Low dietary folate intake has been implicated as a risk factor for depression. However, observational epidemiological studies are plagued by problems of confounding, reverse causality and measurement error. A common polymorphism (C677T) in MTHFR is associated with methyltetrahydrofolate reductase (MTHFR) activity and circulating folate and homocysteine levels and offers insights into whether the association between low folate and depression is causal. We genotyped this polymorphism in 3478 women in the British Women's Heart and Health Study. In these women, we looked at the association between genotype and three indicators of depression; ever diagnosed as depressed, currently taking antidepressants and the EuroQol mood question. We also carried out a systematic review and meta-analysis of all published studies which have looked at the association between MTHFR C677T genotype and depression. In the British Women's Heart and Health Study, we found evidence of an increased risk of ever being diagnosed as depressed in MTHFR C677T TT individuals compared with CC individuals, odds ratio (OR) 1.35(95% CI: 1.01, 1.80). Furthermore, we identified eight other studies, which have examined the association between depression and MTHFR C677T. We were able to include all of these studies in our meta-analysis together with our results, obtaining an overall summary OR of 1.36 (95% CI: 1.11, 1.67, P=0.003). Since this genotype influences the functioning of the folate metabolic pathway, these findings suggest that folate or its derivatives may be causally related to risk of depression.
Biological and epidemiological evidence indicates that folate may be an important risk factor in depression. There are several mechanisms by which folate could affect the central nervous system pathways and lead to depression. Folate is a co-factor in the metabolic pathway leading to the synthesis of serotonin and other neurotransmitters, it decreases homocysteine level, which seems to have a excitotoxic effect via Ca2+ glutamate brain receptors and folate is essential for the biosynthesis of S-adenosylmethionine (SAM), a compound with antidepressant properties.1, 2, 3
Several epidemiological studies have reported associations of folate and its derivatives with depression.3, 4 However, these findings may reflect poor diets among depressed individuals or may be the result of confounding by other factors, such as alcohol intake, socioeconomic position and physical activity, which are associated with depression. Three small clinical trials have also shown preliminary evidence that supplementation with folate may have a therapeutic effect on depression either alone or with standard antidepressants.5, 6
Depression is common and an important cause of morbidity and mortality and supplementation with folate is relatively cheap and safe, therefore the elucidation of folate as a risk factor for depression is likely to have important public health consequences. There were over 25 million antidepressant prescriptions last year in England (www.ppa.org.uk) and since depression is a chronic condition many individuals with depression require repeat episodes of treatment and/or respond poorly to antidepressant medication.7 Thus, any potential means of preventing or improving the treatment of this illness are attractive. Advice on folic acid supplementation is already given to expectant mothers to reduce risks of neural tube defects and could be extended to the general population if shown to be beneficial. There is debate in the UK as to whether there should be compulsory cereal fortification as in the USA and Canada.8, 9 A stronger case for fortification would be made if folate intake were also found to be associated with additional health benefits for the whole population, including the reduction of depression.
The enzyme methyltetrahydrofolate reductase (MTHFR) metabolizes folate and in doing so produces a methyl donor for the synthesis of methionine from homocysteine, and the precursor of S-adenosyl-L-methionine. A common genetic variant in the MTHFR gene has been identified in which a C to T substitution at position 677 leads to an amino-acid change from alanine to valine, resulting in a thermolabile variant of the enzyme, which has approximately 30% of the wild-type enzyme activity. MTHFR C677T TT homozygotes have been shown to have higher homocysteine levels compared with CC homozygotes, levels in heterozygotes appear to be intermediate between the two.10 Several polymorphisms of genes involved in folate/homocysteine metabolism have been investigated for their effect on homocysteine level, but the most consistent and largest effect is observed for MTHFR C677T.11, 12, 13
The association between a polymorphism and disease is not normally subject to the problems of confounding and reverse causation inherent in looking at associations between lifestyle factors and disease,14 thus an association between MTHFR C677T and depression will add to the evidence implicating folate or its metabolites in this condition. Maternal folate intake during pregnancy is now an established risk factor for neural tube defects, and an association between the MTHFR C677T polymorphism and neural tube defects has been found comparing the mothers of cases and mothers of controls, but not when comparing the fathers.15 This suggests that the principle of using MTHFR C677T as a surrogate for measuring folate levels can produce valid inferences.
