Methylation dictates PI.f-specific CYP19 transcription in human glial cells

Molecular and Cellular Endocrinology 452 (2017) 131e137

Contents lists available at ScienceDirect

Molecular and Cellular Endocrinology journal homepage: www.elsevier.com/locate/mce

Methylation dictates PI.f-specific CYP19 transcription in human glial cells Wenjuan Tan a, Zhiping Zhu b, Lan Ye b, *, Lai K. Leung a, c, ** a

Biochemistry Programme, School of Life Sciences, Faculty of Science, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong State Key Laboratory of Reproductive Medicine, Nanjing Medical University, Nanjing 211166, China c Food and Nutritional Sciences Programme, School of Life Sciences, Faculty of Science, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 13 January 2017 Received in revised form 24 May 2017 Accepted 24 May 2017 Available online 27 May 2017

CYP19 is the single copy gene encoding for the estrogen synthetic enzyme aromatase. Alternate splicing of the promoter is the regulatory mechanism of this gene. In the brain, estrogen is synthesized in neuronal and glial cells and the gene is mainly regulated by the alternate promoter PI.f. The hormone produced in this vicinity has been associated with maintaining normal brain functions. Previously, epigenetic regulation has been shown in the promoters PII and I.3 of CYP19 in adipocytes. In the present study, the methylation of PI.f in CYP19 was examined in glial cells. Treatment of the hypomethylating agent 5-aza-20 deoxycytidine increased CYP19 mRNA species in U87 MG cells while little changes were observed in the other glia cell lines. As PI.f is also chiefly used in T98G cells with high expression of CYP19, the methylation statuses of the promoter in these two cell models were compared. Our results showed that treating U87 MG cells with 10 mM 5-aza-20 deoxycytidine significantly induced a ~10-fold increase in CYP19 transcription and ~80% increase in aromatase activity. In contrast, the same treatment did not change either endpoint in T98G cells. Further investigation illustrated the CpGs in PI.f were differentially methylated in the two cell lines; 63% and 37% of the 14 CpG sites were methylated in U87 MG and T98G cells respectively. In conclusion, this study illustrated that the brain-specific PI.f derived CYP19 expression can be regulated by DNA methylation. © 2017 Elsevier B.V. All rights reserved.

Keywords: Methylation Aromatase Glial cells

1. Introduction Human aromatase is a 55-kDa protein encoded by a single copy gene with 10 exons (Means et al., 1989; Toda et al., 1990). The coding region extends from Exon II to IX and many untranslated promoter regions in the upstream of tissue-specific Exon I control the gene regulation. The transcription starts from the promoter region upstream of each Exon I and splices onto a common site in Exon II. Different tissues employ the distinct tissue-specific promoters upstream of Exon I for regulation, and tissue-specific regulation of aromatase expression is achieved through alternate splicing (Simpson et al., 1997). However, the translated aromatase is identical in all transcripts since the protein encoding region falls in

* Corresponding author. State Key Laboratory of Reproductive Medicine, Nanjing Medical University, Nanjing 211166, China. ** Corresponding author. Food & Nutritional Sciences Programme, The Chinese University of Hong Kong, Rm.507C MMW Bldg, Shatin, N.T., Hong Kong. E-mail addresses: [email protected] (L. Ye), [email protected] (L.K. Leung). http://dx.doi.org/10.1016/j.mce.2017.05.029 0303-7207/© 2017 Elsevier B.V. All rights reserved.

the common sequence at Exon II to IX. The role of estrogen in the brain has been contradictory. GarciaSegura et al. (1999a). have observed that reactive astrocytes have increased aromatase expression during brain injury. Since aromatase is responsible for estrogen biosynthesis, estrogen is suggested to be neuroprotective (Saldanha et al., 2009). In contrast, other researchers have shown that estrogen may intensify brain injury in a rat stroke model (Bingham et al., 2005). The hormone's neuroprotective effect remains controversial. Epigenetic regulation represents a major transcriptional control of gene expression. DNA methylation, histone modifications, and non-coding RNAs are the three major epigenetic processes that could lead to expression alternations in genes. Intrinsic and extrinsic signals can be the epigenetic regulators. As reviewed by Singal and Ginder (1999), DNA methylation occurs at the C-5 position of cytosine in a dinucleotide CG-sequence, which is mostly found in clusters known as the CpG islands. The reaction is catalyzed by the enzyme DNA methyltransferase with the substrate sadenosyl-methionine. DNA methylation suppresses transcriptional activity in three different mechanisms. Methylation of cytosine in

