| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
-lysophosphatidylinositol promotes RhoA-dependent Ca2+ signaling and NFAT activation
,1
,2
* Centre for Neuroscience, Ninewells Hospital and Medical School, University of Dundee, Dundee, UK;
Institute of Experimental and Clinical Pharmacology, Medical University of Graz, Graz, Austria; and
Institute of Medical Sciences, University of Aberdeen, Aberdeen, UK
2 Correspondence: A.J.I., Centre for Neuroscience, Ninewells Hospital and Medical School, University of Dundee, Dundee DD1 9SY, UK. E-mail: a.j.irving{at}dundee.ac.uk; M.W., Institute of Experimental and Clinical Pharmacology, Medical University of Graz, Universitätsplatz 4, A-8010 Graz, Austria. E-mail: maria.waldhoer{at}meduni-graz.at
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-lysophosphatidylinositol (LPI) was recently identified as a novel ligand for the orphan G protein-coupled receptor 55 (GPR55). In this study we define the downstream signaling pathways activated by LPI in a human embryonic kidney (HEK) 293 cell line engineered to stably express recombinant human GPR55. We find that treatment with LPI induces marked GPR55 internalization and stimulates a sustained, oscillatory Ca2+ release pathway, which is dependent on G13 and requires RhoA activation. We then establish that this signaling cascade leads to the efficient activation of NFAT (nuclear factor of activated T cells) family transcription factors and their nuclear translocation. Analysis of cannabinoid ligand activity at GPR55 revealed no clear effect of the endocannabinoids anandamide and 2-arachidonoylglycerol; however, the classical CB1 antagonist AM251 evoked GPR55-mediated Ca2+ signaling. Thus, LPI is a potent and efficacious ligand at GPR55, which is likely to be a key plasma membrane mediator of LPI-mediated signaling events and changes in gene expression.—Henstridge, C. M., Balenga, N. A. B., Ford, L. A., Ross, R. A., Waldhoer, M., Irving, A. J. The GPR55 ligand L--lysophosphatidylinositol promotes RhoA-dependent Ca2+ signaling and NFAT activation.
Key Words: GPCR • cannabinoid • LPI • AM251 • transcription • ROCK
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
that may also be activated by L--lysophosphatidylinositol (LPI) (3)
. GPR55 mRNA is expressed widely within the body, with high levels found in the adrenal glands, spleen, central nervous system (CNS), and gut (1)
, suggesting that, like the CB1 receptor, it may play a role in regulating a wide range of physiological processes. Interestingly, LPI exerts mitogenic effects in differentiated thyroid cells via activation of a novel receptor (4)
and may play a role in the growth of certain tumors (5)
. As endogenous lipid amides and phospholipids are key signaling intermediaries that control diverse aspects of cellular function, the proposed identification of GPR55 as a novel molecular target for these ligands suggests a fundamental role for this receptor in both physiological and pathophysiological processes.