A number of studies have examined the association between MTHFR C677T and depression but most have been small and as a consequence have produced imprecise estimates.16, 17, 18, 19, 20, 21, 22, 23 We have looked at the association between the MTHFR C677T polymorphism and three indicators of depression in the British Women's Heart and Health study, a population-based sample of 3487 women. Further, we have carried out a systematic review of all studies to date that have examined the association between MTHFR C677T and indicators of depression and have combined our results from the British Women's Heart and Health Study with those from other studies in a meta-analysis, to provide an overall estimate of the association.
Full details of the selection of participants and measurements used in the British Women's Heart and Health Study have been previously reported.24 Between 1999 and 2001, 4286 women aged 60–79 years, who were randomly selected from 23 British towns were interviewed, examined, completed medical questionnaires and had detailed reviews of their medical records.24
Three indicators of depression were used: current use of antidepressant medication, self-report of ever being diagnosed by a doctor with depression and the EQ5D (EuroQoL – http://www.eur.nl/bmg/imta/eq-net/EQ5d.htm) mood question. Participants brought all of their medications to a nurse interview. Medications were coded using the British National Formulary (http://www.bnf.org/). Participants whose response to the EQ5D mood question was that they were ‘today feeling either moderately or extremely anxious and/or depressed’ were coded as depressed.
Weight and height were measured using standard procedures and were used to calculate body mass index; obesity was defined as a body mass index greater than or equal to 30 kg/m2. Participants were asked to report the longest held occupation of their father during their childhood and their own and husbands longest held occupation. These were used to derive occupational social class of the participant's father and the head of household during her adulthood using the UK Registrar General's Classification (Social class I – professional; II – managerial/technical; IIInm – skilled non-manual; IIIm – skilled manual; IV – partly skilled manual; V – unskilled manual). Data on smoking (classified as never, past, current – including those who said they had quit smoking in the 6 months period prior to assessment), frequency of alcohol consumption (daily or most days, weekends only, once–twice a month, special occasions only, never; in the analyses, the last two categories were combined) and physical activity (categorized as <2, 2–3, ⩾3 h per week of either moderate or vigorous activity) were obtained from the interview or questionnaires.
DNA was extracted by salting out procedure.25 Genotyping was undertaken by KBioscience Ltd (www.kbioscience.co.uk), who use their own form of competitive allele-specific PCR system (KASPar) and Taqman™, for SNP analysis.
Assumptions of Hardy–Weinberg equilibrium were formally tested using a likelihood ratio test. Prevalences (for dichotomous variables) and means (for continuous variables) are presented by genotype. ANOVA was used for testing differences between genotypes for continuous variables, and χ2 tests were used for testing differences between genotypes for dichotomous variables. Logistic regression models were used for assessing genotype associations with depression. We made no a priori assumptions of the genetic model for disease risk. However, our results suggested a co-dominant model and hence we computed a test for trend across the three genotypes. Logistic regression models were performed with and without adjustment for BMI, because MTHFR C677T was found to be associated with obesity in this study. Robust standard errors, which take into account possible non-independence between women from the same town, were used in these logistic regression analyses to estimate confidence intervals and P-values. All analyses were conducted using Stata version 8.
Local ethics committee approvals were obtained for the British Women's Heart and Health Study. Participants were asked for informed consent to review their medical records and for permission to perform anonymized genetic tests on stored blood. Eight women declined to give consent and have not been included in this study.
Systematic review and meta-analysis
Papers published prior to August 2005 were identified through a search of Medline (www.ncbi.nlm.nih.gov), PsycInfo (www.psycinfo.com), ISIS web of knowledge (http://:wok.mimas.ac.uk) and google (www.google.com) using the following search terms: ‘Depression’ and ‘MTHFR’ or ‘methylenetetrahydrofolate reductase’. A cited reference search of retrieved articles was carried out, and publications were also identified by review of the bibliographies of retrieved articles. All articles reporting on MTHFR genotype among individuals with current or past depression and unrelated controls were included in the systematic review. A meta-analysis was carried out using data on genotype distribution relative to an outcome of depression, either from publications or following correspondence with the authors. Unadjusted odds ratios (ORs) were calculated for TT, and then CT, versus CC genotypes. As the different studies were carried out on diverse populations and had used different methods for assessing depression, we anticipated heterogeneity and undertook a random effects meta-analysis.26 The influence of method of defining depression (clinical interview versus self-administered questionnaire) was assessed using meta-regression.26 We assessed small study bias, indicating publication bias, by computing both the Egger and Begg's tests.26 We undertook three meta-analyses including in turn results of the association with each of the three different measurements of depression in the British Women's Heart and Health Study.
