Variants in a cis-regulatory element of TBX1 in conotruncal heart defect patients impair GATA6-mediated transactivation

Background TBX1 (T-box transcription factor 1) is a major candidate gene that likely contributes to the etiology of velo-cardio-facial syndrome/DiGeorge syndrome (VCFS/DGS). Although the haploinsufficiency of TBX1 in both mice and humans results in congenital cardiac malformations, little has been elucidated about its upstream regulation. We aimed to explore the transcriptional regulation and dysregulation of TBX1. Methods Different TBX1 promoter reporters were constructed. Luciferase assays and electrophoretic mobility shift assays (EMSAs) were used to identify a cis-regulatory element within the TBX1 promoter region and its trans-acting factor. The expression of proteins was identified by immunohistochemistry and immunofluorescence. Variants in the cis-regulatory element were screened in conotruncal defect (CTD) patients. In vitro functional assays were performed to show the effects of the variants found in CTD patients on the transactivation of TBX1. Results We identified a cis-regulatory element within intron 1 of TBX1 that was found to be responsive to GATA6 (GATA binding protein 6), a transcription factor crucial for cardiogenesis. The expression patterns of GATA6 and TBX1 overlapped in the pharyngeal arches of human embryos. Transfection experiments and EMSA indicated that GATA6 could activate the transcription of TBX1 by directly binding with its GATA cis-regulatory element in vitro. Furthermore, sequencing analyses of 195 sporadic CTD patients without the 22q11.2 deletion or duplication identified 3 variants (NC_000022.11:g.19756832C > G, NC_000022.11:g.19756845C > T, and NC_000022.11:g. 19756902G > T) in the non-coding cis-regulatory element of TBX1. Luciferase assays showed that all 3 variants led to reduced transcription of TBX1 when incubated with GATA6. Conclusions Our findings showed that TBX1 might be a direct transcriptional target of GATA6, and variants in the non-coding cis-regulatory element of TBX1 disrupted GATA6-mediated transactivation. Supplementary Information The online version contains supplementary material available at 10.1186/s13023-021-01981-4.

Most research has focused on the TBX1 downstream transcription factors in cardiovascular development, such as Isl1, Tbx5, Pax9, Fgf8, Fgf10, Pitx2 and Mef2c [13,14,[24][25][26][27][28][29]. Genetic studies indicate that TBX1 is a dosage-dependent transcriptional regulator [30,31], but the mechanisms of TBX1 gene upstream regulation remain largely unexplored. Previous studies have shown that TBX1 is regulated by sonic hedgehog (Shh) signaling [32,33], vascular endothelial growth factor (VEGF) [34] and Ripply3 [35] or has a feedback loop with retinoic acid (RA) signaling [36], which are involved in pharyngeal arch development. However, how these signals affect the transcriptional regulation of TBX1 remains unclear. Only Forkhead proteins have been reported to regulate the transcription of Tbx1 directly [27,33,37], concerning that Tbx1 is regulated by Shh signaling cascade through a Fox-binding site upstream of Tbx1 in the pharyngeal endoderm [33]. Cis-regulatory elements such as enhancers and promoters are crucial for regulating gene expression [38,39], and cardiac cis-regulatory elements functions are activated by a large number of both commonly expressed and cardiac-specific transcription factors, such as SNAI3, NKX2-5, GATA4, GATA6, SOX9, and NFATC1 [40]. Genome-wide association studies (GWAS) suggest that genetic variants associated with the risk of cardiovascular disease are enriched within annotated candidate cis-regulatory elements (cCREs) [41,42]). The study of the cis-regulatory elements of TBX1 will help to further understand the role of TBX1 in the occurrence of CTD.
In this study, we found that GATA6 could positively activate TBX1 transcription by directly binding to the cis-regulatory element within the TBX1 promoter region, which contains GATA binding sites. We demonstrated that the expression patterns of TBX1 and GATA6 overlapped in the pharyngeal arches of human embryos. In addition, we identified 3 rare TBX1 mutations within the cis-regulatory element in sporadic CTD patients without mutations in the coding region of a known CTD pathogenic gene and 22q11.2 deletions. Luciferase assays showed that these variants impaired the GATA6-mediated transcriptional activation of TBX1. These results provide novel insight into the molecular mechanisms associated with TBX1 during the pathogenesis of CTD.

