Maternal nutrition is paramount not only for the successful outcome of pregnancy, but it also has well proven effects in influencing the health and wellbeing of both mother and child in later years. In this respect, the optimal provision of B-vitamins during pregnancy is vital in building the foundation for normal development of the child and maintaining the health of the mother(Reference McDonald, Thorne-Lyman, Karakochuk, Whitfield, Green and Kraemer1,Reference Caffrey, McNulty and Irwin2) . This review will focus on the metabolically interrelated B-vitamins, folate, vitamin B12, vitamin B6 and riboflavin, and consider the physiological changes in pregnancy affecting the status of these vitamins and the implications for dietary B-vitamin recommendations in pregnancy. Additionally, we explore the emerging evidence for the less well recognised role of B-vitamins in preventing hypertensive disorders in pregnancy and the intergenerational effects of B-vitamins on offspring neurodevelopment and cognitive performance during childhood. The underpinning biological mechanisms are also considered.
Overview of B-vitamins within one-carbon metabolism
B-vitamins are chemically heterogeneous water-soluble organic compounds that cannot be endogenously synthesised by humans but are essential for normal physiological functions which requires their continuous supply in small amounts by the diet. In general, B-vitamins have small body pools at a cellular level and due to their fast turnover, deficiency can be developed rapidly under conditions of restricted dietary intake; in this regard, vitamin B12 is the only exception as the development of deficiency generally takes much longer time(Reference McNulty, Pentieva, Geissler and Powers3). B-vitamins are absorbed in the small intestine through elaborated mechanisms involving high-affinity transporters and/or receptors(Reference Said4). In addition, microbiota in the colon can synthesise some B-vitamins, however their contribution to B-vitamin status in humans is still unclear. B-vitamins act as cofactors in a wide range of diverse metabolic reactions which encompass almost all cellular functions.
This review focusses specifically on four B-vitamins interlinked through one-carbon metabolism (OCM) as detailed below. Folate and vitamin B12 are important for cell division and tissue growth through their involvement in pyrimidine and purine nucleotide synthesis for DNA and RNA, maintenance of genomic stability and methylation reactions(Reference Bailey, Stover and McNulty5–Reference Allen, Miller and De Groot7). Riboflavin is a cofactor in oxidation-reduction reactions essential for energy generation, intermediary metabolism and antioxidant defence system(Reference McNulty, Pentieva and Ward8). Vitamin B-6 is involved in amino acid metabolism, neurotransmitter synthesis, modulation of steroid hormone receptors and Hb synthesis(Reference Stach, Stach and Augoff9,10) .
Metabolic interaction of folate and related B-vitamins
Folate, vitamin B12, vitamin B6 and riboflavin interact by functioning as cofactors within OCM, a network of interrelated cellular pathways essential for numerous biological processes, including the biosynthesis of DNA, amino acid interconversions and methylation reactions(Reference Bailey, Stover and McNulty5). Folate is the core one-carbon donor and it acts synergistically with vitamin B12, vitamin B6 and riboflavin which are cofactors for some of the key enzymes in OCM (Fig. 1). Folate enters OCM as tetrahydrofolate (THF) which acquires a carbon unit from serine in a B6-dependent reaction to form 5,10-methyleneTHF. The latter is then either irreversibly reduced to 5-methylTHF or serves as the one-carbon donor in the synthesis of nucleic acids, where it is required by thymidylate synthetase in the conversion of deoyxuridine to deoxythymidine for pyrimidine biosynthesis or is converted to other folate cofactor forms essential for purine biosynthesis. Methylenetetrahydrofolate reductase (MTHFR) is a riboflavin (FAD)-dependent enzyme that catalyses the irreversible reduction of 5,10-methyleneTHF to 5-methylTHF. Within the methionine cycle, 5-methylTHF acts as a co-substrate in the remethylation of homocysteine to methionine via the B12-dependent enzyme methionine synthase. Methionine is activated by adenosine triphosphate leading to the generation of S-adenosylmethionine (SAM), which is a ubiquitous methyl group donor as it donates methyl groups in transmethylation reactions involved in the synthesis or activation of a wide range of bioactive compounds such as DNA, proteins, neurotransmitters, and membrane phospholipids, all of which are regulators of important physiological processes(Reference Bailey, Stover and McNulty5). As a result, SAM is converted subsequently to S-adenosylhomocysteine and homocysteine which in turn, can be remethylated back to methionine or can be further metabolised in the transsulfuration pathway to form cysteine, in a B6-dependent process(Reference Bottiglieri11).
Overview of the role of B-vitamins in one-carbon metabolism. DHF, dihydrofolate; DHFR, dihydrofolate reductase; DMNT, DNA methyltransferase; dTMP, deoxythymidine monophosphate; dUMP, deoxyuridine monophosphate; MTHFR, methylenetetrahydrofolate reductase; SAH, S-adenosylhomocysteine; SAM, S-adenosylmethionine; THF, tetrahydrofolate.
