Association of MTHFD1 G1958A (rs2236225) gene polymorphism with the risk of congenital heart disease: a systematic review and meta-analysis

Abstract

Background

We did this study to better clarify the correlations of methylenetetrahydrofolate dehydrogenase 1 (MTHFD1)-G1958A (rs2236225) gene polymorphism with the risk of congenital heart diseases (CHD) and its subgroups.

Methods

Relevant articles were searched in PubMed, Web of Science, Cochrane Library, Embase, CNKI, VIP database and Wanfang DATA until October 2023. We will use odds ratios (ORs) and 95% confidence intervals (CIs) to examine the potential associations of MTHFD1- G1958A gene polymorphism with CHD and its subgroups.

Results

We included a total of 9 eligible studies, encompassing 1917 children with CHD, 1863 healthy children, 1717 mothers of the children with CHD and 1666 mothers of healthy children. In our study, the meta-analysis of fetal group revealed no significant association between any of the five genetic models for the MTHFD1-G1958A polymorphism and the risk of CHD. Subgroup analysis showed that associations between the MTHFD1-G1958A polymorphism and Tetralogy of Fallot (TOF) risk in the homozygote model (AA vs. GG, OR = 2.82, 95%CI [1.16, 6.86], P = 0.02) and recessive model (AA vs. GG + GA, OR = 3.09, 95%CI [1.36, 7.03], P = 0.007). In addition, the MTHFD1-G1958A polymorphism was associated with the risk of CHD in racial subgroup, increasing the risk of CHD in Caucasians. In maternal analysis, 2 genetic models of MTHFD1-G1958A polymorphism increased the risk of CHD: the heterozygote model (GA vs. GG, OR = 1.22, 95%CI [1.04, 1.42], P = 0.01), and the dominance model (GA + AA vs. GG, OR = 1.17, 95%CI [1.01, 1.34], P = 0.03).

Conclusions

The fetal MTHFD1-G1958A (rs2236225) gene polymorphism increase their risk of TOF. The maternal MTHFD1-G1958A polymorphism has a strong correlation with the risk of CHD, and there are racial differences in this correlation. Compared with GG genotype, the GA genotype increases the risk of CHD.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12920-024-02052-w.

Background

Congenital heart disease (CHD) refers to a group of disease in which the structure and function of the heart or major blood vessels are abnormal during fetal development, resulting in blood circulation disorders. The prevalence rate is 1–1.2%, making it the most common birth defect in children and a main cause of infant mortality worldwide [1, 2]. Depending on the presence or absence of cyanosis, CHD can be divided into cyanotic heart disease and non-cyanotic heart disease [3]. Cyanotic heart disease is characterized by right-to-left shunt of blood in the heart, resulting from structural abnormalities in both the cardiac and large vascular systems. This category includes conditions such as Tetralogy of Fallot (TOF) and Transposition of Great Artery (TGA), which are classified under Conotruncal defects (CTD). In contrast, non-cyanotic heart disease exhibit a more complex and diverse array of phenotypes, including ventricular septal defect (VSD), atrial septal defect (ASD) and patent ductus arteriosus (PDA). As the most prevalent birth defect, approximately 1/4 of CHD cases necessitate surgical treatment in the first year of life, imposing significant financial burdens on both society and affected families [4]. Research indicates that genetic factors play a dominant role in the pathogenesis of CHD, and its etiology and molecular mechanism are still unclear [5]. Maternal environmental factors, such as folic acid deficiency, maternal malnutrition and unhealthy lifestyle choices, are also involved in the occurrence of CHD [6].

CHD is a birth defect that is sensitive to folic acid levels. The risk of CHD subtypes, including CTD, aortic stenosis and septal defect, is reduced in the offspring of women who received folic acid supplementation during early pregnancy [7–9]. Folic acid plays a momentous role in the development of the cardiovascular system during embryogenesis [10]. Disruption of folic acid metabolism can lead to the blockage of methionine cycle. It not only affects the methylation reaction to affect cellular metabolism and growth, but also causes the metabolic disorder of blood homocysteine (Hcy), resulting in elevated Hcy levels. Elevated Hcy is an independent risk factor for cardiovascular disease, which can adversely affect early cardiovascular growth and development [11, 12]. Therefore, single nucleotide polymorphisms (SNPs) in genes related to Hcy/folate metabolism pathway play a vital role in the etiology of congenital defects, including CHD. Methylenetetrahydrofolate dehydrogenase 1 (MTHFD1) is a key regulatory enzyme in folate-methionine metabolism. It provides single carbon unit in the metabolic pathway of methionine generation, indirectly provides methyl donors for DNA and protein methylation, maintains the methylation cycle from Hcy to methionine, and keeps Hcy in the blood at a low level [9]. The abnormality of MTHFD1 leads to the disorder of Hcy metabolic pathway. High level of Hcy is a teratogenic factor, which is more likely to lead to CHD [13, 14]. The research indicated that abnormal metabolism of MTHFD1 is a risk factor for vascular diseases [15]. The lack of MTHFD1 during embryonic development may potentially impact the occurrence of CHD.

