Abstract
Background
Congenital heart disease (CHD) is a leading cause of neonatal morbidity and mortality, with a complex aetiology involving genetic and environmental factors. Cardiac transcription factors, such as NKX2.5, GATA4 and TBX5, are essential for heart development, and their gene polymorphisms may contribute to CHD. Additionally, maternal periconceptional environmental exposures may influence CHD risk. Limited studies explore gene–gene and gene–environment interactions in CHD pathogenesis.
Methods
A hospital-based case-control study analyzed 585 CHD cases and 600 controls. Associations between transcription factor gene polymorphisms, maternal environmental exposures and CHD risk were assessed using logistic regression, multiplicative interaction models and model-based multifactor dimensionality reduction.
Results
Several transcription factor gene polymorphisms were significantly associated with CHD risk. Specifically, NKX2.5 (rs118026695), GATA4 (rs12458) and TBX5 (rs11067101, rs6489956) increased CHD risk, while NKX2.5 (rs703752) and GATA4 (rs4841588, rs904018) reduced risk. Gene–gene interactions among these loci further modulated CHD risk. Maternal environmental factors, such as gestational diabetes, pre-gestational diabetes, adverse pregnancy history, certain medications and lifestyle factors, were associated with higher CHD risk, certain gene–environment interactions showed suggestive associations with CHD risk, with some combinations reaching statistical significance prior to multiple testing correction.
Conclusion
Infant NKX2.5, GATA4 and TBX5 gene polymorphisms significantly affect CHD risk, and maternal periconceptional environmental exposures may further influence this risk through certain gene–environment combinations. These findings provide insights into CHD aetiology and highlight the need for preventive strategies. Further studies are necessary to validate these results and clarify underlying mechanisms.
Keywords: Congenital heart disease, cardiac transcription factor, gene polymorphisms, gene-gene interaction, gene-environment interaction
Key Messages
Polymorphisms in infant cardiac transcription factor genes NKX2.5, GATA4 and TBX5 are significantly associated with congenital heart disease, manifesting as either risk or protective factors.
Adverse maternal periconceptional environmental exposures, such as history of diabetes, antibiotic use and hair dyeing, are independently associated with an increased risk of congenital heart disease in offspring.
Significant gene–gene and gene–environment interactions, particularly involving NKX2.5 with gestational diabetes and GATA4 with lifestyle factors, suggest a synergistic aetiology for congenital heart disease.
1. Introduction
Congenital heart disease (CHD) is a structural disorder of the heart and great vessels present at birth [1]. It is the congenital disability with the highest disease burden globally [2], accounting for 36.6% of incidents and 39.5% of deaths [3], and is to be close to 2,305.2 cases per 100,000 live births in 2019, with more than 210,000 deaths due to CHD [4]. As in the cause of other complex diseases, the aetiology of CHD is the result of a combination of genes and the environment [5,6], with approximately 90% attributable to the combined effects of genes and environmental factors, 8% to chromosomal or genetic factors and 2% by environmental teratogens [7]. Therefore, early identification of the causes of CHD and intervention targeting its risk factors to reduce the risk of onset and improve the prognosis of children remains the most cost-effective health economics strategy. Human heart development is controlled by several overlapping morphogenetic systems that regulate intricate cardiac transcriptional networks [8], and precise regulation of transcription factors is essential for the correct differentiation and patterning of heart development [9].
Transcription factors regulate the aspects of heart development, including terminal differentiation, cardiac chamber identification, the establishment of patterning boundaries and the formation of transcriptional gradients, and the mutations in their genes may cause CHD [10–12]. Several cardiac transcription factors linked to CHD development have been identified, including the NKX2.5, the GATA, the T-Box, the Forkhead Box, the HAND and the Nuclear Receptor families [13]. Among them, NKX2.5, GATA4 and TBX5 genes are interdependent, coordinating cardiac gene expression, differentiation and morphogenesis, while other transcription factors serve as accessory factors for these core regulators [10,14]. The NKX2.5 gene acts near the top of a large transcriptional cascade, controlling multiple cardiac genes, and is a central regulator of heart development. The GATA4 gene regulates cardiac morphogenesis and myocyte proliferation, while the TBX5 is important for cardiac morphogenesis and conduction systems [13,15,16], and these genes are the most critical transcription factor genes affecting CHD development. Some studies have explored the association between NKX2.5, GATA4 and TBX5 gene polymorphisms and CHD, and they found that the rs2277923 in the NKX2.5 gene [17–19], the rs56166237 in the GATA4 gene [19,20] and rs6489956 in the TBX5 gene [21] have been associated with CHD among different races. However, controversy persists over the association between some loci and CHD, and most current studies on transcription factor gene polymorphisms and CHD have focused on analyzing specific loci, such as those mentioned above, while ignoring the effects of gene–gene interactions.
During the highly ‘plastic’ period of rapid fetal development, fetal development is inextricably linked to the environmental factors to which the pregnant woman is exposed [22]. Pregnancy from 3 to 8 weeks is a critical developmental period for the embryonic heart, during which maternal exposure to adverse factors may lead to the occurrence of CHD in offspring [23]. Multiple studies have shown that maternal exposure risk factors during preconception are associated with the occurrence of CHD [24–26], these factors are often excessive toxic substances or lack of essential nutrients, or alter the uterine environment, acting either directly on the embryo itself or indirectly to cause cardiac malformations in the offspring [27]. Few studies, however, have examined the interaction transcription factor gene polymorphisms and maternal periconceptional environment factors with CHD.
