Inhibition of the PI3K/AKT/mTOR pathway activates autophagy and compensatory Ras/Raf/MEK/ERK signalling in prostate cancer

Abstract

The PI3K/AKT/mTOR pathway is frequently activated in advanced prostate cancer, due to loss of the tumour suppressor PTEN, and is an important axis for drug development. We have assessed the molecular and functional consequences of pathway blockade by inhibiting AKT and mTOR kinases either in combination or as individual drug treatments. In established prostate cancer cell lines, a decrease in cell viability and in phospho-biomarker expression was observed. Although apoptosis was not induced, a G1 growth arrest was observed in PTEN null LNCaP cells, but not in BPH1 or PC3 cells. In contrast, when the AKT inhibitor AZD7328 was applied to patient-derived prostate cultures that retained expression of PTEN, activation of a compensatory Ras/MEK/ERK pathway was observed. Moreover, whilst autophagy was induced following treatment with AZD7328, cell viability was less affected in the patient-derived cultures than in cell lines. Surprisingly, treatment with a combination of both AZD7328 and two separate MEK1/2 inhibitors further enhanced phosphorylation of ERK1/2 in primary prostate cultures. However, it also induced irreversible growth arrest and senescence. Ex vivo treatment of a patient-derived xenograft (PDX) of prostate cancer with a combination of AZD7328 and the mTOR inhibitor KU-0063794, significantly reduced tumour frequency upon re-engraftment of tumour cells. The results demonstrate that single agent targeting of the PI3K/AKT/mTOR pathway triggers activation of the Ras/MEK/ERK compensatory pathway in near-patient samples. Therefore, blockade of one pathway is insufficient to treat prostate cancer in man.

Keywords: AKT; MAPK signalling; RAS; autophagy; mTOR; prostate cancer.

Figure 1. Cell viability decreases following inhibition of AKT and mTOR in patient-derived prostate cultures

(A) Primary cultures derived from cancers (n=5) and normal prostate (n=1) were treated with either 3 μM AZD7328, 3 μM KU-0063794 or a combination of 3 μM AZD7328 + 3 μM KU-0063794 in triplicate for 24 hours and 48 hours. Following treatment, the cells were harvested, stained with trypan blue and counted. The percentage of viable cells was calculated and normalized to the vehicle control (0.04% DMSO). Significant differences (p value < 0.05) in cell viability are indicated on the graphs. (B) Primary prostate cancer samples H252/12, H163/12, H149/12, and H135/11 were treated with 3 μM AZD7328, 3 μM KU-0063297 or a combination of 3 μM AZD7328 + 3 μM KU-0063794 for 72 hours. Following treatment, the cells were harvested, sorted into committed basal (CB), transit amplifying (TA) and stem-like cells (SC) and counted using trypan blue exclusion. Percentage of viable cells was calculated and normalized relative to the vehicle control (0.06% DMSO). Error bars represent the standard deviation. Significant difference (p <0.05) in cell viability was only observed in stem cell fraction treated with the combination of 3 μM AZD7328 + 3 μM KU-0063794 in comparison to vehicle control (indicated on the graph). (C) Cell cycle distribution remains unchanged in primary prostate cultures after treatment with AZD7328 and KU-0063794. Five prostate cultures (2 BPH and 3 cancers) were treated with 5 μM AZD7328, 5 μM KU-0063794 or 5 μM AZD7328 + 5 μM KU-0063794 for 72 hours. Following treatment, non-adherent cells were removed and remaining cells were harvested, fixed with 70% ice-cold ethanol, and stained with propidium iodide and analysed by flow cytometry. Control cells were treated with 0.1% DMSO. The signal was recorded in the PE channel and debris (subG1 phase) were excluded from the analysis.