A balanced and adequate supply of folate, vitamin B12, vitamin B6 and riboflavin as well as some other nutrients (i.e. choline) is required for the effective functioning of OCM. If the provision of some of these vitamins is insufficient, then a compensatory shift in pathways within OCM may occur(Reference Fernàndez-Roig, Cavallé-Busquets and Fernandez-Ballart12) but also it might have a negative impact on SAM synthesis with impairment of transmethylation reactions(Reference Bottiglieri11). In addition, genetic polymorphisms on B-vitamin-dependent enzymes in OCM, can result in disruption of OCM pathways which can lead to adverse phenotypes, even if dietary folate and related B-vitamin intakes are adequate(Reference Newton-Cheh, Johnson and Gateva13). For example, individuals homozygous for the common MTHFR C677T polymorphism have a reduced activity of MTHFR enzyme, impaired folate metabolism and high homocysteine concentrations and may have higher requirements for folate and riboflavin(Reference McNulty, Dowey and Strain14). Meta-analyses showed that the MTHFR C677T polymorphism increases the risk of high blood pressure and hypertension in pregnancy by 24–87 %(Reference Qian, Lu and Tan15–Reference Yang, Fan and Zhi18), which will be discussed in more depth later in this review.
Physiological changes in pregnancy affecting B-vitamin biomarkers
Numerous physiological changes occur in all the systems in the body during normal pregnancy with the aim of providing optimal conditions for the growth and development of the fetus and supporting the placental unit. These changes are predominantly driven by the dramatic increase in the production of progesterone and oestrogen(Reference Macklon and Brosens19). The corpus luteum, that develops from an ovarian follicle during the luteal phase of the menstrual cycle, is initially responsible for the synthesis of these hormones but with the advancement of pregnancy their production is shifted from the corpus luteum to the placenta by the 7th–9th week of gestation(Reference Tal and Taylor20). Most of the oestrogen and progesterone produced by the placenta enter maternal circulation and are exported to the other maternal tissues. The brief overview below is focused on physiological changes in pregnancy which potentially affect B-vitamin status biomarkers in pregnant women.
Plasma volume and red blood cell mass
A measurable expansion of plasma volume already exists at the 6th–8th week of gestation and this increase continues up to the 34th–36th week of gestation(Reference Hytten21). In a singleton pregnancy the total increase of plasma volume accounts to 1200–1400 ml which is around 40 % above the volume of a non-pregnant woman, whereas in a multiple pregnancy this expansion is much bigger(Reference Pinheiro and Stika22). A simultaneous rise of red blood cell (RBC) count by 30 % is also observed in pregnancy, however due to the higher increase of plasma volume than RBC mass these changes lead to ‘physiological anaemia’ of pregnancy which usually becomes evident in the third trimester(Reference Costantine23). The reason for the increased blood volume is still unclear but one of the mechanisms under consideration relates to oestrogen activation of the renin-angiotensin-aldosterone system which promotes water and sodium retention and subsequently leads to hypervolemia(Reference Easterling, Benedetti and Schmucker24). The pregnancy induced hypervolemia generally is considered beneficial for the pregnant woman as it maintains the hemodynamic stability with the blood loss at the time of delivery(Reference Carbillon, Uzan and Uzan25). It also might be advantageous for the fetus as the less viscous blood would have an improved uterine and intervillous perfusion with more efficient transport of nutrients and oxygen to the fetus. However, the hypervolemia in pregnancy could lead to decreased concentrations of various blood constituents including B-vitamins. Moreover, the increased production of RBC is associated with higher requirements for folate, vitamin B12, vitamin B6 and riboflavin due to their involvement either in the generation of new cells or in the synthesis of Hb, which in turn would further compromise the status of these B-vitamins.
Plasma protein binding
The concentrations of different plasma binding proteins are reported to decrease during pregnancy, mainly because of the dilutional effect of plasma volume expansion(Reference Mendenhall26). Plasma albumin concentrations are up to 80 % lower compared with those in non-pregnant women(Reference Pinheiro and Stika22). The fall of plasma albumin is of relevance to the concentrations of folate, vitamin B6 and riboflavin which are normally bound to albumin in the circulation(Reference Bailey, Stover and McNulty5,Reference Ink and Henderson27,Reference Innis, McCormick and Merrill28) . Some of the protein unbound (free) B-vitamins are reported to be prone to hydrolysis and destruction(Reference Anderson, O’brien and Griffin29) which may also contribute to their low blood concentrations in pregnancy.
Renal clearance and glomerular filtration rate
Glomerular filtration rate and renal clearance start to increase in the middle of the first trimester and continue gradually to rise up to the third trimester with a total increase of 30–50 %(Reference Odutayo and Hladunewich30). However, it has been reported that the renal reabsorption process is unchanged in pregnancy(Reference Frederiksen31) suggesting that there is no loss of important nutrients through the urine. In support of this are the results from a small but tightly controlled intervention study with stable isotopes which showed that the excreted amounts of total folate as well as folate catabolites in 24-h urinary samples were similar in second-trimester pregnant women and non-pregnant controls who were maintained for 12 weeks on a diet providing folate at either 450 µg/d or 850 µg/d(Reference Gregory, Caudill and Opalko32). According to our knowledge, there are not such precise investigations with stable isotopes conducted for the other B-vitamins, however a cross-sectional study in age matched (non-supplement users) pregnant, lactating and non-pregnant women with B-vitamin intake within the recommended levels, did not find significant differences among the three groups of women in the 24-hour urinary excretion of vitamin B6, riboflavin and vitamin B12 (Reference Shibata, Fukuwatari and Sasaki33). These findings indicate that despite the considerable increase of glomerular filtration rate, it is unlikely that urinary B-vitamin excretion is affected by any way throughout pregnancy.