Based on the above, a hospital-based case-control study was conducted to explore the following issues: (i) the association between infant transcription factor NKX2.5, GATA4 and TBX5 gene polymorphisms and CHD; (ii) the association between maternal exposure to the periconceptional environment and offspring CHD; (iii) the interaction of infant transcription factor NKX2.5, GATA4 and TBX5 gene polymorphisms with CHD and (iv) the interaction between these gene polymorphisms and maternal periconceptional environmental exposures (particularly maternal disease history, periconceptional medication use and adverse lifestyle factors) with CHD risk. Building on the above content, we propose a hypothesis: specific infant cardiac transcription factor polymorphisms and adverse maternal periconceptional exposures would be associated with CHD risk and that their interaction effects might further affect CHD susceptibility. This research aims to promote the development of primary prevention strategies to reduce the global burden of CHD.
2. Methods
2.1. Study design and participants
This single-center case-control study was performed at Hunan Children’s Hospital (Changsha, China) from December 2018 to June 2021. A total of 1,185 participants were recruited, with 585 children with CHD and their mothers enrolled in the case group from the Cardiothoracic Surgery Department, and 600 healthy children or children without congenital disability or cardiac disease, along with their mothers, were included in the control group from the Child Healthcare Department. The diagnosis of CHD was based on ICD-10 (International Classification of Diseases, Tenth Revision) and confirmed by echocardiography and cardiac operation.
This study was approved by the Xiangya School of Public Health, Central South University (No. XYGW-2018-07). We obtained written informed consent from all mothers.
2.2. Inclusion and exclusion criteria
For the case group, the inclusion criteria were: (i) diagnosis of CHD in children based on echocardiography and cardiac operation; (ii) no familial relationship among all subjects; and (iii) voluntary enrollment of study subjects. The exclusion criteria were: (i) syndromic CHD patients, who were diagnosed with malformations other than cardiac malformation or other congenital disorders; (ii) pregnancies achieved through assisted reproductive technology; (iii) mothers are unable to complete the investigation; (iv) adoptive or stepmothers; and (v) refusing to provide blood samples or written informed consent.
For the control group, the inclusion criteria were: (i) children without CHD and other congenital disability or cardiac disease diagnosed by a medical examination, and (ii) no familial relationship with cases or controls. Except for the first criterion, the exclusion criteria were identical to those for the case groups.
2.3. Questionnaire survey and information/sample collection
To comprehensively assess maternal periconceptional exposures potentially associated with CHD risk in offspring, we used a self-designed structured questionnaire to collect information on the following domains: maternal demographic characteristics (pregnancy age, educational level, residence location and annual income in the past year); reproductive history (parity and adverse pregnancy history); disease and family history (congenital malformation, pre-gestational diabetes, pre-gestational hypertension, consanguineous marriage and family history of congenital malformation); present pregnancy conditions (pre-pregnancy body mass index, gestational diabetes and gestational hypertension); periconceptional health conditions and medication use (influenza exposure, fever exposure, contraceptive pill use, antibiotic use, ovulation-promoting drug use, antibiotic use in the first trimester and folic acid use) and periconceptional person lifestyle (active smoking, passive smoking, alcohol consumption, negative life events, hair dyeing or perming and frequency of cosmetic use). The full questionnaire is provided in Supplementary Questionnaire S1. All information was collected through face-to-face interviews conducted by professionally trained investigators.
About 3–5 millilitres of peripheral venous blood was collected from the child and placed in ethylenediaminetetraacetic acid (EDTA)-treated anticoagulant tubes. The blood was immediately centrifuged into plasma and blood cells and stored at −80 °C for subsequent analysis.
2.4. SNP selection and genotyping
The single nucleitide polymorphisms (SNPs) of the infant NKX2.5 gene, GATA4 gene and TBX5 gene were selected by screening the mutation variants in CHD patients in previous studies and searching the dbSNP database of NCBI and whose minor allele frequency (MAF) was required to be 5% or more. Eventually, 19 SNPs were selected for this study, including 4 SNPs in the NKX2.5 gene (rs6882776, rs118026695, rs2277923 and rs703752), 9 in the GATA4 gene (rs4841588, rs884662, rs804287, rs3203358, rs867858, rs2645457, rs10108052, rs12458 and rs904018) and 6 in the TBX5 gene (rs12426660, rs10850326, rs3782467, rs11067101, rs6489956 and rs883079). DNA extraction and genotyping were performed by BioMiao Biological Technology (Beijing, China) Co., Ltd using QIAamp DNA Mini Kit (Qiagen, Valencia, CA, USA) and time-of-flight mass spectrometry MassArray system (Agena iPLEX Assay, San Diego, CA, USA).
2.5. Statistical analysis
The Pearson χ2 test or Fisher’s exact probability test is used to assess the difference between categorical variables. Hardy-Weinberg equilibrium (HWE) test was used to measure the balance of genotype distribution frequencies in the control populations (significance level at p < 0.01). The logistic regression was used to measure the association between genes, environment, and the risk of CHD. False Discovery Rate P-value (FDR_P) was used for multiple hypothesis testing, and FDR_P < 0.1 was considered statistically significant. The genetic model selection method [28] was used to identify the genetic model of each gene locus by comparing the ZHWDTT value calculated through the HWE trend test with C = 1.645. The effect of gene–gene and gene–environment interactions on CHD was assessed using multiplicative interaction models. In addition, model-based multifactor dimensionality reduction (MB-MDR) was used to explore possible high-dimensional gene–gene and gene–environment interactions (using the ‘mbmdr’ R package) [29]. Sample size was estimated based on preliminary study findings. Using the standard formula for case-control studies, we calculated the minimum sample size required for both groups across representative genetic loci. To accommodate a potential 20% nonresponese or missing data rate, the final target was set to at least 550 subjects per group (Table S1). All tests were performed with a two-sided P value not exceeding 0.05, except where otherwise specified. All analyses were performed using R software version 4.2.2 (R Foundation for Statistical Computing, Vienna, Austria).
