Biochimica et Biophysica Acta 1832 (2013) 2145–2152

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

j ourna l homepage: www.e lsev ie r .com/ locate /bbad is

Down-regulation of EBAF in the heart with ventricular septal defects andits regulation by histone acetyltransferase p300 and transcription factorssmad2 and cited2

Dongmei Su, Qian Li, Lina Guan, Xiaobo Gao, Huiling Zhang, Dandan E, Lei Zhang, Xu Ma ⁎Department of Genetics, National Research Institute for Family Planning, China

⁎ Corresponding author at: Center for Genetics, NationPlanning, 12, Dahuisi Road, Haidian, Beijing 100081, Cfax: +86 10 62179151.

E-mail addresses: jswkysgc@126.com, genetic@263.ne

0925-4439/$ – see front matter © 2013 Elsevier B.V. All rhttp://dx.doi.org/10.1016/j.bbadis.2013.07.013

a b s t r a c t

a r t i c l e i n f o

Article history:Received 7 November 2012Received in revised form 25 June 2013Accepted 1 July 2013Available online 27 July 2013

Keywords:EBAFp300smad2cited2

As a NODAL pathway inhibitor, EBAF plays a critical role duringmammalian cardiac development. As recent teststhat have been conducted on gene-targeted mice indicate, its expression is frequently altered where cardiac de-fects are present. We aimed to explore the EBAF expression pattern and molecular mechanism of EBAF gene forVSD genesis. In this report, we show that the average expression of EBAF in the disease tissues of VSD patientswas lower than the expression in normal fetuses without VSD. Further study showed that the expression patternof EBAF was potentially involved in cardiomyocyte apoptosis by Annexin-V and RT-PCR assays. We also foundthat abnormal activation of NODAL-PITX2C pathwaywas associatedwith down-regulation of EBAF. By luciferasereporter assays, we find that EBAF expression is mediated by transcriptional factors smad2 and cited2. Inaddition, ChIP assays showed that histone acetyltransferase p300 is involved in the activation of EBAFthrough inducing hyperacetylation of histone H4 at the EBAF promoter. Co-immunoprecipitation also indi-cates that the expression of EBAF is regulated by a transcriptional complex including p300, smad2, andcited2. This study revealed a novel regulator mechanism of EBAF, which may be a potential molecular targetfor halting the onset of VSDs. They also indicate that smad2, cited2, and p300 may play important roles inmodulating the confirmation of ventricular septal defects.

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

Congenital heart disease (CHD) is the clinical manifestation ofanomalies in the heart and great vessels; it is the leading cause ofnon-infectious infant mortality, with an incidence about 6/1000 [1,2].As the number of patients with CHD increases, the disease has becomea significant part of the overall disease burden for which governmenthealth agencies are responsible [3]. Ventricular septal defect (VSD) isthe most commonly recognized CHD [4]. VSDs take different forms;around 70–80% take the form of a membranous ventricular septaldefect, about 20% take the formof a supracristal ventricular septal defect,around 5% take the form of an atrioventricular canal defect, and fewerthan 5% of VSDs are muscular defects. VSD genesis is a multifactorialand complex condition in which both genetic and environmental factorsplay important roles. Although the embryology and physiology of VSDsarewidely known, their etiology and pathogenesis are still not very clear.

Recent genetic studies have uncovered a large number of genes inconnection with familial and sporadic forms of CHD [5,6]. Many

al Research Institute for Familyhina. Tel.: +86 1062117712;

t.cn (X. Ma).

ights reserved.

transcription factors have been described as regulators of cardiac-specific gene expression [7,8]. For instance, CITED2 (a cAMP-responsiveelement-binding protein [CBP]/p300-interacting transactivator withED-rich C-terminal domain) has been found to be essential for normaldevelopment of the heart [9], as evidenced by the fact that mice lackingCited2 die in utero showing various cardiac malformations includingatrial and ventricular septal defects, and overriding aortas [10]. Smad2,another transcriptional factor, has been found to have a detrimental ef-fect on embryo development [11,12], but its role in cardiac developmenthas not been clearly clarified.

EBAF, as a Nodal pathway inhibitor [13] has been found to lackany expression [9,14] in Cited2-null embryos with cardiacmalformations. But the correlation between expression patterns ofEBAF and cardiac malformations genesis is still unclear. In thispaper, for the first time, we show that the expression of EBAF is fre-quently down-regulated in patients with supracristal ventricularseptal defects, membranous ventricular septal defects, and atrioven-tricular canal type ventricular septal defects. We also show that thedown-regulation of the EBAF gene may be partially attributed tothe transcriptional modulators smad2 and cited2. Moreover, wefind that histone acetyltransferase p300 is involved in the activationof EBAF by inducing the hyperacetylation of histone H4 in the EBAFpromoter, indicating that histone acetylation may play an importantrole in heart development.