Thus far, a number of inconsistencies have been documented with respect to GPR55 pharmacology and its downstream signaling pathways, despite these studies being performed in the same cellular background (1
2
3)
. A number of groups have reported that GPR55 interacts with cannabinoid ligands (1
, 2)
, including the endocannabinoid anandamide (AEA) (1
, 2
, 6)
; however, this has been contested by another study (3)
, which found no cannabinoid sensitivity and suggested that the natural ligand for GPR55 is LPI. Here, LPI was associated with activation of extracellular regulated kinase (ERK) MAP-kinases and resulted in a modest increase in cytosolic Ca2+ (3)
. Other studies have shown that cannabinoid ligands can increase intracellular Ca2+ levels via GPR55—possibly through the activation of Gq proteins—however, they are not thought to affect the ERK1/2 pathway (6)
. In addition, cannabinoids have been reported to activate G12/G13 proteins and RhoA via this receptor (1
, 6)
. Interestingly, although RhoA and Ca2+ signaling pathways are classically distinct, an interaction at the level of phospholipase C has been suggested (7)
. Thus, we set out to investigate the potential involvement of RhoA, Ca2+, and other downstream signaling partners in LPI-mediated GPR55 activation in a human embryonic kidney (HEK) 293 cell line stably overexpressing recombinant human GPR55. In addition, we evaluated the ability of selected cannabinoid ligands to promote GPR55-mediated Ca2+ signaling.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
13 (Q226L/D294N) were from the cDNA Resource Center, Missouri University of Science and Technology (Rolla, MO, USA). The pNFAT-Luc reporter plasmid was obtained from Stratagene (La Jolla, CA, USA). Nuclear-factor of activated T-cells cytoplasmic, calcineurin-dependent 3 (GFP-NFATc3) was kindly supplied by Dr. Masamitsu Iino (Tokyo University, Tokyo, Japan). AEA, 2-arachidonoyl-glycerol (2-AG), (–)-cis-3-[2-hydroxy-4- (1,1-dimethylheptyl)-phenyl]-trans-4-(3-hydroxypropyl)cyclohexanol (CP55,940), and 1-(2,4-dichlorophenyl)-5-(4-iodophenyl)-4-methyl-N-1-piperidinyl-1H-pyrazole-3-carboxamide (AM251) were obtained from Tocris Cookson (Avonmouth, UK). LPI mixture (58% C16, 42% C18), G418, poly-D-lysine, and all other chemicals were obtained from Sigma-Aldrich (Dorset, UK).
Construction of plasmids
Human GPR55 cDNA was found to have a single amino acid substitution (R124C) with respect to the published sequence (European Molecular Biology Laboratory accession number BC032694). This mutation was corrected using site-directed mutagenesis (QuickChange; Stratagene) and subcloned into a plasmid construct DNA (pcDNA) 3.1 vector. The receptor was tagged with a triple hemagglutinin epitope (HA) at the N terminus (3xHA-GPR55) preceded by an optimized artificial signal sequence from the human growth hormone (HGH; residues 1–33) to ensure efficient surface expression (8)
.
Ca2+ microfluorimetry
A digital epifluorescence imaging system (Perkin Elmer, Emeryville, CA, USA) mounted on an Olympus BX50WI microscope (Olympus, Tokyo, Japan) was used to measure changes in [Ca2+]i. Cells were incubated with the Ca2+-sensitive dye Fura-2-AM (6 µM; 40–60 min at 25°C) in HEPES-buffered saline (HBS; in mM: NaCl 135, HEPES 10, KCl 5, CaCl2 1.8, MgCl2 1.0, D-glucose 25, pH 7.4). Ratiometric images were collected at 5 s intervals with data derived from individual cells. Typically, the effects of ligands were quantified by measuring the mean change in fluorescence ratio. However, in some experiments, intracellular Ca2+ levels were determined following equilibration with Ca2+-free medium containing 1mM EGTA and then perfused with 1.8 mM Ca2+-containing medium in the presence of Triton X-100 (0.01%) to determine Rmin and Rmax values, respectively (9)
. All n values represent data (number of cells) obtained from a minimum of two separate experiments derived from different passages or transfections.
Cell culture, transfection, and stable cell line generation
AD-HEK293 cells (Stratagene) were maintained in Dulbecco’s modified Eagle’s medium DMEM-F12 (Invitrogen) supplemented with 10% calf serum (Invitrogen) at 37°C and 5% CO2. Cell lines stably expressing 3xHA-GPR55 (GPR55-HEK293) were generated by selection with G418 at 800 µg/ml and subsequently maintained at 400 µg/ml. For experimental studies, cells were plated onto poly-D-lysine-coated coverslips (10 µg/ml) and serum starved for 12–48 h. Transient transfections were carried out using Lipofectamine 2000 (Invitrogen).
