A hormone-DNA repair circuit governs the response to genotoxic insult

Abstract

Alterations in DNA repair promote tumor development, but the impact on tumor progression is poorly understood. Here, discovery of a biochemical circuit linking hormone signaling to DNA repair and therapeutic resistance is reported. Findings show that androgen receptor (AR) activity is induced by DNA damage and promotes expression and activation of a gene expression program governing DNA repair. Subsequent investigation revealed that activated AR promotes resolution of double-strand breaks and resistance to DNA damage both in vitro and in vivo. Mechanistically, DNA-dependent protein kinase catalytic subunit (DNAPKcs) was identified as a key target of AR after damage, controlling AR-mediated DNA repair and cell survival after genotoxic insult. Finally, DNAPKcs was shown to potentiate AR function, consistent with a dual role in both DNA repair and transcriptional regulation. Combined, these studies identify the AR-DNAPKcs circuit as a major effector of DNA repair and therapeutic resistance and establish a new node for therapeutic intervention in advanced disease.

Significance: The present study identifies for the fi rst time a positive feedback circuit linking hormone action to the DNA damage response and shows the significant impact of this process on tumor progression and therapeutic response. These provocative findings provide the foundation for development of novel nodes of therapeutic intervention for advanced disease.

Figure 1

Androgen ablation cooperates with IR in HT-sensitive PCa. (A) Top: Cells were cultured in hormone proficient (FBS, full serum control) media for 24 hours then treated with 2 Gy IR, steroid deprived conditions (ADT, androgen deprivation therapy), or a concurrent combination of ADT and 2 Gy IR. Cell number was determined on days 6 and 10 post-treatment and set relative to day 1 (control, untreated). Bottom: Relative cell doubling time was calculated on day 10 for each treatment using the formula T_d= (t₂−t₁)*(ln(2)/ln(q₂/q₁)), t=time and q=quantity. (B) Top: Schematic representing xenograft treatment cohorts. Bottom: LNCaP cells were injected into the flanks of nude mice and randomized into one of the four treatment arms as shown. Tumor volume was determined periodically and relative volume reported. *<0.05 p value compared to all other treatment conditions.

Figure 2

AR suppression sensitizes CRPC to DNA damage. (A) Left: Cells were cultured in hormone proficient media for 24 hours then treated with 2 Gy IR, ADT, or a concurrent combination of ADT and 2 Gy IR. Cell number was determined on days 6 and 10 post-treatment and set relative to day 1 control. Right: Relative cell doubling time was calculated on day 10 for each treatment. (B) Cells were treated as per panel A, but treatments were 2 Gy IR, concurrent ADT and 2 Gy IR, or concurrent ADT supplemented with 0.1nM DHT and 2 Gy IR. (C) Cells were treated as per panel A, but treatments were 2 Gy IR, 1uM MDV3100, or concurrent 1uM MDV3100 and 2 Gy IR. (D) Top: Schematic representing xenograft treatment cohorts. Bottom: C4-2 cells were injected into the flanks of nude mice and randomized into one of the four treatment arms as shown. Tumor volume was determined periodically and relative volume reported. **<0.01 p value compared to all other treatment conditions.

Figure 3

DNA damage induces AR activity. (A) C4-2 cells were cultured in hormone proficient media for 24 hours then treated with 2 Gy IR. (B, C) C4-2 cells were cultured in hormone proficient media for 24 hours then treated with 2 Gy IR, 100 J/m² UV, 1uM doxorubicin, or 1 hour pre-treatment of 1uM N-acetylcysteine followed by 2 Gy IR. (D) Left: C4-2 cells were cultured in hormone deficient or hormone deficient supplemented with 0.1nM DHT media for 24 hours then treated as indicated. Right: C4-2 cells were cultured in hormone proficient media for 24 hours then treated with 1uM MDV3100, concurrent 1uM MDV3100 and 2 Gy IR, or concurrent 1uM MDV3100, 2 Gy IR, and 0.1nM DHT. Relative expression of indicated transcript levels were analyzed and normalized to GAPDH mRNA then set relative to untreated control at indicated time points post treatment. *<0.05; **<0.01 p value.