2146 D. Su et al. / Biochimica et Biophysica Acta 1832 (2013) 2145–2152

2. Materials and methods

2.1. Tissue samples

Fetal myocardial tissue samples were obtained from Shengjing Hospi-tal of China Medical University. Myocardial tissue samples from 16 VSDand 16 normal fetuses at 22–28 weeks of gestationwere obtained duringsurgery for pregnancy termination owing to trauma of the pregnantwomen. The specimens were immediately frozen in liquid nitrogen andthen stored at −80 °C until analysis. The study project about roles ofgene transcriptional regulation in congenital heart genesis was reviewedand the protocol was approved by the Ethics Committee of the NationalResearch Institute for Family Planning (Permit Number: 2012-7). In-formed consentwas obtained fromeach pregnantwoman andher family.The study protocol followed the principles of the Declaration of Helsinki.

2.2. Cell culture, transfection, and luciferase reporter assay

Human cardiac myocytes (HCM) were maintained in Dulbecco'sModified Eagle's Medium supplemented with 10% fetal bovine serum,100 mg/ml penicillin, and 100 mg/ml streptomycin in a humidified at-mosphere containing 5% CO2 at 37 °C. Transient transfection HCM cellswere performed using the Lipofectamine 2000 (Invitrogen) procedures.Transfected cells were assayed for luciferase reporter activity using aPromega dual-luciferase reporter assay system, and the Renilla lucifer-ase control plasmid pREP7-RLuc was co-transfected for normalization.

2.3. Apoptosis assay

Myocardial cells were exposed to deferoxamine mesylate (DFO,8 mM; Sigma) for 72 h, which served as an established hypoxiamodel system and induced cell apoptosis. Cells were harvested usingtrypsin and incubated at room temperature for 5 min in the dark withAnnexin V-FITC and propidium iodide according to the instructions ofthe kit (Roche, Germany). The fluorescence of FITC and PI was analyzedby flow cytometry.

2.4. RNA isolation, reverse transcription and real-time PCR

RNA was extracted from the heart tissues using the TRIzol Reagent(Invitrogen) according to the manufacturer's protocol. cDNA was syn-thesized from 2 μg of RNA using a RNA PCR kit (TaKaRa, Dalian, China).Real-time PCR was performed on an ABI Prism7000 Sequence DetectionSystem (Applied Biosystems), following the manufacturer's protocol,with SYBR Green (TaKaRa, Osaka, Japan) used as a double-stranded

Fig. 1. Three-dimensional ultrasound images of fetal hearts. (A) A representative picture of 16 cthree times). The representative picture showed the size of the defect in the heart shown herewhave no VSDs were used as controls.

DNA-specific fluorescent dye. Primer sequences are reported in theSupplementary material online.

2.5. Western blotting

Frozen heart tissues from patients and controls were lysed in buffer.Total proteinwas applied to a 12% SDS-polyacrylamide gel. After electro-phoresis, polyvinylidene fluoride (PVDF)membranewas incubatedwithanti-EBAF (Santa Cruz), anti-cited2 (Abcam), anti-β-actin (Sigma). Thesignals were visualized by using the Chemiluminescent Substrate meth-od with the SuperSignal West Pico kit provided by Pierce Co.

2.6. CoIP (co-immunoprecipitation) assay

Co-IP was performed in HCM cells as described previously [15]. Totalcell extracts from smad2-GFP transfected cells were pre-cleared with40 μl protein A-agarose at 4 °C for 1 h. The supernatant was incubatedwith the smad2 antibody (Cell Signaling Technology)with gentle shakingfor 4 °C followed by addition of 40 μl of protein A-agarose for another 3 h.The beadswere resuspended in 100 μl of 2× loading buffer and boiled for10 min. The proteins were separated on a 12% SDS-PAGE gel and thentransferred to polyvinylidene fluoride membrane for Western blottingdetection with anti-cited2 (Abcam) or anti-p300 antibody (Santa Cruz).