132

W. Tan et al. / Molecular and Cellular Endocrinology 452 (2017) 131e137

the binding elements may disturb the binding of transcriptional factors. Alternatively, the methylated cytosine may attract the binding of methylcytosine binding proteins (MeCP) and repress the gene transcription. The silencing mechanism may also be caused by the formation of a stable chromatin structure from DNA methylation. DNA methylation may participate in tissue-specific gene expression (Brena et al., 2006), and previous studies have shown that DNA methylation deters CYP19 expression in human breast adipose fibroblasts (Knower et al., 2010), hepatoma (Dannenberg and Edenberg, 2006) and endometrial stromal cells (Izawa et al., 2008). In the present study, we investigated the relationship between the methylation status and transcriptional control in the promoter region PI.f of CYP19 in glial cells. Glial cells are aromataseexpressing cells in the brain and increased expression has been observed in brain injury (Garcia-Segura et al., 1999b). 2. Material and methods 2.1. Chemicals 5-Aza-2’-deoxycytidine and S-(5’-Adenosyl)-L-methionine chloride dihydrochloride (SAM) were obtained from Sigma-Aldrich (St. Louis, MO, USA). Other chemicals, if not stated, were purchased from Sigma-Aldrich. 2.2. Cell culture The glioblastoma cell lines T98G, DBTRG-05MG, U87 MG and U251 were purchased from A.T.C.C., Rockville, MD. T98G was maintained in MEM medium (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (Hyclone Laboratories, South Logan, Utah, USA). U87 MG and U251 were maintained in DMEM medium (Hyclone) supplemented with 10% FBS. DBTRG05MG was maintained in RPMI-1640 (Gibco) supplemented with 10%FBS, 2mM L-Glutamine (Gibco) and 1mM sodium pyruvate (Gibco). Cells were incubated at 37  C, 5% carbon dioxide and routinely sub-cultured when reaching 80% of confluency. Three days before experiments, the cells were incubated in phenol-red free RPMI 1640 medium (Invitrogen Life Technology), and supplemented with 10% charcoal dextran-treated serum (Hyclone). 5-Aza-2’-deoxycytidine and SAM were administered in the solvent vehicle dimethyl sulphoxide (DMSO), and the concentration was limited to 0.1% v/v. 5-Aza-2’-deoxycytidine (0.1, 1, 10 mM) was treated when cells were grown to ~50% confluence and changed daily for 5 days (To et al., 2012), and SAM (0.1, 1, 10 mM) was treated for 24 h (Dello Russo et al., 2000) according to former cell culture studies. Cell density was seeded uniformly at 5  102 cells/mm2 in all experiments. 2.3. “In-cell” aromatase assay The assays were performed as previously described (Wang et al., 2005, 2006) with slight modification. In brief, U87 MG and T98G cells were seeded and allowed 1 day for attachment. After treating with 5-Aza-2’-deoxycytidine for 5 days, assays were started by culturing with serum-free medium containing [1b-3H(N)]-androst -4-ene-3,17-dione. The final concentration of androstenedione was controlled at 25 nM, and the reaction was incubated at 37  C for another 24 h. An aliquot of the medium was then mixed with equal volume of chloroform, followed by a 13,000  g centrifugation at 4  C for 10 min. The aqueous phase was removed into a new tube containing 500 ml of 5% activated charcoal suspension. After 30 min incubation, an aliquot of the supernatant fraction was taken out for

