Association between CBS gene T833C, G919A and 844ins68 polymorphisms in the 8th exon region and coronary artery disease: a meta-analysis

ABSTRACT

Background

Several studies indicate that the cystathionine β-synthase (CBS) gene T833C, G919A and 844ins68 polymorphisms in the 8th exon region may be correlated with coronary artery disease (CAD) susceptibility, but the results have been inconsistent and inconclusive. Thus, a meta-analysis was conducted to provide a comprehensive estimate of these associations.

Methods

On the basis of searches in the PubMed, EMBASE, Cochrane Library, Wanfang, VIP, and CNKI databases, we selected 14 case – control studies including 2123 cases and 2368 controls for this meta-analysis. Pooled odds ratios (ORs) with 95% confidence intervals (CIs) were calculated accordingly using a fixed-effect or random-effect model.

Results

The results indicated an increased risk between the CBS T833C gene polymorphisms and susceptibility to CAD under the dominant model (CC+CT vs. TT: OR = 1.92, 95% CI: 1.11 ~ 3.32), recessive model (CC vs. CT+TT: OR = 1.88, 95% CI: 1.17 ~ 3.03), and homozygous model (CC vs. TT: OR = 2.46, 95% CI: 1.04 ~ 5.83). In these three genetic models, no significant association was identified for CBS G919A (AA+AG vs. GG: OR = 1.48, 95% CI: 0.45 ~ 4.82),(AA vs. AG+GG: OR = 1.58, 95% CI: 0.93 ~ 2.70),(AA vs. GG: OR = 1.66, 95% CI: 0.40 ~ 6.92) or CBS 844ins68 (II+ID vs. DD: OR = 1.04, 95% CI: 0.80 ~ 1.35),(II vs. ID+DD: OR = 1.09, 95% CI: 0.51 ~ 2.36),(II vs. DD: OR = 1.10, 95% CI: 0.51 ~ 2.39).

Conclusions

This meta-analysis suggests that the CBS T833C gene polymorphism is significantly associated with the risk of CAD and it shows a stronger association in Asian populations. Individuals with the C allele of the CBS gene T833C polymorphism might be particularly susceptible to CAD.

KEYWORDS:

• Cystathionine β-synthase
• polymorphism
• coronary artery disease
• meta-analysis

Background

Coronary artery disease (CAD) is a disease involving the formation of atherosclerotic plaques in coronary vessels caused by a variety of factors and may be characterized by stable angina, unstable angina, myocardial infarction, and sudden cardiac death^1,2. CAD has become a public health challenge and is a leading cause of morbidity and mortality in adults in both industrialized and developing countries around the world³. According to the World Health Organization (WHO), approximately 7.3 million people die of CAD each year⁴. Both genetic and environmental factors are involved in the pathogenesis of CAD, and genetic factors can explain 30% to 60% of the CAD risk^5,6. Genome-wide association studies (GWAS) have identified some genetic variants that contribute to the risk of CAD among the general population⁷.

Hyperhomocysteinemia has been associated with a high risk of CAD⁸. Homocysteine (Hcy) damages vascular endothelial cells, stimulates vascular smooth muscle cell proliferation, increases platelet adhesion and aggregation, promotes thrombosis and is an important risk factor for atherosclerosis and cardiovascular disease^9–11. Cystathionine-β-synthase (CBS) is one of the key enzymes in the Hcy metabolic pathway. CBS catalyzes the transformation of serine and homocysteine to cystathionine and water. This enzyme exists in the form of a homotetramer, and each submit is combined with one heme and one pyridoxal 5’-phosphate (LPL) in the active form of CBS¹². Kruger¹³ showed that specific CBS alleles are risk factors for vascular disease, and CBS gene mutations may affect the folding and stability of the encoded protein, reducing the enzyme activity of CBS, which affects the metabolism of Hcy and further leads to CAD¹⁴.

