Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

⁹-Tetrahydrocannabinol is the principal psychoactive ingredient found in the plant Cannabis sativa (marijuana) and it produces its effects by interacting with CB1 and CB2 cannabinoid receptors (Dewey, 1986; Howlett, 1995). CB1 receptors (CB1 and CB1A) are located primarily in the central nervous system, whereas CB2 receptors are found principally in the periphery (Munro et al., 1993; Shire et al., 1995). All subtypes of cannabinoid receptors belong to the large superfamily of G protein-coupled receptors (GPCRs) that traverse the plasma membrane seven times and activate intracellular G proteins. Heterotrimeric G proteins are composed of three distinct subunits,  (39-50 kDa),  (35-36 kDa), and  (6-10 kDa) and their activation by GPCRs produces an exchange of GTP for GDP on the -subunits. This results in the dissociation of the G protein from the receptor and the separation of the -GTP from the -subunits. Both the free -GTP and -subunits then proceed to regulate various downstream effectors (Gudermann et al., 1997). Pertussis toxin (PTX)-sensitive G proteins (i.e., G_i and G_o subtypes) appear to mediate the physiological effects of cannabinoids (Howlett, 1995), although recent studies also suggest a possible role for G_s (Glass and Felder, 1997; Maneuf and Brotchie, 1997; Felder et al., 1998). Regulation of intracellular effectors by cannabinoid receptors includes inhibition of adenylyl cyclase (Howlett, 1984), inhibition of voltage-gated Ca²⁺ channels (Mackie et al., 1995), activation of inwardly rectifying K⁺ channels (Mackie et al., 1995), and activation of mitogen-activated protein kinase (Bouaboula et al., 1995).

CB1 receptors are widely distributed throughout the mammalian central nervous system in relatively high density (Herkenham et al., 1990), particularly in areas involved in mediating the processes affected by marijuana. These brain regions include the cortex (cognition), hippocampus (memory), hypothalamus (body temperature), and cerebellum and basal ganglia (motor function) (Breivogel et al., 1997; Breivogel and Childers, 1998). Although the correlation between regional cannabinoid receptor number and the observed physiological effects is striking, the signal transduction mechanisms underlying these actions have not been determined. Cannabinoid agonists stimulate the binding of the hydrolysis-resistant GTP analog [³⁵S]guanosine-5'-O-(3-thio) triphosphate (GTPS), to G protein -subunits and this can be used as a measure of receptor activation. We have recently developed a method to measure the number of G proteins activated per occupied receptor (i.e., receptor/transducer amplification factors) by calculating the ratio of the apparent B_max of net agonist-stimulated [³⁵S]GTPS binding to the B_max of receptor binding (Breivogel et al., 1997). Using this technique, we found that the amplification factors for cannabinoid receptors differed between several rat brain regions, ranging from 2.0 in the frontal cortex to 7.5 in the hypothalamus. This suggests that although some brain regions contain lower densities of cannabinoid receptors (i.e., hypothalamus), CB1 stimulation results in the activation of an equivalent or greater number of G proteins relative to regions containing higher densities of receptors (i.e., frontal cortex).

