MTHFR 677TT genotype in C₁ metabolism

The MTHFR enzyme requires the B-vitamin riboflavin in its co-enzymatic form flavin adenine dinucleotide to catalyse the conversion of 5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate. This enzyme plays a key role in regulating the availability of methyl groups required for the remethylation of homocysteine into methionine. The common 677C→T variant results in a less active thermolabile MTHFR enzyme typically leading to elevated homocysteine concentration in vivo ⁽ Reference Frosst, Blom and Milos ^{10 )}. In vitro evidence indicates that the reduced activity of the variant enzyme is the result of an increased propensity to dissociate from its FAD co-factor⁽ Reference Guenther, Sheppard and Tran ^{11 ,} Reference Yamada, Chen and Rozen ^{12 )}. The finding of lower erythrocyte folate concentrations in individuals with the homozygous mutant TT genotype compared with those without this genetic variant, has led to the suggestion that folate requirements may be increased in these individuals⁽ Reference Molloy, Daly and Mills ^{13 )}. Furthermore, homocysteine concentrations are found to be highest under conditions of suboptimal folate status in combination with the TT genotype⁽ Reference Jacques, Bostom and Williams ^{14 ,} Reference Brattstrom, Wilcken and Ohrvik ^{15 )}.

The B-vitamins folate and riboflavin (along with the metabolically related B-vitamins, B₁₂ and B₆) are important nutritional determinants of homocysteine concentrations. Supplementation with folic acid, the synthetic form of folate, has been shown to lower elevated homocysteine concentrations by approximately 25% in the population generally^{( 16 )}. In the case of riboflavin, however, the homocysteine-lowering effects are confined to those with the MTHFR 677TT genotype. Riboflavin supplementation specifically in individuals with the TT genotype has been shown to significantly lower plasma homocysteine by 22%, and by as much as 40% in those with the lowest riboflavin status at baseline⁽ Reference McNulty, Dowey and Strain ^{17 )}. This response was not observed in individuals with CC or CT genotypes (even when baseline riboflavin status was suboptimal) despite a significant improvement in riboflavin status post-intervention in both genotype groups. Such evidence confirmed the independent modulating role of riboflavin in determining homocysteine concentrations specifically in individuals with the TT genotype as first suggested by earlier observational studies⁽ Reference Hustad, Ueland and Vollset ^{18 ,} Reference McNulty, McKinley and Wilson ^{19 )}.

Association between the MTHFR 677C→T polymorphism and CVD risk

The MTHFR 677C→T polymorphism previously received much attention as the main genetic determinant of homocysteine⁽ Reference Frosst, Blom and Milos ^{10 )} and is also independently associated with an increased risk of CVD and particularly stroke⁽ Reference Klerk, Verhoef and Clarke ^{1 ,} Reference Holmes, Newcombe and Hubacek ^{2 ,} Reference Lewis, Ebrahim and Smith ^{20 ,} Reference Wald, Law and Morris ^{21 )}. Several meta-analyses to date have demonstrated an increased risk of CVD in individuals with the TT genotype compared to those without this genetic variant⁽ Reference Klerk, Verhoef and Clarke ^{1 ,} Reference Holmes, Newcombe and Hubacek ^{2 ,} Reference Lewis, Ebrahim and Smith ^{20 –} Reference Casas, Bautista and Smeeth ^{22 )} (Table 1). Those studies which assessed the risk of CHD events estimated a 14–16% higher risk in individuals with the TT genotype compared to those with the CC genotype⁽ Reference Lewis, Ebrahim and Smith ^{20 ,} Reference Wald, Morris and Wald ^{23 )}. The most recent meta-analysis considered published and unpublished results in the area and estimated that the MTHFR 677TT genotype carried a 15% excess risk of heart disease (based on published results)⁽ Reference Clarke, Bennett and Parish ^{24 )}.

Table 1. Meta-analyses that investigated the MTHFR 677C→T polymorphism and CVD risk by geographical region

* Reported after national folate supplementation in the USA, Canada, Australia, New Zealand and some but not all European countries.