Gastric changes
Pregnancy is associated with an increase of gastric pH and hypochlorhydria(Reference Gryboski and Spiro34) which theoretically may affect the bioavailability of some B-vitamins by reducing their release from foods where they are trapped in the food matrix bound to macromolecules. Furthermore, the change in the pH may affect the binding ability of some of the carriers/receptors involved in the active transport of B-vitamins in the upper small intestine which in turn might result in reduced B-vitamin bioavailability. Malabsorption of folate and vitamin B12 has been reported in non-pregnant individuals and older adults with hypochlorhydria and atrophic gastritis in different studies(Reference Russell, Krasinski and Samloff35,Reference Carmel36) . However, these chronic conditions are too extreme and the parallel with the situation in pregnancy is probably unreasonable. In order to examine whether the reproductive state has an impact on the response to folate intervention, West et al. (2012) conducted a controlled feeding trial in age matched pregnant, lactating and non-pregnant women maintained for 10–12 weeks on diets containing equivalent amounts of folate (400 µg/d natural folate and a supplement of 750 µg/d) and demonstrated comparable serum folate responses among the three groups, suggesting that there are insignificant changes in the intestinal absorption of folate during pregnancy(Reference West, Yan and Perry37). It is unknown whether the same results would be obtained if the supplemental folic acid intake was in line with the recommendation of 400 µg/d. The same research group also investigated the effects of pregnancy and lactation on vitamin B-12 status response to a controlled vitamin B12 intake of 8·6 µg/d provided for 10–12 weeks(Reference Bae, West and Yan38). The results of this trial showed that at the end of the intervention despite the equivalent B12 intakes pregnant women had 21 % lower serum B12 concentrations compared with the non-pregnant controls but the concentrations of holotranscobalamin (biologically active fraction of B12) were not significantly different between the two groups. However, it is unknown whether these changes in vitamin B12 biomarkers of status are related to altered absorption. According to our knowledge, there are no published human studies investigating the impact of pregnancy on the absorption and bioavailability of ingested vitamin B6 and riboflavin. Of note, a study using labelled pyridoxine in pregnant and non-pregnant control rats did not find significant differences in the intestinal absorption of radioisotope between the two groups(Reference Wang and Trumbo39). Further research is required to clarify whether the absorption of B-vitamins is affected in pregnancy.
Transfer of B-vitamins to placenta and fetus
The placenta is actively involved in the transport of nutrients including B-vitamins from maternal circulation to the fetus. Folate transporters such as folate receptor-α, folate reduced carrier and proton coupled folate transporter are found to be expressed in the syncytiotrophoblast from the early stages of placenta development(Reference Yasuda, Hasui and Kobayashi40,Reference Solanky, Requena Jimenez and D’Souza41) . Transcobalamin receptor which is responsible for the uptake of biologically active fraction of B12, holotranscobalamin, is also expressed in human placenta(Reference Quadros, Nakayama and Sequeira42). RFVT3 protein in placenta has been associated with the transfer of riboflavin derivatives FAD and FMN to the fetus(Reference Jin and Yonezawa43). As a result of the activity of these placental transporters, the concentrations of folate, vitamin B12 and riboflavin derivatives in the intervillous blood and in the fetus are 2–4 times higher than those in maternal blood(Reference Thorand, Pietrzik and Prinz-Langenohl44–Reference Zempleni, Link and Bitsch46). Although carrier proteins and receptors in relation to vitamin B6 have not been identified in placenta yet, studies have reported up to 4 times higher plasma PLP concentrations in the newborn compared with those in their mothers(Reference Contractor and Shane47–Reference Bjørke-Monsen, Varsi and Sakkestad49), suggesting the involvement of some unknown active mechanism that transfers vitamin B6 against its concentration gradient. The substantial flow of B-vitamins from the mother to the fetus might represent a physiological adaptation to satisfy the high demands for these micronutrients during fetal development but with potential implications for maternal B-vitamin status.
Changes of activities of enzymes involved in OCM
Substantial and dynamic changes in the metabolic rates of the pathways in OCM have been reported in women with uncomplicated pregnancy. A study using tracer isotopes of methionine and phenylalanine showed that compared with non-pregnant women, the rate of transsulfuration was higher in the first trimester of pregnancy, however the differences disappeared towards the late gestation(Reference Dasarathy, Gruca and Bennett50). The higher activity in the maternal transsulfuration pathway in early pregnancy is probably a compensatory reaction in response to the high demands for cysteine during the fetal development coupled with the inability of the fetus to synthesise cysteine on its own due to the absence of transsulfuration activity in the fetal liver(Reference Kalhan and Bier51). The study by Dasarathy et al. (2010) also showed a higher rate of transmethylation of methionine in the late gestation compared to non-pregnant women which may be related to high methylation demands by the growing fetus and the placenta(Reference Obeid, Morkbak and Munz45). The changes in the activities of the OCM pathways ultimately suggest that the requirements for folate and metabolically related B-vitamins acting as cofactors within OCM may be also increased.