The MTHFD1 gene is located on 14q24 and encodes a trifunctional enzyme, including 5, 10-methylenetetrahydrofolate dehydrogenase, 5, 10-methylenetetrahydrofolate cyclase and 10-methyltetrahydrofolate synthase. There are multiple SNPs in the coding region of the entire MTHFD1 gene, and G1958A is the most studied site in this gene, which may be related to a variety of diseases (Fig. 1). G1958A is a missense mutation located in the promoter region of the MTHFD1 gene. There is a G > A change at position 1958 of the nucleotide, causing a 653 amino acid mutation, resulting in the conversion of arginine to glutamine at this site [16]. The Arg653Gln enzyme had normal substrate affinity after purification. The substitution of Arg653Gln diminishes the stability of the formyltetrahydrofolate synthase domain and the metabolic activity of MTHFD1, while not affecting the enzyme activity of MTHFD1 [17]. A study using a mouse model has found that the deficiency of MTHFD1 synthase is associated with a high incidence of CHD, specifically VSD [18].

Fig. 1.

The location of G1958A (rs2236225) polymorphism in MTHFD1 gene (SNP2). The genetic variation is named according to the DNA sequence of the human MTHFD1 gene (Genbank database: NG_011992). The transcription start site is located at 1350 in the promoter. Figure 1 is drawn by Illustrator for Biological Sequences (IBS 2.0). Abbreviations: transcriptional start sites (TSS); coding DNA sequences (CDS); single nucleotide polymorphism (SNP)

Hol FA et al. first elucidated the polymorphism of G1958A gene in MTHFD1 gene in 1998 [16], and then, the related research is increasing. Some studies have confirmed that MTHFD1-G1958A gene polymorphism is associated with neural tube defects (NTD), premature delivery, lung cancer, colorectal cancer, ovarian cancer, type II diabetes, Down ‘s syndrome, attention deficit hyperactivity disorder(ADHD) and other diseases [19–24]. Cheng et al. found that the MTHFD1-G1958A mutation in the parental generation (especially the mother) had a protective effect on the offspring and could reduce the risk of ASD. The study of Song and Chen et al. did not find the association between maternal MTHFD1-G1958A and the risk of CHD in offspring, which is inconsistent with the conclusion of Cheng et al. [25, 26].

Recent studies have found that MTHFD1-G1958A mutation in offspring can increase the risk of CHD [9, 27, 28]. Some studies have also shown that this locus is not related to the risk of CHD in offspring [29–34]. Currently, the majority of relevant studies tend to focus on the relationship between MTHFD1- G1958A gene polymorphism and the risk of CHD in children, while there is limited epidemiological evidence regarding the association between maternal MTHFD1-G1958A gene polymorphism and the risk of CHD in offspring.

In order to clarify these relationships, we conducted a comprehensive and systematic review and meta-analysis from multiple perspectives, and analyzed the correlation between MTHFD1-G1958A gene polymorphism in both children and their mothers and the risk of CHD and CHD subgroups in offspring, so as to provide better directions for clinical research.

Materials and methods

Based on the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) 2020 checklist, we organized the meta-analysis. The details were set out in the Supplementary Material 1. Ethical approval was not necessary for the type of the study.

Literature search

We systematically searched 7 databases, including PubMed, Web of Science, EMBASE, the Cochrane Library, China National Knowledge Infrastructure, Wan Fang, and VIP and searched all relevant articles before October 10, 2023. All studies were manually searched and selected. In addition, the reference list of the original article was manually reviewed, and data that could not be retrieved were obtained by contacting the corresponding author to identify other potentially eligible studies. The complete literature selection flow chart is shown in Fig. 2. The complete detailed search strategy in Web of Science is listed in Supplementary Material  2.

Fig. 2.

Flow chart of study selection for the present study

Inclusion and exclusion criteria

Inclusion criteria: (1) All case-control or cohort studies investigated the association between MTHFD1-G1958A polymorphisms and CHD; (2) The case group included at least CHD patients and / or CHD patients ' mothers; (3) The control group was non-CHD patients or healthy people; (4) Population study of MTHFD1-G1958A genotype frequency in case group and control group.

Exclusion criteria: (1) the genotype of the case group was used and published in more than one study; (2) The case group included other congenital cardiovascular diseases ; (3) Research based on incomplete raw data or unreported available data.

Data extraction and risk of bias

The following data were independently extracted according to inclusion and exclusion criteria: surname of first author, publication year, study country and region of study, type of CHD, source of control population, sample size of case and control, genotyping method, genotype frequencies of MTHFD1-G1958A gene polymorphisms in case and control, and HWE test results (Table 1).

Table 1.

Characteristics of studies included

First author	Year	Country	Region	Race	Genotyping method	Type of disease	Control source	PHWE
Cheng	2005	China	East Asian	Chinese Han	PCR-RFLP	ASD VSD PDA TOF	HB	0.804
Xu	2010	China	East Asian	Chinese Han	proteinase K digestion	PDA SD LVOTO	HB	0.199
Wang	2013	China	East Asian	Chinese Han	SNaPShot Sanger	CHD	HB	0.206
Christensen	2008	Canada	North America	N. European	proteinase K digestion	SD CTD VD	HB	0.239
Kuzelicki	2022	Slovenia	South-central Europe	Caucasian	TaqMan probes	SD CTD LVOTO	HB	NA
Khatami	2017	Iran	Southwest Asian	Iranian	PCR-RFLP	ASD VSD TOF	HB	0.137
Song	2022	China	East Asian	Chinese Han	MassARRAY	ASD VSD TOF CTD	HB	0.620
Liu	2023	China	East Asian	Chinese Han	MassARRAY	ASD VSD PDA	HB	0.080
Chen	2022	China	East Asian	Chinese Han	MassARRAY	CHD	HB	0.770

