Simultaneous perturbation of the MAPK and the PI3K/mTOR pathways does not lead to increased radiosensitization

Abstract

Background: The mitogen-activated protein kinases (MAPK) and the phosphatidylinositol-3-kinase (PI3K)/mammalian target of rapamycin (mTOR) pathways are intertwined on various levels and simultaneous inhibition reduces tumorsize and prolonges survival synergistically. Furthermore, inhibiting these pathways radiosensitized cancer cells in various studies. To assess, if phenotypic changes after perturbations of this signaling network depend on the genetic background, we integrated a time series of the signaling data with phenotypic data after simultaneous MAPK/ERK kinase (MEK) and PI3K/mTOR inhibition and ionizing radiation (IR).

Methods: The MEK inhibitor AZD6244 and the dual PI3K/mTOR inhibitor NVP-BEZ235 were tested in glioblastoma and lung carcinoma cells, which differ in their mutational status in the MAPK and the PI3K/mTOR pathways. Effects of AZD6244 and NVP-BEZ235 on the proliferation were assessed using an ATP assay. Drug treatment and IR effects on the signaling network were analyzed in a time-dependent manner along with measurements of phenotypic changes in the colony forming ability, apoptosis, autophagy or cell cycle.

Results: Both inhibitors reduced the tumor cell proliferation in a dose-dependent manner, with NVP-BEZ235 revealing the higher anti-proliferative potential. Our Western blot data indicated that AZD6244 and NVP-BEZ235 perturbed the MAPK and PI3K/mTOR signaling cascades, respectively. Additionally, we confirmed crosstalks and feedback loops in the pathways. As shown by colony forming assay, the AZD6244 moderately radiosensitized cancer cells, whereas NVP-BEZ235 caused a stronger radiosensitization. Combining both drugs did not enhance the NVP-BEZ235-mediated radiosensitization. Both inhibitors caused a cell cycle arrest in the G1-phase, whereas concomitant IR and treatment with the inhibitors resulted in cell line- and drug-specific cell cycle alterations. Furthermore, combining both inhibitors synergistically enhanced a G1-phase arrest in sham-irradiated glioblastoma cells and induced apoptosis and autophagy in both cell lines.

Conclusion: Perturbations of the MEK and the PI3K pathway radiosensitized tumor cells of different origins and the combination of AZD6244 and NVP-BEZ235 yielded cytostatic effects in several tumor entities. However, this is the first study assessing, if the combination of both drugs also results in synergistic effects in terms of radiosensitivity. Our study demonstrates that simultaneous treatment with both pathway inhibitors does not lead to synergistic radiosensitization but causes cell line-specific effects.

Fig. 1

Proliferation assays of SNB19 and A549 after treatment AZD6244 and/or NVP-BEZ235. Effect of 24 h exposure to serial dilutions of AZD6244 (blue squares), AZD6244 with 50 nM NVP-BEZ235 (red triangels), NVP-BEZ235 (green circles) or NVPBEZ235 with 500 nM AZD6244 (black diamonds) on the ATP level in SNB19 or A549 cells, as measured by a standard luciferase assay. The diagram represents the means of two independent experiments, each performed in triplicates, normalized against DMSO-treated controls. Dose–response curves (illustrated in the corresponding color) were generated using the standard four parameter logistic models and error bars indicate SD values

Fig. 2

Expression levels of selected proteins of the MAPK/Erk and the PI3K/mTOR pathways in SNB19 and A549 cells. Representative Western blot analysis of expression levels of selected proteins of the MAPK a and the PI3K/mTOR pathway b in SNB19 and A549 cells. Cells were treated with AZD6244 and/or NVP-BEZ235 before IR with 8 Gy and whole cell lysates were prepared 30 min, 24 h and 48 h after IR. Protein bands were normalized to the β-actin intensity and changes in protein expression are denoted by numbers if applicable. Since the utilized antibody detected p-Erk1 and p-Erk2 the numbers below the blot correspond to the expression levels of total p-Erk1/2

