Search tips
Search criteria 


Logo of plosonePLoS OneView this ArticleSubmit to PLoSGet E-mail AlertsContact UsPublic Library of Science (PLoS)
PLoS One. 2013; 8(4): e59570.
Published online 2013 April 4. doi:  10.1371/journal.pone.0059570
PMCID: PMC3617174

Genetic Variants in the Folate Pathway and the Risk of Neural Tube Defects: A Meta-Analysis of the Published Literature

Osman El-Maarri, Editor



Neural Tube Defects (NTDs) are among the most prevalent and most severe congenital malformations worldwide. Polymorphisms in key genes involving the folate pathway have been reported to be associated with the risk of NTDs. However, the results from these published studies are conflicting. We surveyed the literature (1996–2011) and performed a comprehensive meta-analysis to provide empirical evidence on the association.

Methods and Findings

We investigated the effects of 5 genetic variants from 47 study populations, for a total of 85 case-control comparisons MTHFR C677T (42 studies; 4374 cases, 7232 controls), MTHFR A1298C (22 studies; 2602 cases, 4070 controls), MTR A2756G (9 studies; 843 cases, 1006 controls), MTRR A66G (8 studies; 703 cases, 1572 controls), and RFC-1 A80G (4 studies; 1107 cases, 1585 controls). We found a convincing evidence of dominant effects of MTHFR C677T (OR 1.23; 95%CI 1.07–1.42) and suggestive evidence of RFC-1 A80G (OR 1.55; 95%CI 1.24–1.92). However, we found no significant effects of MTHFR A1298C, MTR A2756G, MTRR A66G in risk of NTDs in dominant, recessive or in allelic models.


Our meta-analysis strongly suggested a significant association of the variant MTHFR C677T and a suggestive association of RFC-1 A80G with increased risk of NTDs. However, other variants involved in folate pathway do not demonstrate any evidence for a significant marginal association on susceptibility to NTDs.


Neural Tube Defects (NTDs) are congenital malformations of the brain and spinal cord in neurulation that occur between 21 and 28 days after conception [1]. The most common subtypes of cases include spina bifida, anencephaly and encephalocele. The disease is one of the most prevalent and most severe of birth defects with a high mortality rate [2]. As reported, the average worldwide prevalence is 1 per 1000 living birth [3], in Whites it is approximately the same [4], and in China, it accounts for 20% to 25% of birth defects. Previous research has revealed that the pathogenesis of NTDs is quite complex involving both environmental factors and genetic components.

Folic acid deficiency is relevant to the risk of the disease which was first demonstrated in seminal work done 36 years ago by Smithells et al. which showed that compared with the control group, women who had given birth to NTD children were significantly deficient in several micronutrients, especially folic acid, in diets and postpartum blood [5]. Following observations confirmed that folic acid fortification can prevent the disease to a large extent [6], [7], [8]. Moreover, folic acid supplement, investigated by Berry et al. [9], can prevent NTDs, reducing the incidence by 50–75% without any adverse effects of folic acid for the doses ranging from 0.36 mg (360 µg) to 4 mg (4000 µg) a day. As De-Regil’s described in his review, it can efficiently decrease not only the first occurrence but also the recurrence of the disease among offspring in NTDs families of which parents have had an affected pregnancy [10], [11].

Emerging views of the evidence have begun to shed light on pathogenic mechanisms. One assumption is that folate transport may be affected by immunological responses and maternal autoantibodies that bind to the folate receptor can block the intracellular uptake of folate might cause NTDs [12]. Later studies support that altered folate metabolism contributes to abnormal development of neural system may involve in the etiology of NTDs that reaffirmed the association between the folic acid and the disease. [13], [14].

