Background
The cannabinoid receptors belong to the G protein-coupled receptor (GPCR) superfamily and include at least two receptor types: CB1 and CB2 [
1‐
3]. The systemic administration of endocannabinoids, such as anandamide, the synthetic agonist [(R)-(+)-[2,3-dihydro-5-methyl-3-(4-morpholinylmethyl) pyrrolo [1,2,3-de]-1,4-benzoxazin-6-yl]-1-naphthalenylmethanone] (WIN55,212-2) or the naturally occurring compound Δ
9-tetrahydrocannabinol (THC), produces analgesia in rodent pain models. Unfortunately, long-term administration of agonists leads to a progressive decrease in the cannabinoid-mediated effects, a process referred to as analgesic tolerance [
4,
5], which persists for as long as 14 days [
6]. The tolerance that follows repeated systemic administration of cannabinoids is caused by down-regulation and/or uncoupling of the receptors from the G proteins [
1]. Most cannabinoids induce rapid internalization of their receptors via clathrin-coated pits [
7], and long-term treatment leads to a significant down-regulation of CB1 receptors (CB1Rs). Indeed, an important fraction of the internalized receptors is transported to the lysosomal compartment for degradation, and the interaction between CB1Rs and G protein-associated sorting protein 1 (GASP1) plays an essential role in this process [
8,
9].
In the brain, CB1R is expressed at high levels in neural cells, whereas CB2R exists only at low levels [
10]. Thus, CB1R appears to mediate the supraspinal effects of cannabinoid agonists. CB1R is found in the cerebral cortex, amygdala, hippocampus, basal ganglia, cerebellum, and brain areas involved in descending pain modulation, such as the periaqueductal gray matter (PAG), rostral ventromedial medulla (RVM), and the spinal cord [
10]. This distribution is consistent with the effects of cannabinoids on emotional responses, cognition, memory, movement, and nociception [
11‐
13]. At the molecular level, CB1R couples to pertussis toxin (PTX)-sensitive Gi/o proteins [
14,
15] and to certain pertussis toxin-insensitive G proteins, probably Gq/11 and Gz [
16,
17]. This receptor regulates the expression of immediate early genes and various cellular effectors, such as adenylyl cyclase, ion channels, mitogen-activated protein kinase, and focal adhesion kinase [
1,
3,
18].
Interestingly, cannabinoids may be useful in controlling pathological pain that is resistant to conventional opioid therapies [
19,
20]. Although cannabinoids act independently of opioids to produce analgesia in rodents, similar brainstem circuitry seems to be involved [
21]. Brain areas such as the caudate putamen, dorsal hippocampus, and substantia nigra are rich in both cannabinoid and opioid receptors, and the co-localization of both types of receptors has been described [
10,
22]. Loss of functional receptors leads to desensitization in both systems. Chronic treatment induces analgesic cross-tolerance between opioids and cannabinoids. This cross-tolerance, however, occurs without any change in the receptors of the other system [
23,
24]. This observation suggests that interactions can take place at the signal transduction/effector level. In nervous tissue, desensitization in response to single doses of opioids, such as morphine, occurs without the loss of surface receptors. However, in response to subsequent doses of morphine, the MOR becomes phosphorylated and undergoes internalization/recycling [
25]. In contrast to what happens with the CB1R, few internalized MORs are destroyed in the lysosomal fraction [
26,
27]. Rather, most of the internalized MORs are re-inserted in the membrane, and tolerance to opioids that induce robust receptor internalization develops slowly [
28,
29].
A series of recent studies have increased our knowledge of the molecular mechanisms implicated in neural MOR signaling and desensitization. In nervous tissue, the C-terminus of MOR interacts with a signaling module consisting of histidine triad nucleotide binding protein 1 (HINT1) that is associated with the Gz regulatory proteins RGS17 (RGSZ2) and RGS20 (RGSZ1). The RGSZ proteins are specific GTPase-activating proteins (GAPs) of receptor-activated Gαz-GTP subunits. Thus, this HINT1-RGSZ signaling module is physically close to the MOR-regulated Gz proteins and helps to deactivate the agonist-activated Gαz-GTP subunits [
30‐
32]. For opioids such as morphine, which induce little MOR internalization, neural-specific Gz proteins are essential to the desensitization of supraspinal antinociception.