The CBS gene consists of 23 exons, is located on chromosome 21q22.3 and is 25–30 kbp long, encoding a protein of 551 amino acids¹⁵. Approximately 80% of CBS alleles are missense mutations, while the remainder are deletion alleles. To date, 164 pathogenic genetic mutations in CBS have been confirmed, most of which are located in the 3rd and 8th exons. Among the above mutations, the frequency of the T833C, G919A and 844ins68 mutations in the 8th exon is very high, and the 844ins68 mutation occurs in approximately 5% of Caucasians¹⁶. The role of CBS gene polymorphisms in the development of CAD has been widely studied in patients, and studies on the correlation between the T833C, G919A and 844ins68 gene mutations and CAD have been reported. Dilley¹⁷ showed that the T833C gene polymorphism exhibited no significant correlation with CAD. Xu and Chen¹⁸ showed that the T833C gene polymorphism was an important genetic factor in CAD. Xu¹⁹ showed that the G919A gene polymorphism presented no significant correlation with CAD. Chen²⁰ considered the G919A mutation to be a risk factor for CAD, and both Iqbal²¹ and Zhang²² found that the 844ins68 gene polymorphism was not associated with CAD.

These gene polymorphisms (T833C, G919A and 844ins68) associated with CAD all occur in the 8th exon region of the CBS gene. Their mutation rates are higher and better studied than those of other polymorphisms, but the results are still controversial and conflicting. It is very possible that a single study may be underpowered for the detection of polymorphisms related to disease risk, especially when the sample size is relatively small. Different types of study populations and study designs may also contribute to the disparate findings. To confirm the correlation between the CBS gene T833C, G919A and 844ins68 polymorphisms and CAD, we performed this meta-analysis by pooling all eligible studies to calculate an estimate of the overall CAD risk.

Methods

Search strategy

All studies on the association between the above three CBS gene polymorphisms (G919A, T833C, 844ins68) and CAD were identified via computer-based searches of the PubMed, Cochrane Library, Web of Science, CNKI, Wanfang and VIP databases, and the time period for the reference searches was from the first available article to January 27, 2024. The following key words were used for the searches: “cystathionine β synthase,” “cystathionine-β-synthase,” “cystathionine-β synthase,” “cystathionine β-synthase,” “cystathionine beta Synthase,” “cystathionine-beta-Synthase,” “cystathionine-beta Synthase,” “cystathionine beta-Synthase,” “CBS,” “myocardial infarction,” “coronary heart disease,” and “coronary artery disease.” The selection process is illustrated in Figure 1.

Figure 1. Flow chart of study selection.

Inclusion and exclusion criteria

Inclusion criteria: (1) study on the association between CBS gene polymorphisms and CAD; (2) case-control design; (3) CAD patients were the case group, and healthy or non-CAD patients were the control group;(4) evaluation standard: the risk of CAD; (5) for repeatedly published studies, we chose the most realistic and accurate study.

Exclusion criteria: (1) multiple publications reporting the same finding; (2) abstracts, reviews, comments and animal studies; (3) a lack of sufficient data.

Data extraction

In this study, two evaluators independently screened the literature and extracted relevant information according to the inclusion and exclusion criteria by using EndNote software. In the case of disagreement, a decision was reached between these two evaluators or made by another evaluator. The information extracted from the studies included the first author’s name, the year of publication, the country of the study, the ethnicity of the subjects, the diagnostic method, the number of cases in the coronary artery disease group and the control group, and the numbers of each genotype identified.

Study quality assessment

The quality of the included studies was independently assessed by the Newcastle-Ottawa Quality Assessment Scale (NOS)²³. The evaluation project consisted of 8 parts, and except for the fifth evaluation criterion of 2 points, the scores of the other items were all 1 point. The total scores of the NOS range from 0 points to 9 points. If the total score of the 8 items is greater than or equal to 7 points, the quality of studies is considered reliable.

Statistical analysis

The statistical analysis was performed by using Review Manager 5.3. Pooled odds ratios and 95% confidence intervals²⁴ were applied to estimate the strength of associations between the CBS gene polymorphisms and CAD risk. The heterogeneity between the included studies was judged on the basis of P and I² values. Values of p > .10 and I² <50% indicated that the fixed effect model could be used in this meta-analysis, however, values of p < .10 and I² >50% indicated the existence of significant heterogeneity²⁵. We identified the source of heterogeneity and performed further analysis, and the random effects model was used after excluding the effects of significant clinical heterogeneity. To further study the effect of sample size, diagnosis methods and ethnicity on heterogeneity, subgroup analysis was performed. The potential publication bias of this literature was investigated by Begg‘s test and Egger’s liner regression test, a p value of less than 0.05 was considered to showed statistically significant publication bias