3. Results
3.1. Baseline characteristics of study participants
From December 2018 to June 2021, 1,185 eligible pairs of children and their mothers, including CHD case group 585 and control group 600, were recruited into this study. All 585 CHD cases were classified as non-syndromic CHD; among them, 555 (94.9%) were sporadic cases (no family history of congenital malformations), while 30 (5.1%) were familial cases with at least one relative affected by congenital malformations. The baseline characteristics of the two groups are summarized in Table 1. Statistically significant differences were found in the following factors: maternal pregnancy ages, maternal educational level, residence location and annual income in the past year. These factors would be considered confounders and adjusted in subsequent analyses. The clinical characteristics of the CHD patients by family history of congenital malformation are summarized in Table S2, maternal education level, adverse pregnancy history, family history of consanguineous marriages, pre-pregnancy BMI and negative life events in the first trimester have statistical significance between two groups, and the difference in clinical characteristics between the two groups is relatively small.
Table 1.
Baseline characteristics in case and control group.
| Baseline characteristics | Control group (n = 600) (%) | Case group (n = 585) (%) | χ2 | P |
|---|---|---|---|---|
| Pregnancy age (years) | ||||
| <25 | 119 (19.8) | 164 (28.0) | 11.014 | 0.004 |
| 25–29.9 | 251 (41.8) | 223 (38.1) | ||
| ≥30 | 230 (38.4) | 198 (33.9) | ||
| Education level | ||||
| Less than primary or primary | 7 (1.2) | 84 (14.4) | 200.800 | <0.001 |
| Junior high school | 117 (19.5) | 250 (42.7) | ||
| Senior middle school | 211 (35.2) | 155 (26.5) | ||
| College and above | 265 (44.2) | 96 (16.4) | ||
| Residence location (urban areas) | 267 (44.5) | 162 (27.7) | 36.230 | <0.001 |
| Annual income in the past 1 year | ||||
| ≤50,000 (RMB) | 174 (29.0) | 463 (79.1) | 302.674 | <0.001 |
| 50,001–100,000 (RMB) | 264 (44.0) | 88 (15.0) | ||
| 100,001–150,000 (RMB) | 51 (8.5) | 12 (2.1) | ||
| >150,001 (RMB) | 111 (18.5) | 22 (3.8) |
3.2. Association of maternal environmental factors with CHD in offspring
Table S3 shows the univariate analysis of the association between maternal environmental factors and CHD in offspring. The results of multivariate regression analysis showed that maternal adverse pregnancy history (OR = 1.50, 95%CI: 1.11–2.05), pre-gestational diabetes history (OR = 3.23, 95%CI: 1.69–6.18), consanguineous marriages history (OR = 13.88, 95%CI: 2.84–67.82), family history of congenital malformation (OR = 5.51, 95%CI: 1.64–18.50), gestational diabetes in present delivery (OR = 3.15, 95%CI: 1.61–6.15), gestational hypertension in present delivery (OR = 2.54, 95%CI: 1.19–5.42), antibiotics use in 6 months before pregnancy (OR = 2.84, 95%CI: 1.41–5.73), ovulation-promoting drugs use in 6 months before pregnancy (OR = 4.38, 95%CI: 1.66–11.54), influenza exposure in the first trimester (OR = 1.57, 95%CI: 1.08–2.28), alcohol consumption in 3 months before pregnancy (OR = 1.93, 95%CI: 1.16–3.23), alcohol consumption in the first trimester (OR = 2.11, 95%CI: 1.12–3.96), hair dyeing or perming in periconceptional period (OR = 2.58, 95%CI: 1.50–4.44) and often use of cosmetics in the periconceptional period (OR = 1.69, 95%CI: 1.14–2.53) were association with higher CHD risk (Figure 1).
Figure 1.
Association of maternal environmental factors with CHD in offspring.
Note: Adjusted for maternal pregnancy age (years), maternal education level, residence location and annual income in the past 1 year.
3.3. Association of infant transcription factor gene polymorphisms with CHD
Table S4 presents the genotype frequencies in the control group for each SNP of the NKX2.5, GATA4 and TBX5 genes, as well as the results of the HWE test. The rs867858 locus of the GATA4 gene does not follow the HWE test, so it was excluded from subsequent analyses. Table S5 shows the genetic model based on the HWE trend test. These genetic models will be used to assess the association between each SNP and CHD risk.
The association of infant transcription factor gene polymorphisms with the risk of CHD is illustrated in Figure 2. After adjustment for the potential confounders, rs118026695 (TC vs. TT, OR = 1.58,95%CI: 1.10–2.28; CC vs. TT, OR = 5.28,95%CI: 1.26–22.14; the additive model, OR = 1.71,95%CI: 1.23–2.37) of the NKX2.5 gene and rs12458 (GA vs. GG, OR = 2.02,95%CI: 1.41–2.89; AA vs. GG, OR = 2.50,95%CI: 1.64–3.83; the additive model, OR = 1.59,95%CI: 1.29–1.97) of the GATA4 gene, as well as rs11067101 (GA vs. GG, OR = 1.50,95%CI: 1.08–2.11), rs6489956 (CT vs. CC, OR = 1.84,95%CI: 1.29–2.62; TT vs. CC, OR = 3.49,95%CI: 1.10–11.08; the recessive model, OR = 2.98,95%CI: 1.94–4.58) of the TBX5 gene were associated with higher CHD risk. In addition to this, rs703752 (CA vs. CC, OR = 0.64, 95%CI: 0.41–0.99) of the NKX2.5 gene and rs4841588 (GG vs. TT, OR = 0.41, 95%CI: 0.23–0.73), rs904018 (AG vs. AA, OR = 0.70, 95%CI: 0.51–0.97; the additive model, OR = 0.78, 95%CI: 0.62–0.98) of the GATA4 gene were associated with a lower risk of CHD.
Figure 2.
