions of binding in the molecular layer (Fig. 4f,h).
307
Fig. 5. Photographs of film images of [3H]PDBu binding to protein kinase C, an element of the PI system, in heterozygous and homozygous mutants of each of the 4 lines. Heterozygous brains are shown in the top panel of each pair; homozygotes at the bottom. Abbreviations as in Fig. 3.
The consequences of homozygous mutant gene expression on patterns of [3H]PDBu binding were rather different. Dense binding confined to the cerebellar
molecular layer was seen in all heterozygotes (Fig. 5a,c, e,g) and was essentially unchanged in all of the homozygotes (Fig. 5b,d,f,h). Moderately dense binding was seen
308 in the deep cerebellar nuclei in all heterozygotes and, similarly, this pattern was not altered in any of the homozygotes. Quantitative analysis of the binding density for each of the ligands in the cerebellar molecular layer of each group is shown in Fig. 6. The data show that [3H]CP55, 940 and [3H]forskolin binding is reduced in the homozygotes for both lines of granule cell-deficient mice. The magnitudes of reduction for the two ligands are similar, with binding reduced by about 50% in the weaver strain and by about 70% in the reeler. In two instances, the binding levels were actually greater, by about 30%, in the homozygotes than in the corresponding heterozygotes - [3H]CP55,940 in the pcd and [3H]PDBu in the weaver.
140
CP 55,940
~20
0
t
~ 100
~ ~
6o
o
40
~
20
N
8o
0 180 160 O 140 N .o 120 100 80 60 O 40 20 0
.x
xx
PCD
Nervous Weaver Reeler
PCD
Nervous Weaver Reeler
140
*
PDBu
o 120 N
~ 100
g ~
8o 6o
0
~
40
~
20 0
PCD
Nervous Weaver Reeler
Fig. 6. Plots of binding densities for each of the groups and ligands used in the study. Data are means and standard deviations (error bars), calculated as % of control, for heterozygous (t~) and homozygous ( I ) mice (n = 3-5). Significance, based on Student's t-test for independent samples: *P < 0.002; *P < 0.0001.
DISCUSSION The present results indicate that significant portions of cannabinoid receptors and forskolin binding sites are localized on elements of granule cells in the mouse cerebellum. Previously 1° we argued that cannabinoid receptors might have a special relationship to second messengers on the basis of several key observations: (1) cannabinoid receptors are among the most numerous of all receptors in the brain; (2) the overall distribution of cannabinoid receptors is similar to that of the two sets of second messengers, though differences are apparent; (3) the functional coupling to the adenylate cyclase system is well-established; cannabinoids inhibit cAMP production in cell lines 13 and in brain slices 1, and they inhibit cGMP production in the cerebellum 15 - - all in a receptormediated fashion; and (4) the binding of [3H]CP55,940 in rat brain is completely inhibited by GTP analogs 9, indicating a close association of cannabinoid receptors with G proteins and the cAMP cascade. The selective reduction of [3H]CP55,940 and [3H]forskolin binding in the molecular layer of the two lines of granule cell-deficient mice and not in the two lines of Purkinje cell-deficient mice indicates that both cannabinoid receptors and components of the adenylate cyclase system are selectively localized on granule cells, more specifically on their axons in the cerebeltar molecular layer. It seems likely, therefore, that cannabinoid receptors and adenylate cyclase are densely colocalized on the same axons and terminals and are mutually absent from other major cerebellar neuronal elements, namely the Purkinje cells. The possibility that some portion of the receptor and enzyme population in the molecular layer might also be localized on stellate and/or basket cells intrinsic to the molecular layer, or on extrinsic axons, cannot be ruled out by these results and in fact seems likely on the basis of remaining binding in the homozygous mutants. The presence of unaltered [3H]PDBu binding in both homozygotes and heterozygotes of all of the mutant lines suggests that protein kinase C is rather evenly distributed among the two major cerebellar neuronal elements. In another study of neuronal localization, cannabinoid receptors in the globus paUidus and substantia nigra disappear after chemical destruction of striatopallidal and striatonigral projection neurons 7"8, indicating that these receptors are also localized on axons and nerve terminals. Thus, the 3 brain areas containing the highest densities of cannabinoid receptors "~ contain them on axons. These areas also selectively contain some of the highest levels of second messengers as revealed by [3H]forskolin and [3H]PDBu binding, and lesion studies show that these too are axonally localized 5"29"3°.
309 A m a j o r difference b e t w e e n the basal ganglia and cerebellar systems is the primary transmitter each utilizes. In the basal ganglia the neurotransmitter in the striatofugal axons containing cannabinoid receptors is G A B A 8'2°, whereas the m a j o r transmitter among granule cell n e u r o n s a p p e a r s to be glutamate 1s'32. The present d a t a do not rule out the possibility that cannabinoid receptors are also localized to basket and stellate cells, which a p p e a r to be G A B A e r g i c 2°'a2. A n association of cannabinoid receptors with the PI system is partially s u p p o r t e d by the present data. Recent work has e m p h a s i z e d the rather exclusive localization of c o m p o n e n t s of this second messenger system with cerebellar Purkinje cells 3°'31. H o w e v e r , the claim that [3H]PDBu binding in cerebellar molecular layer is reduced by 90% in the h o m o z y g o u s nervous mutant 3° is not s u p p o r t e d by the present data showing no differences between levels of binding in the molecular layer of nervous homozygotes versus heterozygotes. The present results instead suggest that both of the m a j o r cerebellar cortical neuronal types - - Purkinje cells and granule cells - - have [3H]PDBu binding.