Results

Study characteristics

A total of 148 studies were identified through database searching. After duplicates removed, leaving 108 studies for screening. Then, 82 studies were excluded since they were reviews, animal experiments and studies not related to this topic. After detailed evaluation, 12 studies were excluded because of insufficient data, no control and duplication of previous data. Ultimately, a total of 14 publications involving 2123 CAD cases and 2368 controls finally met the inclusion criteria and were analyzed. The subjects of 13 studies were Asian, and only 1 study¹⁷, on the CBS T833C gene polymorphism, examined North American subjects. Among the 14 studies, 4 studies involving 612 cases and 500 controls addressed the CBS T833C gene polymorphism, 3 studies including 427 cases and 331 controls addressed the CBS G919A gene polymorphism, and 4 studies comprising 1084 cases and 1537 controls addressed the CBS 844ins68 gene polymorphism. Study characteristics is shown in Table 1.

Table 1. Characteristics of the included studies in the meta-analysis.

	First-author	Year	Country	Ethnicity	Sources of Controls	Diagnosis methods	Coronary artery disease group			Control group
							Disease	Number of cases	HWE balance test	Crowd characteristics	Number of cases	HWE balance test
T833C	Dilley^Citation17	2001	United States	North American	HCC	unclear	MI	107	unclear	healthy	172	unclear
	Xu Haiyan^Citation18	1999	China	Asian	HCC	CAA	CAD	209	unclear	healthy	101	unclear
	Chen Xin^Citation20	2003	China	Asian	HCC	Criteria for CAD	CAD	107	unclear	healthy	96	unclear
	Chen Yulan^Citation26	2006	China	Asian	HCC	CAA	CAD	189	Yes	healthy	131	Yes
G919A	Li Lifan^Citation27	2005	China	Asian	HCC	Criteria for CAD	CAD	65	unclear	healthy	65	unclear
	Xu Haiyan^Citation19	2007	China	Asian	HCC	CAA	MI or CAD	255	unclear	healthy	170	unclear
	Chen Xin^Citation20	2003	China	Asian	HCC	Criteria for CAD	CAD	107	unclear	healthy	96	unclear
844ins68	Iqbal^Citation21	2016	Karachi Pakistan	Asian	PCC	Criteria for AMI	AMI	143	No	healthy	154	No
	Zhang^Citation22	2001	China	Asian	HCC	CAA	MI	73	unclear	healthy	100	unclear
	Yang Linyan^Citation28	2011	China	Asian	HCC	Criteria for CAD	CAD	210	Yes	healthy	249	Yes
	Luo Dan^Citation29	2009	China	Asian	HCC	CAA	DM2+CAD	104	unclear	healthy	91	unclear
	Li Yaohuang^Citation30	2005	China	Asian	HCC	Crieria for CAD	CAD	86	unclear	healthy	143	unclear
	Jiao Kunli^Citation31	2006	China	Asian	HCC	CAA	MI	208	Yes	healthy	500	Yes
	He Guanjun^Citation32	2015	China	Asian	HCC	CAA	CAD	260	unclear	healthy	300	unclear

PCC: population-based case-control study; HCC: hospital-based case-control study; CAA: coronary angiography; Criteria for CAD: diagnosis of coronary heart disease; MI: myocardial infarction; AMI: acute myocardial infarction; T2DM: type 2 diabetes.

Quality evaluation results

According to the NOS quality evaluation system, the quality of the 14 case-control studies was further evaluated in detail. The results showed that the total scores of the 14 studies included were all ≥ 7, which indicated that the 14 included studies were reliable. Detailed information on the studies is listed in Table 2.

Table 2. NOS quality evaluation form.

