Integrated analysis of DNA methylation and mRNA expression profiling reveals candidate genes associated with cisplatin resistance in non-small cell lung cancer

Abstract

DNA methylation plays a critical role during the development of acquired chemoresistance. The aim of this study was to identify candidate DNA methylation drivers of cisplatin (DDP) resistance in non-small cell lung cancer (NSCLC). The A549/DDP cell line was established by continuous exposure of A549 cells to increasing concentrations of DDP. Gene expression and methylation profiling were determined by high-throughput microarrays. Relationship of methylation status and DDP response was validated in primary tumor cell culture and the Cancer Genome Atlas (TCGA) samples. Cell proliferation, apoptosis, cell cycle, and response to DDP were determined in vitro and in vivo. A total of 372 genes showed hypermethylation and downregulation in A549/DDP cells, and these genes were involved in most fundamental biological processes. Ten candidate genes (S100P, GDA, WISP2, LOXL1, TIMP4, ICAM1, CLMP, HSP8, GAS1, BMP2) were selected, and exhibited varying degrees of association with DDP resistance. Low dose combination of 5-aza-2'-deoxycytidine (5-Aza-dC) and trichostatin A (TSA) reversed drug resistance of A549/DDP cells in vitro and in vivo, along with demethylation and restoration of expression of candidate genes (GAS1, TIMP4, ICAM1 and WISP2). Forced expression of GAS1 in A549/DDP cells by gene transfection contributed to increased sensitivity to DDP, proliferation inhibition, cell cycle arrest, apoptosis enhancement, and in vivo growth retardation. Together, our study demonstrated that a panel of candidate genes downregulated by DNA methylation induced DDP resistance in NSCLC, and showed that epigenetic therapy resensitized cells to DDP.

Keywords: 5-aza-2′-deoxycytidine; GAS1; cisplatin; lung cancer; methylation; trichostatin A.

Figure 1. Function and pathway analysis of the 372 hypermethylated genes identified in the A549/DDP cell line. Genes were identified using the Illumina Infinium HumanMethylation450 BeadArray platform. Gene ontology (GO) analysis by three domains: Biological Process (A), Cellular Component (B) and Molecular Function (C). KEGG Pathway analysis (D).

Figure 2. Validation of the methylation status by qMSP (A) and BSP (B). All 10 candidate genes were confirmed to be hypermethylated in A549/DDP cells compared with A549 cells. The amount of methylated DNA was determined by the threshold cycle number (Ct) for each sample and assessed as the percentage of methylation reference (PMR). Each pie in BSP results represented a CpG site; methylation (black) and unmethylation (white) ratios were calculated according to the five clones.

Figure 3. Methylation status and DDP response. (A) By primary tumor cell culture and drug susceptibility testing, 20 NSCLC samples were considered DDP sensitive (IC₅₀ < 5 mg/L) and 20 samples were considered DDP resistant (IC₅₀ > 10 mg/L). Methylation status of 10 candidate genes was tested by qMSP; (B) Kaplan-Meier analysis of candidate gene methylation and overall survival in patients who received platinum-based chemotherapy using TCGA data. Hyper- or hypo-methylation was defined by the median of methylation index β-values.

Figure 4. In vitro effects of the combinatorial treatments with 5-Aza-dC and TSA in the A549/DDP cell line. (A) Minimum effective dose of 5-Aza-dC and TSA determined by MTT; (B) cell proliferation determined by MTT; (C) cell cycle determined by flow cytometry; (D) apoptosis determined by flow cytometry; (E) apoptosis determined by fluorescence microscope; (F) DNMT activity and histone deacetylase H3/H4 activity; (G) mRNA expression of four candidate genes; (H) methylation status of four candidate genes. 1, A549/DDP cells; 2, A549/DDP cells treated with 1 μM 5-Aza-dC; 3, A549/DDP cells treated with 100 nM TSA; 4, A549/DDP cells treated with 1 μM 5-Aza-dC and 100 nM TSA. Cells were cultured in RPMI-1640 medium containing 2 mg/L DDP in these experiments. *P < 0.05 vs control; **P < 0.05 vs. other 3 groups.

Figure 5. Epigenetic therapy in vivo. A549/DDP cells (2 × 10⁶/100 μL PBS) were subcutaneously inoculated into the right flank of BALB/c nu/nu mice and animals were randomly divided into 7 treatment groups as described in Methods (A). The tumor size was monitored every other day (B). Mice were sacrificed and the tumors were isolated after two weeks (C). The mRNA expression (D) and methylation (E) profiling of GAS1, TIMP4, ICAM1 and WISP2 were determined in each group. *P < 0.05 vs. NS group; **P < 0.05 vs. other 6 groups.

Figure 6. Verification of GAS1 function. A549/DDP cells were stably transfected with N-eGFP-NC or N-eGFP-GAS1 vector, and GAS1 expression levels were detected via real-time PCR (A) and western blot (B) 48 h post-transfection. (C) MTT assay revealed that cells with upregulated GAS1 expression were more sensitive to DDP. (D) Colony formation assay revealed suppression of cellular proliferation with upregulated GAS1 expression. (E) Flow cytometric analysis of cell cycle indicated that upregulation of GAS1 significantly induced G1 phase arrest. Flow cytometric analysis (F) and fluorescence microscope (G) indicated that upregulation of GAS1 significantly induced apoptosis. (H) Growth curve of tumors derived from GAS1-transfected A549/DDP cells and NC cells. Each point represents the mean ± SD of 5 mice. (I) Representative photographs of tumors formed 2 wk after the subcutaneous transplantation. Transplanted tumors with H&E staining (J), and immunohistochemistry of GAS1 (K).