Although agonist-induced [³⁵S]GTPS binding provides valuable information concerning activation of total G proteins in a particular brain region or tissue, this technique cannot be used to examine the coupling of GPCRs to individual G subunits. One approach to measure the activation of specific G proteins by receptors is to use agonist-stimulated incorporation of [³²P]azidoanilido-GTP (AA-GTP) into G subunits, followed by immunoprecipitation (IP) and separation with urea/SDS-polyacrylamide gel electrophoresis with subsequent autoradiography (AR) (Prather et al., 1994a,b, 1995; Chakrabarti et al., 1995). With this technique, individual G subunits can be separated into distinct bands on a single gel and thus receptor coupling to individual G proteins simultaneously can be evaluated. In this study, we sought to determine which specific G subunits were activated by CB1 receptors in several rat brain regions and if this coupling might explain the previously observed regional differences in receptor/G protein amplification factors. Furthermore, we examined whether cannabinoid agonists might activate different subtypes of G_i/G_o subunits with varying degrees of efficacy and/or potency. Our results demonstrate that stimulation of CB1 receptors by maximally effective concentrations of WIN 55212-2 results in the activation of a distinct pattern of at least five different G_i/G_o subunits in several brain regions. Furthermore, although the pattern of G proteins activated by WIN 55212-2 appears to be similar across brain regions, slight differences are observed in both the percentage of increase and the amount of the individual G subunits activated. Most importantly, the amount of WIN 55212-2 required to half-maximally activate individual G proteins in the cerebellum varies over a 30-fold range for different G subunits.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. Male Sprague-Dawley rats were purchased from Zivic Miller (Zeleinople, PA). [³²P]GTP (3000 Ci/mmol), [³⁵S]GTPS (1250 Ci/mmol), and antisera (EC2 and GC2) were purchased from NEN (Boston, MA). GDP for membrane [³⁵S]GTPS binding assays and unlabeled GTPS were purchased from Boehringer Mannheim (New York, NY). Antiserum LEP4 was a generous gift from Dr. Ping-Yee Law (University of Minnesota, Minneapolis, MN). Enhanced chemiluminescence (ECL) reagents and Hyperfilm-ECL were purchased from Amersham (Arlington Heights, IL). WIN 55212-2 and AM 281 were obtained from Tocris Cookson, Inc. (Ballwin, MO). All other reagents were purchased from Sigma Chemical Co. (St. Louis, MO).

Membrane Preparations. Brain regions were dissected from fresh rat brains on ice. Tissue samples were pooled and homogenized with a Tissumizer (Tekmar, Cincinnati, OH) in cold assay buffer (50 mM Tris-HCl, pH 7.4; 3 mM MgCl₂; 0.2 mM EGTA; and 100 mM NaCl) and centrifuged at 31,000g for 10 min at 4°C. Pellets were resuspended in membrane buffer, then centrifuged at 31,000g for 10 min at 4°C. Pellets were homogenized in membrane buffer, assayed for protein content (Bradford, 1976), and stored in aliquots at 80°C until being assayed.

Photoaffinity Labeling of G Subunits with [³²P]AA-GTP. The method for synthesis and purification of [³²P]AA-GTP can be found in Prather et al. (1994a). The photoaffinity labeling of G subunits with [³²P]AA-GTP also has been recently reported (Prather et al., 1994a,b, 1995; Chakrabarti et al., 1995). Plasma membranes (25 µg per assay) were incubated in the presence or absence of agonist for 6 min at 30°C in 100 µl of buffer I (50 mM HEPES, pH 7.4; 0.1 mM EDTA; 10 mM MgCl₂; 30 mM NaCl; 30 µM GDP; and 0.04 U/ml adenosine deaminase). After agonist incubation, [³²P]AA-GTP (1 µCi /assay) was added, and samples were incubated for an additional 6 min at 30°C. The reaction was terminated by placing samples on ice. Membranes were then collected by centrifugation at 12,000g for 10 min and resuspended in 100 µl of buffer II (50 mM HEPES, pH 7.4; 0.1 mM EDTA; 10 mM MgCl₂; 30 mM NaCl; and 2 mM dithiothreitol). Resuspended pellets (droplets) were then irradiated at 4°C with 240 milliJoules from an ultraviolet lamp (254 nm; 150 W) at a distance of 15 cm. Samples were centrifuged as before, resuspended in electrophoresis sample buffer, and separated by SDS-PAGE (see below).

o

i1

i2

SDS-PAGE and Immunoblotting. To identify G subunits, membranes were separated on 20-cm separating gels containing 10% acrylamide and 6 M urea (Prather et al., 1994a,b, 1995; Chakrabarti et al., 1995). Before separation, samples were resuspended in 80 µl of electrophoresis loading buffer (65 mM Tris HCl, pH 6.8; 2% SDS; 10% glycerol; and 5% 2-mercaptoethanol), and heated at 90°C for 2 min. The ECL method of immunoblotting was used (Amersham). Gels were transferred to Hybond-ECL nitrocellulose membranes and incubated overnight at 4°C with 10% milk in blotting buffer (TBS-0.1%; 25 mM Tris HCl, pH 7.6; 154 mM NaCl; and 0.1% Tween 20). Blots were then washed three times (5 min each) with TBS-0.1% and incubated with primary antibodies (1:1000) for 1 h at room temperature while shaking. The primary antibodies were then removed and blots were washed as described previously. Secondary antibody (donkey anti-rabbit immunoglobin horseradish peroxidase, 1:5000) was then added and incubated for 30 min, with shaking. The secondary antibody was removed and blots were washed 3× 5-min with TBS-0.3%, followed by 3× 5-min with TBS-0.1%. Blots were then incubated for 1 min with equal volumes of ECL detection reagents 1 and 2, wrapped in plastic wrap, and exposed to Hybond-ECL X-ray film for periods varying between 30 s and 10 min.

