Maternal Methylenetetrahydrofolate Reductase C677T Polymorphism and Down Syndrome Risk: A Meta-Analysis from 34 Studies

Background Methylenetetrahydrofolate reductase (MTHFR) is a key enzyme of folate metabolic pathway which catalyzes the irreversible conversion of 5, 10-methylenetetrahydrofolate to 5-methyltetrahydrofolate. 5-methyltetrahydrofolate donates methyl group for the methylation of homocysteine to methionine. Several studies have investigated maternal MTHFR C677T polymorphism as a risk factor for DS, but the results were controversial and inconclusive. To come into a conclusive estimate, authors performed a meta-analysis. Aim A meta-analysis of published case control studies was performed to investigate the association between maternal MTHFR C677T polymorphism and Down syndrome. Methods PubMed, Google Scholar, Elsevier, Springer Link databases were searched to select the eligible case control studies using appropriate keywords. The pooled odds ratio (OR) with 95%confidence interval were calculated for risk assessment. Results Thirty four studies with 3,098 DS case mothers and 4,852 control mothers were included in the present meta-analysis. The pooled OR was estimated under five genetic models and significant association was found between maternal MTHFR 677C>T polymorphism and Down syndrome under four genetic models except recessive model (for T vs. C, OR = 1.26, 95% CI = 1.09–1.46, p = 0.001; for TT vs. CC, OR = 1.49, 95% CI = 1.13–1.97, p = 0.008; for CT vs. CC, OR = 1.29, 95% CI = 1.10–1.51, p = 0.001; for TT+CT vs. CC, OR = 1.35, 95% CI = 1.13–1.60, p = 0.0008; for TT vs. CT+CC, OR = 0.76, 95% CI = 0.60–0.94, p = 0.01). Conclusion The results of the present meta-analysis support that maternal MTHFR C677T polymorphism is a risk factor for DS- affected pregnancy.


Introduction
Down syndrome (DS) is the most common chromosomal disorder with the prevalence of 1/700-1000 live birth. It is characterized by the trisomy 21, which results from maternal meiotic nondisjunction in majority (90%) of cases. The established risk factor for DS is advanced (.35 years) maternal age at the time of conception. However, a fairly high number of DS children born to younger mothers suggest that risk factors other than advanced maternal age might be involved in predisposing younger mothers to DS-affected pregnancy [1,2]. The molecular and biochemical mechanism of maternal meiotic non-disjunction is still not known. James et al. [3] reported that methylenetetrahydrofolate reductase (MTHFR) C677T polymorphism might be a risk factor for maternal meiotic non-disjunction. Since then several studies have investigated the risk of DS to variants of folate pathway genes like MTHFR, Methionine synthase (MTR) and Methionine synthase reductase (MTRR) in Asian [1,2,4,5] and Caucasian [6][7][8] populations. Folate deficiency and dysfunctional MTHFR causes abnormal DNA methylation [9,10] and chromosomal segregation [11,12]. Hypomethylation of the centromeric DNA has been suggested as the causative mechanism of meiotic non-disjunction. Abnormal DNA methylation of centromere lead to aberrant kinetochore formation that results into abnormal segregation of chromosomes during meiosis [3,13].
MTHFR is a key enzyme in folate metabolism, which catalyzes the reduction of 5, 10-methylenetetrahydrofolate to the predominant circulating form of folate i.e. 5-methyltetrahydrofolate (5-THF). 5-THF donates methyl group for the conversion of homocysteine to methionine, which is further converted into Sadenosylmethionine (SAM). SAM is the main methyl group donor for all cellular methylation reactions. Folate deficiency and/or dysfunctional MTHFR reduces the conversion of 5, 10-methylene THF to 5-methyl THF, and elevates plasma homocysteine concentration. Both folate and MTHFR are involved in many complex biochemical reactions like DNA synthesis, repair and methylation.
There are more than 40 polymorphisms reported in MTHFR gene and among them C677T variant is the most studied and clinically important. The C677T variant (rs 1801133; Ala 222 Val) has been associated with a decreased activity of MTHFR, and increased homocysteine level [14][15][16]. Mutant homozygous (TT) individuals have a decreased enzymatic activity , 70% and the heterozygote by 40%. A dysfunctional MTHFR leads to lower levels of SAM resulting into DNA hypomethylation. DNA hypomethylation increases the risk of many diseases and disorders like-neural tube defects [17], cleft lip and palate [18], Alzheimer disease [19], cardiovascular diseases [14], diabetes [20] and psychiatric disorders [21] etc. Several epidemiological studies have investigated the associations of the maternal MTHFR C677T polymorphism with Down syndrome. However, the results were conflicting and inconclusive. In light of the above facts, we conducted a meta-analysis of published case control studies relating the C677T polymorphism of the maternal MTHFR gene to the risk of having DS offspring.

