Introduction

Top Abstract Introduction Procedures Results Discussion References

It is now well established that ⁹-THC (the main active principle of marijuana), anandamide (an endogenous ligand) (Devane et al., 1992) and synthetic cannabinoid receptor agonists mediate their cellular effects through specific cannabinoid receptors belonging to the large multigene superfamily of the GPCR. To date, two human cannabinoid receptor cDNAs have been identified, designated CB1 and CB2 (Matsuda et al., 1990, Munro et al., 1993). Two splice variants of CB1 mRNA (CB1 and CB1A) have been described in human and rat brain (Shire et al., 1995). Although CB1 mRNA is predominantly expressed in the brain (Matsuda et al., 1990, Westlake et al., 1994), it has also been detected in testis (Gérard et al., 1991), spleen cells (Kaminski et al., 1992) and in leukocytes (Bouaboula et al., 1993). By contrast the CB2 subtype has not been detected in the brain, but found remarkably abundant in immune tissues (Galiègue et al., 1995, Derocq et al., 1995), and may account for cannabinoid-mediated immune responses. Both receptors mediate their effects via a pertussis toxin-sensitive GTP-binding regulatory protein Gi. Upon stimulation, both CB1 and CB2 induced the inhibition of AC (Howlett and Fleming, 1984, Slipetz et al., 1995, Felder et al., 1995) and the activation of the MAPK (Bouaboula et al., 1995a, 1996). In neurons, CB1 has been found to be associated with the inhibition of N-type (Mackie and Hille, 1992) or P/Q-type (Twitchell et al., 1997) calcium channels. In addition, CB1 activation has recently been shown to induce immediate-early gene expression such as Krox 24 through a cAMP-independent pathway (Bouaboula et al., 1995b). These central cannabinoid receptor-mediated effects were all prevented by the potent and selective CB1 receptor antagonist SR 141716 (RinaldiCarmona et al., 1994, 1995, 1996). There is now substantial evidences showing that for some GPCRs, antagonists induce effects that are opposite to those observed by agonists, thereby displaying negative intrinsic activity. This referring to inverse agonism (Lefkowitz et al., 1993, Kenakin, 1996). Such an effect was recently demonstrated for the selective CB1 receptor antagonist SR 141716 (Landsman et al., 1997, Bouaboula et al., 1997) that is therefore defined as an inverse agonist.

It is clearly established that the functionality and the expression of the GPCRs are dynamically regulated after agonist or antagonist exposure. Indeed it has often been observed that cell exposure to agonist causes the desensitization and a down-regulation of the receptors, whereas antagonist treatment leads to their up-regulation. Several in vivo studies have also demonstrated the same behavior for cannabinoid receptor agonist ligands after chronic treatment that induce tolerance to many pharmacological effects: anticonvulsant activity, catalepsy, depression of locomotor activity, hypothermia, hypotension, immunosuppression and static ataxia (Pertwee, 1991, Pertwee et al., 1993, Fan et al., 1994, Fride, 1995). In vivo, tolerance took place with a significant decrease in the density of cannabinoid receptors in striatal structures (Oviedo et al., 1993, Rodriguez De Fonseca et al., 1994) in the cerebellum (Fan et al., 1996) and in the spleen (Massi et al., 1997) as well as in the levels of cannabinoid receptor mRNA (Rubino et al., 1994).

So far no detailed insights on the modulation of CB1 functions by cannabinoid ligands have been described in vitro. Therefore, in our report we have explored the adaptive changes in CB1 receptors after CP 55,940 or SR 141716 treatment in vitro. Their effects on CB1 receptor binding properties, CB1 receptor cell surface expression and their associated effects on CB1 receptor-coupled second messenger were investigated in a CHO cell line stably expressing the CB1 receptor. Our results show that exposure of CHO-CB1 cells to the agonist CP 55,940 leads to a desensitization of the CB1 receptor functions (forskolin-induced cAMP accumulation and the MAP kinase activity). These effects were associated with agonist-promoted receptor internalization without modification of the total binding site number. By striking contrast MAPK signal transduction pathway was dramatically enhanced in SR 141716-pretreated CHO-CB1 cells. An effect that may be related to the observed receptor externalization to the plasma membrane from an intracellular pool in response to its binding with the inverse agonist SR 141716.

