TBX5 genetic variants and SCD-CAD susceptibility: insights from Chinese Han cohorts

Ethics statement

Ethical approval for this study was granted by the Ethical Committee of Soochow University (approval number: ME81772029). Prior to the investigation, written informed consent was acquired from all healthy blood donors and the victim’s family.

Study populations

Our study included a cohort of 553 healthy controls and 201 SCD-CAD cases, all of whom were genetically unrelated Han Chinese from central-southern, eastern, southern and southwestern China, respectively. The SCD-CAD case set was gathered from 2012 to 2020 at Soochow University, Institute of Forensic Science at Guizhou Medical University, Shanghai Key Laboratory of Forensic Medicine at Ministry of Justice, the Medicolegal Expertise Center of Sun Yat-sen University and the Forensic Medical Identification Center of Central South University. The criteria for both sample inclusion and exclusion align with those previously described (Wang et al., 2017; Zou et al., 2019). Rigorous forensic pathological investigations were undertaken to rule out any fatal features except for coronary atherosclerosis of varying degrees, thereby attributing the sudden deaths to coronary heart diseases. Comprehensive toxicological examinations were also conducted to avoid the influence of poisoning and pharmaceutical substances. Age (±5 years) and gender matched 553 healthy controls were enrolled through community nutrition information questionnaires performed during the same time frame within identical areas. Venous blood was taken simultaneously from those healthy controls after written informed consent. These healthy donors had no family history of cardiovascular disease or sudden death. A collection of 23 cardiac tissues from healthy human donors was gathered at the Medicolegal Expertise Center of Soochow University and Guizhou Medical University during traffic accident autopsy. Upon retrieval through medicolegal autopsy, obtained tissues were promptly preserved at −80 °C for following protein, DNA or RNA extraction.

Bioinformatic analysis

Variants within TBX5 gene were sourced from dbSNP database (Sherry et al., 2001). Putative functionality of these variants were examined using RegulomeDB and HaploReg databases (Ward & Kellis, 2012; Boyle et al., 2012). Analysis of splicing quality trait loci (sQTL) was conducted utilizing data from the Genotype-Tissue Expression (GTEx) database (GTEx Consortium, 2013). Linkage disequilibrium (LD) analysis was performed applying R package (version 4.0.3; R Core Team, 2020), and genotype data was supplied by Ensembl Genome Browser (Chan, 2018; Martin et al., 2023). JASPAR was employed for transcription factors binding site (TFBS) prediction within the LD block, and the results were visualized by UCSC genome browser (Castro-Mondragon et al., 2022). Binding sites predicted in the JASPAR CORE collection with scores above 400, indicative of a p-value less than 0.0001, were included. GRNdb and ChIPBase databases were utilized to predict potential target genes of TBX5 (Zhou et al., 2017; Fang et al., 2021). Gene ontology (GO) annotations were performed using the DAVID database (Sherman et al., 2022). Additionally, the STRING database (Version 11.5) was employed to predict protein-protein interaction (PPI) network (Szklarczyk et al., 2023).

DNA extraction and genotyping

Genomic DNA was extracted from both blood samples and human myocardium tissues employing the Blood DNA Kit and Blood spot DNA kit (TIANGEN). A pair of genotyping primers (Forward: ACTTTTCTTCTCCAGTGCCTAC, Reverse: CCATTGTCCTTGAGTGTGAT) designed to amplify the polymorphic region containing rs11278315 were synthesized by Genewiz Company (Suzhou, China). For detailed primer information, please refer to Table S1. The products of the polymerase chain reaction were determined by 7% nondenaturing polyacrylamide gel electrophoresis (PAGE) and observed through silver staining method, during which a double-blinded approach was adopted (Allen, Graves & Budowle, 1989). Regarding quality control, a random selection of 50 samples were subjected to direct sequencing and randomly selected 10% of all DNA samples were verified independently by two investigators.