Differences in B-vitamin status may in some way explain the large geographical variation in the extent of the excess risk, which has previously been reported in relation to this polymorphism (Table 1). Klerk et al. ⁽ Reference Klerk, Verhoef and Clarke ^{1 )} and Lewis et al. ⁽ Reference Lewis, Ebrahim and Smith ^{20 )} found no strong evidence to support an association between the TT genotype and risk of CHD among North American populations; however, the polymorphism was linked to an increased risk of CHD among Asian and Middle Eastern populations. It was concluded that such geographical differences reflected the influence of environmental factors (primarily folate status) on the interaction between the polymorphism and CVD risk⁽ Reference Holmes, Newcombe and Hubacek ^{2 ,} Reference Kauwell, Diaz, Yang and Bailey ^{25 )}. In terms of stroke, the TT genotype was associated with an increased risk in regions of low folate only; whereas in populations of optimal folate status or in regions of mandatory folic acid fortification, the effect of the TT genotype on stroke risk was null⁽ Reference Holmes, Newcombe and Hubacek ^{2 )}. Somewhat surprisingly however, recent prospective evidence from the National Health and Nutrition Examination Survey cohort reported that the MTHFR 677TT genotype was associated with a lower rate of CVD mortality (OR 0·69 (95% CI 0·50, 0. 95)), although stratified analysis found that this association only occurred after the introduction of mandatory folic acid fortification⁽ Reference Yang, Bailey and Clarke ^{26 )}.

Overall, the evidence supports a modulating role of folate on CVD risk in this genetically at-risk group. However caution is required when interpreting the results from some observational studies. Further large-scale genetic studies, particularly among populations with sub-optimal folate status, are required to fully investigate the extent of CVD risk associated with this polymorphism and to consider the role of metabolically related B-vitamins linked with folate.

MTHFR 677C→T polymorphism and blood pressure

Hypertension, defined as a blood pressure (BP) of 140/90 mmHg or greater, is a major risk factor for CVD, with uncontrolled hypertension associated with an almost threefold greater risk of developing CVD⁽ Reference Wang, Lee and Fabsitz ^{27 )}. Furthermore, elevated BP is recognised as an even stronger predictor of stroke. A modest lowering of systolic BP by 2-mmHg is estimated to decrease cardiovascular risk by 10%, whereas a lowering of 10-mmHg is associated with a 40% reduction in stroke mortality⁽ Reference Lewington, Clarke and Qizilbash ^{28 )}.

Approximately 30–60% of BP variability is considered to be inherited⁽ Reference Mongeau ^{29 ,} Reference Levy, DeStefano and Larson ^{30 )} and several studies have focused on investigating genetic variants associated with elevated BP. Recent genome-wide association studies have investigated associations between single-nucleotide polymorphisms and elevated BP⁽ Reference Newton-Cheh, Johnson and Gateva ^{3 ,} Reference Ehret, Munroe and Rice ^{31 )}. One such study, which included data from 34 433 individuals, reported that the MTHFR loci was one of eight loci related to BP; however, the authors noted that there may be other genetic variants linked to the BP located in this region which have not been considered or are still unknown⁽ Reference Newton-Cheh, Johnson and Gateva ^{3 )}. Several subsequent investigations of a genome-wide association studies nature have confirmed the association between the MTHFR gene and BP variability⁽ Reference Liu, Li and Qi ^{32 –} Reference Takeuchi, Isono and Katsuya ^{34 )}.