i3

o

o

i1

i2

i1

i2

Agonist-Stimulated [³⁵S]GTPS Binding Assays. Frozen membranes were thawed and then assayed for protein (Bradford, 1976). All assays included 10 to 20 µg of membrane protein and were conducted at 30°C for 2 h with 0.1% BSA (w/v), 30 µM GDP, and 0.05 nM [³⁵S]GTPS in a final volume of 1 ml. Nonspecific binding was determined with 30 µM unlabeled GTPS. WIN 55212-2 concentration-effect curves were determined by incubating membranes with various concentrations of WIN 55212-2 (0.03-30,000 nM). Reactions were terminated in all tubes simultaneously by rapid filtration under vacuum through Whatman GF/B glass fiber filters, followed by three washes with cold Tris buffer, pH 7.4. Bound radioactivity was determined by liquid scintillation spectrophotometry at 95% efficiency for ³⁵S after overnight extraction of the filters in 4 ml of Scintisate Econo 1 scintillation fluid.

Data Analysis. Unless otherwise stated, data represent the mean ± S.E. from at least three separate experiments that were each performed in triplicate. Data obtained from full concentration-effect curves using WIN 55212-2 were subjected to sigmoidal curve fitting with the Sigmaplot computer program. The minimum and maximum plateau values for the amount of G subunits activated (expressed in mean optical density units) and the amount of agonist required to produce 50% of maximal activation (ED₅₀) were determined from the best-fit curves. The maximum amount of G subunits activated was defined as the difference between the minimum and maximum plateau values. Percentage of increase in G protein activation was defined as the amount of [³²P]AA-GTP incorporated in the presence of agonist, divided by basal incorporation, times 100%. Net agonist-stimulated [³⁵S]GTPS binding values were calculated by subtracting basal binding values (absence of agonist) from agonist-stimulated values. Statistical significance of the data was determined by ANOVA followed by comparison with either the nonpaired two-tailed Student's t test or Tukey's method.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Photoaffinity Label [³²P]AA-GTP Identifies at Least Five Different PTX-Sensitive G Subunits in Cerebellar Membranes. Because brain cannabinoid receptors are primarily coupled to PTX-sensitive G proteins (i.e., G_i and G_o subtypes), we determined the identity of these G subunits in rat cerebellar membranes (Fig. 1). Membranes (25 µg/sample) were incubated with the photoaffinity label [³²P]AA-GTP and proteins were separated by urea/SDS-PAGE. After transfer to nitrocellulose membranes, blots were subjected to AR (Fig. 1, left), followed by immunoblotting with selective antibodies for individual G subunits (IB) (Fig. 1, middle). Additional samples were subjected to photoaffinity labeling with [³²P]AA-GTP and subsequently immunoprecipitated with antisera specific for individual G subunits before separation by urea/SDS-PAGE (IP + AR) (Fig. 1). [³²P]AA-GTP was incorporated into five detectable bands in the 39 to 41 kDa range, designated as proteins 1 to 5 from highest to lowest molecular mass (Fig. 1, left). The molecular mass of these photoaffinity-labeled proteins is consistent with that previously documented for G_i/G_o subunits (Gudermann et al., 1997). Furthermore, pretreatment of cerebellar membranes with PTX altered the mobility of all five bands (data not shown), which is in agreement with evidence of the ability of PTX-catalyzed ADP-ribosylation to decrease the electrophoretic mobility of G_i and G_o proteins (Ribeiro-Neto and Rodbell, 1989).

