Antagonizing Glutamine Bioavailability Promotes Radiation Sensitivity in Prostate Cancer

Abstract

Nearly half of localized prostate cancer (PCa) patients given radiation therapy develop recurrence. Here, we identified glutamine as a key player in mediating the radio-sensitivity of PCa. Glutamine transporters and glutaminase are upregulated by radiation therapy of PCa cells, but respective inhibitors were ineffective in radio-sensitization. However, targeting glutamine bioavailability by L-asparaginase (L-ASP) led to a significant reduction in clonogenicity when combined with irradiation. L-ASP reduced extracellular asparagine and glutamine, but the sensitization effects were driven through its depletion of glutamine. L-ASP led to G2/M cell cycle checkpoint blockade. As evidence, there was a respective delay in DNA repair associated with RAD51 downregulation and upregulation of CHOP, contributing to radiation-induced cell death. A radio-resistant PCa cell line was developed, was found to bypass radiation-induced mitotic catastrophe, and was sensitive to L-ASP/radiation combination treatment. Previously, PCa-associated fibroblasts were reported as a glutamine source supporting tumor progression. As such, glutamine-free media were not effective in promoting radiation-induced PCa cell death when co-cultured with associated primary fibroblasts. However, the administration L-ASP catalyzed glutamine depletion with irradiated co-cultures and catalyzed tumor volume reduction in a mouse model. The clinical history of L-ASP for leukemia patients supports the viability for its repurposing as a radio-sensitizer for PCa patients.

Keywords: asparaginase; glutamine; prostate cancer; radiation therapy.

Figure 1

Glutamine is a conditionally essential amino acid for prostate cancer. (A) Fold change of SLC1A5, SLC38A1, SLC38A2, and SLC38A7 in benign and prostate cancer patients in the gse29079 data set, obtained from R2-Genomics analysis (n = 95). (B) 22RV1 cells were counted 72 h after culturing in media containing the indicated concentrations of glutamine (L-Gln). (C) ARCaP_M cells were counted 72 h after culturing in media containing the indicated concentrations of L-Gln. (D) 22RV1 were irradiated (4 Gy) and counted 72 h after culturing in media containing L-Gln (0 or 2 mM), with and without administration of GPNA (100 µM) or CB839 (1 µM). Cell counts are reported as a mean ± SD of at least three biologic replicates (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, compared to control).

Figure 2

L-ASP reduces viability of PCa cells through depletion of glutamine, not asparagine. (A) 22RV1 and ARCaP_M cultured cells were counted following the indicated dose of irradiation in the context of L-Gln (0 or 2 mM) and/or L-ASP (0 or 2 IU/mL). (B) Clonogenic assays with 22RV1 cells were quantitated following 0 or 2 Gy of radiation, in the presence of increasing levels of L-Gln (0. 0.5, 2, 4, and 10) or 2 mM L-Gln with 2 IU/ mL L-ASP. (C) Clonogenic assays of 22RV1 and ARCaP_M cells were performed 10 days after administering untreated or L-ASP-treated media. L-Gln and L-Asn was added back to L-ASP-treated media. Cell counting shows short-term effects (72 h) under the same conditions. (D) Cell cycle analysis was performed on 22RV1 and ARCaP_M cells following incubation under control conditions or treatment with L-ASP-treated media for 72 h. (E) Graphic shows that L-ASP effects in PCa cells by the depletion of extracellular glutamine, not asparagine. Cell counts are reported as a mean ± SD (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, compared to control).

Figure 3

Glutamine depletion leads to cell cycle arrest and cell death. 22RV1 cells were pre-treated with L-ASP (2 IU/mL), enzalutamide (10 µM), or both for 24 h. Cells were harvested 4 or 24 h after the indicated doses of irradiation (IRR). (A) The mRNA expression of G1/S (CDK4, CDK6, and CCDN1) and G2/M checkpoint markers (p21, CDK1, CCNB1, and PCNA) was determined by RT-PCR. mRNA expression was normalized to ß-actin and untreated samples. (B) Propidium iodide-stained cells were subjected to cell cycle analysis by FACS 4 and 24 h after irradiation. (C) Apoptotic status of cells 4 and 24 h after irradiation was determined by FACS analysis of cell surface annexin V. All studies are reported as a mean of at least three independent experiments (* p < 0.05, ** p < 0.01, **** p < 0.0001).

Figure 4

L-ASP potentiates IR effects by delaying DNA damage repair. 22RV1 cells were pre-treated with L-ASP (2 IU/mL), enzalutamide (10 µM), or both for 24 h. Cells were harvested at the indicated times after irradiation (0 or 4 Gy). (A) Western blots of 22RV1 cell lysates were probed for PARP, cleaved PARP, ATF4, CHOP, p27, p21, survivin, and p-γ-H2Ax. Representative blots are shown with the mean relative quantitation indicated normalized to ß-actin expression. Molecular weights (kDa) are indicated. (B) mRNA expression of DNA damage repair genes was measured by RT-PCR. Data are reported as a mean of three independent experiments.

Figure 5

L-ASP sensitizes radio-resistant PCa to irradiation. (A) Representative images highlight morphological differences in parental ARCaP_M and ARCaP_M-IR cells; Scale bar, 250 μm. (B) Clonogenic assays performed with ARCaP_M and ARCaP_M-IR cells were quantified 10 days after treatment with L-ASP, enzalutamide, or both, with and without irradiation (2 Gy). Solubilized CFUs were normalized to the untreated control. (C) Parental ARCaP_M and ARCaP_M-IR cell were counted 72 h after administering L-ASP, enzalutamide, and/or irradiation (4 Gy). (D) Cell cycle analysis of ARCaP_M and ARCaP_M-IR 24 h after L-ASP, enzalutamide, and/or irradiation (4 Gy) was analyzed by FACS using propidium iodide staining. (E) ARCaP_M and ARCaP_M-IR cell surface CD44 expression was measured by FACS 24 h after L-ASP, enzalutamide, and/or irradiation (4 Gy). (F) mRNA expression of DNA damage repair proteins and reactive oxygen species (ROS) regulators were measured. All studies are reported as a mean of at least three independent experiments (* p < 0.05, ** p < 0.01, **** p < 0.0001).

Figure 6

PCa radio-resistance mediated by CAFs can be countered by administering L-ASP. (A) CAFs were counted after administration of L-ASP (2 IU/mL), CB839 (1 µM), or GPNA (100 µM) 24 h after irradiation (4 Gy). (B) CAFs were counted after incubation with untreated and L-ASP-treated media with or without supplementation of L-Gln, L-Asn, or both. (C) Cells treated under the same conditions were subjected to FACS analysis for annexin V expression. (D) Cell cycle analysis was performed on CAFs following incubation with untreated and L-ASP-treated media with or without supplementation of L-Gln, L-Asn, or both. (E) ARCaP_M cells co-cultured with CAFs in a trans-well were counted following incubation in L-ASP-treated media supplemented with enzalutamide (10 µM) and/or 2 mM L-Gln. Irradiation status of the co-cultures is indicated. (F) Schematic overview of xenograft experiment where athymic nude mice were grafted with ARCaP_M and CAFs in a 1:3 ratio subcutaneously. (G) Mice were administered L-ASP (125 IU) or vehicle prior to and after receiving irradiation (5 Gy) as indicated (n = 6). Tumor volume fold change was normalized to the first dose of L-ASP. Multiple comparison ANOVA was performed. (H) A graphic shows the revealed mechanism of how L-ASP acts as a radio-sensitizer in prostate tumors. All studies are reported as a mean of at least three independent experiments (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001).