scintillation counting. The protein content of the cells, on the other hand, was determined by using a BCA kit (Thermo Scientific Pierce) after dissolving the cells in 0.5 moL/l NaOH. 2.4. Quantitative real-time RT-PCR assay on gene expression The protocol (Wang et al., 2008) and the CYP19 exon-specific probes for I.1(Ia), I.2, I.3, I.4, I.5(2a), I.f and II (Ye et al., 2009; Tan et al., 2013) were previously described. The primers for amplifying exon-specific mRNA species were designed by Customdesigned Taqman® Gene Expression Assays (Assay-by-Demand®, Applied Biosystems, Foster City, CA, USA). The expression of GAPDH (Taqman probe Cat No. Hs99999905_m1, Assay-on- Demand®, Applied Biosystems) was used for normalization. The primer sequences were shown as below. Glial cells were cultured as described above. Total RNA was extracted from the cells using TRIzol reagent (Invitrogen, Carlsbad, CA, USA). The concentration and purity of the isolated RNA were determined by the absorbance reading observed at 260 and 280 nm. 3 mg of total RNA, oligo-dT, and M-MLV Reverse Transcriptase (USB Corporation, Cleveland, Ohio, USA) were used for first strand synthesis. Target fragments were quantified by using StepOne™ Real-Time PCR System (Applied Biosystems®, Life Technologies, Grand Island, NY, USA). Real-time PCR Master Mix Reagent kit was obtained from Takara Bio USA, Inc. and PCR reactions were set up as described in the manual. A typical reaction contained 1ml 20  probes and 4 ml cDNA, and the final reaction volume was 20 ml. The reaction was initiated by preheating at 50  C for 2 min, followed by heating at 95  C for 10 min. Subsequently, 45 amplification cycles were then carried out with 15 s denaturation at 95  C and 1 min annealing and extension at 58  C. Relative gene expression data were analyzed using the 2DDCТmethod (Livak and Schmittgen, 2001). 2.5. Sodium bisulfite sequencing Genomic DNA was isolated from cells using GenElute™ Mammalian Genomic DNA Miniprep Kit (Sigma). DNA was deaminated with sodium bisulfite by using the EZ DNA MethylationGold™ Kit (ZYMO). The online program Primer-BLAST (National Center for Biotechnology Information, Bethesda MD, USA) was used to design PCR primers (as shown below) for amplifying the converted DNA. Ex Taq (Takara Bio USA, Inc.) was used in the PCR and the cycling conditions: 95  C for 5 min; (95  C for 30 s, 55  C for 30 s, 70  C for 1 min) x 35 cycles; 70  C for 10 min. The PCR amplicons were cloned into T-Vector pMD19 (Takara) and were then sent for sequencing. The sequences were analyzed by using the web platform Bisulfite sequencing DNA Methylation Analysis (BISMA) (Rohde et al., 2008).

Primer hCYP19A-I.f-F hCYP19A-I.f-R

Sequence 0

Product size (bp) 0

5 -CTTTTTTTTTTAAGTTAAAT-3 50 -TGTGCAAATATTAAGGTTATT-30

834

2.6. Methylation-sensitive restriction enzyme PCR (msRE-PCR) 250 ng of genomic DNA was isolated and digested with either HapII or MspI (Takara) overnight at 37  C. Both enzymes cleaved (50 -CYCG G-30 ), while only HapII was sensitive to cytosine methylation. PCR was then performed to amplify the target Exon I or II (forward primers) and a common control site (reverse primer)

W. Tan et al. / Molecular and Cellular Endocrinology 452 (2017) 131e137

at Exon II genomic region (Enjuanes et al., 2003). Primer sequences used are as follows:

133

I.f-spliced CYP19 mRNA were the major species expressed in these cells. 3.3. Relative aromatase activity in U87 MG and T98G cells

Target Human Human Human Human Human Human Human

Primer sequence CYP19-I.a-Forward CYP19-I.2-Forward CYP19-I.3-Forward CYP19-I.4-Forward CYP19-I.f-Forward CYP19-II-Forward CYP19-Reverse

50 -GTGGAGGCAAACAGGAAGGT-30 50 -GCAGTTAAGGGCTTCCTGA-30 50 -CCTTGTTTTGACTTGTAACCA-30 50 -GATTGAGGTCACAGAAGGCA-30 50 -CTCCTCTCCCCAAGTCAAAC-30 50 -CCCTTTGATTTCCACAGGACT C-30 50 -CGGGTTCAGCATTTCCAAAACCAT-30

2.7. Western analysis Cells were washed twice with PBS (pH 7.4) and harvested into a 1.5 ml microtube with 0.1 ml lysis buffer (PBS, 1%NP40, 0.5% sodium deoxycholate, 0.1% SDS). The lysis buffer contained protease inhibitors (40 mg/L PMSF, 0.5 mg/L aprotinin, 0.5 mg/L leupeptin, 1.1 mmoL/L EDTA and 0.7 mg/L pepstatin). The harvested cells were then lysed with a cell disruptor (Branson Ultrasonics Corp., Danbury, CT, U.S.A.) on ice for 10 s. The protein concentration of cell lysate was determined by BCA kit (Thermo Scientific Pierce). 50 mg of lysate protein was separated on 10% SDS-PAGE and transferred onto an Immobilon PVDF membrane (Millipore, Bedford, MA, U.S.A.). Anti-aromatase antibody (Abcam, Cambridge, UK), anti-b actin primary (Sigma Chem) and secondary antibodies conjugated to horseradish peroxidase (Santa Cruz Biotechnology) were used for protein detection. An ECL Detection Kit (Amersham, Arlington Heights, IL, U.S.A.) provided the chemiluminescence substrate for HRP, and the targeted protein was visualized by autochemiluminography. 2.8. Statistical analysis The data were analyzed by the software package Prizm5® (GraphPad Software, Inc., San Diego, CA). One Way ANOVA with Dunnett's (Figs. 1, 3 and 5) or Bonferroni's (Fig. 2) post hoc test was used for comparison. The total amounts of CpG methylation in U87 MG and T98G cells (Fig. 6) were compared by c2-test. 3. Results 3.1. Relative CYP19 expressions in different glioblastoma cell lines treated with 5-aza-2'deoxycytidine Relative CYP19 mRNA expressions in four glioblastoma cell lines T98G, DBTRG-05MG (05MG), U87 MG, and U251 were determined and compared after treating with 5-aza-2'deoxycytidine. The basal level of CYP19 expression in T98G cells was 150 times to that expressed in U87 MG cells (data not shown). Compared to the control, expression in U87 MG cells treated with 10 mM 5-aza2'deoxycytidine was increased by 10 fold (p < 0.05) (Fig. 1). No changes were observed in the other three glioblastoma cell lines after treating with 10 mM 5-aza-2'deoxycytidine. 3.2. Specific promoter usage in U87 MG cells By using the Exon-specific probes, the transcripts from Promoters Ia, I.f, I.2, I.3 and II upstream of Exon I were detectable in this cell line (Fig. 2). Promoter I.f is reported as the major regulatory promoter for CYP19 expression in the brain under normal condition (Yague et al., 2006). RT-PCR assay result showed that Exon II, I.a, and