View larger version (43K): [in this window] [in a new window]   Fig. 1.   Identification of PTX-sensitive G subunits in membranes prepared from rat cerebellum. Membranes (25 µg/sample) were incubated with the photoaffinity label [32P]AA-GTP and proteins were separated by urea/SDS-PAGE. After transfer to nitrocellulose membranes, blots were subjected to AR (left), followed by immunoblotting with selective antibodies for individual G subunits (IB) (middle). Additional samples were subjected to photoaffinity labeling with [32P]AA-GTP and subsequently immunoprecipitated with antisera specific for individual G subunits before separation by urea/SDS-PAGE (IP + AR) (right). Specific G subunit antisera used were GC2 (top middle; Go) or LEP4 (bottom middle; Gi1/Gi2). Antibody-protein complexes were visualized with ECL and goat anti-rabbit conjugated with horseradish peroxidase as secondary antibodies. Inset, summary autoradiogram (left) in which the G protein identity (right) of all radioactive bands is depicted.

o

o3

o1

o2

o1

o2

o3

o1

o2

o3

o3

i1

i2

i1

i2

i1

i2

i3

i1

i1

i3

o

o3

o1

o2

i1

i2

i1

i2

o3

i1/

i3

o1

i2

o2

Activation of Cannabinoid Receptors by WIN 55212-2 Produces a Distinct Pattern of G-Protein Activation That Is Similar between Several Brain Regions. Because previous studies had shown brain regional differences in cannabinoid-activated [³⁵S]GTPS binding (Breivogel et al., 1997), we compared the ability of the full cannabinoid agonist WIN 55212-2 (10 µM) to activate individual G subunits in five rat brain regions (cerebellum, hippocampus, striatum, amygdala, and hypothalamus) (Fig. 2). When membranes prepared from each of these regions were incubated with [³²P]AA-GTP in the presence of a maximally effective concentration (10 µM) of the cannabinoid agonist WIN 55212-2, agonist-stimulated [³²P]AA-GTP labeling was observed in all five of the previously identified autoradiographic bands (Fig. 2, upper inset). However, because the relatively high optical density of G_o1 (band 3) interfered with the densitometric quantification of G_i1/G_i3 (band 2) and G_i2 (band 4), agonist-induced increases in photoaffinity labeling were only determined for G_o3 (band 1), G_o1 (band 3), and G_o2 (band 5). With the exception of the cerebellum, WIN 55212-2 produced similar percentage increases (70-108%) in the activation of all G proteins examined (Fig. 2, top). Interestingly, in the cerebellum greater overall percentage of stimulation of G proteins and a slightly different pattern was observed, with the most stimulation for G_o3 (212%) and similar levels of activation of G_o2 (161%) and G_o1 (132%). When the data were presented as amount of G protein stimulated (i.e., increase in optical density units), the pattern of G subunit activation by WIN 55212-2 was similar for all brain regions, with the greatest amount of activation of G_o1, followed by similar stimulation of G_o3 and G_o2 (Fig. 2, bottom).

View larger version (43K): [in this window] [in a new window]   Fig. 2.   Activation of individual G subunits by WIN 55212-2 in membranes prepared from several rat brain regions. Top, autoradiogram of G subunits photoaffinity labeled with [32P]AA-GTP (1 µCi) in the absence (left lanes) or presence (right lanes) of a maximal concentration of the cannabinoid agonist WIN 55212-2 (10 µM) in membranes (25 µg) prepared from cerebellum (CB), hippocampus (HI), striatum (ST), amygdala (AM), or hypothalamus (HY). Photolabeled G subunits were subsequently separated by urea/SDS-PAGE, exposed for AR and quantified by densitometry. To determine the amount of relative radioactivity incorporated by individual G proteins, the area of each band was traced and multiplied by its mean optical density. Data are presented as the percentage increase in G protein labeling from control (middle) and the increase in mean optical density units (bottom) for each brain region examined. The values presented represent the mean ± S.E. from four separate experiments.

