Treatment with 5-azacytidine Increases UCP2 levels in Hep3B and HT-29 Cells but Not in HepG2 Cells


 Background: The function of the uncoupling protein 2 (UCP2) is different for each cancer cell, and the mechanism of production is unclear. DNA methylation affects protein expression and is one factor that transforms normal cells into cancer cells. The hepatocellular carcinoma Hep3B and HepG2 cells and colorectal cancer HT-29 cells were treated with 5-azacytidine (5-aza), a DNA demethylation agent, to observe the modification of UCP2 production and the methylation degree in the UCP2 promoter region.Methods: Promoter basal activity and degree of UCP2 production were measured in Hep3B, HepG2, and HT-29 cells. In addition, methylation-specific PCR (MSP) was performed to investigate the degree of methylation in the UCP2 promoter region. The methylation region in the UCP2 promoter was confirmed based on bisulfite sequencing.Results: In Hep3B cells in which UCP2 mRNA was not transcribed, the promoter basal activity was significantly higher than in HT-29 or HepG2 cells in which UCP2 mRNA was transcribed. Treatment with 5-aza increased UCP2 expression in Hep3B and HT-29 cells; however, the expression in HepG2 cells was unchanged. The UCP2 promoter in Hep3B cells has numerous methylated regions compared with HT-29 and HepG2 cells.Conclusion: The results of the present study revealed that inhibition of UCP2 production in Hep3B cells was due to multiple methylation of the promoter region. Investigating the mechanism that induces UCP2 production in cancer cells is important to understand the function of UCP2, which could aid in cancer treatment.

study, UCP2 inhibited the proliferation of cancer cells by increasing ROS production [16]. Therefore, UCP2 can play different roles depending on the cell type and situation. UCP2 expression also correlates with tumor modi cation [17]. In breast cancer, UCP2 is signi cantly associated with tumor grade; increased UCP2 expression reduced the sensitivity of breast cancer cells to treatment [18,19]. In several studies, UCP2 was shown to play an important role in the resistance of pancreatic cancer to chemotherapy [20]. These results indicate that regulation of UCP2 expression is strongly associated with the growth and treatment of cancer cells. However, the induction of UCP2 expression remains unclear.
In the present study, two types of hepatocellular carcinoma (HCC) cells, Hep3B and HepG2, and colorectal cancer cells, HT-29, were used to determine the regulatory mechanisms underlying UCP2 expression. Each cancer cell type was treated with 5-azacytidine (5-aza), a DNA demethylation agent [21,22]. UCP2 promoter basal activity measurement, methylation-speci c PCR (MSP), and bisul te sequencing were performed to investigate the degree of methylation in the UCP2 promoter region. Consequently, UCP2 expression was suppressed in Hep3B cells, apparently due to multiple methylation of the UCP2 promoter region. In addition, the difference in UCP2 expression level between the HT-29 and HepG2 cells was likely caused by various degrees of methylation in the UCP2 promoter region.