Identification of a novel cis-regulatory element in human TBX1 adjacent to the 5′ flanking region
To identify potential cis-regulatory elements controlling TBX1 transcription, a series of truncated human TBX1 5′-flanking regions were cloned into luciferase reporter vectors (Fig. 1). The subsequent dual-luciferase reporter analysis of these fragments showed that p-138/+ 514, which is located on chromosome 22: 19756565-19757216, significantly increased transcriptional activity by nearly 15-fold compared with the control vector in the C2C12 cell line (Fig. 1); and similar results were obtained in the NIH/3T3 cell line (Additional file 1: Fig. S1). This , deletion analysis of the 5′ human TBX1 flanking region. Relative luciferase activity of different 5′-serially deleted TBX1 reporter constructs transfected in C2C12 cells, which were normalized to Renilla and represented as the fold increase when compared to the pGL3-basic vector. Data are shown as mean ± SEM, statistical significance was calculated by one-way ANOVA with Dunnett's post hoc test, n = 3 independent experiments, ***p < 0.001 vs pGL3-basic suggests that the TBX1-138/+ 514 region around the TBX1 transcription start site (TSS) may be a cis-regulatory element, resulting in increased reporter expression.

GATA6 significantly improved the promoter activity of TBX1 by binding with a cis-regulatory element within the TBX1 promoter
Cardiac cis-regulatory elements play important roles during transcriptional regulation by recruiting cardiacspecific transcription factors [40]. Therefore, to identify specific transcription factor binding sites within TBX1-138/+ 514, a non-exhaustive search for predicted candidate transcription factor binding sites was performed using JASPAR (http:// jaspar. gener eg. net/) [45], and the results showed that this region contained six GATA motifs, three NKX2-5 motifs and one serine response factor (SRF) motif (Fig. 2a). GATA4 and GATA6, two highly conserved transcription factors, are primarily expressed in the myocardium [46,47], and are capable of binding to the same predicted GATA-binding sites [48][49][50]. We cotransfected the p-138/+ 514 reporter gene with individual expression vectors containing GATA4, GATA6, NKX2-5 or SRF in the C2C12 and NIH/3T3 cell lines (Fig. 2b). The results showed that GATA6 significantly increased luciferase activity by nearly sixfold compared with the control pcDNA3.1 vector, NKX2-5 slightly increased luciferase activity by only twofold, and GATA4 and SRF had no effects on luciferase activity (Fig. 2b).

Endogenous overlapping expression of GATA6 and TBX1 in the pharyngeal arches of human embryos
We performed immunohistochemistry (IHC) analyses for TBX1 and GATA6 in human Carnegie Stage 13 (CS13) embryos. As previously reported, TBX1 was predominantly expressed in the pharyngeal arches, whereas GATA6 was widely expressed in the pharyngeal arches, along with all atrial and ventricular chambers (Fig. 3a). To further determine the distinct but overlapping expression patterns of TBX1 and GATA6, we performed double-immunofluorescence staining. The examination of confocal projections revealed that TBX1 and GATA6 expression primarily overlapped in pharyngeal arches, which is a transient vertebrate-specific complex that participates in OFT development 27 (Fig. 3b).

GATA6 directly binds with the TBX1-138/+514 cis-regulatory element within the TBX1 promoter
To investigate whether GATA6 directly binds to the TBX1-138/+ 514 cis-regulatory element within the TBX1 promoter, EMSA was performed using GATA6 protein and a biotin-labeled TBX1 + 115/+ 302 probe containing three candidate GATA binding sites. An in vitro-translated GATA6 protein was verified by immunoblotting (Additional file 1: Fig. S3). The EMSA results showed that the GATA6 protein and the biotin-labeled probe TBX1 + 115/+ 302 formed a protein-DNA supershift, which could be inhibited by the addition of a cold competition probe or an antibody targeting GATA6 (Fig. 4). Together, these results indicated that the human GATA6 protein was able to directly bind to the TBX1-138/+ 514 cis-regulatory element within the TBX1 promoter.