	Inclusion study	Study population selection				Group-to-group	comparison of exposure factors			Total (minutes)
		①	②	③	④	⑤	⑥	⑦	⑧
T833C	Dilley 2001^Citation17	1	1	1	1	2	1	1	0	8
	Xu Haiyan 1999^Citation18	1	1	1	1	2	1	1	1	9
	Chen Xin 2003^Citation20	1	1	1	1	1	1	1	1	8
	Chen Yulan 2006^Citation26	1	1	1	1	2	1	1	1	9
G919A	Li Lifan 2005^Citation27	1	1	1	1	1	1	1	1	8
	Xu Haiyan 2007^Citation19	1	1	1	1	2	1	1	0	8
	Chen Xin 2003^Citation20	1	1	1	1	1	1	1	1	8
844ins68	Iqbal 2016^Citation21	1	1	1	1	2	1	1	1	9
	Zhang 2001^Citation22	1	1	1	1	1	1	1	1	8
	Yang Linyan 2011^Citation28	1	1	1	1	0	1	1	1	7
	Luo Dan 2009^Citation29	1	1	1	1	1	1	1	1	8
	Li Yaohuang2005^Citation30	1	1	1	1	1	1	1	1	8
	Jiao Kunli 2006^Citation31	1	1	1	1	2	1	1	1	9
	He Guanjun 2015^Citation32	1	1	1	1	2	1	1	1	9

①The 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 (eg 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.

CBS T833C polymorphism

Four studies^17,18,20,26 addressing the correlation between 844ins68 and CAD were included in the analyses, among which 3 papers were from China, and 1 paper was from America, involving 612 CAD cases and 500 controls. In these 4 studies, there were 64 CC genotypes and 340 TT genotypes in the CAD group, and there were 28 CC genotypes and 348 TT genotypes in the control group (Table 3). Taking CC as the exposure factor and TT as the nonexposure factor, a random effects model was used to combine the data. The results showed that the T833C CC genotype was associated with susceptibility to CAD (CC vs. TT: OR = 2.46, 95% CI: 1.04 ~ 5.83). There were 272 CC+CT genotypes and 340 TT genotypes in the CAD group, and there were 152 CC+CT genotypes and 348 TT genotypes in the control group. Taking CC+CT as the exposure factor and TT as the nonexposure factor, a random effects model was used to combine the data. The results showed that the T833C CC+CT genotype was associated with susceptibility to CAD (CC+CT vs. TT: OR = 1.92, 95% CI: 1.11 ~ 3.32). There were 64 CC genotypes and 348 CT+TT genotypes in the CAD group, and there were 28 CC genotypes and 472 CT+TT genotypes in the control group. Taking CC as the exposure factor and CT+TT as the nonexposure factor, a fixed effects model was used to combine the data. The results showed that the T833C CC genotype was associated with susceptibility to CAD (CC vs. CT+TT: OR = 1.88, 95% CI: 1.17 ~ 3.03).

Table 3. Distributions of the genotypes of the CBS T833C、G919A、844ins68.

		Cases			Controls
	Inclusion study	CC(%)	CT(%)	TT(%)	CC(%)	CT(%)	TT(%)
CBS T833C	Dilley 2001^[17]	7(6.54)	36(33.65)	64(59.81)	8(4.65)	56(32.56)	108(62.79)
	Xu Haiyan 1999^[18]	21(10.04)	94(44.98)	94(44.98)	2(1.98)	28(27.72)	71(70.30)
	Chen Xin 2003^[20]	9(8.41)	26(24.30)	72(67.29)	2(2.08)	9(9.38)	85(88.54)
	Chen Yulan 2006^[26]	27(14.29)	52(27.51)	110(58.20)	16(12.21)	31(23.66)	84(64.12)
		GG(%)	AG(%)	AA(%)	GG(%)	AG(%)	AA(%)
CBS G919A	Li Lifan 2005^[27]	40(61.54)	19(29.23)	6(9.32)	53(81.54)	10(15.38)	2(3.08)
	Xu Haiyan 2007^[19]	19(7.45)	202(79.22)	34(13.33)	5(2.94)	147(86.47)	18(10.59)
	Chen Xin 2003^[20]	79(73.83)	22(20.56)	6(5.61)	85(88.54)	9(9.38)	2(2.08)
		DD(%)	ID(%)	II(%)	DD(%)	ID(%)	II(%)
CBS 844ins68	Iqbal 2016^[21]	127(88.81)	14(9.79)	2(1.40)	132(85.71)	21(13.64)	1(0.65)
	Zhang 2001^[22]	72(98.63)	1(1.37)	0	95(95.00)	5(5.00)	0
	Yang Linyan 2011^[28]	185(88.10)	24(11.43)	1(0.47)	217(87.15)	30(12.05)	2(0.80)
	Luo Dan 2009^[29]	102(98.07)	2(1.92)	0	85(93.41)	6(6.59)	0
	Li Yaohuang2005^[30]	67(77.91)	16(18.60)	3(3.48)	112(78.32)	27(18.88)	4(2.80)
	Jiao Kunli 2006^[31]	184(88.46)	23(11.06)	1(0.48)	463(92.60)	35(7.00)	2(0.40)
	He Guanjun 2015^[32]	235(90.38)	20(7.69)	5(1.92)	278(92.67)	16(5.33)	6(2.00)