2.7. Chromatin immunoprecipitation (ChIP)

ChIP assays were carried out using a kit supplied by Upstate follow-ing themanufacturer's protocol. Cellswere cross-linkedwith 2% formal-dehyde for 10 min at 37 °C, and then lysed in SDS lysis buffer (1% SDS,10 mM EDTA, 50 mM Tris, pH 8.1) with protease inhibitors. Afterultrasonication, DNA was immunoprecipitated with antibodies againstAc-H3 (Upstate), Ac-H4 (Upstate), p300 (Santa Cruz), cited2 (Abcam)and smad2 (Cell Signaling Technology). The precipitated chromatinwas analyzed by PCR and real time quantitative RCR following themanufacturer's protocol. To calculate the relative abundance of a givengene promoter present in an immunoprecipitate, the following formulawas used: Qgene = IPgene/input gene, to calibrate the variation be-tween different samples. Primer sequences are reported in the Supple-mentary material online. All the experiments were repeated threetimes, and one of the representative results is shown in the text.

2.8. Plasmid constructs

A 600-bp fragment of EBAF enhancer including ASE (asymmetric en-hancer) and a 1300-bp fragment of the PITX2C promoter was amplified

ases of VSDs, whichwere diagnosed using three-dimensional echocardiography (repeatedas 4.9 mm. (B) Fetuses that were revealed by the three-dimensional echocardiography to

Table 1Classification of ventricular septal defects for our study.

Phenotype oftissue donors

Subtypes of the sample Sample no.

Ventricular septaldefect

Supracristal ventricularseptal defect

S144, S162

Membranous ventricularseptal defect

M73,M65,M96,M145,M142,M150,M98, M194, M160, M195, M173,M179, M183

Atrioventricularventricular septal defect

A181

Normal Null N66, N110, N147, N116, N146, N120,N132,N01,N88,N89,N90,N91,N141,N107, N108, N140

S: supracristal ventricular septal defect.M: membranous ventricular septal defect.A: atrioventricular ventricular septal defect.N: normal.

2147D. Su et al. / Biochimica et Biophysica Acta 1832 (2013) 2145–2152

by polymerase chain reaction (PCR) from the human genomic DNAwascloned into the KpnI and XhoI digested luciferase reporter constructPLG3-basic vector, respectively. The open reading frames of smad2,EBAF, nodal or cited2 were amplified by PCR from human cDNA andinserted into the HindIII and XhoI digested GFP vector or pcDNA3-Flagvector to create smad2-GFP, EBAF-GFP, nodal-GFP or Flag-Cited2 fusionprotein, respectively. The pcDNA3.1-p300 plasmid was provided byHuang Baiqu (Institute of Genetics and Cytology, Northeast Normal Uni-versity, China).

2.9. Statistical analysis

Student's t test was used to calculate the statistical significance of theexperimental data. The significance level was set as *p b 0.05; #p b 0.05;**p b 0.01; ##p b 0.01. Error bars denote the standard deviations.

Fig. 2.Down-regulation of EBAFwas associatedwith VSDs genesis. (A) Comparison of EBAF mRby quantitative PCR. Normal represents the average expression of EBAF mRNA inmyocardial tissamples from16 patients with VSDs (combined) is also shown. β-actin was used as the internaltween the normal controls and the different subtypes of VSDs respectively.M-VSD, S-VSD, A-VSand atrioventricular canal type defects, respectively. **p b 0.01 versus normal group. (C) EBAFwith different subtypes of VSDs compared with EBAF proteins in the 10 normal samples whichreference.

3. Results

3.1. Clinical evaluation of ventricular septal defects

Patients underwent a three-dimensional echocardiography to diag-nose whether their fetuses showed any indication of VSDs. As shown inFig. 1A, fetal ventricular septal defect size was 4.9 mm. The three-dimensional echocardiography identified 16 VSD fetuses at 22–28 weeks of gestation. The sizes of the defects of 16 VSDs ranged from4.6–8 mmat 22–28 weeks of gestation. NoVSDswere found in the con-trol group,whohad been identified during the three-dimensional echo-cardiography diagnosis phase (Fig. 1B). Myocardial tissue samples wereobtained during surgery for pregnancy termination owing to trauma ofthe pregnant women. The 16 cases of fetal VSD were confirmed by au-topsy after induced labor. As shown in Table 1, 13 cases were found tobe defects of the membranous septum, 2 were found to be supracristalventricular septal defects, and 1 was found to be an atrioventricularcanal type ventricular septal defect.