Variants within the TBX1 cis-regulatory element in CTD patients impair activation of the TBX1 promoter by GATA6
In an effort to explore whether TBX1 cis-regulatory element variants are related to CTD, we screened the TBX1-138/+ 514 sequence in 195 patients with CTD. The proportions of CTD variations in these patients were as follows: 43.59% TOF, 23.08% DORV, 14.36% pulmonary atresia with ventricular septal defect (PA/VSD), 15.38% transposition of the great arteries (TGA), 2.56% interrupted aortic arch (IAA) and 1.03% persistent truncus arteriosus (PTA) ( Table 1). None of the patients or control individuals presented copy number variants of 22q11.2 or any other chromosomal region (data not shown) [53]. We identified three heterozygous sequence variations in three patients (NC_000022.11:g.19756832C > G, NC_000022.11:g.19756845C > T and NC_000022.11:g. 19756902G > T), and all of the results were verified by The influence of transcription factors on the p-138/+ 514 reporter gene in the C2C12 and NIH/3T3 cell lines. Luciferase analysis after cotransfection with luciferase reporter constructs containing the TBX1 cis-regulatory element TBX1-138/+ 514 and the expression vector of GATA6, GATA4, NKX2-5, SRF or empty vector (pcDNA3.1). The results were normalized to Renilla and presented as the fold induction over the pcDNA3.1 vector. Data are shown as mean ± SEM, significance was calculated by one-way ANOVA with Dunnett's multiple comparisons test, n = 3 independent experiments, **p < 0.01 vs pcDNA3.1, ***p < 0.001 vs pcDNA3.1 in each cell line. c The influence of GATA6 on the TBX1 promoter reporter gene p-1179/+ 514 in the C2C12 and NIH/3T3 cell lines. Luciferase analysis after cotransfection with luciferase reporter p-1179/+ 514 and GATA6 expression vector or empty vector (pcDNA3.1). The results were normalized to Renilla and presented as the fold induction over the pcDNA3.1 vector. Data are shown as mean ± SEM, two-tailed unpaired t test was used for statistical calculation, n = 3 independent experiments, **p < 0.01 vs pcDNA3.1, ***p < 0.001 vs pcDNA3.1 in each cell line. d and e The effects of the transcription factor GATA6 on the TBX1 promoter reporter gene in C2C12 and NIH/3T3 cell lines. Luciferase analysis after cotransfection of luciferase reporter and GATA6 expression construct or empty vector (pcDNA3.1). The results were normalized to Renilla and presented as the fold induction over basal reporter (pGL3-basic + pcDNA3.1). Data are shown as mean ± SEM, n = 3 independent experiments. Significance was calculated by two-way ANOVA with Sidak's post hoc test, ### p < 0.001. Significance was calculated by two-way ANOVA with Tukey's multiple comparisons test, *p < 0.05, **p < 0.01, ***p < 0.001 Sanger sequencing (Fig. 5a). The clinical information and TBX1-138/+ 514 sequence mutations in patients with CTD are shown in Table 2. These three patients had PA, TOF and TGA respectively, combined with other congenital heart diseases (CHDs) such as patent ductus arteriosus (PDA), VSD and pulmonary stenosis (PS). The identified variant (NC_000022.11:g.19756832C > G) is located within the GATA binding site, whereas the other two variants (NC_000022.11:g.19756845C > T; NC_000022.11:g. 19756902G > T) are located between the GATA binding sites (Fig. 5b). None of the three variants were found in the 145 ethnically matched healthy controls, but two variants (NC_000022.11:g.19756845C > T and NC_000022.11:g. 19756902G > T) were reported in the 1000 Genomes database and gnomAD database (Table 2). Several known CHD pathogenic genes (GATA4, GATA6, GATA5, HAND2, FOG2, NKX2-5, MEF2C, TBX1, TBX5, SOX7, PITX2, and MESP1) were also screened in these three patients. There were no nonsynonymous variants identified in these known CHD pathogenic genes, including TBX1.
We constructed vectors containing each mutation individually using the template TBX1 promoter p-1179/+ 514, named M-130, M-143, and M-200. The wild-type (WT) or mutant TBX1 promoter reporter gene were transfected alone into the C2C12 cell line or cotransfected with the GATA6 expression vectors to test whether these variants had any effects on TBX1 transcriptional activation. The luciferase activity measurements showed that the mutation vector alone had no effect on TBX1 promoter activity itself, but when transfected with GATA6, TBX1 promoter activity decreased significantly compared with the WT (Fig. 5c). Taken together, these results revealed that these variants within the TBX1 cis-regulatory element inhibited GATA6-mediated transactivation of TBX1.