As the promoter usage of CYP19 for T98G cells (Tan et al., 2013) is similar to U87 MG cells as shown above, their enzyme and expression profiles were compared and contrasted. The basal aromatase activity in T98G cells was about 10 times to that of U87 MG cells (data not shown). A two-fold increase was observed in U87 MG cultures treated with 10 mM 5-aza-2'deoxycytidine, whereas no alteration was seen in T98G cells with comparable treatment (P < 0.05) (Fig. 3). 3.4. Restriction analysis of DNA methylation As DNA methylation appeared to be a factor in CYP19 expression and both U87 MG and T98G cells were employing a similar promoter usage profile, the methylation status on these promoters was analyzed by restriction enzymes Hap II and Msp I. The restriction enzymes recognized the same sequence with the difference in cutting the internal methylated cytosine. The former enzyme was sensitive to CpG-methylation, while the latter one cleaved disrespect to the methylation. Compared with the Input genomic DNA in the control panels of U87 MG and T98G cells, the HapII digested genomic DNA in U87 MG cells appeared to be in a higher quantity than that in T98G cells (Fig. 4A). However, the 5-aza-2'deoxycytidine -treated DNA looked about the same as the one without the treatment. All DNA samples were highly digested by MspI. The exon-specific fragments were amplified from the digested genomic DNA samples from U87 MG and T98G cells as shown in Fig. 4B. Since HapII could not digest the abovementioned sequence with methylated cytosine, the data suggested that Exon I.2 and I.f in U87 MG cells were methylated. 3.5. Exon-specific CYP19 mRNA expression in U87 MG cells treated with 5-aza-2'deoxycytidine The restriction analysis indicated that Exon I.f and I.2 were likely subjected to DNA methylation. The exon-specific CYP19 mRNA species was quantitated by RT-PCR. 10 mM 5-aza-2'deoxycytidine treatment increased the I.f and I.2 transcript expression by about 3fold in U87 MG cells (P < 0.05) (Fig. 5). Because Exon I.f-spliced transcripts were the greater contributor to the overall pool of CYP19 transcripts, the exon was examined in subsequent methylation investigation. 3.6. Methylation of the PI.f CpGs in U87 MG and T98G cells Fourteen locations of CpGs in PI.f were numbered from 1 to 14 as shown in Fig. 6A. The distribution of methylated CpGs over the 14 locations in 6 cell colonies was exhibited in Fig. 6B. The percentages of methylated CpGs over the 14 locations were displayed in Fig. 6C. After pooling the total methylated CpGs in the two cell types, the methylation in U87 MG cells was significantly (P < 0.05; c2-test) greater than that in T98G cells. 3.7. Differential aromatase protein expression from methylation/ demethylation treatment Increased aromatase protein was observed in U87 GM cells treated with 10 mM 5-aza-2'deoxycytidine (Aza) (Fig. 7A), while the

134

W. Tan et al. / Molecular and Cellular Endocrinology 452 (2017) 131e137

Fig. 1. Relative CYP19 mRNA expression in various glial cell lines treated with 5-aza-2'deoxycytidine. Glial cells T98G, DBTRG-05MG (05MG), U-87 MG and U251 were seeded in 6-well plates and treated with 0.1, 1, 10 mM 5-aza-2'deoxy-cytidine. The amount of CYP19 mRNA was determined by real time PCR and was normalized by GAPDH mRNA. Values are means ± SEM, n ¼ 3 or 4. Means labeled with various letters were significantly different (p < 0.05), and the order is (b > a).