The evidence from case–control studies suggests that there is a graded association between the MTHFR 677T allele and BP risk among hypertensive populations across different ethnic groups. An early case–control study in a Caucasian–Australian population reported a significantly higher frequency of the TT genotype in cases with hypertension compared to controls (after adjustment for BMI)⁽ Reference Heux, Morin and Lea ^{35 )}. Within an Asian–Indian population, a nearly fourfold increased risk of hypertension in individuals with the TT genotype was reported⁽ Reference Poduri, Kumari and Jain ^{36 )}. Interestingly, in a large population-based study of over 3000 Japanese people, the TT genotype was associated with a 42% increased risk of hypertension in women; however no significant association was shown in men⁽ Reference Inamoto, Katsuya and Kokubo ^{37 )}. As the evidence accumulated to support an association between the MTHFR 677C→T polymorphism and hypertension risk, a number of meta-analyses were carried out. One meta-analysis of thirteen case–control studies⁽ Reference Qian, Lu and Tan ^{38 )} confirmed a significant association between the TT genotype and hypertension among the Asian and the Caucasian populations, albeit significant heterogeneity was reported (OR 1·24 (95% CI 1·02, 1·50)). The most recent meta-analysis, which attempted to address the issue of heterogeneity by focusing specifically on Chinese population studies only, reported a 87% increased risk of hypertension in individuals with the TT genotype compared to those with the CC genotype⁽ Reference Niu, You and Qi ^{8 )}. Similarly, case–control studies carried out in hypertensive patients with various co-morbidities such as diabetes⁽ Reference Koubaa, Nakbi and Smaoui ^{39 )}, coronary artery disease⁽ Reference Benes, Kankova and Muzik ^{40 )} and chronic renal failure⁽ Reference Koupepidou, Deltas and Christofides ^{41 )} have also reported significant associations between this polymorphism and hypertension.

In addition to lifestyle and dietary modifications, improvements in the management of BP are largely considered to be due to the development of effective antihypertensive medications. Such medications are broadly categorised into five main classes (diuretics, β-blockers, calcium channel blockers, angiotension converting enzyme inhibitors and angiotension II receptor blockers), all of which have slightly different mechanisms of action. Only a limited number of studies, however, have investigated whether the BP-lowering effect of antihypertensive medications is influenced by the MTHFR 677C→T polymorphism. Jiang et al. ⁽ Reference Jiang, Hsu and Xu ^{42 )} reported an increased diastolic BP response to treatment with an angiotension converting enzyme inhibitor in hypertensive patients with the TT genotype compared to those without the polymorphism. More recently, in a larger cohort of hypertensive patients with the TT genotype (who expressed the phenotype of elevated BP at baseline), an increased diastolic BP response to treatment with an angiotension converting enzyme inhibitor was again reported (P=0·038)⁽ Reference Jiang, Yu and Venners ^{43 )}. Similarly, short-term treatment with a calcium channel blocker resulted in a significantly higher pulse wave velocity response in patients with the TT genotype compared with those with CC and CT genotypes (P=0·018); the polymorphism however did not affect the antihypertensive effects of the calcium channel blocker treatment⁽ Reference Xu, Zheng and Shen ^{44 )}. In contrast, other studies (reviewed later) suggest that the response to antihypertensive medication is generally suboptimal in individuals with the TT genotype compared to those with the CC or CT genotype. Given the limited evidence to date, further work is needed to explore potential interactions between the MTHFR 677C→T polymorphism and the antihypertensive effects of various BP medications.

MTHFR genotype and the modulating potential of B-vitamins in blood pressure control

Although strong evidence shows an association between the MTHFR 677C→T polymorphism and hypertension, the interrelationships between MTHFR genotypes, B-vitamin status and hypertension risk are not generally appreciated. In one small placebo-controlled, cross-over design study (n 41), Williams et al.⁽ Reference Williams, Kingwell and Burke ^{45 )} reported a decreased pulse pressure and arterial stiffness following short-term folic acid supplementation (5 mg/d); an effect that was reported to be independent of homocysteine concentration or MTHFR genotype, but the lack of statistical power (i.e. only five subjects with the TT genotype were available for investigation) makes such conclusions questionable. A subsequent intervention study reported no BP response to short-term folic acid supplementation, but did observe an association between MTHFR T allele frequency and hypertension risk⁽ Reference Markan, Sachdeva and Sehrawat ^{46 )}. Owing to the opportunistic nature of such studies (and therefore the lack of genotype-driven recruitment) such inconsistent results are not surprising. Similarly, in the large-scale trials of CVD risk involving folic acid intervention with or without other B-vitamins, no response of BP has been reported to date; however, no trial has analysed the results of BP (or other CVD outcomes) by MTHFR genotype⁽ Reference Wilson, McNulty and Scott ^{47 ,} Reference McNulty, Strain and Pentieva ^{48 )}.