Discussion
Our results identified a novel non-coding cis-regulatory element within the TBX1 proximal promoter region, which is capable of binding with GATA6; and variants in the cis-element in CTD patients disrupt GATA6mediated TBX1 transactivation. CTD, the common type of cyanotic cardiac disease, is usually defined as malformations of the OFT, and TBX1 participates in OFT development. The findings not only help to deepen the understanding of the molecular mechanism of TBX1 transcription, but also provide ideas for the pathogenesis of CTD.
TBX1, the major candidate gene for 22q11.2DS, is a dosage-sensitive transcription factor contributing to the formation of the OFT during cardiogenesis [15,31]. Understanding the mechanisms of TBX1 upstream transcriptional regulation may uncover novel determinants of CTD. Transcriptional regulation is a significant mechanism that controls the spatial and temporal expression of genes during cardiac development, and non-coding cis-regulatory elements are critical participants in the transcriptional regulation process [54]. A series of truncated portions of the human TBX1 flanking region were used to identify transcriptional cis-regulatory elements associated with TBX1. TBX1-138/+ 514, a 0.65 kb fragment around the TBX1 TSS, was identified as a cis-regulatory element. As the TBX1-138/+ 514 region consists of a cCRE-PLS (ENCODE Accession: EH38E2149905, also named TBX1-112/+ 235), combined with previous reports [51,52], this suggests that TBX1-138/+ 514 may be a cis-regulatory element within the TBX1 promoter.
Cis-regulatory elements can alter transcriptional activity by recruiting DNA-binding transcription factors [40], and we identified potential transcription factor binding sites within the TBX1-138/+ 514 sequence, including GATA/NKX2-5/SRF motifs. Our results revealed a clear requirement for GATA6 to activate the TBX1-138/+ 514 cis-regulatory element within the TBX1 promoter (TBX1-1179/+ 514). GATA6 belongs to the GATA family of zinc finger transcription factors that consists of six known family members (GATA1-6) [55]. GATA6 is capable of binding the GATA consensus motif (A/T)GATA(A/G) through zinc finger domains [48][49][50], which are essential cis-elements in the promoter or enhancer regions of a variety of cardiac-specific genes [49,56]. Accumulating research underscores the important roles played by GATA binding motifs in the promoter regions of several myocardial genes, such as wnt2 [57] and Bmp-4 [48], which were both identified as direct downstream targets for Gata6. Previously, no evidence of a genetic interaction between GATA6 and TBX1 has been reported. Using an EMSA, we showed that GATA6 could directly activate TBX1 transcription by binding with the cis-regulatory element TBX1-138/+ 514. Our results also showed that TBX1 and GATA6, which are two key cardiac regulatory factors, showed an overlapping distribution in the pharyngeal arches of human embryos. Previous reports showed that Semaphorin 3c (Sema3c) transcript was downregulated in the OFT of Tbx1 −/− mouse embryos [17], and SEMA3C was regulated directly by GATA6 in the developing OFT [58,59], suggesting that GATA6 may share a common molecular pathway with TBX1, or at least in part during OFT development [5,59]. Our findings provide evidence of a molecular link between GATA6 and TBX1, confirming the above hypothesis. The conditional inactivation of Gata6 in cardiac neural crest cells results in failed OFT septation [58], and CHD patients with GATA6 mutations have OFT malformations [59][60][61], demonstrating a role for GATA6 in CTD.
TBX1 is critical for normal pharyngeal arch development, and both reductions and increases in TBX1 levels increase the risk of pharyngeal arch birth defects. Any allelic TBX1 variations that cause TBX1 levels to fall below a threshold for proper pharyngeal arch development increase the risk of birth defects [34]. Although TBX1 is the causative gene of 22q11.2, only a few mutations in the coding region of TBX1 were identified in patients with the 22q11.2 syndrome phenotype without detectable deletion of 22q11.2, and it is possible that mutations in the cis-regulatory region of TBX1 also play a role [9,30,37]. The identification of a novel TBX1 cis-regulatory element provided an opportunity to examine whether mutations within the cis-regulatory element occurred in patients with the 22q11DS phenotype but no chromosomal deletion. In our study, three mutations within the TBX1 cis-regulatory element region in CTD patients were consistently associated with significantly reduced transcriptional activation by GATA6. The mutation NC_000022.11:g.19756832C > G is located within the GATA binding site and is therefore expected to impact the GATA6-binding ability. The other two variants NC_000022.11:g.19756845C > T and NC_000022.11:g. 19756902G > T identified in this study were located adjacent to GATA binding sites, which may influence the biological function of the TBX1 gene through other mechanisms. Therefore, further study is warranted. There are also reports of TBX1 cis-regulatory elements being involved in cardiac development [32,33,51,62], such as a mutation of the Fox-binding site upstream of Tbx1 abolishing Tbx1 expression in the pharyngeal endoderm [32,33] and heterozygous variants NC_000022.11:g.19756055C > T and NC_000022.11: g.19756212A > C upstream of the TBX1 TSS in VSD patients significantly decreasing the activity of the TBX1 gene promoter [51]. In addition, studies have reported that variants in the regulatory region of other OFT development-related genes are also associated with CTD, such as WNT5A [63].  Although almost all 22q11.2DS patients are hemizygous for the deletion, highly variable genotypes and phenotypes are commonly observed [13,64]; thus, it has been hypothesized that genes located outside the deleted region may serve as potential modifiers of 22q11.2DS [13,65,66]. Our findings uncover a novel mechanism of the GATA6 transcription factor in the regulation of the TBX1 gene by a cis-regulatory element, and variants in the cis-regulatory element in CTD patients impair TBX1 promoter activity by GATA6, thus providing a basis for the above hypothesis. In addition, with imaging in assessing the phenotype of CTD, computed tomography angiography (CTA) and contrast magnetic resonance angiography (MRA) precisely detect associated aortic, pulmonary venous and coronary anomalies, and other CHDs [67,68], which will contribute to a better

