MicroRNA (miRNA) are non-coding small RNA that regulate gene expression. MiR-328 is reported to influence breast cancer resistance protein (BCRP) expression in cancer cells. As a large inter-individual difference in BCRP levels is observed in various human tissues, the contribution of miR-328 to these differences is of interest. We hypothesized that DNA methylation in the miR-328 promoter region is responsible for the difference in miR-328 levels, leading to inter-individual variability in BCRP levels in human placenta. The association between placental miR-328 and BCRP levels was analyzed, and then DNA methylation in the miR-328 5'-flanking region and regulatory mechanisms causing inter-individual differences in miR-328 and BCRP levels were examined. MiR-328 expression was significantly correlated with BCRP mRNA (Rs = -0.560, P < 0.01) and protein (Rs = -0.730, P < 0.01) levels. It was also up-regulated by the demethylating agent 5-aza-2’-deoxycytidine in BCRP-expressing cells. Luciferase assays with differentially methylated reporter constructs indicated that methylation in the miR-328 5’-flanking region including a predicted CpG island remarkably decreased transcriptional activity compared to that in unmethylated constructs. We selected CCAAT/enhancer binding protein α (C/EBPα), located within the predicted CpG island, by in silico analysis. To elucidate the role of C/EBPα in miR-328 expression, a chromatin immunoprecipitation assay, promoter deletion analysis, and electrophoretic mobility shift assay (EMSA) were performed. C/EBPα-binding site-truncated constructs showed significantly decreased promoter activity, and EMSA indicated that the C/EBPα-binding sites were located in the CpG island. Finally, the methylation patterns of several CpG dinucleotides proximal to two C/EBPα-binding sites in the miR-328 5’-flanking region were correlated negatively with miR-328 levels, and positively with BCRP levels in human placental samples. These results suggest that methylation patterns in the miR-328 5’-flanking region are involved in the inter-individual difference in BCRP levels in human placenta.
Citation: Saito J, Hirota T, Furuta S, Kobayashi D, Takane H, Ieiri I (2013) Association between DNA Methylation in the miR-328 5’-Flanking Region and Inter-individual Differences in miR-328 and BCRP Expression in Human Placenta. PLoS ONE8(8): e72906. https://doi.org/10.1371/journal.pone.0072906
Editor: Alfons Navarro, University of Barcelona, Spain
Received: April 12, 2013; Accepted: July 15, 2013; Published: August 21, 2013
Copyright: © 2013 Saito et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by Japan Society for the Promotion of Science KAKENHI Grant Number 23590186. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Breast cancer resistance protein (BCRP) is constitutively expressed in normal human tissues including the intestine, liver, blood–brain barrier, breast, and placenta, as well as tumor tissues, and is involved in multidrug resistance because BCRP acts as an efflux transporter of anti-cancer drugs . Previous studies have shown that expression levels of BCRP in human placentas were profoundly affected by genetic polymorphisms. Several non-synonymous single nucleotide polymorphisms (SNPs) in the BCRP gene (ABCG2) have been associated with reduced BCRP transport activity [2–7]. Based on these observations, the genetic variation in BCRP is considered to contribute to the inter-individual variability in the pharmacokinetics of BCRP substrate drugs. However, the mechanisms causing inter-individual differences in BCRP levels are not fully understood. Besides SNPs, transcription factors [8–14], epigenetic factors [15–19] and microRNAs (miRNAs)  play important roles in the regulation of BCRP expression.
MiRNA are small noncoding RNAs, approximately 20-25 nucleotides in length, which target the 3’-untranslated region (3’-UTR) of mRNA specifically to prevent translation of mRNA or to degrade mRNA . MiRNA expression levels in normal tissues show large inter-individual variability [21,22]. MiRNA also influence the expression of drug-metabolizing enzymes [23–27] and drug transporters [20,28–34], suggesting that the variations in miRNA expression contribute to the inter-individual differences in drug response. It was reported that miR-328 influenced drug efflux by repressing BCRP mRNA expression in breast cancer cells . Other studies demonstrated that miR-328 plays an inhibitory role in the proliferation of cancer cell lines and indicated that the suppression of miR-328 expression provides an advantage for tumor growth [35–39]. Studies have also showed that miR-328 is important in regulating BCRP expression, but little is known about its contribution to the variability in BCRP levels among individuals.
One-third of all human miRNA may be regulated by DNA methylation in CpG islands upstream . In fact, previous reports showed that miR-127, miR-124a, and miR-34 were down-regulated by epigenetic events such as DNA methylation [41,42]. It was also reported that hypermethylation in the miR-328 upstream region in urine specimens from bladder cancer patients was associated with tumor grade, stage and prognosis, and the reduced expression of miR-328 itself .
In this study, we focused on the inter-individual variability in BCRP and miR-328 levels in the human placenta. We analyzed the relationship between miR-328 and BCRP mRNA and protein levels. Furthermore, we evaluated the importance of methylation patterns to individual miR-328 levels.
The association between miR-328 and BCRP levels in the human placenta
The relationship between miR-328 and BCRP levels (mRNA and protein) was analyzed using 20 human placental samples. It was reported that some SNPs (i.e., C421A, G34A and C376T genotypes) significantly impacted on BCRP function [2–4]. To eliminate the influences of BCRP genetic variants, we selected samples with wild-type homozygotes for C421A, G34A and C376T. BCRP mRNA and protein levels both showed a significantly negative correlation with miR-328 levels (mRNA, Rs = -0.560, P = 0.00526; protein, Rs = -0.730, P = 0.000132) (Figure 1). Furthermore, an over 80-fold inter-individual difference was observed in miR-328 levels in the human placentas.
A single CpG island in the miR-328 5’-flanking region
We focused on DNA methylation in the miR-328 5’-flanking region. CpG islands were identified using CpGplot software . We detected one CpG island in the miR-328 5’-flanking region from bp -2459 to -2343 (Figure 2).
A single CpG island was predicted spanning bp -2459 to -2343 (criteria used: Island size > 100 bp, GC percent > 50.0%, ratio of observed (Obs) CpG sites to expected (Exp) CpG sites > 0.6). The vertical lines on the top horizontal line indicate the cytosine resides of CpGs and the black box indicates the identified CpG island. Numbers in the top panel represent nucleotide positions from miR-328. Second panel, distribution of observed/expected ratios of CpG dinucleotides; third panel, distribution of C+G; lower panel, identification of the putative CpG island within the -4.1 kb of analyzed sequence. The reference sequence was derived from NCBI sequence database (NC_000016.9, complement).