Overall, the results showed a significant association between the CBS T833C gene polymorphism and the risk of CAD under the homozygous model (CC vs. TT: OR = 2.46, 95% CI: 1.04 ~ 5.83), the dominant model (CC+CT vs. TT: OR = 1.92, 95% CI: 1.11 ~ 3.32) and the recessive model (CC vs. CT+TT: OR = 1.88, 95% CI: 1.17 ~ 3.03). (Table 4，Figure S2)

Table 4. Summary of odds ratios (95% CI) in the analysis of the association between the CBS polymorphism and CAD susceptibility in different models.

Polymorphism	Model	Test of association		Test of heterogeneity		Forest plot
		OR (95% CI)	P value	I^2	P
T833C	Homozygote model (CC vs. TT)	2.46(1.04 ~ 5.83)	0.04	57%	0.07	Figure S2-1
	Dominant model (CC+CT vs. TT)	1.92(1.11 ~ 3.32)	0.02	76%	0.006	Figure S2–2
	Recessive model (CC vs. CT+TT)	1.88(1.17 ~ 3.03)	0.01	42%	0.16	Figure S2–3
G919A	Homozygote model (AA vs. GG)	1.66(0.40 ~ 6.92)	0.48	65%	0.06	Figure S2–4
	Dominant model (AA+AG vs. GG)	1.48(0.45 ~ 4.82)	0.52	83%	0.003	Figure S2–5
	Recessive model (AA vs. AG+GG)	1.58(0.93 ~ 2.70)	0.09	0%	0.45	Figure S2–6
844ins68	Homozygote model (II vs. DD)	1.10(0.51 ~ 2.39)	0.81	0%	0.96	Figure S2–7
	Dominant model (II+ID vs. DD)	1.04(0.80 ~ 1.35)	0.76	29%	0.21	Figure S2–8
	Recessive model (II vs. ID+DD)	1.09(0.51 ~ 2.36)	0.82	0%	0.96	Figure S2–9

OR – odds ratio; CI – confidence interval.

CBS G919A polymorphism

Three studies^19,20,27 addressing the correlation between G919A and CAD were included in the analyses, all of which were from China, involving 427 CAD cases and 331 controls. Among these 3 studies, there were 46 AA genotypes and 138 GG genotypes in the CAD group, and there were 22 AA genotypes and 143 GG genotypes in the control group (Table 3). Taking AA as the exposure factor and GG as the nonexposure factor, a random effects model was used to combine the data. The results showed that the G919A AA genotype was not associated with the risk of CAD (AA vs. GG: OR = 1.66, 95% CI: 0.40 ~ 6.92). There were 289 AA+AG genotypes and 143 GG genotypes in the CAD group, and there were 188 AA+AG genotypes and 143 GG genotypes in the control group. Taking AA+AG as the exposure factor and GG as the nonexposure factor, a random effects model was used to combine the data. The results showed that the G919A AA+AG genotype was not associated with susceptibility to CAD (AA+AG vs. GG: OR = 1.48, 95% CI: 0.45 ~ 4.82). There were 46 AA genotypes and 381 AG+GG genotypes in the CAD group, and there were 22 AA genotypes and 309 AG+GG genotypes in the control group. Taking AA as the exposure factor and AG+GG as the nonexposure factor, a fixed effects model was used to combine the data. The results showed that the G919A AG+GG genotype was not associated with susceptibility to CAD (AA vs. AG+GG: OR = 1.58, 95% CI: 0.93 ~ 2.70).