View this table: [in this window] [in a new window]   TABLE 1 Maximal activation of total G proteins by WIN 55212-2 in rat brain regions by [35S]GTPS binding or [32P]AA-GTP photoaffinity labeling Cannabinoid activation of total G proteins was measured by [32P]AA-GTP photoaffinity labeling and [35S]GTPS binding with 10 µM WIN 55212-2 in membranes prepared from various rat brain regions. Results are expressed as net OD units from densitometric analysis of 32P autoradiograms and net picomoles per microgram of agonist-stimulated [35S]GTPS binding, along with percentage of stimulation by agonist in each assay. Data represent the mean ± S.E. from at least three to four separate experiments.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Reversal of WIN 55212-2-induced activation of G proteins by the cannabinoid antagonist AM 281 in membranes prepared from rat cerebellum. Rat cerebellar membranes (25 µg) were incubated with the photoaffinity label [32P]AA-GTP (1 µCi) and a maximal concentration of the cannabinoid agonist WIN 55212-2 (10 µM) in the presence of increasing concentrations (0.001-10 µM) of the cannabinoid antagonist AM 281. Photolabeled G subunits were subsequently separated by urea/SDS-PAGE, exposed for AR, and quantified by densitometry. To determine the amount of relative radioactivity incorporated by individual G proteins, the area of each band was traced and multiplied by its mean optical density. The percentage of maximum response of total G protein activation by 10 µM WIN 55212-2 (i.e., the sum of all individual G subunits) is plotted against the corresponding antagonist concentrations. The values presented for each concentration of AM 281 (0.001-1 µM) are the results of a single experiment, whereas the values presented for the 10 µM concentration of AM 281 represent the mean ± S.E. from four separate experiments.

WIN 55212-2 Activates Multiple G Proteins with Different Potencies in Cerebellar Membranes. WIN 55212-2 produced the greatest activation and a distinct pattern of G protein coupling in the cerebellum. Thus, we chose this brain region to use full WIN 55212-2 concentration-effect curves to examine G protein activation. We have previously observed that concentration-effect curves for cannabinoid-stimulated [³⁵S]GTPS binding are characteristically very shallow, often extending three or four orders of magnitude (Breivogel et al., 1997, 1998). In this study, we compared the concentration-effect curves for WIN 55212-2-induced total G protein activation with both [³²P]AA-GTP photoaffinity labeling and [³⁵S]GTPS binding (Fig. 4). Both methods produced similar shallow curves with Hill slope values of 0.75 for [³⁵S]GTPS and 0.73 for [³²P]AA-GTP. Moreover, for both radioligands, G protein activation was significant beginning with 0.01 µM WIN 55212-2 and not reaching maximal stimulation until 3 to 10 µM. The concentration of WIN 55212-2 required to produce 50% of the maximum stimulation was 1.02 µM for [³²P]AA-GTP photoaffinity labeling and 0.241 µM for [³⁵S]GTPS binding (Table 2).

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Concentration-dependent activation of total G proteins by WIN 55212-2 determined by [35S]GTPS binding and [32P]AA-GTP photoaffinity labeling in membranes prepared from rat cerebellum. The ability of increasing concentrations (0.0003-100 µM) of the cannabinoid agonist WIN 55212-2 to produce increases in [35S]GTPS binding () and [32P]AA-GTP photoaffinity labeling () of total G proteins in membranes prepared from rat cerebellum was evaluated. Data are presented as percentage of maximum response; absolute values for [35S]GTPS were 0.04 pmol/mg basal and 0.27 pmol/mg stimulated, whereas values for [32P]AA-GTP were 873 optical density units basal and 2001 optical density units stimulated. Data represent the mean ± S.E. from a minimum of four experiments.

View this table: [in this window] [in a new window]   TABLE 2 Potency and efficacy of WIN 55212-2-induced activation of G protein -subunits determined by concentration-dependent [32P]AA-GTP labeling in rat cerebellar membranes Full dose-response curves (0.01-100 µM) for the cannabinoid agonist WIN 55212-2 to produce increases in [32P]AA-GTP photoaffinity labeling of individual G subunits were performed. Solubilized membranes containing photolabeled G proteins were either loaded directly onto gels (AR) or first immunoprecipitated with antisera specific for individual G subunits (IP + AR) and separated by Urea SDS/PAGE as described in the Experimental Procedures section. Minimum and maximum plateau values and the amounts of agonist required to produce 50% maximal activation (i.e., ED50) were determined by curve fitting of the sigmoidal dose-response curves. The maximum percentage of increase in G protein activation was defined as the difference between the minimum and maximum plateau values. Data represent the mean ± S.E. from four to seven separate experiments. Statistical significance of the data was determined by ANOVA followed by comparisons using Tukey's method.