Association of infant transcription factor gene polymorphisms with CHD in offspring.
Note: Adjusted for maternal pregnancy ages, maternal educational level, residence location, annual income in the past year, maternal adverse pregnancy history, pre-gestational diabetes history, consanguineous marriages history, family history of congenital malformation, gestational diabetes in present delivery, gestational hypertension in present delivery, antibiotics use in 6 months before pregnancy, ovulation-promoting drugs use in 6 months before pregnancy, influenza exposure in the first trimester, alcohol consumption in 3 months before pregnancy, alcohol consumption in the first trimester, hair dyeing or perming in the periconceptional period and often use of cosmetics in the periconceptional period.
3.4. Gene–gene interaction with the risk of CHD
The interaction between the infant NKX2.5 gene, GATA4 gene and TBX5 gene polymorphisms and the association with CHD are shown in Table 2, selecting gene loci with statistically significant genotypes in multivariate analysis to explore their interaction effect on CHD, and selecting baseline characteristics and environmental factors related to CHD as the potential confounding factors. The interaction terms rs703752-rs6489956 (OR = 0.23, 95%CI: 0.08–0.71), rs4841588-rs6489956 (OR = 0.37, 95%CI: 0.18–0.75) and rs904018-rs6489956 (OR = 0.26, 95%CI: 0.13–0.53) were associated with a lower CHD risk. For the interaction terms rs118026695-rs4841588-rs11067101 (OR = 2.54, 95%CI: 1.01–6.43), rs118026695-rs12458-rs6489956 (OR = 0.13, 95%CI: 0.02–0.83), rs118026695-rs904018-rs11067101 (OR = 1.84, 95%CI: 1.01–3.39) and rs118026695-rs904018-rs6489956 (OR = 1.87, 95%CI: 1.06–3.28) were associated with CHD; nevertheless, this significance vanished from the multiple test corrections (FDR_P > 0.1).
Table 2.
Gene–gene interaction with the risk of CHD.
| Gene–gene interaction term | OR (95%CI)a,b | P | FDR_P | |||
|---|---|---|---|---|---|---|
| NKX2.5-GATA4 | ||||||
| rs118026695 | rs4841588 | 1.99 (0.98–4.03) | 0.057 | 0.180 | ||
| rs12458 | 0.73 (0.34–1.56) | 0.412 | 0.618 | |||
| rs904018 | 1.97 (0.97–3.99) | 0.060 | 0.180 | |||
| rs703752 | rs4841588 | 0.61 (0.25–1.49) | 0.277 | 0.554 | ||
| rs12458 | 0.86 (0.32–2.34) | 0.765 | 0.765 | |||
| rs904018 | 0.84 (0.35–2.05) | 0.707 | 0.765 | |||
| NKX2.5-TBX5 | ||||||
| rs118026695 | rs11067101 | 0.94 (0.82–1.07) | 0.359 | 0.479 | ||
| rs6489956 | 0.66 (0.28–1.55) | 0.335 | 0.479 | |||
| rs703752 | rs11067101 | 0.82 (0.28–2.42) | 0.723 | 0.723 | ||
| rs6489956 | 0.23 (0.08–0.71) | 0.010 | 0.040 | |||
| GATA4-TBX5 | ||||||
| rs4841588 | rs11067101 | 1.13 (0.59–2.14) | 0.712 | 0.766 | ||
| rs6489956 | 0.37 (0.18–0.75) | 0.006 | 0.018 | |||
| rs12458 | rs11067101 | 0.72 (0.34–1.51) | 0.382 | 0.671 | ||
| rs6489956 | 0.74 (0.33–1.62) | 0.447 | 0.671 | |||
| rs904018 | rs11067101 | 0.91 (0.48–1.71) | 0.766 | 0.766 | ||
| rs6489956 | 0.26 (0.13–0.53) | <0.001 | 0.006 | |||
| NKX2.5-GATA4-TBX5 | ||||||
| rs118026695 | rs4841588 | rs11067101 | 2.54 (1.01–6.43) | 0.049 | 0.150 | |
| rs118026695 | rs4841588 | rs6489956 | 1.92 (0.87–4.20) | 0.104 | 0.250 | |
| rs703752 | rs4841588 | rs11067101 | – | 0.999 | 0.999 | |
| rs703752 | rs4841588 | rs6489956 | 0.13 (0.01–1.97) | 0.141 | 0.282 | |
| rs118026695 | rs12458 | rs11067101 | 0.13 (0.01–5.58) | 0.286 | 0.490 | |
| rs118026695 | rs12458 | rs6489956 | 0.13 (0.02–0.83) | 0.031 | 0.150 | |
| rs703752 | rs12458 | rs11067101 | – | 0.999 | 0.999 | |
| rs703752 | rs12458 | rs6489956 | 0.48 (0.02–14.95) | 0.677 | 0.999 | |
| rs118026695 | rs904018 | rs11067101 | 1.84 (1.01–3.39) | 0.050 | 0.150 | |
| rs118026695 | rs904018 | rs6489956 | 1.87 (1.06–3.28) | 0.030 | 0.150 | |
| rs703752 | rs904018 | rs11067101 | – | 0.999 | 0.999 | |
| rs703752 | rs904018 | rs6489956 | 1.05 (0.07–15.50) | 0.973 | 0.999 | |
a Single-nucleotide polymorphisms were classified as wild type and variant genotypes, and using homozygous wild genotypes as a reference group.
b Adjusting with maternal pregnancy ages, maternal educational level, residence location, annual income in the past 1 year, maternal adverse pregnancy history, pre-gestational diabetes history, consanguineous marriages history, family history of congenital malformation, gestational diabetes in present delivery, gestational hypertension in present delivery, antibiotics use in 6 months before pregnancy, ovulation-promoting drugs use in 6 months before pregnancy, influenza exposure in the first trimester, alcohol consumption in 3 months before pregnancy, alcohol consumption in the first trimester, hair dyeing or perming in the periconceptional period and often use of cosmetics in the periconceptional period.