Overall, the results showed that there was no association between the CBS G919A gene polymorphism and the risk of CAD under the homozygous model (AA vs. GG: OR = 1.66, 95% CI: 0.40 ~ 6.92), the dominant model (AA+AG vs. GG: OR = 1.48, 95% CI: 0.45 ~ 4.82) or the recessive model (AA vs. AG+GG: OR = 1.58, 95% CI: 0.93 ~ 2.70). (Table 4，Figure S2)

CBS 844ins68 polymorphism

Seven studies^{21,22,28–32} addressing the correlation between T833C and CAD were included in the analyses, among which 6 papers were from China, and 1 paper was from Pakistan, involving 1084 CAD cases and 1537 controls. Among these seven studies, there were 12 II genotypes and 972 DD genotypes in the CAD group, and there were 15 II genotypes and 1382 DD genotypes in the control group (Table 3). Taking II as the exposure factor and DD as the nonexposure factor, a fixed effects model was used to combine the data. The results showed that the 844ins68 II genotype was not associated with the risk of CAD (II vs. DD: OR = 1.10, 95% CI: 0.51 ~ 2.39). There were 112 II+ID genotypes and 972 DD genotypes in the CAD group, and there were 155 II+ID genotypes and 1382 DD genotypes in the control group. Taking II+ID as the exposure factor and DD as the nonexposure factor, a fixed effects model was used to combine the data. The results showed that the 844ins68 II+ID genotype was not associated with susceptibility to CAD (II+ID vs. DD: OR = 1.04, 95% CI: 0.80 ~ 1.35). There were 12 II genotypes and 1072 ID+DD genotypes in the CAD group, and there were 15 II genotypes and 1522 ID+DD genotypes in the control group. Taking II as the exposure factor and ID+DD as the nonexposure factor, a fixed effects model was used to combine the data. The results showed that the 844ins68 II genotype was not associated with susceptibility to CAD (II vs. ID+DD: OR = 1.09, 95% CI: 0.51 ~ 2.36).

Overall, the results showed that there was no association between the CBS G919A gene polymorphism and the risk of CAD under the homozygous model (II vs. DD: OR = 1.10, 95% CI: 0.51 ~ 2.39), the dominant model (II+ID vs. DD: OR = 1.04, 95% CI: 0.80 ~ 1.35) or the recessive model (II vs. ID+DD: OR = 1.09, 95% CI: 0.51 ~ 2.36). (Table 4，Figure S2)

Subgroup analysis results

We performed a subgroup analysis of the sample size, diagnosis methods and ethnicities included in the studies, and we found a significant association between T833C gene polymorphism and CAD in Chinese subjects but not in Americans under the homozygous, dominant and recessive genetic models (CC vs. TT: OR = 2.38, 95% CI: 1.38 ~ 4.10) (CC+CT vs. TT: OR = 2.12, 95% CI: 1.56 ~ 2.87) (CC vs. CT+TT: OR = 2, 95% CI: 1.17 ~ 3.43). In addition, a significant association between the CBS T833C gene polymorphism and the risk of CAD was observed in a large sample (CC vs. TT: OR = 1.93, 95% CI: 1.16 ~ 3.21) (CC+CT vs. TT: OR = 1.6, 95% CI: 1.21 ~ 2.10) (CC vs. CT+TT: OR = 1.69, 95% CI: 1.02 ~ 2.79) (Table 5). We also performed subgroup analysis on G919A and 844ins68, but there was no association between the G919A or 844ins68 gene polymorphism and CAD.

Table 5. Meta-analysis of the association between CBS T833C polymorphism and risk of CAD.

Indexes		No. of studies	Homozygote model (CC vs. TT)			Dominant model (CC+CT vs. TT)			Recessive model (CC vs. CT+TT)
			OR	95%CI	p value	OR	95%CI	p value	OR	95%CI	p value
Sample size	Large(>250)	3	1.93	(1.16，3.21)	0.01	1.6	(1.21，2.10)	0.0009	1.69	(1.02，2.79)	0.04
	Small(≤250)	1	5.31	(1.11，25.38)	0.04	3.76	(1.78，7.93)	0.0005	4.32	(0.91，20.50)	0.07
Diagnosis methods	CAA	2	2.08	(1.16,3.73)	0.01	1.86	(1.33,2.61)	0.0003	1.76	(0.99,3.15)	0.06
	Others	2	2.39	(1.04,5.52)	0.04	1.68	(1.13,2.51)	0.01	2.16	(0.94,4.97)	0.07
Ethnicity	Chinese	3	2.38	(1.38,4.10)	0.002	2.12	(1.56,2.87)	<0.00001	2	(1.17,3.43)	0.01
	American	1	1.48	(0.51，4.26)	0.47	1.13	(0.69，1.86)	0.62	1.44	(0.50，4.08)	0.5

Publication bias

The Begg’s funnel plot and Egger’s linear regression test were performed to assess publication biases. As shown in Figure S3 and Table 6, this bias was observed under two genetic models (G919A’s Homozygote model and 844ins68’s Dominant model), the other showed no publication bias. Therefore, there was no strong statistical evidence of publication bias.