All these changes in pregnancy may lead to a decline in the concentrations of B-vitamin biomarkers in maternal circulation. The understanding of these changes and their profound impact on B-vitamin status and requirements in pregnancy is essential for optimising maternal and fetal health.
B-vitamin biomarkers during pregnancy
Longitudinal studies of non-supplemented pregnant women at different stages in pregnancy generally show a decline of B-vitamin status with the advancing gestation (Fig. 2). Serum and RBC folate concentrations were reported to decrease by up to 50 % of those in non-pregnant state(Reference Ek and Magnus52–Reference Milman, Byg and Hvas54). In contrast serum folate in the umbilical cord vein was reported to be around two times higher than in the maternal circulation(Reference Obeid, Munz and Jäger55).
B-vitamin status biomarkers in maternal and umbilical cord blood in unsupplemented pregnant women. Levels of B-vitamin biomarkers in pregnancy are presented as percentage of the concentrations of these biomarkers in non-pregnant, non-lactating unsupplemented women (punctured line). Data are based on: Kerr et al. (2005)(Reference Kerr, McNulty and Scott133); Milman et al. (2006a)(Reference Milman, Byg and Hvas54) and Obeid et al. (2005)(Reference Obeid, Munz and Jäger55) for serum folate; Kerr et al. (2005)(Reference Kerr, McNulty and Scott133); McNulty et al. (2013)(Reference McNulty, McNulty and Marshall124); Molloy et al. (2013)(Reference Molloy, Pentieva and McNulty134) for serum vitamin B12; Bjorke-Monsen et al. (2023)(Reference Bjørke-Monsen, Varsi and Sakkestad49) and Zempleni et al. (1992)(Reference Zempleni, Link and Kübler48) for plasma pyridoxal 5´-phosphate (PLP); Bates et al. (1981)(Reference Bates, Prentice and Paul65) and Shaw et al. (1993)(Reference Shaw67) for erythrocyte glutathione reductase activation coefficient (EGRac).
Several research groups have found a progressive fall of approximately 30 % in serum vitamin B12 during pregnancy(Reference Koebnick, Heins and Dagnelie56–Reference Milman, Byg and Bergholt58) which returns to pre-pregnancy concentrations shortly after delivery(Reference Milman, Byg and Bergholt58). Importantly, holotranscobalamin, the fraction of vitamin B12 available for tissue uptake, was reported to remain relatively stable throughout pregnancy after a decline in early gestation(Reference Murphy, Molloy and Ueland57,Reference Greibe, Andreasen and Lildballe59) . The maintenance of stable holotranscobalamin in pregnancy is considered to be a result of mobilisation of maternal B12 pools to safeguard the adequate B12 supply to the fetus. In support of this is the fact that the concentrations of biomarkers of B12 function, methylmalonic acid and homocysteine increase in the later stages of pregnancy(Reference Murphy, Molloy and Ueland57,Reference Milman, Byg and Bergholt58) suggesting some level of metabolic B12 depletion in pregnant women.
Studies investigating vitamin B6 status throughout pregnancy have reported a dramatic decline of plasma pyridoxal phosphate (PLP) concentration with up to 70 % lower values in late pregnancy in comparison with the concentrations postpartum(Reference Bjørke-Monsen, Varsi and Sakkestad49) or in age matched non-pregnant controls(Reference Trumbo and Wang60). Some investigators found a compensatory increase of other vitamin B6 derivatives in maternal circulation corresponding to the fall of PLP(Reference Barnard, de Kock and Vermaak61) and unchanged urinary excretion of the catabolic product 4-pyridoxic acid(Reference Trumbo and Wang60) leading to the conclusion that there are no substantial changes in vitamin B6 status in pregnancy, however, this remains uncertain as others have failed to confirm these findings(Reference Shane and Contractor62). A recent study reported an increase of HKr index, a functional biomarker of vitamin B6 status based on the activities of B6-dependent enzymes in the kynurenine pathway, which indicates that there are some B6 insufficiency at the metabolic level during pregnancy(Reference Bjørke-Monsen, Varsi and Sakkestad49). The intensive transfer of vitamin B6 to the fetus demonstrated by up to 4 times higher concentrations of PLP in the umbilical cord vein than in maternal circulation in the late gestation(Reference Shane and Contractor62) is still considered the main reason for the decline of B6 status in pregnancy.
Riboflavin status, measured by the functional biomarker erythrocyte glutathione reductase coefficient (EGRac), has been found to decrease only marginally, up to 10 % during pregnancy, however this was reported to occur together with a progressive increase in the prevalence of riboflavin deficiency, especially towards the end of the third trimester(Reference Heller, Salkeld and Körner63–Reference Bates, Prentice and Paul65). The changes in riboflavin status have shown to be more pronounced with around 60 % prevalence of clinical deficiency in women with habitually low riboflavin intake in the range 0·5–1 mg/d(Reference Bates, Prentice and Paul65,Reference Kuizon, Natera and Alberto66) . Compared with the maternal EGRac values, the measurements of EGRac in the umbilical cord vein have indicated a better riboflavin status(Reference Shaw67).