The high-dimensional interactions among infant NKX2.5, GATA4 and TBX5 gene polymorphisms and the association between CHD were investigated using the MB-MDR method, as shown in Table S6. The MB-MDR method identified five high-dimensional interaction models involving the three genes, all of which have interplayed the development of CHD.
3.5. Gene–environment interaction with the risk of CHD
The interaction between the gene and environmental factors and the association with CHD in offspring are shown in Table S7-S9. Genetic loci whose genotypes were related to CHD and environmental factors related to CHD were selected to explore their interactions, with baseline characteristics considered confounders. The results show that for the NKX2.5 gene, rs118026695-gestational diabetes (OR = 2.44, 95%CI: 1.60–3.72) were related to a higher risk of CHD. For the GATA4 gene, rs4841588-hair dyeing or perming in periconceptional period (OR = 4.03, 95%CI: 1.34–12.10) and rs904018-antibiotics use in 6 months before pregnancy (OR = 6.49, 95%CI: 1.70–24.83) were related to a higher CHD risk and rs4841588-pre-gestational diabetes history (OR = 0.08, 95%CI: 0.02–0.32) and rs12458-hair dyeing or perming in periconceptional period (OR = 0.05, 95%CI: 0.01–0.27) was associated with a lower CHD risk; however, the effect of the interaction term rs4841588-antibiotics use in 6 months before pregnancy, rs12458-adverse pregnancy history and rs904018-adverse pregnancy history lost statistical significance after FDR correction. For the TBX5 gene, rs6489956-adverse pregnancy history (OR = 2.44, 95%CI: 1.12–4.48) and rs6489956-influenza exposure in the first trimester (OR = 2.86, 95%CI: 1.24–6.60) were related to a higher CHD risk; in addition, rs11067101-pre-gestational diabetes history (OR = 0.05, 95%CI: 0.01–0.25), and rs6489956-alcohol consumption in the first trimester (OR = 0.09, 95%CI: 0.02–0.37) were related to a lower CHD risk.
The high-dimensional interactions among infant NKX2.5, GATA4 and TBX5 gene polymorphisms and maternal periconceptional environment factors associated with CHD were investigated using the MB-MDR method, as shown in Table S10. The MB-MDR method identified five high-dimensional interaction models involving the three genes, all of which have interplayed the development of CHD.
3.6. Sensitivity analysis
To further assess the potential confounding effect of strong genetic background to gene–environment interaction, we performed a sensitivity analysis by excluding 30 CHD cases with a family history of congenital malformations. After excluding participants with having family history of congenital malformation, the effect estimates of gene-environment interaction were generally consistent with the primary analyses, with similar association strength and direction for the main findings (Table S11-S14).
4. Discussion
Few studies have examined the gene–gene and gene–environment interactions between the infant cardiac transcription factors NKX2.5, GATA4 and TBX5 gene and maternal periconceptional environmental factors in relation to CHD, and the majority of studies concentrate on the comparison of genotype and allele frequencies between CHD patients and control groups. In this study, we further explored the gene–gene and gene–environmental factor interactions based on the analysis of polymorphisms at specific loci of the infant NKX2.5, GATA4 and TBX5 genes and the association of maternal environmental factors with CHD. Expanding from the simple association of single genes and single environmental factors to the exploration of multi-gene environmental interaction effects, we aim to explore the impact of infant transcription factor genes comprehensively, and maternal periconceptional environmental factors on the pathogenesis of CHD from multiple perspectives.
Our study found that rs118026695 (NKX2.5), rs12458 (GATA4), rs11067101 (TBX5) and rs6489956 (TBX5) are significantly associated with higher CHD risk. Moreover, rs703752 (NKX2.5), rs4841588 (GATA4) and rs904018 (GATA4) were associated with a lower risk of CHD. Previous studies have reported that the rs118026695 [30], rs12458 [31] and rs6489956 [21] are associated with a higher CHD risk, while the rs703752 [18,32] were associated with lower CHD risk, and our study reaffirms these findings. In addition, Yin et al. [19] and our study observed that the rs2277923 was not associated with CHD in the Chinese population, whereas this association was significant in other races [20,33]. A meta-analysis [18] also reported that rs2277923 was associated with CHD, besides the Chinese population, and that finding demonstrated the variability of the genetic background of CHD among different races. It is noteworthy that our study reported for the first time the association of the rs4841588 (GATA4), rs904018 (GATA4) and rs11067101 (TBX5) with CHD, which has not been reported in previous studies[31,34], and these findings provide new clues for further comprehensive exploration of the association between transcription factor gene polymorphisms and CHD.
Building on the aforementioned foundation, we used two statistical models to delve deeper into the association of the interaction between the three transcription factors with CHD. The multiplicative interaction model shows that the interaction terms rs703752 (NKX2.5)-rs6489956 (TBX5), rs4841588 (GATA4)-rs6489956 (TBX5) and rs904018 (GATA4)-rs6489956 (TBX5) were associated with a lower CHD risk, this is an intriguing finding, in our previous steps analysis, we found that rs4841588, rs904018 and rs703752 are the protective factors for CHD, rs6489956 is a risk factor for CHD. This seems to imply that gene–gene interaction may mitigate the effect of rs6489956 in precipitating CHD. Interaction between cardiac transcription factors influences cardiac development, TBX5 and GATA4 cooperatively interact on DNA throughout the genome to regulate heart development [14,35], and GATA4 and NKX2.5 are co-expressed in cardiac progenitors, synergistically activating several cardiac genes [36]. Heterozygous mutations in the TBX5 gene cause Holt-Oram syndrome and CHD [37,38], and reducing the TBX5 gene can impair the differentiation and function of cardiomyocytes [38,39]. Considering the coordinated interaction between TBX5 gene and other transcription factors, which collectively generate tissue- and context-specific gene expression [37,40,41], we speculate that there may be some mechanism between GATA4 and TBX5 genes that attenuate the cardiac structural variability induced by mutations in the TBX5 genes, which needs to further be explored in future studies. In addition, high-dimensional gene–gene interactions suggest that interactions between the cardiac transcription factor genes are associated with the development of CHD. Currently, research on the interaction between transcription factor genes is minimal, and our results will provide the epidemiologic clues for future research.