REFERENCES 1 Bidaut-Russell, M., Devane, W.A. and Howlett, A.C., Cannabinoid receptors and modulation of cyclic AMP accumulation in the rat brain, J. Neurochem., 55 (1990) 21-26. 2 Chaudry, A., Thompson, R.H., Rubin, R.P. and Laycheck, S.G., Relationship between d-9-tetrahydrocannabinoi-induced arachadonic acid release and secretagogue-evoked phosphoinositide breakdown and Ca2+ mobilization of exocrine pancreas, Mol. Pharmacol., 34 (1988) 543-548. 3 Devane, W.A., Dysarz, EA.I., Johnson, M.R., Melvin, L.S. and Howlett, A.C., Determination and characterization of a cannabinoid receptor in rat brain, Mol. Pharmacol., 34 (1988) 605-613. 4 Furuichi, T., Yoshikawa, S., Miyawaki, A., Wada, K., Maeda, N. and Mikoshiba, K., Primary structure and functional expression of the inositol 1,4,5-triphosphate-binding protein P4oo, Nature, 342 (1989) 32-34. 5 Gehlert, D.R., Dawson, T.M., Filloux, EM., Sanna, E., Hanbauer, I. and Wamsley, J.K., Evidence that [3H]forskolin binding in the substantia nigra is intrinsic to a striato-nigra projection; an autoradiographic study, Neurosci. Lett., 73 (1987) 114-118. 6 Hamburg, M., Analysis of the postnatal development of 'reeler', a neurological mutation in mice. A study in developmental genetics, Dev. Biol., 8 (1960) 165-185. 7 Herkenham, M., Lynn, A.B., De Costa, B. and Richfield, E.K., Neuronal localization of cannabinoid receptors in the basal ganglia, Soc. Neurosci. Abstr., 15 (1989) 905. 8 Herkenham, M., Lynn, A.B., De Costa, B. and Richfield, E.K., Neuronal localization of cannabinoid receptors in the basal ganglia of the rat, Brain Research, (1991) in press. 9 Herkenham, M., Lynn, A.B., Johnson, M.R., Melvin, L.S., De Costa, B.R. and Rice, K.C., Characterization and localization of cannabinoid receptors in rat brain: a quantitative in vitro autoradiographic study, J. Neurosci., 11 (1991) 563-583. 10 Herkenham, M., Lynn, A.B., Little, M.D., Johnson, M.R., Melvin, L.S., De Costa, B.R. and Rice, K.C., Cannabinoid receptor localization in brain, Proc. Natl. Acad. Sci. U.S.A., 87
Physiological studies do not support a coupling of cannabinoid receptors to the PI system. Cannabinoids are
reported to influence levels of phospholipids, among them PI, but the doses r e q u i r e d ~ M ) are higher than would be expected for r e c e p t o r m e d i a t i o n , and the relative potencies do not show p r o p e r s t r u c t u r e - a c t i v i t y relationship 2'23. T h e p r e s e n t a n a t o m i c a l studies show at least one m a j o r dissociation of localization of cannabinoid receptors and the PI system - - m a r k e r s for the PI system are localized in P u r k i n j e cell bodies and axons, whereas cannabinoid receptors are not found on this neuronal population. [3H]PDBu binding in the deep cerebellar nuclei, which appears to be selectively r e d u c e d in the homozygous pcd and nervous mutants (Fig. 4b,d) and not in the weaver and reeler mutants (Fig. 4f,h), suggests localization of protein kinase C to the axons and terminals of Purkinje cells projecting there. C a n n a b i n o i d receptors are virtually u n d e t e c t a b l e in the d e e p cerebellar nuclei. The discrete localization of the inositol 1,4, 5-triphosphate (IP3) r e c e p t o r to P u r k i n j e cells4'31 further differentiates the c a n n a b i n o i d r e c e p t o r distribution from components of the PI system.
(1990) 1932-1936. 11 Howlett, A.C., Cannabinoid inhibition of adenylate cyclase: relative activity of constituents and metabolites of marijuana, Neuropharmacology, 26 (1987) 507-512. 12 Howlett, A.C., Bidaut-Russell, M., Devane, W.A., Melvin, L.S., Johnson, M.R. and Herkenham, M., The cannabinoid receptor: biochemical, anatomical and behavioral characterization, Trends Neurosci., 13 (1990) 420-423. 13 Howlett, A.C., Johnson, M.R., Melvin, L.S. and Milne, G.M., Nonclassical cannabinoid analgesics inhibit adenylate cyclase: development of a cannabinoid receptor model, Mol. Pharmacol., 33 (1988) 297-302. 14 Johnson, M.R. and Melvin, L.S., The discovery of nonclassical cannabinoid anaigetics. In: R. Mechoulam (Ed.), Cannabinoids as Therapeutic Agents, CRC Press, Boca Raton, FL, 1986, pp. 121-145. 15 Koe, B.K., Milne, G.M., Weissman, A., Johnson, M.R. and Melvin, L.S., Enhancement of brain [3H]flunitrazepam binding and analgesic activity of synthetic cannabimimetics, Eur. J. Pharmacol., 109 (1985) 201-212. 16 Landis, S.C., Ultrastructural changes in mitochondria of cerebellar Purkinje cells of nervous mutant mice, J. Cell Biol., 57 (1973) 782-797. 17 Little, EJ., Compton, D.R., Johnson, M.R. and Martin, B.R., Pharmacology and stereoselectivity of structurally novel cannabinoids in mice, J. Pharmacol. Exp. Ther., 247 (1988) 10461051. 18 Liu, C.-J., Grandes, P., Matute, C., Cu6nod, M. and Str¢it, P., Glutamate-like immunoreactivity revealed in rat olfactory bulb, hippocampus and cerebellum by monoclonal antibody and sensitive staining method, Histochemistry, 90 (1989) 427-445. 19 Martin, B.R., Balster, R.L., Razdan, R.K., Harris, L.S. and Dewey, W.L., Behavioral comparisons of the stereoisomers of tetrahydrocannabinols, Life Sci., 29 (1981) 565-574. 20 Mugnani, E. and Oertel, W.H., An atlas of the distribution of GABAergic neurons and terminals in the rat CNS as revealed by GAD immunohistochemistry. In: A. Bj6rklund and T. H6kfelt (Eds.), Handbook of Chemical Neuroanatomy, Vol. 4 (GABA and Neuropeptides in the CNS, Part I), Elsevier, Amsterdam,
