The endocannabinoids anandamide and virodhamine modulate the activity of the candidate cannabinoid receptor GPR55

Abstract

The role of cannabinoid receptors in inflammation has been the topic of many research endeavors. Despite this effort, to date the involvement of the endocannabinoid system (ECS) in inflammation remains obscure. The ambiguity of cannabinoid involvement may be explained by the existence of cannabinoid receptors, other than CB(1) and CB(2), or a consequence of interaction of endocannabinoids with other signaling systems. GPR55 has been proposed to be a cannabinoid receptor; however the interaction of the endocannabinoid system with GPR55 remains elusive. Consequently this study set about to examine the effects of the endocannabinoids, anandamide (AEA) and virodhamine, on GPR55 mediated signaling. Specifically, we assessed changes in β-arrestin2 (βarr2) distribution and GPR55 receptor internalization following activation by lysophosphatidylinositol (LPI), the synthetic cannabinoid ligand SR141716A, and new selective synthetic GPR55 agonists. Data obtained from the experiments presented herein demonstrate that AEA and virodhamine modulate agonist-mediated recruitment of βarr2. AEA and virodhamine act as partial agonists; enhancing the agonist effect at low concentrations and inhibiting it at high concentrations. Furthermore, both virodhamine and AEA significantly attenuated agonist-induced internalization of GPR55. These effects are attributed to the expression of GPR55, and not CB(1) and CB(2) receptors, as we have established negligible expression of CB(1) and CB(2) in these GPR55-transfected U2OS cells. The identification of select endocannabinoids as GPR55 modulators will aide in elucidating the function of GPR55 in the ECS.

Fig. 1. AEA inhibits agonist mediated βarr2-GFP distribution in U2OS cells co-expressing GPR55E and βarr2-GFP

(A) (i–iv) Cells were pre-incubated with increasing concentrations of the antagonist (AEA) for 15 min prior to agonist application (3 μM LPI or 30 μM SR141716A). The response was normalized to the averaged response obtained by the agonist. Data are mean ± SEM from three independent experiments performed in duplicate (*P<0.05, **P<0.01, ***P<0.001). (a) AEA inhibition of LPI. (b) AEA inhibition of SR141716A. (c) Representative images, captured with confocal microscopy at 40x magnification depicting βarr2-GFP recruitment pattern are illustrated. Cells treated with vehicle and 30 μM AEA show that βarr2-GFP is homogenously distributed in the cytoplasm. Treatment with 3 μM LPI and 30 μM SR141716A resulted in cytosolic distribution βarr2-GFP which was largely attenuated following treatment with AEA.

Fig. 2. Virodhamine inhibits agonist mediated βarr2-GFP distribution in U2OS cells co-expressing GPR55E and βarr2-GFP

Cells were pre-incubated with increasing concentrations of virodhamine for 15 min prior to agonist application (3 μM LPI or 30 μM SR141716A). The response was normalized to the averaged response obtained by the agonist. Data are mean ± SEM from three independent experiments performed in duplicate (*P<0.05, **P<0.01, ***P<0.001). (a) Virodhamine inhibition of LPI. (b) Virodhamine inhibition of SR141716A. (c) Representative images, captured with confocal microscopy at 40x magnification depicting βarr2-GFP recruitment pattern are illustrated. Cells treated with vehicle and 30 μM virodhamine show that βarr2-GFP is homogenously distributed in the cytoplasm. Treatment with 3 μM LPI and 30 μM SR141716A resulted in cytosolic distribution of β-arr2-GFP which was largely attenuated following treatment with virodhamine.

Fig. 3. AEA and Virodhamine inhibit agonist-mediated GPR55E internalization

U2OS cells expressing HAGPR55E were pre-labeled with anti-HA antibody and Alexa Fluor 568 antibody. Cells were then pre-incubated with virodhamine (3 μM) for 15 min, followed by additional of 40 min incubation along with the agonist. Membrane staining was captured by confocal microscopy at 63x magnification. Upon treatment with vehicle (a) 10 μM AEA (g) and 10 μM virodhamine (m), cells show primarily membrane localization of GPR55. Treatment with 3 μM LPI (b), 30 μM SR141716A (c), 5 μM ML184 (d), 5 μM ML185 (e) and 5 μM ML186 (f), resulted in receptor internalization, evident by loss of plasma membrane receptor staining. Treatment with 10 μM AEA and 10 μM virodhamine largely attenuated receptor internalization induced by 3 μM LPI (h,n), 30 μM SR141716A (i,o), 5 μM ML184 (j,p), 5 μM ML185 (k,q) and 5 μM ML186 (l,r). Data are representative examples of at least three individual experiments.

Fig. 4. Expression levels of CB₁ and CB₂ in U2OS and GPR55E U2OS cells

Levels of CB₁ and CB₂ mRNA expression were assessed by RT-PCR and qRT-PCR. (a) Representative 3% agarose gel showing PCR products for eukaryotic 18S RNA after 30 cycles of RT-PCR. (b) Representative 3% agarose gel showing PCR products for CB₁ and CB₂ after 30 cycles of RT-PCR. (c) Average Ct values for CB₁ and CB₂ from qRT-PCR analysis. Data are means from three independent experiments performed in triplicate. 55E, GPR55E U2OS cells; CB1, HEKCB1-293 cells; CB2, CB2-CHO cells, NTC, no template control.