Findings from this study suggested that maternal periconceptional environmental factors were associated with higher CHD risk in offspring, consistent with previous studies [24,25,42,43]. Both interaction models indicated that there may be interactions between the maternal periconceptional environment factor and three cardiac transcription factors that could affect CHD. It is worth noting that several gene–environment interaction terms lost statistical significance after FDR correction. Therefore, the observed interaction patterns should be interpreted as preliminary evidence requiring replication in larger studies. Animal experimental studies [44–46] have shown that the expression of cardiac transcription factor genes is significantly down-regulated in embryos of animal models exposed to adverse peri-pregnancy environmental factors, suggesting that maternal exposure to adverse environmental factors may interfere with the expression of transcription factors during cardiac morphogenesis, increasing CHD risk. There are few existing studies related to the association of cardiac transcription factor gene–environment interactions with CHD, and our study provides some clues to understanding the aetiology of CHD pathogenesis. From the translational perspective, our findings underscore the importance of targeted periconceptional counseling. Specific preventive measures, such as: optimizing glycemic control for pre-gestational diabetes, cautious prescribing of antibiotics and advising against alcohol consumption and hair dyeing during the periconceptional period, could realistically mitigate CHD risk, particularly for genetically susceptible populations.
Given the influence of family history on the incidence of CHD, we excluded patients with a family history of congenital malformations for sensitivity analysis to verify the robustness of gene–environment interactions. Sensitivity analysis showed that the gene environment interaction effect remained stable after excluding patients with a family history. This indicates that the incidence of CHD was not explained solely by familial aggregation, and strongly suggests that maternal periconceptional environmental exposures and infant polymorphisms may jointly contribute to the development of non-syndromic CHD. However, it is important to note that our data set records family histories of congenital malformations, rather than specifically family histories of CHD. Therefore, although sensitivity analyses support the reliability of the main findings, they still need to be validated in a larger population.
There are some limitations. First, this was a single-center hospital-based case-control study, which may limit the generalizability of the findings. Second, because cases and controls were recruited from the same hospital setting, selection bias cannot be completely excluded. Third, some maternal exposure data were collected retrospectively by questionnaire, and recall bias was therefore unavoidable. Fourth, although we considered a broad range of maternal periconceptional exposures, residual or unmeasured confounding may still exist. Fifth, we did not include an independent replication cohort, and validation in larger multi-center populations is needed. Finally, the observed SNP associations may also be influenced by residual population stratification or linkage disequilibrium with nearby causal variants. Therefore, future research needs to verify the found association in a large multi-ethnic cohort. In addition, functional studies are still needed in the future to elucidate the potential biological mechanisms of CHD pathogenesis. For example, the synergy between GATA4 and TBX5 gene in the process of cardio-genesis could be verified by using multi-omics methods or animal models.
5. Conclusion
This is the first study to comprehensively investigate the association of infant cardiac transcription factor gene polymorphisms and maternal periconceptional environmental factors with CHD. This study found that NKX2.5, GATA4 and TBX5 gene polymorphisms were significantly associated with CHD risk. Gene–gene interaction analyses revealed that locus-specific interactions between genes may modulate the risk of CHD. In addition, adverse maternal environmental exposures during the periconceptional period significantly increased the risk of CHD in the offspring. Gene–environment interaction analyses suggest that exposure to specific genes and environments may affect the occurrence of CHD. However, concerning the limitations of our study, more convincing population-based studies and experimental research are necessary to verify the findings and further elucidate the potential mechanism comprehensively.
Supplementary Material
Acknowledgements
The authors thank all pediatricians and parents who participated in the study.
Funding Statement
This study was supported by National Natural Science Foundation Program of China (82473644, 82073653 and 81803313), Hunan Outstanding Youth Fund Project (2022JJ10087), National Key Research and Development Project (2018YFE0114500), China Postdoctoral Science Foundation (2020M682644), Hunan Provincial Science and Technology Talent Support Project (2020TJ-N07), Hunan Provincial Key Research and Development Program (2018SK2063), Open Project from NHC Key Laboratory of Birth Defect for Research and Prevention (KF2020006), Natural Science Foundation of Hunan Province (2018JJ2551), Natural Science Foundation of Hunan Province of China (grant numbers 2022JJ40207) and Changsha Municipal Natural Science Foundation (grant numbers kq2202470).
Ethics declarations
The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee for Clinical Research of Xiangya School of Public Health of Central South University (no. XYGW-2018-07).
Informed consent statement
Informed consent was obtained from all subjects involved in the study.
Disclosure statement
The authors declare no conflict of interest.
Data availability statement
The data presented in this study are available on request from the corresponding author.