CHD – congenital heart disease; HWE – Hardy Weinberge equilibrium; TGA – transposition of the great arteries; ASD-Atrial Septal Defect; VSD- Ventricular Septal Defect; SD- Septal Defect; PDA- Patent Ductus Arteriosus; TOF – tetralogy of fallot; LVOTO- left ventricular outflow tract obstruction. CTD – conotruncal heart defects; PCR-RFLP- polymerase chain reaction-restriction fragment length polymorphism; PB – population-based; HB – hospital-based; NA – notavailable; PHWE: P value of HWE of the control group, PHWE > 0.05 means that the control group conforms to Hardy–Weinberg equilibrium, and the research is meaningful

We assessed the risk of bias in the included literature using the Newcastle-Ottawa scale scoring standard. The scoring system evaluated the included studies from 3 aspects, including case-to-control selectivity, case-to-control comparability and exposure of the risk factors. Scores ranged from 0 to 9, with scores ≥ 7 classified as low risk of bias studies. Table 2 is the quality evaluation results of the Newcastle-Ottawa scale included in the 9 studies.

Table 2.

Results of Newcastle-Ottawa scale quality evaluation included in the study

Our two researchers independently completed research selection, data extraction, bias risk assessment and other work, and discussed the differences encountered together to find solutions.

(1) The case definition is adequate with independent validation; (2) Consecutive or obviously representative series of cases; (3) Community controls; (4) Controls with no history of disease (endpoint); (5) Cases and controls with comparable ages and comparability on any other factors; (6) Ascertainment of exposure using secure records (e.g.: surgical records) or structured interviews with blinding to case/control; (7) Ascertainment of exposure using the same method for cases and controls; (8) Ascertainment of exposure with non-response rate for both groups.

Statistical analysis

All data analysis was performed by RevMan5.3 software. In the five genetic models of heterozygous model (GA vs. GG), homozygous model (AA vs. GG), dominant model (GA + AA vs. GG), recessive model (AA vs. GA + GG) and allele model (A vs. G), I² test and P value were used to analyze the statistical heterogeneity of the included literature. When P ≥ 0.1 and I² ≤ 50%, we believe that the heterogeneity was acceptable and the fixed effect model can be used for merger analysis. If P < 0.1 or I² > 50%, the heterogeneity is considered to be large, and the random effect model is used for analysis, and the sensitivity analysis is carried out by excluding the literature one by one. OR value and 95% confidence interval (95% CI) were used to describe the effect size. When P < 0.05, we considered that the difference was statistically significant.

Results

Characteristics of included studies

The literature search identified 91studies, and 74 remained after removing duplicates. We removed 61 irrelevant articles by reading the title and abstract. We read the full text of the remaining 13 articles in detail. Among them, 4 articles from the same team or research on MTHFD rs2236225C > T were excluded. As a result, a total of 9 studies. Among them, 4 studies included fetal cases [27, 29–31], 2 studies included maternal cases [25, 26], and 3 studies included both fetal cases and maternal cases [9, 28, 35]. After summarizing the data, our study contained 1917 fetal cases, 1863 fetal controls, 1718 maternal cases, and 1666 maternal controls. It is worth noting that the research data of Khatami, Kuzelicki et al. are incomplete and can only analyze the dominant model [27, 28]. The genotype characteristics of the included studies are summarized in the following table (Tables 3, 4 and 5).

Table 3.

Genotype characteristics of included studies

First author	Case				GA+AA	Control				GA+AA	Allele frequencies cases		Allele frequencies controls
	Total	GG	GA	AA		Total	GG	GA	AA		G	A	G	A
Fetal group
Cheng	179	108	64	7	71	122	76	40	6	46	0.782	0.218	0.787	0.213
Xu	502	291	195	16	211	527	289	210	28	238	0.774	0.226	0.748	0.252
Wang	160	94	53	13	66	188	109	64	15	79	0.753	0.247	0.75	0.25
Christensen	158	50	78	30	108	110	43	56	11	67	0.563	0.437	0.645	0.355
Kuzelicki	196	74	–	–	122	198	54	–	–	144	–	–	–	–
Khatami	102	57	32	13	45	98	73	21	4	25	0.716	0.284	0.852	0.148
Liu	620	393	201	26	227	620	403	185	32	217	0.796	0.204	0.799	0.201
Maternal group
Cheng	177	96	73	8	81	119	62	47	10	57	0.749	0.251	0.718	0.281
Christensen	199	59	106	34	140	105	34	52	19	71	0.563	0.437	0.571	0.429
Kuzelicki	195	67	–	–	128	198	68	–	–	130	–	–	–	–
Song	464	286	158	20	178	504	332	152	20	172	0.787	0.213	0.81	0.19
Chen	683	420	238	25	263	740	495	219	26	245	0.789	0.211	0.817	0.183

Table 4.