Conclusions
In conclusion, our data provide the first evidence of a genetic link between TBX1 and GATA6 signaling. A GATA6-binding cis-regulatory element within the TBX1 promoter was identified, and variants in the cis-regulatory element impaired GATA6-mediated TBX1 transactivation. However, our study is preliminary. We do not have sufficient evidence to show that TBX1 expression is directly dependent on the GATA6-binding cis-regulatory element during cardiovascular development in vivo, and the mutations in the TBX1 cis-regulatory element reported in this paper are causal for CTD. This is the limitation of our study. Further functional assays for the GATA6-TBX1 pathway in CTD patients need to be performed. Our study can be used as the first evidence for future research on the pathogenic non-coding variants in cis-regulatory elements in CTD. This is hoped to be further confirmed in other populations.

Human samples
Human embryonic samples were obtained at Department of Obstetrics and Gynecology of Xinhua Hospital according to the procedure we previously reported [69]. In brief, the pregnant women who could not continue pregnancy for certain reasons, and elective medical termination of pregnancy were performed under the Maternity Protection Law of China. All the donors signed the informed consents from Xinhua Hospital, Affiliated to Shanghai Jiao Tong University School of Medicine. The study was approved by the Medical Ethics Committee of Xinhua Hospital (no. XHEC-C-2012-018). The aborted embryos were isolated with an Olympus stereomicroscope and categorized in accordance with the Carnegie stage (CS). The well-preserved specimens evaluated as normal, were fixed in 4% paraformaldehyde (PFA) overnight, embedded in paraffin sagitally, and then sectioned at the thickness of 4 μm.
Blood samples from 195 sporadic CTD patients (75 female, 120 male; median age 2.3 years) of Han ethnicity, which included 85 TOF, 45 DORV, 28 PA/VSD, 30 TGA, 5 IAA and 2 PTA patients, were collected in Shanghai Children's Medical Center (SCMC) from November 2011 to January 2014 (Table 1) [53]. Clinical records like echocardiography, computed tomography angiocardiography and magnetic resonance angiocardiography [67,68] were reviewed by pediatric cardiologists before recruitment, and patients who had extracardiac anomalies were excluded. None of which harbored 22q11.2 deletion or duplication tested by CNVplex [53]. A total of 145 healthy control individuals (57 female, 88 male, median age 6.43 years) were ethnically gender-matched and recruited from the same geographical area. DNA was extracted from the whole blood samples using the QIAamp DNA Blood Mini Kit (Qiagen, Duesseldorf, Germany). The human experimentations were approved by the Medical Ethics Committee of Xinhua Hospital (no. XHEC-C-2012-018) and Shanghai Children's Medical Center (no. SCMC-201015). Fully written informed consent was obtained from all participants or their guardians. All procedures were in accordance with the ethical standards of the institutional and national research committees and the Declaration of Helsinki.  