310 1985, pp. 436-608. 21 Mullen, R.J., Eicher, E.M. and Sidman, R.L., Purkinje cell degeneration, a new neurological mutation in the mouse, Proc. Natl. Acad. Sci. U.S.A., 73 (1976) 208-212. 22 Razdan, R.K., Structure-activity relationships in cannabinoids, Pharmacol. Rev., 38 (1986) 75-149. 23 Reichman, M., Nen, W. and Hokin, L.E., d9-Tetrahydrocan nabinol increases arachadonic acid levels in guinea pig cerebral cortex slices, Mol. Pharmacol., 34 (1988) 823-828. 24 Sidman, R.L., Development of interneuronal connections in brains of mutant mice. In: ED. Carlson (Ed.), Physiological and Biochemical Aspects of Nervous Integration, Prentice-Hall, Englewood, Cliffs, NJ, 1968, pp. 163-193. 25 Sidman, R.L. and Green, M.C., Nervous, a new mutant mouse with cerebellar disease. In: M. Sabourdy (Ed.), Les Mutants Pathologique chez l'Animal -- Leur Intdr~t dans la Recherche Biom~dicale, CNRS, Paris, 1970, pp. 69-79. 26 Sidman, R.L., Green, M.C. and Appel, S.H., Catalog of the Neurological Mutants of the Mouse, Harvard University Press, Cambridge, MA, 1965. 27 Speth, R.C. and Yamamura, H.I., Benzodiazepine receptors:
28
29
30
31
32
alterations in mutant mouse cerebellum, Eur. J. Pharmacol., 54 (1979) 397-399. Wilson, R.S., May, E.L., Martin, B.R. and Dewey, W.L., 9-Nor-9-hydroxyhexahydrocannabinols. Synthesis, some behavioral and analgesic properties, and comparison with the tetrahydrocannabinols, J. Med. Chem., 19 (1976) 1165-1167. Worley, P.F., Baraban, J.M., De Souza, E.B., and Snyder, S.H., Mapping second messenger systems in the brain: differential localization of adenylate cyclase and protein kinase C, Proc. Natl. Acad. Sci. U.S.A., 83 (1986) 4053-4057. Worley, P.F., Baraban, J.M. and Snyder, S.H., Heterogeneous localization of protein kinase C in rat brain: autoradiographic analysis of phorbol ester binding, J. Neurosci., 6 (1986) 199-207. Worley, P.E, Baraban, J.M. and Snyder, S.H., Inositol 1, 4,5-triphosphate receptor binding: autoradiographic localization in rat brain, J. Neurosci., 9 (1989) 339-346. Zhang, N., Walberg, F., Laake, J.H., Meldrum, B.S. and Ottersen, O.P., Aspartate-like and glutamate-like immunoreactivities in the inferior olive and climbing fibre system: a light microscope and semiquantitative electron microscopic study in rat and baboon (Papio anubus), Neuroscience, 38 (1990) 61-80.
Brain Research, 552 (1991) 301-310 Elsevier ADONIS 000689939116730T BRES 16730
Neuronal localization of cannabinoid receptors and second messengers in mutant mouse cerebellum M. Herkenham 1, B.G.S. Groen 1, A.B. Lynn 1, B.R.
D e C o s t a 2 a n d E . K . Richfield 1'*
1Section on Functional Neuroanatomy, National Institute of Mental Health, and 2Laboratory of Medicinal Chemistry, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD (U.S.A.) (Accepted 22 January 1991) Key words: Tetrahydrocannabinol; [3H]CP 55,940; Forskolin; Phorbol ester; Adenylate cyclase; Phosphoinositide
Four lines of mutant mice were used to investigate (1) the neuronal localization of cannabinoid receptors in the cerebellar molecular layer and (2) the anatomical association of these receptors with elements of the two second messenger systems in the brain. Two of the mutant lines Purkinje cell degeneration and nervous - - are selectively deficient in Purkinje cells; the other two - - weaver and reeler - - are deficient in granule cells. In the heterozygous mice, [3H]CP 55,940 binding to cannabinoid receptors was discretely and densely localized to the molecular layer, as was [3H]forskolin binding to adenylate cyclase and [3H]phorbol 12,13-dibutyrate binding to protein kinase C, a component of the phosphoinositide cycle. [3H]CP 55,940 and [3H]forskolin binding was selectively reduced in weaver and reeler homozygous mice but unchanged in Purkinje cell deficient and nervous homozygotes. No decreases in [3H]phorpbol 12,13-dibutyrate binding were found in any of the homozygous mutants relative to the heterozygous littermates. The results suggest that cannabinoid receptors and adenylate cyclase are localized to granule cell axons in the molecular layer, whereas protein kinase C is equally distributed in parallel fibers and Purkinje cell dendrites.