Figure 4

AR promotes DNA double strand break resolution. (A) C4-2 cells were cultured in hormone proficient media for 24 hours then treated as indicated. At indicated time points cells were pre-treated with BrdU for 2 hours then fixed. Cells were incubated with BrdU antibody and propidium iodide and processed for FACS analysis. Data is graphed as normalized to untreated control. (B, C) C4-2 cells were cultured in hormone proficient media for 24 hours then treated as indicated. Cells were fixed at the indicated time points, stained with α-γH2AX and α-53BP1 antibodies, and imaged by confocal microscopy. Foci were counted and plotted as average foci per cell set relative to control at each time point. (D) C4-2 cells were cultured in hormone proficient, hormone deficient, or hormone deficient supplemented with 0.1nM DHT media for 24 hours then treated with 15 Gy IR and harvested at indicated timepoints. Single-cell gel electrophoresis was performed and tail moments assessed. *<0.05; **<0.01 p value.

Figure 5

AR activity regulates genes required for DNA damage repair. (A) Left: Schematic showing experimental design. Right: Array results depicting genes regulated by androgen after 2 Gy IR in C4-2 and 22Rv1 cells. (B–D) C4-2 cells were cultured in hormone proficient, hormone deficient, or hormone deficient supplemented with 0.1nM DHT media for 24 hours then treated as indicated. Samples were harvested for ChIP-qPCR analysis and percent (input) occupancy of AR set relative to control at each time point is reported. Arrows and numbers on diagrams depict location of identified AROR in relation to transcriptional start site (TSS).

Figure 6

AR induces DNAPKcs expression and activity. (A, B) C4-2 cells were cultured in hormone proficient, hormone deficient, or hormone deficient supplemented with 0.1nM DHT media for 24 hours then treated. Cells were harvested and expression of DNAPKcs was analyzed and quantified. Representative image of at least 3 independent experiments is shown. (C–D) Same as (A, B), but expression of phospho (Ser2056)-DNAPKcs was analyzed and quantified. DNAPKcs specific activity was determined by dividing phospho-DNAPKcs by total DNAPKcs. (E) C4-2 cells were transfected with a pool of siRNAs directed against the XRCC6 transcript or control siRNA for 96 hours then treated with 2 Gy IR. Cells were harvested and expression of DNAPKcs, phospho-DNAPKcs, and Ku70 analyzed and quantified. *<0.05 p value.

Figure 7

DNAPK is required for DNA repair in the presence of androgen. (A) C4-2 cells were transfected with a pool of siRNAs directed against the PRKDC transcript or control siRNA for 96 hours then treated with 2 Gy IR. Cells were fixed at the indicated timepoints, stained with α-γH2AX and α-53BP1 antibodies, and analyzed by confocal microscopy. Foci were counted and plotted as average foci per cell set relative to control at each time point. (B) Same as (A), but cells were counted 5 days post treatment and cell number set relative to control. (C) Same as (A), but relative expression of transcript levels were analyzed and normalized to GAPDH mRNA at 24 hours post treatment (D) C4-2 cells were cultured in hormone proficient media for 24 hours then treated with combinations of DHT (1 hour pre-treatment), 2 Gy IR, and 1uM DNAPKcs inhibitor. Cells were fixed at the indicated timepoints, stained with α-γH2AX and α-53BP1 antibodies, and analyzed by confocal microscopy. Foci were counted and plotted as average foci per cell set relative to control at each time point. (E) Same as (D), but cells were counted 5 days post treatment and cell number set relative to control. (F) C4-2 cells transfected with linearized pEYFP plasmid were treated as indicated and activity assessed. (G) Model describing the feedback circuit of AR and DNAPKcs expression and activity controlling DNA repair and tumor cell survival *<0.05; **<0.01 p value.