Western blotting
Confluent 10-cm dishes of GPR55-HEK293 and HEK293 cells were washed twice with ice-cold PBS and lysed on ice in 800 µl of lysis buffer containing 10 mM Tris-HCl, pH 7.4; 150 mM NaCl; 25 mM KCl; 1 mM CaCl2; 0.1% Triton X-100; and protease inhibitors (Roche Applied Science, Indianapolis, IN, USA). After centrifugation, supernatants were collected and resolved on SDS-PAGE. After transfer to polyvinylidene difluoride membrane (Millipore, Bedford, MA, USA), the blot was blocked in TBST buffer (1 mM CaCl2, 136 mM NaCl, 2.5 mM KCl, 25 mM Tris-HCl, 0.1% (v/v) Tween 20) containing 10% nonfat dry milk and incubated with anti-HA antibody (1:1000; Covance Research Products) followed by anti-mouse HRP-conjugated antibody (1:10000; Jackson ImmunoResearch). Bands were visualized by Pierce enhanced chemiluminescence (ECL) Western blotting substrate (Thermo Fisher Scientific, Pittsburgh, PA, USA).
RhoA activation assay
GPR55-HEK293 or HEK293 cells were seeded on 10-cm cell-culture plates, grown to 80% confluency, and serum starved overnight. RhoA was stimulated with 1 µM LPI for 7 or 10 min at 37°C, then washed twice with ice-cold TBS and harvested on ice with 500 µl of 1x MLB lysis buffer (Upstate Biotechnology Inc.). Lysates were then clarified by centrifugation at 14,000 g for 5 min. The cell lysates were incubated with 30 µg of GST-RBD-agarose beads to precipitate GTP-bound RhoA. The beads were pelleted by brief centrifugation (14,000 g for 30 s), and washed 3 times with lysis buffer. Samples were denatured in sample buffer, boiled for 5 min at 90°C, and resolved by SDS-PAGE using a 12% gel. Bound RhoA was detected by Western blot using a monoclonal anti-RhoA antibody (1:200), followed by IRDye 800 secondary antibody (1:10,000).
Immunofluorescence and confocal microscopy
Cells were fixed in 4% paraformaldehyde in HBS for 20 min and permeabilized with 0.3% Triton X-100 for 15 min. A laser-scanning confocal imaging system (Zeiss LSM510) incorporating an upright Axioskop FS2 microscope (Carl Zeiss, Oberkochen, Germany) was used for image acquisition and processing. HA-GPR55 was detected using rabbit anti-HA antibody (Abcam) and visualized using a Cy3-(ImmunoResearch Laboratories) or Alexa Fluor 488-conjugated secondary antibody (Invitrogen). In some experiments, full Z-stack images were generated using a series of 2.5-µm sections. Agonist-induced trafficking studies were carried out in HBS at 37°C. Following agonist exposure, cells were immediately fixed and processed for HA immunolabeling. Live imaging experiments with GFP-NFATc3 were carried out in HBS at 30°C, with laser exposure limited in order to minimize photobleaching and phototoxicity. Single plane images were obtained every 2–5 min.
Antibody feeding experiments
Cells were plated onto glass coverslips and grown in serum-free medium for 16 h. Living cells were fed monoclonal anti-HA.11 antibody (Covance Research Products) for 30 min to label receptors. Subsequently, cells were treated with ligand for the time points indicated, fixed in 3% paraformaldehyde, and permeabilized with 0.1% Triton X-100, essentially as described elsewhere (10
, 11)
. Receptors were labeled with Alexa Fluor 488-conjugated goat anti-mouse IgG1a antibody (Invitrogen). Following staining, cells were mounted in Vectashield Mounting Medium (Vector Laboratories Inc., Burlingame, CA, USA) and analyzed using a Zeiss LSM510 META Axioplan confocal microscope.
NFAT reporter gene assay
GPR55-HEK293 and HEK293 cells (20,000 cells/well) were seeded in 96-well plates coated with 1% poly-D-lysine. Cells were transiently transfected with a pNFAT-Luc reporter plasmid (PathDetect NFAT cis-Reporter; Stratagene) at 200 ng/well alone or in combination with HA-RhoA (T19N) or pcDNA3.1 plasmids at 200 ng/well using Lipofectamine 2000. Cells were incubated with increasing amounts of LPI as indicated for 6 h in serum-free OptiMEM medium at 37°C. NFAT-luciferase responses were detectable as soon as 1 h after LPI treatment (data not shown); however, 6 h of LPI-treatment gave the best detection window. NFAT-reporter-luciferase activity was visualized by using the LucLite Kit (Packard Instrument Company, Meriden, CT, USA) as described previously (12)
. Luminescence was measured in a TopCounter (Top Count NXT; Packard) for 5 s. Luminescence values are given as relative light units (RLU).