RNA extraction and quantitative real-time PCR (qRT-PCR) analysis

Total RNA from human myocardial tissues, extracted by TRIzol reagent (Invitrogen), was subsequently quantified with the NanoDrop (Thermo Scientific). The A260/A280 ratio was 1.8−2.3. According to NCBI database, we identified three experimentally validated isoforms of TBX5: NM_000192.3 (variant 1), NM_080717.4 (variant 3), and NM_181486.4 (variant 4). Variant 1 as the longest variant encodes the same protein as variant 4, whereas variant 3 lacks an exon containing the transcription start site, resulting a protein with a shorter N-terminus. Due to the high sequence similarity among three isoforms and the limited suitability of specific fragments for primer design, we finally selected a pair of universal primers capable of amplifying all three isoforms, as well as a pair of primers that specifically amplify variant 4, a representative isoform suggested by Matched Annotation from the NCBI and EMBL-EBI (Morales et al., 2022). The primer sequences designed for isoforms and GAPDH were given below and the specificity of primers were confirmed with Primer-BLAST:

TBX5-Forward (variant 1, 3, 4): TAGATTACACATCGTGAAAGCGG, TBX5-Reverse (variant 1, 3, 4): GGAAAGACGTGAGTGCAGAAC;

TBX5-Forward (variant 4): AGTAAACCCCGCATAAACCCC, TBX5-Reverse (variant 4): GCAAGGTTCTGCTCTCGTTCG;

GAPDH-Forward: CTCTCTGCTCCTCCTGTTCGAC, GAPDH-Reverse: TGAGCGATGTGGCTCGGCT.

Quantitative real-time PCR (qRT-PCR) analysis

Following reverse transcription with the HiScrip III All-in-one RT SuperMix Perfect for qRT-PCR (Vazyme), acquired cDNA was used immediately for ChamQ Universal SYBR qRT-PCR Master Mix (Vazyme) employing the LightCycler96 instrument (Roche) with a minimum of twice replicates. Remaining RNA and cDNA were stored at −80 °C. Reaction mixes with a total volume of 20ul were prepared from 10 µl ChamQ Universal SYBR qRT-PCR Master Mix, 7.84 µl RNase-free water, 0.08 µl forward primer (50µM), 0.08 µl reverse primer (50µM) and 2 µl cDNA per well. The thermal cycling conditions was as specified by the manufacturer: initial denaturation at 95 °C for 30 s, followed by 40 cycles of 95 °C for 10 s and 60 °C for 30 s. Melting curve analysis was performed as follow: 95 °C for 15 s, 60 °C for 60 s and 95 °C for 15 s. Standard curve was generated to test amplification efficiency of over 90%. Set GAPDH as internal control and samples with ins/ins plus ins/del genotype as control group for the 2^−ΔΔCT algorithm-based calculation of quantitative results, where Cq refers to cycle threshold. The formulas used was as follow: 2^−ΔΔCT = 2^{−[(Cq of target gene−Cq of GAPDH)experiment−(Cq of target gene−Cq of GAPDH)control]}

Western blot assay

Total protein in human myocardium tissues were extracted using Phenylmethanesulfonyl fluoride and Radio Immunoprecipitation Assay. Homogenized protein(60ug) was separated via 10% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and subsequently transferred onto a PVDF membrane (Millipore). After a two hours block at room temperature and overnight incubation with the primary antibodies Anti-TBX5 (Rabbit, Cat# 13178-1-AP, 1:500, Proteintech) and Anti-GAPDH (Mouse, Cat# AB-M-M001, 1:1000, Goodhere Biotechnology) at four degrees centigrade, the probed proteins were then treated with the secondary antibody (1:2000, Biyotime Biotechnology) for one hour. The hypersensitive Enhanced Chemiluminescence (ECL) device was used to visualize bands. The Western blot results were finally analyzed using ImageJ analysis software (National Institutes of Health).

Cell cultures

Human embryonic kidney 293T (HEK 293T) cell lines were cultured in Dulbecco’s Modified Eagle Medium (Procell) supplemented with 10% fetal bovine serum (Inner Mongolia Opcel Biotechnology Co., Ltd) and 1% penicillin-streptomycin (Invitrogen) at 37 °C in a humidified incubator with 5% CO2. The cell lines were originated from Shanghai Cell Bank of Chinese Academy of Sciences. The cell lines were characterized by Genetic Testing Biotechnology Corporation (Suzhou, China) using short tandem repeat (STR) markers.

