ARNTL hypermethylation promotes tumorigenesis and inhibits cisplatin sensitivity by activating CDK5 transcription in nasopharyngeal carcinoma

Abstract

Background: Increasing evidence support an important role for DNA methylation in nasopharyngeal carcinoma (NPC). Here, we explored the role of circadian clock gene Aryl Hydrocarbon Receptor Nuclear Translocator-Like (ARNTL) methylation in NPC.

Methods: We employed bisulfite pyrosequencing to determine the epigenetic change of ARNTL in NPC cell lines and tissues. ARNTL mRNA and protein expression in cell lines and tissues were detected by real-time PCR and western blotting. Then, we constructed cell lines overexpressing ARNTL and knocked down ARNTL to explore its function and effect on chemotherapy sensitivity of NPC cell lines to cisplatin in vitro and vivo. Finally, we investigated the potential molecular mechanism of ARNTL by gene set enrichment analysis (GSEA), dual Luciferase reporter assay and chromatin immunoprecipitation assay.

Results: ARNTL was hypermethylated, and its mRNA and protein were significantly down-regulated in NPC cell lines and tissues. When treated by 5-aza-2'-deoxycytidine, mRNA expression was up-regulated. Overexpression of ARNTL could suppress NPC cells proliferation in vitro and vivo while silencing of ARNTL using shRNA achieved opposite results. GSEA assay found that ARNTL was associated with cell cycle and ectopic ARNTL overexpression could induce G2-M phase arrest. Then, we identified and validated cyclin-dependent kinase 5 (CDK5) as the targeting gene of ARNTL by dual Luciferase reporter assay and chromatin immunoprecipitation assay. When transiently infected ARNTL-overexpression cells with PENTER-vector or PENTER-CDK5 plasmids, the later could reverse the suppressive effects of ARNTL on NPC cell proliferation. Moreover, ARNTL significantly enhanced sensitivity to cisplatin in NPC cells.

Conclusions: ARNTL suppresses NPC cell proliferation and enhances sensitivity to cisplatin by targeting CDK5. ARNTL may represent a novel therapeutic target for NPC.

Keywords: ARNTL; CDK5; Chemotherapy sensitivity; Methylation; Nasopharyngeal carcinoma; Proliferation.

Fig. 1

ARNTL is hypermethylated in nasopharyngeal carcinoma. a ARNTL promoter CpG islands and bisulfite pyrosequencing region. Blue region, CpG islands; Red region, input sequence; TSS, transcription start site; cg15603424, CG site identified in our previous genome-wide methylation analysis; red text, CG sites for bisulfite pyrosequencing; bold red text, the most methylated CG sites in ARNTL. b Bisulfite pyrosequencing analysis of the ARNTL promoter region in 8 pairs of normal and nasopharyngeal carcinoma tissues. c Average methylation levels of ARNTL in 8 pairs of normal and nasopharyngeal carcinoma tissues. d Methylation levels of ARNTL in NP69 and nasopharyngeal carcinoma cell lines. e Correlation between promoter methylation and mRNA expression of ARNTL in the TCGA head and neck cancer dataset. ** indicated P < 0.01

Fig. 2

ARNTL is downregulated in nasopharyngeal carcinoma due to its promoter hypermethylation. a Quantitative RT-PCR analysis of ARNTL mRNA expression in NP69 and nasopharyngeal carcinoma cell lines. b ARNTL mRNA is downregulated in the GSE12452 nasopharyngeal carcinoma dataset. c Western blotting assay of ARNTL protein expression in nasopharyngeal carcinoma cell lines and normal NP69. d Western blotting assay of ARNTL protein expression in nasopharyngeal carcinoma tissues (T, n = 4) and normal nasophrynx epithelial tissues (T, n = 4). e ARNTL methylation levels before and after DAC treatment in NP69 and nasopharyngeal carcinoma cell lines. f ARNTL mRNA expression before and after DAC treatment in NP69 and nasopharyngeal carcinoma cell lines. * indicated P < 0.05

Fig. 3

Effects of ARNTL overexpression or knocking down on nasopharyngeal carcinoma cell viability and colony formation ability in vitro. a Quantitative RT-PCR assay of ARNTL mRNA expression in SUNE1 and HONE1 cells stably overexpressing ARNTL. b Western blotting assay of ARNTL protein expression in SUNE1 and HONE1 cells stably overexpressing ARNTL. c, d The CCK-8 assay demonstrated that overexpression ARNTL reduced the viability of SUNE1 and HONE1 cells. e The colony formation assay showed that overexpression of ARNTL suppressed colony-forming ability of SUNE1 and HONE1 cells. f Quantitative RT-PCR analysis of ARNTL mRNA expression in SUNE1 and HONE1 cells knocking down by ShRNAs. g Western blotting analysis of ARNTL protein expression in SUNE1 and HONE1 cells knocking down by ShRNAs. h, i The CCK-8 assay demonstrated that knocking down ARNTL promoted the viability of SUNE1 and HONE1 cells. j The colony formation assay showed that knocking down ARNTL enhanced colony-forming ability of SUNE1 and HONE1 cells. * indicated P < 0.05; ** indicated P < 0.01

