Combination small molecule MEK and PI3K inhibition enhances uveal melanoma cell death in a mutant GNAQ- and GNA11-dependent manner

Abstract

Purpose: Activating Q209L/P mutations in GNAQ or GNA11 (GNAQ/11) are present in approximately 80% of uveal melanomas. Mutant GNAQ/11 are not currently therapeutically targetable. Inhibiting key down-stream effectors of GNAQ/11 represents a rational therapeutic approach for uveal melanomas that harbor these mutations. The mitogen-activated protein/extracellular signal-regulated kinase/mitogen-activated protein kinase (MEK/MAPK) and PI3K/AKT pathways are activated in uveal melanoma. In this study, we test the effect of the clinically relevant small molecule inhibitors GSK1120212 (MEK inhibitor) and GSK2126458 (pan class I PI3K inhibitor) on uveal melanoma cells with different GNAQ/11 mutation backgrounds.

Experimental design: We use the largest set of genetically annotated uveal melanoma cell lines to date to carry out in vitro cellular signaling, cell-cycle regulation, growth, and apoptosis analyses. RNA interference and small molecule MEK and/or PI3K inhibitor treatment were used to determine the dependency of uveal melanoma cells with different GNAQ/11 mutation backgrounds on MEK/MAPK and/or PI3K/AKT signaling. Proteomic network analysis was done to unveil signaling alterations in response to MEK and/or PI3K small molecule inhibition.

Results: GNAQ/11 mutation status was not a determinant of whether cells would undergo cell-cycle arrest or growth inhibition to MEK and/or phosphoinositide 3-kinase (PI3K) inhibition. A reverse correlation was observed between MAPK and AKT phosphorylation after MEK or PI3K inhibition, respectively. Neither MEK nor PI3K inhibition alone was sufficient to induce apoptosis in the majority of cell lines; however, the combination of MEK + PI3K inhibitor treatment resulted in the marked induction of apoptosis in a GNAQ/11 mutant-dependent manner.

Conclusions: MEK + PI3K inhibition may be an effective combination therapy in uveal melanoma, given the inherent reciprocal activation of these pathways within these cells.

Figure 1

Effect of GNAQ or GNA11 knockdown on cell growth and MAPK/AKT signaling in GNAQ mutant, GNA11 mutant or wild-type uveal melanoma cells. A, GNAQ mutant [Q209L (92.1 and MEL202), Q209P (MEL 270)] or GNA11 mutant [Q209L (OMM1, UPMD1, and UPMD2)] or wild-type (MEL285, MEL290) uveal melanoma cells were transfected with two unique siRNAs specific to GNAQ or GNA11. Cell growth was determined 72 hrs after transfection relative to cells that were untreated (untransfected), or transfected with lipid alone, non-targeting siRNA (negative control), or siRNA targeting PlK1 (positive control, siRNA-P). Combined results of two independent experiments, bars represent 1 SEM. B, Western-blot analysis of GNAQ, GNA11, P-MAPK (threonine 202, tyrosine 204), MAPK, P-AKT (serine 473), AKT and PTEN expression in GNAQ mutant (Q209L (92.1)), GNA11 mutant (Q209L (UPMD2)), versus GNAQ/11 wild-type (MEL285) uveal melanoma cells following transfection with two unique GNAQ or GNA11 siRNAs relative to non-targeting control.

Figure 2

Effect of small molecule MEK or PI3K inhibitors on GNAQ mutant, GNA11 mutant, or wild-type uveal melanoma cell growth and MAPK/AKT signaling. GNAQ mutant [Q209L (92.1, MEL202)] or GNA11 mutant [Q209L (OMM1, UPMD2)] or GNAQ/11 wild-type (MEL285, MEL 290) uveal melanoma cells were used for the following experiments: A, uveal melanoma cells were treated with increasing doses of GSK1120212 (MEK inhibitor = MEKi) or GSK2126458 (PI3K inhibitor = PI3Ki) for 3 hrs and expression of P-MAPK, MAPK, P-AKT, and AKT determined by western-blot analysis. B, uveal melanoma cells were treated with increasing doses of MEKi or PI3Ki for 72 hrs and percent of maximal cell growth inhibition determined. (representative experiments, bars represent SD).