4. Discussion

Fig. 2. Exon-specific CYP19 expression in U87 MG cells. U87 MG cells were seeded in 6-well plates and maintained in culture medium as described in Methods. The amounts of exon-specific spliced mRNA species were determined by real time PCR and were normalized by GAPDH mRNA. Values are means ± SEM, n ¼ 3. Means labeled with different letters were significantly different (p < 0.05) among the alternate spliced species, and the order is (b > a > c).

same protein was reduced by 10 mM s-adenosyl- methionine (SAM) treatment (Fig. 7B). SAM which provided the substrate of cellular methylation reaction was added to promote DNA methylation. A similar result was observed when T98G cells were treated with SAM in Fig. 7C. These results verified that DNA methylation could affect the aromatase protein expression.

CpG islands (CGI) can be found at transcription start sites (TSS), intragenic and intergenic locations. These CGI-containing locations are possible transcription initiation sites, while TSS encompasses ~50% of the total CGIs in the human genome (Deaton and Bird, 2011). Nevertheless, transcription regulated by methylation in non-CGI promoter has also been reported (Lee et al., 2012; Fujii et al., 2006; Jones and Chen, 2006). In the present study, the methylation status of CpGs at the TSS of CYP19 exon I.f was analyzed. Because of the low CG content (<50%) and CpG frequency (<0.6), this gene stretch can be classified as a non-CGI promoter as defined previously (Brena et al., 2006). We demonstrated that the demethylation of the CpG sites in TSS was associated with increased PI.f-specific CYP19 mRNA species and aromatase activity. Nevertheless, the CpG statuses in the intragenic and intergenic regions were not determined. The nucleoside antimetabolite azacytidine is capable of reactivating gene expression without removing methylation in the promoter as shown in a previous study (Flotho et al., 2009). This property might bring in interference in the present study, since 5aza-20 deoxycytidine is an analogue of azacytidine. The issue was addressed by administering SAM in a separate experiment, and the result was consistent with the state of DNA methylation. The tissue-specific expression of CYP19 attributed by alternate splicing in the promoter region has been well-established; yet, the mechanisms driving the differential expression remain unknown. Based on the studies performed in sheep or cattle placentae, Vanselow's group suggests that CpG methylation can be a factor partially responsible for the tissue-specific CYP19 expression (Vanselow et al., 2008; Furbass et al., 2008). In the present study,

W. Tan et al. / Molecular and Cellular Endocrinology 452 (2017) 131e137

135

Fig. 3. Aromatase activity was induced by 5-aza-2'deoxycytidine treatment in U87 MG cells. U87 MG and T98G cells were seeded in six-well plates and maintained in phenol red-free RPMI medium supplemented with 10% charcoal dextranetreated serum. Aromatase activities were determined in U87 MG (Fig. 3A) and T98G (Fig. 3B) cultures treated with 0.1, 1, 10 mM 5-aza-2'deoxycytidine. The data represents the means ± SE of 6 samples isolated from independent cultures. Means labeled with different letters were significantly different (p < 0.05), and the order is (b > a).

Fig. 5. 5-Aza-2'deoxycytidine increased Exon I-specific CYP19 mRNA species in U87 MG cells. U87 MG cells were seeded in 6-well plates and maintained in phenol-red free RPMI 1640 medium supplemented with 10% charcoal dextran-treated FBS. Cells were treated with 5-aza-2'deoxycytidine for 24 h. The amounts of Exon I.f and I.2spliced CYP19 transcripts were determined by real time PCR and were normalized by GAPDH mRNA. Values are means ± SEM, n ¼ 3. Means labeled with various letters were significantly different (p < 0.05), and the order is (b > a).

Fig. 4. Restriction analysis on the DNA methylation status of Exon I region of CYP19. Brain cells were seeded in 6-well plates and maintained in phenol-red free RPMI 1640 medium supplemented with 10% charcoal dextran-treated FBS. Genomic DNA was isolated and incubated with Hap II or Msp I. Subsequently, the digested DNA was amplified by PCR with primers designed for Exon I subtype-specific regions. The digested DNA is shown in Fig. 4A and the amplicons of various Exon I sequences are shown in Fig. 4B.

