Methylenetetrahydrofolate Reductase Gene rs1801133 and rs1801131 Polymorphisms and Essential Hypertension Risk: A Comprehensive Analysis

Background Essential hypertension (EH) is a common and multifactorial disorder that is likely to be influenced by multiple genes. The methylenetetrahydrofolate reductase (MTHFR) gene rs1801133 and rs1801131 polymorphisms influence MTHFR enzyme activity and plasma homocysteine concentration. In addition, variations in MTHFR functions likely play roles in the etiology of EH. Thus far, a large number of studies investigating the associations between the MTHFR polymorphisms and EH have provided controversial or inconclusive results. To better assess the purported relationship, we performed a comprehensive analysis of 52 published studies. Objective and Methods. Eligible studies were identified by searching the PubMed, Wanfang, and China National Knowledge Infrastructure (CNKI) databases. Odds ratios (ORs) with 95% confidence intervals (CIs) were estimated to assess the potential association between the MTHFR rs1801133 polymorphism and EH. Results Overall, 10712 patients and 11916 controls were involved; we observed significantly increased association between the MTHFR rs1801133 polymorphism and EH risk (such as T vs. C: OR = 1.38, 95% CI = 1.25 − 1.54, P ≤ 0.001), with similar results evident within race subgroups (such as Asian: T vs. C: OR = 1.47, 95% CI = 1.30 − 1.67, P ≤ 0.001; compared to Chinese: T vs. C: OR = 1.54, 95% CI = 1.33 − 1.79, P ≤ 0.001). Similar associations were also found in subgroups defined by the source of controls and genotype methods. To our regret, based on the limited studies, no association was detected for rs1801131 polymorphism. Conclusions Our study provides evidence that the MTHFR rs1801133 null genotype may increase EH risk. Future studies with larger sample sizes are warranted to evaluate this association in more detail.


Introduction
Essential hypertension (EH) has a high prevalence rate worldwide and is considered to derive from complex interactions between diverse genes and environmental conditions [1,2]. The present evidence-based treatment of EH is a critical intervention in reducing cardiovascular (CV) morbidity and mortality [3]. A contemporary meta-analysis of 123 studies with 613,815 hypertensive participants showed that, for every 10 mmHg reduction in systolic blood pressure, there is a significantly decreased risk of major CV disease events (relative risk 0.80 and 95% confidence interval (CI) 0.77-0.83), coronary heart disease (0.83 and 0.78-0.88), stroke (0.73 and 0.68-0.77), and heart failure (0.72 and 0.67-0.78) [4]. EH accounts for 90%-95% of all patients with hypertension, and approximately 20%-60% of its etiology is influenced by genetic factors [5].
Most people with hypertension do not present with typical symptoms, making the condition easy to ignore and reducing the opportunities for early medical intervention. Therefore, the identification of high-risk patients is particularly meaningful, since identifying such patients with high blood pressure as early as possible allows for timely monitoring of blood pressure, regular follow-ups, and options for improving unhealthy lifestyles. In known high-risk patients, rising blood pressure can be immediately detected and treatment can be provided timeously, thus avoiding complications and, ultimately, reducing the incidence of hypertension and improving quality of life [3,6,7].
Studies have shown that a high plasma concentration of homocysteine (Hcy) may injure the vascular endothelium, which results in hypertension and a predisposition toward atherosclerosis. Elevated Hcy levels have been identified as an independent risk factor for hypertension [8][9][10][11]. Methylenetetrahydrofolate reductase (MTHFR) plays an important role in the metabolism of Hcy. The rs1801133 polymorphism of the MTHFR gene is a C to T transition at nucleotide position 677 (C667T) in exon 4, which results in a change from alanine to valine at amino acid 222. This mutation may lead to a decrease in MTHFR activity and heat tolerance and thus metabolic damage of Hcy, which may moderately increase plasma Hcy levels [12,13]; moreover, 1298 (A to C) also may cause a significant reduction in enzyme activity. This information suggests that MTHFR two polymorphisms may be related to EH development and susceptibility. It is therefore worthwhile to demonstrate whether there is an association between this polymorphism and EH risk, as it may provide guidance for the prevention and diagnosis of EH in the clinic. Thus far, numerous studies have reported the association between the MTHFR two com-mon polymorphisms and EH risk. We therefore performed a comprehensive analysis of 52 different case-control studies to derive a convincing conclusion regarding this apparent association [14].