Table 6. The effect of publication bias was assessed by Begg’s and Egger’s linear regression tests.

Polymorphism	Model	Test of Begg(P value)	Test of Egger(P value)		Figure
T833C	Homozygote model (CC vs. TT)	0.308	0.091	FIG.S3-A
	Dominant model (CC+CT vs. TT)	0.734	0.311	FIG.S3-B
	Recessive model(CC vs. CT+TT)	0.308	0.069	FIG.S3-C
G919A	Homozygote model (AA vs. GG)	0.296	0.027	FIG.S3-D
	Dominant model (AA+AG vs. GG)	0.296	0.118	FIG.S3-E
	Recessive model (AA vs. AG+GG)	0.296	0.054	FIG.S3-F
844ins68	Homozygote model (II vs. DD)	0.826	0.779	FIG.S3-G
	Dominant model (II+ID vs. DD)	0.035	0.031	FIG.S3-H
	Recessive model (II vs. ID+DD)	0.806	0.747	FIG.S3-I

Discussion

We performed a meta-analysis involving 14 case-control studies comprising 2123 cases and 2368 controls to comprehensively evaluate the effect of CBS gene polymorphism on CAD susceptibility. The results of our studies suggested that the CBS T833C gene polymorphism might be associated with the risk of CAD. There was no significant relationship between the CBS G919A and 844ins68 gene polymorphisms and the risk of CAD in the three genetic models examined in our study.

In the subgroup analysis of ethnicity, we found a significant association with Chinese ethnicity under the homozygous, dominant and recessive genetic models (OR = 2.38, 95% CI: 1.38 ~ 4.10) (OR = 2.12, 95% CI: 1.56 ~ 2.87) (OR = 2, 95% CI: 1.17 ~ 3.43). The subgroup analysis results suggested that CBS T833C may be an ethnicity-dependent factor in CAD progression. In addition, a significant association between the CBS T833C gene polymorphism and the risk of CAD was observed in a large sample under a homozygous genetic model (OR = 1.93, 95% CI: 1.16 ~ 3.21). Four studies on the CBS T833C gene polymorphism were included in the analyses, but the subjects of all but one of these studies were Chinese, while the fourth was conducted in American subjects. Based on the large sample, the results are likely to be more accurate and more credible.

According to some related studies, several possible molecular mechanisms can explain the above conclusion. T833C refers to a T to C transition at nucleotide position 833 in exon 8, causing a substitution of threonine for isoleucine at codon 278 (I278T). This mutation may change the conformation of the pyridoxal 5’-phosphate (coenzyme) binding site in CBS to affect the binding of the coenzyme to the enzyme, resulting in low CBS enzyme activity and high plasma homocysteine levels, leading to CAD^14,33. Many previous studies have resulted in similar conclusions, Nienaber’s research demonstrated that CBS T833C had no significant impact on homocysteine levels among black South Africans³⁴. Nan’s study³⁵, on the other hand, revealed that individuals with homozygous and heterozygous mutant genotypes of CBS T833C had higher homocysteine levels compared to healthy individuals with wild genotypes, specifically in patients with cerebral arterial thrombosis. Additionally, for children with idiopathic mental retardation, those with the wild homozygous CBS T833C genotype had higher homocysteine levels compared to those with the heterozygous mutant genotype³⁶. Furthermore, Chang’s study illustrated that individuals carrying the CBS T833C allele, specifically those with the CT and CC genotypes, had a higher prevalence compared to individuals with the TT genotype³⁷. Furthermore, the identification of the T833C mutation in the CBS gene revealed a notable disparity in both genotype and allele frequency distribution when compared to the control group. These findings provide evidence to suggest that the T833C mutation in the CBS gene, which is associated with elevated homocysteine levels, may serve as a significant genetic risk factor for cerebral infarction³⁸. In a case-control study conducted by Song, it was observed that plasma homocysteine levels varied across different genotypes of the CBS gene T833C in a cohort of 570 patients with ischemic cerebrovascular disease and healthy individuals³⁹, their results are consistent with our results. G919A is a G to A transition at nucleotide position 919 in exon 8 of CBS, resulting in the substitution of serine for glycine at codon 307 (G307S). This mutation may be located at the homocysteine binding site, which makes it difficult for the CBS enzyme to bind homocysteine, causing a high level of plasma homocysteine¹⁴. It is generally believed that the G919A mutation leads to low activity of CBS and hypercysteinemia, but the results of our meta-analysis do not support this conclusion, and studies with a large sample size still need to be carried out. The 844ins68 variant is a 68-bp insertion consisting of an exact duplication at the intron-exon boundary of exon 8, and it is a common mutation among normal individuals. Since this mutation produces a new splice site, it may lead to the introduction of another splicing method that still forms the wild-type sequence of CBS. The mRNA size is still correct and does not affect transcription and translation, so CBS is not affected by this mutation. The activity and amount of the CBS enzyme remain normal, which will not lead to a high plasma homocysteine level or CAD⁴⁰. The results of studies on 844ins68 are different, with many studies showing that 844ins68 is a protective factor for CAD, which is not consistent with the results of our study; therefore, additional studies will be needed to reach more convincing conclusions.