the methylation of CYP19 was analyzed in glial cells and the process appeared to be a regulatory mechanism in PI.f-driven CYP19 expression. Previous studies have shown that some signaling molecules, such as TNFa, (Sullivan et al., 2007), c-fos (Chandrasekhar and Raman, 1997), c-jun and c-myc (Tao et al., 2000), can modulate the epigenetic regulation of the alternate promoter PI.4. It is possible that the methylation of PI.f is also under a comparative system, and following-up studies should be performed in this regard. A review article has described that PI.f and PII are the promoter

sequences driving the expression of CYP19 in the brain (Bulun et al., 2005). In the present study, we demonstrated that PI.f rather than PII was subjected to the control of DNA methylation. Contrasting to the high aromatase-expressing cells such as T98G, the methylation of PI.f appeared to play a significant role in CYP19 expression regulation in the low aromatase-expressing cells U87 MG. This result supports the claim that DNA methylation is only one of many factors in governing tissue-specific CYP19 expression. Among all the other regulatory mechanisms of CYP19 expression, the mitogen-activated protein kinase ERK and its associated pathways have received a considerable amount of attention. ERK induces PI.f transcription in glial cells (Tan et al., 2013) and PI.3 and PII in MCF-7 cells (Li et al., 2011). It may also increase the stability of total CYP19 transcripts in cells (Chan et al., 2010). The kinase activates the transcription factor CCAAT/enhancer binding protein (C/ EBP) (Li et al., 2007) and initiates the transactivation through binding to the response elements at PI.f. It was possible that the CpG methylation in PI.f as shown in the present study interfered with this regulatory pathway. A similar study performed in human breast adipose fibroblasts has shown that DNA methylation can suppress overall CYP19 transcription; however, the abundances of PI.4 and I.3/II-derived

136

W. Tan et al. / Molecular and Cellular Endocrinology 452 (2017) 131e137

Fig. 6. Prevalency of methylation in the CpG sites of PI.f. The locations of CpG sites on PI.f of CYP19 are shown in Fig. 6A. The distribution of methylated CpG sites in the 14 locations in PI.f of U87 MG and T98G cells is displayed in Fig. 6B. Result exhibit in Fig. 6C is the percentage of methylation calculated from individual CpG sites. Data represent the percentage distribution from 6 colonies.

demethylation of the CpGs within PI.f was correlated to the specific mRNA species. Dysregulation of gene methylation has been associated with some brain diseases, such as Alzheimer's, Parkinson's, and other dementia-causing diseases (Delgado-Morales and Esteller, 2017). In addition, a reduction in 5-methylcytosine can be an indicator of global epigenetic dysregulation in diffuse intrinsic pontine glioma (Ahsan et al., 2014). The significance of PI.f methylation in the context of human diseases remains unclear. In summary, this study illustrated that DNA methylation was a regulatory mechanism of PI.f-driven CYP19 expression in glioblastoma cells. Although the physiological relevance has not been established, the methylation potentially disturbs the estrogen balance in the brain. Conflict of interest statement The authors do not have any conflict of interest in this study. Acknowledgements

Fig. 7. 5-Aza-2'deoxycytidine increased aromatase protein in U87 MG cells. U87 MG cells were seeded in 6-well culture dishes and treated with various concentrations of 5-aza-2'deoxycytidine (Aza) (Fig. 7A) or s-adenosylmethionine (SAM) (Fig. 7B) for 24 h. Fig. 7C is the result of T98G cells treated with s-adenosylmethionine. Amounts of CYP19 in protein extracts were determined by western blot analysis. The images shown are a representation of two independent experiments.

mRNA transcripts are not dependent on the methylation of CpGs within the promoters (Knower et al., 2010). Since PI.4 and I.3/II are promoters used in adipocytes, the result implies that an alternate pathway exists. In contrast, the present study illustrated that the

This study was supported by The Chinese University of Hong Kong Direct Grant #4053184, Innovative and Entrepreneurial Program of Jiangsu Province, and Natural Science Foundation of Jiangsu Province (15KJA180006). References Ahsan, S., Raabe, E.H., Haffner, M.C., Vaghasia, A., Warren, K.E., Quezado, M., Ballester, L.Y., Nazarian, J., Eberhart, C.G., Rodriguez, F.J., 2014. Increased 5hydroxymethylcytosine and decreased 5-methylcytosine are indicators of global epigenetic dysregulation in diffuse intrinsic pontine glioma. Acta Neuropathol. Commun. 2, 59.