3.2. Down-regulation of EBAF in the heart tissues of patients with VSDs

We first assessed the expression of EBAF in myocardial tissue sam-ples from 32 patients (16 patients with different subtypes of VSDs and16 normal fetuses without VSD) using quantitative RT-PCR. The averageexpression of EBAF in the patients with VSDs was lower than the aver-age value of EBAF expression in 16 normal fetuses without VSD(Fig. 2A). Specifically,as shown in Fig. 2B, EBAF was frequently down-regulated in membranous ventricular septal defects (one subtype ofVSD). Though the expression of EBAF was also down-regulated insupracristal ventricular septal defects and atrioventricular ventricularseptal defect (another two subtypes of VSD) in our study (Fig. 2B), theresults should be further confirmed by increasing the numbers of thesamples. As confirmed by Western blotting, the level of EBAF protein

NA expression in 32 samples (16 samples with VSDs and 16 samples without), as revealedsue samples from 16 normal fetuses. The average expression of EBAF in myocardial tissuereference. **p b 0.01 versus normal group. (B) Comparison of EBAF mRNA expression be-D representmembranous ventricular septal defects, supracristal ventricular septal defects,proteins, as detected by Western blotting, were found to be lower in the 10 VSD patientswere selected from the 32 sample pairs mentioned above, β-actin was used as the internal

Fig. 3.Down-regulation of EBAFwas associatedwith cardiomyocyte apoptosis. (A) DFO-induced apoptosis in cardiomyocytes. UntreatedHCM cells (left panel) andHCM cells treatedwithDFO at 8 mM for 72 h (right panel)were examined for apoptotic response by Annexin V-FITC assay. The lower left population of cells in each plot represents viable cells which excluded PIand didn't bind Annexin V, the upper left population cells comprise necrotic cells, which did not exclude PI and were not stained with FITC-labeled Annexin V. The lower and upper rightpopulations correspond to early apoptotic and late apoptotic cells, respectively. The percentages of early and late apoptotic cells, non-apoptotic and necrotic cells are shown in the rightlower quadrant, right upper quadrant, left lower quadrant and left upper quadrant, respectively. (B) RT-PCR analysis of the mRNA expression of EBAF during DFO-inducedapoptosis. β-actin was used as the internal reference.

2148 D. Su et al. / Biochimica et Biophysica Acta 1832 (2013) 2145–2152

was significantly down-regulated in the 10 VSD patients with differentsubtypes of VSDs comparedwith the 10normal sampleswhichwere se-lected from the 32 samples mentioned above (Fig. 2C). The results

Fig. 4. Cited2 participated in regulating the expression of EBAF gene. (A) Western blotting contissues. (B) Low expression of CITED2 is associatedwith down-regulation of EBAF inDFO-inducactivity of EBAF. HCM cells were transfected with EBAF reporter plasmid, together with cited2was observed 30 h after transfection andwas normalized to the Renilla activity. The vector was

indicated that the expression of EBAF was, on average, down-regulated in the disease tissues of VSD patients, and this may play a sig-nificant role in the development of human VSD.

firmed that low expression of CITED2 was associated with low expression of EBAF in VSDed apoptosis in cardiomyocytes. (C) Transcriptional regulator cited2 elevated the promoterexpression plasmids; they were also lysed for luciferase reporter assay. Luciferase activityused as a negative control. ∗p b 0.05; ∗∗p b 0.01 comparedwith the control group (n = 3).

Fig. 5. cited2 and smad2 function cooperatively in transcriptional activation of the EBAF gene. (A) Characterization of EBAF gene responsive regulatory sequences. ASE regulatory regionand sequence location of smad2 are shown in the sub-figure. In addition, the location of the EBAF promoter region for ChIP assay has been underlined for specification in the sub-figure. (B) Transcriptional regulators smad2 elevated the promoter activity of EBAF. HCM cells were transfected with EBAF reporter plasmid, together with smad2 expressionplasmids; they were also lysed for luciferase reporter assay. (C) cited2 and smad2 cooperatively up-regulated EBAF promoter activity. HCM cells were transfected with EBAFreporter plasmid, together with smad2 and cited2 expression plasmids, and lysed for luciferase reporter assay. (D) smad2 or cited2 over-expression potentiated their bindingactivities at the EBAF promoter. Samples were immunoprecipitated (IP) with smad2 or cited2 antibodies and amplified by PCR.

2149D. Su et al. / Biochimica et Biophysica Acta 1832 (2013) 2145–2152

3.3. Down-regulation of EBAF expression in apoptotic cardiomyocytes

We therefore aimed to identify the biological characteristics of theEBAF gene in myocardial cells. The flow cytometry analysis showedDFO-induced apoptosis in cardiomyocytes (Fig. 3A). Then we assessedthe expression of EBAF by comparing control cells and DFO-treatedcells using quantitative RT-PCR. As shown in Fig. 3B, the average ex-pression of EBAF in DFO-induced apoptotic cells was lower than theEBAF expression in untreated cells. The results indicated that down-regulation of EBAF was associated with cardiomyocyte apoptosis.Thus, down-regulation of EBAF may be potentially involved in heartmalformations.

