NRF2 mutation confers malignant potential and resistance to chemoradiation therapy in advanced esophageal squamous cancer

Abstract

Esophageal squamous cancer (ESC) is one of the most aggressive tumors of the gastrointestinal tract. A combination of chemotherapy and radiation therapy (CRT) has improved the clinical outcome, but the molecular background determining the effectiveness of therapy remains unknown. NRF2 is a master transcriptional regulator of stress adaptation, and gain of-function mutation of NRF2 in cancer confers resistance to stressors including anticancer therapy. Direct resequencing analysis revealed that Nrf2 gain-of-function mutation occurred recurrently (18/82, 22%) in advanced ESC tumors and ESC cell lines (3/10). The presence of Nrf2 mutation was associated with tumor recurrence and poor prognosis. Short hairpin RNA-mediated down-regulation of NRF2 in ESC cells that harbor only mutated Nrf2 allele revealed that themutant NRF2 conferred increased cell proliferation, attachment-independent survival, and resistance to 5-fluorouracil and γ-irradiation. Based on the Nrf2 mutation status, gene expression signatures associated with NRF2 mutation were extracted from ESC cell lines, and their potential utility for monitoring and prognosis was examined in a cohort of 33 pre-CRT cases of ESC. The molecular signatures of NRF2 mutation were significantly predictive and prognostic for CRT response. In conclusion, recurrent NRF2 mutation confers malignant potential and resistance to therapy in advanced ESC, resulting in a poorer outcome. Molecular signatures of NRF2 mutation can be applied as predictive markers of response to CRT, and efficient inhibition of aberrant NRF2 activation could be a promising approach in combination with CRT.

Figure 1

NRF2 mutation in EC. (A) Somatic NRF2 mutations are clustered in the KEAP1 binding domain. Schematic presentation of NRF2 protein indicates the location of Neh1∼6 (Nrf2-ech homology) domains and functional annotations. Somatic mutations identified in this study (blue) affected amino acids in the DLG or ETGE motifs. (B) Sequence chromatography of NRF2 mutations in ESC cell lines with corresponding normal sequences in the bottom panel. KYSE70 and KYSE180 cells harbor homozygous mutations. (C) Kaplan-Meier plot showing overall survival of ESC patients segregated according to NRF2 mutation status.

Figure 2

Down-regulation of gain-of-function NRF2 mutant in KYSE70 cells by shRNA. (A) Transcriptional activities of wild-type and ESC-associated mutant NRF2 were measures in the absence or presence of the Keap1 gene. (B) Down-regulation of NRF2 protein in KYSE70 cells treated with NRF2-targeting shRNAs. Left, Immunoblot analysis of NRF2 protein in the nuclear fraction of KYSE70 cells infected with GFP shRNA (cont) or two different NRF2-shRNAs (sh-1 and sh-2). HEK293 cells lack NRF2 accumulation and were loaded as a negative control. Lamin B1 was used as a loading control. Right, Relative NRF2 expression in each clone was presented. (C) NRF2-dependent transcriptional activities of KYSE clones and HEK293. (D) Cell proliferation plot of KYSE70 clones infected with control or NRF2-shRNAs (shRNA-1 and shRNA-2). Data are mean ± SD of three independent experiments. *P < .05, **P < .01. (E) KYSE70 clones were cultured on an ultra low-attachment plate for 72 hours, and whole proteins were then extracted. Expression of cleaved PARP was examined by immunoblot analysis. β-Actin was used as a loading control. Relative PARP1 expression compared with the control shRNA was shown below. Molecular marker is indicated on the right (kDa).

Figure 3

Mutant NRF2 confers resistance to chemotherapeutic reagent and radiation. (A) KYSE70 clones infected with control shRNA or NRF2-shRNAs were treated with various concentrations of 5-FU. (B) KYSE180 and KYSE110 cells were treated with control or NRF2 siRNA and various concentrations of 5-FU, as in A. Data represent cell viability after 48 hours relative to vehicle (DMSO)-treated control. Data are mean ± SD of three independent experiments; statistical difference to control siRNA or shRNA, **P < .01, ***P < .001. (C) Left, A representative plate showing colony formation by each clone with control (0 Gy) and 2-Gy irradiation. Top right, Number of KYSE70 clones infected with control shRNA and NRF2-shRNAs that formed colonies after control (0 Gy) and 2-Gy irradiation. Data are mean ± SD of eight independent plates. Bottom right, The relative number of colonies after 2-Gy irradiation compared with the control (0 Gy). Note that the difference between control shRNA and shRNA-1 clones was not significant (n.s.). (D) KYSE70 clones were irradiated with different doses, and the expression of GCLC and GSR mRNAs was then examined. Data are mean ± SD of three independent experiments, **P < .01. ***P < .001.

Figure 4

Pathway signatures of mutant NRF2 are associated with ESC patients' survival after CRT. (A) Two-dimensional hierarchal clustering of KYSE cells by genome-wide gene expression. NRF2 mutation status is shown at the bottom. GSEA extracted characteristic gene sets enriched in NRF2-mutated or wild-type ESC cells, and their prognostic significance was evaluated based on gene expression data for biopsies from pre-CRT cases. (B) Representative panels showing results of GSEA of genes enriched in NRF2-mutated ESC cell lines. (C) Kaplan-Meier plot showing overall survival of ESC patients after CRT segregated according to the NRF2 mutant signature.