Folic acid must first convert to its naturally bioactive formtetrahydrofolate (THF) and then it can accomplish the methylation cycle in order to function in folate metabolism. Inhibition in the folate metabolism pathway may induce a neural tube defect. Thus, the folate pathway genes that regulate the function of this cycle are widely investigated. Observations showed that some key genes involved in the methylation cycle of THF, such as the methylenetetrahydrofolate reductase (MTHFR), the reduced folate carrier (RFC) and the methionine synthase reductase (MTRR), combine with vitamin B12-dependent methionine synthase (MTR) function and transfer the methyl group to homocysteine to accomplish the circle [15], [16]. The folate pathway is shown in Figure 1. The association between genetic variance and NTDs was not found until 1995, when the first literature on single-nucleotide polymorphisms (SNPs) appeared [17]. Since then, many articles have shown that the aberrant gene mutations that inhibit cellular folate transportation in folate metabolism have the strongest association with NTDs [10], [18], [19], [20], [21], [22]. SNPs, C677T and A1298C in MTHFR, A2756G in MTR, A66G in MTRR, A80G in RFC-1, have attracted most attention and may represent a substantial proportion of the risk of developing a neural tube defect as their key role in folate metabolism pathway [17]. Increasing evidence from epidemiological case-control studies has revealed that up to 70% in NTD prevalence may result from genetic factors [23].

Figure 1
Simplified overview of folate metabolism pathway, highlighting enzymes with polymorphisms investigated in this study.

Interestingly, even though a number of studies investigated the correlation of the NTDs and the polymorphisms, no consensus has been reached. Some observations showed that folate pathway gene polymorphisms might be capable of inhibiting the folate pathway [24], [25], [26], [27], [28]. However, several follow-up studies failed to replicate the association [29], [30], [31], [32]. We conducted this comprehensive meta-analysis integrating previous publications to study the association between key polymorphisms in the major folate pathway genes and NTDs.

Materials and Methods

To ensure the rigour of this current meta-analysis, we designed and reported it according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) [33] statement (

Search Strategy and Identification of Relevant Studies

We searched PubMed, EMBase, ISI Web of Science, and Chinese Wan Fang Data databases for published articles from June 1996 to May 2011, which investigated at least one of the polymorphisms of MTHFR, MTRR, MTR and RFC associated with NTDs. The search strategy was based on combinations of the English and/or Chinese keywords, “MTHFR”, “MTRR”, “MTR”, “RFC”, “folate pathway” “polymorphism”, or “SNP” and “NTDs or Neural Tube Defects or spina bifida” without language restrictions. References of reviews and retrieved studies were also scanned.

The following inclusion criteria had to be fulfilled: (1) case-control study and cohort study design; (2) data on any, some or all polymorphisms in MTHFR, MTRR, MTR and RFC; (3) presentation of data necessary for calculating odds ratios (ORs); (4) clear definition of NTDs. Animal studies, mother studies, reviews, and no specific data reported were excluded. Studies that duplicated other studies were eliminated, and only those whose design was complete were finally selected.

Data Extraction

All the data were extracted independently by two reviewers (T. Zhang & R. Zhong). The following information was extracted from the eligible literature: year of publication, first author’s name, country, ethnicity, genotyping method, source of control, and matching variables of controls with cases. Counts of alleles or genotypes in both case and control groups were extracted or calculated from published data to re-calculate crude ORs and their 95% confidence intervals (95% CIs) for assessing the association of the polymorphisms in MTHFR, MTRR, MTR and RFC with NTDs.

Statistical Analysis

Data from the case-control studies were summarized in two-by-two tables. In each table, crude ORs and their 95% confidence intervals (CIs) were calculated for each individual study based on the genotype data using the method as described by Zintzaras et al. [34]. The Cochran’s χ2-based Q statistic test was adopted to assess the case-between heterogeneity and, and heterogeneity was considered significant when P<0.05 for Q statistic or when I2 was above 75%. Data from the studies were combined in the Mantel-Haenszel Chi-square test by a random-effects model where heterogeneity was significantly present; otherwise, a fixed-effects model was applied. Pooled frequency analysis was performed in the method described by Thakkinstian [35]. Egger’s test and Begg’s test described by Egger et al. [36] for funnel plot asymmetry were applied to evaluate the evidence for publication bias. Models were chosen based on the method described by Thakkinstian [37], briefly, calculating and comparing the ORs of AA vs aa, Aa vs aa and AA and Aa, checking the heterogeneity and significance, then determining the best model. To explore sources of heterogeneity across studies, a meta-regression model was employed [38]. The pre-specified characteristics for assessment of heterogeneity sources were: ethnicity of population (Europe, Native America, Asia, Blacks and Other), source of control (population and hospital based controls), genotyping (PCR-RFLP, PCR-Taqman, Other and NR) and publication year. Stratified analysis, if feasible, was performed in a dominant model based separately on the source of the control group, ethnicity, and genotyping to investigate the reason for heterogeneity. The control group was drawn from three sources: population-based, hospital-based and NR (not reported in literature); by ethnicity, it was divided into 5 groups: European, Native American, Asian, African and Mixed; and by genotyping, it was divided into 4 groups, PCR-RFLP, PCR-Taqman, others and NR. Sensitivity analysis was also conducted to assess the influence of each study on the overall estimate. Cumulative meta-analysis was initially performed by date of publications to determine the dynamic trends as studies accumulated over time. All P values are two-tailed with a significance level at 0.05. All statistical analyses were done using Stata Version 10. (College Station, TX: StataCorp LP).