Data analysis
Statistical analyses were performed using ANOVA for comparisons between multiple groups, followed by a Bonferroni’s post hoc analysis using GraphPad Prism (GraphPad, San Diego, CA, USA) and Mirocal Origin software (Mirocal, Northampton, MA, USA); P < 0.05 was considered to be significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
37 kDa that was not present in lysates from control HEK293 cells and is consistent with the predicted molecular mass for GPR55 (Fig. 1A
). A common feature of many G protein-coupled receptors (GPCRs) is receptor endocytosis following ligand binding (13)
. Thus, we evaluated whether LPI was capable of inducing receptor internalization. Live cells were treated with LPI and subsequently fixed, permeabilized, and stained for receptor expression with an HA antibody (Fig. 1B
). Analysis of HA immunoreactivity using confocal microscopy showed that under control conditions GPR55 was predominantly expressed at the plasma membrane (Fig. 1B
, left panel). Nevertheless, the treatment with 1 µM LPI for 60 min resulted in a pronounced redistribution of GPR55 into intracellular vesicles (Fig. 1B
, right panel). This approach reveals the total amount of receptor expressed in a given cell; however, it does not differentiate between receptors that have been internalized or are merely trapped within the protein synthesis and delivery pathways. Thus, we conducted antibody feeding experiments (10)
on living cells (Fig. 1C
), a method that allowed us to assess the fate of cell-surface receptors that were accessible to the HA antibody during a 30-min preincubation period. Antibody feeding experiments verified that GPR55 undergoes LPI-induced internalization and translocation into intracellular vesicles after 60 min (Fig. 1C
, middle panel) and 90 min (Fig. 1C
, right panel).
|
Activation of GPR55 induces oscillatory and prolonged Ca2+ release from intracellular stores via activation of phospholipase C (PLC)
Next, we evaluated the effects of LPI on Ca2+ homeostasis in GPR55-HEK293 cells. Brief exposure to LPI (5 min) resulted in a concentration-dependent induction of oscillatory Ca2+ transients that persisted for up to 45 min following agonist removal (Fig. 2A
), with an EC50 of 49 ± 4 nM (Fig. 2B
). In calibrated experiments, a supramaximal concentration of LPI (1 µM) increased intracellular Ca2+ levels by 929 ± 46 nM (n=20) above basal (237±75 nM). In control HEK293 cells, LPI failed to elevate Ca2+ levels; however, responses to 1 µM lysophosphatidic acid (LPA) were observed (Fig. 2C, D
). LPA activates endogenous LPA receptors in HEK293 cells (3)
, but unlike LPI, LPA responses in GPR55-HEK cells normally comprise a single Ca2+ transient. The magnitude of responses to LPI and LPA were similar in GPR55-HEK293 cells (Fig. 2E
).
|
To establish whether activation of GPR55 by LPI promoted Ca2+ release from intracellular stores, we investigated the effects of Ca2+-free medium and irreversible blockade of the ER Ca2+-ATPase with 2 µM of thapsigargin (14)
. Recently, thapsigargin has been shown to block GPR55-mediated responses to cannabinoids in HEK293 cells and dorsal root ganglia (DRG) neurons (6)
. Following removal of extracellular Ca2+, only a single peak was observed after treatment with LPI (Fig. 3A
). However, readdition of extracellular Ca2+ rescued the oscillatory Ca2+ signals (Fig. 3A
), suggesting that a store-operated Ca2+ influx mechanism is required to replenish intracellular stores and enable further Ca2+ release (15)
. Indeed, depletion of Ca2+ stores following a brief exposure to thapsigargin (5 min) resulted in sustained elevation in Ca2+ levels and abolished subsequent responses to 1 µM LPI (Fig. 3B
). Given that LPI evokes Ca2+ release from intracellular stores, we then investigated whether phospholipase C is part of this signaling pathway. Indeed, treatment of cells with 1 µM of the PLC inhibitor U73122 for 20 min markedly attenuated subsequent responses to LPI (Fig. 3C, D
).