	Experimental Procedures

Top Abstract Introduction Procedures Results Discussion References

Materials. Fatty acid-free BSA, IBMX and forskolin were from Sigma Chemical Co. (St. Louis, MO). Biofluor liquid scintillant and [³H]-CP 55,940 (103.4 Ci/mmol) were purchased from New England Nuclear Corporation (Paris, France). [³H]-SR 141716A (42 Ci/mmol), ³³P ATP (112.9 Ci/mmol), cAMP scintillant proximity assay, the enhanced chemiluminescence detection system and Biotrack p42/p44 MAPK kits were from Amersham (Les Ulis, France). RO 20-1724{4-[(3-butoxy-4-methoxyphenyl)methyl]-2-imidazolidinone} was purchased from Research Biochemicals Incorporated (Illkirch, France). SR 141716 [N-piperidino-5-(4-chlorophenyl)-1-(2,4-dichlorophenyl)-4-methylpyrazole-3-carboxamide] and CP 55,940 {[1, 2-(R)-5-(1,1-dimethylheptyl)-2-[5-hydroxy-2-(3-hydroxypropyl)cyclohexyl]-phenol} were synthesized at Sanofi Recherche. Drugs were dissolved in dimethyl sulfoxide. The solvent was limited to a 0.1%-concentration (v/v) that does not interfere in radioligand binding and second messenger assays. Tissue culture reagents were from Gibco (Eragny, France). Culture flasks (Costar and Falcon) were purchased from Dutscher (Brumath, France) and Becton Dickinson (Le Pont de Claix, France).

Construction of expression vectors. The CB1 coding sequence (Shire et al., 1995) was amplified by polymerase chain reaction with a sense primer bearingHindIII sites and Kozak consensus sequences 5'-CCACACAAGCTTGCCGCCACCATGAAGTCGATCCTAGATGGC and an antisense primer carrying an EcoRI site, 5'-CCACTCGAATTCTCATCACAGAGCCTCGGCAG. The amplicons were digested with HindIII/EcoRI to generate unique cloning sites and inserted into p658, an expression plasmid derived from p7055 (Miloux and Lupker, 1994) in which a polylinker replaced the IL-2 coding sequence. The expression vectors were of 7816 base pairs for CB1. Constructs were all verified by dideoxy sequencing. The expression vector of the CB1 carrying the supplementary 13-amino acid NH2-terminal C-myc (MEQKLISEEDLKL) was generated as described in (Shire et al., 1996).

Expression of human CB1 receptor in CHO cells. Vectors were transfected into CHO dihydrofolate-reductase-negative cells by a modified CaPO₄ precipitation method (Graham and Van der Eb, 1973). CHO cells were treated with trypsin 48 hr after transfection and seeded at a density of 5 × 10⁵ cells/dish onto culture medium containing minimum essential medium-glutamine medium, heat-inactivated dialyzed FCS (10%), gentamicin (20 mg/liter), L-proline (40 mg/liter), pyruvate sodium (0.5 mM) and anti-PPLO agent (1%). After 10 days surviving clones were recovered and cultivated in the same medium, selection consisting in binding assays on isolated membranes (Rinaldi-Carmona et al., 1996). For all assays, cells were seeded in 24-well cluster plates (10⁵ cells/well). Cells were used between the 3rd and the 20th passage.

