Differential modulation of AP-1- and CRE-driven transcription by cannabinoid agonists emphasizes functional selectivity at the CB1 receptor

Abstract

Background and purpose: Long-term adaptations to pharmacological stimuli frequently originate from modulation of complex intracellular signalling pathways. We previously reported that HU210 and CP55940, two CB1 cannabinoid receptor agonists, induced opposite effects on TH expression. Herein, we characterized their influence on cAMP response element (CRE) and activator protein 1 (AP-1)-mediated regulation of gene transcription.

Experimental approach: The activity of the agonists was examined on transfected N1E-115 cells in which expression of the luciferase reporter gene was controlled by transcription promoters consisting of repeats of either CRE or AP-1 elements. In addition, the implication of classical signalling pathways was investigated using a variety of kinase inhibitors.

Key results: Consistent with the CB1-mediated reduction of cAMP accumulation, both ligands decreased CRE-driven luciferase expression with similar potencies. HU210 also exhibited a concentration-dependent reduction of luciferase activity in cells engineered to examine AP-1-controlled transcription, whereas such response was not obtained with CP55940. Responses were all inhibited by SR141716A and were modified in Pertussis toxin-treated cells, suggesting agonist-selective regulations of distinct Gi/o-dependent mechanisms through CB1 receptor activation. Finally, PKC inhibitors efficiently inhibited the paradoxical effect of HU210 on AP-1-mediated transcription, indicating selective regulation of PKC-dependent responses.

Conclusions and implications: Together, our results demonstrate that two cannabinoid ligands, commonly used as reference agonists acting on the same receptor with similar affinities, differentially modulate gene transcription through distinct controls of AP-1. This could reflect activation of distinct subsets of Gi/o-proteins, supporting the concept of functional selectivity at CB1 receptors.

Figure 1

Regulation of cAMP accumulation in N1E-115 neuroblastoma cells. Cells were exposed to HU210 or CP55940 (0.1 μM), in the absence (a) or presence (b) of 1 μM forskolin. The responses to both agonists were also measured after overnight treatment of the cells with PTx (100 ng mL⁻¹). Results are expressed as percentage of responses in control cells (exposed to vehicle or forskolin, which caused a sevenfold increase in basal cAMP accumulation). Data are mean values with s.e.mean from three separate experiments performed in triplicate. ^**P<0.01, ^*P<0.05 relative to control condition; ^##P<0.01, ^#P<0.05 relative to responses in the presence of PTx.

Figure 2

Cannabinoid-mediated regulation of CRE-dependent transcriptional activity in N1E-115 neuroblastoma cells. Luciferase activity was measured in transiently transfected N1E-115 cells carrying the pCRE-Luc construct. Cells were exposed to increasing concentrations of HU210 or CP55940 in the absence or presence of 1 μM forskolin (a) and luciferase activity was monitored 5 h later. The responses to both agonists (used at 0.1 μM) were also measured in the presence of 1 μM SR141716A (b). The role of G_i/o-type G proteins was investigated by repeating these measures in the presence of forskolin on cells treated overnight with PTx (100 ng mL⁻¹) (c). Results (mean values with s.e.mean from at least three separate experiments performed in triplicate) are given as the percentages of relative luciferase activity (firefly luciferase relative to Renilla luciferase activity) compared with untreated or forskolin-treated cells. Forskolin increased the basal level of luciferase by 8.5-fold. ^**P<0.01 and ^*P<0.05 denote significant difference as compared with control conditions.

Figure 3

HU210-mediated regulation of AP-1-dependent transcriptional activity in N1E-115 neuroblastoma cells. Luciferase activity was measured in N1E-115 cells transiently transfected with pAP-1-Luc construct. (a) Cells were exposed to increasing concentrations of HU210 or CP55940 and luciferase activity was monitored 5 h later. The responses to both agonists (used at 0.1 μM) were also measured in the presence of 1 μM SR141716A (b). Results shown are mean values with s.e.mean from at least three separate experiments performed in triplicate and are given as the percentages of relative luciferase activity (firefly luciferase relative to Renilla luciferase activity) compared with untreated cells. ^**P<0.01 and ^*P<0.05 denote significant difference as compared with control conditions.

Figure 4

CP55940 competitively antagonizes HU210-induced inhibition of AP-1-dependent luciferase activity. Luciferase activity was measured in N1E-115 cells transiently transfected with pAP-1-Luc construct. The apparent potency of HU210 in decreasing luciferase activity was evaluated in the presence of different concentrations of CP55940. Inset: Schild plot analysis allowing estimation of the pA₂ value of CP55940 at inhibiting the response to HU210. Results shown are mean values with s.e.mean from at least three separate experiments performed in triplicate and are given as the percentage of relative luciferase activity (firefly luciferase relative to Renilla luciferase activity) as compared with untreated cells.

Figure 5

Involvement of G_i/o-type G proteins in the modulation of CRE and AP-1-dependent transcriptional activities by cannabinoid ligands. Luciferase activity was measured in N1E-115 cells transiently transfected with either pCRE (a) or pAP-1-Luc (b) constructs. Cells were treated overnight with 100 ng mL⁻¹ PTx. Thereafter, cells were exposed to increasing concentrations of HU210 or CP55940 and luciferase activity was monitored after 5 h. Results shown are mean values with s.e.mean from at least three separate experiments performed in triplicate and are given as the percentage of relative luciferase activity (firefly luciferase relative to Renilla luciferase activity) compared with control. ^*P<0.05 denotes significant difference as compared with control conditions.

Figure 6

Putative signalling pathways associated with regulation of CRE and AP-1-controlled transcription by the CB₁ receptor. Through stabilization of different active conformations of the receptor, HU210 and CP55940 could selectively promote coupling with several types of G proteins, supporting activation/inhibition of distinct signalling pathways. On the one hand, the high affinity agonists HU210 (striped arrow) and CP55940 (open arrow) mediate inhibition of CRE-dependent transcription through activation of Pertussis toxin-sensitive G proteins and related inhibition of AC and PKA activity, as confirmed by the use of PKA inhibitors. On the other hand, only HU210 regulates AP-1-dependent transcription via an unresolved mechanism involving the activity of PKC, as suggested by the use of appropriate inhibitors. In addition, experimental observations demonstrate the possible interaction of CB₁ receptors with PTx-insensitive G proteins. Although these alternative couplings exert a modest influence with comparison to the response mediated by G_i/o proteins, activation of PTx-insensitive G proteins by both HU210 and CP55940 is thought to equally promote CRE and AP-1-controlled transcription.