Gβγ subunits inhibit Epac-induced melanoma cell migration

Reagents were purchased from Sigma unless otherwise specified. G-protein βγ activating peptide, myr-SIRKALNILGYPDYD-OH (mSIRK), and its control peptide, myr-SIRKALNIAGYPDYD-OH (L9A), 2-aminoethoxydiphenyl borate (2-APB), xestopongin C, ryanodine, N-(6-aminohexyl)-1-naphthalenesulfonamide (W-5) were purchased from EMD Chemicals. Guanosine 5', α-,β-methylene, triphosphate (Gp(CH2)pp) was purchased from Biomol International. 8-(4-Methoxyphenylthio)-2'-O-methyladenosine-3',5'-cyclic monophosphate (8-pMeOPT) was purchased from Biolog. Antibodies against Epac1, Gβ1, Gγ2, C-terminal domain of β-adrenergic receptor kinase (βARK-CT) were purchased from Santa Cruz Biotechnology. Anti-α-tubulin antibody was purchased from Abcam. SK-Mel-2 and SK-Mel-24 melanoma cell lines were obtained from the American Type Culture Collection. SK-Mel-187 cell line was kindly provided by Dr. Alan Houghton. C8161 cell line was provided by Dr. Mary JC Hendrix. WM1552C, WM115, WM3248, WM1361A were obtained from Dr. Meenhard Herlyn. SK-Mel-2 and SK-Mel-24 cells were maintained in MEM containing 10% FBS, 1% penicillin-streptomycin. All other melanoma cells were maintained in RPMI containing 10% FBS, 1% penicillin-streptomycin.

Measurement of intracellular Ca ²⁺ release was performed as we previously described [18]. Cells were incubated with HEPES buffer containing 4 μmol/L of Fluo-4AM followed by washing and incubation with HEPES buffered saline containing 1.8 mmol/L of CaCl₂. An iXon+ 885 charge-coupled device camera (Andor Technology) was used to monitor fluorescence changes. Full images were collected every 4 seconds. The field was illuminated by two wavelengths in rapid succession. Fluo-4 was excited at 488 nm, and data were expressed as normalized changes in background-corrected fluorescence emission (F/F₀). Data were analyzed using Imaging Workbench (INDEC BioSystems). Representative Ca ²⁺ signals averaged from 5 individual cells were shown in the figures.

Western blot analysis was performed as we previously described [22, 23]. Briefly, cells were lysed and sonicated in lysis buffer. Equal amounts of protein were subjected to SDS-PAGE. After protein separation by electrophoresis, samples were transferred to Millipore Immobilon-P membrane and immunoblotting with antibodies was performed.

Migration assay was performed using the 24-well Boyden chambers (8 μm pores, BD Biosciences) as we previously described [17]. The cells were plated at a density of 1 × 10⁶ cells/100 μl of medium in the inserts, and incubated for 3 h at 37°C followed by staining using the Diff-Quick kit (Dade Behring). Pictures were taken with a microscope and migrated cells were counted with Image J software using 10 randomly chosen fields.

Adenoviral overexpression of Epac1, βARK-CT or LacZ was performed as we previously described [18]. βARK-CT adenovirus was kindly provided by Dr. Koch. The recombinant vector was introduced into human embryonic kidney cells (HEK-293) to recover infectious adenovirus. Cells were infected with adenovirus for 24 h followed by confirmation of overexpression of target proteins.

Plasmid constructs harboring Gβ1 or Gγ2 subunit were kindly provided by Dr. Simonds [24]. Plasmid transfection was performed as we previously described [22, 23]. Briefly, 750 μl of serum- and antibiotics-free OPTI-MEM was mixed with both 13.5 μl of Lipofectamine2000 and 6.25 μl plasmid DNA. The mixture was then diluted with 5 ml MEM supplement with 5% FBS and gently overlaid onto the cells, followed by incubation for 16 h. The overexpression was confirmed by western blot analyses, and the transfected cells were used for cell migration or cytosolic Ca ²⁺ measurement assays.

Statistical comparisons among groups were performed using one factor ANOVA with Bonferroni post hoc test. Statistical significance was set at the 0.05 level.

We have previously demonstrated that Epac increases melanoma cell migration by modification of heparan sulfate [17]. In addition, since it has been reported that Gβγ plays a role in cell migration of endothelial cells and breast cancer cells [4‐6], we examined the effects of Gβγ on melanoma cell migration. We found that mSIRK, a cell-membrane permeable activator of Gβγ [7], decreased basal cell migration only at the highest dose (50 μM). In contrast, mSIRK inhibited Epac1 overexpression-induced cell migration in a dose-dependent manner (Figure 1a). These data suggest that the inhibitory effect of Gβγ is obvious under Epac-activated conditions, but not clear under basal conditions (Figure 1a). Gp(CH)₂pp, a constitutively active GTP analogue which dissociates Gβγ from Gα, also inhibited Epac-induced cell migration. In addition, mSIRK inhibited cell migration induced by 8-pMeOPT, an Epac-specific agonist, suggesting that Gβγ also inhibits cell migration induced by endogenous Epac (Figure 1b). Further, we examined the specificity of Gβγ in the inhibition of Epac-induced cell migration. Co-overexpression of Gβ1 and Gγ2 subunits (Figure 1c), which is the major combination of Gβγ [25], inhibited Epac1-induced cell migration (Figure 1e). By contrast, overexpression of βARK-CT (Figure 1d), an inhibiting-peptide for Gβγ [26], abolished the mSIRK's inhibitory effect (Figure 1e). These data suggest that there is cross talk between Gβγ and Epac, which affects melanoma cell migration. Since activation of Epac was achieved by artificial overexpression or the Epac-agonist which does not exist in nature, we tested whether hormonal control of GPCR increases cell migration via Epac. Stimulation of β-adrenergic receptor increased cell migration in melanoma, and it was inhibited by ablation of Epac1 (Figure 1f), suggesting that Epac increases cell migration upon activation of hormone receptors.