In the British Women's Heart and Health Study of the 4278 participants who gave consent for genetic testing, 15 (five Afro-Caribbean, eight South Asian and two other) were defined by the examining nurse as being of non-European origin and have been excluded from further analysis. Of the remaining 4263 women, 441 had either no blood samples taken or had insufficient blood taken for all analyses and did not have a K-EDTA sample for DNA extraction. Of the remaining 3822 women, DNA was successfully extracted on 3545 (93%). Genotyping reactions failed or were unclear for just 58 (1.6%) subjects. X–Y plots of allele 1 and allele 2 fluorescence signal for each subject fell distinctly into three clusters for the remaining 3487 women (who comprised 81.5% of the total participants). Mean age and the prevalence of the three indicators of depression did not differ between the 3487 with genotypic data and those women without these data (all P-values>0.5).
In total, 1565 (44.9%) of the women were homozygous for the major allele (CC), 1520 (43.6%) were heterozygous (TC) and 402 (11.5%) were homozygous for the minor allele (TT). The relative frequency of the T allele was 0.33. There was no evidence of departure from Hardy–Weinberg equilibrium. Table 1 shows the relationship of characteristics that could potentially confound an association between folate and depression to genotype. As one would expect from the random allocation of genotype at gamete formation, genotype was not associated with mean age, childhood or adulthood social class, smoking, physical activity or alcohol consumption. Mean BMI did not vary by genotype but possession of a T allele was associated with increased probability of being obese. In an analysis of genotype frequency by longitude and latitude of place of birth, the MTHFR C677T genotype was shown not to vary by geographical location in the UK (results not shown). The prevalence of ever having been diagnosed with depression in this study was 15.6% (n=545), 11.5% (n=401) were using antidepressants at the time of interview, and 22.7% (n=790) responded positively to the Euroquol mood question. The overlap between the three indicators of depression is shown in Figure 1 with 4.2% (n=147) of women responding positively to all three.
There was a linear trend of increasing prevalence of reporting ever being diagnosed with depression with each additional T allele (Table 2), with a crude OR of 1.20 (95% CI: 0.99, 1.46) for CT versus CC genotype and an OR of 1.35 (95% CI: 1.01, 1.80) for TT versus CC genotype. The prevalences of responding positively to the Euroquol mood question and of taking antidepressants were also marginally higher among women with the TT genotype than among those who were homozygous for the major C allele. However, aside from the association of the CT genotype with the Euroquol mood question, the genotype associations for these outcomes were weak and the 95% confidence intervals included the null value. The OR for TT versus CC for individuals who responded positively to all three indicators of depression (N=147) was 1.56 (95% CI: 0.94, 2.59). Adjustment for BMI made no meaningful difference to any of the genotype depression associations.
Eight other, mostly small studies (Table 3) totaling 696 cases have looked at associations between the MTHFR C677T polymorphism and risk of depression, determined by various different methods.16, 17, 18, 19, 20, 21, 22, 23 The largest of these studies found an OR of 1.63 (1.05–2.53) for risk among TT versus CC homozygotes after adjustment for age and sex,21 findings which were supported by two further small studies. A random effects meta-analysis of TT versus CC homozygotes gave an OR of 1.37 (1.03–1.82, P=0.03) for these eight studies.
