Building radiation-resistant model in triple-negative breast cancer to screen radioresistance-related molecular markers

Background: To build the triple-negative breast cancer (TNBC) radiation resistance model in vitro and vivo, and screen the molecular markers that related to radiation resistance. Methods: We used X-ray to irradiate MDA-MB-231 cells repeatedly to build radioresistant cell (231-RR), then select one gemcitabine-resistance of MDA-MB-231 cell (231-GEM). We screen differentially expressed genes of these cell lines. Then, we would select 2 genes of them associated with DNA damage repair or cell cycle, and build RNAi lentivirus vector to knock down related gene. We also used X-rays repeatedly exposure TNBC tumor xenograft to build tumor with radioresistance properties, and then verify previously screening differentially expressed genes using IHC. Finally, we used The Cancer Genome Atlas (TCGA) database to validate the relationships between radioresistance related genes and the prognosis of breast cancer. Results: We got 161 up-regulated genes and 156 down-regulated genes from three cell lines. Cellular results show the 231-cell with knock-down CDKN1A or SOD2 gene, its radiation sensitivity was significantly enhanced. We successfully got the TNBC xenograft tumor with radioresistance properties. Immunohistochemical results show that the radioresistance of tumor tissue with higher p21 (CDKN1A encoding protein) and SOD2 expression (P<0.01). The prognosis of patients with low SOD2 expression is better than that of high expression, but have no statistical significance (P=0.119); patients with low CDKN1A expression is significantly better than high expression (P=0.000). Multivariate cox analysis manifest that CDKN1A gene expression level is an independent prognostic factor in breast cancer patient (P=0.008). Conclusions: Construction of radiation resistance cell and xenograft tumor with radio-resistant properties model for radiation biology research is feasible. High SOD2 and CDKN1A is associated with the poor prognosis in breast cancer patients. These two genes could be used as a predicted makers of breast cancer radiation sensitivity.