Given the emerging evidence linking hypertension with this polymorphism⁽ Reference Newton-Cheh, Johnson and Gateva ^{3 )} and the established genotype-specific response of homocysteine to riboflavin supplementation⁽ Reference McNulty, Dowey and Strain ^{17 )}, recent research from our Centre has focused on investigating the BP-lowering potential of riboflavin in hypertensive individuals with the TT genotype. In a placebo-controlled trial, premature CVD patients (pre-screened for the MTHFR 677C→T polymorphism) were randomised within all three genotype groups to receive either riboflavin (1·6 mg/d) or placebo over a 16-week intervention period. Baseline BP was significantly higher in patients with the TT genotype compared to those with other genotypes; an effect that was most pronounced in those with the lowest baseline riboflavin status (as determined by the biomarker erythrocyte glutathione reductase activation co-efficient). Systolic and diastolic BP decreased significantly (by 13 and 8 mmHg, respectively) in those individuals with the TT genotype but no response was observed in the other genotype groups⁽ Reference Horigan, McNulty and Ward ^{49 )}. A follow-up, cross-over, randomised controlled trial reinvestigated the effect of riboflavin on BP in the same cohort of high-risk CVD patients 4 years after the original investigation⁽ Reference Wilson, Ward and McNulty ^{50 )}. Although there were marked changes in prescribed antihypertensive therapy over the 4-year period, the mean BP of those with the TT genotype (who originally received riboflavin supplementation for the 16-week period) had returned to pre-intervention levels. When the original treatment groups were reversed, the genotype-specific BP-lowering effect of riboflavin was again confirmed⁽ Reference Wilson, Ward and McNulty ^{50 )} with reductions of 9·2 (sd 12·8) mmHg (P=0·001) and 6·0 (sd 9·9) mmHg (P=0·003) in systolic and diastolic BP, respectively. Most recently, the responsiveness of BP to riboflavin supplementation in a separate cohort of treated hypertensive individuals (without overt CVD) with the TT genotype (n 91) was investigated⁽ Reference Wilson, McNulty and Ward ^{51 )}. At baseline, 60% of the participants had failed to achieve target BP levels (≤140/90 mmHg), despite taking three or more antihypertensive medications. Following riboflavin intervention (1·6 mg/d for 16 weeks), systolic BP decreased significantly, a finding which confirmed that the BP-lowering effects of riboflavin in individuals with the MTHFR 677TT genotype were not confined to high-risk CVD patients but also applied to hypertensive individuals generally⁽ Reference Wilson, McNulty and Ward ^{51 )}.

Although accumulating evidence highlights the BP-lowering effects of riboflavin in hypertensive individuals with the MTHFR 677C→T polymorphism, the mechanism to explain the role of this gene–nutrient interaction in BP remains unclear. It is possible however that the potent vasodilator nitric oxide may be involved. Decreased vascular concentrations of 5-methlytetrahydrolfolate, a nitric oxide regulator, have been reported in those with the MTHFR 677TT genotype⁽ Reference Antoniades, Shirodaria and Leeson ^{52 )}. By stabilising the variant enzyme via riboflavin supplementation, it is possible that normal concentrations of 5-methyltetrahydrofolate would be restored in vascular cells, thus improving the bioavailability of nitric oxide and resulting in BP-lowering specifically in those with the TT genotype. Further investigations are required to explore this and other potential mechanisms as well as the role of other B-vitamins involved in C₁ metabolism. Given the known metabolic interdependencies between the relevant B-vitamins, work is underway at this Centre to investigate the independent and combined effects of riboflavin and folic acid on BP in hypertensive individuals with the TT genotype in a factorial study design. In addition, although convincing evidence has demonstrated the BP-lowering effects of low-dose riboflavin over a relatively short period of time (1·6 mg/16 weeks)⁽ Reference Horigan, McNulty and Ward ^{49 –} Reference Wilson, McNulty and Ward ^{51 )}, it remains to be seen how variations in dose and duration would influence the BP-lowering effects of riboflavin in these genetically at-risk patients.