Plasmid construct
The plasmids for the transient expression assay to examine basal promoter activity of the human UCP2 gene were constructed using the SEAP reporter system (TaKaRa, Tokyo, Japan) following the manufacturer's protocol. The following primers with appropriate restriction sites were used to amplify the promoter regions [23]: sense primer sequence, 5′-GGTACCTCAAGATAACTGGTATGCCTTGT-3′, and antisense primer sequence, 5′-GAATTCTCATACTATGTGTCCGAGCCGCA-3′. PCR conditions were 40 cycles of 30 s at 94°C, 30 s at 60°C, and 3 min at 72°C, with a nal extension of 10 min at 72°C. The PCR product size of UCP2 promoter was 2,960 bp. The PCR product was ligated into the KpnI/EcoRI site of the polylinker region of the SEAP2 basic vector.
Analysis of basal promoter activity of the human UCP2 gene To examine basal promoter activity of the human UCP2 gene, transient expression assay of the UCP2 promoter SEAP construct was performed in Hep3B, HT-29, and HepG2 cell lines. The cells were cultured at a density of 1 × 10 5 cells in 35-mm dishes and DMEM containing 10% FCS. After seeding, the dishes were washed extensively to remove non-adherent cells and the medium was replaced. On the second day, transfection was performed with FuGENE 6 transfection reagent (Boehringer Mannheim, Mannheim, Germany) according to the manufacturer's protocol. The plasmid (1 μg) consisting of the human UCP2 promoter region was fused to the SEAP basic vector and 1 μg of the SEAP control vector. After transfection, the medium was replaced and the cells cultured for an additional day. The supernatant was collected from each sample culture and SEAP activity measured.
Total RNA extraction and cDNA synthesis Cells were homogenized with 1 mL of TRIzol Reagent (Ambion, Carlsbad, NM, USA), mixed with 0.3 mL of chloroform, and centrifuged at 12,000 rpm for 15 min at 4°C. The aqueous layer was transferred to a new tube and 0.6 mL of isopropanol added. The tubes were inverted several times and centrifuged at 12,000 rpm for 10 min at 4°C. The supernatant was removed and the RNA pellet washed with 70% alcohol. The RNA pellet was brie y air-dried and dissolved in diethyl dicarbonate-treated water. The total RNA concentration was measured using a NanoDrop (Molecular Devices, LLC, Sunnyvale, CA, USA).
Complementary DNA (cDNA) was synthesized using the PrimeScript 1st Strand cDNA Synthesis Kit (TaKaRa). PCR PCR was performed using premix Taq (TaKaRa) and speci c primers. Primer sequences used to amplify UCP2 were designed based on GenBank sequences: sense primer sequence, 5′-GCCCGGGCTGGTGGTGGTC-3′ and antisense primer sequence, 5′-CCCCGAAGGCAGAAGTGAAGTGG-3′. PCR UCP2 ampli cation consisted of denaturation at 94°C for 2 min, followed by 25 cycles of 30 s at 95°C, 30 s at 58°C, and 30 s at 72°C, with a nal extension for 7 min at 72°C. The PCR products were analyzed using 2% agarose gel electrophoresis; the UCP2 PCR product size was 290 bp. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the control housekeeping gene. GAPDH ampli cation consisted of denaturation at 94°C for 30 s, followed by 30 cycles of 30 s at 94°C, 30 s at 50°C, and 30 s at 72°C, with a nal extension for 10 min at 72°C. The GAPDH sense primer sequence was 5′-ACCACAGTCCATGCCATCAC-3′ and the antisense primer sequence was 5′-TCCACCACCCTGCTGTA-3′. The PCR products were analyzed using 2% agarose gel electrophoresis; the GAPDH PCR product size was 450 bp.

Protein extraction and Western blot analysis
The 5-aza-treated cells were homogenized using a sonicator with lysis buffer containing protease inhibitors. Lysates were centrifuged at 12,000 rpm for 20 min at 4°C. Then, the protein lysates were transferred to a new tube. Total protein concentration was assessed using the BCA Protein Assay Kit (Thermo Fisher Scienti c, Rockford, IL, USA). Protein samples were boiled at 95°C for 5 min after adding 5× sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis loading buffer (25 mM Tris-HCL pH 6.8, 10% SDS, 50% glycerol, 0.5 M dithiothreitol, 0.5% bromophenol blue). After electrophoresis, the proteins were transferred to a nitrocellulose membrane and blocked with 3% skim milk (Sigma-Aldrich, St.

Measurement of DNMT activity
Nuclear proteins were isolated using the EpiQuik™ Nuclear Extraction Kit I (Epigentek, Brooklyn, NY, USA) from 5-aza-treated cells. After measuring the protein concentration with the BCA Protein Assay Kit (Thermo Fisher Scienti c), total DNMT activity was analyzed using EpiQuik™ DNA Methyltransferase Activity/Inhibition Assay (Epigentek). gDNA puri cation and bisul te modi cation Genomic DNA (gDNA) was isolated from 5-aza-treated cells using the Wizard® Genomic DNA Puri cation Kit (Promega Corp., Madison, WI, USA) according to the manufacturer's instructions. Extracted gDNA was modi ed using the EpiTect® Bisul te Kit (QIAGEN, Hilden, Germany) according to the manufacturer's instructions. During the experiment, DNA concentration was measured using a NanoDrop (Molecular Devices). Sodium bisul te-modi ed DNA was stored at −15 to −30°C.

MSP and direct PCR sequencing assays
Speci c methylated and unmethylated primers are required for MSP analysis. The primer design for nonsulfurized processing sequences is available on the MetPrimer site (http://www.urogene.org/methprimer2/). Speci c PCR conditions such as speci c methylated and unmethylated primer sequences, combined temperature, and number of cycles of target genes are presented in Table 1. Sodium bisul te-modi ed DNA was analyzed using the EpiScope® MSP Kit (TaKaRa) according to the manufacturer's instructions. The PCR products were analyzed using 3% agarose gel electrophoresis. The MSP samples were sent to Macrogen Corporation (Seoul, South Korea) for direct PCR sequencing.