-
-
INTRODUCTION The actions of d9-tetrahydrocannabinol (A9-THC),the major psychoactive ingredient of marijuana (Cannabis sativa) 22, appear to be mediated by the recently characterized cannabinoid receptor 3'1°. A9-THC and other natural and synthetic cannabinoids produce characteristic m o t o r effects in animals. Structure-activity relationships and localization patterns 1° suggest that cannabinoid receptors in the basal ganglia and cerebellum mediate the catalepsy and ataxia elicited by systemic administration of cannabinoids in mice and dogs 17A9"2°. Recently, we have shown that the receptors in the basal ganglia are localized tO striatal neurons, very densely to their axons and terminals in the globus pallidus, entopeduncular nucleus, and substantia nigra pars reticulata 7'8. The other area in the brain containing very dense cannabinoid receptor binding is the cerebellum, where receptors are discretely localized to the molecular layer. We wondered whether this too might represent axonal localization. This possibility was reinforced by observations of axonal binding of ligands marking second messengers, in patterns that match the cannabinoid binding patterns in both the basal ganglia and cerebellum 5'29'3°. Cannabinoids are known to modulate the adenylate cyclase second messenger system via the
cannabinoid receptor 11-13, but evidence for modulation of the phosphoinositide (PI) cycle is scarce 2'23. Specific questions arise regarding the anatomical correlations. First, are cannabinoid receptors in the cerebellum neutonally localized, and if so, to which cerebellar elements are they confined? Second, do they reside on the same elements that are enriched in either of the second messengers? These questions have been addressed using section binding assays and autoradiography performed on lines of mutant mice showing selective losses of either Purkinje cells or granule cells. The neuronal localization patterns are compared with patterns of [3H]forskolin and [3H]phorbol 12,13-dibutyrate ([3H]PDBu) binding to the adenylate cyclase and PI systems, respectively. MATERIALS AND METHODS Animals Litters of mice were obtained from Jackson Laboratories (Bar Harbour, ME). Four lines of cerebeUar mutants were obtained (Table I). Upon arrival, each litter, including the mother, was housed in one cage. The litters contained animals homozygous and heterozygous for the mutant gene. Homozygotes showed slow weight gain and ataxia, whereas heterozygotes showed normal weight gains and normal motor behavior. Animals were given tap water and dry food ad libitum, and were housed at approximately 24 °C with lights on from 06.00 h to 18.00 h. After the appropriate survival time, to ensure maximal expression of the phenotype (Table I), the mice were decapitated, and the
*Present address: University of Rochester, Monroe Community Hospital, Rochester, NY 14620, U.S.A. Correspondence: M. Herkenham, Section on Functional Neuroanatomy, NIMH, Bldg. 36, Rm. 2D-15, Bethesda, MD 20892, U.S.A.
302 brains were removed, frozen in isopentane at -30 °C, and stored at -40 °C until use. Whole brains were mounted onto cryostat pedestals with embedding matrix (M-l, Lipshaw). Serial 20 pm thick sections were cut in a cryostat at -18 °C through the whole brain in one and through the whole cerebellum in the remaining 4 brains in every group. At levels interspaced by 100/~m, 7 serial sections were saved. The sections were thaw-mounted onto separate gelatincoated slides, desiccated overnight at 0 °C, and stored at -40 °C until use. One set of sections from each series was fixed in formalin, defaned, and stained in Cresyl violet for examination of cytoarchitecture; the remaining sets were saved for binding assays. Binding assays [3H]CP 55,940 is one of a series of synthetic cannabinoids whose structure and biological activity have been documented 3A°'14.Assay conditions were described previously1°. Slide-mounted sections taken from 5 animals in each group were brought to room temperature and incubated at 37 °C for 2.5 h in 50 mM Tris-HC1 buffer, pH 7.4, with 5% bovine serum albumin (BSA) and 10 nM [3H]CP 55,940 (sp. act. 48 Ci/mmol; custom radiolabeled at New England Nuclear). Sections were washed at 0 °C for 2 x 2 h in the same buffer with 1% BSA and dried. Nonspecific binding was determined in adjacent sections by the addition of 10 pM CP 55,244 and represented about 20% of the total. Slides were exposed to 3H-sensitive film (LKB or Amersham) for 2 (nervous, reeler) or 3 (pcd, weaver) weeks and developed (D19, Kodak) at 20 °C for 4 min. [3H]Forskolin binding to adenylate cyclase associated with guanine nucleotide binding proteins was carried out as described previously29. Sections were incubated in cytomailers containing 50 mM Tris-HCl, pH 7.7, 100 mM NaCl, 5 mM MgCI2, and 10 nM [3H]forskolin (sp. act. 38.6 Ci/mmol; New England Nuclear) for 10 min at 24 °C. Slides were washed 2 x 2 rain in same buffer at 4 °C, briefly dipped in 4 °C water, and dried. Nonspecific binding was assessed in adjacent sections by the addition of 10 pM forskolin (Sigma) and represented < 5% of the total. Slides were exposed to film for 12 weeks before developing. [3H]PDBu binding to protein kinase C was carried out as described previously3°. Sections were incubated in cytomailers containing 50 mM Tris-HC1, pH 7.7, 100 mM NaCl, 1 mM CaCl2, and 2.5 nM [3H]PDBu (sp. act. 19.1 Ci/mmol; New England Nuclear) for 60 min at 33 °C. Following incubation, slides were washed 2 x 1 min in ice cold buffer and dried. Nonspecific binding was assessed by the addition of 1 /~M PDBu (Sigma) and represented < 5% of the total. Slides were exposed to film for 2 weeks before developing. Data analysis Sections from heterozygotes and homozygotes from each mutant line were co-exposed together along with tritium standards on the same films. Developed films were digitized with a solid state video camera (Sierra Scientific) and Macintosh II computer-based system for quantitative densitometry using IMAGE software (Wayne Rasband, Research Services Branch, NIMH). Transmittance levels were converted to fmol/mg tissue using the tritium standards (Amersham 3H Microscales). Light transmittance was measured in sections at levels containing the deep cerebellar nuclei. A threshold function was applied to each section to select and measure transmittance confined to the molecular layer. For each animal, structure, and ligand, measures from 2-4 sections containing the structure were averaged. The process was performed on images from adjacent sections incubated for total and for nonspecific binding, and the difference was computed to determine specific binding for the molecular layer. RESULTS
TABLE I Strains of mice used and times of sacrifice Mutant type
Strain
Deficiency
Time of onset of maximal defect (d) °
Age at sacrifice (d)
Purkinje cell degeneration ( p c d ) Nervous Weaver Reeler
C57BL C3HeB B6CBA B6C3
Purkinje cell Purkinje cell Granule cell Granule cell
>28 23-50 20-22 37-56
31 74-77 59-68 21-26
a Literature sources: pcd21; nervous16'25'27; weaver26"27;reeler 6'24'27.