Up-regulation of miR-328 expression by treatment with a DNA demethylating agent in BCRP-expressing cell lines
To select BCRP-expressing cells, BCRP mRNA was measured. LS174T, HeLa, Caco-2, BeWo, HepG2 and K562 cells were treated with 5 µM 5-aza-2’-deoxycytidine (5-aza-dC, an inhibitor of DNA methylation) for 72 hours. All cell lines showed significantly increased miR-328 levels after the treatment (Figure 3).
Transcriptional activity in the differentially methylated miR-328 5’-flanking region
To identify important regions and to evaluate the effect of methylation on the transcriptional activity, luciferase assays using methylated and unmethylated constructs containing various segments of the miR-328 5’-flanking region were conducted (Figure 4). Activity levels were markedly higher in unmethylated constructs containing the regions (−4187 to +1), (−3928 to +1) and (-2428 to +1) than in methylated constructs. The construct with (-1395 to +1) showed significantly higher activity than that with (-482 to +1) in both a methylated and unmethylated state, suggesting that the region (-1395 to -482) was important for regulating the expression independent of DNA methylation. The activity in the unmethylated vector with (-3928 to +1) was significantly higher than with (-2428 to +1) (P < 0.05), and the unmethylated vector containing (-2428 to +1) showed significantly higher activity than that with (-2197 to +1) (P < 0.05). These results suggest that methylation between positions -3928 and -2197 was important for transcriptional activity.
Luciferase activity of unmethylated reporter constructs (opened bars) compared with the methylated constructs (filled bars) in BeWo cells. Results represent the mean ± SD for three independent experiments. Data are expressed as fold-increase in pGL4.10 activity, which was assigned a value of 1.00. *P < 0.05 vs. methylated control.
Inhibition of C/EBPα expression down-regulates miR-328 expression
Transcription factors capable of binding to the sequence from bp -3928 to -2197 of the miR-328 5’-flanking region were screened using TFSEARCH , TRANSFAC ver. 8.3 [46,47] and TFBIND . CCAAT/enhancer binding protein α (C/EBPα) was selected by all three programs. To confirm the involvement of C/EBPα in miR-328 expression, C/EBPα siRNA was transfected into BeWo cells. The C/EBPα mRNA level was completely reduced after transfection (Figure 5A). C/EBPα siRNA also significantly diminished miR-328 levels (Figure 5B).
Methylation-dependent C/EBPα-binding to the miR-328 5’-flanking region
To investigate the association between C/EBPα-binding frequency and methylation patterns in the miR-328 5’-flanking region, a chromatin immunoprecipitation (ChIP) analysis was performed. Figure 6A shows the C/EBPα-binding sites predicted using Transfac ver. 8.3. All the predicted sites were proximal to the CpG dinucleotide. C/EBPα-binding frequency was analyzed in the miR-328 5’-flanking region in BeWo cells using 5-aza-dC. For real-time PCR, ten primer sets (ChIP primers I to X, see Figure 6A) that amplified 5’-flanking regions including predicted C/EBPα-binding sites were designed. After treatment with 5-aza-dC, the C/EBPα-binding frequency was increased in four regions (ChIP primers IV, V, VI and VII) compared to DMSO-treated cells as a negative control (Figure 6B). C/EBPα-binding was also detected in ChIP primer X amplifying the region from -1196 to -1096 bp, but independent of the methylation. When we tested the C/EBPα-binding frequency in human placental samples, large inter-individual variability was observed in ChIP primers VI and X (n = 20) (Figure 6C).
(A) Schematic representation of predicted binding sites of C/EBPα in the miR-328 5’ flanking region. The position numbers indicate the 5’ end of the C/EBPα motif. Primers for the ChIP assay are also indicated.
(B) Comparison of C/EBPα-binding frequency in the miR-328 5’ flanking region. Open and closed squares indicate the results of 5-aza-dC and DMSO (control) treatment, respectively, in BeWo cells. Immunoprecipitated DNA was amplified with ten specific primers (I-X described as double-headed lines).
(C) C/EBPα-binding frequency in human placentas (n = 20).
Input, DNA isolated from the lysate before immunoprecipitation; IP, DNA immunoprecipitated with anti-C/EBPα antibody. Data are expressed as the mean ± SD for three independent experiments.
Promoter deletion analysis in the miR-328 5’-flanking region
To identify the functionally important C/EBPα-binding sites, we constructed a series of deletion plasmids for seven C/EBPα-binding sites (position -2871, -2810, -2429, -2329, -2293, -1173, and -1128, Figure 6A), and evaluated their transcriptional activity using luciferase assays. Deletion of (-2870 and -2609), (-2293 and -2282), and (-1135 and -1072) had no effect on the transcriptional activity. However, constructs lacking (-2440 to -2390), (-2372 to 2317) and (-1179 to -1151) showed extremely low transcriptional activity compared with the full-length vector (Figure 7).
Confirmation of C/EBPα-binding to predicted motif sequences
A gel mobility shift assay (EMSA) was performed to confirm that C/EBPα binds to the predicted motif sequences. We used biotin-labeled double-stranded DNA corresponding to the C/EBPα-binding sequences. The results showed that all three probes formed complexes with nuclear proteins from BeWo nuclear extracts (Figure 8). The specificity of these protein-DNA complexes was confirmed in competition assays using each unlabeled oligonucleotide. Anti-C/EBPα antibodies led to the formation of supershifted complexes, indicating that C/EBPα binds to the predicted motifs.
Correlation between C/EBPα-binding frequency and miR-328 levels in the human placenta
We analyzed the association between binding frequencies in the regions around the three C/EBPα-binding sites (Figure 9) and miR-328 levels in the human placenta (n = 20). Binding frequencies by ChIP primer VI (-2429 and -2329) and primer X (-1173) were significantly correlated with miR-328 levels (Figure 10). Notably, a significantly higher correlation was observed in the region analyzed using ChIP primer VI. These results suggest that the C/EBPα-binding frequency is a determinant of human placental miR-328 levels.