3.4. Cited2 played an important role in the expression regulation of EBAFgene

To explore the molecular mechanism of EBAF under-expression inVSD, we first identified the EBAF regulator. Western blotting showed

that low expression of cited2 was associated with low expression ofEBAF in VSD tissues (Fig. 4A). Furthermore, we also find that low ex-pression of CITED2 is associated with down-regulation of EBAF incardiomyocyte apoptosis (Fig. 4B). In addition, luciferase reporter as-says showed that cited2 over-expression up-regulated EBAF promoteractivity (Fig. 4C). Data above showed that cited2 participated in regulat-ing the expression of EBAF gene.

3.5. Smad2 cooperates with cited2 to activate the EBAF promoters

The promoter fragments of the EBAF gene contain several GCAT orGTCTmotifs, which are potential BR-Smad binding sites [16]. Moreover,through sequence analysis and reference basis, we detected several pu-tative binding sites for the transcription factor smad2 in the ASE regionof the EBAF promoter (Fig. 5A) [20,21]. Therefore, these data promptedus to examine the possibility of the cooperation of smad2with cited2 onupregulation of EBAF gene. Firstly, luciferase reporter assays showedthat smad2 and cited2 overexpression increased EBAF promoter activity

Fig. 6. p300 over-expression up-regulates the expression of EBAF by enhancing the acety-lation levels of histone H4 at the EBAF promoter. (A) p300 up-regulated EBAF promoteractivity by luciferase reporter assay. (B) Over-expression of p300 increased the enrich-ment of p300 at the EBAF promoter. HCM cells transfected with p300 expression plasmidwere subjected to ChIP assays. (C) The acetylation levels of histones H4 were assessedusing ChIP assays. HCM cells were transfected with p300 expression plasmid andharvested for ChIP assays. Samples were immunoprecipitated using anti-Ac-H3 or Ac-H4antibodies, and the precipitated DNAs were amplified using PCR.

2150 D. Su et al. / Biochimica et Biophysica Acta 1832 (2013) 2145–2152

by 2.1-fold and 1.9-fold, respectively (Figs. 5B and 4C). Then we foundthe promoter activity of the EBAF gene was up-regulated significantlyhigher upon the co-expression of cited2 and smad2 than transfectionwith smad2 and cited2, respectively (Fig. 5C). Furthermore, chromatinimmunoprecipitation (ChIP) assays showed that smad2 was able tobind to the EBAF promoter, and that the over-expression of smad2 in-creased its enrichment at the EBAF promoter (Fig. 5D, left). Likewise,cited2 was also present at the EBAF promoter (Fig. 5D, right). This indi-cates that both smad2 and cited2 participate in regulating the expres-sion of EBAF.

3.6. The role of p300 in inducing histone H4 hyperacetylation at the EBAFpromoter

P300 has a domainwith intrinsic histone acetyltransferase (HAT) ac-tivity, and it plays an important role in chromatin remodeling [17]. p300was reported to interact physically with cited2. We examined whetherp300 regulated the expression of EBAF. Luciferase reporter assaysshowed that the promoter activity of EBAF was up-regulated by trans-fection with p300 expression plasmid (Fig. 6A). ChIP assays showed

Fig. 7. smad2, cited2, and p300 coordinate to up-regulate the expression of EBAF. (A) cited2transfectedwith EBAF reporter plasmid, together with smad2, p300, and cited2 expression plasin the same complex by Co-IP assays. HCM cells transfected with the smad2-GFP expression plWestern blotting with anti-p300 or anti-cited2 antibody.

that over-expression of p300 increased the enrichment of p300 at theEBAF promoter (Fig. 6B and Supplemental Fig. S1A). Furthermore, H4acetylation levels at the EBAF promoter were also found to increasewhen there was p300 over-expression.Meanwhile, there was no signif-icant increase in the H3 acetylation levels (Fig. 6C, SupplementalFig. S1B and C). These results clearly indicate that the presence ofp300 at the EBAF promoter enhances the acetylation levels of histoneH4 at the EBAF promoter.

3.7. p300, smad2, and cited2 in the same transcriptional complex

We next examined whether smad2, cited2, and p300 cooperated toregulate the expression of EBAF. Luciferase reporter assays showed thatthe promoter activity of EBAF genewas significantly increased by 7-fold,which is higher than the increased folds of EBAF promoter activity bytransfection with p300, Smad2 and Cited2, respectively (Fig. 7A). Wesought to investigate, using Co-IP assays, whether there was a physicalinteraction between these transcriptional regulators. In this experi-ment, HCM cells were transfected with smad2-GFP expression vectors.The immunoprecipitationwas conductedwith the anti-smad2 antibodyand the precipitated proteinswere assayed usingWestern blottingwiththe antibody against cited2 and p300, respectively. As can be seen inFig. 7B, smad2, cited2, and p300 were found to be present in the samecomplex and worked together to contribute to the regulation of theEBAF expression.

