Human liver cancer cell lines HepG2 and HuH7 were obtained from the Japanese Cancer Research Resources Bank. DMEM and 100X antibiotic-antimycotic solution were purchased from Thermo Fisher Scientific, Inc. FBS, PBS and DAC were purchased from Sigma-Aldrich; Merck KGaA. DMSO, βNF and proteinase K were purchased from Fujifilm Wako Pure Chemical Corporation. DAC and βNF were dissolved in PBS and DMSO, respectively.

HepG2 and HuH7 cells were cultured in DMEM containing 10% FBS and 1X antibiotic-antimycotic solution at 37˚C in a humidified atmosphere under 5% CO₂. Thereafter, CYP1B1 expression was induced by treatment with 10 µM βNF at 37˚C for 4 h. To assess DNA demethylation, cells were exposed to 0.5 µM DAC at 37˚C for 72 h and the medium was changed every 24 h, prior to βNF treatment. PBS and 0.1% DMSO were used as solvent controls. After treatment, cells were further assessed using various assays.

From the results of preliminary experiments using reverse transcription-quantitative PCR analysis and bisulfite direct sequencing, it was revealed that 10 µM (between 0.01 and 100 µM) βNF and 0.5 µM (between 0.5 and 5 µM) DAC were the minimum concentration levels that caused substantial CYP1B1 gene induction and gene demethylation, respectively (data not shown). In addition, these concentration levels fell within the range of those reported by previous studies (5,16). Therefore, this dose combination was selected for the detection of DNA demethylation.

Total RNA was isolated from cells and purified using the RNeasy Mini kit and DNase Set (Qiagen GmbH). First-strand cDNA was prepared from total RNA (3 µg) using the PrimeScript™ II 1st strand cDNA Synthesis kit (Takara Bio, Inc.). The following temperature protocol was used for reverse transcription: 30˚C for 10 min followed by 42˚C for 60 min and the reaction was terminated by 95˚C for 5 min. Thereafter, the levels of CYP1B1, AHR, ARNT and ACTB (β-actin) transcripts were measured using qPCR and the FastStart™ Universal SYBR^®-Green Master (ROX) kit (Roche Diagnostics) in the ABI 7500 System (Thermo Fisher Scientific, Inc.). The following thermocycling conditions were used for qPCR: Initial denaturation for 10 min at 95˚C; followed by 45 cycles for 15 sec at 95˚C and 60 sec at 60˚C. The primer sequences used for qPCR are listed in Table I. Relative mRNA expression levels were determined using the standard curve method and a 10-fold dilution series, followed by normalization to ACTB mRNA expression levels (17). In all qPCR experiments, three independent replicates were analyzed for each treatment and the data were expressed as the mean ± standard deviation. P<0.05 was considered to indicate a statistically significant difference. This was determined using Shaffer's multiple comparison test using R statistical language and RStudio (ver. 1.2.5042; RStudio, Inc.).

In total, eight potential XRE sites are located within 2.3 kb of the 5'-flanking region of the human CYP1B1 gene (18). To assess the original DNA methylation status of these XREs in HepG2 and HuH7 cells without any treatment, genomic DNA samples were isolated from cells using the standard proteinase K/phenol-chloroform extraction method (19) and subjected to bisulfite sequencing.

Bisulfite modification was performed using the EpiTect Plus DNA Bisulfite kit (Qiagen GmbH) according to the manufacturer's protocol. A total of four genomic DNA segments covering the XRE1, XRE2/XRE3, XRE4/XRE5/XRE6 or XRE7/XRE8 regions were amplified by PCR using the Takara EpiTaq HS kit (Takara Bio, Inc.) and the primer sets listed in Table I. Thermocycling conditions were as follows: Initial denaturation at 94˚C for 5 min; followed by 35 cycles of denaturation at 94˚C for 30 sec, annealing at 55˚C (XRE1 and XRE4/XRE5/XRE6) or 60˚C (XRE2/XRE3 and XRE7/XRE8) for 30 sec and extension at 72˚C for 30 sec; and final extension at 72˚C for 7 min. These primers were originally designed based on bisulfite-modified sequences covering the XRE1-8 sites as described previously (18). After amplification and purification using the NucleoSpin Gel and PCR Clean-up kit (Takara Bio, Inc.), the methylation status of these fragments was detected by Sanger direct sequencing, which was performed using the BigDye™ Terminator v3.1 Cycle Sequencing kit (Thermo Fisher Scientific, Inc.) and 3130xl Genetic Analyzer (Thermo Fisher Scientific, Inc.). DNA methylation status of each XRE site was visibly determined according to the peak area ratio of C (methylated) and T (unmethylated) in the electropherogram and categorized as follows: Fully methylated; highly methylated (C>T); lowly methylated (C<T); and rarely methylated (fully unmethylated). The categorization was confirmed to be identical between two independent experiments.