Putative C/EBPα-binding sites and CpG dinucleotides are indicated in bold capital letters. All numbers indicate the relative distance from the 5’ end of the miR-328 transcript. Each vertical bar represents the CpG dinucleotide closest to the C/EBPα-binding site (circle). The three probes for EMSA and two primers (VI and X) for ChIP assays are indicated by horizontal lines and double-headed lines, respectively.
Influence of methylation patterns and C/EBPα levels on C/EBPα-binding frequency in the human placenta
The association between the methylation levels in the CpG dinucleotides proximal to the C/EBPα-binding sites and C/EBPα levels was evaluated. We focused on the six CpG dinucleotides (-2456CG, -2421CG, -2344CG, -2306CG, -1184CG and -1155CG) close to identified C/EBPα-binding sites (Figure 9). Methylation patterns at three CpG dinucleotides (-2421CG, -2344CG and -2306CG) were negatively correlated with C/EBPα-binding frequency (Figure 11A). The role of C/EBPα levels in C/EBPα-binding was also evaluated. No correlation was found in the region amplified by ChIP primer VI (Figure 11B), but a significant correlation was observed using primer X (Figure 11C).
Influence of methylation patterns and C/EBPα levels on miR-328 levels in the human placenta
Methylation patterns at the three CpG dinucleotides (-2421CG, -2344CG and -2306CG) were also negatively correlated with miR-328 levels (Figure 12A). C/EBPα levels showed no correlation with miR-328 levels, indicating they were not crucial to the individual differences in miR-328 levels in the human placenta (Figure 12B).
Relationship between methylation patterns and placental BCRP mRNA and protein levels
The association between the methylation patterns of the three CpG dinucleotides (-2421CG, -2344CG and -2306CG) which correlated with C/EBPα-binding frequency and miR-328 levels, and BCRP levels was analyzed. The methylation level of -2421CG and -2344CG but not -2306CG positively correlated with BCRP mRNA levels (Figure 13). With regard to the protein levels, significantly positive correlations were observed at all three CpG dinucleotides. These results suggest that methylated CpG dinucleotides in the miR-328 5’-flanking region suppressed miR-328 expression, leading to BCRP up-regulation in the human placenta.
Human placenta shows the highest BCRP expression of all tissues . Thus, much effort has been made to describe the pharmacological roles of BCRP at the maternal-fetal interface [49,50]. BCRP is located mainly on the apical membrane in syncytiotrophoblasts where it pumps substrates into the maternal circulation. In mice, Bcrp was shown to limit fetal exposure to substrate drugs [51–54]. In human placental trophoblasts, BCRP gene polymorphisms play a limited role in the inter-individual variability of BCRP levels . In this study, to explore the mechanisms behind the variability in placental BCRP levels, we focused on the association between miR-328 and BCRP levels. As shown in Figure 1, miR-328 levels were significantly correlated with BCRP mRNA and protein levels. These observations are consistent with a previous report showing BCRP mRNA degradation by miR-328 [20,55].
Although it is now quite clear that BCRP plays an important role in protecting the fetus against potential toxicity of drugs and xenobiotics, few studies have been done to demonstrate the physiological role of BCRP in the placenta. Evseenko et al. showed that inhibition of BCRP activity by a BCRP-specific inhibitor Ko143 increased cytokine-induced apoptosis in primary trophoblasts and BeWo cells, and BCRP expression in placenta from pregnancies with idiopathic fetal growth restriction was significantly lower than that from normal pregnancies [56,57]. These authors proposed that BCRP may play an unrecognized role in the placenta, protecting trophoblasts against apoptosis induced by cytokines, and BCRP is likely a survival factor in differentiation of placental trophoblasts. In human placenta, the role of DNA methylation has been studied extensively. Several recent studies have examined the association between placental methylation change and neonatal outcome , and showed that placental miRNA expression profiles were associated with fetal growth . Decreased BCRP expression associated with high miR-328 expression resulting from the demethylation in 5’-flanking region may result in placental function deficit, thus contributing to fetal growth restriction. However, analyzed placental samples in this study had no information about the demographics, because the purpose of this study focused on the association analysis between BCRP and miR-328 expression. Further studies are needed to confirm this hypothesis.
A single CpG island was predicted in the miR-328 5’-flanking region. Since methylation in CpG islands is known to regulate gene expression, we analyzed methylation patterns of CpG dinucleotides in the island. Before the luciferase analysis, the effect of demethylation on miR-328 expression was tested using BCRP-expressing cells. In luciferase reporter assays in BeWo cells, we found that the region (-1395 to -482) regulates the expression, independent of DNA methylation. On the other hand, methylation in (-3928 to -2197), the region containing the CpG island, strongly suppressed the miR-328 expression. Next, we selected C/EBPα as a candidate transcription factor which binds to the CpG island, and confirmed its involvement in miR-328 expression using C/EBPα siRNA-transfected BeWo cells (Figure 5). C/EBPα knock-down was associated with an extremely low expression of miR-328, suggesting that C/EBPα is the major activator for miR-328 expression. To investigate the role of methylation in the binding of C/EBPα to the miR-328 5’-flanking region, the C/EBPα-binding frequency was determined in BeWo cells. Treatment of BeWo cells with 5-aza-dC increased the C/EBPα-binding frequency in four regions (ChIP primers IV, V, VI and VII) in the miR-328 5’-flanking region (Figure 6). The region with relatively high binding frequency under demethylation was approximately -3167 to -2184 (see Table S1). Interestingly, this region was close to (-3928 to -2197) showing significantly higher luciferase activity in unmethylated constructs (Figure 4). In human placental samples, the C/EBPα-binding frequency varied inter-individually in this region (Figure 6C). These results suggest that methylation from bp -3167 to 2184 in the miR-328 5’-flanking region is important for C/EBPα-mediated miR-328 expression.
By deletion analysis (Figure 7) and EMSA (Figure 8), positions -2429, -2329 and -1173 bp within the region showing relatively high C/EBPα-binding frequencies in BeWo cells were identified as binding sites for C/EBPα. In human placenta, a strong association between C/EBPα-binding frequencies and miR-328 levels was also observed for DNA fragments (including the C/EBPα-binding sites at -2429 and -2329) obtained using ChIP primer VI (Figure 10A, Rs = 0.869, P = 4.89E-07).