3.8. Down-regulation of EBAF caused abnormal activation of NODAL-PITX2C pathway

Finally, we explore the down-stream events of EBAF down-regulation that may influence the VDS genesis. Quantitative RT-PCR re-sults showed that PITX2Cwas up-regulated in the disease tissues of VSDpatients (Fig. 8A, B and Supplemental Fig. S3B). Meanwhile, we foundthat overexpression of NODAL gene upregulated PITX2C gene promoteractivity by luciferase reporter assay. However, the upregulation ofPITX2C promoter activity was attenuated when EBAF expression plas-mid was co-transfected with NODAL expression plasmid (Fig. 8C).These results indicated down-regulation of EBAF may partly participat-ed in occurrence of VSD by abnormally activating the NODAL-PITX2Cpathway.

4. Discussion

Observational epidemiological studies of CHD have demonstratedthe existence of gene predispositions. However, the genetic basis ofthemajority of CHD remains unknown, for most CHDs, the basic patho-genetic mechanism of VSD is still unclear. However, to date, very few

, p300, and smad2 cooperatively up-regulated EBAF promoter activity. HCM cells weremids, and lysed for luciferase reporter assays. (B) smad2, cited2, and p300were presentedasmid were prepared and precipitated with the anti-smad2 antibody, then detected using

Fig. 8. EBAF down-regulation led to abnormal activation of NODAL-PITX2C pathway. (A) Comparison of PITX2C mRNA expression in 32 samples (16 samples with VSDs and 16 sampleswithout), as revealed by quantitative PCR. **p b 0.01 versus normal group. (B) Comparison of PITX2CmRNA expression between the normal controls (combined) and the different sub-types of VSDs, respectively. M-VSD, S-VSD, A-VSD representmembranous ventricular septal defects, supracristal ventricular septal defects, and atrioventricular canal type defects, respec-tively. **p b 0.01 versus normal group. (C) The up-regulation of PITX2C promoter activity by NODAL overexpression was attenuated when EBAF expression plasmid was co-transfectedwith NODAL expression plasmid. HCM cells were transfected with EBAF reporter plasmid, together with NODAL and EBAF expression plasmids, and lysed for luciferase reporter assays.*p b 0.05 versus transfection of vector group; **p b 0.01 versus transfection of vector group; #p b 0.05 versus transfection of NODAL expression plasmid only group. Data are fromthree independent experiments. (D) A sketched diagram shows the regulatory mechanisms driven by smad2, cited2, p300, EBAF and NODAL-PITX2C pathway in heart development.

2151D. Su et al. / Biochimica et Biophysica Acta 1832 (2013) 2145–2152

VSD-causing genes have been identified. Thus, the identification andcharacterization of novel genes and proteins associated with VSD re-mains an important issue. Here, we find that the expression of EBAFwas frequently down-regulated in VSD tissues for the first time (Fig. 2and Supplemental Fig. S3A). Furthermore, we also found that down-regulation of EBAF was associated with cardiomyocyte apoptosis byAnnexin V-FITC and RT-PCR assays (Fig. 3A and B).

Then,we explored the down-streamevents of EBAF down-regulationthat may influence the VDS genesis. EBAF, a feedback inhibitor of NODALsignaling, blocks NODAL signaling and rapidly terminates the expressionof both NODAL and its target gene such as PITX2C [18]. PITX2C up-

regulation was recently reported to be associated with the occurrenceof ventricular septal defect [19]. Thus, we speculate that EBAF down-regulation may partly lead to VSD by abnormal activation of NODALpathway. Luciferase reporter assay and quantitative RT-PCR results indi-cated down-regulation of EBAF may partly participate in occurrence ofVSD by abnormally activating NODAL-PITX2C pathway (Fig. 8 and Sup-plemental Fig. S3B). This study established the association of EBAF withVSD genesis.

Here, we report that the expression of EBAF was frequentlydown-regulated in VSD tissues for the first time (Fig. 2A), furthermore,we found that EBAF was frequently down-regulated in membranous

2152 D. Su et al. / Biochimica et Biophysica Acta 1832 (2013) 2145–2152

ventricular septal defects specifically (Fig. 2B). Though the expression ofEBAF was also down-regulated in supracristal ventricular septal defectsand atrioventricular ventricular septal defect (two other subtypes ofVSD) in our study (Fig. 2B), the results should be further confirmed byincreasing the numbers of the samples. However, we were not able toconfirm this now because of low incidence of the two subtypes of VSDand so no tissues could be acquired right now. The findings of thisstudy indicate that decreased EBAF expression may play an importantrole in the onset of different subtypes of ventricular septal defects(Table 1, Fig. 2 and Supplemental Fig. S3A).

