TET2 repression contributes to EGFR TKI resistance in EGFR-mutant non-small cell lung cancer through regulating AXIN2 methylation

Abstract

Through targeted next-generation sequencing of 83 non-small cell lung cancer (NSCLC) patients with first-generation epidermal growth factor receptor tyrosine kinase inhibitor (EGFR TKI) resistance, we detected 11% TET2 mutations in the T790M-negative subgroup. To explore the molecular mechanism of TET2 in EGFR TKI resistance, reduced representation bisulfite sequencing (RRBS) was adopted to analyze the global genomic methylation profiles and detect the differentially methylated genes in the TET2-knockdown (KD) PC9 and control PC9 cell lines, following bioinformatics analysis of gene ontology (GO) functions and kyoto encyclopedia of genes and genomes (KEGG) signaling to screen for genes associated with drug resistance. TET2 KD attenuated gefitinib-induced apoptosis and decreased the sensitivity of EGFR-mutant lung cancer cells to gefitinib. Forty-three drug resistance genes with hypermethylated promoter regions were identified via RRBS and bioinformatic analysis in PC9^{TET2 KD} cells. Then, 10 candidate genes were screened for further validation. RT‒PCR demonstrated that the expression of AXIN2 and CSK was significantly lower in PC9^{TET2 KD} cells than in control cells. Furthermore, AXIN2 KD attenuated gefitinib-induced apoptosis and decreased the sensitivity of PC9 cells to gefitinib. Importantly, we found that the demethylation drug decitabine (DCA) could reverse gefitinib resistance in PC9^{TET2 KD} cells and mouse models. These results indicate that the methylation of AXIN2 induced by TET2 repression is a novel resistance mechanism of EGFR TKIs in EGFR-mutant NSCLC. Demethylation drugs have the potential to overcome EGFR TKI resistance induced by loss-of-function TET2 mutations.

Keywords: AXIN2; DNA methylation; EGFR TKI resistance; EGFR-mutant non-small cell lung cancer; TET2 repression.

Fig. 1

Suppression of TET2 induces resistance to gefitinib in EGFR-mutant NSCLC cells. (A, B). Western blotting was used to determine the knockdown efficiency of TET2 (TET2-KD) in the PC9 and H1650 cell lines. (C, D). An Annexin V/PI assay was used to monitor cell apoptosis in different groups 72 h after treatment with gefitinib (1 µM). (E, F). The quantification of cell apoptosis in TET2-KD and control cells (*p < 0.05). (G, H). Western blotting were used to determine the overexpression efficiency of TET2 (TET2 OE) in the PC9 and H1650 cell lines (**p < 0.01, ***p < 0.001). (I, J). A CCK-8 assay was used to monitor the proliferation of different groups of cells treated with various doses of gefitinib (*p < 0.05, **p < 0.01, ***p < 0.001, CONT KD vs. TET2 KD; #p < 0.05, ##p < 0.01, ###p < 0.001, and TET2 KD vs. TET2 KD + TET2). (K). Western blotting was used to detect phospho-EGFR and phospho-ERK in the different groups subjected to gefitinib treatment. The 2-sample t test was used to evaluate significance.

Fig. 2

The R methyl Kit package was used to analyze the differential methylation status and functional annotation. (A). Annotation of different feature areas of CpG revealed that the distribution of CpG in the genome is similar in PC9^{TET2 KD} and PC9^{CONT KD} cells. (B). The horizontal bar plots show the number of hyper- and hypomethylation events per chromosome as a percentage of the sites with the minimum coverage and differential. Hyper- or hypo-DMRs were defined as those with a ≥ 5% change in methylation and a p value < 0.05; PC9^{TET2 KD} vs. PC9^{CONT KD}.

Fig. 3

AXIN2 and CSK were significantly differentially expressed between PC9^{TET2 KD} and PC9^{CONT KD} cells. (A). Comparison of the mRNA level of AXIN2 between PC9^{TET2 KD} and PC9^{CONT KD} cells. (B). Comparison of the mRNA level of CSK between PC9^{TET2 KD} and PC9^{CONT KD} cells. The 2-sample t test was used to evaluate significance.