Among six CpG dinucleotides (-2456, -2421, -2344, -2306, -1184, and -1155) within the C/EBPα-binding sites, the methylation levels of the three CpG dinucleotides (-2421CG, -2344CG and -2306CG) proximal to the two C/EBPα-binding sites (-2429 and -2329 bp) were negatively correlated with C/EBPα-binding frequencies and miR-328 levels, but not for -1184CG and -1155CG dinucleotides proximal to the C/EBPα-binding site (-1173 bp). The absence of an association for the -1184CG and -1155CG dinucleotides was consistent with the results of luciferase assays suggesting that miR-328 expression is regulated by C/EBPα independently of DNA methylation. Since the C/EBPα-binding frequency for DNA fragments (including the C/EBPα-binding site at -1173 bp) obtained with ChIP primer X was significantly correlated with C/EBPα levels, there are two factors, methylation patterns and C/EBPα levels, behind the C/EBPα-binding frequency. However, as shown in Figure 12B, miR-328 levels were not directly affected by C/EBPα levels.
Therefore, the DNA methylation pattern of three CpG dinucleotides (-2421CG, -2344CG and -2306CG) plays a major role in regulating the binding of C/EBPα to the miR-328 5’-flanking region, leading to inter-individual differences in miR-328 levels in the human placenta.
Finally, we showed that methylation levels at the three CpG dinucleotides (-2421CG, -2344CG and -2306CG) were significantly correlated with BCRP protein levels in human placentas (Figure 13). Methylation status in the human placenta is profoundly affected by environmental factors and drug exposure, due to its role at the interface between the maternal and fetal circulations [60–63]. Interestingly, it was reported that drug-induced hyper-methylation involves resistance to azidothymidine , etoposide, nalidixic acid, doxorubicin, vinblastine, colchicine, cisplatin, 5-fluorouracil, 5-fluorodeoxyuridine, and methotrexate in human cancer cell lines . DNA methylation in the miR-328 5’-flanking region is suggested to be associated with an increase of BCRP expression, leading to drug resistance in cancers. In this study, analyzed placental samples were relatively small size (n = 20). Additional study using larger sample size is needed to determine the role of DNA methylation in the inter-individual differences in miR-328 and BCRP expression. However, our results are important first step to investigate how DNA methylation in the miR-328 5’-flanking region, and to identify the association between miR-328 and BCRP expression levels in human placenta.
Materials and Methods
Twenty human placental samples were obtained from patients who underwent a normal pregnancy at Tottori University Hospital. All placental samples were excised from each individual placenta. Placenta tissues were macroscopically separated from decidua. Tissues were frozen in liquid nitrogen immediately after delivery and stored at -80°C.
Each patient gave written informed consent for her sample to be used for scientific research, which was approved by the Tottori University Ethics Committee.
Genotyping of BCRP
The procedure used to genotype BCRP G34A, C376T, and C421A mutations was described previously .
RNA Extraction and cDNA Synthesis
Total RNA extraction and RT-PCR procedures for placental samples were described elsewhere . RT was performed in a 20-µL reaction mixture containing 5 µg of total RNA in 1×First-strand Buffer, 25 mM DTT, 0.5 µg of the random primers (Promega, Madison, WI), 2 mM of each deoxynucleoside-5’-triphosphate, and SuperScript II RNase H- reverse transcriptase (Invitrogen, Carlsbad, CA). Samples were incubated at 42°C for 1 h. As a negative control, template RNA was processed without the reverse transcriptase.
Quantification of BCRP mRNA and protein levels
The quantitative real-time PCR for BCRP mRNA levels and western blotting for BCRP protein levels were performed as described previously .
TaqMan-based real-time quantification of miRNA
The expression of mature miR-328 was quantified by quantitative real-time PCR using TaqMan miRNA assays according to the manufacturer’s protocols (Applied Biosystems, Foster City, CA). U6 small nuclear RNA (RNU6B; Applied Biosystems) served as an internal normalized reference. The relative quantity for miR-328 was determined by the comparative Ct method (2 –ΔΔCt).
The LS174T, Caco-2, BeWo, HepG2 and K562 cell lines were obtained from Dainippon Pharmaceuticals (Osaka, Japan). The HeLa cell line was obtained from the RIKEN, Cell Bank (Tsukuba, Japan). All cell lines were grown at 37°C in an incubator with 5% CO2. The LS174T cell line was maintained in Eagle’s minimum essential medium. The Caco-2, HeLa and HepG2 cell lines were maintained in Dulbecco’s modified Eagle’s medium. The BeWo cell line was maintained in Ham’s F12 Kaighn’s modification medium. The K562 cell line was maintained in RPMI 1640 supplemented with 2.5 mM glutamine (Life Technologies, Carlsbad, CA). All media were supplemented with 10% fetal bovine serum.
To block DNA methylation, all cells were treated with 5 µM 5-aza-2’-deoxycytidine (5-aza-dC; Sigma Aldrich, St Louis, MO) and DMSO (0.01%, final concentration) as vehicle control for 72 h. The cells were dissociated enzymatically using TrypLE Express (Invitrogen), washed, and resuspended in phosphate-buffered saline (PBS). Small RNA was immediately extracted from the collected cells using the mirVana isolation Kit (Ambion, Austin, TX) according to the manufacturer’s protocol.
DNA extraction and bisulfite sequencing
Genomic DNA was extracted from 20 placental tissues and then subjected to sodium bisulfite treatment using an Epitect bisulfite modification kit (Qiagen, Valencia, CA). We amplified and sequenced the proximal 5’-flanking miR-328 for methylation analysis. These regions were amplified using two primer sets (primer 1 for bp -2510 to -2145 and primer 2 for bp -1287 to -918, see Table S2). The PCR product amplified with primer 1 was used for the nested second PCR (primer 1s). The initial denaturation was for 9 min at 95°C, followed by 45 cycles for 40 s at 98°C, 45 s at 54°C and 20 s at 72°C and a final elongation for 5 min at 72°C. The PCR products were separated by 3% agarose gel electrophoresis, extracted and then cloned into the pGEM-T easy vector (Promega). After bacterial amplification of the cloned PCR fragments by standard procedures, 10 clones were subjected to DNA sequencing.
Construction of 5’-flanking reporter plasmids
To explore the transcriptional regulation of miR-328, we generated a longer fragment of the miR-328 5’-flanking region (-4230 to +36 relative to the transcription start site of miR-328) using two sets of primers (described as ins1 and ins2 in Table S3). pGL4-(-4230/+36) was used as a template to generate a series of vectors of various lengths. Each amplified fragment was digested and inserted into the pGL4.10 basic vector (Promega) between the KpnI and XhoI sites (primer sets described in Table S3). These plasmids were then sequenced for confirmation.