A meta-analysis of all studies, including the British Women's Heart and Health study (BWHHS), using ever diagnosed with depression as the depression indicator, included 1241 cases of depression and resulted in an OR of 1.36 (95% CI: 1.11, 1.67, P=0.003) for TT versus CC genotypes and depression, with the BWHHS providing 49% of the weight in this pooled estimate. There was no evidence of between-study heterogeneity (χ2=5.49, P=0.70, I2=0.0%) (Figure 2). The OR for CT versus CC genotypes was 1.14 (95% CI: 0.89–1.47, P>0.05), with evidence of between-study heterogeneity (χ2=17.5 P=0.02, I2=54.2%). When a positive score on the Euroquol mood question was used as the indicator for the BWHHS, the OR for the TT versus CC genotype was 1.23 (95% CI: 1.02, 1.49, P=0.04), and for CT versus CC the OR was 1.14 (95% CI: 0.89–1.46, P>0.05). When taking antidepressants was used as the indicator, the OR for TT was 1.23 (95% CI: 0.99, 1.53, P=0.06) and for CT it was 1.08 (95% CI: 0.85,1.39, P>0.05). An Egger test and a Begg test provided no strong evidence that TT effect estimates, using diagnosed depression as the outcome in the BWHHS, were related to study size (P=0.95 and P=0.85 respectively), providing some reassurance that publication bias has not distorted the findings. There is strong evidence from meta-analyses of genetic association studies that the first published study typically indicates a far greater effect size than is subsequently reported. This appears to be the case here, and so the meta-analysis was re-run, using ever diagnosed with depression as the depression indicator in the BWHHS, and excluding the first published study. The results did not change substantially the OR for CC versus TT genotype, which was 1.31 (95% CI: 1.06–1.61) and for CT versus TT genotype was 1.21 (95% CI: 0.91–1.61). A meta-regression analysis showed no influence of method of defining depression on outcome (P=0.89) (this was essentially a comparison of studies which used clinical interview versus those which used self-report). Further, a meta-analysis including only studies which included a psychiatrist diagnosis, although based on a small number of cases (n=412), gave an OR of very similar magnitude 1.40 (95% CI: 0.92, 2.13).
We found evidence of an association between the MTHFR C677T TT genotype and risk of depression both in the British Women's Heart and Health Study and in a meta-analysis of all such studies published to date. The greatest increase in risk associated with depression in the BWHHS occurred when the diagnosis of depression was most stringent and limited to those responding positively to three measures of depression. The risk associated with being heterozygous at this locus is intermediate between the risk for TT and CC homozygotes, suggesting a co-dominant mode of inheritance, and is consistent with the differences in homocysteine levels between these genotypes.10, 15 Aside from the finding that this gene may confer susceptibility to depression, this result points towards folate or its derivatives as important factors in predicting risk of depression, since the MTHFR C677T influences the function of the folate metabolic pathway, and results in differences in serum homocysteine and folate levels.10
Genotypic data were available on 81% of the total cohort but those without these data did not differ with respect to age or indicators of depression from those with these data. In addition, the proportion of current and past smokers in this study, together with the distributions of other characteristics including body mass index, cardiovascular disease and social class, are consistent with those of a nationally representative sample of English women of a similar age, as reported in the health survey for England, and therefore indicate that this study is representative of the target population.27 Our study is cross-sectional, but reverse causality could not explain the association between genotype and depression and the similarity of mean age across genotypes suggests that survivor bias is unlikely to have influenced our results. Because our study did not set out a priori to collect cases of depression, our assessment of depression was based on self-report of past diagnoses, current mood and current medication use rather than clinical assessment with international diagnostic criteria. Thus, we are likely to have included some women who are not clinically depressed in our case groups and missed some true cases. However, a large proportion (68.3%) of women who had reported that they had been diagnosed with depression, were also taking antidepressants at baseline and/or answered positively to the mood questionnaire. All our indicators of depression show an association with MTHFR C677T in the same direction, which suggests that the results reflect real associations and not false positives due to multiple testing. This interpretation is further strengthened, by the finding that the greatest increase in risk by genotype was apparent among women who responded positively to all three indicators, and these are the women who are most likely to be clinically depressed. The association with use of antidepressants was particularly weak, which could be because antidepressants in this age group are used to manage other conditions in addition to depression, such as generalized fatigue, agitation or drooling, which would attenuate the association. Similarly, the Euroquol mood questionnaire also captures general unhappiness in addition to anxiety and depression. A further indication that the results for the BWHHS are not importantly biased by our outcome measure is that our results are consistent with those of other studies using international diagnostic criteria (Figure 1).
