Recovery of phospho-ERK activity allows melanoma cells to escape from BRAF inhibitor therapy

Abstract

Background: Resistance to BRAF inhibitors is an emerging problem in the melanoma field. Strategies to prevent and overcome resistance are urgently required.

Methods: The dynamics of cell signalling, BrdU incorporation and cell-cycle entry after BRAF inhibition was measured using flow cytometry and western blot. The ability of combined BRAF/MEK inhibition to prevent the emergence of resistance was demonstrated by apoptosis and colony formation assays and in 3D organotypic cell culture.

Results: BRAF inhibition led to a rapid recovery of phospho-ERK (pERK) signalling. Although most of the cells remained growth arrested in the presence of drug, a minor population of cells retained their proliferative potential and escaped from BRAF inhibitor therapy. A function for the rebound pERK signalling in therapy escape was demonstrated by the ability of combined BRAF/MEK inhibition to enhance the levels of apoptosis and abrogate the onset of resistance.

Conclusion: Combined BRAF/MEK inhibition may be one strategy to prevent the emergence of drug resistance in BRAF-V600E-mutated melanomas.

Figures

Figure 1

PLX4720 inhibits the growth of melanoma cells harbouring the BRAF-V600E mutation. (A). Increasing concentrations of PLX4720 reduced the growth of melanoma cell lines harbouring the BRAF-V600E mutation (WM35, 1205Lu and WM164), whereas melanoma cell lines that were BRAF wild type were relatively resistant (WM1346, WM1361A and WM1366). Cells were treated with drug (3 nM–30 μM) for 72 h, and cell numbers were quantified using the MTT assay. Bars show s.e. mean. (B) Low doses of PLX4720 are cytostatic in melanoma cells harbouring the BRAF-V600E mutation. 1205Lu cells were treated were either 0.3 or 3 μM PLX4720 for 24 h before being fixed, stained with propidium iodide and analysed by flow cytometry. (C) PLX4720 inhibits MAPK signalling in BRAF-V600E-mutated melanoma cells. Cells were treated with increasing concentrations of PLX4720 (0.03–30 μM, 1 h); proteins were extracted and probed for expression of phospho-ERK (pERK). Blots were stripped once and reprobed for total-ERK to show even protein loading. (D) PLX4720 induces a concentration-dependent reduction in the phosphorylation of the retinoblastoma protein (phospho-RB), induces the cleavage of PARP, stabilises p27 and suppresses the expression of cyclin D1 in WM164 BRAF-V600E-mutated melanoma cells. Cells were treated with increasing concentrations of PLX4720 (3 nM–30 μM) for 24 h, after which time, protein was extracted and resolved by western blotting (C=vehicle control). Blots were stripped once and probed for actin to show equal protein loading.

Figure 2

PLX4720 induces apoptosis in BRAF-V600E-mutated melanoma cell lines. (A) PLX4720 induces apoptosis in three BRAF-mutated melanoma cell lines. Cultures were treated with increasing concentrations of PLX4720 (0.03–30 μM, 48 h), before staining for FITC-annexin-V and flow cytometry. Data show mean of three experiments. (B) PLX4720 reduces viability and invasion of 1205Lu cells grown as 3D collagen-implanted spheroids. Preformed 1205Lu spheroids were implanted into collagen and overlayed with media. Cells were treated with PLX4720 (0.3–30 μM for 72 h) before being treated with calcein-AM and ethidium bromide. Green, viable cells; red, dead cells. Lack of green staining also indicates a loss of cell viability. Magnification × 10. *P<0.05, Significant difference from control. The colour reproduction of this figure is available on the html full text version of the manuscript.