Given the relevance of brain CB1Rs as analgesics, in the present study we sought to analyze supraspinal antinociception and its desensitization following icv administration of cannabinoid agonists. The results indicate that CB1Rs, like MORs [
32], interact with the HINT1-RGSZ signaling module. Both receptors regulate Gz proteins to produce supraspinal analgesia, and this G protein is implicated in their desensitization and cross-desensitization.
Discussion
This study has shown that acute icv administration of cannabinoids promotes a long-lasting and robust desensitization of supraspinal CB1Rs, which could be mediated by post-receptor events. The analgesic tolerance that is observed after ip or sc administration of these substances usually requires repeated injections and affects both CB1 and CB2 receptors at spinal and supraspinal levels [
10,
61]. This long-term cannabinoid administration produces CB1R desensitization and down-regulation [
6]. Specifically, supraspinal CB1R expression diminishes, as does the ability of systemic cannabinoids to induce hypoactivity, hypothermia, and antinociception [
5,
8,
9]. Up to 2 weeks are required to recover the initial levels of CB1Rs in the synaptosomal membrane as well as the analgesic response to cannabinoids [
5,
6,
9,
62]. Therefore, the analgesic tolerance that follows the repeated systemic administration of cannabinoids can be explained in terms of the loss of surface CB receptors.
Brain CB1Rs, however, desensitize in response to acute doses of agonists; this cannot be explained merely in terms of a permanent loss of receptors. The effect of a single icv-injection of ACEA or WIN55,212-2 on surface CB1Rs is certainly brief. During the analgesic time-course of these agonists, the CB1Rs decreased in the PAG membrane by 60–70%. Most of the internalized CB1Rs bind to GASP1 and are then degraded in the lysosomal compartment [
8,
9]. After the analgesic effects of single doses of the cannabinoids cease, the CB1Rs are gradually restored to the surface, probably by both the recycling of a portion of the internalized receptors and the insertion of newly synthesized receptors. As a result, 24 h or 48 h later, the presence of CB1Rs in the membrane is comparable to that seen before the agonist challenge. During this time, the restored CB1Rs become coupled to G proteins, but the analgesic response takes a significantly longer time to be restored: about 14 days. Most relevant, this tolerance is also promoted by cannabinoids such as methanandamide, which cause almost no loss of surface CB1Rs. Thus, it is likely that the analgesic desensitization promoted after several days of systemic treatment with cannabinoids primarily affects receptors at the spinal and peripheral levels, and the associated downregulation of the supraspinal CB1Rs, about 30–50% [
5,
62], may be secondary to these effects.
There is compelling evidence that the CB1R couples to and regulates both PTX-sensitive Gi/o proteins and PTX-insensitive Gq/11 and Gz proteins. Thus, the endocannabinoid, 2-arachidonoylglycerol, protects neurons by limiting cyclooxygenase-2 expression, an effect mediated by PTX-sensitive G proteins [
15]. WIN55,212-2 shows a more complex pattern of receptor activation. Whereas this agonist affects acetylcholine release in the hippocampus through a PTX-sensitive mechanism [
63], in cultured hippocampal neurons it promotes increases in intracellular calcium via CB1Rs and the PTX-insensitive Gq protein. Interestingly, the latter effect is not reproduced by other cannabinoids, such as THC, CP55,940, 2-arachidonoylglycerol, or methanandamide [
16]. These results indicate that after binding the CB1R, cannabinoids may determine the class(es) of G proteins to be activated. Indeed, in a cell line derived from human trabecular meshwork, which is an ocular tissue, WIN55,212-2 was shown to increase intracellular calcium via CB1R and Gq/11 proteins and to increase ERK1/2 phosphorylation via PTX-sensitive Gi/o proteins. In this system, CP55,940 and methanandamide produced the same effects, but they acted via PTX-sensitive Gi/o proteins [
17]. Therefore, the CB1R, like the MOR, couples to a series of PTX-sensitive and -insensitive G proteins, and the agonists determine the pattern of G protein activation [
56,
58,
64].