Figure 2. Inhibition of AKT and mTOR kinase reduces phospho-biomarker expression and induces cell cycle arrest in LNCaP cells

(A) Prostate cell lines BPH1, LNCaP and PC3 were treated with either 1 γM AZD7328 or 1 γM KU-0063794 for 2 hours to measure acute response in change of biomarkers phosphorylation status. Following treatment, cells were lysed on ice, spun down and supernatant was collected. 20 γg of protein was loaded and run on 4-12% Bis-Tris polyacrylamide gel. Resolved protein was then electrotransferred to PVDF membranes and immunostained with antibodies against Phospho-AKT (S473), Phospho-FoxO1/FoxO3a (T24/T32), Phospho-PRAS40 (T246), Phospho-S6 (S235/236), total AKT, FoxO3a, total S6 and Vinculin, which was used as a loading control. Dashed line separating lane 1 (0.1% DMSO control) from the remaining 2 lanes indicates where images were cropped. (B) Cell cycle distribution was determined in BPH1, LNCAP and PC3 cells. Cells were treated with either 1 γM AZD7328 or 1 γM KU-0063794 for 72 hours. Following treatment, cells were washed with PBS, then harvested, fixed with 70% ice-cold ethanol, and stained with propidium iodide and analysed by flow cytometry. Control cells were treated with 0.1% DMSO. The signal was recorded in the PE channel, and debris were excluded from the analysis.

Figure 3. Autophagy is induced and phospho-ERK1/2 levels increase in patient-derived cultures after treatment with AKT inhibitor

(A) Phase contrast images of primary cells (H282/13) from a Gleason 7 (4+3) prostate cancer, treated with 0.5% DMSO (left), 25 μM AZD7328 (middle) and 25 μM KU-0063794 (right) for 72 hours. Increased vacuolization of AZD7328-treated cells and membrane blebbing of KU-0063794-treated cells are indicated (red arrows). Control cells were treated with 0.5% DMSO. (B) LC3 B expression increases in primary prostate cancer cells H310/13 (GL7) following treatment with AZD7328. Expression of the autophagy marker, an LC3 B, was determined using Immunofluorescence. The cells were seeded into collagen I-coated 8-well chamber slides and treated with either 25 μM AZD7328 or 25 μM KU-0063794 for 96 hours and then fixed in 4% PFA. Control cells were treated with 0.25% DMSO. Scale bar on the phase contrast images represents 200μm and on the fluorescent images 63 μm. (C) Phospho-ERK1/2 levels increase in prostate primary cultures following treatment with AZD7328 and KU-0063794. Primary BPH (H277/13) and cancer cells (H278/13) were treated with 5 μM AZD7328, 5 μM KU-0063794 or 5 μM AZD7328 + 5 μM KU-0063794 for 72 hours. Westerns were carried out and biomarkers detected (phospho-AKT (S473), total AKT, phospho-ERK1/2 (Thr202/Tyr204), total ERK1/2 and LC3 B). Vehicle control cells were treated with 0.1% DMSO. Staining with β-actin antibody was used as a loading control. The bands were quantified using Image J software and the expression normalized to the vehicle control; the values are shown below the blot. (D) Primary prostate cancer cells H282/13 were treated for 72 hours with up to 50 μM of AZD7328 and KU-0063794 or a 10 μM AZD7328 + 10 μM KU-0063794 combination of both. Westerns were carried out and biomarkers detected (phospho-AKT, total AKT, phospho-ERK1/2, total ERK1/2, phospho-S6, total S6, cleaved PARP and LC3 B). Vehicle control cells were treated with 0.5% DMSO. Staining with β-actin antibody was used as a loading control. The bands were quantified using Image J software and the expression normalized to β-actin; the values are shown below the blot.

Figure 4. Senescence is induced following combination treatment with AZD7328 and MEK1/2 inhibitors

Primary cancer cultures were treated with a range of concentrations of AZD6244, AZD7328 or combinations of AZD6244 + AZD7328 (A) or RO-512 + AZD7328 (B) for 72 hours. Westerns were carried out and biomarkers detected (phospho-AKT, total AKT, phospho-ERK1/2 and total ERK1/2). Vehicle control cells were treated with 0.5% DMSO (A) or 0.2 % DMSO (B). The bands were quantified using Image J software and the expression normalized to β-actin; the values are shown below the blot. (C) Primary BPH culture H373/13 was treated with either 1 μM AZD6244, 1 μM RO-512 or a combination of 1 μM AZD6244 + 1 μM AZD7328 or 1 μM RO-512 + 1 μM AZD7328 for 72 hours. Following treatment, cells were fixed and stained according to the manufacturer's instructions and phase contrast images were taken. β-galactosidase – positive cells were observed (red arrows) as well as β-galactosidase negative cells (green arrows). Vehicle control cells were treated with 0.02% DMSO. Scale bar represents 400 μm.