Identification of Eligible Studies.
Searches were performed in the PubMed, Wanfang, and China National Knowledge Infrastructure (CNKI) databases (updated on Dec. 10, 2021) using the following related keywords: polymorphism/variant/mutation, hypertension/essential hypertension, and MTHFR/methylenetetrahydrofolate reductase. We included all studies that described a relationship between the MTHFR two polymorphisms and EH susceptibility. All of the included studies met the following criteria: (1) association between MTHFR two polymorphisms and EH risk; (2) case-control study; (3) each genotype frequency is shown in tables; and (4) genotype distributions of the control were consistent, with a Hardy-Weinberg equilibrium (HWE) more than 0.05. Otherwise, studies should be excluded with the following issues: (1) no control, (2) incomplete genotype frequency data, (3) duplication studies, and (4) not according with HWE in control groups.

Quality Assessment.
In this meta-analysis, the quality was assessed using the Newcastle-Ottawa Scale (NOS) for cross-sectional study quality assessment. The methodological quality of each study (sampling strategy, response rate, and representativeness of the study), comparability, and outcome were checked using the NOS tool. Studies with a score of more than 7 out of 10 were considered as good quality. This cut-off point was determined after reviewing relevant meta-analyses from the literature [15][16][17]. Moreover, to assess the quality of each qualified articles by quality score to explore other potential sources of heterogeneity, another score of quality assessment was applied. The quality scores of the studies ranged from 0 (lowest) to 15 (highest). Studies with scores ≤ 9 were categorized into low quality, while those with scores > 9 were considered as high quality [18].

Statistical Analysis.
The extracted data were imported into the Stata software program (version 10.0, Corporation, College Station, Texas) for analysis. Odds ratios (ORs) with 95% CIs were used to measure the strength of the association [19,20]. The subgroup analysis stratified by race was performed first. Race was categorized as European, Asian, Mixed, Chinese, and non-Chinese (all people who are not Chinese) subtypes. The source of the control subgroups was defined based on two classifications: hospital-based (HB) and population-based (PB). For MTHFR rs1801133 and rs1801131, we investigated the relationship between genetic variants and EH risk in five different models (T-allele vs. C-allele or A-allele vs. C-allele, TT vs. CC or AA vs. CC, TC vs. CC or AC vs. CC, TT+TC vs. CC or AA+AC vs. CC, and TT vs. TC+CC or AA vs. AC+CC).
The evaluation of heterogeneity within the included studies was assessed using Cochrane's Q test (chi-square) and I 2 (%) statistical analysis. A fixed effect model was applied when the effects were assumed to be homogenous (P > 0:05, I 2 ≤ 50%); otherwise, the random-effect model was adopted (P < 0:05, I 2 ≥ 50%) [21,22]. When heteroge-neity was observed, the source of the heterogeneity was explored via subgroup analysis, performed using the ethnicity, publication year, study design, and genotype methods.
The presence of potential publication bias was determined using the Egger/Begg's test and presented graphically in the form of a funnel plot [23]. In addition, the departure of MTHFR polymorphism frequencies from expectations under HWE were assessed in controls using the Pearson chi-square test [24]. Another sensitivity analysis was conducted to assess the stability of the results. Finally, the power and sample size analysis of our meta-analysis was calculated using a program called PS: Power and Sample Size Calculation (http://biostat.mc.vanderbilt.edu/wiki/Main/ PowerSampleSize#Windows) [25].

Genotyping
Methods. Genotyping to identify single nucleotide polymorphisms (SNPs) in the MTHFR gene was conducted using the following analyses: polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP), TaqMan, sequencing, PCR, amplification refractory mutation system-PCR (ARMS-PCR), and high-resolution melt (HRM) genotyping.
2.6. Metaregression. Random-effect metaregression analysis was performed to define the source of publication bias, with the subgroups of publication year, ethnicity, source of control, and genotype method set as independent variables and the log values considered as dependent variables [26].

Gene Interaction Network Analysis of the MTHFR Gene.
In order to fully understand the role of MTHFR and its potential functional partners in EH, we used the STRING online server (http://string-db.org/) to construct a MTHFR gene-gene interaction network [27,28].