Selection of studies
Electronic searches were conducted using PubMed, Google Scholar, Elsevier and Springer link and all published manuscripts up to January, 2014 were considered in present meta-analysis. The following index terms were used for search 'MTHFR' 'Methylenetetrahydrofolate reductase', and 'C677T polymorphism', 'maternal risk' and 'Down syndrome'. In addition, bibliographies of all articles and reviews were hand searched for additional suitable studies.

Inclusion criteria
Included studies had to meet the following criteria: (1) article should be published; (2) article should have sufficient data to calculate the odds ratio with 95% CI; (3) article should be case control association study; and (4) author should describe the genotyping protocols.

Data extraction
The following data were extracted from each study: first author's name, publication year, journal name, country name, genotyping method, and different MTHFR genotype numbers.

Meta-analysis
Statistical analysis of maternal MTHFR C677T polymorphism and DS risk was estimated by Odds ratio (ORs) with 95% confidence intervals (CIs). The heterogeneity was tested by the Qstatistics with p-values ,0.05. Subgroup analysis was done to know the source of heterogeneity. If higher heterogeneity (I 2 . 50%) would be observed, the random effect model [22] would be applied. Otherwise, fixed-effect model [23] was applied to obtain the summary OR and 95% CI. All p values were two-sided and a p value of less than 0.05 was considered statistically significant. All analyses were performed using the computer program MIX version 1.7 [24]. The control genotypes were tested for Hardy-Weinberg equilibrium (HWE) using the Goodness of fit Chisquare test. The quality of the included studies was measured according to the scoring system for randomized controlled association studies proposed by Clark and Baudouin [25]. Case control studies scoring ,5 were defined as low quality study and those $5 were defined as high quality study.

Publication bias
Funnel plots of precision by log (OR) and standard error by log (OR) were plotted to determine publication bias and asymmetrical funnel plots represent publication bias. Begg and Mazumdar rank correlation [26] and Egger's regression intercept [27] tests were adopted to assess the publication bias.
In all thirty four studies, total cases were 3,098 with CC (1,396), CT (1,326) and TT (376), and controls were 4,852 with CC (2,329), CT (2,015), and TT (508) genotypes. In controls genotypes, percentage of CC, CT and TT were 48.00%, 41.53%, and 10.47% respectively. In total cases, genotype percentage of CC, CT, and TT was 45.06%, 42.8% and 12.14% respectively. Frequencies of CC and CT genotypes were highest in both cases and controls (Table 2). In cases and controls, the allele C was the most common. All five genetic models; -allele contrast (T vs C) homozygote (TT vs CC), codominant (CT vs CC), dominant (TT+CT vs CC) and recessive (TT vs CT+CC) models were used to evaluate C677T polymorphism as DS risk.