Low expression of EBAF and its regulatory mechanism of under-expression in patients with VSD have been clarified in this study. West-ern blotting has indicated that cited2 expression is associated withunder-expression of EBAF (Fig. 4). Further investigation using luciferasereporter assays and ChIP assays suggested that smad2 and cited2 didcontribute to the regulation of EBAF expression (Fig. 5). Moreover, onesignificant finding that has emerged from this study is the role that his-tone acetylase p300 inmediating cardiac development, through acetylat-ing the histone H4 at the promoter regions of EBAF by luciferase reporterand ChIP assays (Fig. 6A, B and C). cited2 protein, the transactivator, hasno DNA-binding domain. p300, an acetyl transferase, has no DNA-binding domain neither. They should be recruited by transcriptional fac-tor to the promoter of target gene to coactivate gene transcription. Co-IPassays showed that the interaction of smad2 with p300 and cited2(Fig. 7), may facilitate the recruitment of p300 and cited2 to the promot-er of EBAF gene. These data suggest that the cooperation of p300, cited2,and smad2 plays a critical role in transcriptional regulation of EBAF inventricular septal confirmation in the process of cardiac development.

CITED2 plays an important role in the normal development of theheart based on previous findings from gene-targeted mice. However,the correlation between CITED2 gene and congenital cardiacmalforma-tion genesis needs to be further validated with human sample. A recentstudy showed that CITED2 mutations were associated with congenitalheart defects genesis by using human blood samples [22]. Our studyalso found in abnormal expression of CITED2 was correlative with VSDgenesis by using patients' tissues (Fig. 2). Further study showed thatCITED2 participated in EBAF gene transcriptional regulation (Fig. 3).Thus, our findings provide another important evidence that CITED2 isessential for congenital heart defects genesis. Moreover, we also foundthe responsive gene of CITED2 in cardiac development.

To conclude, this study established the association of EBAFwith VSDgenesis with patients' tissues. Further study showed that the expressionpattern of EBAFwas potentially associated with cardiomyocyte apopto-sis by Annexin-V and RT-PCR assays. As shown in Fig. 8D, We alsoshowed that down-regulation of EBAF may partly participate in the oc-currence of VSD by abnormally activating the NODAL-PITX2C pathway.Furthermore, our data revealed that histone acetyltransferase p300wasinvolved in the activation of EBAF through inducing hyperacetylation ofhistone H4 at the EBAF promoter. Luciferase reporter assays and co-immunoprecipitation revealed that EBAF is regulated by a transcription-al complex including p300, smad2, and cited2. The sketched diagramshowed the regulatory mechanisms driven by smad2, cited2, p300,EBAF and NODAL-PITX2C pathway in heart development. Generallyspeaking, our findings reveal not only a novel regulatory mechanismof EBAF; they also indicate that smad2, cited2, p300 and EBAF play im-portant roles in modulating the confirmation of the ventricular septal.

Acknowledgments

This research was supported by grants from the Central Public-interest Scientific Institution Basal Research Fund (Grant No.

2010GJSSJKA07, 2012GJSSJKA02). We are grateful to all the peoplewho helped us successfully complete the research. We thank theShengjing Hospital of China Medical University for its support inthe myocardial tissue collection.

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.bbadis.2013.07.013.

References

[1] J.I. Hoffman, S. Kaplan, The incidence of congenital heart disease, J. Am. Coll. Cardiol.39 (2002) 1890–1900.

[2] E. Garne, C. Stoll, M. Clementi, Evaluation of prenatal diagnosis of congenital heartdiseases by ultrasound: experience from 20 European registries, Ultrasound Obstet.Gynecol. 17 (2001) 386–391.

[3] B.G. Bruneau, Transcriptional regulation of vertebrate cardiac morphogenesis, Circ.Res. 90 (2002) 509–519.

[4] C. Zhu, Z.B. Yu, X.H. Chen, Y. Pan, X.Y. Dong, L.M. Qian, S.P. Han, Screening for dif-ferential methylation status in fetal myocardial tissue samples with ventricularseptal defects by promoter methylation microarrays, Mol. Med. Rep. 4 (2011)137–143.

[5] D. Srivastava, E.N. Olson, A genetic blueprint for cardiac development, Nature 407(2000) 221–226.

[6] J. Bentham, S. Bhattacharya, Genetic mechanisms controlling cardiovascular develop-ment, Ann. N. Y. Acad. Sci. 1123 (2008) 10–19.