Cell pretreatment. Cells were incubated at 37°C in culture medium supplemented with 10% FCS (control) or supplemented with 30 nM CP 55,940 or 10 nM SR 141716 for different time periods. Cells were washed thoroughly four times with 1 ml of washing buffer [25 mM HEPES/Tris (pH 7.5), 140 mM NaCl, 5 mM KCl, 1.8 mM CaCl₂, 0.9 mM MgCl₂, 4.5 g/liter glucose and 0.5% acid-free BSA], then tested for binding or cAMP assays. The washing protocol after incubation of the cells with the ligands, was optimized as checked by binding experiments with [³H]-CP 55,940. For MAPK assays, cells were maintained in culture medium containing 0.5% FCS for 24 hr before beginning of the treatment with CP 55,940 or SR 141716. For long-time exposure cells were first exposed to the drug for 24 or 48 hr in the presence of 10% FCS and then for 24 hr in the presence of 0.5% FCS.

Binding experiments. For binding assays, treated and untreated CHO-CB1 cells were washed and incubated at 37°C with [³H]-CP 55,940 (0.1 to 30 nM) or [³H]-SR 141716A (0.3 to 30 nM) in 0.5 ml of binding buffer [25 mM HEPES/Tris (pH 7.5), 140 mM NaCl, 5 mM KCl, 1.8 mM CaCl₂, 0.9 mM MgCl₂, 4.5 g/l glucose and 0.2% acid-free BSA] for 30 min. Rapid aspiration of the assay medium stopped the reaction. Cells were then rinsed (three times) with 1 ml of binding buffer. After addition of 0.1 M NaOH (2 × 0.5 ml), extracts were transferred and the radioactivity bound to the cells was counted with 4 ml of biofluor liquid scintillant. Nonspecific binding was determined in the presence of 1 µM of their respective unlabeled ligand.

Immunolocalization of C-myc-CB1. Immunocytochemical fluorescence labeling of C-myc-CB1 was analyzed by FACS and confocal laser microscopy. For fluorescence activated cell sorter analysis, the CHO-C-myc-CB1 cells were grown for 24 hr in a 6-well plate, before the addition of cannabinoid ligands for various periods of time. At the end of treatment, the cells were washed in PBS and harvested with 5 ml ethylenediamine tetraacetic acid. Washed cells were incubated for 30 min at 4°C with 1 µg/ml of 9E10 monoclonal antibody directed against C-myc epitope (Santa-Cruz). Cells were washed, incubated with 1/200 dilution of fluorescein isothiocyanate (FITC)-conjugated anti-mouse IgG (Sigma Immunochemical) for 30 min at 4°C, then washed and the fluorescence was analyzed by flow cytometry.

cAMP measurements. cAMP measures were carried out as already described (Rinaldi-Carmona et al., 1996). Briefly, cells were incubated for 5 min at 37°C in PBS containing 0.25% acid-free BSA, IBMX (0.1 mM), RO 20-1724 (0.2 mM), with or without CP 55,940. Forskolin was added (3 µM final concentration) and cells were incubated for 20 min at 37°C. The reaction was stopped by rapid aspiration of the assay medium and addition of 1.5 ml of ice-cold 50 mM Tris-HCl, pH 8, 4 mM ethylenediamine tetraacetic acid. Dishes were cooled on ice for 5 min and then the extracts were transferred to a glass tube. Extracts were boiled and then centrifuged for 10 min at 3500 × g to eliminate cellular debris. Aliquots from supernatants were dried and the cAMP concentration was determined by radioimmunoassay.

MAPK activity. MAPK activity was measured as described (Bouaboula et al., 1995a). Briefly, 80% confluence grown cells were maintained in culture medium containing 0.5% FCS for 24 hr before the addition of ligands. Ligand-treated and vehicle-treated CHO-CB1 cells were washed, then incubated at 37°C with or without CP 55,940 or SR 141716 for 10 min. Cells were then washed at 4°C with 0.5 ml of buffer A [50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM ethyleneglycol-bis-(-aminoethyl ether) N,N,N',N-tetraacetic acid, 1 mM Na₃ VO₄] and lysed for 15 min in buffer A supplemented with 1% triton X-100, 10 µg/ml aprotinin, 10 µg/ml, leupeptin, 1 mM dithiothreitol and 1 mM phenylmethylsulfonyl fluoride. Solubilized cell extracts were then clarified by centrifugation at 14,000 × g for 15 min at 4°C. Aliquots (15 µl) were removed and stored at 80°C until use. Phosphorylation assays were carried out at 30°C for 30 min (linear assay conditions) with ³³P ATP using the Biotrack p42/p44 MAPK enzyme system. The radioactivity incorporated was determined by liquid scintillation counting.