Reporter plasmid vector construction and dual Luciferase reporter assay

DNA segments containing 326 or 318 base pairs, centered around rs11278315, were directly synthesized and subcloned into XbaI and FseI sites of pGL3-control vector by Genewiz Company (Suzhou, China). This process yielded vectors containing wild-type (pGL3-WT) with an insertion allele or mutant-type (pGL3-MT) with a deletion allele, with orientation and sequence authenticated by direct sequencing. After 24 h of cell seeding in 24-well plate (1 × 105 cells per well), a co-transfection by the jetPRIME transfection reagent (Polyplustransfection) was completed by adding 400 ng of the reconstructed pGL3-WT or pGL3-MT vector along with 20 ng of the pRL-SV40 vector (Promega). Empty pGL3-control vector was also transfected as a negative control. Cell lysis was performed at 24 h post transfection, followed by disruption in 100 µL of Passive Lysis Buffer (Promega). Measurement of Firefly and Renilla luciferase activity was performed using SpectraMax iD3 (China), with a minimum of eight wells per conditions measured.

Statistical analysis

IBM SPSS Statistics Subscription (version 27.0.1) and GraphPad Prism (version 9.0.0) were carried out for statistical analysis. Hardy-Weinberg equilibrium assessed by chi-square test was utilized to ensure the representativeness of genotype distribution in control samples. The odds ratio (OR) and 95% confidence interval (95% CI) were calculated through unconditional logistic regression, with age and gender adjusted, so as to look into the relationships between rs11278315 and the risk of SCD-CAD. Student’s t-test was used for comparing the means of two groups for functional evaluation. In our study, we set up p < 0.05 as statistical significance threshold and all statistical tests were two-sided.

TBX5 gene downstream candidate variants selection

Figure 1 depicts the flowchart employed in our current study. Given the potential regulatory role and the presence of various functional elements of downstream non-coding regions in gene expression, we conducted a screening of the 3′ UTR and 500-bp downstream region of TBX5. In Table S2, we have listed all 11 variants with a minor allele frequency (MAF) > 0.05. Based on the compatibility with the widely used capillary electrophoresis platforms in forensic genetics, two selected indel sites, rs35534655 and rs11278315, which exhibited length polymorphism characteristics analogous to STR markers, were picked as candidate variants. Nonetheless, rs35534655 is a multiple adenine (A) duplicate that make genotyping complicated. Therefore, we chose rs11278315 as our candidate variant. Through LD analysis within 10 kb of rs11278315, we discovered that rs11278315 shared a LD block with six other sites (Fig. 2A), which were listed in Table S3 . Among these sites, rs11278135 is strongly associated with rs1895597 and rs3825214, which was associated with electrocardiographic traits and atrial fibrillation in GWAS analysis.

Database retrieval and functional prediction analysis

Database annotation suggested that rs11278315 was enriched with H3K36me3 (associated with active gene transcription) or H3K27me3 (Inhibitory promoter related), H3K4me1 (only found in enhancers) (Fig. 2B). Besides, we predicted 17 TFBS (Table S2) within the LD block of rs11278315 by using JASPAR database. Then we performed the GO enrichment analysis on these predicted sites by querying DAVID database. Significance was defined as p < 0.05, and the most enriched GO terms are presented as the -log10 of the uncorrected p-values. GO analysis was divided into three aspects: biological process (BP), and cellular component (CC) and molecular function (MF). In terms of BP level, the top terms primarily pertained to regulation of transcription from polymerase II promoter and regulation of gene expression, whereas the most enriched terms consist of chromatin, nucleus, and transcription factor complex in CC. At the MF level, the enriched terms included transcription factor, activator activity, and RNA polymerase II regulatory region sequence-specific DNA binding (Fig. 3A).