Figure 3

Melanoma cells escape PLX4720 and become resistant. (A) Representative photomicrograph of WM164 and 1205Lu melanoma cells treated with PLX4720 (3 μM) for either 14 or 28 days. (B) Western blot showing levels of pERK expression in PLX4720 naive (N) and resistant (R) (8 weeks, 3 μM) WM164 and 1205Lu cell lines. Note that the resistant cell lines were maintained continuously in the presence of PLX4720 (3 μM). Total-ERK demonstrates even protein loading. (C) Resistant 1205Lu and WM164 cell lines continue to incorporate BrdU in the continual presence of PLX4720 (3 μM). Panel shows either treatment-naive WM164 and 1205Lu cell lines (control) or resistant (chronically treated with PLX4720 for 8 weeks) treated with either PLX4720 (3 μM) or the MEK inhibitor U0126 (3 and 10 μM). Cells were stained for BrdU (20 μM, 1 h) uptake and the cell viability marker 7-AAD and were analysed by flow cytometry. (D) Representative sequencing trace from Exon 3 of MEK1 of 1205Lu cells chronically treated with PLX4720 for 8 weeks, arrow indicates site of P124L mutation identified previously in Emery et al (2009).

Figure 4

pERK signalling recovers after PLX4720 treatment. (A) Naive WM164 melanoma cells were treated with PLX4720 (3 μM, every 24 h) for increasing periods of time (0–48 h) and probed for pERK and total-ERK (tERK). (B) Recovery of pERK is observed in three naive BRAF-V600E-mutated melanoma cell lines. Cells were treated with PLX4720 for 0, 8, 24, 48 h (3 μM) and analysed as in (A). (C) Most cells remain growth arrested even when pERK recovers. WM793 cells were treated with PLX4720 (3 μM) for 0–72 h. Cells were harvested, fixed and stained with propidium iodide before being analysed by flow cytometry. (D) p27 expression levels remain high even when pERK signalling recovers. WM793 cells were treated with PLX4720 as for (C); protein lysates were probed for expression of pERK, total-ERK (tERK), phospho-RB (p-pRB), total retinoblastoma protein (t-RB) and p27. Equal protein loading was confirmed by stripping the blot once and probing for GAPDH expression. (E) PLX4720 treatment leads to a drop in cell numbers followed by stabilisation of the population. WM793, 1205Lu and WM164 melanoma cells were treated with PLX4720 (3 μM) for 0–120 h. At each time point, the cells were removed from the plate and counted. Data show the mean±s.e.mean of three independent experiments.

Figure 5

The function of rebound pERK signalling in the escape from PLX4720 treatment. (A) U0126 blocks the rebound increase in pERK after PLX4720 treatment. Melanoma cells were either treated with vehicle (0), PLX4720 (3 μM) or PLX4720 + U0126 (both 3 μM) for 48 h, protein was then probed for expression of pERK and tERK. (B) Melanoma cells were treated with increasing concentrations of U0126 for 1 h before being probed for pERK and tERK expression. (C) Cells were treated with increasing concentrations of PLX4720 (30 nM–30 μM) for 24 h in the absence or presence of U0126 (3 μM), after which time, protein was extracted and resolved by western blotting and probed for either cleaved PARP (cl-PARP), phospho-ERK (pERK), cyclin D1 (Cyclin D1), p27 or cleaved caspase-3 (cl-casp-3). Blots were stripped once and probed for actin to show equal protein loading. (D) Combined BRAF and MEK inhibition leads to enhanced apoptosis. WM164 cells were treated with either vehicle, U0126 (3 μM, 3U0), PLX4720 (3PLX, 3 μM) or the two inhibitors in combination for 48 h. Levels of apoptosis were measured by annexin-V staining and flow cytometry. Data show the mean of three experiments. *P<0.05.

Figure 6

Dual BRAF/MEK inhibition prevents escape from PLX4720 therapy. (A) WM164, WM793 and 1205Lu melanoma cells were treated with vehicle (1 week), PLX4720 (3 μM), U0126 (3 μM) or the two inhibitors in combination (both 3 μM) for 4 weeks. After this time, colonies were fixed and stained with crystal violet. Photographs are representative of three independent experiments. Photomicrographs show the detail of one colony each on the WM793 plate ( × 4). (B) Combined PLX4720 and U0126 treatment reduce growth of melanoma cells and enhance cell death in a 3D spheroid model. WM164 spheroids were implanted in collagen and treated with PLX4720 (3 μM), U0126 (3 μM) or the two drugs in combination for 72 h. After this time, plates were washed and cells were stained with a cell viability kit. Red=dead cells, green=live cells. The colour reproduction of this figure is available on the html full text version of the manuscript.