Statistical analysis
All experiments were repeated ve to eight times. The data were expressed as the means ± standard error of the mean. All statistical analyses were performed using analysis of variance with the Statistical Analysis System (SAS) software (SAS Institute, Cary, NC, USA); each treatment was compared using the least-squares or Duncan method. A P-value < 0.05 indicated signi cant differences among treatments.

Results
UCP2 mRNA expression and basal promoter activity of human UCP2 in cancer cells UCP2 mRNA is not transcribed in Hep3B cells but is transcribed in HT-29 and HepG2 cells [11,24]. In the present study, the transcriptional difference of UCP2 mRNA among the three cancer cell lines was identi ed. To determine the difference in the transcription degree of UCP2 mRNA in these cells, total RNA was isolated without any treatment and reverse transcription PCR was performed. A SEAP reporter plasmid containing the UCP2 promoter region was constructed, transformed into each cell line, and basal promoter activity was measured. The SEAP activity in the Hep3B cells in which UCP2 mRNA was not expressed was 900 ± 100 SEAP activity. Conversely, SEAP activities in HT-29 and HepG2 cells in which UCP2 mRNA was expressed were 525 ± 25 and 350 ± 50 SEAP activity, respectively (Fig. 1b). The basal activity of the mouse UCP2 promoter was approximately two-fold higher in Hep3B cells than in HT-29 and HepG2 cells, indicating that transcription of UCP2 mRNA in Hep3B cells is inhibited by some other factor. The HT-29 cells showed slightly higher, albeit nonsigni cant, promoter activity than HepG2 cells.

Effects of 5-aza on UCP2, DNMT level, and DNMT activity in cancer cells
To investigate whether methylation in the UCP2 promoter affects UCP2 transcription, cells were treated with 5 or 10 µM 5-aza (a DNA demethylation agent) for 24, 48, or 72 h. In Hep3B cells, UCP2 was not expressed in the control group; however, UCP2 expression increased depending on the concentration and treatment time in the 5-aza-treated groups. In addition, UCP2 expression was highest in cells stimulated for 72 h with 10 µM 5-aza (Fig. 2a). DNMT is an enzyme that methylates DNA and affects gene expression [6,25]. In the present study, DNMT level and activity were measured to investigate whether the expression of UCP2 and DNMT are correlated. DNMT1 and DNMT3a expression decreased depending on the concentration and time of 5-aza treatment (Fig. 2a). DNMT1 and DNMT3a expression were not observed in Hep3B cells treated with 10 µM 5-aza for 48 h (Fig. 2a). The DNMT activity in Hep3B cells decreased depending on the concentration and treatment time in the 5-aza-treated groups, and the DNMT activity in cells treated with 10 µM 5-aza for 72 h was reduced approximately 2-fold compared with the control group (Fig. 2b).
The results showed that UCP2 was produced in the control group of HT-29 cells (Fig. 3a). The UCP2 expression in HT-29 cells was signi cantly increased depending on the 5-aza concentration and treatment time (Fig. 3a). DNMT1 and DNMT3a expression were signi cantly reduced depending on the 5-aza concentration and treatment time and were not detected in cells treated with 10 µM 5-aza (Fig. 3a). DNMT activity was signi cantly decreased in HT-29 cells treated with 10 µM 5-aza for 72 h (Fig. 3b).
HepG2 cells showed no difference in UCP2 expression level in the treatment group compared with the control group (Fig. 4a). However, DNMT1 and DNMT3a expression were signi cantly reduced depending on the 5-aza concentration and treatment time and were not detected in cells treated with 10 µM 5-aza (Fig. 4a). DNMT activity was also reduced approximately 2-fold in the 5-aza treatment groups compared with the control group ( Fig. 4a and b).