Table S1) and 3′ PCR primers (common 3′ primer + 514 R, -1179/-1 R, Additional file 2: Table S1) were used to amplify serial deletion fragments of the 5′ end region of TBX1 by the template TBX1 (NC_000022.11). Numbering shows the position relative to the TSS of the TBX1 gene at g.19756703 (+1) of NC_000022.11. Then, these fragments were subcloned into the KpnI and XhoI sites of the pGL3-basic Table 2 Clinical information and mutations of TBX1 cis-regulatory element in patients with CTD  Table S1.

Cell culture and transfection
C2C12 (mouse myoblast) and NIH/3T3 (mouse embryo fibroblast) cells were obtained from Cell Bank of Chinese Academy of Sciences (Shanghai, China) and cultured at 37℃ in a 5% CO 2 humidified incubator in DMEM (Gibco, Portland, OR, USA) supplemented with 10% fetal bovine serum (Gibco, Portland, OR, USA).
For luciferase reporter assay, cells were transfected with firefly luciferase reporter plasmid and Renilla luciferase expression plasmid pRL-TK (Promega, Madison, Wisconsin, USA) as an internal control per well using Fugene HD. The firefly and Renilla luciferase activities were measured 40 h after transfection by the Dual Luciferase reporter assay system (Promega, Madison, Wisconsin, USA). The transcriptional activities were represented as ratios of firefly luciferase activities over Renilla luciferase activities.

Electrophoretic mobility shift assays (EMSA)
Plasmids pcDNA-3.1 and pCDNA3.1-GATA6 were used for in vitro protein synthesis with the TNT ® T7 Quick Coupled transcription/translation system (Promega, Madison, Wisconsin, USA) according to the manufacturer's instructions. The oligonucleotide probes (+ 115/+ 302 bp) acquired from the TBX1 cis-regulatory element were synthesized by PCR using 5′-end biotinlabeled primers (GATA-biotin + 115/+ 302, Additional file 2: Table S1). Unlabeled consensus oligonucleotides were used as competitors (GATA cons, Additional file 2: Table S1) [70]. The EMSA was performed using the LightShift Chemiluminescent EMSA kit (Thermo Scientific, USA) according to the manufacturers' protocol. Briefly, for the binding reactions with volume 20 μl, 4uL TNT protein sample were incubated with 500 fmol of 5′-labeled biotin probes in the following buffer: 2 μl of 10 × binding buffer, 1 μl of 50% Glycerol, 1 μl of 1% NP-40, 1 μl of 1 M KCl, 1 μl of 100 mM MgCl2, 1 μl of 200 mM EDTA and 1 μl Salmon DNA. All reactions were incubated at room temperature for 20 min. For competitor lanes, a 120-fold excess of competitor oligonucleotides was added and incubated at room temperature for 10 min before the addition of biotin-labeled probe. For antibody lanes, assays were conducted in a similar manner with the following exceptions: addition of ∼5 μg of antibodies against GATA6 (sc-9055 X, Santa Cruz Biotechology Inc., USA) and incubation at room temperature for 20 min before the addition of the biotin-labeled DNA probe. The products were then separated by 6.5% nondenaturing polyacrylamide gels in 0.5 × TBE. The DNA-protein complex was electrotransferred to a positively charged of nylon membrane (Pierce), and the signal was detected by the chemiluminescent nucleic acid detection.