|
LPI-mediated Ca2+ signaling involves G13 and activation of RhoA
Previous studies have implicated G13 in GPR55 signaling (1)
. We therefore overexpressed the catalytically inactive form of G13 (Q226L/D294N) (16)
to perturb endogenous G13 function. Transient transfection of GPR55-HEK293 cells with G13 (Q226L/D294N) resulted in a marked reduction in the duration of LPI responses (Fig. 4A, top trace; B, gray bar
), with respect to untransfected cells in the field (Fig. 4A, bottom trace; B, black bar
).
|
RhoA is a downstream effector of G12/G13 family G proteins and has also been implicated in GPR55- mediated signaling following treatment with cannabinoids (1)
. Thus, we examined the effects of LPI on RhoA activation in GPR55-HEK293 cells. As demonstrated in Fig. 5A, B
, LPI (1 µM) elicited RhoA activation at both 7 and 10 min, with optimal stimulation achieved at 10 min (n=3; P<0.05). Next, we used a dominant-negative RhoA mutant (T19N) (6)
to investigate the role of RhoA in LPI-induced Ca2+ transients in GPR55-HEK293 cells. Overexpression of RhoA (T19N) abolished or strongly inhibited LPI-induced Ca2+ signaling (Fig. 6A, top trace; D, gray bar
) with respect to untransfected cells in the field (Fig. 6A, bottom trace; D, black bar
). In contrast, responses to LPA were unaffected by overexpression of this construct (Fig. 6A
, both traces). Given that many of the cellular actions of RhoA are mediated by Rho kinase (ROCK) (17)
, we next investigated the effects of the ROCK inhibitor Y-27632 on LPI responses. Indeed, 50 µM of Y-27632 proved to be highly effective in preventing LPI-mediated Ca2+ signaling in GPR55-HEK293 cells (inhibited to 6.96±1.82% control; n=30; P<0.01; Fig. 6B, C
, gray bar). Partial inhibition of LPI responses was observed with 10 µM of Y-27632 (29.5±9.2% control; n=30; P<0.01).
|
|
Activation of GPR55 induces NFAT transcription factor activation via RhoA
Well-established downstream effects of oscillatory increases in cytosolic Ca2+ levels include the activation of transcription factors and changes in gene expression (18
, 19)
. Thus, we tested the ability of GPR55 to induce NFAT as an example of a Ca2+/calcineurin-dependent regulator of transcription (20)
. Treatment of GPR55-HEK293 cells with increasing amounts of LPI for 6 h resulted in a dramatic, dose-dependent induction of NFAT-luciferase reporter activity (Fig. 7A
, solid circles) with an EC50 of 1.37 ± 0.06 µM. In control HEK293 cells, no effect of LPI on NFAT-luciferase activity was observed (Fig. 7A
, open circles). In addition, overexpression of the dominant-negative RhoA (T19N) markedly inhibited the effects of 1 µM LPI in GPR55-HEK293 cells (Fig. 6B
, gray bar). Next, we used live confocal imaging to investigate the ability of LPI to promote the nuclear translocation of GFP-NFATc3 (19)
transiently expressed in GPR55-HEK293 cells. Before agonist treatment, GFP-NFATc3 was predominantly localized in the cytosol (Fig. 7C
, left panel), but following exposure to 3 µM LPI, a rapid redistribution to the nucleus was observed (Fig. 7C
, right panel; n=7/9 cells). LPI (3 µM) was without effect on GFP-NFATc3 translocation when expressed in control HEK293 cells (n=10 cells).
|
Do cannabinoids promote GPR55-mediated Ca2+ signaling?