Confounding by population admixture is a potential problem in genetic association studies, and is particularly relevant to this analysis as the propensity to report depressive symptoms and be treated for depression varies across ancestral and social groups, and small effects could result from bias such as this. However, the analysis of the BWHHS excluded individuals of non-European origin and found no association between MTHFR C677T genotype and place of birth. In addition, there was no evidence that the distribution of this genotype differed by socioeconomic status.
In the meta-analysis, we combined studies that were carried out in diverse populations and that measured depression using different methods. The total number of cases in the previous studies was small, the largest of which, like the BWHHS, did not use a clinical diagnosis of depression. Limiting the meta-analysis to those studies which used psychiatrist diagnosis reduces the power of the meta-analysis to detect an effect, although the association with genotype is in the expected direction and of similar magnitude.
Is folate causally related to depression?
Although associations between low folate levels and depression have been reported previously in epidemiological studies,3, 4 associations may be confounded as individuals with low folate may also be more likely to be from less favourable socioeconomic backgrounds, be heavy alcohol consumers or be less likely to exercise. The observed association between low folate and depression could also be due to reverse causality as depressed people may have a reduced dietary folate intake. Further, there is substantial error in the measurement of folate, and intake at a particular time point may not be a good indicator of average lifetime exposure to folate. In two studies of breast cancer and folate intake, the intraclass correlation coefficient for folate intake measured on two occasions was 0.36,28 and for folate measured by food frequency questionnaire compared with 24 h recall was 0.26.29 There are promising results from small preliminary randomized controlled trials of folate supplementation and depression. Two trials assessed the use of folate in addition to other treatment, and a meta-analysis of these studies found that adding folate reduced Hamilton Depression Rating Scale (HDRS) scores on average by a further 2.65 points (95% CI: 0.38, 4.93).5, 6
In contrast to traditional epidemiology, reverse causality cannot explain the association between genotype and depression, since genotype is fixed at birth.14 Also, as genotype is determined randomly at conception, the factors that commonly confound associations with environmental or behavioural exposures, such as dietary intake of folate, do not confound genotype-depression associations.14 This principle of Mendelian randomization14 is demonstrated in Table 2, which shows that MTHFR genotype is not associated with socioeconomic position or behavioural risk factors. The association with obesity, which has been demonstrated in a previous study, may be a chance finding or may be a biological effect of MTHFR C677T.30 The MTHFR C677T genotype will give an indication of lifetime differences in exposure to folate. Further, the error rate in the measurement of genotypes is extremely low.31 Assay data clustering indicates high precision of data calling, with only 1.6% of reactions being of uncertain genotype or failing completely and therefore excluded from analysis.
Although it is unclear how the folate metabolic pathway may lead to depression and whether this is due to effects of folate, homocysteine or SAM, homocysteine is a good biomarker of pathway functioning. Individuals possessing the C677T CC genotype have homocysteine levels which are on average 2.6 μmol/l lower than those with the TT genotype.32 A 3–4 μmol/l reduction in homocysteine levels has been shown to be achievable by supplementation with at least 0.5 mg folic acid.33 On the basis that the association between genotype and depression is causal, and taking the effect of genotype on homocysteine as an indicator of its biological effect, the summary OR of 1.36 from the meta-analysis of the TT versus CC genotype is equivalent to an OR of 1.43 for a 3 μmol/l increase in homocysteine (calculated by raising 1.36 to power 3/2.6), or a 43% increase in risk. A difference in homocysteine of this magnitude could potentially be corrected with folate supplementation, thus if the biological effect of MTHFR genotype with respect to depression is of similar magnitude to the effect of MTHFR genotype on homocysteine, a meaningful reduction in depression should be produced by folate supplementation. Although confidence intervals around our estimates are relatively wide, and are subject to the limitations described above, this represents an important public health finding as it may be possible to reduce the prevalence of depression simply by nutrient supplementation. Further large genetic and intervention studies are required to confirm this finding and provide more precise estimates of this relationship.34 The role of folate supplementation should be evaluated with respect to overall predicted health effects of such supplementation.
