Inhibition of the PI3K but not the MEK/ERK pathway sensitizes human glioma cells to alkylating drugs

Abstract

Background: Intrinsic chemoresistance of glioblastoma (GBM) is frequently owed to activation of the PI3K and MEK/ERK pathways. These signaling cascades are tightly interconnected however the quantitative contribution of both to intrinsic resistance is still not clear. Here, we aimed at determining the activation status of these pathways in human GBM biopsies and cells and investigating the quantitative impact of both pathways to chemoresistance.

Methods: Receptor tyrosine kinase (RTK) pathways in temozolomide (TMZ) treatment naive or TMZ resistant human GBM biopsies and GBM cells were investigated by proteome profiling and immunoblotting of a subset of proteins. Resistance to drugs and RTK pathway inhibitors was assessed by MTT assays. Apoptotic rates were determined by Annexin V staining and DNA damage with comet assays and immunoblotting.

Results: We analyzed activation of RTK pathways by proteome profiling of tumor samples of patients which were diagnosed a secondary GBM and underwent surgery and patients which underwent a second surgery after TMZ treatment due to recurrence of the tumor. We observed substantial activation of the PI3K and MEK/ERK pathways in both groups. However, AKT and CREB phosphorylation was reduced in biopsies of resistant tumors while ERK phosphorylation remained unchanged. Subsequent proteome profiling revealed that multiple RTKs and downstream targets are also activated in three GBM cell lines. We then systematically describe a mechanism of resistance of GBM cell lines and human primary GBM cells to the alkylating drugs TMZ and cisplatin. No specific inhibitor of the upstream RTKs sensitized cells to drug treatment. In contrast, we were able to restore sensitivity to TMZ and cisplatin by inhibiting PI3K in all cell lines and in human primary GBM cells. Interestingly, an opposite effect was observed when we inhibited the MEK/ERK signaling cascade with two different inhibitors.

Conclusions: Temozolomide treatment naive and TMZ resistant GBM biopsies show a distinct activation pattern of the MEK/ERK and PI3K signaling cascades indicating a role of these pathways in resistance development. Both pathways are also activated in GBM cell lines, however, only the PI3K pathway seems to play a crucial role in resistance to alkylating agents and might serve as drug target for chemosensitization.

Keywords: Cisplatin; Drug resistance; Glioblastoma; PI3K; Temozolomide.

Fig. 1

The MEK/ERK and PI3K pathways are activated in GBM in vivo and in vitro. a Human proteome-profiler phospho-MAPkinase arrays of GBM tumor biopsies before and after TMZ treatment. Shown are biopsies derived from patients, who were newly diagnosed a secondary GBM and underwent surgery before TMZ treatment (P1–P4) and of patients which underwent a second surgery after TMZ treatment due to recurrence of the tumor (P5–P8). Phosphorylated ERK2 (T185/Y187, AKTpan (S473, S474, S472) and CREB (S133) were detected in all biopsies. b Western blotting reveals that EGF-R, IGF1-R, and PDGF-Rβ RTKs are constitutively overexpressed. β-actin Western blot was performed to control for loading (left panel). Human proteome-profiler phospho-antibody arrays were used to assess the activation status of RTKs (middle panel) and downstream targets ERK2 (T185/Y187), AKTpan (S473, S474, S472) and CREB (S133) (right panel)

Fig. 2

GBM cell lines show high intrinsic tolerance to alkylating cytostatics cisplatin and TMZ. a Cisplatin resistance status of U87, U251, and U373 cell lines was assessed by MTT assays (n = 4). b Intrinsic resistance of GBM cell lines is demonstrated in comparison to cell pairs of non-resistant and cells with acquired resistance. GBM cell line IC₅₀ concentrations of cisplatin parallel the ones of resistant MCF-7 breast-, and A2780 ovarian cancer cell lines (determined by MTT assays; n = 3). c GBM cell lines were treated by weekly cycles of cisplatin and IC₅₀ values were assayed after each treatment cycle in MTT assays. IC₅₀ values were obtained from sigmoidal concentration–response curves and dotted in a time-dependent manner. d Western blots showing MRP-2 and P-gp expression in GBM cell lines untreated and after 24 cycles of weekly intermittent cisplatin exposure. β-actin Western blot was performed to control for loading. e TMZ resistance status in U87, U251 and U373 cells was assessed by MTT assays (upper graph; n ≥ 6). The respective IC₅₀ values for each cell line are shown in the table below the graph

Fig. 3

PI3K as a key component of TMZ resistance in GBM cells. a PI3K inhibition (20 µM LY294002) sensitizes all GBM cells to TMZ (1000 µM) (n = 3). b MTT concentration response curve showing concentration dependency of sensitization by LY294002 in U251 cells (upper graph). Respective TMZ IC₅₀ values at increasing LY294002 concentrations are displayed in the table below the graph (n ≥ 3). c Both, LY294002 (5 µM) and Wortmannin (30 nM), sensitize GBM cell lines to cisplatin. Presented IC₅₀ values were derived from MTT assays (n = 3). *P < 0.05, ***P < 0.001

Fig. 4

LY290042 increases apoptosis and DNA damage of TMZ treated U251 cells. a Representative FACS blots of U251 cells treated with either vehicle (DMSO), TMZ (1000 µM), LY294002 (20 µM) or the combination for 72 h. Annexin V positive cells were regarded as apoptotic cells. b Quantification of apoptotic cells from Annexin V assays (n = 5). c DNA damage was assayed by comet assay and is displayed as tail moment in U251 cells treated with LY290042, TMZ or vehicle (DMSO) for 1 h as indicated. d Representative Western blots showing P–P53 (S15), P53 and P21 expression in U251 cells treated with LY290042 (20 µM), TMZ (1000 µM) or vehicle (DMSO) for 24 h as indicated. β-actin Western blot was performed to control for loading (left panels). Densitometric analysis of the respective (phospho-) proteins relative to β-actin (right bars, n = 3). *P < 0.05, ***P < 0.001

Fig. 5

MEK inhibition negatively impacts on intrinsic chemoresistance. MTT assays showing that a U0126 (5 and 20 µM) and b PD98059 (5 and 20 µM) desensitize U251 cells for TMZ treatment (n = 3). c U0126 (5 and 20 µM) and d PD98059 (5 and 20 µM) lead to a strong right shift of TMZ concentration response curves in U373 cells (n = 3). e U0126 (10 µM) and PD98059 (20 µM) do not sensitize GBM cell lines to cisplatin. Presented IC₅₀ values were derived from MTT assays (n = 3)

Fig. 6

PI3K but not MEK inhibition sensitizes human primary GBM cells to TMZ. a PI3K inhibition (20 µM LY294002) leads to a left shift of the TMZ concentration response curve (n = 3). b Neither, U0126 (5 and 20 µM; n ≥ 3) c nor PD98059 (5 and 20 µM; n ≥ 3) sensitize primary cells to TMZ. d Derived mean IC₅₀ values for TMZ of respective primary GBM cells treated with LY294002 and MEK inhibitors (20 µM each; n ≥ 3). *P < 0.05