o3

o1

o2

o3

o1

o2

View larger version (59K): [in this window] [in a new window]   Fig. 5.   Concentration-dependent [32P]AA-GTP labeling of individual PTX-sensitive G protein -subunits by the cannabinoid agonist WIN 55212-2 in rat cerebellar membranes (AR only). Top, autoradiogram of G subunits photoaffinity labeled with [32P]AA-GTP (1 µCi) in the presence of increasing concentrations (0.01-100 µM) of the cannabinoid agonist WIN 55212-2 in rat cerebellar membranes (25 µg), separated by urea/SDS-PAGE. Bottom, to determine the amount of relative radioactivity incorporated by individual G subunits, the area of each band was traced and multiplied by its mean optical density. The activation of each G subunit activated, expressed as percentage of increase from control, is plotted against the corresponding WIN 55212-2 concentrations. The values presented for each concentration represent the mean ± S.E. from seven separate experiments.

o1

o3

o1

o1

i1/

i3

i2

o

i1

i2

o

i1/

i2

o

o3

o1

o2

o2

o2

o

i1

i2

View larger version (39K): [in this window] [in a new window]   Fig. 6.   Concentration-dependent [32P]AA-GTP labeling of individual PTX-sensitive G protein -subunits by the cannabinoid agonist WIN 55212-2 in rat cerebellar membranes (IP + AR). Top, autoradiogram of G subunits photoaffinity labeled with [32P]AA-GTP (1 µCi) in the presence of increasing concentrations (0.03-10 µM) of the cannabinoid agonist WIN 55212-2 in rat cerebellar membranes (25 µg), first immunoprecipitated with antisera specific for Go (left) or Gi1/Gi2 (right) before separation by urea/SDS-PAGE. Bottom, to determine the amount of relative radioactivity incorporated by individual G subunits, the area of each band was traced and multiplied by its mean optical density. The activation of each G subunit activated, expressed as percentage of increase from control, is plotted against the corresponding WIN 55212-2 concentrations. The values presented for each concentration represent the mean ± S.E. from four or five separate experiments.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

It is generally accepted that cannabinoid receptors relay intracellular signals by coupling primarily to PTX-sensitive G proteins (Howlett, 1995). However, because CB1 receptors have the potential to interact with several isoforms of G_i/G_o proteins expressed in the brain (i.e., G_i1, G_i2, G_i3, G_o1, G_o2, and G_o3), the potential exists for significant diversity in intracellular signaling produced by cannabinoid agonists. The initial finding of this study, that the full cannabinoid agonist WIN 55212-2 activates at least five different G subunits in the G_i/G_o family, confirms previous findings that GPCRs can activate the full spectrum of G proteins that are available in the relevant cell types (Goetzl et al., 1994; Liu et al., 1994; Prather et al., 1994a,b,1995; Chakrabarti et al., 1995), and represents the first report of the simultaneous activation of several individual G subunits in brain by endogenous cannabinoid receptors. The finding that G_o activation by cannabinoids predominates in brain compared with G_i activation is predictable, based on previous reports that G_o subunits are in significant excess over G_i in brain (Sternweis and Robishaw, 1984; Spicher et al., 1992). Indeed, in agreement with these and other studies, the basal labeling patterns of [³²P]AA-GTP in this study suggest that G_o1 is the predominant G protein subtype expressed in rat brain.

This study also represents the first direct comparison between cannabinoid activation of brain G proteins by two different approaches: [³²P]AA-GTP photoaffinity labeling and [³⁵S]GTPS binding. There are a number of important similarities between the results of these two methods, including the concentration-effect curves for WIN 55212-2 and the overall regional distribution of agonist stimulation. The differences noted in the relative regional distribution of agonist stimulation between the two assays is probably related to the considerable differences in the level of percentage stimulation observed. This difference is inherent in the basic properties of the two procedures and is the result of a number of factors. First, the affinity of GTPS for the agonist-activated form of G is higher than that of [³²P]AA-GTP (D. E. Selley and S.R.C, unpublished observations). Second, the concentrations of the two radioligands are very different: 0.05 nM for [³⁵S]GTPS and 3 nM for [³²P]AA-GTP. Finally, the fold stimulation is clearly affected by the ratio of radioligand to GDP, which is used in both assays to reduce basal radioligand binding.