In summary, the physiological changes in pregnancy with associated decline of B-vitamin status are well recognised, but specific reference ranges for B-vitamin biomarkers in pregnancy so far have not been developed. This greatly limits the timely diagnosis and management of B-vitamin deficiency in pregnant women.
Recommendations for B-vitamin intake in pregnancy
The recommendations for B-vitamin intake in pregnant women vary substantially between countries as different standards and criteria have been adopted (Table 1). Depletion-repletion studies, which are traditionally used for deriving dietary recommendations are inappropriate in pregnancy for health safety and ethical issues, thus other types of studies and indirect approaches need to be utilised.
Recommendations for dietary B-vitamin intakes during pregnancy
DH, Department of Health; RNI, recommended nutrient intake; IOM, Institute of Medicine; RDA, recommended dietary allowance; EFSA, European Food Safety Authority; DRV, dietary reference value.
RNI, RDA, DRV are defined as the average daily intake that is sufficient to meet the nutrient requirements of almost all (97·5 %) healthy pregnant women. Figures in brackets indicate the increment added to the value for non-pregnant non-lactating women.
* DFE, dietary folate equivalents.
† µg/d.
‡ Adequate Intake (AI).
In view of the intensive tissue growth during pregnancy, health authorities worldwide have established higher folate intake recommendations for pregnant women than for non-pregnant/non-lactating women. Red blood cell folate concentration has been used as the main criterion to assess intake adequacy but not all authorities have expressed the recommended intake values as dietary folate equivalents (DFE) which considers the differences in the bioavailability of food folates and folic acid, the synthetic folate derivative used for supplementation and food fortification. Most of the authorities recommended an intake of 600 µg/d DFE based on a controlled metabolic study(Reference Caudill, Cruz and Gregory68). However, the UK Department of Health (1991) has derived its recommendations for folate intake in pregnancy at 300 µg/d(69) using evidence available at that time from several historic intervention studies with different folic acid doses(Reference Hansen and Rybo70,Reference Chanarin, Rothman and Ward71) but there are some concerns that the requirements might be underestimated as folate intake through the diet in these studies has not been accurately measured. Of note is that these recommendations do not include the required amount of folic acid supplementation for neural tube defect (NTD) prevention.
The requirements for vitamin B12 intake in pregnancy set by most official authorities are based on the recommendations for adults with an increment (0·2–0·5 μg/d) to compensate for B12 accumulated by the fetus(72–74). That extra amount was estimated by liver B12 content in infants born to mothers with an adequate B12 status and assuming that liver contains half of B12 in the body(Reference Baker, Jacob and Rajan75–Reference Loría, Vaz-Pinto and Arroyo77). However, other authorities, have not considered any B12 increment in pregnancy with the view that the requirement for B12 in adults should be also sufficient for pregnant women who are with adequate body stores at the start of pregnancy(69).
Most, but not all, authorities consider that pregnant women have increased requirements for vitamin B6 as a result of the intensive transfer of B6 to the fetus. The required increment of 0·4–0·5 mg/d added to the recommended value for non-pregnant women has been estimated based on the gestational weight gain, B6 tissue content and the vitamin bioavailability from food(10,72) . Some have taken the view that B6 requirements in pregnancy are not different from those in non-pregnant women since the observed changes of vitamin B6 biomarkers in pregnancy are physiological and are not related to any obvious signs of deficiency(69).
The increased requirements for riboflavin intake in pregnancy are generally based on the progressive decline of riboflavin status throughout gestation(Reference Bates, Prentice and Paul65) and rely heavily on the evidence from one small intervention study in Filipino pregnant and non-pregnant women(Reference Kuizon, Madriagà and Perlas78). The amount of 0·3 mg/d in addition to the requirements of non-pregnant women has been estimated to be sufficient for the normal growth of maternal tissues and the fetus(69,72,73,79) .
Considering the lack of standardised approach for deriving dietary recommendations for B-vitamin intakes in pregnancy, and the fact that most existing recommendations are based on extrapolation from values for non-pregnant women and estimated needs for tissue growth and fetal transfer, there is a need for well-designed controlled feeding trials to be conducted in pregnancy. These will facilitate the generation of evidenced-based B-vitamin requirements during pregnancy to help support successful pregnancy outcomes without compromising maternal health.