Characteristics of Included Studies

Figure 2 shows the procedure by which literature was selected. A comprehensive search yielded 172 references. After the removal of duplicate literature and articles containing unspecific data that did not meet our criteria, a total of 47 publications was finally included in this meta-analysis. Table S1 illustrates the characteristics of all the literature included in this research. From the table it can be seen that the studies that we investigated consisted of 85 case-control studies, including 42 studies of MTHFR C677T, 22 studies of MTHFR A1298C, 8 studies of MTRR A66G, 9 studies of MTR A2756G and 4 studies of A80G. These studies enrolled 4374 cases and 7232 controls of MTHFR C677T, 2602 cases and 4070 controls of MTHFR A1298C, 703 cases and 1572 controls of MTRR A66G, 843 cases and 1006 controls of MTR A2756G and 1107 cases and 1585 controls of RFC-1 A80G. (Table S1).

Figure 2
Flow chart of the literature search.

Frequency of Risk Allele in the Control Population

To estimate the pooled frequency, we combined case-control studies of MTHFR C677T and A1298C, and extracted data only from the control group. Figure 3 shows the pooled frequency of the variant alleles of MTHFR C677T and A1298C that yielded the most publications, in controls stratified by ethnicity. Based on all these samples, the frequency of risk T allele varied among different ethnicities: high in Native American and European healthy populations 43.8% (34.7%–52.9%) and 34.2% (30.7%–37.8%); low in Asian healthy populations 20.7% (11.1%–30.3%). The frequency of risk C allele in A1298C also revealed differences–high in the Asian controls 42.8% (40.3–45.4) and low in Native American and European populations 19.6% (14.4%–24.7%) and 27.2% (24.1%–30.4%) (Figure 3).

Figure 3
Pooled frequencies of the MTHFR C677T alleles and MTHFR A1298C alleles in controls stratified by ethnicity.

Results of the Overall Meta-analysis

Table 1 summarizes the ORs with corresponding 95% CIs for the association between genetic polymorphisms in the folate metabolism pathway and the risk for NTDs in the dominant, recessive and allelic models. Where significant difference was found in the three genetic models, a random-effects model was chosen according to the p values for heterogeneity. A fixed-effects model was applied to the allelic and recessive models in the study of MTR A2756G and to the recessive model in the study of RFC-1 A80G, while a strict random-effects model was chosen for the rest of the studies in which the p values of heterogeneity(<0.05) showed significance. According to the method of choosing genetic model, we first calculated the ORs (OR1 = 1.335 p<0.001, OR2 = 1.175 p = 0.011, OR3 = 1.177 p = 0.063), and OR1≠OR2, so that we excluded the recessive model. Considering that OR3 was not significant and the heterogeneity was more significant in an allele model, a dominant model was finally determined. In the same way, we chose a dominant model for the remaining four SNPs. Among all the combined studies of SNPs and NTDs, MTHFR C677T showed an association with NTDs (OR 1.23; 95%CI 1.07–1.42). There was no association observed between these four SNPs and NTDs and the results of other SNPs that we performed were negative. Specific ORs, I2 and p values are presented below. The results are presented in Figure 4.

Figure 4
The overall forest plot of OR with 95%CI for MTR A2756G, MTRR A66G and RFC-1 A80G polymorphism and Neural tube defects risk in dominant model.
Table 1
Summarized odds ratios with 95% confidence intervals for the association between genetic polymorphisms in the folate pathway and Neural Tube Defects risk.