W. Tan et al. / Molecular and Cellular Endocrinology 452 (2017) 131e137 Bingham, D., Macrae, I., Carswell, H., 2005. Detrimental effects of 17 beta-oestradiol after permanent middle cerebral artery occlusion. J. Cereb. Blood Flow Metabolism 25, 414e420. Brena, R.M., Huang, T.H., Plass, C., 2006. Toward a human epigenome. Nat. Genet. 38, 1359e1360. Bulun, S.E., Lin, Z., Imir, G., Amin, S., Demura, M., Yilmaz, B., Martin, R., Utsunomiya, H., Thung, S., Gurates, B., Tamura, M., Langoi, D., Deb, S., 2005. Regulation of aromatase expression in estrogen-responsive breast and uterine disease: from bench to treatment. Pharmacol. Rev. 57, 359e383. Chan, M.Y., Huang, H., Leung, L.K., 2010. 2,3,7,8-Tetrachlorodibenzo-para-dioxin increases aromatase (CYP19) mRNA stability in MCF-7 cells. Mol. Cell Endocrinol. 317, 8e13. Chandrasekhar, K., Raman, R., 1997. De novo methylation of the proto-oncogene, cfos, during development occurs step-wise and directionally in the laboratory mouse. Mol. Reprod. Dev. 48, 421e432. Dannenberg, L.O., Edenberg, H.J., 2006. Epigenetics of gene expression in human hepatoma cells: expression profiling the response to inhibition of DNA methylation and histone deacetylation. BMC Genomics 7, 181. Deaton, A.M., Bird, A., 2011. CpG islands and the regulation of transcription. Genes Dev. 25, 1010e1022. Delgado-Morales, R., Esteller, M., 2017. Opening up the DNA methylome of dementia. Mol. Psychiatry 22 (4), 485e496. Dello Russo, C., Tringali, G., Ragazzoni, E., Maggiano, N., Menini, E., Vairano, M., Preziosi, P., Navarra, P., 2000. Evidence that hydrogen sulphide can modulate hypothalamo-pituitary-adrenal axis function: in vitro and in vivo studies in the rat. J. Neuroendocrinol. 12, 225e233. Enjuanes, A., Garcia-Giralt, N., Supervia, A., Nogues, X., Mellibovsky, L., Carbonell, J., Grinberg, D., Balcells, S., Diez-Perez, A., 2003. Regulation of CYP19 gene expression in primary human osteoblasts: effects of vitamin D and other treatments. Eur. J. Endocrinol. 148, 519e526. Flotho, C., Claus, R., Batz, C., Schneider, M., Sandrock, I., Ihde, S., Plass, C., Niemeyer, C.M., Lubbert, M., 2009. The DNA methyltransferase inhibitors azacitidine, decitabine and zebularine exert differential effects on cancer gene expression in acute myeloid leukemia cells. Leukemia 23, 1019e1028. Fujii, G., Nakamura, Y., Tsukamoto, D., Ito, M., Shiba, T., Takamatsu, N., 2006. CpG methylation at the USF-binding site is important for the liver-specific transcription of the chipmunk HP-27 gene. Biochem. J. 395, 203e209. Furbass, R., Selimyan, R., Vanselow, J., 2008. DNA methylation and chromatin accessibility of the proximal Cyp 19 promoter region 1.5/2 correlate with expression levels in sheep placentomes. Mol. Reprod. Dev. 75, 1e7. Garcia-Segura, L., Naftolin, F., Hutchison, J., Azcoitia, I., Chowen, J., 1999. Role of astroglia in estrogen regulation of synaptic plasticity and brain repair. J. Neurobiol. 40, 574e584. Garcia-Segura, L.M., Wozniak, A., Azcoitia, I., Rodriguez, J.R., Hutchison, R.E., Hutchison, J.B., 1999. Aromatase expression by astrocytes after brain injury: implications for local estrogen formation in brain repair. Neuroscience 89, 567e578. Izawa, M., Harada, T., Taniguchi, F., Ohama, Y., Takenaka, Y., Terakawa, N., 2008. An epigenetic disorder may cause aberrant expression of aromatase gene in endometriotic stromal cells. Fertil. Steril. 89, 1390e1396. Jones, B., Chen, J., 2006. Inhibition of IFN-gamma transcription by site-specific methylation during T helper cell development. EMBO J. 25, 2443e2452. Knower, K.C., To, S.Q., Simpson, E.R., Clyne, C.D., 2010. Epigenetic mechanisms regulating CYP19 transcription in human breast adipose fibroblasts. Mol. Cell Endocrinol. 321, 123e130. Lee, S.T., Xiao, Y., Muench, M.O., Xiao, J., Fomin, M.E., Wiencke, J.K., Zheng, S., Dou, X., de Smith, A., Chokkalingam, A., Buffler, P., Ma, X., Wiemels, J.L., 2012. A global DNA methylation and gene expression analysis of early human B-cell