In recent years, there have been many studies on homocysteine metabolic enzyme gene polymorphisms and CAD susceptibility. A study by Singh and Lele⁴¹ indicated MTRR and BHMT gene polymorphisms may not be associated with CAD, while Chen⁴² showed that the MTR A2756G gene polymorphism is associated with the risk of CAD in a European population. Our study shows that the CBS T833C gene polymorphism is associated with CAD. The family aggregation phenomenon of CAD and the observed racial differences suggest that genes play an important role in the risk of CAD, and the prevalence of CAD differs widely among different populations. These studies provide objective medical evidence for the initial identification of individuals susceptible to CAD and suggest comprehensive diagnostic and prevention measures. In the future, we can use technology such as gene chip or gene sequencing approaches to help identify groups showing a high risk for CAD and perform early interventions to reduce the burden of CAD patients on society.

This study has some limitations. It involved a meta-analysis of only 14 case-control studies, so the sample size was relatively small, and false positive phenomena may have occurred. Most of the included studies did not provide detailed basic information that could be used for subgroup analysis, so additional factors affecting the prevalence of CAD were not accurately identified⁴³. In the 14 included studies, 13 consisted of Asian subjects, and only one consisted of an American population; therefore, no more accurate ethnic correlation could be performed. In addition, there was a difference in the severity of coronary artery disease among the patients in the control group. Therefore, additional well-designed studies are needed to further verify the association of CBS polymorphism with the risk of CAD, which may be helpful for formulating individualized prevention and treatment strategies for CAD in the future⁴⁴.

Conclusion

In conclusion, the CBS T833C gene polymorphism is significantly associated with the risk of CAD and shows a stronger association in Asian populations. Individuals with the C allele of the CBS gene T833C polymorphism might be particularly susceptible to CAD. Neither CBS G919A nor 844ins68 present an association with CAD in the three applied models. Considering the limitations indicated above, more studies with larger sample sizes are needed to definitively determine the association between the CBS T833C, G919A and 844ins68 gene polymorphisms and the risk of CAD.

Author’ contributions

Conceptualization: ZJ Z, YH G, LM.

Data curation: ZJ Z, YH G, LM, YM M, KY Y.

Formal analysis: RT C, JL L, KY Y.

Fundingacquisition: ZJZ.

Investigation: ZJ Z, YH G, LM, RT C, JL L.

Methodology: ZJ Z, YH G.

Supervision: ZJ Z.

Writing – original draft: ZJ Z, YHG.

Writing – review& editing: ZJ Z, YH G, LM.

Acknowledgments: ZJ Z, YH G, LM, KY Y, YM M, RT C, JL L.

Consent for publication

All data relevant to the study are included in the article or uploaded as online supplementary information.

Ethics approval and consent to participate

Ethical approval was not necessary for this work.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/10641963.2024.2328147

Additional information

Funding

This work was supported by a grant from the Natural Science Foundation of Gansu Province (1606RJZA168).