Meta-regression Analysis and Stratified Analysis

To explore the potential sources of across study heterogeneity, a meta-regression analysis of MTHFR C677T and A1298C was performed respectively. An empty regression was firstly run to estimate the baseline value for τ212 = 0.098 and τ22 = 0.244), and then a series univariate model was conducted by adding single covariates including ethnicity of populations, source of controls, publication year and genotyping. In the univariate analysis, only the model including ethnicity and source of controls slightly reduced the τ2 value. Then we added the both two covariates, the τ12 value reduced to 0.089, R2 = 9.47% (P1 = 0.103) and τ22 value reduced to 0.196, R2 = 22.52% (P2 = 0.196), suggesting ethnicity and source of control cannot explain the major between-study heterogeneity. Studies of MTHFR C677T and A1298C were stratified to address heterogeneity while other SNPs were not stratified due to the importance priority and data availability. After stratification by sources of the controls, heterogeneity for MTHFR C677T was reduced in the substratification of the hospital based control group. In the population based control group, however, heterogeneity remained and the variant allele still conferred a significant increased risk. After stratification by ethnicity, heterogeneity in the Native American subgroup decreased with P = 0.139, I2 = 33.6%. Although heterogeneity remained, it should be noted that 22 studies of European subgroups also provided a significant correlation (OR = 1.21; 95%CI 1.03–1.42). For MTHFR A1298C, heterogeneity was reduced in the hospital based and Native American subgroups. Nonetheless, no significant association between MTHFR A1298C and NTDs was found (Table 2 & 3).

Table 2
Stratified analysis of the association between MTHFR C677T polymorphism and Neural Tube Defects in dominant model.
Table 3
Summarized odds ratios with confidence intervals of stratified studies for MTHFR A1298C polymorphism.

Sensitivity Analysis

We performed a sensitivity meta-analysis to assess the effects of individual studies on pooled ORs. Table 4 shows the results of this sensitivity analysis. None of the studies showed a strong enough influence to affect the combined results in MTHFR C677T and A1298C, MTRR A66G and MTR A2756G. After eliminating the results of Marco (2002) [24] in A1298C, heterogeneity decreased (P = 0.162, I2 = 23.4), which indicated that this study may be the main origin of the heterogeneity. Nevertheless, our results did not change despite removing the data in this study. Relton’s study [27] of A80G affected the association between A80G and NTDs that the results showed a significant correlation with the overall ORs (OR 1.55; 95%CI 1.24–1.92) and no statistical heterogeneity was observed; thus we dropped this study. We will have a more detailed discussion in the following part. The results after the removal were performed in fixed-effects model and identified an overall OR of 1.55(95%CI 1.24–1.92) (Table 4).

Table 4
Sensitive analysis of pooled OR for Genetic polymorphisms in MTHFR in the folate pathway.

Cumulative Meta-analyses

Cumulative meta-analyses were performed using a dominant model for C677T and A1298C, which were the most widely reported in the research. We sorted the literature in chronological order as shown in Figure 3. Remarkably, a statistically significant effect of a positive association between MTHFR C677T and NTDs was consistently observed with a narrowing of the 95% confidence interval through publication of the study in 2011. However, association study result of MTHFR A1298C and the risk for NTDs was negative. (Figure 5).

Figure 5
The cumulative forest plot of OR with 95%CI for MTHFR C677T polymorphism, MTHFR A1298C and Neural tube defects risk in dominant model.

Publication Bias

As demonstrated by the funnel plot and the Egger’s test, there was no significant publication bias in any overall meta analysis. Specific PEgger’s test results are presented below. The funnel plots showed in Figure 6.

Figure 6
The funnel plot of natural logarithm of OR against inverse standard error in each study.


This current study, to our knowledge, was the first to combine previous studies of key SNPs in the folate metabolism pathway underlying NTDs pathogenesis. Our results demonstrated a significant association between MTHFR C677T and NTDs in an overall meta-analysis of case-control studies. Moreover, the association was well supported by the subsequent cumulative meta-analysis. Our overall meta-analysis also integrate studies on A1298C in MTHFR, MTRR A66G, MTR A2756G and RFC-1 A80G. No significant evidence of correlation btween these SNPs and the NTDs was observed in our study initially. However, after removing one study of RFC-1 A80G, which was considered the origin of heterogeneity, the results showed a suggestive association.