Brain CB1Rs mediate the production of analgesia via PTX-sensitive and PTX-insensitive G proteins [[
53] and present study]. The spinal-mediated analgesic action of cannabinoids is mostly mediated via Gi/o proteins. Intrathecal administration of PTX also abolishes the analgesia evoked by icv cannabinoids, indicating that the descending pathways triggered by these substances act at the spinal level through receptors coupled to Gi/o proteins [
65]. Signaling via the neural-specific PTX-insensitive Gz protein appears to occur more at the supraspinal level [
55,
66]. In fact, supraspinal analgesia mediated by MORs has an important Gz component [
56]; at the spinal level, in contrast, PTX abolishes most MOR-mediated analgesia [
66,
67]. Consistent with this observation, the levels of expression of specific regulators of activated Gαz subunits, GAPs, RGSZ1, and RGSZ2, are lower in the spinal cord than in the midbrain [
47,
54].
Activation of Gz proteins mediates long-lasting analgesic desensitization of supraspinal CB1Rs. Cannabinoid agonists, such as methanandamide, which apparently do not activate Gq/11 proteins [
16,
17], produced desensitization of CB1Rs via activation of Gz proteins. Therefore, it seems that agonists that trigger activation of Gi/o proteins via CB1Rs also activate the PTX-insensitive Gz protein. The unique biochemical and regulatory properties of Gαz subunits account for their strong ability to desensitize GPCR signaling events. The Gz transducer protein, like Gi/o proteins, regulates adenylyl cyclase activity and the gating of certain K
+ channels. Gαz, however, is predominantly confined to neuronal cells. The rate of Gαz-GTP hydrolysis is as much as 200-fold slower than that of Gαs-GTP and Gαi-GTP. Therefore, Gz may be resistant to inactivation after receptor activation unless external factors accelerate the rate of Gαz-GTP hydrolysis, much the same way that the GAPs do for many Ras-like proteins. Therefore, inadequate control of Gαz signaling may easily lead to over-regulation of target effectors and subsequent desensitization [
47]. Thus, deactivation requires the assistance of specific GAPs to augment the rate of hydrolysis and thus release effector(s) from continuous regulation.
The RGS-Rz subfamily bears the primary responsibility for regulating Gz, and the C-terminus of the MOR associates with a signaling module consisting of HINT1-RGSZ, which helps deactivate MOR-activated Gαz-GTP subunits [
30]. This study has shown that brain CB1Rs regulate Gz proteins and associate with the HINT1-RGSZ signaling module, which is involved in the zinc-mediated recruitment of PKCγ [
32]. Indeed, PKC has been implicated in the desensitization of CB1Rs by phosphorylation of a serine residue (S317) in the third internal loop [
68]. We have observed the in vivo recruitment of PKCγ toward the HINT1-RGSZ module at the C-terminus of CB1R during the intervals when the receptor is uncoupled from regulated transduction. However, the inhibition of this kinase did not prevent the development of acute tolerance, suggesting that other post-receptor mechanisms operate in this process. Most relevant, deregulation of this module brings about increased Gz signaling at the corresponding effector(s) and the development of profound analgesic desensitization of brain MORs [
47,
50,
54] and CB1Rs (present study). In contrast, depletion of Gz proteins reduces the analgesic desensitization produced by icv injection of various doses of morphine [
50] and also abolishes acute desensitization of brain CB1Rs and cross-tolerance with morphine. A single icv injection of morphine produces desensitization that lasts for approximately 3 days; however, the cannabinoid agonists studied here desensitized CB1Rs for more than 14 days. Because both of these effects were mediated by the activation of Gz proteins, this observation indicates that CB1R-activated Gz proteins are controlled less efficiently than those activated by the MOR. In agreement with this idea, disruption of the HINT1-RGSZ module led acutely administered morphine to promote a profound and long-lasting desensitization of brain MORs, and most relevant, impaired the analgesic activity of CB1R agonists. Therefore, disruption of Gz regulation brings about a bidirectional supraspinal cross-tolerance between acute doses of morphine and cannabinoids, similar to that attained through chronic and systemic administration of the respective MOR or CB1R agonists [
23,
24].