137

development reveals a demethylation signature and transcription factor network. Nucleic Acids Res. 40, 11339e11351. Li, H., Gade, P., Xiao, W., Kalvakolanu, D.V., 2007. The interferon signaling network and transcription factor C/EBP-beta. Cell Mol. Immunol. 4, 407e418. Li, F., Ye, L., Lin, S.M., Leung, L.K., 2011. Dietary flavones and flavonones display differential effects on aromatase (CYP19) transcription in the breast cancer cells MCF-7. Mol. Cell Endocrinol. 344, 51e58. Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods 25, 402e408. Means, G.D., Mahendroo, M.S., Corbin, C.J., Mathis, J.M., Powell, F.E., Mendelson, C.R., Simpson, E.R., 1989. Structural analysis of the gene encoding human aromatase cytochrome P-450, the enzyme responsible for estrogen biosynthesis. J. Biol. Chem. 264 (32), 19385e19391. Rohde, C., Zhang, Y., Jurkowski, T.P., Stamerjohanns, H., Reinhardt, R., Jeltsch, A., 2008. Bisulfite sequencing Data Presentation and Compilation (BDPC) web serverea useful tool for DNA methylation analysis. Nucleic Acids Res. 36, e34. Saldanha, C., Duncan, K., Walters, B., 2009. Neuroprotective actions of brain aromatase. Front. Neuroendocrinol. 30, 106e118. Simpson, E.R., Michael, M.D., Agarwal, V.R., Hinshelwood, M.M., Bulun, S.E., Zhao, Y., 1997. Cytochromes P450 11: expression of the CYP19 (aromatase) gene: an unusual case of alternative promoter usage. FASEB J. 11 (1), 29e36. Singal, R., Ginder, G.D., 1999. DNA methylation. Blood 93, 4059e4070. Sullivan, K.E., Reddy, A.B., Dietzmann, K., Suriano, A.R., Kocieda, V.P., Stewart, M., Bhatia, M., 2007. Epigenetic regulation of tumor necrosis factor alpha. Mol. Cell Biol. 27, 5147e5160. Tan, W., Wong, T.Y., Wang, Y., Huang, J., Leung, L.K., 2013. CYP19 expression is induced by 2,3,7,8-tetrachloro-dibenzo-para-dioxin in human glioma cells. Mol. Cell Endocrinol. 375, 106e112. Tao, L., Yang, S., Xie, M., Kramer, P.M., Pereira, M.A., 2000. Hypomethylation and overexpression of c-jun and c-myc protooncogenes and increased DNA methyltransferase activity in dichloroacetic and trichloroacetic acid-promoted mouse liver tumors. Cancer Lett. 158, 185e193. To, S.Q., Takagi, K., Miki, Y., Suzuki, K., Abe, E., Yang, Y., Sasano, H., Simpson, E.R., Knower, K.C., Clyne, C.D., 2012. Epigenetic mechanisms regulate the prostaglandin E receptor 2 in breast cancer. J. Steroid Biochem. Mol. Biol. 132, 331e338. Toda, K., Terashima, M., Kawamoto, T., Sumimoto, H., Yokoyama, Y., Kuribayashi, I., Mitsuuchi, Y., Maeda, T., Yamamoto, Y., Sagara, Y., et al., 1990. Structural and functional characterization of human aromatase P-450 gene. Eur. J. Biochem. 193 (2), 559e565. Vanselow, J., Selimyan, R., Furbass, R., 2008. DNA methylation of placenta-specific Cyp19 promoters of cattle and sheep. Exp. Clin. Endocrinol. Diabetes 116, 437e442. Wang, Y., Chan, F.L., Chen, S., Leung, L.K., 2005. The plant polyphenol butein inhibits testosterone-induced proliferation in breast cancer cells expressing aromatase. Life Sci. 77, 39e51. Wang, Y., Lee, K.W., Chan, F.L., Chen, S., Leung, L.K., 2006. The red wine polyphenol resveratrol displays bilevel inhibition on aromatase in breast cancer cells. Toxicol. Sci. 92, 71e77. Wang, Y., Man Gho, W., Chan, F.L., Chen, S., Leung, L.K., 2008. The red clover (Trifolium pratense) isoflavone biochanin A inhibits aromatase activity and expression. Br. J. Nutr. 99, 303e310. Yague, J.G., Munoz, A., de Monasterio-Schrader, P., Defelipe, J., Garcia-Segura, L.M., Azcoitia, I., 2006. Aromatase expression in the human temporal cortex. Neuroscience 138, 389e401. Ye, L., Chan, M.Y., Leung, L.K., 2009. The soy isoflavone genistein induces estrogen synthesis in an extragonadal pathway. Mol. Cell. Endocrinol. 302, 73e80.