So far we have known that sufficient folate supplyment during the first four weeks of pregnancy can decrease the risk for NTDs by more than 50% [39] and that folate metabolism and homocysteine status are relevant for the etiology. Mutations of genes in key enzymes in folate metabolism regulate folate transportation and metabolism meanwhile may interfere with its original function, thus leading to birth defects. Studies suggest that several mutations can severely impair MTHFR activity, lowering the folate status which could explain a quater of the NTDs occurrence [40], [41]. Fosst et al. [42] first demonstrated that TT substitution at nucleotide(nt) 677 can reduce the MTHFR activity by more than 65%, same effect was found in A1298C with less power [43]. This overall meta-analysis indicated that MTHFR C677T might be a risk factor of NTDs. Similar results were reported by Motulsky earlier [39] and some follow-up meta-analyses [44], [45]. It’s worth mentioning that the analysis of MTHFR C677T included approximately nine times as many participants as N. van der Put’s study (1997) [44] and 9 more studies than M. Amorim (2007) [45], which not only concordant with the ealier two studies but also extended the association to different populations. Although the ORs were not as high as the earliest study reported by Ou [46], with over 4300 cases and 7200 controls, our current analysis would have sufficient statistical power to detect a small size effect in the association between MTHFR C677T and NTDs based on current limited knowledge of the exact mechanism.

In addition, the cumulative meta-analysis of MTHFR C677T shows a consistantly positive trend with objective facticity. Furthermore, sensitivity analyses have shown that none of the literature we included would influence the results negatively enough to reverse the results.

Nonetheless, the obvious evidence of between-study heterogeneity should be issued in our meta-analysis. The τ2 value of meta-regression reduced to 0.089, R2 = 9.47% (P = 0.103), suggesting ethnicity and source of control cannot explain the major between-study heterogeneity but it may be part of the origin of the heterogeneity. Further, we stratified all the studies into subgroups classified according to source of controls, ethnicity and genotyping. Reduced heterogeneity in Native America subgroup and hospital based subgroup was observed, and significant association was also observed among studies of Europe and Mixed populations. The result of the overall meta-analysis and cumulative meta-analysis for MTHFR A1298C did not support the MTHFR A1298C as an independent risk factor of NTDs. Considering that C677T and A1298C are both in MTHFR and are likely to interact. Also, lack of consensus in the results of individual studies as MTHFR C677T and A1298C may be due to the different environment background. The influence of dietary intake, especially folate intake, varied in different ethnic populations, which may well be relevant to the difference in prevalence and cannot be excluded in the this study. The between-study heterogeneity would also affect the results. However, the meta-regression analysis suggested that ethnicity and source of control may not be the major origin of the heterogeneity. In stratified analysis, we observed the heterogeneity reduced sharply only in Native American group and Taqman group. After checking all possible sources of errors, with the avalaible data we failed to exclude the influence of the between-study heterogeneity that existed in a relatively large meta-analysis. The heterogeneity might be due to many reasons, such as differences in maternal folate status and recruitment procedures of the study population.

Under the hypothesis that loss-of-function mutations in MTR and mutations of the chaperone MTRR related to the activity of MTR may influence homocysteine levels resulting in severe disease phenotypes [47] [48], we combined the data of the key variants in MTHFR, MTRR and MTR that are in folate metabolisms to analyse the association. However, the available evidence did not support MTRR A66G or MTR A2756G as an independent risk factor of NTDs. Some explanation might be responsible for the lack of correlation. First, the sample size of studies was relatively small, so that to detect a very small size effect may require much larger sample size. Second, the outcome of an NTD patient varies from livebirth to stillbirth as the severity varies in different subtypes. Thus, the effect of genetic variants on risk of NTDs may be underestimated if studies only collect livebirths and less severity cases. Third, potential gene-gene, gene-environment interaction may affect the current results.