Morphine poorly internalizes MORs and promotes strong analgesic desensitization by stimulating permanent transfer of a part of the receptor-regulated G proteins toward RGS proteins belonging to the R7 and Rz subfamilies [
25,
47,
48,
50]. In contrast, DAMGO produces a robust internalization and recycling of MORs, a transient transfer of G proteins toward the RGS proteins, and a low level of analgesic tolerance [
25]. Because icv-injected cannabinoids facilitated a reversible transfer of Gα subunits toward RGSZ2 proteins and because the membrane levels of CB1Rs were almost restored within 24 h of their initial challenge, one should expect the resensitization of the analgesic response to these substances, as with DAMGO. However, our results reveal a long-term supraspinal analgesic tolerance even after the CB1R reassociates with these G proteins. This apparent divergence between DAMGO and cannabinoids in the production of tolerance can be explained in terms of the classes of G proteins activated by these agonists after binding to their respective supraspinal receptors. Thus, the analgesic effects of WIN55,212-2 are mediated mostly by Gz proteins, whereas those of DAMGO require Gi/o proteins and, to a lesser extent, Gz [
51,
56,
58,
64]. In the absence of Gz activation, the cannabinoids behave as DAMGO, promoting low levels of analgesic tolerance. Therefore, the desensitizing capacity of Gz proteins on post-receptor events predominates over the resensitization caused by reinsertion and G protein-coupling of the internalized CB1Rs in the neural membrane. Thus, it is the coincidence of WIN55,212-2 and morphine at Gz proteins that accounts for their cross-desensitization, whereas the poor regulation of this Gz by DAMGO explains why WIN55,212-2 fails to impair DAMGO-evoked analgesia.
Thus, an inefficient Gαz-GTP deactivation results in desensitization of brain MORs and CB1Rs, suggesting a post-receptor mechanism that appears to be regulated by Gz proteins. However, at the spinal level CB1Rs primarily regulate Gi/o proteins, and in the absence of Gz proteins, tolerance is primarily achieved by reducing the density of active surface receptors. Indeed, this is seen after repeated systemic treatment with cannabinoids (see Introduction).
Methods
Drugs and production of antibodies
Arachidonyl-2¡-chloroethylamide (ACEA, Tocris #1319), WIN55,212-2 mesylate (Tocris #1038), N-(piperidin-1-yl)-5-(4-iodophenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide (AM251, Tocris #1117), and Δ9-tetrahydrocannabinol (THC, Pharm, Frankfurt, Germany) were dissolved in 1:1:18 (v/v/v) mixture of ethanol: cremophor EL (Sigma Chemical Co., Madrid): physiologic saline. (R)-(+)-Methanandamide in Tocrisolve™ 100 (Tocris #1782). [D-Ala2, N-Me-Phe4, Gly-ol5]-enkephalin (DAMGO, Tocris #1171), clonidine hydrochloride (Tories #0690), [D-Ala2, D-Leu5]-encephalin acetate salt (DADLE, Sigma #E7131), and morphine sulphate (Merck, Darmstadt Germany) were prepared in saline. Pertussis toxin (#516562) and Gö7874 (#365252) were purchased from Calbiochem. The antibodies against CB1R used in this study were produced in New Zealand white rabbits (Sigma Genosys). The antiserum CB1-Nt was raised against amino acid residues 53–66 of the receptor (FRGSPFQEKMTAGD), and the antiserum CB-1EL was raised against residues 177–188 of the murine CB1 receptor (DFHVFHRKDSPN; accession code NP_031752). Anti-CB1R IgGs were purified by affinity chromatography using these synthetic peptides coupled to NHS-activated Sepharose 4 Fast Flow (#17-0906-01, GE Healthcare Biosciences) and labeled with biotin (Sigma #B1022) according to the manufacturer's instructions.