Clinical implications of B-vitamins in pregnancy
B-vitamins play an important role in maintaining maternal health, fetal development and successful delivery. Irrefutable evidence from over 30 years ago showed that maternal folic acid supplementation in early pregnancy could protect against NTD(80,Reference Czeizel and Dudás81) , leading to global recommendations for supplementation of women with folic acid at 400 µg/d from preconception to the end of the first pregnancy trimester. NTD are major birth defects that occur when the neural tube fails to close properly in the first four weeks of gestation, leading to death of the fetus or newborn, or to various disabilities involving the spinal cord, the most common form of which is spina bifida. The exact mechanism for the protective effect of folic acid against NTD is unknown but it is believed that it is through OCM. In support of this hypothesis there is evidence from several reports, although only observational, indicating that other methyl donors as well as B-vitamins involved in OCM are also implicated in the risk of NTD. For example, low maternal B12 status was reported to be associated with an increased risk of NTD-affected pregnancy independently of folate(Reference Molloy, Kirke and Troendle82,Reference Ray, Wyatt and Thompson83) . Furthermore, a recent population-based, case-control study of more than 40 000 US pregnancies showed that the concurrent high consumption of vitamin B6, vitamin B12, choline, betaine and methionine in addition to folic acid was associated with a higher NTD reduction compared with the increased intake of a single nutrient(Reference Petersen, Smith-Webb and Shaw84). Thus, the effectiveness of NTD prevention by folic acid supplementation could be further improved with the intakes of B-vitamins and other methyl donors involved in OCM. B-vitamin deficiencies are also associated with various adverse pregnancy outcomes and complications including anaemia(Reference Metz85–Reference Finkelstein, Fothergill and Venkatramanan89), early pregnancy loss(Reference Finkelstein, Fothergill and Venkatramanan89–Reference Diao, Luo and Li91), hypertension in pregnancy and preeclampsia(Reference Liu, Liu and Wang92–Reference Duffy, McNulty and Ward94), low birth weight and intrauterine growth restriction(Reference Relton, Pearce and Parker95). Furthermore, there is emerging evidence for the intergenerational effects of maternal B-vitamin status during pregnancy on the offspring development(Reference Finkelstein, Fothergill and Venkatramanan89,Reference McNulty, Rollins and Cassidy96,Reference Caffrey, McNulty and Rollins97) . The impact of B-vitamins on hypertensive disorders of pregnancy (HDP) and offspring neurodevelopment during childhood are considered below.
Impact on hypertensive disorders of pregnancy (HDP)
HDP are serious pregnancy complications that include chronic hypertension, gestational hypertension, preeclampsia/eclampsia, and preeclampsia superimposed on chronic hypertension(Reference Garovic, Dechend and Easterling98). These conditions affect up to 15 % of pregnancies and are recognised as one of the leading causes of maternal and fetal death globally(Reference Kassebaum, Barber and Bhutta99). In addition, HDP are also related to an increased risk for cardiovascular and metabolic diseases in later life of both mother and child(Reference Garovic, Dechend and Easterling98,Reference Huang, Wei and Lee100) . The treatment of HDP is challenging as certain antihypertensive medications are associated with serious neonatal complications (intrauterine growth restriction, fetal bradycardia and distress and neonatal hypoglycaemia)(Reference Kattah and Garovic101); therefore, the identification of effective nondrug approaches for managing hypertension in pregnancy could be particularly beneficial.
There is considerable observational evidence linking B-vitamins involved in OCM with the risk of HDP. The use of folic acid supplements in pregnancy has been related to a decreased risk of gestational hypertension and preeclampsia(Reference Liu, Liu and Wang92,Reference De Ocampo, Araneta and Macera102) , whereas a systematic review of case-control studies showed that women with preeclampsia had significantly lower serum vitamin B12 concentrations than normotensive pregnant women(Reference Mardali, Fatahi and Alinaghizadeh103). A large observational study from our centre of over 2200 pregnancies found that riboflavin deficiency was associated with an almost 3-fold greater risk of developing hypertension during pregnancy(Reference Duffy, McNulty and Ward94). However, the evidence so far from RCTs for a causal effect of B-vitamins is weak (Table 2)(Reference Charles, Ness and Campbell104–Reference Salam, Zuberi and Bhutta108). Inconsistencies in the literature might be explained to some extent by genetic differences among populations.
Randomised trials investigating the effect of maternal B-vitamin supplementation on blood pressure in pregnancy
FA, folic acid; GW, gestational week; PE, preeclampsia; MMN, multiple micronutrient.
* Combined daily multivitamin comprised of 20 mg thiamine, 20 mg riboflavin, 25 mg vitamin B6, 50 µg vitamin B12, 500 mg vitamin C, 30 mg vitamin E, 0·8 mg FA v. Vitamin A (30 mg beta-carotene + 5000 μg preformed vitamin A/d).
† Combined daily MMN comprised of vitamin A 1000 μg, vitamin B6 2·2 mg, vitamin B12 2·2 μg, vitamin C 20 mg, vitamin E 400 μg, FA 0·4 mg, N-acetylcysteine 200 mg, Cu 2 mg, Zn 5 mg, Mn 0·5 mg, Iron 30 mg, calcium 800 mg, selenium 100 μg v. Control (Iron 30 mg, FA 0·4 mg).