Figure 3

Effect of small molecule MEK and/or PI3K inhibitors on signaling pathways in GNAQ mutant versus wild-type uveal melanoma. A, Mutant GNAQ [Q209L (92.1, MEL202)] or wild-type (MEL285, MEL290) uveal melanoma cells were treated with MEKi alone (10nM), PI3Ki alone (100nM), or the combination of MEKi + PI3Ki for 4, 12 or 24 hrs. Reverse Phase Protein Array was performed to assess the expression of proteins in the MEK/MAPK, PI3K/AKT, cell cycle and apoptosis pathways. The heatmap represents the mean centered log₂ expression value for each protein after the indicated time of treatment (red = greater, green = lesser). Each treatment was performed in triplicate; therefore there are three distinct heatmap boxes per treatment condition. B, The highest scoring network interactions (edge flux values) from the NetWalk analysis of directly measured proteins following 4 hours of MEKi and/or PI3Ki treatment. Nodes represent proteins, edges represent network interactions among nodes, node-coloring represents relative increase in protein level. Asterisks (*) indicate phosphorylated proteins.

Figure 4

Effect of c-jun knockdown on combination MEKi + PI3Ki treatment of uveal melanoma cells. A, GNAQ Q209L mutant uveal melanoma cells (92.1) were treated with combination MEKi (10nM) + PI3Ki (100nM) for 4 hours and levels of phosphorylated c-jun determined relative to untreated cells. B, Cells were transfected with two distinct siRNAs targeting c-jun, then levels of phosphorylated c-jun, total c-jun or beta-actin were determined by western blot. C, Cells were transfected with two distinct siRNAs targeting c-jun, then treated with combination MEKi (10nM) + PI3Ki (100nM). The effect of c-jun knockdown in MEKi + PI3Ki treated cells was compared to the effect of MEKi + PI3Ki treatment alone. Asterisks (*) indicate p value <0.05.

Figure 5

Effect of small molecule MEK and/or PI3K inhibitors on cell cycle regulation in GNAQ mutant, GNA11 mutant or wild-type uveal melanoma cells. The indicated mutant GNAQ mutant GNA11 or wild-type uveal melanoma cells were treated with MEKi alone, PI3Ki alone, or the combination of MEKi + PI3Ki for 24 hrs. Cell cycle regulation was assessed by single cell flow cytometry analysis and the percent of cells in the S/G2/M phases determined. Cell index was determined as the percent of cells in S/G2/M phase normalized to untreated baseline cultures, bars represent 1 SEM. Means of triplicate experiments were compared with repeated measures one-way ANOVA and Bonferroni's Multiple Comparison Test using Graphpad Prism. Significance indicates a p-value < 0.05 given a confidence interval of 95% of difference.

Figure 6

Effect of small molecule MEK and/or PI3K inhibitors on apoptosis induction in GNAQ mutant, GNA11 mutant or wild-type uveal melanoma cells. A, Representative flow cytometry experiment assessing levels of annexin V and propidium iodide in GNAQ mutant cells (MEL202) after MEKi alone, PI3Ki alone, or the combination of MEKi + PI3Ki treatment at 24, 48 and 72 hours. B, Mutant GNAQ [Q209L (92.1, MEL202), Q209P (MEL270)], mutant GNA11 [Q209L (OMM1, UPMD1, UPMD2)] or wild-type (MEL285, MEL290) uveal melanoma cells were treated with MEKi alone, PI3Ki alone, or the combination of MEKi + PI3Ki for 48 hrs. Induction of apoptosis was assessed by flow cytometry using Annexin V and Propidium Iodide staining. The percent of cells in early apoptosis was determined by gating the high annexin V/low propidium iodide cell population following treatment. Data represents n=3-6 experiments, bars represent 1 SEM. C, Light microscopy image of uveal melanoma spheroids (bar = 50 um). D, Confocal microscopic imaging of caspase 3/7 activity in cells within uveal melanoma spheroids (UPMD2) following treatment with increasing doses of MEKi and/or PI3Ki. Heatmap represents caspase 3/7 positive cells per spheroid area.