Western-blot analysis. After stimulation, cells were washed and lysed in Laemmli's loading buffer containing 6 M urea. Proteins were run on sodium dodecyl sulfate/polyacrylamide gel electrophoresis and blotted onto nitrocellulose filters. Nonspecific binding of antibody was prevented by incubating filters in 5% dried milk powder in buffer A (10 mM Tris-HCl pH 7.6, 150 mM NaCl, 0.05% Tween 20). Phosphorylated MAPK isoforms (P-p42 and P-p44) were immunostained for 3 hr at room temperature using purified anti phospho Erk1/phospho Erk2 antiserum (1 µg/ml) (Biolabs, Beverly, MA) in buffer A saturated with 5% BSA. After extensive washings, the blot was subsequently incubated for 45 min at room temperature with a peroxidase-labeled anti-rabbit-IgG antibody. After washing, immuno-stained MAPKs were visualized using the enhanced chemiluminescence detection system according to the supplier's instructions and subjected to autoradiography.

Data analysis. EC₅₀, IC₅₀, K_d, B_max, K_i and Hill coefficient (nH) were analyzed using a nonlinear curve fitting (Marquardt-Levenberg least-squares method) on a Compaq Desk Pro 4/66i computer. Experiments were all performed in duplicate (binding assays) or in triplicate (cAMP and MAPK activity measures).

	Results

Top Abstract Introduction Procedures Results Discussion References

Binding and functional properties of CP 55,940 and SR 141716 in CHO-CB1 cells. Specific binding of [³H]-CP 55,940 and [³H]-SR 141716A performed on whole CHO-CB1 cells at 37°C was time-dependent and reached a steady state within a 20-min incubation (data not shown). Binding of these two radioligands was concentration dependent, and the nonspecific binding, measured in the presence of 1 µM of their respective unlabeled ligand, was linearly dependent on the concentration of the radioligand used (10-20 and 45-50% of the total binding at 0.1 and 30 nM, respectively). The specific binding was saturable and reached a maximum of 10 to 15 nM for both ligands (fig. 1, A and B). The nonlinear regression analysis of the saturation curves revealed the presence of one class of binding sites exhibiting high affinity for both ligands. An apparent equilibrium dissociation constant (K_d) value of 5.47 ± 0.28 nM (13 experiments) and 4.84 ± 0.66 nM (10 experiments) was found for [³H]-CP 55,940 and [³H]-SR 141716A, respectively. The total binding site number (B_max) varied from 1.5 to 3.5 × 10⁶ receptors/cell. The specific [³H]-CP 55,940 binding on whole CHO-CB1 cells was displaced in a concentration-dependent manner by unlabeled CP 55,940 (K_i = 16.9 ± 1.1 nM; nH = 1.06 ± 0.11, 7 experiments) and SR 141716 (K_i = 11.5 ± 1.3 nM; nH = 1.13 ± 0.15, 14 experiments). Similar results were obtained for [³H]-SR 141716A (SR 141716, K_i = 9.0 ± 1.1 nM; nH = 1.15 ± 0.12, 5 experiments, CP 55,940, K_i = 39.2 ± 5.6 nM; nH = 0.80 ± 0.01, 6 experiments).