Emerging evidence from genome-wide association studies (GWAS) and epidemiological research has linked the C677T polymorphism in the gene encoding the folate-metabolising enzyme MTHFR with an increased risk of hypertension and HDP by up to 87 %, as mentioned earlier in this review(Reference Qian, Lu and Tan15–Reference Yang, Fan and Zhi18). Individuals with the homozygous variant TT genotype in MTHFR appear to have reduced MTHFR enzyme activity and thus impaired folate metabolism and OCM. Riboflavin, in the form of FAD, acts as a cofactor of MTHFR and molecular studies show that the variant enzyme becomes inactive as a result of an increased propensity to dissociate from FAD(Reference Pejchal, Campbell and Guenther109,Reference Yamada, Chen and Rozen110) . New evidence points to a novel gene-nutrient interaction of MTHFR with riboflavin in relation to blood pressure. Notably, our study of over 6000 Irish non-pregnant adults found that a deficient riboflavin status exacerbated the genetic risk of hypertension throughout adulthood, with a three-fold excess risk when the TT genotype occurred in combination with deficient riboflavin status compared to those without this polymorphism and normal riboflavin status(Reference Ward, Hughes and Strain111). Furthermore, in three previous RCTs of non-pregnant hypertensive patients, with and without CVD, we demonstrated that intervention with low-dose riboflavin (1·6 mg/d) for 16 weeks resulted in a lowering of systolic blood pressure by up to 13 mmHg, specifically in those with the TT genotype(Reference Horigan, McNulty and Ward112–Reference Wilson, Mcnulty and Ward114). Together, these studies show that riboflavin plays an important role in modulating blood pressure phenotype in individuals with the variant TT genotype in MTHFR. This novel gene-nutrient interaction with implications for blood pressure management is important in public health terms, given that the frequency of the variant MTHFR 677TT genotype is 10–12 % in the UK and Ireland but it varies according to the geographical area with the highest frequency of 32–44 % is reported in Mexico(Reference Wilcken, Bamforth and Li115–Reference Reyes, Godfrey and Ming117). Moreover, low to deficient riboflavin status is emerging as a much more widespread issue across the developed world than has previously been recognised(Reference McNulty, Pentieva and Ward8,Reference Troesch, Hoeft and McBurney118) . Given the increased risk of HDP in pregnant women with MTHFR TT genotype, and the blood pressure-lowering effect of riboflavin in non-pregnant adults with this genotype, it is worth exploring further whether targeted supplementation with riboflavin in pregnancy has benefits as an effective preventive strategy to reduce the risk of HDP and decrease the burden of related adverse pregnancy outcomes in genetically at-risk women.
Impact on offspring neurodevelopment
There is emerging evidence linking maternal folate and vitamin B12 during pregnancy with offspring neurodevelopment and cognitive function in childhood. In terms of folate, two systematic reviews of observational studies concluded that maternal folate status or self-reported folate intake or usage of folic acid supplements were associated with improved cognitive function and reduced risk of behavioural and language problems of the child(Reference Veena, Gale and Krishnaveni119,Reference Chen, Qin and Gao120) . Of note is also that the supplementation with folic acid at doses higher than 1000 µg/d was negatively associated with global verbal scores in children(Reference Valera-Gran, Navarrete-Muñoz and Garcia de la Hera121). These findings are in contrast with the results of a noteworthy RCT of mother-child pairs which showed that periconceptional supplementation with a multivitamin product containing folic acid did not have impact on the child cognitive function at 11 months, 2 and 6 years(Reference Czeizel and Dobo122,Reference Dobo and Czeizel123) (see Table 3 for the summary of the RCTs in this area). However, a RCT from this centre, the Folic Acid Supplementation in the Second and Third Trimesters (FASSTT) trial, investigated the effects of extended maternal folic acid supplementation (at the recommended dose of 400 μg/d) in trimesters 2 and 3 of pregnancy, and thus beyond the periconceptional period recommended for preventing NTD(Reference McNulty, McNulty and Marshall124). FASSTT trial found direct evidence for positive effects of folic acid on various cognitive domains in the children at age 3, 7 and 11 years and particularly in the verbal domain at age 7 and 11 years(Reference McNulty, Rollins and Cassidy96,Reference Caffrey, McNulty and Rollins97) . In addition, an objective assessment of the brain function by magnetoencephalography of the FASSTT offspring at 11 years suggested more efficient processing of language in children from folic acid-supplemented mothers(Reference Caffrey, McNulty and Rollins97). Thus, maintaining optimal maternal folate throughout pregnancy, well beyond the early period known to be protective against NTD, may be beneficial for brain development and functioning in the child. Furthermore, the dose of folic acid supplementation appears to be of importance as higher doses than recommended might have a negative impact on offspring neurodevelopment.
Randomised trials investigating the effect of maternal B-vitamin supplementation and cognitive performance of the offspring
FA dosage is 0·4 mg/d, unless otherwise stated. FA, folic acid; MMN, multiple micronutrient; GW, gestational week; UNIT, Universal Nonverbal Intelligence Test; MABC, Movement Assessment Battery for Children; 5-MTHF, 5-methyltetrahydrofolate; K-ABCKM, Kaufman Assessment Battery for Children; EEG, electroencephalography; WPPSI, Weschler Preschool and Primary Scale of Intelligence; WISC, Weschler Intelligence Scale for Children; MEG, magnetoencephalography; BSID, Bayley Scales of Infant and Toddler Development; ERP, Event-related potentials; MMA, methylmalonic acid.