At present the ability of cannabinoids to activate GPR55 is controversial, with evidence both for (1
, 2
, 6)
and against an interaction (3)
. We therefore profiled the activity of selected cannabinoid ligands in GPR55-HEK293 and control HEK293 cells using Ca2+ signaling as a reporter of receptor activation. The principle endocannabinoids AEA and 2-AG have been suggested to activate GPR55 (1)
. Although these endocannabinoids were able to elevate intracellular Ca2+ levels in a concentration range of 3–30 µM, no difference in maximal response, duration, and/or potency was observed when GPR55-HEK293 cells (Fig. 8A, B
, solid squares) were compared to control HEK293 cells (Fig. 8A, B
, open squares). In addition, the temporal profile of the Ca2+ responses were also different from those induced by LPI in GPR55-HEK293 cells, with less oscillatory activity (mostly single transients) and much larger Ca2+ elevations above basal levels (for example, 5.59±1.57 µM with 30 µM AEA and 1.27±0.81 µM with 30 µM 2-AG in control HEK293 cells). Thus, at concentrations at or below 3 µM, we find no evidence for a selective and LPI-like GPR55 activation with the endocannabinoids AEA and 2-AG.
|
The nonclassical cannabinoid agonist CP55,940 has also been suggested to activate GPR55 (1)
. Here we find that CP55,940 also had no effect on intracellular Ca2+ levels in GPR55-HEK293 cells when applied alone at concentrations up to 3 µM. However, on addition of 3 µM CP55,940, the concentration-response curve for LPI (Fig. 8C
, solid squares) was shifted to the right by a factor of 10 (Fig. 8C
, open squares), suggesting that CP55,940 can act as a competitive antagonist at GPR55 in this assay.
Finally, we evaluated the CB1 antagonist/inverse agonist AM251 as this ligand also has been suggested to activate GPR55 (1)
. At 3–10 µM, AM251 had no effect on control HEK293 cells (Fig. 8D
); however, in GPR55-HEK293 cells concentrations of 0.3–10 µM were able to induce sustained and oscillatory Ca2+ responses reminiscent of LPI responses (Fig. 8D
), with a peak change in Ca2+ of 586 ± 48 nM using a maximally effective concentration of AM251 (3 µM). However, AM251 was less potent than LPI, with an EC50 of 612 ± 86 nM and a slightly lower maximal response (0.21± 0.01 vs. 0.25±0.01 ratio units; P<0.05).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
13) and small GTPases (RhoA) with oscillatory and prolonged Ca2+ signaling and downstream NFAT activation. Our findings and those of Oka et al. (3)
suggest that the endogenous phospholipid LPI can activate GPR55, but the endocannabinoids AEA and 2-AG have no clear effect at equivalent concentrations. Despite this, the CB1 antagonist AM251 also efficiently activated GPR55, which is supportive of the sensitivity of GPR55 to LPI and a subset of cannabinoid ligands.
GPR55 promotes Ca2+ release from intracellular stores
LPI has previously been shown to induce a modest elevation in intracellular Ca2+ levels (50 nM) in cuvette-based assays with HEK293 cells expressing GPR55 (3)
. Here we demonstrate that the pathway linking GPR55 with a Ca2+ response is novel, involving G13 and RhoA-mediated Ca2+ mobilization from intracellular stores. In addition, the use of single-cell imaging reveals a sustained, oscillatory Ca2+ response profile distinct from that induced by activation of endogenous LPA receptors, which can engage Gq (21)
. LPA receptors are also thought to activate G12/G13 family G proteins (7)
; however, we did not see any functional evidence for this in terms of promoting RhoA-dependent Ca2+ signaling. The magnitude of the responses to LPI within individual cells was much larger than that reported previously using a population-based assay (3)
, which is likely to reflect the oscillatory nature of the response. Although cannabinoid ligands have also been shown to interact with GPR55, they had no effect of on Ca2+ activity in GPR55-expressing HEK293 cells assessed using a FLIPR (1)
, which again, may reflect the complex, temporal profile of the Ca2+ response.