Study Selection and Characteristics in our Meta-analysis.
Using appropriate keywords (see Materials and Methods), we identified 364 articles from PubMed, 56 from CNKI, and 362 from the Wanfang database. In total, 552 articles were excluded after review of the abstract, leaving 230 articles for full article evaluation. Among them, 39 articles featured systematic analysis/meta-analysis/review; 24 covered only case T o t a l E u r o p e a n A f r i c a n A f r i c a n A m e r i c a n A s i a n E a s t A s i a n S o u t h A s i a n MAF of A-allele from 1000 Genomes Ethnicity (a) T o t a l E u r o p e a n A f r i c a n A f r i c a n A m e r i c a n A s i a n E a s t A s i a n S o u t h A s i a n     Figure 1). After exclusion of the above studies by full article review, we were left with 55 articles covering 60 case-control studies, of which 8 case-control studies were not consistent with HWE and were excluded. In total, 49 case-control studies about rs1801133 and 3 case-control studies about rs1801131 were included in our current analysis. All essential information is listed in Table 1, including first author, publishing year, race, the numbers of cases and controls, HWE, genotype numbers in cases/controls, study design, and genotype methods. Our study comprised 9 European, 36 Asian, and 4 Mixed-race case-control studies. The T frequency was 43.59% in the Asian group, 35.2% among Europeans, and 47.7% in the Mixed-race group, which indicated that the European group had lower frequency of Tallele than the Asian and Mixed groups (P < 0:05). The distribution of genotypes in all the controls was in agreement with HWE. In addition, we confirmed the minor allele frequency (MAF) reported for the seven main worldwide populations     (Figure 2(a)). When we examined the frequency of T-and C-alleles both in the case and control groups, our analyses showed similar allele frequencies in each group (Figure 2(b)). Finally, we used The Cancer Genome Atlas (TCGA) database to search for trends in the frequency of rs1801133 polymorphisms. Our results indicated that the TT (AA) frequency was relatively low compared to other genotypes (Figure 2(c)). This polymorphism is associated with the coronary artery, rather than the aorta artery left ventricle and the tibial artery (https://www.gtexportal.org/home/ ) (Figure 2(d)). We also showed the corresponding information about rs1801131 polymorphism (Figures 3(a)-3(d)). Table 2 shows the summary of odds ratios of MTHFR, based on 10,712 EH cases and 11,916 matched controls. We observed an increased association between the MTHFR rs1801133 polymorphism and EH in total population groups (for example, T-allele vs. C-allele: OR = 1:38, 95% CI = 1:25 − 1:54, P h < 0:001, P < 0:001, I 2 = 81:3%, Figure 4). Similar trends were observed   Figure 4; OR = 1:28, 95% CI = 1:05 − 1:57, P h = 0:004, P = 0:014, I 2 = 64:2% for Europeans, Figure 4; OR = 1:54, 95% CI = 1:33 − 1:79, P h < 0:001, P < 0:001, I 2 = 85:6% for Chinese, Figure 5; and OR = 1:20, 95% CI = 1:06 − 1:37, P h < 0:001, P = 0:004, I 2 = 66:3% for non-Chinese). In order to analyze the source of controls and determine the source of heterogeneity, the ratios were calculated for the HB and PB subgroups. The results showed significantly increased relationships in these groups (T-allele vs. C-allele: OR = 1:49, 95% CI = 1:28 − 1:75, P h < 0:001, P < 0:001, I 2 = 85:4% for HB and OR = 1:29, 95% CI = 1:13 − 1:47, P h < 0:001, P < 0:001, I 2 = 73:3% for PB) ( Figure 6). As different methods for detecting this polymorphism were applied in all of the included studies, we considered whether positive results were associated with particular genotyping methods. Our analyses revealed some significant findings, such as PCR (T-allele vs. C-allele: OR = 1:51, 95% CI = 1:14 − 2:01, P h < 0:001, P = 0:004, I 2 = 86:1%), PCR-RFLP (T-allele vs. C-allele: OR = 1:47, 95% CI = 1:29 − 1:68, P h < 0:001, P < 0:001, I 2 = 64:3%, Figure 7), and HRM analyses (T-allele vs. C-allele: OR = 1:32, 95% CI = 1:15 − 1:51, P h < 0:001, P < 0:001, I 2 = 47:5%, Figure 8). To our regret,  10 Cardiovascular Therapeutics no positive associations were observed for rs1801131 polymorphism (Table 2).

Publication Bias and Sensitivity Analysis.
Begg's funnel plot and Egger's test were performed to determine the publication bias of the included studies. Significant obvious evidence of publication bias was detected in five genetic model analyses (such as Figures 9(a) and 9(b) regarding T-allele vs. C-allele) ( Table 3).
To remove studies which may influence the power and stability of the current meta-analysis, sensitivity analysis was applied, but no sensitive case-control studies were found for this SNP among the above five models (such as Figure 9(c) regarding T-allele vs. C-allele).

Metaregression.
The metaregression analysis indicated that there is a significant relationship for the allele model (T-allele vs. C-allele) with respect to ethnicity, source of control, and genotype methods, with a regression coefficient of 0.001, 0.004, 0.010, and 0.002, respectively. There was no association with publication year, which suggests that the heterogeneity from the rs1801133 polymorphism in EH  11 Cardiovascular Therapeutics may be due to the ethnicity, source of control, and genotype method subgroups (Figures 10(a)-10(e)).