Genotype characteristics of included studies.(children with CHD and its subgroups)

First author	Total	GG	GA	AA	GA+AA	Total	GG	GA	AA	GA+AA	Allele frequencies cases		Allele frequencies controls
											G	A	G	A
ASD
Cheng	22	17	5	0	5	122	76	40	6	46	0.886	0.114	0.787	0.213
Xu	41	21	17	3	23	527	289	210	28	238	0.72	0.28	0.748	0.252
Christensen	11	2	6	3	9	110	43	56	11	67	0.455	0.545	0.645	0.355
Khatami	24	15	–	–	9	98	73	–	–	25	–	–	–	–
VSD
Cheng	88	52	32	4	36	122	76	40	6	46	0.773	0.227	0.787	0.213
Xu	257	145	103	9	112	527	289	210	28	238	0.765	0.235	0.748	0.252
Christensen	43	13	25	5	30	110	43	56	11	67	0.593	0.407	0.645	0.355
Khatami	63	33	–	–	30	98	73	–	–	25	–	–	–	–
SD
Cheng	110	69	37	4	41	122	76	40	6	46	0.795	0.205	0.787	0.213
Xu	298	166	120	12	132	527	289	210	28	238	0.758	0.242	0.748	0.252
Christensen	63	19	36	8	44	110	43	56	11	67	0.587	0.413	0.645	0.354
Kuzelicki	90	31	–	–	59	198	54	–	–	144	–	–	–	–
Khatami	87	48	–	–	39	98	73	–	–	25	–	–	–	–
TOF
Cheng	22	16	4	2	6	122	76	40	6	46	0.818	0.182	0.787	0.213
Christensen	35	11	14	10	24	110	43	56	11	67	0.514	0.486	0.645	0.355
Khatami	15	9	–	–	6	98	73	–	–	25	–	–	–	–
yellow race
Cheng	179	108	64	7	71	122	76	40	6	46	0.782	0.218	0.787	0.213
Xu	502	291	195	16	211	527	289	210	28	238	0.774	0.226	0.748	0.252
Wang	160	94	53	13	66	188	109	64	15	79	0.753	0.247	0.75	0.25
Liu	620	393	201	26	227	620	403	185	32	217	0.8	0.2	0.8	0.2
white race
Christensen	158	50	78	30	108	110	43	56	11	67	0.563	0.437	0.645	0.355
Kuzelicki	196	74	–	–	122	198	54	–	–	144	–	–	–	–
Khatami	102	57	32	13	45	98	73	21	4	25	0.716	0.284	0.852	0.148
Asia
Cheng	179	108	64	7	71	122	76	40	6	46	0.782	0.218	0.787	0.213
Xu	502	291	195	16	211	527	289	210	28	238	0.774	0.226	0.748	0.252
Wang	160	94	53	13	66	188	109	64	15	79	0.753	0.247	0.75	0.25
Khatami	102	57	32	13	45	98	73	21	4	25	0.716	0.284	0.852	0.148
Liu	620	393	201	26	227	620	403	185	32	217	0.796	0.204	0.799	0.201

Table 5.

Genotype characteristics of included studies.(mothers of children with CHD and their subgroups)

First author	Total	GG	GA	AA	GA+AA	Total	GG	GA	AA	GA+AA	Allele frequencies cases		Allele frequencies controls
											G	A	G	A
SD
Cheng	106	59	43	4	47	119	62	47	10	57	0.759	0.241	0.718	0.282
Christensen	75	26	40	9	49	105	34	52	19	71	0.613	0.387	0.571	0.429
Kuzelicki	89	30	–	–	59	198	68	–	–	130	–	–	–	–
yellow race
Cheng	177	96	73	8	81	119	62	47	10	57	0.749	0.251	0.718	0.282
Song	464	286	158	20	178	504	332	152	20	172	0.787	0.213	0.81	0.19
Chen	683	420	238	25	263	740	495	219	26	245	0.789	0.211	0.817	0.183
white race
Christensen	199	59	106	34	140	105	34	52	19	71	0.563	0.437	0.571	0.429
Kuzelicki	195	67	–	–	128	198	68	–	–	130	–	–	–	–

Association of MTHFD1-G1958A gene polymorphism and its subgroups with CHD in offspring (Table 6)

Table 6.

Meta-analysis of fetal MTHFD1 gene G1958A polymorphism and its risk of congenital heart disease