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Equilibrium binding of [3H]-CP 55,940 or [3H]-SR 141716A to CHO-CB1 cells. Confluent cells (105 cells/well) were incubated for 30 min at 37°C with increasing concentrations of [3H]-CP 55,940 (A) or [3H]-SR 141716A (B) as described in "Experimental Procedure." Specific binding was defined as the amount of radioligand bound in the presence of 1 µM of their respective unlabeled ligand substract from total radioligand bound at each concentration. Inset, Represents Scatchard plots of [3H]-CP 55,940 () or [3H]-SR 141716A () specific binding. Data are from one out of ten to thirteen experiments performed in duplicate.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Inverse agonist properties of SR 141716. A, Effect of SR 141716 on forskolin-stimulated cAMP production. CHO-CB1 cells (105 cells/well) were pretreated with increasing concentrations of SR 141716 for 10 min before a 20-min stimulation with forskolin (3 µM) at 37°C. cAMP levels were measured as described in "Experimental Procedure." The data are expressed as the percentage of values from cells stimulated with forskolin alone (control cells, 100%). Data represent mean values ± S.E.M. of one determination performed in triplicate. The basal level of cAMP in control cells was 2.5 ± 1.9 pmol of cAMP/well. Forskolin-stimulated cAMP levels in control cells were 50 ± 2.4 pmol of cAMP/well. B, Effect of SR 141716 on basal MAPK activity in CHO-CB1 cells. Growth-arrested cells (105 cells/well) were treated for 10 min with increasing concentrations of SR 141716 at 37°C. The mitogen-activated protein kinase activity was measured in cell lysates as described in "Experimental Procedure." Results are expressed as percentage of basal MAPK activity in the vehicle-treated cells (100%). The basal activity of the MAP kinase in control cells was 2400 ± 210 cpm/well. Data represent mean values ± S.E.M. of one determination performed in triplicate.

Effect of receptor ligands on modulation of CB1 expression. We used two approaches to examine the effects of CP 55,940 or SR 141716 exposure on CB1 expression: by measuring the ability of CHO-CB1 cells to bind either [³H]-CP 55,940 or [³H]-SR 141716 and by analyzing CB1 cell surface expression. The effects of treatment by CP 55,940 and SR 141716 on binding parameters (K_d, B_max and K_i) of [³H]-CP 55,940 and [³H]-SR 141716A were evaluated after exposure of confluent CHO-CB1 cells to vehicle, CP 55,940 or SR 141716. To guarantee a complete removal of all drugs, we limited to 30 and 10 nM the maximum agonist (CP 55,940) or inverse agonist (SR 141716) concentration, respectively, and we optimized the washing protocol after incubation of the cells with the ligands, as checked by binding experiments with [³H]-CP 55,940. Figure 3 shows that no significant differences (P < .05) in the specific binding, measured with [³H]-CP 55,940 or [³H]-SR 141716A, were observed in CHO-CB1 cells exposed to either CP 55,940 or SR 141716 for different periods ranging from 30 min to 72 hr, compared to untreated control cells. The results obtained for the K_d and B_max of [³H]-CP 55,940 and [³H]-SR 141716A, after short or long exposure with CP 55,940 or SR 141716 were reported in table 1. The K_d values were unchanged after treatment of CHO-CB1 cells with CP 55,940 or SR 141716. Similarly, the K_i value and the nH values of CP 55,940, measured at 0.4 nM [³H]-CP 55,940 in washed CHO-CB1 cells, were unchanged after 1 hr exposure with unlabeled CP 55,940 or SR 141716 (CP 55,940-treated cells, K_i of CP 55,940 = 10.41 ± 0.81 nM, nH = 0.85 ± 0.02; SR 141716-treated cells, K_i of CP 55,940 = 14.98 ± 1.50 nM, nH = 0.92 ± 0.08 vs. control cells, K_i of CP 55,940 = 13.20 ± 0.51 nM, nH = 1.07 ± 0.03, three experiments).

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Effect of pretreatment with CP 55,940 or SR 141716 on binding of [3H]-CP 55,940 and [3H]-SR 141716A to CHO-CB1 cells. Confluent cells (105 cells/well) were incubated with 30 nM CP 55,940 () or 10 nM SR 141716 () at 37°C as described in "Experimental Procedure." After the indicated periods of time, cells were washed with washing buffer and 10 nM of [3H]-CP 55,940 (A) or [3H]-SR 141716A (B) was added for 30 min at 37°C. Specific binding was defined as the amount of radioligand bound in the presence of 1 µM of their respective unlabeled ligand substract from total radioligand bound at 10 nM concentration. The data are expressed as the mean ± S.D. from two independent determinations performed in triplicate. Results are expressed as the percentage of the binding measured in untreated cells taken as 100%.

