Survival advantage of AMPK activation to androgen-independent prostate cancer cells during energy stress

Abstract

Androgen-independent prostate cancer usually develops as a relapse following androgen ablation therapy. Removing androgen systemically causes vascular degeneration and nutrient depletion of the prostate tumor tissue. The fact that the malignancy later evolves to androgen-independence suggests that some cancer cells are able to survive the challenge of energy/nutrient deprivation. AMP-activated protein kinase (AMPK) is an important manager of energy stress. The present study was designed to investigate the role of AMPK in contributing to the survival of the androgen-independent phenotype. Most of the experiments were carried out in the androgen-dependent LNCaP cells and the androgen-independent C4-2 cells. These two cell lines have the same genetic background, since the C4-2 line is derived from the LNCaP line. Glucose deprivation (GD) was instituted to model energy stress encountered by these cells. The key findings are as follows. First, the activation of AMPK by GD was much stronger in C4-2 cells than in LNCaP cells, and the robustness of AMPK activation was correlated favorably with cell viability. Second, the response of AMPK was specific to energy deficiency rather than to amino acid deficiency. The activation of AMPK by GD was functional, as demonstrated by appropriate phosphorylation changes of mTOR and mTOR downstream substrates. Third, blocking AMPK activation by chemical inhibitor or dominant negative AMPK led to increased apoptotic cell death. The observation that similar results were found in other androgen-independent prostate cancer cell lines, including CW22Rv1 abd VCaP, provided further assurance that AMPK is a facilitator on the road to androgen-independence of prostate cancer cells.

Fig. 1

Effect of glucose or amino acid deprivation on cell survival and AMPK activation in LNCaP and C4-2 cells. (A) MTT cell survival data following glucose deprivation. The results (mean ± SD, n=3) are expressed as % surviving cells compared to the value of the untreated control, which is set as 100%. *significantly different from each other, P<0.05. (B) Western blot of phosphorylated or total AMPK in cells following glucose deprivation (GD). (C) MTT cell growth data following amino acid deprivation (AAD). (D) Western blot of phosphorylated or total AMPK following AAD.

Fig. 2

Effect of glucose deprivation (GD) on phospho-ACC, phospho-mTOR, phospho-S6 and phospho-p70S6K. The substrates in (A) are the direct targets of AMPK, while the substrates in (B) are the direct targets of mTOR.

Fig. 3

Effect of Compound C on AMPK activation (A) and cell survival (B) in LNCaP and C4-2 cells subjected to glucose deprivation. The cell survival data were obtained at day 3. The results from Compound C-treated cells are expressed as a percentage of the value from DMSO (vehicle) treated cells (set as 100%). *significantly different from each other, P<0.05.

Fig. 4

Effect of dominant negative AMPK (DN-AMPKα) on AMPK activity (A) and cell death (B) in C4-2 cells following glucose deprivation (GD). (A) Western blot of phospho-AMPK, phospho-ACC, and phospho-mTOR. (B) Cell death data as determined by flow cytometric analysis of propidium iodide-stained cells. Effect of AMPK knockdown on cell death is expressed as fold of increase compared to the value observed in the absence of knockdown. *significantly different than the GD only value, P<0.05. (C) Western blot of phospho-AMPK, total AMPK and myc of mycDN-AMPKα transfected cells.

Fig. 5

Effect of AMPK-siRNA on AMPK activity (A) and cell death (B) in C4-2 cells following glucose deprivation (GD). (A) Western blot of phospho-AMPK, phospho-ACC, and phospho-mTOR. (B) Cell death data as determined by trypan blue exclusion analysis. Effect of AMPK knockdown on cell death is expressed as percent of trypan blue positive cells. *significantly different than the GD only value, P<0.05.

Fig. 6

Effect of glucose deprivation (GD) on cell survival and cell death in CW22Rv1 and VCaP cells with or without AMPK activation knockdown by DN-AMPKα. (A) MTT cell survival data after 5 days of GD. The results are expressed as % surviving cells compared to the value of the untreated control, which is set as 100%. *significantly different than the LNCaP value, P<0.05. (B) Cell death data as determined by flow cytometric analysis of propidium iodide-stained cells. The effect of AMPK activation knockdown on cell death is expressed as fold of increase compared to the value observed in the absence of knockdown. *significantly different than the GD only value, P<0.05. (C) Confirmation of decreases of phospho-AMPKα by DN-AMPKα.

Fig. 7

Effect of AMPK knockdown by DN-AMPKα on cell death induction following glucose deprivation (GD). (A) LDH release as an indicator of necrotic cell death. The data are expressed as fold of increase compared to the value from untreated control cells (set as 1). (B) PARP cleavage as an indicator of apoptotic cell death. The Western blot shows both full length and cleaved PARP with or without DN-AMPKα transfection.