MTHFR genotype, B-vitamins and pregnancy outcomes

Well-established evidence has accumulated to suggest a link between the MTHFR 677C→T polymorphism and the risk of certain adverse pregnancy outcomes including NTD as well as hypertensive disorders such as preeclampsia and gestational hypertension; some of which will be the focus of the next section of this review.

The link between the B-vitamins and neural tube defects

NTD are of multi-factorial origin and involve both genetic and environmental factors. Landmark studies in the early 1990s confirmed the protective effects of periconceptional folic acid supplementation against the first occurrence⁽ Reference Czeizel and Dudas ^{4 )} and reoccurrence^{( 5 )} of NTD. Of note, Daly et al. ⁽ Reference Daly, Kirke and Molloy ^{53 )} illustrated a dose–response inverse relationship between maternal red cell folate (RCF) and NTD risk and identified maternal RCF concentrations of ≥906 nmol/l as the level associated with the lowest risk of an NTD affected pregnancy. Subsequently, folic acid supplementation levels of 400, 200 or 100 μg/d were reported to produce RCF concentrations, which predicted reductions in NTD by 47, 41 and 22%, respectively⁽ Reference Daly, Mills and Molloy ^{54 )}.

Present recommendations, in place in most countries worldwide, are that women planning a pregnancy should take a folic acid supplement of 400 μg/d for at least 1 month before conception and during the first trimester of pregnancy⁽ Reference Pitkin ^{55 )}. Reports have confirmed, however, that the uptake of this recommendation among women of child bearing age is typically poor⁽ Reference Nilsen, Vollset and Gjessing ^{56 ,} Reference McNulty, Pentieva and Marshall ^{57 )} and so it is not surprising that the NTD rates have changed very little across several European countries as a result of these folic acid recommendations⁽ Reference Botto, Lisi and Robert-Gnansia ^{58 )}. In contrast, the introduction of mandatory fortification of food with folic acid in the USA has successfully increased both serum and RCF concentrations among women of childbearing age^{( 59 )} which has coincided with a 27–50% reduction in the reported prevalence of NTD in the USA and Canada⁽ Reference Honein, Paulozzi and Mathews ^{60 ,} Reference De Wals, Tairou and Van Allen ^{61 )}. At a global level, folic acid fortification of food in countries worldwide has led to an estimated 46% reduction in the prevalence of NTD since 2008⁽ Reference Blencowe, Cousens and Modell ^{62 )}.

The role of the MTHFR 677C→T polymorphism in neural tube defects

To date, no single gene or genetic variant has been identified as a risk factor for NTD. Given the established protective effects however, of folic acid supplementation on the occurrence and reoccurrence of NTD, polymorphisms in genes encoding the proteins that are directly involved in folate metabolism, uptake and transport have been investigated. The reduced activity of the MTHFR enzyme in individuals with the MTHFR 677TT genotype, decreases the availability of 5-methyltetrahydrofolate which plays a key role in methylation⁽ Reference Frosst, Blom and Milos ^{10 )} and decreases global DNA methylation⁽ Reference Friso and Choi ^{63 ,} Reference Castro, Rivera and Ravasco ^{64 )}; therefore, it is no surprise that this specific mutant gene has been extensively investigated as a genetic risk factor for NTD⁽ Reference van der Put, Steegers Theunissen and Frosst ^{65 ,} Reference Blom, Shaw and den Heijer ^{66 )}.