In relation to vitamin B12, a systematic review of observational studies (n 7) showed inconsistent results with investigations from developing countries demonstrating reduced cognitive function in children of mothers with deficient B12 status or with low B12 intakes whereas there was no evidence of an association between self-reported maternal B12 intake and offspring cognitive performance in the cohorts from developed countries(Reference Veena, Gale and Krishnaveni119). In support of these findings are the results from a systematic review that focused on studies conducted in India, where the prevalence of vitamin B12 deficiency among pregnant women is up to 70 % and is considered one of the highest in the world(Reference Behere, Deshmukh and Otiv125). The observational studies (n 4) included in this review supported associations of lower maternal B12 and poorer cognitive functions in the child, whereas one RCT showed a beneficial effect of B12 supplementation on offspring language abilities at 30 months but no effect at 9 months(Reference Behere, Deshmukh and Otiv125) (Table 3). Similar results were obtained in a recent Cochrane systematic review on RCTs (three ancillary analyses of one trial) which concluded that vitamin B12 supplementation during pregnancy improved offspring B12 status that may have important health impacts in later life but its effect on child cognitive function is still inconclusive and requires further research in larger scale trials(Reference Finkelstein, Fothergill and Venkatramanan89).
The biological mechanism linking folate and B12 during pregnancy with offspring neurodevelopment is likely to be via the roles of these vitamins in OCM and the generation of SAM, which in turn is essential in donating methyl groups in numerous transmethylation reactions including the synthesis of neurotransmitters and DNA methylation(Reference Bailey, Stover and McNulty5,Reference Allen, Miller and De Groot7) . DNA methylation is one of the key epigenetic mechanisms for regulation of gene expression which is currently a subject of intensive research and it was reported that maternal deficiency of folate and B12 could result in aberrant gene expression with consequential health outcomes(Reference Irwin, Pentieva and Cassidy126,Reference Monasso, Hoang and Mancano127) . By using a candidate gene approach, FASSTT trial, conducted in this centre, demonstrated that folic acid supplementation through trimesters 2 and 3 of pregnancy led to significant changes in DNA methylation in cord blood which also involved two genes related to brain development, IGF2 and BDNF(Reference Caffrey, Irwin and McNulty128). The subsequently applied epigenome-wide screening on these samples identified differentially methylated regions(Reference Irwin, Thursby and Ondičová129), where a high proportion of the affected genes were associated with brain function as shown by examination through gene ontology analysis(Reference Ondičová, Irwin and Thursby130). Furthermore, a study using a Mendelian randomisation approach provided evidence that DNA methylation plays a mediating and causal role in associations between maternal vitamin B12 status and offspring cognition(Reference Caramaschi, Sharp and Nohr131). These findings offer a biological basis to link maternal B-vitamin status with neurodevelopment of the offspring, but more thorough investigation of these diet-epigenome-brain relationships is currently ongoing within the international EpiBrain project, involving partners in the UK, Canada and Spain, with the aim of providing substantiation and translation of the evidence for health improvement strategies(Reference Caffrey, Lamers and Murphy132).
Conclusions
Pregnancy, a period of intense development with increased demands for folate and related B-vitamins (vitamins B12, B6 and riboflavin), is associated with substantial temporal changes in B-vitamin status. Despite this, there are no pregnancy-specific reference ranges for relevant B-vitamin biomarkers. This greatly limits the ability to identify and appropriately manage pregnant women most at risk of B-vitamin deficiency. The development of reference ranges for plasma B-vitamin concentrations in pregnancy to support placental and fetal growth and maintain maternal health is an urgent priority. Another challenge is the limited evidence and lack of standardised approach for deriving dietary recommendations for B-vitamin intakes in pregnancy, with current recommendations mostly based on extrapolation from reference values for non-pregnant women. Therefore, well-designed controlled feeding trials for defining evidenced-based B-vitamin requirements during pregnancy are needed, that will help support successful pregnancy outcomes without compromising maternal health.
Apart from the well-established protective effect of periconceptional maternal folic acid supplementation against NTD, the driver of international policy in this area, these B-vitamins are associated with a plethora of health impacts. Exciting new evidence indicates that the effectiveness of NTD prevention could be further enhanced with the simultaneous provision of folic acid along with other B-vitamins and methyl donors involved in OCM. Furthermore, continued supplementation with folic acid beyond the first trimester of pregnancy (the period covered by current folic acid recommendations and at the same dose of 400 µg/d), may have intergenerational effects by improving cognitive performance and neuronal functioning in the offspring into the second decade of life. If these results are confirmed in other populations, they may provide the basis for revising recommendations for folic acid usage during pregnancy with the aim of optimising brain health in the child. Emerging evidence also suggests that optimal riboflavin status may help to maintain healthier blood pressure in pregnancy and reduce the prevalence of HDP and related adverse consequences, particularly in pregnant mothers who are genetically at a higher risk of these conditions owing to impaired folate metabolism. However, this area of research is still at an early stage and requires further investigations using carefully designed randomised trials in pregnant women.
Ethics of human subject participation
N/A.
Acknowledgements
None.
Financial support
No funding agency was involved in the writing of this review article.
Conflicts of interest
There are no conflicts of interest.
Authorship
K.P. drafted the manuscript; A.C., B.D., M.W., M.C., M.K. and H. McN. critically revised the manuscript for important intellectual content. All the authors have read and approved the final manuscript.
Open access