Fig. 4

TET2 repression could contribute to gefitinib resistance in EGFR-mutant NSCLC through DNA methylation-dependent downregulation of AXIN2.(A, B). Western blotting were used to determine the knockdown efficiency of AXIN2/CSK in PC9 cell lines. Three different sgRNAs were designed for different sites of the AXIN2 and CSK proteins, which resulted in different frameshift mutation efficiencies and KD efficiencies. Finally, the cell line with the highest knockdown efficiency was selected for subsequent study. (C, D). CCK-8 assay was used to monitor the proliferation of different groups of cells treated with various doses of gefitinib (*p < 0.05, **p < 0.01, and ***p < 0.001). (E). An Annexin V/PI assay was used to monitor cell apoptosis in different groups 72 h after treatment with gefitinib (1 µM). (F). Relative apoptosis rates of AXIN2-KD and control PC9 cells (G).Western blotting were used to determine the overexpression efficiency of AXIN2 (AXIN2 OE) in the PC9 cell line. (H). Western blotting was used to detect AXIN2 protein expression in the PC9^{CONT KD}, PC9^{AXIN2 KD}, PC9^{TET2 KD}, and PC9^{TET2 KD+AXIN2 OE} cell lines. (I). Proliferation rates of PC9^{CONT KD}, PC9^{TET2 KD}, and PC9^{TET2 KD+AXIN2 OE} cells treated with different concentrations of gefitinib (0.001–10 µM) for 48 h (*p < 0.05, and ***p < 0.001). (J). Apoptosis rates of PC9^{CONT KD}, PC9^{TET2 KD} and PC9^{AXIN2 KD} cells treated with 1 µM gefitinib for 72 h. (K). Relative apoptosis rates of PC9^{TET2 KD}, PC9^{TET2 KD+AXIN2 OE} and control PC9 cells (*p < 0.05). The 2-sample t test was used to evaluate significance.

Fig. 5

Decitabine reverses gefitinib resistance in TET2-KD EGFR-mutant NSCLC cells. BSAS analysis on PC9TET2 KD and PC9TET2 KD + DCA treated cells (A) Methylation mean of AXIN2 in PC9TET2 KD and PC9TET2 KD + DCA cell line. (B) The DNA methylation levels of CpG sites in the AXIN2 gene in PC9TET2 KD and PC9TET2 KD + DCA cell line (X-axis: CpG sites; Y-axis: DNA methylation level; Each line represents a group of samples, *p < 0.05). (C) The CCK-8 assay was used to monitor the proliferation of different groups of cells treated with various doses of gefitinib (0.0001-1 µM) ± DCA (*p < 0.05, **p < 0.01, ***p < 0.001, and TET2 KD vs. TET2 KD + DCA) (D). An Annexin V/PI assay was used to monitor the degree of cell apoptosis in the different groups after gefitinib ± DCA treatment. (E). The relative apoptosis rate is shown in Fig. 5B (*p < 0.05, ***p < 0.001). (F). Western blotting was used to detect AXIN2 and phospho-ERK in the different groups after gefitinib ± DCA treatment. (G). The expression of 5-hmC in different groups of PC9 cells was detected via a DNA hydroxymethylation (5-hmC) ELISA kit (*p < 0.05). The 2-sample t test was used to evaluate significance.

Fig. 6

Decitabine enhances the sensitivity of PC9TET2KD cells to gefitinib in vivo. (A). Timeline of animal experiments. (B). Photographs of tumors from each group. (C). Tumor size was monitored every three days from treatment initiation (Day 21) until the endpoint (Day 32) in the different treatment groups (**p < 0.01, TET2 KD + gefitinib vs. TET2 KD + gefitinib + DCA). (D). Tumor weight was monitored in the different treatment groups.