Analysis of methylation level in the UCP2 promoter region using MSP
To investigate the level of methylation in the region from − 842 to − 696 bp in the UCP2 promoter, a speci c unmethylated primer con rming the unmethylation degree and a speci c methylated primer con rming the methylation degree were created [26]. Among the UCP2 promoters, the − 842 to − 696 bp region was selected as the most appropriate according to the MetPrimer manual. Table 1 shows the primer sequences and PCR conditions. Because a speci c unmethylated primer was used in the control group of Hep3B cells, gDNA was not ampli ed in PCR (Fig. 5a). However, the degree of gDNA ampli cation was increased depending on the 5aza concentration. UCP2 gDNA was ampli ed when a speci c methylated primer was used in the control group of Hep3B cells. These results indicated the methyl group gradually disappeared due to 5-aza treatment and numerous methyl groups existed in the − 842 to − 696 bp region in the UCP2 promoter of the control group in Hep3B cells. Because a speci c unmethylated primer was used in the control group of HT-29 cells, gDNA was ampli ed and the increase was 5-aza concentration dependent (Fig. 5b). UCP2 gDNA was ampli ed when a speci c methylated primer was used in the control group of HT-29 cells. No signi cant difference was observed in the gDNA ampli cation results in the control group compared with the 5-aza treatment groups in HepG2 cells (Fig. 5c). The results in Fig. 5 show that the − 842 to − 696 bp region in the UCP2 promoter of Hep3B cells is multi-methylated and the degree of methylation in HT-29 cells appears to be lower than that in Hep3B cells. In addition, the degree of methylation in HepG2 cells was lower than that in HT-29 cells.
Analysis of the methylated sequence in the UCP2 promoter region using direct PCR sequencing MSP samples were sent to Macrogen Corporation and analyzed to examine the methylated sequence in the UCP2 promoter region. Subsequently, nucleotide sequence analysis of Hep3B cells showed that the − 714, −702, and − 697 bps were cytosine because they were not converted into uracil due to the presence of methyl groups (Fig. 6a). However, in Hep3B cells treated with 10 µM 5-aza, the − 714, −702, and − 697 bps in the UCP2 promoter were all thymine (Fig. 6a). Subsequent nucleotide sequence analysis of HT-29 cells showed the − 702 bp was cytosine (Fig. 6b). However, in HT-29 cells treated with 10 µM 5-aza, the − 714, −702, and − 697 bps in the UCP2 promoter were all thymine (Fig. 6b). Sequencing analysis results of HepG2 cells showed the − 714, −702, and − 697 bps were all thymine (Fig. 6c). However, in HepG2 cells treated with 10 µM 5-aza, the − 714, −702, and − 697 bps in the UCP2 promoter were all thymine (Fig. 6c). These results indicated a methyl group was bound to a total of three cytosines at − 714, −702, and − 697 bp in the UCP2 promoter region from − 842 to − 696 bp in Hep3B cells. In HT-29 cells, only the cytosine at − 702 bp was bound to a methyl group. However, in HepG2 cells, the methyl group was not bound to cytosines at − 714, −702, or − 697 bp.