Discussion
While the precise causes of hypertension are still unknown, the risk factors include genetic factors, age, and unhealthy lifestyle practices, with 70-80% of hypertensive cases resulting from unhealthy lifestyle practices. As the risk factors for high blood pressure accumulate, the risk of high blood pressure increases [30,31].
The detection of significant polymorphisms may be a suitable method to predict the risk of hypertension in susceptible individuals. Our current study focused on EH, a common type of hypertension, and included 10,712 patients with EH and 11,916 healthy individuals. In the overall analysis, we observed that individuals carrying TT or a T-allele may have an increased risk of developing EH compared to those with CC or a C-allele (between 37% and 89%). In other   words, individuals carrying T-allele or TT genotype may have more possibility to suffer from hypertension in a current or in future time. This can give these people warnings, such as regular changes in blood pressure, changing bad habits, moderating physical exercise, or early medical intervention. Significant heterogeneity was indicated in all of the genetic models. To determine the source of the variation, we analyzed the associations in other subgroups, such as ethnicity, source of control, and genotype method. In parallel, significant relationships were also observed with the ethnicity, source of control, and genotype method subgroups, which provided further support for hypothesis that the rs1801133 polymorphism is a risk factor for EH. In addition, when we used metaregression analysis to evaluate the source of heterogeneity, the aforementioned three subgroups emerged as significant sources of variation. The power of our study was 1, suggesting that our conclusions were accurate. Several meta-analyses regarding the rs1801133 polymorphism and hypertension have been published to date. For example, Wu et al. included 30 case-control studies, and their findings supported a role for the rs1801133 polymorphism in the risk of developing EH [32]. Kosmas et al. identified 23 comparisons relating to hypertensive disease in pregnancy, and they concluded that the T-allele of the rs1801133 polymorphism may increase the risk of hypertension in pregnancy by 1.21-fold [33]. Qian et al. combined 26 published studies of both hypertension and hypertension in pregnancy, suggesting that the rs1801133 polymorphism may be one independent risk factor [34]. Finally, Yang et al. performed a meta-analysis of 27 studies, including 5,418 EH and 4,997 controls, and their findings supported the evidence that carriers of the rs180113 T-allele were susceptible to EH [35]. However, the above studies were subject to some disadvantages. First, in several studies, the HWE values were not consistently above 0.05, which may have increased the heterogeneity and reduced the power of the conclusions. Second, each study omitted other related casecontrol studies, whereas our current study is a relative comprehensive analysis. Third, some articles did not distinguish the types of hypertension, which may have introduced variability because different kinds of hypertension have different etiologies, pathogenesis, and genetic backgrounds. For these reasons, we focused on one form of hypertension for our analysis. Fourth, our analysis increased genotype subgroup and evaluation of power. Fifth, we analyzed gene-gene interactions between MTHFR and related genes to elucidate potential functional interactions. Despite these disadvantages, it should be noted that the conclusions from our current study are similar to those from previous meta-analyses.
A key feature of our study was the identification of genegene interactions for MTHFR. The average scores of the ten most significant genes were more than 0.9. The top three genes that were suggested to interact with MTHFR were MTR (0.995), TYMS (0.992), and MTHFD1 (0.989). MTHFR and MTR both participate in homocysteine metabolism, regulating different stages of the process. MTHFR converts 5,10-methylene-THF into 5-methyl-THF, while MTR catalyzes the demethylation of 5-methyl-THF to THF and the remethylation of homocysteine to methionine [36,37]. The MTR 2756 A/G polymorphism is also associated with the risk of hypertension [38].
Some limitations in our meta-analysis should be considered. First, our analyses indicated that the heterogeneity identified in our study emerged from ethnicity, source of control, and genotype methods. Future studies should optimize the design of both retrospective and prospective research projects to overcome this deficiency. Second, EH is a complex disease including genetic and other factors (such as environment, diet, and concomitant disease) [39], and future studies should analyze the gene-gene or geneenvironment interactions using larger sample sizes. Third, further meta-analyses should cover all kinds of hypertension, analyzing the associations and determining the genetic background of each type individually. Fourth, the specific mechanism underpinning the impact of the rs1801133 or rs1801131 polymorphism should be explored, for example, using animal models of EH.

Conclusion
Our meta-analysis provides evidence that the MTHFR 677T null genotype is associated with increased risk of EH, suggesting that further, well-designed large studies are necessary to confirm this relationship. Furthermore, it will be important to focus on the mechanism of action of the rs1801133T-allele to explain the complete chain of evidence for the prevention of EH in the future.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.

Disclosure
We stated it has been presented as preprint in research square according to the following link: https://www .researchsquare.com/article/rs-176622/v1.

Conflicts of Interest
The authors declare no conflict of interest.