Reduction of G protein and RGS protein expression
To interfere with the expression of the proteins of interest we used synthetic end-capped phosphorothioate antisense oligodeoxynucleotides (ODNs) which have previously been characterized. These were synthesized by Sigma-Genosys Ltd. (Cambridge, UK). In the following sequences, the nucleotides containing the phosphorothioate linkage are marked with an asterisk: 5'-C*T*CGAATCAGTTCG*C*T-3' (16 nt), corresponding to nucleotides 1044–1059 of the murine RGS9-2 mRNA expressed in the CNS (AF125046) [
48,
52]; 5'-C*C*GAAGAGTCTCCTC*T*T-3' (17 nt), corresponding to nucleotides 281–297 of the murine RGSZ2 gene (AF191555) [
47]; 5'-T*G*TAATCTCACCCTTGCTCTCTGCTGGGCCA*G*T (33 nt), corresponding to nucleotides 1033–1065 of the murine Gαz subunit gene (NM_010311) [
49,
50]; 5'-G*T*GGTCAGCCCAGAGCCTCCGGATGACGCCC*G*A (33 nt), corresponding to nucleotides 477–502 of the murine Gαi2 subunit gene (NM_008138) [
49,
50]; 5'-C*C*ATGCGGTTCTCATTGTC*T*G-3' (21 nt), corresponding to nucleotides 725–745 of the Gαq/Gα11 gene sequences (NM_008139/NM_010301) [
51]; 5'-T*T*GAGCCTTGGCAAT*C*T-3' (17 nt), corresponding to nucleotides 11–27 of the murine PKCI/HINT1 gene (NM_008248) [
32]. These sequences showed no homology to any other relevant cloned proteins (GenBank database). Antisense ODN controls consisted of mismatched sequences in which five bases were switched without altering the remaining sequence.
Animals, icv injection, and evaluation of antinociception
Male albino CD-1 mice (Charles River, Barcelona, Spain) weighing 22–25 g were housed and used strictly in accordance with the guidelines of the European Community for the Care and Use of Laboratory Animals (Council Directive 86/609/EEC). Animals were kept at 22°C and were on a 12-h light/dark cycle (lights on from 8 a.m. to 8 p.m.). Food and water were provided
ad libitum. Animals were lightly anaesthetized with ether, and the different substances were injected into the lateral ventricle in a volume of 4 μL as previously described [
49]. The response of the animals to nociceptive stimuli was assessed using the warm water (52°C) tail-flick test. Baseline latencies ranged from 1.5 to 2.2 seconds, and they were not significantly affected by the kinase inhibitor Gö7874, its solvent, or the solvent used for the cannabinoid agonists: Gö7874 in DMSO, 1.7 ± 0.1 seconds; DMSO alone, 1.8 ± 0.12 seconds (n = 10); saline, 1.8 ± 0.2 seconds; and ethanol/cremophor EL/physiologic saline (1:1:18), 1.9 ± 0.2 seconds (n = 10). A cut-off time of 10 seconds was used to minimize the risk of tissue damage. Treatment with the selected active and mismatched ODNs did not alter the baseline latencies. Since the mismatched ODNs produced no changes in cannabinoid/opioid activity compared to saline-treated mice, the results obtained with these ODNs are presented as controls. Antinociception is expressed as a percentage of the maximum possible effect (MPE = 100 × [test latency-baseline latency]/[cut-off time-baseline latency]). Groups of 10–15 mice received a dose of cannabinoid agonist and antinociception was assessed at different time intervals thereafter.
ODN solutions were prepared in the appropriate volume of sterile water immediately prior to use. Animals received either vehicle (control), the mismatched sequence ODN, or the antisense oligo. These compounds were injected into the right lateral ventricle. On days 1 and 2, 1 nmol was injected; on days 3 and 4, 2 nmol was injected; and on day 5, 3 nmol was injected. On day 6, the drugs were injected icv and antinociceptive effects were evaluated using the warm water tail-flick test. This schedule of administration did not alter the normal behavior of the mouse [
69].
Production of acute tolerance
Animals were injected icv into the right lateral ventricle with a dose of cannabinoid agonists sufficient to produce 70–80% of the maximum analgesic effect (priming dose). The controls were given vehicle in the same manner. The development of acute tolerance was monitored once the priming dose was observed to have no effect on baseline latencies. Thus, at 24 h an identical dose of agonist (test dose) was administered icv to all the mice – both the treatment and the control groups. Acute tolerance assays were performed when the compound reached its peak effect after 10 min.
Astroglial cell cultures
Primary mixed glial cultures were prepared from 1-day-old Wistar rat cortex following the procedure described previously [
70]. The cultures were maintained for 12 days in DMEM + 10% FCS in a moist 5% CO
2 atmosphere at 37°C. Enriched astrocytes cultures were obtained after overnight shaking to minimize oligodendrocyte and microglial contamination. For immunoprecipitation studies, astrocytes grown in 75 cm
2 flasks were pooled and homogenized in 10 volumes of 25 mM Tris-HCl (pH 7.4), 1 mM EGTA, and 0.32 M sucrose supplemented with a protease inhibitor cocktail (Sigma #P8340, Madrid, Spain). The homogenate was centrifuged at 1000
g for 10 min (Sorvall RC5C, rotor SS-34, Newton, CT, USA). The supernatant (S1) was removed and centrifuged at 20,000
g for 20 min to obtain the crude membrane pellet and the cytosolic fraction.
