Dependence of p53-deficient cells on the DHX9 DExH-box helicase

Abstract

DHX9 is a DExH-box helicase family member with key regulatory roles in a broad range of cellular processes. It participates at multiple levels of gene regulation, including DNA replication, transcription, translation, RNA transport, and microRNA processing. It has been implicated in tumorigenesis and recent evidence suggests that it may be a promising chemotherapeutic target. Previous studies have determined that DHX9 suppression elicits an apoptotic or senescence response by activating p53 signaling. Here, we show that DHX9 inhibition can also have deleterious effects in cells lacking functional p53. Loss of DHX9 led to increased cell death in p53-deficient mouse lymphomas and HCT116 human colon cancer cells, and G0/G1 cell cycle arrest in p53-deficient mouse embryonic fibroblasts. Analysis of mRNA levels for p53 transcriptional targets showed that a subset of p53 targets in the p53-null lymphomas and HCT116 cells were activated despite the absence of functional p53. This implies an alternative pathway of DHX9-mediated activation of cell death and cell cycle arrest in p53-deficient cells and supports the feasibility of targeting DHX9 in p53-deficient tumors.

Keywords: DHX9; apoptosis; drug target; helicase; p53.

Figure 1. DHX9 suppression reduces cellular fitness in both p53-wildtype and p53-null systems

Ex vivo competition assay with (A) TSC2^+/−Eμ-Myc (p53^+/+) and p53^−/− Eμ-Myc lymphomas, (B) INK4A^−/− (p53^+/+) and p53^−/− MEFs, and (C) HCT116 p53^+/+ and HCT116 p53^−/− cells. Cells were infected with shRNAs targeting DHX9 or a neutral control (shRLuc.713) and the relative %GFP monitored over time. The experiment was started 48 hours after the final infection (t = Day 0). N = 3 ± SEM. (D–F) Western blot analysis of extracts from the indicated cell lines. Membranes are probed with antibodies indicated to the left. Solid bar indicates that a different set of Western blots were probed.

Figure 2. Context-dependent consequences of DHX9 suppression

PI staining of (A) TSC2^+/−Eμ-Myc (p53^+/+) and p53^−/− Eμ-Myc lymphomas, (B) INK4A^−/− (p53^+/+) and p53^−/− MEFs, and (C) HCT116 p53^+/+ and HCT116 p53^−/− cells expressing the indicated shRNAs 7 days post-infection. N = 3 ± SEM. ^#p ≤ 0.05, ^§p ≤ 0.01, *p ≤ 0.005, ^##p ≤ 0.001, ^§§p ≤ 0.0005, ^*p ≤ 0.0001, NS – not significant. Cell cycle analysis of (D) TSC2^+/−Eμ-Myc (p53^+/+) and p53^−/− Eμ-Myc lymphomas, (E) INK4A^−/− (p53^+/+) and p53^−/− MEFs, and (F) HCT116 p53^+/+ and HCT116 p53^−/− cells expressing the indicated shRNAs 10 days post-infection. N = 3 ± SEM.

Figure 3. Consequences of DHX9 knockdown on p53 targets in TSC2^+/−Eμ-Myc (p53^+/+) and p53^−/− Eμ-Myc lymphoma cells

(A) Quantitative RT-PCR analysis documenting DHX9 knockdown and p53 levels in TSC2^+/−Eμ-Myc (p53^+/+) and p53^−/− Eμ-Myc lymphomas. The indicated cell lines were transduced with control shRLuc.713 or DHX9 shRNAs and harvested 6 days post-infection. mRNA levels were normalized to GAPDH and the mRNA levels of the shDHX9 samples were then normalized to that of the shRLuc.713 sample for each cell line. N = 3 ± SEM. (B) Quantitative RT-PCR analysis of p53 transcriptional targets in TSC2^+/−Eμ-Myc (p53^+/+) and p53^−/− Eμ-Myc lymphomas. The analysis was performed 6 days post-transduction with control shRLuc.713 or DHX9 shRNAs. mRNA levels for each cell line and target were normalized as in (A). N = 3 ± SEM. Significant differences between shDHX9 and shRLuc.713 samples (where p ≤ 0.05) are indicated as follows: ^#p ≤ 0.05, ^§p ≤ 0.01, *p ≤ 0.005, ^##p ≤ 0.001, ^§§p ≤ 0.0005, ^*p ≤ 0.0001.

Figure 4. Consequences of DHX9 knockdown on p53 targets in INK4A^−/− (p53^+/+) and p53^−/− MEFs

(A) Quantitative RT-PCR analysis showing DHX9 knockdown and p53 levels in INK4A^−/− (p53^+/+) and p53^−/− MEFs. The indicated cell lines were transduced with control shRLuc.713 or DHX9 shRNAs and harvested 6 days post-infection. mRNA levels were normalized to GAPDH and the mRNA levels of the shDHX9 samples were then normalized to that of the shRLuc.713 sample for each cell line. N = 3 ± SEM. (B) Quantitative RT-PCR analysis of p53 transcriptional targets in INK4A^−/− (p53^+/+) and p53^−/− MEFs. The analysis was performed 6 days post-transduction with control shRLuc.713 or DHX9 shRNAs. mRNA levels for each cell line and target were normalized as in (A). N = 3 ± SEM. Significant differences between shDHX9 and shRLuc.713 samples (where p ≤ 0.05) are indicated using the same key as in Figure 3.

Figure 5. Consequences of DHX9 knockdown on p53 targets in HCT116 p53^+/+ and p53^−/− cells

(A) Quantitative RT-PCR analysis showing DHX9 knockdown and p53 levels in HCT116 p53^+/+ and p53^−/− cells. The indicated cell lines were transduced with control shFLuc.1309 or DHX9 shRNAs and harvested 6 days post-infection. mRNA levels were normalized to GAPDH and the mRNA levels of the shDHX9 samples were then normalized to that of the shFLuc.1309 sample for each cell line. N = 3 ± SEM. (B) Quantitative RT-PCR analysis of p53 transcriptional targets in HCT116 p53^+/+ and p53^−/− cells. The analysis was performed 6 days post-transduction with control shFLuc.1309 or DHX9 shRNAs. mRNA levels for each cell line and target were normalized as in (A). N = 3 ± SEM. Significant differences between shDHX9 and shFLuc.1309 samples (where p ≤ 0.05) are indicated using the same key as in Figure 3.

Figure 6. Consequences of DHX9 knockdown on protein levels of p53 targets in HCT116 p53^+/+ and p53^−/− cells

Western blot analysis of p53 transcriptional targets in HCT116 p53^+/+ and p53^−/− cells. Cells were harvested 6 days post-transduction and extracts were fractionated on (A) 15% and (B) 8% SDS-PAGE gels. In each panel ((A) and (B)), all probings were performed on the same blot. Actin and eEF2 are used as loading controls. Quantitation of intensity levels of the proteins in the shDHX9 samples relative to the shFLuc.1309 samples are indicated beneath each band.