To assess potential AhR binding to the XRE2/XRE3 sequence of the CYP1B1 gene, the CUT&RUN method was applied to HepG2 and HuH7 cells treated with βNF at 37˚C for 4 h using the antibody-targeted digestion of chromatin, which causes lower background signals than those obtained using the ChIP method (15). AhR-DNA complexes were precipitated from cells using an anti-AhR rabbit monoclonal antibody (cat. no. 83200; Cell Signaling Technology, Inc.) and a CUT&RUN Assay kit (cat. no. 86652S; Cell Signaling Technology, Inc.), according to the manufacturer's protocol.

Briefly, 10 µl concanavalin A magnetic beads suspended in 100 µl concanavalin A bead activation buffer were added to 1x10⁵ cell solution (suspended in 100 µl 1X Wash Buffer) and were mixed on a rotator for 5 min at room temperature. The cell pellets separated on magnetic rack for 2 min at room temperature were mixed with 100 µl antibody binding buffer and 3 µl 1:33 diluted anti-AhR rabbit monoclonal antibody on the rotator for 12 h at 4˚C. The cell pellets were then bound to pAG-MNase enzyme by gentle mixing on the rotator for 1 h at 4˚C with 50 µl digitonin buffer (including 5 µl 10X Wash Buffer, 0.5 µl 100X Spermidine, 0.25 µl 200X Protease Inhibitor Cocktail, 1.25 µl Digitonin Solution, and 43 µl water) and 1.5 µl pAG-MNase Enzyme. After cooling for 5 min on ice with 150 µl digitonin buffer, the cell pellets were subjected to MNase activation by adding 3 µl cold CaCl₂. DNA digestion was stopped by adding 150 µl of 1X Stop Buffer (including 3.75 µl Digitonin Solution and 0.75 µl RNaseA) and incubating at 37˚C for 10 min. DNA purification was performed using DNA Purification Buffers and Spin columns (cat. no. 14209; Cell Signaling Technology, Inc.). Briefly, digested DNA was mixed with 1.5 ml DNA Binding Buffer and was transferred to a DNA spin column in a collection tube. After centrifugation at 18,500 x g for 30 sec at 20˚C, DNA was washed with 750 µl DNA wash buffer. Finally, DNA was eluted from the column using 50 µl of DNA Elution Buffer. The eluate was subjected to PCR using the Takara EpiTaq HS kit (Takara Bio, Inc.). To confirm that this assay was optimized, positive control using 3 µl 1:33-diluted Tri-Methyl-Histone H3 (Lys4) (C42D8) Rabbit monoclonal antibody (cat. no. 9751; Cell Signaling Technology, Inc.) and negative control using 5 µl 1:20-diluted Rabbit (DA1E) monoclonal antibody IgG XP Isotype Control (CUT&RUN; cat. no. 66362; Cell Signaling Technology, Inc.) were tested instead of using the anti-AhR rabbit monoclonal antibody. Input DNA without antibody precipitation (genomic DNA prepared for DNA methylation analysis as aforementioned) was used as an internal control. The primers used for amplifying the CYP1B1 enhancer region covering the XRE2/XRE3 (-872 to -723) sequences are listed in Table I. Thermocycling conditions were as follows: Initial denaturation at 94˚C for 5 min; followed by 35 cycles of denaturation at 94˚C for 30 sec, annealing at 62˚C for 30 sec and extension at 72˚C for 30 sec; and final extension at 72˚C for 7 min. The sizes of the PCR products were confirmed by electrophoresis on a 2% agarose gel with ethidium bromide staining at room temperature for 20 min.

After the CUT&RUN assay, DNA methylation at the individual CpG sites within the XRE2/XRE3 sequences was examined for anti-AhR antibody-precipitated complexes using COBRA (17,20). The template DNA was purified from the AhR-DNA complexes using proteinase K digestion at 55˚C for 2 h, phenol/chloroform extraction and ethanol precipitation to achieve the complete conversion of unmethylated cytosines to uracils by bisulfite treatment (21). After bisulfite modification, the XRE2/XRE3 sequence was amplified by PCR using the same kit and primers as that used for DNA methylation analysis. For COBRA, the amplified products were digested with the restriction endonuclease HpyCH4IV (New England Biolabs, Inc.) at 37˚C for 3 h, which cleaves only methylated CpG sites within both XRE sequences. Subsequently, the digested DNA was electrophoresed on a 3% agarose gel and visualized by ethidium bromide staining.

For in-depth analysis, the CpG methylation status of a single copy of the CYP1B1 gene was determined through dideoxy sequencing after subcloning the amplified products into the pCR4-TOPO vector using a TOPO™ TA Cloning™ kit for Sequencing (Thermo Fisher Scientific, Inc.) (22). The sequencing reaction was performed using the M13 primer set: Forward, 5'-GTAAAACGACGGCCAG-3' and reverse, 5'-CAGGAAACAGCTATGAC-3' (reverse). In total, ≥10 clones of each amplified product were sequenced. In addition, the input genomic DNA was subjected to COBRA and subcloning as a control to indicate the original DNA methylation status of HepG2 and HuH7 cells.