In addition, given that TBX5 is a transcription factor, we predicted 43 downstream target genes from databases intersection (Fig. 3B) and observed following degrees of enrichment: cardiac conduction system development, regulation of cardiac muscle contraction, ventricular cardiac muscle tissue morphogenesis at the BP level; sarcomere and neuromuscular junction at the CC level; epidermal growth factor receptor binding, protein serine/threonine/tyrosine kinase activity, ATP binding at MF level (Fig. 3C). In the context of protein expression, we analyzed the functional and physical protein associations related to TBX5, and generated a set of protein-protein interaction (PPI) network based on the STRING database with a high confidence score of ≥ 0.7 (Fig. 3D).

The associations of rs11278315 with SCD-CAD susceptibility

Table 1 summarized the demographic characteristics of individuals who participated in current investigation. The detailed features and forensic autopsy findings regarding the SCD-CAD cases were listed in Table S4. Of note, cases exhibited a higher prevalence in males, having a 9.05:1 male to female ratio. Among all cases, 44 cases (24.31%) occurred following strenuous exercise or significant physical exertion, while 60 cases (33.15%) took place in stress-inducing conditions such as emotional events, arguments, alcohol consumption, injury, and surgical treatments, categorizing them as stress-related stimuli. Additionally, 16 cases (8.84%) transpired during sleep, and 81 cases (44.75%) occurred in a tranquil state or lacked eyewitnesses, which were classified as nonspecific.

As illustrated in Fig. 4, the genotyping and sequencing results for rs11278315 confirmed its polymorphic character. The observed rs11278315 genotype frequencies in control group were consistent with Hardy-Weinberg equilibrium. Allele and genotype frequencies of rs11278315, and odds ratio (OR) with its 95% confidence interval (CI) for both groups, were listed in Table 2. According to codominant model, the del/del genotype was linked with a significantly reduced risk of SCD-CAD when compared to individuals with the ins/ins homozygous allele (OR: 0.47, 95% CI [0.29–0.75], p = 0.0014), in line with recessive and additive models (OR: 0.64, 95% CI [0.44–0.93], p = 0.0194; adjusted OR: 0.70, 95% CI [0.55–0.88] p = 0.0019). As described, these findings indicated a significant association between rs11278315 and susceptibility to SCD-CAD, with the deletion allele lowering the risk of SCD-CAD in its carriers.

The correlation between rs11278315 genotype and expression of TBX5

Based on GTEx database, we identified these variants rs11278315, rs7977083, rs10744823, rs12367410, rs3825214 within the LD block as sQTL variants of TBX5 (Fig. 5). After genotyping cardiac tissues from healthy human donors, we identified 11 tissue samples with del/del, four with ins/ins and eight with ins/del genotype. Our qRT-PCR analysis using these 24 same normal myocardial samples showed that individuals carrying deletion allele had significant lower expression level compared to those carrying insertion allele (Fig. 6A). Intriguingly, variant 4, when carried with the deletion allele, exhibited relatively lower expression level compared to results from total isoforms. As the results from total isoforms showed a 0.73-fold lower expression (p = 0.0002), while results from variant 4 showed a 0.41-fold lower expression (p < 0.0001), with the ins/ins and ins/del genotype set as reference. For another, western blot analysis also correspondingly reflected considerably higher-level trend of TBX5 protein with the ins/ins genotype of rs11278315 compared to the ins/del and del/del genotypes (p >  0.05) (Fig. 6B). Although the results were not statistically significant, this may be due to our small sample size and inner differences between samples. Collectively, those indicates that rs11278315 has a regulatory effect on TBX5 expression.

The influence of rs11278315 on gene transcription activity

With respect to the correlation between rs11278315 and gene transcription activity, we performed a dual luciferase reporter assay to assess relative fluorescence intensity at 24 h after transfection. It is noteworthy that either pGL3-WT or pGL3-MT transfected group presented higher luciferase activity than negative pGL3-control vector co-transfected concurrently. Likewise, the group transfected with pGL3-MT displayed a significantly lower luciferase activity in comparison to the group transfected with pGL3-WT, highlighting the regulatory properties of rs11278315 variant (Fig. 7).