of the cerebellum of each of the homozygous mutants, with a normal appearance of the brains of the heterozygous littermates. In every case, reduced cerebellar size was seen only in those animals that had showed reduced body weight and difficulties in m o v e m e n t and posture. The gross as well as the microscopic appearance of the cerebella noted in these brains conformed to previous detailed descriptions of the m u t a n t s 6A6'21'24-27. The normal cerebellar molecular layer consists of Purkinje and Golgi cell dendrites, granule cell axons (parallel fibers), scattered stellate and basket cells, and extrinsic afferent axons (climbing fibers) from the inferior olive. In each of the 4 homozygous m u t a n t types, the depth of the molecular layer a n d the arrangements of neurons in the Purkinje and granule cell layers is characteristically altered (Figs. 1 and 2). Two of the m u t a n t lines show selective losses of Purkinje cells. Mice heterozygous for the pcd gene show a n o r m a l appearing cerebellum (Fig. l a , c). Mice homozygous for the pcd gene show a dramatic loss of virtually all of their Purkinje neurons (Fig. ld) and a c o m m e n s u r a t e reduction in the width of the molecular layer and volume of the deep cerebellar nuclei (Fig. l b , d ) , reflecting the loss of Purkinje cell dendrites and axons in these locations, respectively. The nervous gene m u t a t i o n produces a phenotypic defect similar to that of the pcd, although surviving Purkinje cells are evident (Fig. lh). The two other m u t a n t lines examined in this study are characterized by relatively greater losses of granule than Purkinje cells; in these, the cerebellar deformation is much more dramatic (Fig. 2). T h e m a j o r defects in the mice homozygous for the weaver and reeler genes are a greatly s h r u n k e n molecular layer and an intermingling of Purkinje cells in the granule layer, which is reduced in size (Fig. 2). In both lines, however, the size of the deep cerebellar nuclei is normal in the homozygotes (Fig. 2b,
f). Gross appearances of the brains at the time of brain removal showed dramatic differences in the appearance
Figure 3 shows the pattern of [3H]CP55,940 binding in the cerebellum of each of the m u t a n t lines. As in all other
303
i .....
.
.
.
.
.
.
.
.
lib
•
I
Illm •
II
I
'~ '1)
, , • .-,
_ • -...
~ ,-
..
::,, ,,,
,,..
• ~ ,~.....
.. .i, ~
,
~%.,
"*, .~,
*
8
~e v
o
t O 'P ~11
etb s* a *
.¢
,-~
~p
0 . io " , D
~ t
~,
lie
Fig. 1. Photographs of Nisslostained sections of heterozygous and homozygous gene mutations leading to losses of Purkinje cells in the cerebellum. The heterozygous Purkinje cell-deficient (+/pc,d) mouse has a normal appearance of lamination and organization (a and c), but the homozygous (ped/pcd) mouse shows nearly complete loss of Purkinje cells (b and d). Similarly, the normal appearance of the heterozygous nervous (+/nr) mouse (e and g) is contrasted with the major depletion of Purkinje neurons in the homozygous (nr/nr) mouse (f and h). Large arrow in the low-power photographs point to the deep cerebellar nuclei, which are reduced in size in the homozygotes. Small arrows in the high-magnification photomicrographs point to Purkinje cell bodies. Scale bar in b = 1 mm; in c = 100/tm.
mammalian species examined 1°, cannabinoid receptors in the heterozygous control mice are discretely and densely
localized to the cerebellar molecular layer, with very low binding in the granule cell layer, the cerebellar white
304
b ,el
G ~d
wvlwv
~
9
•
Q
"
I , , , ,
¸ ,
4
• I,
Fig. 2. Photographs of Nissl-stained sections of heterozygous and homozygous gene mutations leading to losses of granule cells in the cerebellum. The heterozygous weaver (+/wv) mouse has a normal appearance of lamination and organization (a and c), but the homozygote (wv/wv) shows greatly reduced molecular and granule layers and intermingling of Purkinje cells with granule cells (b and d). The normal appearance of the heterozygous reeler (+/rl) mouse (e and g) contrasts with tremendously reduced cortical size in the homozygous (rl/rl) mouse (f). High magnification reveals near absence of a molecular layer and Purkinje cells intermingled in a greatly reduced granule layer (h). Large arrows in the low-power photographs point to the deep cerebellar nuclei, which retain their comparative size in the homozygotes. Curved arrow in f points to a visible portion of the molecular layer, where densitometric measurements were taken. Small arrows in the high-magnification photomicrographs point to Purkinje cell bodies. Same scale as in Fig. 1.
