Genotoxic therapy stimulates error-prone DNA repair in dormant hepatocellular cancer stem cells

Abstract

Previous studies have described distinct dormant and proliferating populations of cancer stem cells in hepatocellular carcinoma. The CD13 protein is involved in the scavenging of reactive oxygen species through the glutathione reductase pathway and is associated with resistance to chemotherapy. Whereas CD13(-) proliferating cancer stem cells are sensitive to chemotherapy, CD13(+) dormant cancer stem cells are associated with the development of resistance to chemotherapy. CD13(+) cells in hypoxic areas of the tumour survive chemotherapy, leading to subsequent disease relapse and metastasis. Whether CD13(+) dormant cells simply resume proliferation following therapy or whether they also acquire greater malignant potential, remains unknown. The mechanisms involved also remain unclear. In the present study, we investigated the repair of DNA damage in CD13(+) dormant and CD13(-) proliferating cells. Total RNA was extracted from tissues, and quantitative real-time polymerase chain reaction (PCR) was performed for specific genes and GAPDH following PCR. Products were then subjected to a temperature gradient of 55-95°C with continuous fluorescence monitoring to generate a melting curve. Cells were incubated with primary antibodies, washed twice, incubated with fluorescent-labelled secondary antibodies for 30 min on ice and analyzed by flow cytometry. The results revealed that the repair of DNA damage in CD13(+) dormant cells occurs predominantly through non-homologous end-joining, a repair process that is error-prone, whereas CD13(-) proliferating cells primarily utilise high-fidelity homologous recombination for DNA repair. These data indicate that not only is dormancy a protective mechanism for cancer stem cells to survive therapy, but it also enhances the generation and accumulation of mutations following DNA damage. Therefore, the CD13(+) dormant cancer stem cells must be eradicated fully to achieve complete remission of cancer.

Figure 1.

Quantitative analysis of mRNAs encoding proteins involved in HR and NHEJ. A total of 24 h after exposure to 7 Gy radiation, cells were separated by FACS into pools of dormant CD13⁺ CSCs and actively dividing (non-dormant) CD13-CSCs. The RNAs were extracted and subjected to quantitative analysis of PCR by utilising specific primers. For HR, Rad51, Rad54, Rpa, Smc6, Xrcc2 and Xrcc3 were evaluated, whereas Ku80, Ku70 and Prkdc were assessed for NHEJ. HR, homologous recombination; NHEJ, non-homologous end-joining; CSCs, cancer stem cells; PCR, polymerase chain reaction.

Figure 2.

The time course experiment of expression of NHEJ and HR proteins. Cells were separated by surface markers, dormant CSCs (CD13⁺) and non-dormant CSCs (CD13-) at the indicated time following 7 Gy radiation. cDNAs were synthesised and subjected to PCR. The expression of Rad51 (HR) and Prkdc (NHEJ) were evaluated by qRT-PCR. HR, homologous recombination; NHEJ, non-homologous end-joining; CSCs, cancer stem cells; qRT-PCR, quantitative real-time polymerase chain reaction.

Figure 3.

FACS analysis of CD13⁺ dormant CSCs following exposure to radiation (7 Gy). Following exposure to radiation, CD13⁺ cells are shown as a percentage. FACS, fluorescence-activated cell sorting; CSCs, cancer stem cells.