The Ca2+ response to LPI primarily reflects Ca2+ release from intracellular stores, as responses were observed in the absence of extracellular Ca2+ and blocked by prior depletion of intracellular stores with thapsigargin. The sustained oscillatory activity did however require extracellular Ca2+, presumably because a store-operated Ca2+ influx pathway is necessary to replenish the ER (15)
. Indeed, thapsigargin induced a sustained elevation in intracellular Ca2+ levels consistent with the existence of store-operated Ca2+ influx pathways in HEK293 cells (22)
. The mechanism responsible for triggering Ca2+ release is likely to be IP3 formation, as responses were blocked in the presence of the PLC inhibitor U73122, which has previously been reported to prevent GPR55-mediated Ca2+ responses to JWH015 (6)
.
Role of G13 and Rho in GPR55-mediated Ca2+ signaling
GPR55 is thought to couple to G12/G13 family G proteins and possibly Gq (6)
, but not Gi or Gs, as was determined using antibody and peptide blocking approaches (1)
. Here we confirm the involvement of G12/G13 family G proteins in GPR55 signaling using overexpression of a catalytically inactive G13. We also show that RhoA can be activated by LPI in our GPR55-expressing cell lines. These data are consistent with a previous study demonstrating RhoA activation following treatment of GPR55-expressing HEK293 cells with cannabinoids (1)
.
Interestingly, using dominant-negative RhoA or Y-27632, a well-established ROCK inhibitor, we were able to block very effectively LPI-mediated Ca2+ transients. These data are in agreement with a recent study by Lauckner et al. (6)
, who demonstrated that Ca2+ signaling associated with cannabinoid activation of GPR55 involves RhoA. This group also observed that dominant-negative Gq inhibited GPR55 responses, indicating the possibility of dual signaling routes to PLC via GPR55.
NFAT is a key molecular target for GPR55
NFAT family proteins are regulated primarily via an increase in intracellular Ca2+, which activates the serine/threonine phosphatase calcineurin, which in turn dephosphorylates NFAT, allowing its nuclear translocation. NFAT can then bind DNA and regulate the transcriptional activity of a number of genes (20)
. The ability of GPR55 to signal via NFAT is likely to be a consequence of its effects on Ca2+ signaling. The oscillatory nature of the response would make this much more effective, by allowing for the efficient recruitment of Ca2+-dependent processes while minimizing metabolic stress/toxicity associated with a continuous Ca2+ elevation (19)
. Oscillatory Ca2+ signals show a frequency-dependence in their ability to promote changes in gene expression, and the temporal characteristics of LPI-induced oscillatory activity appears to be well tuned to calcineurin-dependent NFAT activation (18
, 19)
, suggesting that this may be a key molecular target for GPR55. The ability of RhoA to induce NFAT activation is novel and likely to be a consequence of its role in promoting Ca2+ release in HEK293 cells. In contrast, a previous study has shown a RhoA-dependent inhibition of transcriptional activity of NFAT at the IL-2 promoter in human T cells (23)
. Thus, the effects of RhoA activation on NFAT function may be cell-type specific, and depend on the signaling profile downstream of RhoA. In comparison with acute Ca2+ signaling, LPI was less potent in inducing NFAT-responses. It is not clear what accounts for this difference, but a possible scenario may be that efficient NFAT activation requires a more prolonged Ca2+ stimulus and analysis of the initial peak Ca2+ level may not directly equate with the ability of LPI to increase Ca2+ levels over more prolonged periods (6 h). However, as LPI is an endogenous molecule it may also be subject to metabolism/degradation during the NFAT assay period, thereby reducing its apparent potency.
