Downregulation of the DNA Repair Gene DDB2 by Arecoline Is through p53's DNA-Binding Domain and Is Correlated with Poor Outcome of Head and Neck Cancer Patients with Betel Quid Consumption

Abstract

Arecoline is the principal alkaloid in the areca nut, a component of betel quids (BQs), which are carcinogenic to humans. Epidemiological studies indicate that BQ-chewing contributes to the occurrence of head and neck cancer (HNC). Previously, we have reported that arecoline (0.3 mM) is able to inhibit DNA repair in a p53-dependent pathway, but the underlying mechanism is unclear. Here we demonstrated that arecoline suppressed the expression of DDB2, which is transcriptionally regulated by p53 and is required for nucleotide excision repair (NER). Ectopic expression of DDB2 restored NER activity in arecoline-treated cells, suggesting that DDB2 downregulation was critical for arecoline-mediated NER inhibition. Mechanistically, arecoline inhibited p53-induced DDB2 promoter activity through the DNA-binding but not the transactivation domain of p53. Both NER and DDB2 promoter activities declined in the chronic arecoline-exposed cells, which were consistent with the downregulated DDB2 mRNA in BQ-associated HNC specimens, but not in those of The Cancer Genome Atlas (TCGA) cohort (no BQ exposure). Lower DDB2 mRNA expression was correlated with a poor outcome in HNC patients. These data uncover one of mechanisms underlying arecoline-mediated carcinogenicity through inhibiting p53-regulated DDB2 expression and DNA repair.

Keywords: DDB2; DNA repair; arecoline; betel quid; head and neck cancer; p53.

Figure 1

Arecoline specifically downregulates the expression of DDB2. (A) RT-qPCR showed that arecoline treatment (0.3 mM, 24 h) decreased DDB2 mRNA level in HEp-2 cells. The mRNA levels of DDB1, XPB, and XPC were not affected apparently. The relative mRNA expression in vehicle control (H₂O) was set as one by using GAPDH as an internal control. (B) RT-qPCR showed that DDB2 mRNA level was downregulated in arecoline treated KB, SAS, HSC3, and SCC9 cells. (C) The mRNA expression of XPC was not affected by arecoline treatment in KB, SAS, HSC3, and SCC9 cells. (D) Western blot analyses showed that arecoline treatment (0.3 mM, 24 h) decreased DDB2 protein level in HEp-2 and KB cells. The protein level of XPC was not affected. (E) The expression of DDB2 mRNA was downregulated in ANE-treated human gingival fibroblasts (hGF). The results were extracted from the Gene Expression Omnibus (GSE59414) and the expression level of DDB2 mRNA was log2 transformed. FC, fold-changed (ANE versus H₂O). (F) Host cell reactivation (HCR) assay showed that overexpression of DDB2 restored arecoline-mediated inhibition of nucleotide excision repair in HEp-2 cells. The expression of flag-tagged DDB2 was detected by Western blot analysis using an anti-flag antibody. All data are shown as mean ± standard deviation (n = 3–5). Ctrl, vehicle control (H₂O); Arec, arecoline. * p < 0.05 versus control; ** p < 0.01 versus control. The full-length blots for Figure 1D,F can found at Figure S1.

Figure 2

Arecoline inhibits the recruitment of p53 and RNA polymerase II to the promoters of DDB2 and p21^Cip1 (CDKN1A) in vivo. The HEp-2 cells were treated with arecoline (0.3 mM) or vehicle (H₂O) for 24 h and then were harvested for chromatin immunoprecipitation assays using anti-p53 (A), anti-RNA polymerase II (B), and control IgG (C) antibodies followed by quantitative PCR. The PCR amplicons cover the p53-binding sites on the DDB2 (around the transcription start site, TSS) and p21^Cip1 (at 2.3 kilobase upstream to TSS) promoters. (D) Chromatin immunoprecipitation assays show no specific binding of p53 and control IgG to the XPC promoter. Data are shown as mean ± standard deviation (n = 3). Ctrl, H₂O; Arec, arecoline. ** p < 0.01 versus control.

Figure 3

Arecoline inhibits p53-induced DDB2 promoter activity in HEp-2 cells. (A) Schematic illustration of the DDB2 promoter-luciferase construct (pDDB2-Luc). (B) The wild-type p53 (p53-WT) could activate pDDB2-Luc but not pDDB2-p53x-Luc, in which the p53 binding site was mutated. (C) The mutations in DNA-binding domain (p53-175m/R175H, p53-273m/R273H) abolished p53-mediated transactivation of pDDB2-Luc. (D)The mutations of multiple phosphorylation sites in the N-terminal transactivation domain (p53-N/S6A, S9A, S15A, S18A, S20A, S33A, and S37A) or the C-terminal regulatory domain (p53-C/S315A, S371A, S376A, S378A, and S392A) did not affect p53-mediated transactivation of pDDB2-Luc. (E) Arecoline inhibited DDB2 promoter activity in a dose-dependent manner. (F) Overexpression of p53 restored arecoline-mediated inhibition of DDB2 promoter activity. Data are shown as mean ± standard deviation (n = 3). Ctrl, vehicle control (H₂O); Arec, arecoline; * p < 0.05 versus control; ** p < 0.01 versus vector (B–D) or control (E,F).