Over the past two decades, several reports have shown a 2 to 4-fold increased risk of NTD if the infant or the mother has the MTHFR 677TT genotype⁽ Reference van der Put, Steegers Theunissen and Frosst ^{65 ,} Reference Ou, Stevenson and Brown ^{67 –} Reference Whitehead, Gallagher and Mills ^{70 )}. van der Put et al. ⁽ Reference van der Put, Trijbels, Hol, Mato and Caballero ^{71 )} reported an even higher risk of an NTD pregnancy if both the mother and the fetus were homozygous for the MTHFR 677C→T polymorphism. The large meta-analysis conducted by Botto and Yang⁽ Reference Botto and Yang ^{9 )} considered the frequency of the MTHFR 677C→T polymorphism and the risk of NTD across a number of different countries and ethnic groups. Overall, a 1·8 to 2-fold increased risk of spina bifida was established regardless of whether the infant or the mother was homozygous for the MTHFR 677C→T polymorphism⁽ Reference Botto and Yang ^{9 )}. However, other reports have indicated that the presence of the MTHFR 677TT genotype in the infant, and not in the mother, was the relevant consideration in NTD risk⁽ Reference Shields, Kirke and Mills ^{72 )}. Although Shaw et al. ⁽ Reference Shaw, Rozen and Finnell ^{73 )} established an association between the TT genotype and spina bifida risk, only a weak interaction between the infant genotype and maternal folic acid usage in the occurrence of spina bifida was reported, possibly suggesting that other contributing factors may also need to be considered. Studies focusing on maternal rather than infant MTHFR genotype have found that maternal folate status and/or maternal homozygosity for the MTHFR 677C→T genotype are important determinants of NTD risk⁽ Reference Martinez de Villarreal, Delgado-Enciso and Valdez-Leal ^{74 ,} Reference Volcik, Blanton and Tyerman ^{75 )}. Evidence from the aforementioned meta-analysis suggested that the excess NTD risk associated with the TT genotype may depend on folate status and/or a variant of other folate-related genes⁽ Reference Botto and Yang ^{9 )}. Christensen et al. ⁽ Reference Christensen, Arbour and Tran ^{76 )} demonstrated a 13-fold increased risk for spina bifida where the TT genotype was combined with RCF concentrations within the lowest quartile, a finding that strongly supports a gene–nutrient interactive effect being implicated in NTD risk, and also may explain some of the inconsistencies within the literature in this area.

Limited studies have investigated the influence of the MTHFR TT genotype in relation to present folic acid recommendations for the prevention of NTD. Preliminary evidence from a study conducted by our group reported that among women with the TT genotype, sampled at the fourteenth gestational week, significantly increased homocysteine and lower RCF concentrations were observed if folic acid supplementation (400 μg/d) was commenced prior to conception (i.e as recommended) compared with post-conception (i.e. more typical practice); such differences were not evident in the other genotype groups⁽ Reference Reilly, Ward and McNulty ^{77 )}. Furthermore, a large randomised controlled trial reported that despite 6 months supplementation with 4000 μg/d folic acid, women of childbearing age with the TT genotype achieved lower serum folate concentrations than did those with the CC genotype⁽ Reference Crider, Zhu and Hao ^{78 )}. Such evidence strongly suggests that folate requirements are higher in women with the MTHFR TT genotype and highlights the much greater need for women with this genotype to adhere to the specific recommendation of commencing folic acid prior to conception in order to optimise folate status for the prevention of NTD, although these findings need to be confirmed in larger studies among different populations.

Role of the MTHFR genotype in determining the risk of hypertensive disorders of pregnancy

Hypertensive disorders of pregnancy, such as gestational hypertension and preeclampsia affect 5–7% of all low-risk pregnancies worldwide and are the leading cause of maternal and fetal morbidity and mortality⁽ Reference Also-Rallo, Lopez-Quesada and Urreizti ^{79 )}. Gestational hypertension occurs without proteinuria (a predominant feature of preeclampsia), but increases the risk of developing preeclampsia by 15–26%⁽ Reference Saudan, Brown and Buddle ^{80 )}. Historical investigations reported an association between gestational hypertension and folate deficiency⁽ Reference Hibbard ^{81 –} Reference Stone ^{84 )}.

In the past two decades, a plethora of studies has investigated the effect of B-vitamin supplementation on hypertensive disorders of pregnancy. The evidence, however, remains inconsistent. Bodnar et al. ⁽ Reference Bodnar, Tang and Ness ^{6 )} and Hernandez et al. ⁽ Reference Hernandez-Diaz, Werler and Louik ^{7 )} reported that regular users of a multivitamin containing folic acid compared with non-users, had a reduced risk of developing preeclampsia and gestational hypertension by 45 and 55%, respectively. However, Ray and Mamdani⁽ Reference Ray and Mamdani ^{85 )} compared the rate of preeclampsia before and after population-wide fortification of food with folic acid in Canada and reported only a small reduction in the rate of preeclampsia (OR 0·96 (95% CI 0·94, 0·98)). It is possible that such inconsistencies may be explained to some extent by different frequencies of the MTHFR 677C→T genotype among study populations.