Fig. 1 demonstrates the levels of CYP1B1 expression in the two liver cancer cell lines. βNF significantly increased CYP1B1 expression in both HepG2 and HuH7 cells compared with the PBS-treated group (Fig. 1A and B). By contrast, cell treatment with DAC alone conferred no significant effects on CYP1B1 expression compared with those in the PBS group. The βNF-induced upregulation of CYP1B1 expression was significantly potentiated by pre-treating the HepG2 cells with DAC compared with that in the βNF group (βNF vs. D + β; Fig. 1A). However, no such potentiation could be observed in HuH7 cells (βNF vs. D + β; Fig. 1B). Treatment with DAC alone had no significant effects on the AHR and ARNT expression levels in HepG2 cells (Fig. 1C and E). However, AHR and ARNT expression was slightly but significantly upregulated in HuH7 cells by DAC treatment alone compared with that in cells treated with PBS (Fig. 1D and F). These results suggested that DNA demethylation induced by DAC treatment enhanced the response to the βNF-activated AhR pathway in HepG2 cells, which increased CYP1B1 expression without increasing AHR or ARNT expression. In HuH7 cells, DNA demethylation upregulated AHR and ARNT expression, but this had little effect on CYP1B1 gene expression.

Fig. 2 illustrates the DNA methylation status of the eight XRE sites detected in the two liver cancer cell lines without any treatment. XRE4, XRE5 and XRE6 were found to be rarely methylated (mostly unmethylated) in both cell lines. This finding indicates that DAC treatment-enhanced the βNF-induced CYP1B1 expression in the HepG2 cells in a manner independent of the methylation status of these XRE sequences. By contrast, highly methylated XRE7 and fully methylated XRE8 sequences were observed in both cell lines. Since DAC treatment (inhibition of methylation) did not markedly enhance βNF-induced CYP1B1 expression in HuH7 cells, demethylation of the XRE7/8 is unlikely to exert any effects on the enhanced inducibility in HepG2 cells.

Subsequently, the potential binding of AhR to the XRE2 and XRE3 sites was next analyzed for two reasons. These two XREs have been reported to serve a pivotal role in TCDD-mediated induction of CYP1B1 expression (18,23). In addition, lowly methylated XRE2 and highly methylated XRE3 were observed in HepG2 cells (Fig. 2), where methylated and unmethylated target sequences coexist. Therefore, XRE2 and XRE3 were selected for evaluating whether AhR preferentially binds to unmethylated or methylated XREs.

COBRA using input DNA revealed that amplified DNA covering the XRE2/XRE3 sequences was partially digested by the restriction enzyme HpyCH4IV in βNF-treated HepG2 cells but not in HuH7 cells (Fig. 3). Therefore, the CpG sites within the XRE2/XRE3 sequence were partially methylated in the HepG2 cells but not in the HuH7 cells. These results are consistent with those from the direct sequencing for the assessment of methylation status without any treatment (Fig. 2). In addition, this partial methylation in the HepG2 cells was mostly eliminated after combined treatment with DAC and βNF (Fig. 3), indicating successful DNA demethylation. To determine if AhR preferentially bounds to unmethylated XRE sequences, AhR-DNA complexes were precipitated from cells using the CUT&RUN method with the anti-AhR antibody. Specific DNA amplification of the CYP1B1 XRE2/XRE3 sequence was confirmed by electrophoresis and direct sequencing. The AhR-DNA complexes were then subjected to bisulfite modification and methylation analysis. Unlike input DNA, the precipitated AhR-bound DNA from these cell lines exhibited no detectable methylation in either cell line (Fig. 3).

To determine the CpG methylation status in a single copy of the gene, the input and precipitated DNA used for COBRA in Fig. 3 were subcloned and sequenced. Fig. 4A shows the nucleotide sequence around the CYP1B1 XRE2/XRE3 sequence whereas Fig. 4B shows the representative electropherograms of methylated and unmethylated XRE sequences. Fig. 4C demonstrates the methylation status of the individual clones. In HepG2 cells, of the 10 clones derived from input DNA, eight exhibited XRE3 methylation and three revealed simultaneous XRE2 methylation (Fig. 4C). However, such methylation could not be detected in the clones isolated from the precipitated AhR-bound DNA, whereas methylation outside the XRE2/XRE3 sequences was frequently retained even after precipitation (Fig. 4C). This finding that AhR-bound DNA clones always revealed the unmethylated XRE2/3 status was consistent with the COBRA results (Fig. 3). In HuH7 cells, clones with methylated XRE2 and XRE3 could not be obtained from input or AhR-precipitated DNA. The results suggested that βNF-induced AhR recognizes and selectively bind to unmethylated XRE2 and XRE3 sequences.