Type	OR(95%CI)	z	P	Test of heterogeneity		Analysis model
				I2	P *
Overall(7)
GA VS GG	1.07[0.92,1.24]	0.89	0.37	3%	0.40	Fixed-effects model
AA VS GG	1.15[0.67,1.96]	0.51	0.61	63%	0.02	Random-effects model
GA+AA VS GG	1.05[0.83,1.32]	0.38	0.70	61%	0.02	Random-effects model
AA VS GG+GA	1.09[0.67,1.78]	0.36	0.72	58%	0.04	Random-effects model
A VS G	1.13[0.91,1.41]	1.08	0.28	68%	0.008	Random-effects model
ASD(4)
GA VS GG	1.01[0.60,1.70]	0.03	0.97	10%	0.33	Fixed-effects model
AA VS GG	1.58[0.64,3.95]	0.99	0.32	31%	0.23	Fixed-effects model
GA+AA VS GG	1.34[0.86,2.07]	1.30	0.19	39%	0.18	Fixed-effects model
AA VS GG+GA	1.52[0.64,3.63]	0.95	0.34	0%	0.37	Fixed-effects model
A VS G	1.10[0.53,2.26]	0.25	0.80	62%	0.07	Random-effects model
VSD(4)
GA VS GG	1.06[0.82,1.37]	0.44	0.66	0%	0.59	Fixed-effects model
AA VS GG	0.83[0.47,1.49]	0.62	0.54	0%	0.50	Fixed-effects model
GA+AA VS GG	1.34[0.86,2.10]	1.29	0.20	63%	0.04	Random-effects model
AA VS GG+GA	0.80[0.45,1.40]	0.78	0.44	0%	0.67	Fixed-effects model
A VS G	0.99[0.81,1.20]	0.15	0.88	0%	0.52	Fixed-effects model
SD(5)
GA VS GG	1.04[0.82,1.33]	0.33	0.74	0%	0.59	Fixed-effects model
AA VS GG	0.90[0.53,1.53]	0.38	0.70	0%	0.45	Fixed-effects model
GA+AA VS GG	1.14[0.80,1.63]	0.71	0.48	60%	0.04	Random-effects model
AA VS GG+GA	0.87[0.52,1.45]	0.54	0.59	0%	0.63	Fixed-effects model
A VS G	1.00[0.83,1.20]	0.04	0.97	0%	0.49	Fixed-effects model
TOF(3)
GA VS GG	0.73[0.37,1.46]	0.89	0.37	0%	0.33	Fixed-effects model
AA VS GG	2.82[1.16,6.86]	2.28	0.02	0%	0.43	Fixed-effects model
GA+AA VS GG	1.16[0.68,1.99]	0.55	0.59	20%	0.29	Fixed-effects model
AA VS GG+GA	3.09[1.36,7.03]	2.69	0.007	0%	0.53	Fixed-effects model
A VS G	1.27[0.62,2.59]	0.66	0.51	54%	0.14	Random-effects model
yellow race(4)
GA VS GG	1.02[0.87,1.19]	0.26	0.79	0%	0.72	Fixed-effects model
AA VS GG	0.77[0.54,1.08]	1.51	0.13	0%	0.70	Fixed-effects model
GA+AA VS GG	0.99[0.85,1.15]	0.17	0.86	0%	0.69	Fixed-effects model
AA VS GG+GA	0.76[0.54,1.07]	1.56	0.12	0%	0.74	Fixed-effects model
A VS G	0.96[0.84,1.08]	0.72	0.47	0%	0.68	Fixed-effects model
white race(3)
GA VS GG	1.46[0.97,2.20]	1.80	0.07	23%	0.26	Fixed-effects model
AA VS GG	2.84[1.47,5.49]	3.11	0.002	0%	0.43	Fixed-effects model
GA+AA VS GG	1.08[0.82,1.43]	0.55	0.59	85%	0.001	Random-effects model
AA VS GG+GA	2.44[1.31,4.55]	2.82	0.005	0%	0.49	Fixed-effects model
A VS G	1.74[1.09,2.78]	2.31	0.02	58%	0.12	Random-effects model
Asia(5)
GA VS GG	1.06[0.91,1.23]	0.74	0.46	19%	0.29	Fixed-effects model
AA VS GG	0.97[0.57,1.65]	0.10	0.92	54%	0.07	Random-effects model
GA+AA VS GG	1.09[0.86,1.38]	0.71	0.48	54%	0.07	Random-effects model
AA VS GG+GA	0.88[0.64,1.21]	0.77	0.44	45%	0.13	Fixed-effects model
A VS G	1.08[0.85,1.37]	0.65	0.52	68%	0.01	Random-effects model

Overall analyses for MTHFD1-G1958A gene polymorphisms in fetal analysis

For the CHD group, 7 studies were summarized, including 1917 CHD cases and 1863 healthy controls [9, 27–31, 35]. Homozygous model (I² = 63%, P = 0.61), dominant model (I² = 61%, P = 0.70), recessive model (I² = 58%, P = 0.72) and allele model (I² = 68%, P = 0.28) had high heterogeneity, which were analyzed by random effect model. The heterogeneity of the heterozygote model was low, so we used a fixed-effect model for analysis. The results of the meta-analysis of the association between MTHFD1-G1958A gene polymorphism and CHD risk were shown in the Table 6.

The results showed that there was no correlation between MTHFD1-G1958A polymorphism and offspring CHD in the five models (Fig. 3).

Fig. 3.

Forest plot of published correlation between the fetal MTHFD1-G1958A polymorphism and CHD risk

Subgroup analyses for MTHFD1G1958A gene polymorphisms in fetal analysis

Analysis of MTHFD1-G1958A gene polymorphism and TOF

In the TOF group, a total of 3 studies were included, including 72 TOF cases and 330 healthy controls [9, 27, 35]. The heterogeneity of the five models is low, and the fixed effect model is used for analysis. The results showed that two genetic models of MTHFD1-G1958A gene polymorphism in TOF group: homozygous model (AA vs. GG, OR = 2.82,95% CI [1.16, 6.86], P = 0.02) and recessive model (AA vs. GG + GA, OR = 3.09,95% CI [1.36, 7.03], P = 0.007) increased the risk of TOF, but no significant correlation was found in other models (Fig. 4).

Fig. 4.

Forest plot of published correlation between the fetal MTHFD1-G1958A gene polymorphism and TOF risk

Analysis of MTHFD1-G1958A gene polymorphism and septal defect (SD)

A total of 5 studies were included in the septal defect group, including 648 cases of septal defect and 1055 healthy controls [9, 27–29, 35]. The heterogeneity of the dominant model (I² = 60%, P = 0.48) was high, and the random effect model was used for analysis. The other four genetic models have low heterogeneity, so we use the fixed effect model for analysis. The results showed that in the septal defect group, MTHFD1-G1958A gene polymorphism was not found in the five genetic models. (Fig. 5)

Fig. 5.

Forest plot of published correlation between the fetal MTHFD1-G1958A gene polymorphism and SD risk

Four studies were included in the ASD group, including 98 ASD cases and 857 controls [9, 27, 29, 35]. Among them, the heterogeneity in the allele model (I² = 62%, P = 0.80) was high, and the random effect model was used for analysis. The other four models have low heterogeneity and are analyzed using a fixed effect model. In the ASD group, MTHFD1-G1958A gene polymorphism was not found to be significantly correlated in the five genetic models (Fig. 6).