References
- 1.Zhao L, Chen L, Yang T, et al. Birth prevalence of congenital heart disease in China, 1980-2019: a systematic review and meta-analysis of 617 studies. Eur J Epidemiol. 2020;35(7):631–642. doi: 10.1007/s10654-020-00653-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bai Z, Han J, An J, et al. The global, regional, and national patterns of change in the burden of congenital birth defects, 1990-2021: an analysis of the global burden of disease study 2021 and forecast to 2040. EClinicalMedicine. 2024;77:102873. doi: 10.1016/j.eclinm.2024.102873. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kang L, Cao G, Jing W, et al. Global, regional, and national incidence and mortality of congenital birth defects from 1990 to 2019. Eur J Pediatr. 2023;182(4):1781–1792. doi: 10.1007/s00431-023-04865-w. [DOI] [PubMed] [Google Scholar]
- 4.Roth GA, Mensah GA, Johnson CO, et al. Global Burden of Cardiovascular Diseases and Risk Factors, 1990-2019: update From the GBD 2019 Study. J Am Coll Cardiol. 2020;76(25):2982–3021. doi: 10.1016/j.jacc.2020.11.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Nie X, Liu X, Wang C, et al. Assessment of evidence on reported non-genetic risk factors of congenital heart defects: the updated umbrella review. BMC Pregnancy Childbirth. 2022;22(1):371. doi: 10.1186/s12884-022-04600-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.van der Bom T, Zomer AC, Zwinderman AH, et al. The changing epidemiology of congenital heart disease. Nat Rev Cardiol. 2011;8(1):50–60. doi: 10.1038/nrcardio.2010.166. [DOI] [PubMed] [Google Scholar]
- 7.Sadowski SL. Congenital cardiac disease in the newborn infant: past, present, and future. Crit Care Nurs Clin North Am. 2009;21(1):37–48, vi. doi: 10.1016/j.ccell.2008.10.001. [DOI] [PubMed] [Google Scholar]
- 8.Zubrzycki M, Schramm R, Costard-Jackle A, et al. Cardiac development and factors influencing the development of congenital heart defects (CHDs): part I. Int J Mol Sci. 2024;25(13):7117. doi: 10.3390/ijms25137117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Bruneau BG. Signaling and transcriptional networks in heart development and regeneration. Cold Spring Harb Perspect Biol. 2013;5(3):a008292–a008292. doi: 10.1101/cshperspect.a008292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Olson EN. Gene regulatory networks in the evolution and development of the heart. Science (1979). 2006;313(5795):1922–1927. doi: 10.1126/science.1132292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Srivastava D. Making or breaking the heart: from lineage determination to morphogenesis. Cell. 2006;126(6):1037–1048. doi: 10.1016/j.cell.2006.09.003. [DOI] [PubMed] [Google Scholar]
- 12.Bruneau BG. The developmental genetics of congenital heart disease. Nature. 2008;451(7181):943–948. doi: 10.1038/nature06801. [DOI] [PubMed] [Google Scholar]
- 13.Williams K, Carson J, Lo C.. Genetics of congenital heart disease. Biomolecules. 2019;9(12):879. doi: 10.3390/biom9120879. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Luna-Zurita L, Stirnimann CU, Glatt S, et al. Complex interdependence regulates heterotypic transcription factor distribution and coordinates cardiogenesis. Cell. 2016;164(5):999–1014. doi: 10.1016/j.cell.2016.01.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.McCulley DJ, Black BL.. Chapter nine - Transcription factor pathways and congenital heart disease. In: Bruneau BG, editor. Current topics in developmental biology. Vol 100: Waltham, MA, USA: Academic Press; 2012. p. 253–277. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Mubeen H, Farooq M, Rehman AU, et al. Gene expression and transcriptional regulation driven by transcription factors involved in congenital heart defects. Ir J Med Sci. 2023;192(2):595–604. doi: 10.1007/s11845-022-02974-3. [DOI] [PubMed] [Google Scholar]
- 17.Ashiq S, Ashiq K, Sabar MF.. The role of NKX2-5 gene polymorphisms in congenital heart disease (CHD): a systematic review and meta-analysis. Egypt Heart J. 2021;73(1):72. doi: 10.1186/s43044-021-00199-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.González-Castro TB, Tovilla-Zárate CA, López-Narvaez ML, et al. Association between congenital heart disease and NKX2.5 gene polymorphisms: systematic review and meta-analysis. Biomark Med. 2020;14(18):1747–1757. doi: 10.2217/bmm-2020-0190. [DOI] [PubMed] [Google Scholar]
- 19.Yin J, Qian J, Dai G, et al. Search of somatic mutations of NKX2-5 and GATA4 genes in Chinese patients with sporadic congenital heart disease. Pediatr Cardiol. 2019;40(1):17–22. doi: 10.1007/s00246-018-1955-z. [DOI] [PubMed] [Google Scholar]
- 20.Behiry EG, Al-Azzouny MA, Sabry D, et al. Association of NKX2-5, GATA4, and TBX5 polymorphisms with congenital heart disease in Egyptian children. Mol Genet Genomic Med. 2019;7(5):e612. doi: 10.1002/mgg3.612. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Wang F, Liu D, Zhang RR, et al. A TBX5 3′UTR variant increases the risk of congenital heart disease in the Han Chinese population. Cell Discov. 2017;3(1):17026. doi: 10.1038/celldisc.2017.26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Limperopoulos C, Wessel DL, Du Plessis AJ.. Understanding the maternal-fetal environment and the birth of prenatal pediatrics. J Am Heart Assoc. 2022;11(2):e023807. doi: 10.1161/JAHA.121.023807. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Huhta J, Linask KK.. Environmental origins of congenital heart disease: the heart-placenta connection. Semin Fetal Neonatal Med. 2013;18(5):245–250. doi: 10.1016/j.siny.2013.05.003. [DOI] [PubMed] [Google Scholar]