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Effect of CP 55,940 or SR 141716 on CB1 cell surface expression. A, Kinetics of receptor modulation. CHO C-myc-CB1 were incubated in the absence () or presence of 30 nM CP 55,940 (...) or 50 nM SR 141716 (- -) at 37°C for the time periods indicated. Immunofluorescence staining of the receptor with monoclonal antibody against the epitope C-myc and analyzed by flow cytometry is described in "Experimental Procedure." The units on the x-axis were arbitrary unit of fluorescence. B, Confocal immunofluorescence microscopy of epitope-tagged CB1 expressed in CHO. CHO-C-myc-CB1 were incubated in the absence (A) or presence of 30 nM CP 55,940 (B) or 50 nM SR 141716 (C). Cells were imaged by confocal fluorescence microscopy.

Effect of agonist or inverse agonist exposure on cAMP accumulation. We examined the effect of a permanent presence of agonist or inverse agonist on the coupling efficiency of CB1 to AC. First of all, we checked whether such a treatment per se affected the forskolin-induced cAMP production. Confluent CHO-CB1 cells were exposed to either vehicle, CP 55,940 or SR 141716 for variable periods of time ranging from 15 min to 72 hr. After treatment, the cells were extensively washed, and forskolin-stimulated adenylyl cyclase activity was determined. Pretreatment of CHO-CB1 cells with CP 55,940 led to a significant decrease in the ability of the adenylyl cyclase to be stimulated by forskolin. This effect which reflected the agonist effect of CP 55,940, began at 15 min (54% of inhibition), leveled off within 1 hr (69% of inhibition) and lasted at least 3 hr (33% of inhibition). After 6 hr, the stimulation exerted by forskolin was equivalent to that reached in untreated cells. In contrast, in SR 141716-treated CHO-CB1 cells a marked increase in the level of activation of the forskolin-stimulated adenylyl cyclase activity was observed in agreement with the blockage of autoactivated CB1 receptors by SR 141716. This effect was transient and was observed within the first 30 min of incubation with SR 141716 (56% of stimulation). Thereafter, the level of activation of forskolin was similar to that obtained for untreated cells.

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Effect of pretreatment with CP 55,940 and SR 141716 on cAMP production. A, Effect on the ability of 108 M CP 55,940 to inhibit the forskolin-stimulated cAMP levels in CHO-CB1 cells. Confluent cells (105 cells/well) were pretreated with vehicle (), 30 nM CP 55,940 () or 10 nM SR 141716 () at 37°C for 15 min to 72 hr, cells were washed with washing buffer and the effect of 10 nM CP 55,940 on cAMP levels stimulated by 3 µM forskolin was measured as described in "Experimental Procedure." Data are in pmol of cAMP/well. Forskolin-stimulated cAMP levels at 3 µM were 63 ± 4.1 pmol of cAMP/well in control cells. Data represent mean values ± S.D. from two independent determinations performed in triplicate. Statistical significance of difference between vehicle and CP 55,940-treated cells was determined by use of Student's t test and *P < .05. B, Schedule of the experiment.

View larger version (27K): [in this window] [in a new window]   Fig. 6.   Concentration-dependent effects of pretreatment with CP 55,940 and SR 141716 on CP 55,940-modulated cAMP accumulation in CHO-CB1 cells. Confluent cells (105 cells/well) were pretreated with vehicle (), 0.3 (), 1 (), 3 (), 10 () or 30 () nM of CP 55,940 or 10 () nM of SR 141716 at 37°C and after 1 hr, cells were washed with washing buffer. The effect of CP 55,940 (1011-108 M) on cAMP levels stimulated by 3 µM forskolin was then measured as described in "Experimental Procedure." Data are expressed as the percentage of forskolin effect at 3 µM in their respective control cells (100%). Data represent mean values ± S.D. from two independent determinations performed in triplicate.