SJL undertook the literature review, carried out the meta-analysis and coordinated the writing of the paper. DAL is a co-director of the BWHHS, carried out the analysis using the BWHHS data and contributed to the interpretation of the analysis. GDS contributed to the study design and conduct, and to the interpretation of the analysis. RA provided advice on the interpretation of depression outcomes. NT and INMD coordinated the genotyping and DNA extraction. SE is the principle investigator of the BWHHS and contributed to the interpretation of the analysis. All authors contributed to the final version of this paper.
Bottiglieri T, Hyland K, Reynolds EH . The clinical potential of ademetionine (S-adenosylmethionine) in neurological disorders. Drugs 1994; 48: 137–152.
Mattson M, Shea T . Folate and homocysteine metabolism in neural plasticity and neurodegenerative disorders. Trends Neurosci 2003; 23: 137.
Paul RTP, McDonnell AP, Kelly CB . Folic acid: neurochemistry, metabolism and relationship to depression. Hum Psychopharmacol Clin Exp 2004; 19: 477–488.
Morris MS, Fava M, Jacques PF, Selhub J, Rosenberg IH . Depression and folate status in the US population. Psychother Psychosom 2003; 72: 80–87.
Guaraldi GP, Fava M, Mazzi F, la Greca P . An open trial of methyletrahydrofolate in elderly depressed patients. Ann Clin Psychiatry 1993; 5: 101–105.
Taylor MJ, Carney SM, Goodwin GM, Geddes JR . Folate for depressive disorders: systematic review and meta-analysis of randomized controlled trials. J Psychopharmacol 2004; 18: 251–256.
Nemeroff C . Progress in the battle with the black dog: advances in the treatment of depression. Am J Psychiatry 2001; 158: 1555–1557.
Ray JG, Meier C, Vermeulen MJ, Boss S, Wyatt PR, Cole DE . Association of neural tube defects and folic acid food fortification in Canada. Lancet 2002; 360: 2047–2048.
Honein MA, Paulozzi LJ, Mathews TJ, Erickson JD, Wong LY . Impact of folic acid fortification of the US food supply on the occurrence of neural tube defects. JAMA 2001; 285: 2981–2986.
Frosst P, Blom HJ, Milos R, Goyette P, Sheppard CA, Matthews RG et al. A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nat Genet 1995; 10: 111–113.
Kluijtmans L, Young IS, Boreham C, Murray L, McMaster D, McNulty H et al. Genetic and nutritional factors contributing to hyperhomocysteinemia in young adults. Blood 2003; 101: 2483–2488.
Sunder-Plassmann G, Fodinger M . Genetic determinants of the homocysteine level. Kidney Int Suppl 2003; 84: S141–S144.
Geisel J, Hubner U, Bodis M . The role of genetic factors in the development of hyperhomocysteinemia. Clin Chem Lab Med 2003; 41: 1427–1434.
Davey Smith G, Ebrahim S . ‘Mendelian randomization’: can genetic epidemiology contribute to understanding environmental determinants of disease? Int J Epidemiol 2003; 32: 1–22.
Botto LD, Yang Q . 5,10-Methylenetetrahydrofolate reductase gene variants and congenital anomalies: a HuGE review. Am J Epidemiol 2000; 151: 862–877.
Reif A, Pfuhlmann B, Lesch KP . Homocysteinemia as well as methylenetetrahydrofolate reductase polymorphism are associated with affective psychoses. Prog Neuropsychopharmacol Biol Psychiatry 2005; 29: 1162–1168.
Almeida OP, Flicker L, Lautenschlager NT, Leedman P, Vasikaran S, van Bockxmeer FM . Contribution of the MTHFR gene to the causal pathway for depression, anxiety and cognitive impairment in later life. Neurobiol Aging 2005; 26: 251–257.
Kelly CB, McDonnell AP, Johnston TG, Mulholland C, Cooper SJ, McMaster D et al. The MTHFR C677T polymorphism is associated with depressive episodes in patients from Northern Ireland. J Psychopharmacol 2004; 18: 567–571.
Tan EC, Chong SA, Lim LC, Chan AO, Teo YY, Tan CH et al. Genetic analysis of the thermolabile methylenetetrahydrofolate reductase variant in schizophrenia and mood disorders. Psychiatr Genet 2004; 14: 227–231.
Bjelland I, Tell GS, Vollset SE, Refsum H, Ueland PM . Folate, vitamin B12, homocysteine, and the MTHFR 677C-T polymorphism in anxiety and depression. Arch Gen Psychiatry 2003; 60: 618–626.