The sensitivity analysis of RFC-1 A80G showed that the study of Relton (2004) [29] affected the results which should be figured out. Under review of this report, Relton et al. indicated a contradictory result to other included studies that the 80A allele, not the 80G allele, increased risk of NTDs. Then we excluded the possibility that they report the allele for the reverse strand. Additionally, after removing this study, the heterogeneity reduced sharply, which revealed it was the main origin of the heterogeneity. The result of RFC-1 A80G in fixed-effects model was 1.55(95%CI 1.24–1.92). Since the sample size was quite limited (901 cases and 983 controls), we must be cautious of the association and more studies should be required to add in to improve the precision of the result.

The current study strongly supports the association of MTHFR C677T alleles with NTD risk by performing a cumulative meta-analysis of 42 studies of MTHFR C677T that demonstrated results with a relatively narrow 95% confidence interval. We believe that as more studies are added to our meta-analysis, results would remain stable. Also, we observed a suggestive association beween RFC-1 A80G and the risk of NTDs. However we failed to find a correlation in the remaining SNPs.

Some limitations merit serious consideration in our meta analysis. Firstly, as the pressent meta-analysis was primarily based on unadjusted effect estimates, the confounding factors were not controlled. Additionally, with the eligible information and methodological limitations we cannot excluded the between-study heterogeneity that remained. Secondly, the effects of gene-gene and gene-environment interaction was not addressed in this study. Thirdly, we systematically searched a variaty of databases for published literature, however, we cannot excluded the possiblity of missing some.

In summary, our studies demonstrated a significant correlation of MTHFR C677T, and a suggestive association of RFC-1 A80G and the increased risk of NTDs while the other SNPs in our study failed to support an evidence of the association. As the low edge of 95%CI nearly touched the null value, our research should be viewed with caution. Further large and well-designed studies will be needed to clarify the association of the polymorphisms in the folate pathway genes.

Supporting Information

Table S1

Characteristics of studies on genetic polymorphisms in the folate pathway and Neural Tube Defects risk included in the meta-analysis.


Checklist S1

PRISMA Checklist.


Funding Statement

The authors have no support or funding to report.