We have previously shown that the amplification factors for cannabinoid receptors, defined as the number of total G proteins activated per occupied receptor, differed between several rat brain regions (Breivogel et al., 1997). If different brain regions possess unique stoichiometric compositions of cannabinoid receptors and G proteins, it is possible that cannabinoid receptors might activate different subtypes of G_i/G_o subunits with varying degrees of efficiency that would differ between brain regions. To test this hypothesis, this study found that activation of CB1 receptors by WIN 55212-2 resulted in coupling to a distinct pattern of at least five different G_i/G_o subunits in several brain regions. Interestingly, with the exception of slight differences in the cerebellum, the pattern of G_o activation by cannabinoid receptors was similar between all brain regions examined. Additionally, when total G protein activation was compared across brain regions, WIN 55212-2 produced the greatest percentage of increase in the cerebellum, followed by the hypothalamus > hippocampus = striatum = amygdala. This was similar to the regional distribution of WIN 55212-2-stimulated [³⁵S]GTPS binding previously reported in brain membranes with 10 µM WIN 55212-2 (Selley et al., 1996). Because the overall pattern of individual G labeling by WIN 55212-2 was similar between regions previously shown to have significant differences in receptor/G protein amplification by [³⁵S]GTPS and [³H]WIN 55212-2 binding (Breivogel et al., 1997), it is unlikely that the differences in receptor/G protein amplification are primarily mediated by different populations of G activation by CB1 receptors.

When activation of individual G subunits by WIN 55212-2 was examined using [³²P]AA-GTP photoaffinity labeling, it was demonstrated that the agonist activated different subtypes of G_i/G_o subunits with varying degrees of potency. For example, low agonist concentrations (0.10-0.62 µM) preferentially activated several G protein subtypes (i.e., G_o1, G_o3, G_i1, and G_i2), whereas G_o2 was stimulated at higher concentrations of agonist (i.e., 3.7 µM). It was also evident from the autoradiograms that several minor bands were activated at higher agonist concentrations like G_o2, but the inability of the IP procedures to adequately separate and identify these bands precluded their precise identification. Interestingly, the overall EC₅₀ values for WIN 55212-2 in stimulating [³⁵S]GTPS binding (0.24 µM) and [³²P]AA-GTP photoaffinity labeling (1 µM; see Fig. 4) are intermediate between these values of WIN 55212-2 in activating G_o2 and the other G subunits. Moreover, the potencies reported for G protein activation in this study correlate well with those for regulation of some intracellular effectors by WIN 55212-2. For example, the potency of WIN 55212-2 in inhibiting cAMP levels in intact cerebellar granule cells is 0.41 µM (Pacheco et al., 1993). However, the potency of WIN 55212-2 (<10 nM) in modulating calcium conductance in these cells (Gruol et al., 1996) is much higher than the >100 nM potencies observed in this study; it is possible that some effectors require less than full receptor occupancy to produce full responses.

Although the number of studies using full concentration-effect curves to examine the activation of individual G subunits by GPCRs are few, review of the current literature indicates that our observations are unique. Offermanns et al. (1994) observed a 20-fold difference in the concentration of carbachol required to activate G_i1/G_i3, relative to G_q/11, by m1 and m3 muscarinic receptors. However, these findings are not as surprising because G_i1/G_i3, and G_q/11 are members of completely different classes of G subunits. In contrast, the potency of µ-, -, and -opioid agonists required for half-maximal activation of individual G subunits within the same G_i/G_o class have been shown to differ by only by 2- to 3-fold (Prather et al., 1994a,b, 1995; Chakrabarti et al., 1995). Therefore, the 36-fold difference in the amount of WIN 55212-2 required to activate individual G subunits within the same G protein class (i.e., G_i/G_o) observed in this study may indicate an unique property of cannabinoid receptor signal transduction and/or of the agonist WIN 55212-2.