- 24.Zhang TN, Wu QJ, Liu YS, et al. Environmental risk factors and congenital heart disease: an umbrella review of 165 systematic reviews and meta-analyses with more than 120 million participants. Front Cardiovasc Med. 2021;8:640729. doi: 10.3389/fcvm.2021.640729. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Yang L, Liu X, Chen Y, et al. An update on the CHDGKB for the systematic understanding of risk factors associated with non-syndromic congenital heart disease. Comput Struct Biotechnol J. 2021;19:5741–5751. doi: 10.1016/j.csbj.2021.10.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Gomersall JC, Moore VM, Fernandez RC, et al. Maternal modifiable factors and risk of congenital heart defects: systematic review and causality assessment. BMJ Open. 2024;14(8):e082961. doi: 10.1136/bmjopen-2023-082961. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kalisch-Smith JI, Ved N, Sparrow DB.. Environmental risk factors for congenital heart disease. Cold Spring Harb Perspect Biol. 2020;12(3):a037234. doi: 10.1101/cshperspect.a037234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Zheng G, Ng HK.. Genetic model selection in two-phase analysis for case-control association studies. Biostatistics. 2008;9(3):391–399. doi: 10.1093/biostatistics/kxm039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Calle ML, Urrea V, Malats N, et al. mbmdr: an R package for exploring gene-gene interactions associated with binary or quantitative traits. Bioinformatics. 2010;26(17):2198–2199. doi: 10.1093/bioinformatics/btq352. [DOI] [PubMed] [Google Scholar]
- 30.Pang S, Shan J, Qiao Y, et al. Genetic and functional analysis of the NKX2-5 gene promoter in patients with ventricular septal defects. Pediatr Cardiol. 2012;33(8):1355–1361. doi: 10.1007/s00246-012-0346-0. [DOI] [PubMed] [Google Scholar]
- 31.Reamon-Buettner SM, Cho SH, Borlak J.. Mutations in the 3′-untranslated region of GATA4 as molecular hotspots for congenital heart disease (CHD). BMC Med Genet. 2007;8(1):38. doi: 10.1186/1471-2350-8-38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Cao Y, Wang J, Wei C, et al. Genetic variations of NKX2-5 in sporadic atrial septal defect and ventricular septal defect in Chinese Yunnan population. Gene. 2016;575(1):29–33. doi: 10.1016/j.gene.2015.08.033. [DOI] [PubMed] [Google Scholar]
- 33.Ketharnathan S, Koshy T, Sethuratnam R, et al. Investigation of NKX2.5 gene mutations in congenital heart defects in an Indian population. Genet Test Mol Biomarkers. 2015;19(10):579–583. doi: 10.1089/gtmb.2015.0112. [DOI] [PubMed] [Google Scholar]
- 34.Su W, Zhu P, Wang R, et al. Congenital heart diseases and their association with the variant distribution features on susceptibility genes. Clin Genet. 2017;91(3):349–354. doi: 10.1111/cge.12835. [DOI] [PubMed] [Google Scholar]
- 35.Ang YS, Rivas RN, Ribeiro AJS, et al. Disease model of GATA4 mutation reveals transcription factor cooperativity in human cardiogenesis. Cell. 2016;167(7):1734–1749 e22. doi: 10.1016/j.cell.2016.11.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Kinnunen S, Välimäki M, Tölli M, et al. Nuclear receptor-like structure and interaction of congenital heart disease-associated factors GATA4 and NKX2-5. PLoS One. 2015;10(12):e0144145. doi: 10.1371/journal.pone.0144145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Steimle JD, Moskowitz IP.. TBX5: a key regulator of heart development. Curr Top Dev Biol. 2017;122:195–221. doi: 10.1016/bs.ctdb.2016.08.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Kathiriya IS, Rao KS, Iacono G, et al. Modeling human TBX5 haploinsufficiency predicts regulatory networks for congenital heart disease. Dev Cell. 2021;56(3):292–309 e9. doi: 10.1016/j.devcel.2020.11.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Mori AD, Zhu Y, Vahora I, et al. Tbx5-dependent rheostatic control of cardiac gene expression and morphogenesis. Dev Biol. 2006;297(2):566–586. doi: 10.1016/j.ydbio.2006.05.023. [DOI] [PubMed] [Google Scholar]
- 40.Maitra M, Schluterman MK, Nichols HA, et al. Interaction of Gata4 and Gata6 with Tbx5 is critical for normal cardiac development. Dev Biol. 2009;326(2):368–377. doi: 10.1016/j.ydbio.2008.11.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Rao KS, Kameswaran V, Bruneau BG.. Modeling congenital heart disease: lessons from mice, hPSC-based models, and organoids. Genes Dev. 2022;36(11-12):652–663. doi: 10.1101/gad.349678.122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Xiao D, Li W, Zhang WH, et al. Maternal periconceptional environmental exposure and offspring with congenital heart disease: a case-control study in Guangzhou, China. BMC Pregnancy Childbirth. 2023;23(1):57. doi: 10.1186/s12884-023-05355-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Saad H, Sinclair M, Bunting B.. Maternal sociodemographic characteristics, early pregnancy behaviours, and livebirth outcomes as congenital heart defects risk factors - Northern Ireland 2010-2014. BMC Pregnancy Childbirth. 2021;21(1):759. doi: 10.1186/s12884-021-04223-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Liu X, Gao Q, Feng Z, et al. Protective effects of spermidine and melatonin on deltamethrin-induced cardiotoxicity and neurotoxicity in zebrafish. Cardiovasc Toxicol. 2021;21(1):29–41. doi: 10.1007/s12012-020-09591-5. [DOI] [PubMed] [Google Scholar]
- 45.Moumne O, Chowdhurry R, Doll C, et al. Mechanism sharing between genetic and gestational hypoxia-induced cardiac anomalies. Front Cardiovasc Med. 2018;5:100. doi: 10.3389/fcvm.2018.00100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Doll CF, Pereira NJ, Hashimi MS, et al. Gestational intermittent hyperoxia rescues murine genetic congenital heart disease in part. Sci Rep. 2021;11(1):6608. doi: 10.1038/s41598-021-85569-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data presented in this study are available on request from the corresponding author.