Mutation of putative GRK phosphorylation sites in the cannabinoid receptor 1 (CB1R) confers resistance to cannabinoid tolerance and hypersensitivity to cannabinoids in mice

Abstract

For many G-protein-coupled receptors (GPCRs), including cannabinoid receptor 1 (CB1R), desensitization has been proposed as a principal mechanism driving initial tolerance to agonists. GPCR desensitization typically requires phosphorylation by a G-protein-coupled receptor kinase (GRK) and interaction of the phosphorylated receptor with an arrestin. In simple model systems, CB1R is desensitized by GRK phosphorylation at two serine residues (S426 and S430). However, the role of these serine residues in tolerance and dependence for cannabinoids in vivo was unclear. Therefore, we generated mice where S426 and S430 were mutated to nonphosphorylatable alanines (S426A/S430A). S426A/S430A mutant mice were more sensitive to acutely administered delta-9-tetrahydrocannabinol (Δ(9)-THC), have delayed tolerance to Δ(9)-THC, and showed increased dependence for Δ(9)-THC. S426A/S430A mutants also showed increased responses to elevated levels of endogenous cannabinoids. CB1R desensitization in the periaqueductal gray and spinal cord following 7 d of treatment with Δ(9)-THC was absent in S426A/S430A mutants. Δ(9)-THC-induced downregulation of CB1R in the spinal cord was also absent in S426A/S430A mutants. Cultured autaptic hippocampal neurons from S426A/S430A mice showed enhanced endocannabinoid-mediated depolarization-induced suppression of excitation (DSE) and reduced agonist-mediated desensitization of DSE. These results indicate that S426 and S430 play major roles in the acute response to, tolerance to, and dependence on cannabinoids. Additionally, S426A/S430A mice are a novel model for studying pathophysiological processes thought to involve excessive endocannabinoid signaling such as drug addiction and metabolic disease. These mice also validate the approach of mutating GRK phosphorylation sites involved in desensitization as a general means to confer exaggerated signaling to GPCRs in vivo.

Keywords: CB1; GPCR; THC; cannabinoid; desensitization; tolerance.

Figure 1.

Generation of S426A/S430A knock-in mice. A, Mice expressing a desensitization-resistant form of CB₁R were produced using a targeting vector designed to mutate two putative GRK phosphorylation sites, serines 426 and 430, to nonphosphorylatable alanines. Additionally, the targeting vector introduced an N-terminal HA tag into CB₁R and contained a NeoR gene flanked by FLP recombinase sites (blue triangles). B, Correct integration of the targeting vector was verified in ES cells by Southern blot analysis. A genomic DNA probe located outside of the targeting vector was used to detect a 5.2 kb WT fragment and a 4.2 kb mutant fragment after digestion of ES cell DNA with HindIII. C, PCR analysis of tail or ear DNA was used to determine the genotypes of offspring from heterozygous matings. NcoI digestion of the PCR of the mutant allele (KI) product produced two fragments, while the WT product produced a single band, and heterozygotes produced the predicted three bands. The no template control is shown in the lane labeled “B.”

Figure 2.

CB₁R protein levels are largely unchanged in S426A/S430A mutant brains. A–D, Western blot analysis of CB₁R protein levels in forebrain (A), cerebellum (B), striatum (C), and hippocampus (D) of S426A/S430A mutant, WT, and CB₁R KO mice. In A and B, the top bands are CB₁R as detected by the L15 rabbit polyclonal antibody and bottom bands are β-tubulin (tubulin), used to normalize protein loading of individual samples. Average CB₁R/tubulin density ratios shown for the forebrain (A), cerebellum (B), striatum (C), and hippocampus (D) were analyzed with unpaired t tests (WT, N = 2–5; S426A/S430A, N = 4–5). Also shown in A and B is the absence of CB₁R immunoreactivity in equivalent brain regions from a CB₁R KO mouse, demonstrating the specificity of the L15 antibody.

Figure 3.

Desensitization of CB₁R-mediated G-protein activation is attenuated in S426A/S430A mutants. A–C, Desensitization of CB₁R-mediated G-protein activation was assessed using CP55,940-stimulated [³⁵S]GTPγS binding in membranes prepared from PAG (A), spinal cord (B), or hippocampus (C) from S426A/S430A mutant and WT mice. Data points represent mean net-stimulated [³⁵S]GTPγS binding ± SEM (n = 4–6).

Figure 4.