Hernandez-Diaz et al. ⁽ Reference Hernandez-Diaz, Wu and Hayes ^{86 )} found a significant association between the MTHFR 677TT genotype and risk of gestational hypertension and noted that this association may be weakened among women taking folic acid supplementation. Convincing evidence from meta-analyses has also reported a link between this polymorphism and the risk of hypertensive disorders of pregnancy. Qian et al. ⁽ Reference Qian, Lu and Tan ^{38 )} conducted a meta-analysis among Caucasian and Asian populations and reported a 1·66-fold increased risk of hypertension-in-pregnancy in T allele carriers; however, significant heterogeneity was noted. Subgroup analysis found that the T allele was associated with an increased risk of hypertension in Asians only, whereas no significant association was found in Caucasians. In general agreement with these findings, a subsequent meta-analysis that included studies in Chinese populations only, reported a significant association between the MTHFR 677C→T polymorphism and risk of hypertension-in-pregnancy⁽ Reference Niu, You and Qi ^{8 )}.

Apart from gestational hypertension, the closely related hypertensive-disorder of pregnancy, preeclampsia, has separately been investigated in relation to the MTHFR 677C→T polymorphism. In 1997, two separate case–control studies independently reported a significant association between the TT genotype and preeclampsia⁽ Reference Sohda, Arinami and Hamada ^{87 ,} Reference Grandone, Margaglione and Colaizzo ^{88 )}. Since then, numerous studies have investigated this association; however, conflicting results have emerged (Table 2). Worldwide, Mexico has the highest frequency of the TT genotype⁽ Reference Wilcken, Bamforth and Li ^{89 )} as well as a high prevalence of preeclampsia⁽ Reference Villar, Say and Shennan ^{90 )}. Within populations from two regions of Mexico, however, Perez-Mutul et al. ⁽ Reference Perez-Mutul, Gonzalez-Herrera and Sosa-Cabrera ^{91 )} and Davalos et al. ⁽ Reference Davalos, Moran and Martinez-Abundis ^{92 )} reported no significant association between the polymorphism and preeclampsia in pregnant women. In contrast, Canto et al. ⁽ Reference Canto, Buixeda and Palau ^{93 )} reported from another group of pregnant Mexican women, that the TT genotype (compared to the CC and CT genotypes) was associated with a decreased risk of preeclampsia, following adjustment for age and BMI. Such discrepancies between studies may suggest that the effect of the MTHFR 677TT genotype may be modulated by relevant dietary and environmental factors that could vary considerably among different populations.

Table 2. Studies that investigated the association between the MTHFR 677C→T polymorphism and preeclampsia risk

The association between the MTHFR 677C→T polymorphism and preeclampsia has been addressed in several meta-analyses. Kosmas et al. ⁽ Reference Kosmas, Tatsioni and Ioannidis ^{94 )} reported that the TT genotype compared with the CC genotype was associated with a 33% increased risk of preeclampsia, however significant between-study heterogeneity was found. Most recently, the meta-analysis by Xia et al. ⁽ Reference Xia, Chang and Cao ^{95 )} reported that the TT genotype compared with the CC and CT genotypes carried a significantly greater risk of preeclampsia (by 76%) among Asian women only, whereas in Caucasian women this increased risk was not evident. It is worth noting however that a much greater number of large-scale studies in this area have been conducted within Asian populations with far fewer studies conducted in Caucasian populations; a factor that limits the extent to which different populations can be compared. An additional limitation is that genotype-driven recruitment is generally not undertaken in these studies, a feature that is reflected by the relatively small numbers of pregnant women with the TT genotype being investigated and these raised the concern that many such studies may be statistically underpowered. It is not surprising therefore that in some cases the association between the MTHFR polymorphism and gestational hypertension is found to be relatively weak. In addition, the modulating effects of the polymorphism in relation to hypertension may be masked by the presence of other risk factors, such as reduced physical activity, increased BMI and gestational diabetes, which can develop during pregnancy and increase BP.