Immunohistochemistry
Paraffin-embedded embryo sections underwent deparaffinization, hydration, and antigen retrieval with citrate buffer (Servicebio technology, China). Then quenched endogenous peroxidase activity with 3% H 2 O 2 , blocked in 3% BSA for 1 h at room temperature, followed by incubating with rabbit polyclonal antibody against TBX1 (Abcam Inc., ab18530, Cambridge, UK) or GATA6 (Santa Cruz Biotechnology Inc., sc-9055, USA) overnight at 4℃. Next, the secondary antibody, HRP-labelled antirabbit immunoglobulin (Gene Tech, Shanghai, China) was applied for 45 min at room temperature. Immunoreactions were visualized by diaminobenzidine substrate (DAB), and nuclei were stained with hematoxylin. Finally, sections were rinsed with water, dehydrated, cleared and mounted with neutral balsam. Immunohistochemical images were photographed with microscope slide scanner Pannoramic MIDI (3D HISTECH).

Western blot analysis
Cells were lysed by RIPA buffer, mixed with SDS loading buffer, and denatured at 100 °C for 10 min. Proteins were separated by 10% SDS-PAGE gels (Beyotime Biotech Inc., Shanghai, China) and transferred electrophoretically onto polyvinylidene difluoride membrane (Merck Millipore, Darmstadt, Germany). Then the membranes were blocked with 5% skim milk for 2 h at room temperature, then incubated overnight at 4 °C with anti-GATA6 (Sigma-Aldrich, G2298) and anti-GAPDH (Sigma-Aldrich, G9545) antibodies. After the membranes were rinsed, they were incubated with the HRP-conjugated secondary antibody for 2 h at room temperature. Detection was performed by an enhanced chemiluminescence western blot detection system (Millipore, Billerica, MA, USA) according to the manufacturer's instructions.

Gene sequencing and mutation characterization
Mutation screening analysis of the TBX1-138/+ 514 cisregulatory element sequence was performed in patients and control individuals by Sanger sequencing. The referential genomic DNA sequence of TBX1 was derived from a nucleotide (NC_000022.11). Mutations in TBX1-138/+ 514 with a prevalence of under 0.1% in control populations (under 0.1% in both the 145 healthy controls and public variant databases, including 1000 Genomes and gnomAD) were regarded as candidate variants. The variants were validated by Sanger sequencing, and the primers used in the present study are summarized in Additional file 2: Table S1.
To validate the potential function of identified TBX1-138/+ 514 cis-regulatory element variants, the coding sequences and splicing sites of these known CHD pathogenic genes (GATA4, GATA6, GATA5, HAND2,  FOG2, NKX2-5, MEF2C, TBX1, TBX5, SOX7, PITX2,  MESP1) were also screened in CTD patients and controls using the EasyTarget ® amplification kit (Genesky Biotechnologies Inc, Shanghai, China). Nonsynonymous mutations in the coding and splicing regions with a prevalence of under 0.1% in control populations (under 0.1% in both the 145 healthy controls and public variant databases, including 1000 Genomes, ExAC and gnomAD) were regarded as CHD pathogenic gene mutations.

Statistical analysis
Data were analyzed by PRISM 6.0 (GraphPad Software Inc., La Jolla, CA, USA) software and were presented as the mean ± standard error of the mean (S.E.M.). The experiments were repeated at least three times independently. Two-tailed unpaired t-test was used to determine significant differences between two groups. One-way ANOVA with Dunnett's post hoc test was used to analyze differences among three or more groups. Differences between two independent variables were analyzed by two-way ANOVA with Sidak's multiple comparisons test, Tukey's multiple comparisons test or Dunnett's multiple comparisons test. P < 0.05 were considered statistically significant.
Additional file 1. Fig. S1 Deletion analysis identifies a 0.65-kb region around the TBX1 TSS essential for transcriptional activity in the NIH/3T3 cell line. Fig. S2 UCSC Genome Browser views of cCREs around the TBX1 TSS and the underlying DNase and ChIP data (hg38 human genomes). Fig. S3 Verification of in vitro-translated GATA6 by reticulocyte lysates.

Additional file 2. Table S1
Primers used for PCR, EMSA and Site-directed mutagenesis.