For immunocytochemistry, the astrocytes were plated onto poly-D-lysine-coated 10-mm glass coverslips at a density of 20,000 cells/well. After replating, cultures were maintained in DMEM + 10% FCS for 6 h and the serum was reduced to 1% for no longer than 72 h. To label surface CB1 receptors in living astrocytes, cells were incubated with WIN55,212-2 for 1 h at 37°C., after which CB1R antibodies (1:500) labeled with Alexa-488 were added to the cultures. After 30 min the coverslips were rinsed several times with 0.1 M phosphate-buffered saline (PBS) and fixed with 4% paraformaldehyde in PBS for 7 min.
To label internalized CB1 receptors the cells were first incubated with WIN55,212-2 for 1 h at 37°C. Afterwards, the coverslips were rinsed several times with 0.1 M phosphate-buffered saline (PBS) and fixed with 4% paraformaldehyde in PBS for 7 min. This was followed by 45 min incubation with 0.5% NGS, 1% BSA and 0.1% Triton X100. The CB1R antibodies were incubated in this solution for 2 h at room temperature. Cells were observed with a Leica DMIII 6000 CS confocal fluorescence microscope equipped with a TCS SP5 scanning laser. Selectivity of the immunosignal was confirmed by incubating the antibodies with the antigenic peptide.
Preparation of membranes from neural cells and subcellular fractionation
This procedure has been described elsewhere [
25,
48]. Briefly, synaptosomal membranes were obtained from groups of 6 to 10 mice that were sacrificed by decapitation at various intervals after receiving icv injection of the compounds. The PAGs were collected and homogenized in 10 volumes of 25 mM Tris-HCl (pH 7.4), and 0.32 M sucrose supplemented with a phosphatase inhibitor mixture (Sigma, P2850), H89 (Sigma, B1427) and a protease inhibitor cocktail (Sigma, P8340). The homogenate was centrifuged at 1000
g for 10 min to remove the nuclear fraction. The supernatant (S1) was centrifuged twice at 20,000
g for 20 min to obtain the crude synaptosomal pellet (P2). The final pellet was diluted in Tris buffer supplemented with a mixture of protease inhibitors (0.2 mM phenylmethylsulphonyl fluoride, 2 μg/mL leupeptin, and 0.5 μg/mL aprotinin) before aliquoting and freezing at 80°C. The supernatant (S2) was centrifuged at 105,000
g for 1 h to obtain the crude microsomal pellet (P3) (Beckman XL-70 ultracentrifuge, rotor Type 70 Ti). The S3 supernatant was concentrated on an Amicon Ultra-4 centrifugal filter device (nominal molecular weight limit [NMWL] of 10,000; #UFC8 01024, Millipore Iberica S.A., Madrid, Spain), and it was then loaded on a 10–40% (w/v) continuous sucrose gradient and centrifuged at 225,000
g for 18 h. Ten fractions (4 mL each) were collected, the proteins concentrated, and the CB1Rs immunoprecipitated and analyzed by Western blotting.
Glycoprotein purification by wheat germ lectin affinity chromatography
Solubilization and wheat germ lectin (WGL) affinity chromatography were carried out at 4°C on neural membranes resuspended in buffer A with 2% Triton X-100 (20 mM Tris-HCl, pH 7.5, 1 mM EGTA, supplemented with protease inhibitor cocktail). The mixture was incubated at 4°C for 16 h with agitation and then centrifuged at 100,000 g for 1 h. The clear supernatant obtained was applied at a rate of 1.5 mL/min to a WGL-Sepharose 4B column (GE Healthcare Biosciences, #17-0444) previously equilibrated with 20 bed volumes of buffer A containing 1% Triton X100, 1 mM CaCl2, and 1 mM MnCl2 (buffer B). The retained glycoproteins were then eluted with 0.25 M N-acetyl-D-glucosamine in buffer B and were collected in siliconized tubes in 1-mL fractions.