In vitro DNA methylation of luciferase reporter constructs
The series of vectors were methylated in vitro using the bacterial methylase SssI (New England Biolabs, Ipswich, MA). To generate completely-methylated vectors, each vector was digested with KpnI and XhoI to release the inserted sequence. Each fragment (4 µg) was incubated at 37°C for 4 h with 40 U of SssI supplemented with 160 µM S-adenosylmethionine. After methylation, each fragment was religated with the pGL4.10 vector. The unmethylated insert sequence was extracted from E. coli JM110 (a dam and dcm DNA methylase-deficient strain).
Cell transfection and luciferase assays
BeWo cells were transfected using Lipofectamine 2000 (Invitrogen) with 400 ng of each methylated or unmethylated vector. In each case, 100 ng of pRL-TK (Promega) was used to correct for the transfection efficiency. Luciferase activity was measured with the Dual-Luciferase Reporter Assay System (Promega). 5’-flanking activities were expressed as the ratio of Firefly luciferase to Renilla luciferase activity.
Chromatin immunoprecipitation assay
Chromatin immunoprecipitation (ChIP) assays were performed using the immunoprecipitation kit (Roche Diagnostics, Germany) according to the manufacturer’s instructions. Briefly, BeWo cells were grown to 80% confluence in Ham’s F12 Kaighn’s modification medium supplemented with 10% fetal bovine serum. Cells were washed with PBS twice. DMSO (control) and 5-aza-dC-treated cells were fixed using 1% formaldehyde on ice for 5 min. Cells were incubated for 10 min in a 2.5 M glycine solution. After being re-suspended in SDS lysis buffer, the cells were sonicated until crosslinked chromatin was sheared to an average DNA fragment length of 200–500 bp. The sonicated lysate (200 µL) was used to quantify the total amount of DNA before immunoprecipitation (inputs). Immunoprecipitation was performed for 12-16 h at 4°C with either 1 µg of anti-C/EBPα antibody (Santa Cruz Biotechnology, Santa Cruz, CA) or a non-specific antibody (IgG). Precipitates were then washed three times with 1 mL of TE buffer and extracted with a DNeasy Mini kit (Qiagen). The average cycle threshold (Ct) was analyzed according to the 2(−ΔΔCt) method. The resulting data were divided by their corresponding input DNA. Fold enrichment was calculated with the ratio from the IgG control as a negative control to adjust the amplification efficiency of each primer. The primers and amplified regions are shown in Table S1.
siRNA knock-down assay
Cells were plated in a 24-well plate at a density of 1.0× 105 cells/mL/well. The transfection mix was prepared in Opti-MEM with siRNA and lipofectamine 2000 according to the manufacturer’s directions. The final concentrations of siRNA and lipofectamine added to the cells were 20-50 nM and 2 µL/ml, respectively. Cells were cultured in the presence of the transfection mixture for 48 h. After siRNA transfection, cells were harvested to analyze target genes expression levels and miR-328 levels by real-time PCR. The primers and amplified regions are shown in Table S4.
Nuclear protein extraction
Nuclear extracts were prepared from cells using the NE-PER Nuclear and Cytoplasmic Extraction Reagents (Pierce Biotechnology, Rockford, IL). The protein concentration of each extract was measured at OD 595 nm. Each nuclear extract (5 µg) was subjected to an electrophoretic mobility shift assay.
Electrophoretic mobility shift assay (EMSA)
3’ end biotin-labeled double-stranded oligonucleotides containing the binding sites for C/EBPα were purchased from Hokkaido System Science (Sapporo, Japan). The oligonucleotide sequences used for EMSA are listed in Table S5. EMSA was performed using a LightShift Chemiluminescent EMSA kit (Pierce Biotechnology). For each gel shift reaction (20 µL), a total of 20 fmol of labeled probe was combined with 4 µg of nuclear extract prepared from BeWo cells. Binding reactions were performed by incubating the samples for 20 min at room temperature. Protein-DNA complexes were separated from the free DNA probes by electrophoresis through 6% native polyacrylamide gels containing 0.5×Tris borate/EDTA. The gel was transferred to a nylon membrane (Thermo scientific, Lafayette, CO) and fixed for 15 min by use of a UV trans-illuminator (UVP Inc., Upland, CA). The biotin-labeled DNA was detected by addition of a streptavidin-horseradish peroxidase conjugate and chemiluminescent substrate. The competition assay was performed using a 250-fold amount of unlabeled probes with the reaction mixture before addition of biotin-labeled probes. For the super-shift assay, samples were further incubated with anti-C/EBPα antibody (sc-61X; Santa Cruz).
Data from at least three independent experiments are expressed as the mean ± standard error of the mean. Standard error bars are included for all data points except for scatter plots. The differences between groups were analyzed using Student’s t-test when only two groups were present. The correlation analysis was performed using the Spearman’s correlation test (Rs). All tests performed were one-side. Data were considered significant if P < 0.05 (indicated by ‘*’).
Primers and positions for ChIP assay.
Primers for bisulfite sequencing in the miR-328 5'-flanking region.
Primers for construction of 5’-flanking reporter plasmids.
siRNA sequences for knock-down assay.
Conceived and designed the experiments: TH II. Performed the experiments: JS TH SF DK HT. Analyzed the data: JS TH. Contributed reagents/materials/analysis tools: JS TH HT. Wrote the manuscript: JS TH II.
- 1. Allen JD, Schinkel AH (2002) Multidrug resistance and pharmacological protection mediated by the breast cancer resistance protein (BCRP/ABCG2). Mol Cancer Ther 1: 427-434. doi:10.4161/cbt.1.4.25. PubMed: 12477055.
- 2. Imai Y, Nakane M, Kage K, Tsukahara S, Ishikawa E et al. (2002) C421A polymorphism in the human breast cancer resistance protein gene is associated with low expression of Q141K protein and low-level drug resistance. Mol Cancer Ther 1: 611-616. PubMed: 12479221.
- 3. Kobayashi D, Ieiri I, Hirota T, Takane H, Maegawa S et al. (2005) Functional assessment of ABCG2 (BCRP) gene polymorphisms to protein expression in human placenta. Drug Metab Dispos 33: 94-101. PubMed: 15475413.