Hickie I, Scott E, Naismith S, Ward PB, Turner K, Parker G et al. Late-onset depression: genetic, vascular and clinical contributions. Psychol Med 2001; 31: 1403–1412.
Kunugi H, Fukuda R, Hattori M, Kato T, Tatsumi M, Sakai T et al. C677T polymorphism in methylenetetrahydrofolate reductase gene and psychoses. Mol Psychiatry 1998; 3: 435–437.
Arinami T, Yamada N, Yamakawa-Kobayashi K, Hamaguchi H, Toru M . Methylenetetrahydrofolate reductase variant and schizophrenia/depression. Am J Med Genet 1997; 74: 526–528.
Lawlor DA, Bedford C, Taylor M, Ebrahim S . Geographic variation in cardiovascular disease, risk factors and their control in older women: British Women's Heart and Health Study. J Epidemiol Community Health 2003; 57: 134–140.
Miller SA, Dykes DD, Polesky HF . A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 1988; 16: 1215.
Sterne JAC, Bradburn MJ, Egger M . Meta-analysis in Stata. In: Egger M, Davey Smith G and Altman D (eds). Systematic Reviews in Health Care: Meta-Analysis in Context, 2nd edn. BMJ: London, 2001; 347–369.
Erens B, Primatesta P . Health Survey for England 1998: Cardiovascular Disease. The Stationary Office: London, 1999.
Shrubsole MJ, Jin F, Dai Q, Shu XO, Potter JD, Hebert JR et al. Dietary folate intake and breast cancer risk: results from the Shanghai Breast Cancer Study. Cancer Res 2001; 61: 7136–7141.
Sellers TA, Kushi LH, Cerhan JR, Vierkant RA, Gapstur SM, Vachon CM et al. Dietary folate intake, alcohol, and risk of breast cancer in a prospective study of postmenopausal women. Epidemiology 2001; 12: 420–428.
Wilcken DE, Wang XL, Sim AS, McCredie RM . Distribution in healthy and coronary populations of the methylenetetrahydrofolate reductase (MTHFR) C677T mutation. Arterioscler Thromb Vasc Biol 1996; 16: 878–882.
Knight JC, McGuire W, Kortok MM, Kwiatkowski D . Accuracy of genotyping of single-nucleotide polymorphisms by PCR-ELISA allele-specific oligonucleotide hybridization typing and by amplification refractory mutation system. Clin Chem 1999; 45: 1860–1863.
Brattstrom L, Wilcken DE, Ohrvik J, Brudin L . Common methylenetetrahydrofolate reductase gene mutation leads to hyperhomocysteinemia but not to vascular disease: the result of a meta-analysis. Circulation 1998; 98: 2520–2526.
Homocysteine Lowering Trialists' Collaboration. Lowering blood homocysteine with folic acid based supplements: meta-analysis of randomized trials. Br Med J 1998; 316: 894–898.
Colhoun HM, McKeigue PM, Davey Smith G . Problems of reporting genetic associations with complex outcomes. Lancet 2003; 361: 865–872.
The British Women's Heart and Health Study is co-directed by Professor Shah Ebrahim, Dr Debbie Lawlor, Professor Peter Whincup and Dr Goya Wannamethee. We thank Carol Bedford, Alison Emerton, Nicola Frecknall, Karen Jones, Rita Patel, Mark Taylor and Katherine Wornell for collecting and entering data, all of the general practitioners and their staff who have supported data collection, and the women who have participated in the study. We also thank Professor Ian Hickie, Dr Sharon Naismith and Professor Osvaldo Almeida who provided data from their studies for the meta-analysis. The British Women's Heart and Health Study is funded by the UK Department of Health and the British Heart Foundation. DAL is funded by a UK Department of Health Career Scientist Award. NT is funded by a UK Medical Research Council studentship. The views expressed in this publication are those of the authors and not necessarily those of any of the funding bodies. The funding bodies have had no influence over the scientific work or its publication.
About this article
Folic acid supplementation during pregnancy may protect against depression 21 months after pregnancy, an effect modified by MTHFR C677T genotype
European Journal of Clinical Nutrition (2012)