Fig. 6.

Forest plot of published correlation between the fetal MTHFD1-G1958A gene polymorphism and ASD risk

A total of 4 studies were included in the VSD group, including 451 VSD cases and 857 controls [9, 27, 29, 35]. The heterogeneity of the dominant model (I² = 63%, P = 0.20) was high, and the random effect model was used for analysis. The other four models have low heterogeneity and are analyzed using a fixed effect model. In the VSD group, no significant correlation was found in the five genetic models (Fig. 7).

Fig. 7.

Forest plot of published correlation between the fetal MTHFD1-G1958A gene polymorphism and VSD risk

Analysis of MTHFD1-G1958A gene polymorphism in different races

Our study included yellow and white races, and the pooled data were from seven studies, including 1719 cases and 1863 controls. Four studies were included in the yellow race subgroup, including 1461 cases and 1457 controls [29–31, 35]. The heterogeneity of the 5 models was low, and the fixed effect model was used for analysis. Three studies were included in the white race subgroup, including 456 cases and 406 controls [9, 27, 28]. The heterogeneity of the dominant model (I² = 85%, P = 0.59) and the allele model (I² = 58%, P = 0.02) was high, and the random effect model was used for analysis. The other three models have low heterogeneity, so the fixed effect model is used for analysis.

The results showed that the MTHFD1-G1958A gene polymorphism in the yellow race subgroup was not significantly associated with CHD in the five basic genetic models (Fig. 8).

Fig. 8.

Forest plot of published correlation between the fetal MTHFD1-G1958A polymorphism and CHD risk in yellow race

In the white race subgroup, MTHFD1-G1958A polymorphism increased the risk of CHD in three genetic models: allele model (A vs. G, OR = 1.74, 95% CI [1.09, 2.78], P = 0.02), recessive model (AA vs. GG + GA, OR = 2.44, 95% CI [1.31, 4.55], P = 0.005), homozygous model (AA vs. GG, OR = 2.84, 95% CI [1.47, 5.49], P = 0.002). No significant correlation was found in other models (Fig. 9).

Fig. 9.

Forest plot of published correlation between the fetal MTHFD1-G1958A polymorphism and CHD risk in white race

Analysis of MTHFD1-G1958A gene polymorphism in different regions

The geographical subgroup pooled five studies, all from Asia, including 1563 cases and 1555 controls [27, 29–31, 35]. The heterozygote model (I² = 19%, P = 0.46) and the recessive model (I² = 45%, P = 0.44) had low heterogeneity, and the fixed effect model was used for analysis. The other three models had high heterogeneity and were analyzed using a random effect model. The results showed that there was no significant correlation between MTHFD1-G1958A gene polymorphism in the five genetic models (Fig. 10).

Fig. 10.

Forest plot of published correlation between the fetal MTHFD1-G1958A polymorphism and CHD risk in Asia

Association of maternal MTHFD1-G1958A gene polymorphism and its subgroups with CHD in offspring (Table 7)

Table 7.

Meta-analysis of maternal MTHFD1 gene G1958A polymorphism and the risk of congenital heart disease in offspring

Type	OR(95%CI)	z	P	Test of heterogeneity		Analysis model
				I2	P *
Overall(5)
GA VS GG	1.22[1.04,1.42]	2.45	0.01	0%	0.84	Fixed-effects model
AA VS GG	1.02[0.72,1.43]	0.09	0.93	0%	0.55	Fixed-effects model
GA + AA VS GG	1.17[1.01,1.34]	2.15	0.03	0%	0.71	Fixed-effects model
AA VS GG + GA	0.94[0.68,1.31]	0.35	0.73	0%	0.61	Fixed-effects model
A VS G	1.12[0.99,1.26]	1.74	0.08	0%	0.44	Fixed-effects model
SD(3)
GA VS GG	0.98[0.64,1.49]	0.10	0.92	0%	0.92	Fixed-effects model
AA VS GG	0.53[0.25,1.12]	1.67	0.10	0%	0.62	Fixed-effects model
GA + AA VS GG	0.93[0.68,1.28]	0.43	0.67	0%	0.90	Fixed-effects model
AA VS GG + GA	0.54[0.27,1.08]	1.73	0.08	0%	0.62	Fixed-effects model
A VS G	0.82[0.61,1.11]	1.26	0.21	0%	0.90	Fixed-effects model
yellow race(3)
GA VS GG	1.22[1.04,1.44]	2.38	0.02	0%	0.67	Fixed-effects model
AA VS GG	1.01[0.69,1.49]	0.05	0.96	6%	0.35	Fixed-effects model
GA + AA VS GG	1.20[1.02,1.40]	2.24	0.03	0%	0.47	Fixed-effects model
AA VS GG + GA	0.95[0.65,1.39]	0.28	0.78	0%	0.4	Fixed-effects model
A VS G	1.13[0.99,1.29]	1.79	0.07	19%	0.29	Fixed-effects model
white race(2)
GA + AA VS GG	1.05[0.76,1.45]	0.31	0.76	0%	0.7	Fixed-effects model

Overall analyses for MTHFD1-G1958A gene polymorphisms in maternal analysis

For the mother group of CHD patients, five studies were summarized, including 1718 mothers and 1666 healthy controls [9, 25, 26, 28, 35]. The heterogeneity of the five models was low and the fixed effect model was used for analysis. The results showed that MTHFD1-G1958A gene polymorphism increased the risk of CHD in the two genetic models: heterozygous model (GA vs. GG, OR = 1.22,95% CI [1.04, 1.42], P = 0.01) and dominant model (GA + AA vs. GG, OR = 1.17,95% CI [1.01, 1.34], P = 0.03). No significant correlation was found in other models (Fig. 11).