- 4. Kondo C, Suzuki H, Itoda M, Ozawa S, Sawada J et al. (2004) Functional analysis of SNPs variants of BCRP/ABCG2. Pharm Res 21: 1895-1903. doi:10.1023/B:PHAM.0000045245.21637.d4. PubMed: 15553238.
- 5. Mizuarai S, Aozasa N, Kotani H (2004) Single nucleotide polymorphisms result in impaired membrane localization and reduced ATPase activity in multidrug transporter ABCG2. Int J Cancer 109: 238-246. doi:10.1002/ijc.11669. PubMed: 14750175.
- 6. Yoshioka S, Katayama K, Okawa C, Takahashi S, Tsukahara S et al. (2007) The identification of two germ-line mutations in the human breast cancer resistance protein gene that result in the expression of a low/non-functional protein. Pharm Res 24: 1108-1117. doi:10.1007/s11095-007-9235-2. PubMed: 17373578.
- 7. Tamura A, Wakabayashi K, Onishi Y, Takeda M, Ikegami Y et al. (2007) Re-evaluation and functional classification of non-synonymous single nucleotide polymorphisms of the human ATP-binding cassette transporter ABCG2. Cancer Sci 98: 231-239. doi:10.1111/j.1349-7006.2006.00371.x. PubMed: 17297656.
- 8. Ee PL, Kamalakaran S, Tonetti D, He X, Ross DD et al. (2004) Identification of a novel estrogen response element in the breast cancer resistance protein (ABCG2) gene. Cancer Res 64: 1247-1251. doi:10.1158/0008-5472.CAN-03-3583. PubMed: 14973083.
- 9. Wang X, Wu X, Wang C, Zhang W, Ouyang Y et al. (2010) Transcriptional suppression of breast cancer resistance protein (BCRP) by wild-type p53 through the NF-kappaB pathway in MCF-7 cells. FEBS Lett 584: 3392-3397. doi:10.1016/j.febslet.2010.06.033. PubMed: 20600004.
- 10. Pradhan M, Bembinster LA, Baumgarten SC, Frasor J (2010) Proinflammatory cytokines enhance estrogen-dependent expression of the multidrug transporter gene ABCG2 through estrogen receptor and NFκB cooperativity at adjacent response elements. J Biol Chem 285: 31100-31106. doi:10.1074/jbc.M110.155309. PubMed: 20705611.
- 11. Nakamichi N, Morii E, Ikeda J, Qiu Y, Mamato S et al. (2009) Synergistic effect of interleukin-6 and endoplasmic reticulum stress inducers on the high level of ABCG2 expression in plasma cells. Lab Invest 89: 327-336. doi:10.1038/labinvest.2008.157. PubMed: 19139722.
- 12. Krishnamurthy P, Ross DD, Nakanishi T, Bailey-Dell K, Zhou S et al. (2004) The stem cell marker Bcrp/ABCG2 enhances hypoxic cell survival through interactions with heme. J Biol Chem 279: 24218-24225. doi:10.1074/jbc.M313599200. PubMed: 15044468.
- 13. Vore M, Leggas M (2008) Progesterone acts via progesterone receptors A and B to regulate breast cancer resistance protein expression. Mol Pharmacol 73: 613-615. PubMed: 18086804.
- 14. Wang H, Lee EW, Zhou L, Leung PC, Ross DD et al. (2008) Progesterone receptor (PR) isoforms PRA and PRB differentially regulate expression of the breast cancer resistance protein in human placental choriocarcinoma BeWo cells. Mol Pharmacol 73: 845-854. PubMed: 18042733.
- 15. To KK, Zhan Z, Bates SE (2006) Aberrant promoter methylation of the ABCG2 gene in renal carcinoma. Mol Cell Biol 26: 8572-8585. doi:10.1128/MCB.00650-06. PubMed: 16954373.
- 16. Turner JG, Gump JL, Zhang C, Cook JM, Marchion D et al. (2006) ABCG2 expression, function, and promoter methylation in human multiple myeloma. Blood 108: 3881-3889. doi:10.1182/blood-2005-10-009084. PubMed: 16917002.
- 17. Calcagno AM, Fostel JM, To KK, Salcido CD, Martin SE et al. (2008) Single-step doxorubicin-selected cancer cells overexpress the ABCG2 drug transporter through epigenetic changes. Br J Cancer 98: 1515-1524. doi:10.1038/sj.bjc.6604334. PubMed: 18382425.
- 18. Nakano H, Nakamura Y, Soda H, Kamikatahira M, Uchida K et al. (2008) Methylation status of breast cancer resistance protein detected by methylation-specific polymerase chain reaction analysis is correlated inversely with its expression in drug-resistant lung cancer cells. Cancer 112: 1122-1130. doi:10.1002/cncr.23285. PubMed: 18219662.
- 19. To KK, Polgar O, Huff LM, Morisaki K, Bates SE (2008) Histone modifications at the ABCG2 promoter following treatment with histone deacetylase inhibitor mirror those in multidrug-resistant cells. Mol Cancer Res 6: 151-164. doi:10.1158/1541-7786.MCR-07-0175. PubMed: 18234970.
- 20. Pan YZ, Morris ME, Yu AM (2009) MicroRNA-328 negatively regulates the expression of breast cancer resistance protein (BCRP/ABCG2) in human cancer cells. Mol Pharmacol 75: 1374-1379. doi:10.1124/mol.108.054163. PubMed: 19270061.
- 21. Hirota T, Date Y, Nishibatake Y, Takane H, Fukuoka Y et al. (2012) Dihydropyrimidine dehydrogenase (DPD) expression is negatively regulated by certain microRNAs in human lung tissues. Lung Cancer 77: 16-23. doi:10.1016/j.lungcan.2012.05.027. PubMed: 22306127.
- 22. Yokoi T, Nakajima M (2011) Toxicological implications of modulation of gene expression by microRNAs. Toxicol Sci 123: 1-14. doi:10.1093/toxsci/kfr168. PubMed: 21715665.
- 23. Komagata S, Nakajima M, Takagi S, Mohri T, Taniya T et al. (2010) Human CYP24 catalyzing the inactivation of calcitriol is post-transcriptionally regulated by miR-125b. Mol Pharmacol 76: 702-709. PubMed: 19570947.