S426A/S430A hippocampal neurons have enhanced DSE and desensitize more slowly. A, Sample DSE time courses in WT and S426A/S430A autaptic neurons in response to a 1 s depolarization. Inset shows sample EPSCs for S426A/S430A and WT neurons before and after the 1 s depolarization. B, Depolarization response curves in WT (black diamonds, solid line) and S426A/S430A neurons (red circles) show that the response in S426A/S430A neurons is shifted to the left. In addition, S426A/S430A neurons (red triangles, dotted lines) desensitize to a lesser extent after overnight treatment with the CB₁ agonist WIN55,212-2 (100 nm) compared with WT neurons (black squares, dotted line). Brackets on right indicate grouping for statistical comparisons: top (*) compares WT-WIN-treated to S426A/S430A-WIN-pretreated; lower (#) compares WT to S426A/S430A-WIN-pretreated. *p < 0.05, **p < 0.01, #p < 0.05, ##p < 0.01, two-way ANOVA with Bonferroni post hoc test. C, Bar graph shows ED₅₀ for “depolarization response curves” in WT vs S426A/S430A neurons derived from the solid curves in B. The ED₅₀ refers to the duration of depolarization that is required to obtain a 50% maximal DSE inhibition. Values are in seconds with 95% CIs.

Figure 5.

S426A/S430A mutants are more sensitive to the antinociceptive and hypothermic effects of Δ⁹-THC. A, B, S426A/S430A mutants (red bar or red lines with circles) exhibit increased antinociceptive (A) and hypothermic (B) responses to 30 mg/kg Δ⁹-THC relative to WT mice (black bar and black line with squares). C, D, The dose–response curves (1, 10, 30, and 50 mg/kg) for the antinociceptive (C) and hypothermic (D) effects of Δ⁹-THC were shifted to the left for S426A/S430A mutants (red lines and circles) relative to WT littermates (black lines and squares). Sample sizes for each group are in parentheses. Error bars represent the SEM, and data analyses were performed using unpaired Student's t tests (A) or two-way ANOVA with Bonferroni post hoc tests (B–D). *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 6.

S426A/S430A mutants are more sensitive to endocannabinoids. A, B, S426A/S430A mutants (red bars or lines with circles) exhibit increased antinociceptive (A) and hypothermic (B) responses to 50 mg/kg AEA that was administered 30 min after 10 mg/kg URB597 compared with WT mice (black bars or lines with squares). C, D, The dose–response curves for the antinociceptive (C) and hypothermic (D) effects of 10 mg/kg URB597 combined with 1, 3, 10, or 30 mg/kg AEA were shifted to the left in the S426A/S430A mutant mice. E, F, The dose–response curves for the antinociceptive (E) and hypothermic (F) effects of the dual FAAH and MAGL inhibitor JZL195 were also shifted to the left for S426A/S430A mutant mice. Error bars represent the SEM and data analyses were performed using unpaired Student's t test (A) and two-way ANOVA with Bonferroni post hoc tests (B–F). *p < 0.05, **p < 0.01, ****p < 0.0001. Sample sizes for each group are in parentheses.

Figure 7.

The development of tolerance is delayed in S426A/S430A mutants. A, B, S426A/S430A (red circles and line) and WT (black squares and line) mice were treated (intraperitoneal injection) daily with 30 mg/kg Δ⁹-THC for 7 d. A, C, Tail-flick antinociception was measured daily at 55 min after administration of Δ⁹-THC. B, D, Body temperature was measured daily at 60 min following Δ⁹-THC. C, D, S426A/S430A and WT mice were daily with 10 and 30 mg/kg Δ⁹-THC, respectively, and tail-flick antinociception and body temperatures were measured as above. Error bars represent the SEM and data were analyzed by two-way ANOVA and Bonferroni post hoc tests. *p < 0.05, **p < 0.01, ***p < 0.001. Sample sizes for each group are in parentheses.

Figure 8.

Δ⁹-THC dependence is increased in S426A/S430A mutant mice. A–C, Paw tremors (A), jumps (B), and diarrhea (C) were measured in S426A/S430A mutants (red bars) and WT (black bars) littermates to evaluate Δ⁹-THC dependence. Dependence to Δ⁹-THC was induced by 5.5 d of twice daily subcutaneous injections of vehicle (veh) or 50 mg/kg Δ⁹-THC. Withdrawal was precipitated by administering 10 mg/kg rimonabant (SR1). Withdrawal symptoms were scored for the following conditions: Veh/Veh (WT, N = 5; S426A/S430A, N = 8), Veh/SR1 (WT, N = 6; S426A/S430A, N = 8), and SR1/THC (WT, N = 10; S426/S430A; N = 11). *p < 0.05, ***p < 0.001. Error bars represent the SEM, and data were analyzed by two-way ANOVA and Bonferroni post hoc tests.