Discussion
UCP2 expression is inhibited in Hep3B cells, a human-derived liver cancer cell line [11]. However, UCP2 production normally occurs in HepG2 cells but not in other liver cancer cells and colon cancer cells such as HT-29 cells [11,24]. In the present study, the UCP2 mRNA transcription and protein expression levels differed among the cell lines. Various factors can affect UCP2 production in cells; however, the activity of transcriptional stimulating factors is considered important. In the present study, a plasmid containing the promoter region of UCP2 was constructed, transformed into three cancer cell lines, and the activity of the UCP2 promoter measured. The basal promoter activity in Hep3B cells in which UCP2 was not produced was signi cantly higher compared with HT-29 and HepG2 cells in which UCP2 was normally produced. These results indicate that transcription factors involved in UCP2 production are functioning normally in Hep3B, HT-29, and HepG2 cells. However, because UCP2 production was suppressed only in Hep3B cells, another factor likely interferes with transcription in the UCP2 promoter in these cells.
DNA methylation, an epigenetic mechanism, does not alter the DNA sequence but usually attaches the methyl group to the cytosine base of CpG [5]. Methylation in the promoter region of a speci c gene prevents transcription by interfering with the binding of transcription stimulating factors [4,9,10]. DNMT is an enzyme that regulates DNA methylation [6] and catalyzes the transfer of methyl groups to cytosine in the CpG sequence [27]. In mammals, there are three DNMTs, DNMT1, DNMT3a, and DNMT3b [8]. DNMT1 (maintenance DNMT) is an enzyme that methylates the CpG region synthesized during the cell differentiation process [28,29]. DNMT3a and DNMT3b (de novo DNMTs) are enzymes that methylate DNA CpG dinucleotides in early cells before cellular differentiation [30,31]. Intracellular methylation can also be affected by a decrease in the activity of one of the three DNMTs [30,31]. Many studies have investigated DNMT function. Reportedly, DNMT1 can function as DNMT3a and DNMT3b, and DNMT3a and DNMT3b can also act as DNMT1 [32,33]. The transcription and protein production of DNMT increases during cell proliferation and is more pronounced in tumor cells than in normal cells. In addition, the DNMT production level in cancer cells is high [34]. DNMT affects the methylation of several genes, including p53, a tumor suppressor gene [35][36][37][38]. p53 protein is not expressed in Hep3B cells; however, inhibition of DNMT production induces p53 protein expression [39]. DNA demethylation caused by 5-aza activates the p53 signaling pathway and induces apoptosis [37]. Furthermore, inhibition of DNMT activity leads to demethylation of the p53 promoter, resulting in apoptosis [40]. In addition, DNA demethylation of p15 and p16 genes can inhibit HCC proliferation [6,41].
Based on the results presented in Fig. 1, we hypothesized that UCP2 expression in Hep3B cells is inhibited by DNA methylation, and that the UCP2 expression level in HT-29 and HepG2 cells may differ depending on the degree of methylation. Therefore, each cell line was treated with the DNMT inhibitor 5-aza [20,21] to observe UCP2 expression level and methylation changes in the UCP2 promoter region. In Hep3B and HT-29 cells, UCP2 expression was increased depending on the concentration of 5-aza; however, DNMT expression and activity were decreased (Fig. 2, 3). These results indicate that UCP2 and DNMT expression levels in Hep3B and HT-29 cells are inversely correlated. However, treatment with 5-aza signi cantly decreased DNMT expression and activity in HepG2 cells but did not affect UCP2 expression (Fig. 4), indicating that UCP2 and DNMT expression levels are not correlated in HepG2 cells.
The MSP test results showed that UCP2 gDNA in the control group of Hep3B cells was not ampli ed due to a methyl group at the UCP2 promoter in the − 842 to − 696 bp region; however, treatment with 5-aza signi cantly ampli ed UCP2 gDNA (Fig. 5a). In the control group of HT-29 cells, UCP2 gDNA was ampli ed in the UCP2 promoter in the − 842 to − 696 bp region, and 5-aza treatment further ampli ed gDNA (Fig. 5b). In the control group of HepG2 cells, UCP2 gDNA was ampli ed in the UCP2 promoter in the − 842 to − 696 bp region, and treatment with 5-aza did not signi cantly amplify gDNA (Fig. 5c). The results indicate that the UCP2 promoter region in Hep3B cells undergoes signi cant methylation compared with HT-29 cells, and the UCP2 promoter region in HT-29 cells is more methylated than in Hep2G cells. Future studies are necessary to investigate the degree of methylation for the entire UCP2 promoter region; however, the methylation of the sequences from − 842 to − 696 bp in the UCP2 promoter region in Hep3B cells likely inhibits UCP2 production.
MSP samples were analyzed to examine the methylated sequence. When gDNA is treated with sodium bisul te, the unmethylated cytosine sequence is transformed into uracil. When PCR is performed with this gDNA, the uracil sequence is converted into thymine (Fig. 7). The control group of Hep3B cells was treated with sodium bisul te and methyl groups were observed in cytosines at − 714, −702, and − 697 bp (Fig. 6a). However, in HT-29 cells, the methyl group was only observed in the cytosine at − 702 bp, and in HepG2 cells, the methyl group was not present (Fig. 6b, c). In future studies, methylation in the entire promoter sequence should be analyzed to con rm that the nucleotide sequences in the UCP2 promoter region from − 842 to − 696 bp are important sites for transcription stimulatory factors. The results of the present study showed that the UCP2 promoter region in Hep3B cells has numerous methylated sites compared with other cancer cells and that UCP2 expression is inhibited by methylation. Furthermore, treatment with 5-aza increased UCP2 level in Hep3B and HT-29 cells but not in HepG2 cells.
Several UCP2 functions remain unclear. In cancer cells, UCP2 may play a role in causing apoptosis [15,16]. Conversely, UCP2 has been shown to help tumor cells proliferate by inhibiting ROS production [13,14] and promote pancreatic cancer proliferation [43]. In the present study, the mechanism that induces the production of UCP2 in cancer cells was investigated and the results can be expanded in further studies on the function of UCP2 in cancer cells and possible use in treating cancer cells.

Conclusion
The UCP2 promoter in Hep3B cells is multi-methylated and the degree of UCP2 promoter methylation in HT-29 cells is lower than that in Hep3B cells. In addition, the degree of methylation in HepG2 cells is lower than that in HT-29 cells. Therefore, 5-aza treatment can increase the UCP2 expression level in Hep3B and HT-29 cells but not in HepG2 cells.