- 24. Tsuchiya Y, Nakajima M, Kyo S, Kanaya T, Inoue M et al. (2006) Human CYP1B1 is regulated by estradiol via estrogen receptor. Cancer Res 15: 9090-9098. PubMed: 15126349.
- 25. Mishra PJ, Song B, Mishra PJ, Wang Y, Humeniuk R et al. (2010) MiR-24 tumor suppressor activity is regulated independent of p53 and through a target site polymorphism. PLOS ONE 24: e8445. PubMed: 20041160.
- 26. Boni V, Bitarte N, Cristobal I, Zarate R, Rodriguez J et al. (2010) miR-192/miR-215 influence 5-fluorouracil resistance through cell cycle-mediated mechanisms complementary to its post-transcriptional thymidilate synthase regulation. Mol Cancer Ther 9: 2265-2275. doi:10.1158/1535-7163.MCT-10-0061. PubMed: 20647341.
- 27. Rayner KJ, Esau CC, Hussain FN, McDaniel AL, Marshall SM et al. (2010) Inhibition of miR-33a/b in non-human primates raises plasma HDL and lowers VLDL triglycerides. Science 18: 1570-1573. PubMed: 22012398.
- 28. Ikemura K, Yamamoto M, Miyazaki S, Mizutani H, Iwamoto T et al. (2013) Microrna-145 Post-Transcriptionally Regulates the Expression and Function of P-Glycoprotein In Intestinal Epithelial Cells. Mol Pharmacol 83: 399-405. doi:10.1124/mol.112.081844. PubMed: 23166305.
- 29. Liang Z, Wu H, Xia J, Li Y, Zhang Y et al. (2009) Involvement of miR-326 in chemotherapy resistance of breast cancer through modulating expression of multidrug resistance-associated protein 1. Biochem Pharmacol 15: 817-824. PubMed: 19883630.
- 30. Zhu H, Wu H, Liu X, Evans BR, Medina DJ et al. (2008) Role of MicroRNA miR-27a and miR-451 in the regulation of MDR1/P-glycoprotein expression in human cancer cells. Biochem Pharmacol 76: 582-588. doi:10.1016/j.bcp.2008.06.007. PubMed: 18619946.
- 31. Bao L, Hazari S, Mehra S, Kaushal D, Moroz K et al. (2012) Increased expression of P-glycoprotein and doxorubicin chemoresistance of metastatic breast cancer is regulated by miR-298. Am J Pathol 180: 2490-2503. doi:10.1016/j.ajpath.2012.02.024. PubMed: 22521303.
- 32. Kovalchuk O, Filkowski J, Meservy J, Ilnytskyy Y, Tryndyak VP et al. (2008) Involvement of microRNA-451 in resistance of the MCF-7 breast cancer cells to chemotherapeutic drug doxorubicin. Mol Cancer Ther 7: 2152-2159. doi:10.1158/1535-7163.MCT-08-0021. PubMed: 18645025.
- 33. Pogribny IP, Filkowski JN, Tryndyak VP, Golubov A, Shpyleva SI et al. (2010) Alterations of microRNAs and their targets are associated with acquired resistance of MCF-7 breast cancer cells to cisplatin. Int J Cancer 15: 1785-1794. PubMed: 20099276.
- 34. Malzkorn B, Wolter M, Liesenberg F, Grzendowski M, Stühler K et al. (2010) Identification and functional characterization of microRNAs involved in the malignant progression of gliomas. Brain Pathol 20: 539-550. doi:10.1111/j.1750-3639.2009.00328.x. PubMed: 19775293.
- 35. Li WQ, Li YM, Tao BB, Lu YC, Hu GH et al. (2010) Downregulation of ABCG2 expression in glioblastoma cancer stem cells with miRNA-328 may decrease their chemoresistance. Med Sci Monit 16: 27-30. PubMed: 20885358.
- 36. Xu XT, Xu Q, Tong JL, Zhu MM, Nie F et al. (2012) MicroRNA expression profiling identifies miR-328 regulates cancer stem cell-like SP cells in colorectal cancer. Br J Cancer 106: 1320-1330. doi:10.1038/bjc.2012.88. PubMed: 22453125.
- 37. Tokarz P, Blasiak J (2012) The role of MiRNA in metastatic colorectal cancer and its significance in cancer prognosis and treatment. Acta Biochim Pol 59: 467-474. PubMed: 23173124.
- 38. Wu Z, Sun L, Wang H, Yao J, Jiang C, Xu W et al. (2012) MiR-328 expression is decreased in high-grade gliomas and is associated with worse survival in primary glioblastoma. PLOS ONE 7: e47270. doi:10.1371/journal.pone.0047270. PubMed: 23077581.
- 39. Zhou X, Ruan J, Wang G, Zhang W (2007) Characterization and identification of microRNA core promoters in four model species. PLOS Comput Biol 9: e37. PubMed: 17352530.
- 40. Lujambio A, Esteller M (2007) CpG island hypermethylation of tumor suppressor microRNAs in human cancer. Cell Cycle 6: 1455-1459. PubMed: 17581274.
- 41. Corney DC, Hwang CI, Matoso A, Vogt M, Flesken-Nikitin A et al. (2010) Frequent downregulation of miR-34 family in human ovarian cancers. Clin Cancer Res 16: 1119-1128. doi:10.1158/1078-0432.CCR-09-2642. PubMed: 20145172.
- 42. Dudziec E, Miah S, Choudhry HM, Owen HC, Blizard S et al. (2011) Hypermethylation of CpG islands and shores around specific microRNAs and mirtrons is associated with the phenotype and presence of bladder cancer. Clin Cancer Res 17: 1287-1296. doi:10.1158/1078-0432.CCR-10-2017. PubMed: 21138856.
- 43. Ceckova M, Libra A, Pavek P, Nachtigal P, Brabec M (2006) Expression and functional activity of breast cancer resistance protein (BCRP, ABCG2) transporter in the human choriocarcinoma cell line BeWo. Clin Exp Pharmacol Physiol 33: 58-65. doi:10.1111/j.1440-1681.2006.04324.x. PubMed: 16445700.
- 44. Rice P, Longden I, Bleasby A (2000) EMBOSS: the European Molecular Biology Open Software Suite. Trends Genet 16: 276-277. doi:10.1016/S0168-9525(00)02024-2. PubMed: 10827456.