Fig. 4

Overexpression of ARNTL suppressed tumorigenicity of nasopharyngeal carcinoma cells in vivo. a Xenograft tumors of BALB/c nude mice formed at 30 days after injecting with SUNE1 cells stably overexpressing ARNTL or Vector. b Growth curves of xenograft tumor volume. c Average xenograft tumor weights at 30 days after injecting with SUNE1 cells. d Immunohistochemistry assay of ARNTL protein expression in xenograft tumors. e Immunohistochemistry assay of CDK5protein expression in xenograft tumors. ** indicated P < 0.01

Fig. 5

Overexpression of ARNTL induced cell cycle arrest at G2-M phase. a GSEA enrichment plots showed that enrichment of G2/M checkpoint and mitotic spindle pathways was associated with ARNTL downregulations. b, c Flow cytometry analysis of cell cycle distribution in SUNE1 and HONE1 cells stably overexpressing ARNTL or Vector. d, e Flow cytometry analysis of cell cycle distribution in SUNE1 and HONE1 cells after silencing ARNTL

Fig. 6

CDK5 is a direct targeting gene of ARNTL. a ARNTL motif; b Putative ARNTL-binding sequence in promoter of CDK5 mRNA; c Spearman correlation analysis between ARNTL and CDK5 mRNA expression in GSE12452 nasopharyngeal carcinoma dataset; d Quantitative RT-PCR analysis of CDK5 mRNA expression in ARNTL-overexpression SUNE1 and HONE1 cells; e Quantitative RT-PCR analysis of CDK5 mRNA expression in ARNTL-silencing SUNE1 and HONE1 cells; f Western blotting analysis of CDK5 protein expression in ARNTL-overexpression or ARNTL-silencing SUNE1 and HONE1 cells;. g Relative luciferase activity of ARNTL-overexpression or Vector-expression SUNE1 and HONE1 cells after transfecting with wild type or mutant CDK5 3’UTR reporter genes. h ChIP real-time PCR assay for assessing the enrichment of ARNTL in the CDK5 promoter regimen in SUNE1-ARNTL and HONE1-ARNTL NPC cells (anti-RNA Pol II serves as positive control).; i Quantitative RT-PCR analysis of CDK5 mRNA expression in ARNTL-overexpression SUNE1 and HONE1 cells after transiently transfecting with CDK5-overexpression plasmid. j Western blotting analysis of CDK5 protein expression in ARNTL-overexpression SUNE1 and HONE1 cells after transiently transfecting with CDK5-overexpression plasmid. k The CCK-8 assay revealed overexpression of CDK5 could reverse ARNTL-mediated viability suppression in SUNE1 cells; l The CCK-8 assay revealed overexpression of CDK5 could reverse ARNTL-mediated viability suppression in HONE1 cells; (m-n) Colony formation assay revealed that CDK5 could reverse ARNTL-mediated colony-forming ability suppression in SUNE1 and HONE1 cells. * indicated P < 0.05; ** indicated P < 0.01

Fig. 7

ARNTL increased sensitivity of nasopharyngeal carcinoma cells to cisplatin in vitro and vivo. a-d The CCK-8 assay revealed that overexpression of ARNTL increased cisplatin sensitivity while knocking down ARNTL decreased cisplatin sensitivity of SUNE1 and HONE1 cells in vitro. e, f Xenograft tumors of BALB/c nude mice at 18 days after injecting with Vector-overexpression or ARNTL-overexpression HONE1 cells and treating with cisplatin or saline. g Average tumor volume of BABL/c nude mice injecting with Vector-overexpression or ARNTL-overexpression HONE1 cells and treating with cisplatin or saline. h Tumor volume remission rates between Vector-overexpression and ARNTL-overexpression nude mice treated by cisplatin. i Average xenograft tumor weight of Vector-overexpression and ARNTL-overexpression nude mice treated by cisplatin or saline. j Tumor weight remission rates between Vector-overexpression and ARNTL-overexpression nude mice treated by cisplatin. * indicated P < 0.05; ** indicated P < 0.01