Effect of agonist or inverse agonist treatment on MAP kinase activation. We also studied the effect of sustained CP 55,940 or SR 141716 treatment of CHO-CB1 on MAPK activity. We first examined the effect of such a treatment on the basal level of MAPK. Growth arrested CHO-CB1 cells were pretreated for different periods ranging from 30 min to 72 hr in the presence of vehicle, CP 55,940 or SR 141716. After treatment, cells were extensively washed, and the basal MAPK activity was determined. As shown in figure 7A, pretreatment of CHO-CB1 cells with 30 nM CP 55,940 causes a significant 2-fold enhancement of the basal activity of the enzyme. This effect was maximal at 1 hr and lasted 3 hr. This reflects the transient activation of MAPK by the cannabinoid agonist. In contrast, the treatment of the cells by 10 nM of SR 141716 had no significant effect on the basal activity of the enzyme. Then we assessed the effect of agonist on receptor responsiveness to stimulate MAPK on cells that have been pretreated with CP 55,940 or SR 141716. As reported in table 3, the cannabinoid receptor agonist CP 55,940 is able to increase the MAP kinase activity by about 3-fold in untreated cells with an EC₅₀ value varying from 2.3 to 4.7 nM. Exposure of the cells to CP 55,940 resulted in a marked decrease (P < .05) in CP 55,940 efficiency to stimulate the MAP kinase activity (no stimulation at 10⁷ M). This effect was time-dependent with the addition of the agonist, and maximal 1 hr after the beginning of the experiment and maintained unchanged after up to 72 hr. By contrast incubation of cells with SR 141716 led to sensitization of CB1-coupled MAPK as indicated by an increase in maximal activities. This effect which was maximal 1 hr after the treatment by SR 141716, was maintained up to 72 hr. This result suggested an enhancement of CB1 receptor response by the agonist (table 3). As shown in figure 7 changes in CP 55,940 or SR 141716-modulated MAPK activation in CHO-CB1 cells were concentration dependent, with a maximal effect measured at 30 nM for CP 55,940 (fig. 7B) and 10 nM for SR 141716 (fig. 7C). The activated forms of MAPKs were phosphorylated on both Tyr and Thr residues (Anderson et al., 1990) and were readily detectable on immunoblotting using an antiserum directed against these phosphorylated forms of MAPKs. Figure 7D shows that CP 55,940 pretreatment induced a complete desensitization of CB1 receptors whereas the pretreatment with SR 1417176 induced an up-regulation of CP 55,940 induced phosphorylated p42 and p44 MAPKs.

View larger version (34K): [in this window] [in a new window]   Fig. 7.   Effect of pretreatment with CP 55,940 and SR 141716 on the basal MAPK activity in CHO-CB1 cells. A, Effect on the basal MAPK activity. Growth-arrested cells (105 cells/well) were incubated with 30 nM CP 55,940 () or 10 nM SR 141716 () at 37°C. After the indicated periods of time, cells were washed with washing buffer and the MAPK activity was measured in cell lysates as described in "Experimental Procedure." Results are expressed as percentage of basal MAPK activity in the vehicle-treated cells (100%). Each is expressed as the mean ± S.D. from two independent determinations performed in triplicate. Student's t test was used to determine the level of significance between control and treated cells (*P < .05). B and C, Concentration effects on CP 55,940-modulated MAPK activation in CHO-CB1 cells. Growth-arrested cells (105 cells/well) were pretreated with vehicle (), 0.3 (), 1 (), 3 (), 10 () or 30 () nM of CP 55,940 or 1 (), 3 (), 10 () nM of SR 141716 at 37°C. After 1 hr, cells were washed with washing buffer and the effect of CP 55,940 (3 × 1010-107 M) on the MAPK activity was determined in cell lysates as described in "Experimental Procedure." Data for vehicle and drug conditions are from one of two independent determinations performed in triplicate. D, Western-blot analysis of phosphorylated MAPKs isoforms. Growth-arrested cells (3 × 105 cells/well) were pretreated with vehicle (lanes 1, 2 and 4) or 30 nM (lane 3) of CP 55,940 or 10 nM (lane 5) of SR 141716. After 1 hr, cells were washed with washing buffer (four times) and treated with 10 nM of CP 55,940 (lanes 2, 3 and 5) or 10 nM of SR 141716 (lanes 2 and 4) for 10 min. Reaction was ended by the addition of Leammli's buffer and protein extracts were processed as described in "Experimental Procedure."