It is possible that the shallow concentration-effect curves for G protein activation by WIN 55212-2 is produced by these differences in agonist potencies in activating individual G subunits. However, this is not likely for several reasons. First, the contribution of G_o2 compared with the other G subunits is relatively low. Even if other minor bands are activated with the same lower agonist potency as G_o2, it still appears likely that the overall stimulation of G subunits would be dominated by G_o1, as well as the other bands activated with higher agonist potencies. Second, previous studies (Breivogel et al., 1998) have demonstrated the importance of multiple-affinity states for agonist binding to cannabinoid receptors themselves in producing the overall activation of brain G proteins. Therefore, although differences in agonist potencies at individual G subunits may contribute somewhat to shallow concentration-effect curves for agonists, it is likely that this phenomenon is produced by a complex interrelationship between agonist, receptor and G protein.

This study has not addressed the potential mechanisms that would produce different potencies of a single cannabinoid agonist in activating different G subunits. One possibility is that WIN 55212-2 may be acting at different cannabinoid receptor subtypes in brain with differential affinities for different G subunits. A detailed pharmacological analysis comparing the potencies and efficacies of various cannabinoid agonists in activating individual G subunits should help address this question, and is currently underway in our laboratories. Another possibility is that the quantity of individual G subunits present in relevant neurons provides a major determinant driving receptor/G protein coupling. It is possible that the receptor interacts more readily with G_o subunits because of their relative abundance compared with G_i, even though the affinity of the receptor for G_i might be higher. However, this hypothesis does not explain why in this study the potency of WIN 55212-2 to activate G_o1 and G_o3 is similar, despite the vast differences in their abundance as determined by [³²P]AA-GTP labeling. Finally, a recent study found that the stability of the ternary complex formation of A₁-adenosine receptors was determined by the dissociation rate of interacting G proteins (Waldhoer et al., 1999). They hypothesize that although agonists may promote interaction of G protein-coupled receptors with several different types of G subunits, only those G proteins that dissociate from the receptor slowly would allow sufficient stability of the ternary complex to result in productive signal transduction. Their results suggest that although low concentrations of agonist are required to produce efficient coupling of receptors to physiological G proteins, higher concentrations may provide sufficient ternary complex stability to allow activation of nonphysiological G proteins. Therefore, it is possible that higher concentrations of WIN 55212-2 are required to provide sufficient stability of the cannabinoid receptor ternary complex to observe activation of G_o2.

Our findings enhance the level of current understanding of cannabinoid-coupled signal transduction and may have significant ramifications for potential development of new therapeutic agents. It is well known that cannabinoid receptors regulate the activity of several intracellular effectors, including inhibition of adenylyl cyclase (Howlett, 1984), inhibition of voltage-gated Ca²⁺ channels (Mackie et al., 1995), activation of inwardly rectifying K⁺ channels (Mackie et al., 1995), and activation of mitogen-activated protein kinase (Bouaboula et al., 1995). It also has been established that distinct G subunits can couple GPCRs to specific effectors. For example, G_i2 couples -opioid receptors to adenylyl cyclase (McKenzie and Milligan, 1990), whereas muscarinic and somatostatin receptors produce inhibition of Ca²⁺ channels through G_o1 and G_o2, respectively (Kleuss et al., 1991). Therefore, data from this study suggest that cannabinoid receptors may produce distinct intracellular signals by activation of a specific pattern of G proteins responsible for regulation of a unique blend of intracellular effectors in a concentration-dependent manner. Thus, at low concentrations of agonist certain intracellular effectors might preferentially be activated by specific G subunits, and coupling to additional effectors would be recruited at higher concentrations of agonist required to activate additional G protein subtype(s). These data may help to explain why cannabinoid agonists inhibit adenylyl cyclase and modulate calcium conductance with different potencies in the same cells (Pacheco et al., 1993; Gruol et al., 1996). The complexity, and hence, potential flexibility of cannabinoid-mediated signal transduction is increased further given that different regions of the brain posses different stoichiometric compositions of cannabinoid receptors, G proteins, and effectors (Breivogel et al., 1997). Last, with the observation that different types of opioid- (Standifer and Pasternak, 1997) and cannabinoid- (Cook et al., 1995) induced antinociception may be mediated in part by specific G subtypes, it might be possible to develop agonists that at optimal concentrations preferentially activate G subunits responsible for the therapeutic effects of cannabinoids (i.e., antinociception), while avoiding activation of other G subunits potentially mediating undesirable actions (i.e., disruption of short-term memory).