1. Christianson A (2006) Global Report on Birth Defects. March of Dimes Birth Defects Foundation.
2. Pitkin RM (2007) Folate and neural tube defects. Am J Clin Nutr 85: 285S–288S [PubMed]
3. Copp AJ, Greene ND, Murdoch JN (2003) The genetic basis of mammalian neurulation. Nat Rev Genet 4: 784–793 [PubMed]
4. Detrait ER, George TM, Etchevers HC, Gilbert JR, Vekemans M, et al. (2005) Human neural tube defects: developmental biology, epidemiology, and genetics. Neurotoxicol Teratol 27: 515–524 [PMC free article] [PubMed]
5. Smithells RW, Sheppard S, Schorah CJ (1976) Vitamin dificiencies and neural tube defects. Arch Dis Child 51: 944–950 [PMC free article] [PubMed]
6. Wolff T, Witkop CT, Miller T, Syed SB (2009) Folic acid supplementation for the prevention of neural tube defects: an update of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med 150: 632–639 [PubMed]
7. Laurence KM (1985) Prevention of neural tube defects by improvement in maternal diet and preconceptional folic acid supplementation. Prog Clin Biol Res 163B: 383–388 [PubMed]
8. Christiansen M, Garne E (2005) [Prevention of neural tube defects with periconceptional folic acid supplementation in Europe]. Ugeskr Laeger 167: 2875–2876 [PubMed]
9. Berry RJ, Li Z, Erickson JD, Li S, Moore CA, et al. (1999) Prevention of neural-tube defects with folic acid in China. China-U.S. Collaborative Project for Neural Tube Defect Prevention. N Engl J Med 341: 1485–1490 [PubMed]
10. Kibar Z, Capra V, Gros P (2007) Toward understanding the genetic basis of neural tube defects. Clin Genet 71: 295–310 [PubMed]
11. De-Regil LM, Fernandez-Gaxiola AC, Dowswell T, Pena-Rosas JP (2010) Effects and safety of periconceptional folate supplementation for preventing birth defects. Cochrane Database Syst Rev: CD007950. [PubMed]
12. Rothenberg SP, da Costa MP, Sequeira JM, Cracco J, Roberts JL, et al. (2004) Autoantibodies against folate receptors in women with a pregnancy complicated by a neural-tube defect. N Engl J Med 350: 134–142 [PubMed]
13. Yates JR, Ferguson-Smith MA, Shenkin A, Guzman-Rodriguez R, White M, et al. (1987) Is disordered folate metabolism the basis for the genetic predisposition to neural tube defects? Clin Genet 31: 279–287 [PubMed]
14. van der Put NM, Thomas CM, Eskes TK, Trijbels FJ, Steegers-Theunissen RP, et al. (1997) Altered folate and vitamin B12 metabolism in families with spina bifida offspring. QJM 90: 505–510 [PubMed]
15. Bassuk AG, Kibar Z (2009) Genetic basis of neural tube defects. Semin Pediatr Neurol 16: 101–110 [PubMed]
16. Ulrich CM, Robien K, McLeod HL (2003) Cancer pharmacogenetics: Polymorphisms, pathways and beyond. Nature Reviews Cancer 3: 912–920 [PubMed]
17. Molloy AM, Brody LC, Mills JL, Scott JM, Kirke PN (2009) The search for genetic polymorphisms in the homocysteine/folate pathway that contribute to the etiology of human neural tube defects. Birth Defects Res A Clin Mol Teratol 85: 285–294 [PubMed]
18. Richter B, Stegmann K, Roper B, Boddeker I, Ngo ET, et al. (2001) Interaction of folate and homocysteine pathway genotypes evaluated in susceptibility to neural tube defects (NTD) in a German population. J Hum Genet 46: 105–109 [PubMed]
19. Collin SM, Metcalfe C, Zuccolo L, Lewis SJ, Chen L, et al. (2009) Association of Folate-Pathway Gene Polymorphisms with the Risk of Prostate Cancer: a Population-Based Nested Case-Control Study, Systematic Review, and Meta-analysis. Cancer Epidemiology Biomarkers & Prevention 18: 2528–2539 [PubMed]
20. van der Linden IJM, Afman LA, Heil SG, Blom HJ (2007) Genetic variation in genes of folate metabolism and neural-tube defect risk. Proceedings of the Nutrition Society 65: 204–215 [PubMed]
21. Beaudin AE, Stover PJ (2007) Folate-mediated one-carbon metabolism and neural tube defects: balancing genome synthesis and gene expression. Birth Defects Res C Embryo Today 81: 183–203 [PubMed]
22. Bjorke-Monsen AL, Ueland PM, Schneede J, Vollset SE, Refsum H (1997) Elevated plasma total homocysteine and C677T mutation of the methylenetetrahydrofolate reductase gene in patients with spina bifida. QJM 90: 593–596 [PubMed]
23. Copp AJ, Greene ND (2010) Genetics and development of neural tube defects. J Pathol 220: 217–230 [PubMed]
24. Boduroğlu K (2005) Analysis of MTHFR 1298A>C in addition to MTHFR 677C>T polymorphism as a risk factor for neural tube defects in the Turkish population. The Turkish Journal of Pediatrics 47: 327–333 [PubMed]
25. Kirke PN (2004) Impact of the MTHFR C677T polymorphism on risk of neural tube defects: case-control study. British Medical Journal 328: 1535–1536 [PMC free article] [PubMed]