Cannabinoid activity at GPR55
GPR55 is thought to be sensitive to a range of cannabinoid ligands (1
, 2
, 6)
; however, there are a number of inconsistencies in the pharmacological profiles of cannabinoids that interact with GPR55. Here we find that the endocannabinoids anandamide and 2-AG have no specific GPR55-mediated Ca2+ signaling activity below 3 µM and at higher concentrations elicit a marked Ca2+ response in HEK293 cells devoid of GPR55. It should be noted that the clonal background of HEK293 cells can differ markedly between laboratories, and this additional cannabinoid Ca2+ signaling pathway may not be observed in all studies/conditions (6)
. It is noteworthy that non-CB1/CB2-receptor-mediated effects of the endocannabinoids AEA and 2-AG on Ca2+ homeostasis have been reported in a variety of cell types (3
, 25
26
27
28)
. AEA and 2-AG have previously been reported to be partial agonists at GPR55 (1)
; thus, in the present case they may not be able to generate sufficient IP3 to trigger a LPI-like Ca2+ response. Some of these discrepancies may also reflect ligand-dependent signaling through GPR55 and/or host-dependent effects. Thus, the eicosanoids may be less effective than LPI (and AM251) in driving RhoA-dependent Ca2+ responses. In addition, AEA can promote GPR55-mediated Ca2+ responses in a human endothelial cell line; however, this effect is complex, being constrained by CB1 receptor activation, which is in turn regulated by integrin clustering (24)
. On balance, the present data suggest that 2-AG and AEA can affect Ca2+ homeostasis in a variety of cell types; however, these actions do not necessarily involve GPR55. It is notable that the lack of appreciable activation of GPR55-mediated Ca2+ responses observed in this study does not preclude activation of differential GPR55-mediated signaling events by these endogenous eicosanoids.
In agreement with a previous study, we observed that CP55,940 does not induce Ca2+ signaling via GPR55 (6)
. However, the presence of CP55,940 resulted in a rightward shift of the LPI concentration-response curve, suggesting that this compound can behave as a competitive antagonist at GPR55. The pharmacological actions of CP55,940 at GPR55 may however be protean, as it is suggested to behave as a partial agonist in a different assay involving GTP
S binding (1)
. Thus, competitive antagonism observed in the present study may reflect a lack of efficacy of this compound in promoting Ca2+ signaling. Finally, the widely used CB1 receptor antagonist/inverse agonist AM251, elicited a significant agonist response via GPR55. These data are consistent with a previous study reporting G protein activation with AM251 (1)
, although notably this was at a similar level to that observed with AEA and 2-AG.
In summary, GPR55 displays a sensitivity to certain cannabinoid ligands, but due to the lack of clear, potent effects of endocannabinoids in the present Ca2+ signaling assay, the status of GPR55 as a novel cannabinoid receptor requires further validation. Also, it will be interesting to establish whether any of the physiological effects of CB1 antagonists that have been previously ascribed to CB1 inhibition involve GPR55 activation.
Despite its widespread distribution, the physiological function of GPR55 remains to be established. The prominent expression of GPR55 within the brain, DRG neurons, and immune cells suggests a potential role in these tissues and in regulation of inflammatory/neuropathic pain (6
, 29)
, which is supported by recent data using GPR55 knockout animals (31)
. The potent, endogenous GPR55 ligand LPI is also known to have mitogenic activity and may influence the growth of certain tumors (5)
, further highlighting a role for GPR55 under pathological conditions. At least some of the cellular consequences of GPR55 activation are likely to involve Ca2+ mobilization from intracellular stores; for example, in DRG neurons (6)
and endothelial cells (24)
as well as activation of Rho family G proteins, which typically play a role in regulating cell morphology and mobility as well as influencing gene expression, cell–cell adhesion, proliferation, and membrane trafficking (17
, 30)
. A link between RhoA activation and Ca2+ signaling is also indicated, and this noncanonical pathway may be an important mediator of GPR55 actions (Fig. 9
). Thus, GPR55 is quite distinct from other GPCRs and represents an intriguing and unique therapeutic target.
|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication March 18, 2008. Accepted for publication August 7, 2008.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Kapur, P. Zhao, H. Sharir, Y. Bai, M. G. Caron, L. S. Barak, and M. E. Abood Atypical Responsiveness of the Orphan Receptor GPR55 to Cannabinoid Ligands J. Biol. Chem., October 23, 2009; 284(43): 29817 - 29827. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Cinar and M. Szucs CB1 Receptor-Independent Actions of SR141716 on G-Protein Signaling: Coapplication with the {micro}-Opioid Agonist Tyr-D-Ala-Gly-(NMe)Phe-Gly-ol Unmasks Novel, Pertussis Toxin-Insensitive Opioid Signaling in {micro}-Opioid Receptor-Chinese Hamster Ovary Cells J. Pharmacol. Exp. Ther., August 1, 2009; 330(2): 567 - 574. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