305
Fig. 3. Photographs of film images of [3H]CP55,940 binding to cannabinoid receptors in hetereozygous and homozygous mutants of each of the 4 lines. Heterozygous brains are shown in the top panel of each pair; homozygotes at the bottom. Abbreviations: pcd, Purkinje cell degeneration; nr, nervous; wv, weaver; rl, reeler. Same scale as in Fig. lb.
matter, and the deep cerebeilar nuclei. This pattern was also seen in the two lines of homozygous mutants with selective losses of Purkinje cells (pcd, Fig. 3b; nervous,
Fig. 3d). In contrast, the homozygous mutants with reductions of granule cells showed markedly reduced [3H]CP55,940 binding (weaver, Fig. 3f, reeler, Fig. 3h).
306
Fig. 4. Photographs of film images of [3H]forskolin binding to adenylate cyclase in heterozygous and homozygousmutants of each of the 4 lines. Heterozygous brains are shown in the top panel of each pair; homozygotesat the bottom. Abbreviations as in Fig. 3.
A nearly identical normal appearance and mutant pattern were seen with [3H]forskolin binding. In all of the heterozygotes, binding was selectively dense in the cerebellar molecular layer and very sparse in the granule
cell layer, white matter, and deep cerebellar nuclei (Fig. 4a,c,e,g). Among the homozygotes, only the weaver and reeler mutants showed reductions of binding in the molecular layer (Fig. 4f,h).
307
Fig. 5. Photographs of film images of [3H]PDBu binding to protein kinase C, an element of the PI system, in heterozygous and homozygous mutants of each of the 4 lines. Heterozygous brains are shown in the top panel of each pair; homozygotes at the bottom. Abbreviations as in Fig. 3.
The consequences of homozygous mutant gene expression on patterns of [3H]PDBu binding were rather different. Dense binding confined to the cerebellar
molecular layer was seen in all heterozygotes (Fig. 5a,c, e,g) and was essentially unchanged in all of the homozygotes (Fig. 5b,d,f,h). Moderately dense binding was seen
308 in the deep cerebellar nuclei in all heterozygotes and, similarly, this pattern was not altered in any of the homozygotes. Quantitative analysis of the binding density for each of the ligands in the cerebellar molecular layer of each group is shown in Fig. 6. The data show that [3H]CP55, 940 and [3H]forskolin binding is reduced in the homozygotes for both lines of granule cell-deficient mice. The magnitudes of reduction for the two ligands are similar, with binding reduced by about 50% in the weaver strain and by about 70% in the reeler. In two instances, the binding levels were actually greater, by about 30%, in the homozygotes than in the corresponding heterozygotes - [3H]CP55,940 in the pcd and [3H]PDBu in the weaver.
140
CP 55,940
~20
0
t
~ 100
~ ~
6o
o
40
~
20
N
8o
0 180 160 O 140 N .o 120 100 80 60 O 40 20 0
.x
xx
PCD
Nervous Weaver Reeler
PCD
Nervous Weaver Reeler
140
*
PDBu
o 120 N
~ 100
g ~
8o 6o
0
~
40
~
20 0
PCD
Nervous Weaver Reeler
Fig. 6. Plots of binding densities for each of the groups and ligands used in the study. Data are means and standard deviations (error bars), calculated as % of control, for heterozygous (t~) and homozygous ( I ) mice (n = 3-5). Significance, based on Student's t-test for independent samples: *P < 0.002; *P < 0.0001.
DISCUSSION The present results indicate that significant portions of cannabinoid receptors and forskolin binding sites are localized on elements of granule cells in the mouse cerebellum. Previously 1° we argued that cannabinoid receptors might have a special relationship to second messengers on the basis of several key observations: (1) cannabinoid receptors are among the most numerous of all receptors in the brain; (2) the overall distribution of cannabinoid receptors is similar to that of the two sets of second messengers, though differences are apparent; (3) the functional coupling to the adenylate cyclase system is well-established; cannabinoids inhibit cAMP production in cell lines 13 and in brain slices 1, and they inhibit cGMP production in the cerebellum 15 - - all in a receptormediated fashion; and (4) the binding of [3H]CP55,940 in rat brain is completely inhibited by GTP analogs 9, indicating a close association of cannabinoid receptors with G proteins and the cAMP cascade. The selective reduction of [3H]CP55,940 and [3H]forskolin binding in the molecular layer of the two lines of granule cell-deficient mice and not in the two lines of Purkinje cell-deficient mice indicates that both cannabinoid receptors and components of the adenylate cyclase system are selectively localized on granule cells, more specifically on their axons in the cerebeltar molecular layer. It seems likely, therefore, that cannabinoid receptors and adenylate cyclase are densely colocalized on the same axons and terminals and are mutually absent from other major cerebellar neuronal elements, namely the Purkinje cells. The possibility that some portion of the receptor and enzyme population in the molecular layer might also be localized on stellate and/or basket cells intrinsic to the molecular layer, or on extrinsic axons, cannot be ruled out by these results and in fact seems likely on the basis of remaining binding in the homozygous mutants. The presence of unaltered [3H]PDBu binding in both homozygotes and heterozygotes of all of the mutant lines suggests that protein kinase C is rather evenly distributed among the two major cerebellar neuronal elements. In another study of neuronal localization, cannabinoid receptors in the globus paUidus and substantia nigra disappear after chemical destruction of striatopallidal and striatonigral projection neurons 7"8, indicating that these receptors are also localized on axons and nerve terminals. Thus, the 3 brain areas containing the highest densities of cannabinoid receptors "~ contain them on axons. These areas also selectively contain some of the highest levels of second messengers as revealed by [3H]forskolin and [3H]PDBu binding, and lesion studies show that these too are axonally localized 5"29"3°.