We next explored the mechanism by which Gβγ inhibits Epac-induced cell migration. Since our previous report have shown that Epac1-induced cell migration is mediated by Ca ²⁺ release from the ER [18], we examined whether mSIRK affects Epac1-induced cytosolic Ca ²⁺ elevation. When cells were pretreated with mSIRK, but not with L9A, 8-pMeOPT failed to increase Ca ²⁺ signal (Figure 2a and 2b). In addition, co-overexpression of Gβ1 and Gγ2 also inhibited 8-pMeOPT-induced Ca ²⁺ elevation (Figure 2c). Further, inhibition of Gβγ with βARK-CT, or guanosine 5'-O-(2-thiodiphosphate) (GDPβS), a GDP analogue that inactivates Gβγ, restored 8-pMeOPT-induced Ca ²⁺ elevation (Figure 2d and 2e). These data suggest that Gβγ inhibits Epac-induced Ca ²⁺ release.

Since IP3 receptors are inactivated by prior cytosolic Ca ²⁺ elevation [19‐21], we hypothesized that Ca ²⁺ elevation induced by Gβγ inhibits Epac-induced Ca ²⁺ elevation. mSIRK, but not L9A, increased Ca ²⁺ signal in SK-Mel-2 cells (Figure 3a and 3b). mSIRK-induced Ca ²⁺ elevation similarly observed in various melanoma cells including WM1552C, a regional growth pattern primary melanoma cell line (RGP), WM115, WM1361A, WM3284, vertical growth pattern primary melanoma cells (VGP), and SK-Mel-24, SK-Mel-187 and C8161, metastatic melanoma cell lines (MM) (data not shown), suggesting mSIRK's effect on Ca ²⁺ is universal in different types of melanoma. After the elevation of Ca ²⁺ signal by mSIRK, 8-pMeOPT did not show an additional elevation of Ca ²⁺ signal (Figure 3b). GDPβS inhibited mSIRK-induced cytosolic Ca ²⁺ elevation, suggesting the specificity of mSIRK for Gβγ (Figure 3c). We next examined the source of Gβγ-induced Ca ²⁺. When extracellular Ca ²⁺ was depleted by removal of Ca ²⁺ or by addition of EGTA, a Ca ²⁺-chelating agent, mSIRK did not increase Ca ²⁺ signal (Figure 3d and 3e, respectively). By contrast, inhibition of IP3 receptors or ryanodine receptor did not affect mSIRK-induced Ca ²⁺ signal (Figure 3f, g and 3h). Further, depletion of Ca ²⁺ store in the ER with thapsigargin did not inhibit mSIRK-induced Ca ²⁺ elevation (Figure 3i). Taken together, these data suggest that Gβγ induces Ca ²⁺ from the extracellular space, but not from intracellular store.

Since IP3 receptor is known to be inactivated by Ca ²⁺/calmodulin [19‐21], we examined whether calmodulin mediates Gβγ's inhibitory effect on Epac-induced Ca ²⁺ elevation. W-5, a calmodulin inhibitor, restored Epac-mediated Ca ²⁺ signal, and cell migration even in the presence of mSIRK (Figure 4a and 4c, respectively). In order to identify Ca ²⁺ channel in which Gβγ activates Ca ²⁺ influx from the extracellular space, we tested Ca ²⁺ inhibitors in mSIRK-induced Ca ²⁺ elevation. We then found that, 4-methyl-4' -[3,5-bis(trifluoromethyl)-1H-pyrazol-1-yl]-1,2,3-thiadi azole-5-carboxanilide (YM58483), which is known as an antagonist for Ca ²⁺-activated Ca ²⁺ channels (CRAC) [27] or canonical transient receptor potential (TRP) channels [28], inhibited mSIRK-induced Ca ²⁺ signal (Figure 4b). In addition, YM58483 attenuated Gβγ's inhibitory effect on Epac-induced cell migration (Figure 4c). Since YM583483 did not change basal cell migration (Figure 4c), Gβγ-activation is necessary to inhibit Epac-induced cell migration. In contrast, neither verapamil nor nifedipine inhibit mSIRK-induced Ca ²⁺ (data not shown), suggesting that dihydropyridine Ca ²⁺ channels are not involved in Gβγ-induced Ca ²⁺ signal. Put together, Gβγ inhibits Epac-induced Ca ²⁺ elevation, and cell migration by Ca ²⁺/calmodulin pathway.