26. Marco PD (2002) Study of MTHFR and MS polymorphisms as risk factors for NTD in the Italian population. J Hum Genet 47: 319–324 [PubMed]
27. Parle-McDermott A, Mills JL, Kirke PN, O’Leary VB, Swanson DA, et al. (2003) Analysis of the MTHFR 1298A>C and 677C>T polymorphisms as risk factors for neural tube defects. Journal of Human Genetics 48: 190–193 [PubMed]
28. Shaw GM, Rozen R, Finnell RH, Wasserman CR, Lammer EJ (1998) Maternal vitamin use, genetic variation of infant methylenetetrahydrofolate reductase, and risk for spina bifida. Am J Epidemiol 148: 30–37 [PubMed]
29. Relton CL (2004) Gene-gene interaction in folate-related genes and risk of neural tube defects in a UK population. Journal of Medical Genetics 41: 256–260 [PMC free article] [PubMed]
30. Godbole K, Gayathri P, Ghule S, Sasirekha BV, Kanitkar-Damle A, et al. (2011) Maternal one-carbon metabolism, MTHFR and TCN2 genotypes and neural tube defects in India. Birth Defects Res A Clin Mol Teratol 91: 848–856 [PubMed]
31. Franchis Rd (1998) The C677T mutation of the 5,10-methylenetetrahydrofolate reductase gene is a moderate risk factor for spina bifida in Italy. J Med Genet 35: 1009–1013 [PMC free article] [PubMed]
32. Mornet E, Muller F, Lenvoise-Furet A, Delezoide AL, Col JY, et al. (1997) Screening of the C677T mutation on the methylenetetrahydrofolate reductase gene in French patients with neural tube defects. Hum Genet 100: 512–514 [PubMed]
33. Moher D, Liberati A, Tetzlaff J, Altman DG (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol 62: 1006–1012 [PubMed]
34. Zintzaras E, Doxani C, Koufakis T, Kastanis A, Rodopoulou P, et al. (2011) Synopsis and meta-analysis of genetic association studies in osteoporosis for the focal adhesion family genes: the CUMAGAS-OSTEOporosis information system. BMC Medicine 9: 9. [PMC free article] [PubMed]
35. Thakkinstian A, McEvoy M, Minelli C, Gibson P, Hancox B, et al. (2005) Systematic review and meta-analysis of the association between {beta}2-adrenoceptor polymorphisms and asthma: a HuGE review. Am J Epidemiol 162: 201–211 [PubMed]
36. Egger M, Davey Smith G, Schneider M, Minder C (1997) Bias in meta-analysis detected by a simple, graphical test. BMJ 315: 629–634 [PMC free article] [PubMed]
37. Thakkinstian A, McElduff P, D’Este C, Duffy D, Attia J (2005) A method for meta-analysis of molecular association studies. Stat Med 24: 1291–1306 [PubMed]
38. Thompson SG, Higgins JP (2002) How should meta-regression analyses be undertaken and interpreted? Stat Med 21: 1559–1573 [PubMed]
39. Motulsky AG (1996) Nutritional ecogenetics: homocysteine-related arteriosclerotic vascular disease, neural tube defects, and folic acid. Am J Hum Genet 58: 17–20 [PubMed]
40. Friso S, Choi SW, Girelli D, Mason JB, Dolnikowski GG, et al. (2002) A common mutation in the 5,10-methylenetetrahydrofolate reductase gene affects genomic DNA methylation through an interaction with folate status. Proc Natl Acad Sci U S A 99: 5606–5611 [PubMed]
41. Kondo A, Kamihira O, Ozawa H (2009) Neural tube defects: prevalence, etiology and prevention. Int J Urol 16: 49–57 [PubMed]
42. Frosst P, Blom HJ, Milos R, Goyette P, Sheppard CA, et al. (1995) A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nat Genet 10: 111–113 [PubMed]
43. Weisberg I, Tran P, Christensen B, Sibani S, Rozen R (1998) A second genetic polymorphism in methylenetetrahydrofolate reductase (MTHFR) associated with decreased enzyme activity. Mol Genet Metab 64: 169–172 [PubMed]
44. van der Put NM, Eskes TK, Blom HJ (1997) Is the common 677C–>T mutation in the methylenetetrahydrofolate reductase gene a risk factor for neural tube defects? A meta-analysis. QJM 90: 111–115 [PubMed]
45. Amorim MR, Lima MAC, Castilla EE, Orioli IM (2007) Non-Latin European descent could be a requirement for association of NTDs andMTHFR variant 677C > T: A meta-analysis. American Journal of Medical Genetics Part A 143A: 1726–1732 [PubMed]
46. Ou CY (1996) 5,10 Methylenetetrahydrofolate reductase genetic polymorphism as a risk factor for Neural Tube Deffects. American Journal of Medical Genetics 63: 610–614 [PubMed]
47. Li YN, Gulati S, Baker PJ, Brody LC, Banerjee R, et al. (1996) Cloning, mapping and RNA analysis of the human methionine synthase gene. Hum Mol Genet 5: 1851–1858 [PubMed]
48. Selzer RR, Rosenblatt DS, Laxova R, Hogan K (2003) Adverse effect of nitrous oxide in a child with 5,10-methylenetetrahydrofolate reductase deficiency. N Engl J Med 349: 45–50 [PubMed]

Articles from PLoS ONE are provided here courtesy of Public Library of Science