309 A m a j o r difference b e t w e e n the basal ganglia and cerebellar systems is the primary transmitter each utilizes. In the basal ganglia the neurotransmitter in the striatofugal axons containing cannabinoid receptors is G A B A 8'2°, whereas the m a j o r transmitter among granule cell n e u r o n s a p p e a r s to be glutamate 1s'32. The present d a t a do not rule out the possibility that cannabinoid receptors are also localized to basket and stellate cells, which a p p e a r to be G A B A e r g i c 2°'a2. A n association of cannabinoid receptors with the PI system is partially s u p p o r t e d by the present data. Recent work has e m p h a s i z e d the rather exclusive localization of c o m p o n e n t s of this second messenger system with cerebellar Purkinje cells 3°'31. H o w e v e r , the claim that [3H]PDBu binding in cerebellar molecular layer is reduced by 90% in the h o m o z y g o u s nervous mutant 3° is not s u p p o r t e d by the present data showing no differences between levels of binding in the molecular layer of nervous homozygotes versus heterozygotes. The present results instead suggest that both of the m a j o r cerebellar cortical neuronal types - - Purkinje cells and granule cells - - have [3H]PDBu binding.
REFERENCES 1 Bidaut-Russell, M., Devane, W.A. and Howlett, A.C., Cannabinoid receptors and modulation of cyclic AMP accumulation in the rat brain, J. Neurochem., 55 (1990) 21-26. 2 Chaudry, A., Thompson, R.H., Rubin, R.P. and Laycheck, S.G., Relationship between d-9-tetrahydrocannabinoi-induced arachadonic acid release and secretagogue-evoked phosphoinositide breakdown and Ca2+ mobilization of exocrine pancreas, Mol. Pharmacol., 34 (1988) 543-548. 3 Devane, W.A., Dysarz, EA.I., Johnson, M.R., Melvin, L.S. and Howlett, A.C., Determination and characterization of a cannabinoid receptor in rat brain, Mol. Pharmacol., 34 (1988) 605-613. 4 Furuichi, T., Yoshikawa, S., Miyawaki, A., Wada, K., Maeda, N. and Mikoshiba, K., Primary structure and functional expression of the inositol 1,4,5-triphosphate-binding protein P4oo, Nature, 342 (1989) 32-34. 5 Gehlert, D.R., Dawson, T.M., Filloux, EM., Sanna, E., Hanbauer, I. and Wamsley, J.K., Evidence that [3H]forskolin binding in the substantia nigra is intrinsic to a striato-nigra projection; an autoradiographic study, Neurosci. Lett., 73 (1987) 114-118. 6 Hamburg, M., Analysis of the postnatal development of 'reeler', a neurological mutation in mice. A study in developmental genetics, Dev. Biol., 8 (1960) 165-185. 7 Herkenham, M., Lynn, A.B., De Costa, B. and Richfield, E.K., Neuronal localization of cannabinoid receptors in the basal ganglia, Soc. Neurosci. Abstr., 15 (1989) 905. 8 Herkenham, M., Lynn, A.B., De Costa, B. and Richfield, E.K., Neuronal localization of cannabinoid receptors in the basal ganglia of the rat, Brain Research, (1991) in press. 9 Herkenham, M., Lynn, A.B., Johnson, M.R., Melvin, L.S., De Costa, B.R. and Rice, K.C., Characterization and localization of cannabinoid receptors in rat brain: a quantitative in vitro autoradiographic study, J. Neurosci., 11 (1991) 563-583. 10 Herkenham, M., Lynn, A.B., Little, M.D., Johnson, M.R., Melvin, L.S., De Costa, B.R. and Rice, K.C., Cannabinoid receptor localization in brain, Proc. Natl. Acad. Sci. U.S.A., 87
Physiological studies do not support a coupling of cannabinoid receptors to the PI system. Cannabinoids are
reported to influence levels of phospholipids, among them PI, but the doses r e q u i r e d ~ M ) are higher than would be expected for r e c e p t o r m e d i a t i o n , and the relative potencies do not show p r o p e r s t r u c t u r e - a c t i v i t y relationship 2'23. T h e p r e s e n t a n a t o m i c a l studies show at least one m a j o r dissociation of localization of cannabinoid receptors and the PI system - - m a r k e r s for the PI system are localized in P u r k i n j e cell bodies and axons, whereas cannabinoid receptors are not found on this neuronal population. [3H]PDBu binding in the deep cerebellar nuclei, which appears to be selectively r e d u c e d in the homozygous pcd and nervous mutants (Fig. 4b,d) and not in the weaver and reeler mutants (Fig. 4f,h), suggests localization of protein kinase C to the axons and terminals of Purkinje cells projecting there. C a n n a b i n o i d receptors are virtually u n d e t e c t a b l e in the d e e p cerebellar nuclei. The discrete localization of the inositol 1,4, 5-triphosphate (IP3) r e c e p t o r to P u r k i n j e cells4'31 further differentiates the c a n n a b i n o i d r e c e p t o r distribution from components of the PI system.