- 45. Heinemeyer T, Wingender E, Reuter I, Hermjakob H, Kel AE et al. (1998) Databases on Transcriptional Regulation: TRANSFAC, TRRD, and COMPEL. Nucleic Acids Res 26: 364-370.
- 46. Messeguer X, Escudero R, Farré D, Núñez O, Martínez J et al. (2002) PROMO: detection of known transcription regulatory elements using species-tailored searches. Bioinformatics 18: 333-334. doi:10.1093/bioinformatics/18.2.333. PubMed: 11847087.
- 47. Farré D, Roset R, Huerta M, Adsuara JE, Roselló L et al. (2003) Identification of patterns in biological sequences at the ALGGEN server: PROMO and MALGEN. Nucleic Acids Res 31: 3651-3653. doi:10.1093/nar/gkg605. PubMed: 12824386.
- 48. Tsunoda T, Takagi T (1999) Estimating Transcription Factor Bindability on DNA. Bioinformatics 15: 622-630. doi:10.1093/bioinformatics/15.7.622. PubMed: 10487870.
- 49. Mao Q (2008) BCRP/ABCG2 in the placenta: expression, function and regulation. Pharm Res 25: 1244-1255. doi:10.1007/s11095-008-9537-z. PubMed: 18202831.
- 50. Hahnova-Cygalova L, Ceckova M, Staud F (2011) Fetoprotective activity of breast cancer resistance protein (BCRP, ABCG2): expression and function throughout pregnancy. Drug Metab Rev 43: 53-68. doi:10.3109/03602532.2010.512293. PubMed: 20854129.
- 51. Jonker JW, Buitelaar M, Wagenaar E, Van Der Valk MA, Scheffer GL et al. (2002) The breast cancer resistance protein protects against a major chlorophyll-derived dietary phototoxin and protoporphyria. Proc Natl Acad Sci U S A 99: 15649-15654. doi:10.1073/pnas.202607599. PubMed: 12429862.
- 52. Zhang Y, Wang H, Unadkat JD, Mao Q (2007) Breast cancer resistance protein 1 limits fetal distribution of nitrofurantoin in the pregnant mouse. Drug Metab Dispos 35: 2154-2158. doi:10.1124/dmd.107.018044. PubMed: 17785426.
- 53. Enokizono J, Kusuhara H, Sugiyama Y (2007) Effect of breast cancer resistance protein (Bcrp/Abcg2) on the disposition of phytoestrogens. Mol Pharmacol 72: 967-975. doi:10.1124/mol.107.034751. PubMed: 17644650.
- 54. Zhou L, Naraharisetti SB, Wang H, Unadkat JD, Hebert MF et al. (2008) The breast cancer resistance protein (Bcrp1/Abcg2) limits fetal distribution of glyburide in the pregnant mouse: an Obstetric-Fetal Pharmacology Research Unit Network and University of Washington Specialized Center of Research Study. Mol Pharmacol 73: 949-959. PubMed: 18079276.
- 55. Li X, Pan YZ, Seigel GM, Hu ZH, Huang M et al. (2010) Breast cancer resistance protein BCRP/ABCG2 regulatory microRNAs (hsa-miR-328, -519c and -520h) and their differential expression in stem-like ABCG2+ cancer cells. Biochem Pharmacol 15: 783-792. PubMed: 21219875.
- 56. Evseenko DA, Murthi P, Paxton JW, Reid G, Emerald BS et al. (2007) The ABC transporter BCRP/ABCG2 is a placental survival factor, and its expression is reduced in idiopathic human fetal growth restriction. FASEB J 21: 3592-3605. doi:10.1096/fj.07-8688com. PubMed: 17595345.
- 57. Evseenko DA, Paxton JW, Keelan JA (2007) The xenobiotic transporter ABCG2 plays a novel role in differentiation of trophoblast-like BeWo cells. Placanta 28 (Suppl A): S116-S120. doi:10.1016/j.placenta.2006.12.003. PubMed: 17275084.
- 58. Novakovic B, Saffery R (2012) The ever growing complexity of placental epigenetics - role in adverse pregnancy outcomes and fetal programming. Placenta 33: 959-970. doi:10.1016/j.placenta.2012.10.003. PubMed: 23102655.
- 59. Maccani MA, Padbury JF, Marsit CJ (2011) miR-16 and miR-21 expression in the placenta is associated with fetal growth. PLOS ONE, 6: e21210. PubMed: 21698265.
- 60. Jansson T, Powell TL (2007) Role of the placenta in fetal programming: underlying mechanisms and potential interventional approaches. Clin Sci 113: 1-13. doi:10.1042/CS20060339. PubMed: 17536998.
- 61. Yuen RK, Avila L, Peñaherrera MS, von Dadelszen P, Lefebvre L et al. (2009) Human placental-specific epipolymorphism and its association with adverse pregnancy outcomes. PLOS ONE, 4: e7389. PubMed: 19838307.
- 62. Novakovic B, Yuen RK, Gordon L, Penaherrera MS, Sharkey A et al. (2011) Evidence for widespread changes in promoter methylation profile in human placenta in response to increasing gestational age and environmental/stochastic factors. BMC Genomics 12: 529. doi:10.1186/1471-2164-12-529. PubMed: 22032438.
- 63. Novakovic B, Gordon L, Robinson WP, Desoye G, Saffery R (2012) Glucose as a fetal nutrient: dynamic regulation of several glucose transporter genes by DNA methylation in the human placenta across gestation. J Nutr Biochem 24: 282-288. PubMed: 22901689.
- 64. Nyce J, Leonard S, Canupp D, Schulz S, Wong S (1993) Epigenetic mechanisms of drug resistance: drug-induced DNA hypermethylation and drug resistance. Proc Natl Acad Sci U S A 90: 2960-2964. doi:10.1073/pnas.90.7.2960. PubMed: 8464912.
- 65. Nyce J (1989) Drug-induced DNA hypermethylation and drug resistance in human tumors. Cancer Res 49: 5829-5836. PubMed: 2790794.
- 66. Tanabe M, Ieiri I, Nagata N, Inoue K, Ito S et al. (2001) Expression of P-glycoprotein in human placenta: relation to genetic polymorphism of the multidrug resistance (MDR)-1 gene. J Pharmacol Exp Ther 297: 1137-1143. PubMed: 11356939.