	Discussion

Top Abstract Introduction Procedures Results Discussion References

In this report, we studied CB1 receptor regulation induced by the agonist CP 55,940 or the antagonist SR 141716 in stably transfected CHO cells expressing CB1. Direct binding experiments of [³H]-CP 55,940 and [³H]-SR 141716A to CHO cells expressing the human CB1 (CHO-CB1 cells), showed that both ligands bind to one class of specific binding sites (B_max = 1.5 to 3.5 × 10⁶ receptors/cell) in a saturable manner with high affinity (K_d = 5.47 ± 0.28 nM and 4.84 ± 0.66 nM for [³H]-CP 55,940 and [³H]-SR 141716A, respectively). Expression of the cloned receptor CB1 in CHO cells results in agonist-independent activation that is decreased by the antagonist SR 141716 that has negative intrinsic activity and is referred to as an inverse agonist. These conclusions emerged from observation on two independent-signaling pathways: G-mediated MAPK and G_i-mediated AC responses. First, we showed that the basal level of MAPK activity was enhanced in CHO cells after transfection of CB1 receptors (Bouaboula et al., 1997) and this increase was reversed by treatment with SR 141716. Second, we showed that SR 141716 prevented the inhibition of AC mediated by autoactivated CB1 receptors. In addition, we observed that the coupling of the CB1 receptors with the adenylyl cyclase through G_i is more efficient than the coupling with the MAP kinase signaling pathway depending on G.

Exposure to CP 55,940 induced CB1 desensitization. To study receptor localization we used a chimeric protein designed with the CB1 and a peptide derived from C-myc. We verified that the presence of fusion of the N-terminus of the CB1 to C-myc peptide did not alter receptor ligand binding affinity and signal transduction (Rinaldi-Carmona et al., 1996).

Exposure to SR 141716 induced CB1 sensitization. Incubation of cells expressing CB1 with SR 141716 did not affect CB1 receptor density and affinity constant measured by binding assays. Despite this similarity with CP 55,940, treatment with SR 141716 induced a completely different scenario. In CHO cells expressing the epitope-tagged CB1, exposure to inverse agonist causes a substantial alteration in the cellular distribution profile of the receptor by up-regulating the density of cell surface receptors. The rate of externalization of CB1 is rapid with a maximal enhancement by about 30 min, as determined by the increase in mean surface receptor fluorescence intensity measured using flow cytometry. Such an increase in receptor density at the cell surface requires either synthesis of new receptors or translocation of intracellular pools of intact receptors. We favored this latter possibility because treatment with SR 141716 had no significant effect on the total population of CB1. We also showed that exposure of cells to SR 141716 resulted in a rapid and pronounced sensitization of the agonist-induced MAPK activity whereas the ability of the agonist-inhibited AC was not affected. The inverse agonist-induced sensitization was concentration-dependent with an EC₅₀ = 2 nM. This demonstrated that 1) up-regulation of the G protein-coupled receptor can occur in response to binding with inverse agonist and 2) inverse agonist can enhance subsequent responsiveness to agonist stimulation. The explanation why sensitization was only apparent on MAPK and not on AC may be that maximal effect was already reached in control CHO-CB1 cells in which agonist induced 100% inhibition in cAMP accumulation. Another possibility that remains to be explored could be that inverse agonist pretreatment switches the coupling of the receptor to a different G protein leading to a stronger response through the subunit.