Figure 4

Arecoline inhibits p53-regulated promoters through p53’s DNA-binding domain in HEp-2 cells. (A) Schematic diagram shows the DNA-binding (DB) and transactivation (TA) domains of wild-type p53 and the p53DB-VP16TA chimeric construct; (B–D) Arecoline (0.3 mM, 24 h) inhibited p53DB-VP16TA-mediated transactivation of the p53 binding site-containing DDB2 promoter (B), p21^Cip1 promoter (C), and p3PREc-Luc (D). The p3PREc-Luc contains only 3 copies of consensus p53-responsive elements and a TATA box [38]. (E,F) Schematic illustration of pGAL-p53TA and pGAL-VP16TA chimeric constructs (E) and the pFR-Luc reporter that contains 5 copies of GAL4-binding sites (F). (G) Arecoline (0.3 mM, 24 h) did not inhibit p53’s or VP16’s TA domain-mediated transactivation of pFR-Luc. Data are shown as mean ± standard deviation (n = 3). Ctrl, H₂O; Arec, arecoline; * p < 0.05 versus control; ** p < 0.01 versus control.

Figure 5

Long-term arecoline treatment leads to suppression of DDB2 promoter and nucleotide excision repair (NER) activity. The long-term arecoline-treated HA60d cells were obtained by repetitive treatment of arecoline (0.3 mM for 6–8 h/day) for 60 days. (A) MTT assays show the cell sensitivity to arecoline treatment for 48 h; (B) HCR assay showed an impaired NER activity in HA60d cells; (C) DDB2 promoter activity was decreased in HA60d cells. Data are shown as mean ± standard deviation (n = 3–4). *, p < 0.05 versus HEp-2 cells; **, p < 0.01 versus HEp-2 cells.

Figure 6

DDB2 is downregulated in oral submucous fibroblasts (OSFs) and head and neck cancer (HNC) specimens collected from betel quid (BQ)-epidemic areas. (A) The expression of DDB2 mRNA was downregulated in 8 out of 10 OSFs in the GSE20170 dataset. FC, fold-changed (OSFs versus normal tissues). (B) The DDB2 mRNA expression in the specimens of BQ-associated HNC versus that in adjacent non-tumor tissues was examined by RT-qPCR and is shown as a ratio in a box plot. The box represents upper and lower quartiles and the horizontal line in the box represents the median expression among the 92 HNC cases of the Kaohsiung Medical University Hospital (KMUH) cohort. (C,D) The DDB2 mRNA expression in the HNC specimens of The Cancer Genome Atlas (TCGA) cohort. The level 3 RNA sequencing data was acquired from TCGA data portal and was checked for the expression of DDB2 mRNA in (C) 43 pairs of tumor/normal samples (shown as a ratio of tumor versus normal) and (D) all HNC samples (n = 521, shown by fragments per kilobase of transcript per million mapped reads, FPKM).

Figure 7

Kaplan–Meier and multivariate Cox regression analyses of overall survival (OS). (A) The OS curves of BQ-associated HNC patients (KMUH cohort). (B) Patients’ 5-yr OS rate. (C) Hazard rate ratio (HRR) of T-, N-stage, and DDB2 mRNA expression calculated by multivariate Cox model. The patients were sub-grouped based on the expression of DDB2 mRNA in tumor tissues versus that in adjacent non-tumor tissues (cutoff: 0.24). N: patient number.

Figure 8

A schematic model of the roles of DDB2 in activating DNA repair and in suppressing epithelial-mesenchymal transition (EMT). DDB2 cooperates with DDB1, XPC, and RAD23 in the recognition of the DNA damage site to initiate the global genome-NER (GG-NER). The following GG-NER steps include DNA unwinding (by XPB and XPD), excision (by XPF, ERCC1, XPG, and XPA), synthesis (by DNA polymerase δ and ε), and ligation (by DNA ligase 1) [34]. This DNA repair function plays a critical role in preventing genome instability and cancer formation [40,41]. DDB2 can also suppress metastasis [65,66] and chemoresistance [67] of cancer cells through inhibiting the expression of EMT activators Sanil, Zeb1, and VEGF. The expression of DDB2 is positively regulated by p53, which binds to the promoter of the DDB2 gene [28,29]. Arecoline suppresses DDB2 gene expression through inhibiting p53’s DNA-binding domain (p53-dbd), which may ultimately facilitate tumorigenesis, cancer metastasis, and chemoresistance.