Meta-analysis
Meta-analysis with allele contrast showed significant association between maternal 677T allele and DS with both fixed effect (OR TvsC = 1.22; 95% CI = 1.13-1.31; p = ,0.0001) and random effect models (OR TvsC = 1.26; 95% CI = 1.09-1.45; p = 0.001) ( Figure 2) (Table 3). In cumulative meta-analysis using random effect model, the association of maternal T allele with DS turned statistically significant with the addition of study of Wang et al. (2008) and remained significant thereafter. Table 3 summarizes the ORs with corresponding 95% CIs for association between maternal C677T polymorphism and risk of DS in dominant, recessive, homozygote and co-dominant models. With our primary analysis, there was an increased risk of DS among mutant homozygote variants (TT), with both fixed (OR TTvs.CC = 1.44; 95% CI = 1.2221.69, p = ,0.0001) and ran-

Discussion
In 1999, James et al [3] reported that genetic polymorphism of folate and homocysteine pathway enzymes predispose a woman to abnormal chromosome segregation, which act as risk factor for DS pregnancy. In subsequent years, several in vivo studies in humans suggested that chronic folate deficiency has been associated with abnormal DNA methylation [11,53,54], and aberrant chromosome segregation [6,55259]. Population-based studies have shown that folic acid intake during fetal development has a protective effect, resulting in a significant reduction in the   occurrence of developmental defects, like neural tube defects (NTD), congenital heart defects, limb defects, and orofacial clefts [60].
Meta-analysis is a powerful tool for analyzing cumulative data with small and low power studies. Several meta-analyses were published accessing MTHFR as risk factor to various diseases/  disorders like-neural tube defects [61,62], cleft lip and palate [63], stroke [64], psychiatric disorders [65]. During literature search, we identified four meta-analyses [66][67][68][69] published between 2007 and 2013. They examined the effect of maternal MTHFR C677T as DS risk, but no consistent conclusion was achieved. Zintzaras [66] performed a meta-analysis based on eleven studies and did not find any significant association between the maternal MTHFR polymorphisms and DS risk. Medica et al. [67] aggregated sixteen studies and reported significant relationship between the maternal mutant genotypes (TT+CT vs CC) and risk of DS child. Recently, Wu et al. [68] published a meta-analysis (included twenty eight studies with 2806 cases/4597 controls), and found statistical association with dominant model (OR = 1.305, 95% CI = 1.125-1.514, p = 0, p = 0.003). Yang et al. [69] performed a metaanalysis which was based on twenty six studies (2458 cases/3144 controls) and found statistically significant association in allele contrast model (OR = 1.28; 95% CI: 1.11-1.47) ( Table 4). Several newly published studies were not included in the previous published meta-analyses. So authors conducted a comprehensive meta-analysis with the largest number of studies (34 studies). In the present meta-analysis significant association was found between maternal C677T polymorphism and DS risk in total 34 studies using all five genetic models. Whereas in stratified analysis, except allele contrast model in American population, no significant association was observed in European and American population but significant higher risk was found in Asian population. Such phenomenon probably could be ascribed to the folate metabolism profile and dietary structure of different regions.
There are few limitations of the present meta-analysis like-i) we used crude ORs in the pooled analysis without adjustment; ii) the relatively small sample size in some of the included studies, especially those from Asia; iii) we considered only one gene polymorphism (MTHFR C677T) of folate pathway. Present metaanalysis had several advantages/strength to the previous published meta-analyses like-(i) the publication bias was not detected in present meta-analysis, (ii) pooled number of cases and controls from different studies significantly increased the statistical power of the analysis, (iii) largest number of studies (34 studies) with largest sample size (3,098 cases and 4,852 controls) was included in the present meta-analysis, (iv) controls included in the present metaanalysis was mothers of healthy child, (v) distribution of genotypes in control mothers except two studies was in Hardy-Weinberg equilibrium, (vi) significant association was found between maternal MTHFR C677T polymorphism and DS risk in allelic contrast, homozygote, co-dominant and dominant genetic models and (vii) in addition we did sub-group analysis according to geographical regions.
In conclusion, results of present meta-analysis suggest that the maternal MTHFR 677T allele is a risk factor for development of DS pregnancy. However the results of present meta-analysis were based on single gene polymorphism and significant heterogeneity was also observed; hence results should be interpreted with caution.

Supporting Information
Checklist S1 PRISMA checklist.