Co-immunoprecipitation of signaling proteins
Samples were sonicated (two cycles of 5 s each) in a volume of 400 μL containing 50 mM Tris-HCl (pH 7.7), 50 mM NaCl, 1% Nonidet P-40, and 50 μL of protease and phosphatase inhibitor mixtures and H89. CB1Rs were immunoprecipitated as described for MORs [
25,
32,
48]. Pilot assays were carried out to optimize the amount of IgGs and sample protein, as well as the period of incubation needed to precipitate the desired protein in a single run. At the end of the procedure, proteins in the soluble fraction were concentrated in centrifugal filter devices (Amicon Microcon YM-10 #42407, Millipore) and solubilized in 2× Laemmli buffer containing mercaptoethanol by heating at 100°C for 3 min. After the samples cooled, proteins were resolved by 10–16% SDS/PAGE.
Detection of signaling proteins in mouse brain: electrophoresis and immunoblotting
Western blots were probed with affinity-purified IgGs: antibodies directed against peptide sequences in the murine CB1R, i.e., CB1-Nt and CB1-1EL (diluted 1:1000); anti-Gαi1, anti-Gαi2, anti-Gαz (1:2000), anti-Gαo and anti-Gαq/11 (diluted 1:1000) [
49,
69]; anti-RGS20(Z1) (1:1000) [
54], anti-RGS17(Z2) [
47], anti-MOR and anti-DOR (1:3000) [
45]; anti-PKCI/HINT1 (1:1000) [
71]; and anti-PKCγ (1:1000; BD Biosciences). The antibodies were diluted in TBS + 0.05% Tween 20 (TTBS) and incubated with the PVDF membranes for 24 h at 6°C. The primary antibodies were detected using the corresponding secondary antibodies conjugated to horseradish peroxidase (diluted 1:10,000 in TTBS). Antibody binding was visualized with Immobilon Western Chemiluminescent HRP substrate (Millipore #WBKLS0100), and the chemiluminescence was recorded with a ChemiImager IS-5500 (Alpha Innotech, San Leandro, California) equipped with a Peltier-cooled CCD camera that provided a real-time readout of 30 frames per second (-35°C; high signal-to-noise ratio; dynamic range of up to 3.4 optical density units). Densitometry was performed using Quantity One Software (BioRad) and expressed as the mean ± S.E. of the integrated volume (average optical density of the pixels within the object area/mm
2). The assays were typically performed two to three times on samples obtained from independent groups of mice (n = 12), and the results were always similar.
[35S]GTPγS Binding Assays
Agonist-stimulated [
35S]GTP
γ S binding was assayed as described previously [
48]. Briefly, synaptosomal membranes from mouse PAG (5
μ g of protein) were incubated for 120 min at 30°C in assay buffer containing 50 mM Tris-HCl (pH 7.4), 100 mM NaCl, 10 mM MgCl
2, 1 mM EDTA, 1 mM dithiothreitol, 10 μM GDP, 0.1% bovine serum albumin, and 0.1 nM [
35S]GTP
γ S, together with varying concentrations of WIN55,212-2 (0.1–10 μM). Nonspecific [
35S]GTP
γ S binding was assessed by carrying out the above reactions in the presence of 20 μM unlabeled GTP
γ S. The incubation was terminated by rapid filtration under vacuum through Whatman GF/B filters, followed by three washes with 3 ml of ice-cold 50 mM Tris-HCl (pH 7.2). Bound radioactivity was determined by liquid scintillation spectrophotometry using a Beckman LS-6500 scintillation counter.
Statistics
All statistical analyses were performed using ANOVA with a Student-Newman-Keuls posthoc test (SigmaStat, SPSS Science Software, Erkrath, Germany), and significance was defined as P < 0.05.
Competing interests
The authors declare that, except for income received from our primary employer "Ministerio de Ciencia y Tecnología", no financial support or compensation has been received from any individual or corporate entity over the past three years for research or professional service and there are no personal financial holdings that could be perceived as constituting a potential conflict of interest.
Authors' contributions
JG conceived the study, participated in its design, and performed the characterization of the CB1R antibodies. MRM and ETM executed the molecular studies, and contributed to the analysis of the data. PSB performed the behavioral studies and assisted with the data analysis and interpretation. JG and PSB wrote and revised the manuscript. All authors have read and approved the final manuscript.