Fig. 11.

Forest plot of published correlation between the maternal MTHFD1-G1958A polymorphism and CHD risk in offspring

Analysis of maternal MTHFD1-G1958A gene polymorphism in septal defect

A total of 3 studies were included in the septal defect group, including 270 mothers of patients with septal defects and 422 healthy controls [9, 28, 35]. The heterogeneity of the five genetic models is low, so we use the fixed effect model for analysis. The results showed that in the septal defect group, MTHFD1-G1958A gene polymorphism was not found in the five genetic models (Fig. 12).

Fig. 12.

Forest plot of published correlation between the maternal MTHFD1-G1958A gene polymorphism and SD risk in offspring

Analysis of maternal MTHFD1-G1958A gene polymorphism in different races

Our study included yellow and white races, and the pooled data were from five studies, including 1718 cases and 1666 controls. A total of 3 studies were included in the yellow race subgroup, including 1324 cases and 1363 controls [25, 26, 35]. The heterogeneity of the 5 models was low, and the fixed effect model was used for analysis. Two studies were included in the white race subgroup, including 394 cases and 303 controls [9, 28]. Due to the lack of some data, only explicit model data analysis can be performed, and the heterogeneity is low, and the fixed effect model is used.

The results showed that in the yellow race subgroup, the MTHFD1-G1958A gene polymorphism in the mother was in two genetic models: heterozygous gene model (GA vs. GG, OR = 1.22, 95% CI [1.04,1.44], P = 0.02), dominant gene model (GA + AA vs. GG, OR = 1.20, 95% CI [1.02,1.40], P = 0.03) increased the risk of CHD in their offspring, and no significant correlation was found in other models (Fig. 13). The MTHFD1-G1958A gene polymorphism in the white race subgroup was not found to be significantly associated with CHD in the dominant genetic model (Fig. 14).

Fig. 13.

Forest plot of published correlation between the maternal MTHFD1-G1958A polymorphism and CHD risk in yellow race

Fig. 14.

Forest plot of published correlation between the maternal MTHFD1-G1958A polymorphism and CHD risk in white race

Publication Bias and Heterogeneity Test

Since the number of included articles was less than 10, we did not assess the publication bias; Due to the low heterogeneity of the included studies, we did not perform sensitivity analysis.

Discussion

CHD is a significant public health issue that affects children’s health and population quality. Based on literature retrieval and analysis, this study is the first meta-analysis to investigate the correlation between MTHFD1-G1958A polymorphism and CHD and its subgroups. We have tracked all relevant literature and attempted to summarize and analyze whether mutations at this site would increase the risk of CHD. The systematic review aims to elucidate the current research landscape in this field, providing valuable insights for future clinical investigations.

In the sub-analysis of CHD, MTHFD1-G1958A polymorphism was associated with the risk of TOF in offspring. Compared with GG genotype, AA genotype increased the risk of TOF by 182%. Compared with the GG + GA genotype, the risk of TOF in the AA genotype increased by 209%, suggesting that the genotype AA in the MTHFD1-G1958A gene polymorphism may increase the susceptibility to TOF. In the overall analysis of CHD children, no association was found between MTHFD1-G1958A polymorphism and CHD in offspring, which was consistent with the results of Xu, Wang et al. [29, 30]. The results of this meta-analysis showed that the MTHFD1-G1958A gene polymorphism may increase the risk of TOF in offspring.

Researching the association between maternal MTHFD gene polymorphism and CHD in offspring can provide a scientific basis for judging whether the parents are high-risk individuals. Previous studies reported that maternal MTHFD1-G1958A gene polymorphism is associated with CHD susceptibility in offspring [35]. Some studies showed that the above loci are not related to CHD [25, 26], and most of the published studies not accurately elucidated the correlation with CHD in offspring. In the overall analysis of the maternal group, five studies were summarized, including 1718 mothers of CHD patients and 1666 healthy controls. The maternal MTHFD1-G1958A polymorphism was associated with the risk of CHD in offspring. Compared with GG genotype, the GA genotype increased the risk of CHD by 22%, and the GA + AA genotype increased the risk by 17%. In the sub-analysis of CHD, no association was found between the maternal MTHFD1-G1958A polymorphism and the risk of offspring with septal defect. The results of this study indicate that the presence of the MTHFD1-G1958A variant genotype in mothers is associated with an increased risk of CHD in their offspring.

Through a comprehensive literature review, we identified that the number of studies on MTHFD1-G1958A in CHD is relatively scarce. In addition to rs2236225 (G1958A), rs1950902, rs2236222, rs1256142 and rs11849530 are also relatively common mutations in MTHFD1 gene. It has been reported that the polymorphisms of maternal MTHFD1 gene rs1256142, rs2236222 and rs11849530 are significantly associated with the risk of CHD in offspring, while the correlation between rs1950902 polymorphism and CHD is controversial [25, 26]. There is an interaction between MTHFD1 gene and environment. The rs2236222 is a synonymous polymorphism of MTHFD1. Maternal MTHFD1 rs2236222 gene polymorphism increases the risk of CHD in offspring. Maternal tobacco exposure and dietary habits during pregnancy interact with rs2236222, which will further increase the risk of CHD in offspring [25, 36]. Of course, there is also an interaction between SNPs. It has been reported that the interaction between maternal MTHFD1 and MTHFD2 gene polymorphisms, and the interaction between two loci (MTHFD1 rs1950902, MTHFD1 rs2236222) and three loci (MTHFD1 rs1950902, MTHFD1 rs1256142, MTHFD2 rs1095966) are related to the occurrence of CHD in offspring [26]. However, there are few studies on gene interactions and the effects of environment on genes, which limits the analysis of interactions. It can be seen that the research on the association between maternal MTHFD1 gene and offspring CHD is limited to several common SNP loci, and the results are inconsistent. The heterogeneity between different populations and the selection of controls may be the reasons for the differences in the above results. In the future, the research sites on this gene can be expanded, and the role of population and region in these aspects can be investigated.