Fig. 3

Putative interactions and feedback loops in the MAPK and PI3K/mTOR signaling cascades. Putative signaling diagram of the MAPK and PI3K/mTOR pathways derived from the Western blot analysis shown Fig. 2 and literature data specified in the text. Treatment with the allosteric MEK inhibitor AZD6244 causes desphosphorylation of Erk resulting in phenotypic changes (e.g. changes in the proliferation rate, cell cycle phase distribution and radiosensitivity). However, simultaneously the MAPK pathway is activated through a feedback loop from Erk to the upper part of the signaling cascade resulting in hyperphosphorylation of MEK. Inhibiting the PI3K/mTOR pathway with NVP-BEZ235 caused a dephosphorylation of the transcription and translation regulators S6 and 4E-BP1 also inducing the aforementioned phenotypic changes. However, also for the PI3K/mTOR pathway we observed a feedback loop, as indicated by increased Akt phosphorylation after PI3K and mTOR inhibition in SNB19 cells. Furthermore, we validated a crosstalk between the two signaling cascades since treatment with the dual PI3K/mTOR inhibitor also resulted in decreased Raf-1 expression levels. Stimulation is indicated by black normal arrows, whereas inhibition is indicated by red blunt arrows

Fig. 4

Colony forming abilities of SNB19 and A549 cancer cells as functions of drug and IR exposure. Control (DMSO-treated, empty circles), AZD6244- (filled squares), NVP-BEZ235- (filled triangles) and AZD6244 + NVP-BEZ235-treated cells (filled diamonds) were irradiated with single graded doses up to 8 Gy. Two weeks after IR colonies were fixed and stained using standard protocols. Experiments were performed in triplicates and repeated at least four times. Colonies containing at least 50 cells were scored as survivors

Fig. 5

Effects of AZD6244, NVP-BEZ235 and IR on cell cycle phase distributions. Cell cycle phase distribution of SNB19 and A549 tumor cells treated with AZD6244 and/or NVP-BEZ235 before IR with 8 Gy. The cells were fixed 30 min, 24 or 48 h after IR, permeabilized, treated with RNase, stained with PI and analyzed for DNA content by flow cytometry. Data are presented as means (±SD) of G1- (light grey), S- (white) and G2/M-phase (dark grey) cells of at least three independent experiments for each cell line. Statistical significant changes assessed as specified in the Methods section are indicated by # P < 0.05

Fig. 6

Effects of AZD6244, NVP-BEZ235 and IR on the induction of apoptosis and autophagy. Mean percentage of cells with hypodiploid DNA content and cellular debris in cells treated with AZD6244 (red striped columns), NVP-BEZ235 (blue columns) or a combination of both inhibitors (red and blue striped columns) 24 and 48 h after IR. DMSO treated cells (blank columns) served as controls. Cells were detached with trypsin, fixed, permeabilized, treated with RNase, stained with PI and then analyzed for fluorescence by flow cytometry. The columns display means (±SD) of hypodiploid cells and cellular debris of at least three independent experiments. Statistical significant differences are indicated as follows: # P < 0.05; ## P <0.01. Statistical significant differences between sham-irradiated and irradiated control samples are not depicted for reasons of clarity. Representative Western blot analysis of expression levels of PARP, cleaved PARP, LC3-I and LC3-II in SNB19 and A549 cells c. Cells were treated with AZD6244 and/or NVP-BEZ235 before IR with 8 Gy and whole cell lysates were prepared 24 and 48 h after IR as described previously. Protein bands were normalized to the β-actin intensity and changes in protein expression are denoted by numbers if applicable. The experiment was repeated at least three times

Fig. 7

Effects of AZD6244, NVP-BEZ235 and IR on cell cycle related proteins. Representative Western blot analysis of expression levels of selected proteins associated with the cell cycle in SNB19 and A549 cells. Cells were treated with AZD6244 and/or NVP-BEZ235 before IR with 8 Gy and whole cell lysates were prepared 30 min a, 24 b and 48 h c after IR as described previously. Protein bands were normalized to the β-actin intensity and changes in protein expression are denoted by numbers if applicable. The experiment was repeated at least three times