(1990) 1932-1936. 11 Howlett, A.C., Cannabinoid inhibition of adenylate cyclase: relative activity of constituents and metabolites of marijuana, Neuropharmacology, 26 (1987) 507-512. 12 Howlett, A.C., Bidaut-Russell, M., Devane, W.A., Melvin, L.S., Johnson, M.R. and Herkenham, M., The cannabinoid receptor: biochemical, anatomical and behavioral characterization, Trends Neurosci., 13 (1990) 420-423. 13 Howlett, A.C., Johnson, M.R., Melvin, L.S. and Milne, G.M., Nonclassical cannabinoid analgesics inhibit adenylate cyclase: development of a cannabinoid receptor model, Mol. Pharmacol., 33 (1988) 297-302. 14 Johnson, M.R. and Melvin, L.S., The discovery of nonclassical cannabinoid anaigetics. In: R. Mechoulam (Ed.), Cannabinoids as Therapeutic Agents, CRC Press, Boca Raton, FL, 1986, pp. 121-145. 15 Koe, B.K., Milne, G.M., Weissman, A., Johnson, M.R. and Melvin, L.S., Enhancement of brain [3H]flunitrazepam binding and analgesic activity of synthetic cannabimimetics, Eur. J. Pharmacol., 109 (1985) 201-212. 16 Landis, S.C., Ultrastructural changes in mitochondria of cerebellar Purkinje cells of nervous mutant mice, J. Cell Biol., 57 (1973) 782-797. 17 Little, EJ., Compton, D.R., Johnson, M.R. and Martin, B.R., Pharmacology and stereoselectivity of structurally novel cannabinoids in mice, J. Pharmacol. Exp. Ther., 247 (1988) 10461051. 18 Liu, C.-J., Grandes, P., Matute, C., Cu6nod, M. and Str¢it, P., Glutamate-like immunoreactivity revealed in rat olfactory bulb, hippocampus and cerebellum by monoclonal antibody and sensitive staining method, Histochemistry, 90 (1989) 427-445. 19 Martin, B.R., Balster, R.L., Razdan, R.K., Harris, L.S. and Dewey, W.L., Behavioral comparisons of the stereoisomers of tetrahydrocannabinols, Life Sci., 29 (1981) 565-574. 20 Mugnani, E. and Oertel, W.H., An atlas of the distribution of GABAergic neurons and terminals in the rat CNS as revealed by GAD immunohistochemistry. In: A. Bj6rklund and T. H6kfelt (Eds.), Handbook of Chemical Neuroanatomy, Vol. 4 (GABA and Neuropeptides in the CNS, Part I), Elsevier, Amsterdam,
310 1985, pp. 436-608. 21 Mullen, R.J., Eicher, E.M. and Sidman, R.L., Purkinje cell degeneration, a new neurological mutation in the mouse, Proc. Natl. Acad. Sci. U.S.A., 73 (1976) 208-212. 22 Razdan, R.K., Structure-activity relationships in cannabinoids, Pharmacol. Rev., 38 (1986) 75-149. 23 Reichman, M., Nen, W. and Hokin, L.E., d9-Tetrahydrocan nabinol increases arachadonic acid levels in guinea pig cerebral cortex slices, Mol. Pharmacol., 34 (1988) 823-828. 24 Sidman, R.L., Development of interneuronal connections in brains of mutant mice. In: ED. Carlson (Ed.), Physiological and Biochemical Aspects of Nervous Integration, Prentice-Hall, Englewood, Cliffs, NJ, 1968, pp. 163-193. 25 Sidman, R.L. and Green, M.C., Nervous, a new mutant mouse with cerebellar disease. In: M. Sabourdy (Ed.), Les Mutants Pathologique chez l'Animal -- Leur Intdr~t dans la Recherche Biom~dicale, CNRS, Paris, 1970, pp. 69-79. 26 Sidman, R.L., Green, M.C. and Appel, S.H., Catalog of the Neurological Mutants of the Mouse, Harvard University Press, Cambridge, MA, 1965. 27 Speth, R.C. and Yamamura, H.I., Benzodiazepine receptors:
28
29
30
31
32
alterations in mutant mouse cerebellum, Eur. J. Pharmacol., 54 (1979) 397-399. Wilson, R.S., May, E.L., Martin, B.R. and Dewey, W.L., 9-Nor-9-hydroxyhexahydrocannabinols. Synthesis, some behavioral and analgesic properties, and comparison with the tetrahydrocannabinols, J. Med. Chem., 19 (1976) 1165-1167. Worley, P.F., Baraban, J.M., De Souza, E.B., and Snyder, S.H., Mapping second messenger systems in the brain: differential localization of adenylate cyclase and protein kinase C, Proc. Natl. Acad. Sci. U.S.A., 83 (1986) 4053-4057. Worley, P.F., Baraban, J.M. and Snyder, S.H., Heterogeneous localization of protein kinase C in rat brain: autoradiographic analysis of phorbol ester binding, J. Neurosci., 6 (1986) 199-207. Worley, P.E, Baraban, J.M. and Snyder, S.H., Inositol 1, 4,5-triphosphate receptor binding: autoradiographic localization in rat brain, J. Neurosci., 9 (1989) 339-346. Zhang, N., Walberg, F., Laake, J.H., Meldrum, B.S. and Ottersen, O.P., Aspartate-like and glutamate-like immunoreactivities in the inferior olive and climbing fibre system: a light microscope and semiquantitative electron microscopic study in rat and baboon (Papio anubus), Neuroscience, 38 (1990) 61-80.