Our study included both the yellow race (all Chinese Han) and the white race, and sub-analysis was performed according to different races. The MTHFD1-G1958A polymorphism was associated with the risk of CHD in the white race of the fetal group. The risk of CHD in the allele A increased by 74%, and the risk of homozygous AA genotype increased by 184%, but this risk was not found in the yellow race. In the yellow subgroup of maternal group, MTHFD1-G1958A polymorphism was associated with the risk of CHD in offspring. Compared with GG genotype, GA increased the risk of CHD by 22%, and GA + AA increased the risk by 20%. In the white race subgroup, due to the lack of research data, only the dominant model was analyzed, and other genetic model risks could not be analyzed.

We tried to explore the reasons for the ethnic differences in the association between MTHFD1-G1958A gene polymorphism and CHD risk. The study found that the genotype frequency of GG / GA / AA in the western population is the highest proportion of GA, and the mutation frequency is about 50%. In the northern Chinese population, GG accounts for the vast majority, and the mutation frequency is not more than 25% [16, 30, 37]. The mutation frequency of MTHFD1-G1958A locus in northern Chinese population is significantly lower than that observed in western population, indicating potential racial differences in gene polymorphism at this locus. Although this study shows that racial differences may affect the incidence of CHD, the precise genetic and / or environmental factors remain undetermined. Individuals from diverse racial backgrounds exhibit genomic differences, suggesting that certain variants or susceptibility loci for CHD may be more prevalent in specific ethnic groups. However, little is known about the causes of racial differences in CHD, and it is necessary to further study its genetic factors. It is conceivable that clinical DNA sequencing, gene chip technology, and karyotyping could be extensively applied in patients with CHD, thereby continuously accumulating cases to establish a substantial sample size for future research validation. In addition, the racial differences that affect the occurrence of CHD may also be attributed to socio-economic differences, cultural differences, dietary habits and other environmental exposure factors. Different races have different diets, so the intake and level of folic acid are also different. The resulting maternal health differences may affect the fetal development environment and lead to birth defects.

Our meta-analysis has some limitations. First, this study is the systematic review of the only case-control studies, indicating that the recruited participants may not adequately represent the general population, and selection bias cannot be entirely excluded. Therefore, a large-scale population-based study is needed to clarify the role of MTHFD1-G1958A gene polymorphism and CHD susceptibility. Secondly, MTHFD1 polymorphism and insufficient maternal folic acid supplementation in early pregnancy have a strong synergistic effect on the development of CHD. The subjects of this study were CHD patients and their mothers, but they lacked information such as maternal folic acid supplementation during pregnancy and folic acid levels in the body. It was difficult to analyze whether CHD was related to maternal folic acid levels during pregnancy. Therefore, it is a challenge to explore gene-gene, gene-environment and maternal-fetal genotype interactions. Finally, the sample size of this study is relatively small, and the data extraction of the original literature is limited. All the included studies are retrospective studies, and there may be other bias factors. Especially in the CHD subgroup, a larger sample size study is needed to confirm and quantify the risk of CHD.

Conclusions

In summary, the MTHFD1-G1958A gene polymorphism in the offspring was only observed to increase its risk in TOF, and the risk of other CHD types was not clear. The maternal MTHFD1-G1958A gene polymorphism is associated with an increased risk of CHD in the offspring. In addition, our study showed that there were racial differences in this correlation. The MTHFD1-G1958A polymorphism was associated with the risk of CHD in Caucasians. The etiology of congenital heart disease is complex, and more research is needed in the future to provide stronger evidence and scientific basis for disease prevention.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

CHD

Congenital heart diseases

MTHFD1

Methylenetetrahydrofolate dehydrogenase 1

SNPs

Single nucleotide polymorphisms

TOF

Tetralogy of Fallot

TGA

Transposition of Great Artery

CTD

Conotruncal defects

VSD

Ventricular septal defect

ASD

Atrial septal defect

PDA

Patent ductus arteriosus

Hcy

Homocysteine

NTD

Neural tube defects

ADHD

Attention deficit hyperactivity disorder

Author contributions

Kang Yi, Shao-E He, Tao You and Tao Guo wrote the main manuscript text. Zi-Qiang Wang, Xin Zhang, Jian-Guo Xu, Wei-Guo Liu and Hao-Yue Zhang prepared figures and tables. All authors reviewed the manuscript.

Funding

The new fund projects are as follows: Medicine Research Fund Project of Gansu Provincial Hospital (23GSSYF-30); Natural Science Foundation of Gansu Province (22JR5RA655); Health Industry Research Project of Gansu Province of China (GSWSKY-2019-76).

Data availability

Data is provided within the manuscript or supplementary information files.

Declarations

Ethical approval

As a systematic review, there is no ethical approval is required.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.