[7] S. Hammer, M. Toenjes, M. Lange, J.J. Fischer, I. Dunkel, S. Mebus, C.H. Grimm, R.Hetzer, F. Berger, S. Sperling, Characterization of TBX20 in human hearts and its reg-ulation by TFAP2, J. Cell. Biochem. 104 (2008) 1022–1033.

[8] M. Nemer, Genetic insights into normal and abnormal heart development, Cardiovasc.Pathol. 17 (2008) 48–54.

[9] S.D. Bamforth, J. Braganca, C.R. Farthing, J.E. Schneider, C. Broadbent, A.C. Michell, K.Clarke, S. Neubauer, D. Norris, N.A. Brown, R.H. Anderson, S. Bhattacharya, Cited2controls left-right patterning and heart development through a Nodal-Pitx2c path-way, Nat. Genet. 36 (2004) 1189–1196.

[10] S.D. Bamforth, J. Braganca, J.J. Eloranta, J.N. Murdoch, F.I. Marques, K.R. Kranc, H.Farza, D.J. Henderson, H.C. Hurst, S. Bhattacharya, Cardiac malformations, adrenalagenesis, neural crest defects and exencephaly in mice lacking Cited2, a newTfap2 co-activator, Nat. Genet. 29 (2001) 469–474.

[11] S. Tang, P. Snider, A.B. Firulli, S.J. Conway, Trigenic neural crest-restricted Smad7over-expression results in congenital craniofacial and cardiovascular defects, Dev.Biol. 344 (2010) 233–247.

[12] P.A. Roest, D.G. Molin, C.G. Schalkwijk, L. van Iperen, P. Wentzel, U.J. Eriksson, A.C.Gittenberger-de Groot, Specific local cardiovascular changes of Nepsilon-(carboxymethyl)lysine, vascular endothelial growth factor, and Smad2 in the devel-oping embryos coincide with maternal diabetes-induced congenital heart defects,Diabetes 58 (2009) 1222–1228.

[13] H. Juan, H. Hamada, Roles of nodal-lefty regulatory loops in embryonic patterning ofvertebrates, Genes Cells 6 (2001) 923–930.

[14] W.J. Weninger, K. Lopes Floro, M.B. Bennett, S.L. Withington, J.I. Preis, J.P. Barbera, T.J.Mohun, S.L. Dunwoodie, Cited2 is required both for heart morphogenesis and estab-lishment of the left-right axis in mouse development, Development 132 (2005)1337–1348.

[15] S. Han, J. Lu, Y. Zhang, C. Cheng, L. Han, X.Wang, L. Li, C. Liu, B. Huang, Recruitment ofhistone deacetylase 4 by transcription factors represses interleukin-5 transcription,Biochem. J. 400 (2006) 439–448.

[16] A. Zwijsen, K. Verschueren, D. Huylebroeck, New intracellular components ofbone morphogenetic protein/Smad signaling cascades, FEBS Lett. 546 (2003)133–139.

[17] X. Wang, L. Pan, Y. Feng, Y. Wang, Q. Han, L. Han, S. Han, J. Guo, B. Huang, J. Lu, P300plays a role in p16(INK4a) expression and cell cycle arrest, Oncogene 27 (2008)1894–1904.

[18] C. Meno, J. Takeuchi, et al., Diffusion of nodal signaling activity in the absence of thefeedback inhibitor Lefty2, Dev. Cell 1 (1) (Jul 2001) 127–138.

[19] A. Saxena, C.J. Tabin, miRNA-processing enzyme Dicer is necessary for cardiac out-flow tract alignment and chamber septation, Proc. Natl. Acad. Sci. U. S. A. 107 (1)(Jan 5 2010) 87–91.

[20] D. Besser, Expression of nodal, lefty-a, and lefty-B in undifferentiated human embry-onic stem cells requires activation of Smad2/3, J. Biol. Chem. 279 (43) (Oct 22 2004)45076–45084.

[21] K. Yashiro, Y. Saijoh, R. Sakuma, M. Tada, N. Tomita, K. Amano, Y. Matsuda, M.Monden, S. Okada, H. Hamada, Distinct transcriptional regulation and phylogeneticdivergence of human LEFTY genes, Genes Cells 5 (5) (May 2000) 343–357.

[22] S. Sperling, C.H. Grimm, I. Dunkel, S. Mebus, H.P. Sperling, A. Ebner, R. Galli, H.Lehrach, C. Fusch, F. Berger, S. Hammer, Identification and functional analysis ofCITED2 mutations in patients with congenital heart defects, Hum. Mutat. 26 (6)(Dec 2005) 575–582.