Skip to main content
NCBI home page
The Journal of Nutrition logoLink to The Journal of Nutrition
. 2025 Jul 23;155(11):3781–3790. doi: 10.1016/j.tjnut.2025.07.013

Contrasting Effects of Phosphatidylcholine and Betaine Supplementation on Embryonic Development in a Mouse Model of the MTHFD1 R653Q Variant

Karen E Christensen 1,2, Marie-Lou Faquette 1,2, Aaron T Gebert 3, Vafa Keser 1,2, Teodoro Bottiglieri 3, Rima Rozen 1,2,⁎
PMCID: PMC12799447  PMID: 40714174

Abstract

Background

Folate and choline are interconnected nutrients important for embryonic development. Genetic variants in folate enzymes, such as MTHFD1 R653Q (c.1958 G>A, rs2236225), may increase dietary choline requirements.

Objectives

To evaluate the impact of phosphatidylcholine or betaine supplementation on embryonic development and maternal choline metabolism in the Mthfd1S+/- mouse model for the MTHFD1 653QQ genotype.

Methods

Female Mthfd1S+/+ (wild type [WT]) and Mthfd1S+/- (heterozygous [HET]) mice were fed amino acid-defined control (Ctrl), phosphatidylcholine-supplemented (PtdCho), or betaine-supplemented (BTN) diets for 4–6 wk before mating and during pregnancy (18–33 mice/group). Ctrl contained 2.5 g choline bitartrate/kg diet (based on AIN-93G). Supplemented diets contained additional phosphatidylcholine or betaine to obtain the molar equivalent of 3 times the free choline of Ctrl. Embryos were examined for anatomical developmental delays and morphological defects at embryonic day 10.5. One-carbon metabolites (homocysteine, choline, betaine, methionine, cystathionine, S-adenosylmethionine, S-adenosylhomocysteine) were measured in maternal plasma and liver by liquid chromatography-mass spectrometry (LC-MS/MS). The MTHFR protein in maternal liver was assessed by Western blot. Data were analyzed by binary logistic regression or 2-way analysis of variance.

Results

PtdCho increased developmental delays, compared with Ctrl, in litters of WT (170%) and HET (125%) mothers (Pdiet < 0.05). BTN decreased delays 53% compared with Ctrl in HET mother’s litters (Pdiet (HET) < 0.05). Defects decreased 71% in BTN-fed HET mother’s litters compared to WT (P < 0.05). BTN increased total pregnancy losses 360% (Pdiet < 0.01); PtdCho had no effect. Both diets decreased maternal plasma homocysteine (30–50%, Pdiet < 0.05), increased liver BTN (220–290%, Pdiet < 0.0001), and decreased S-adenosylmethionine/S-adenosylhomocysteine (WT: 20–30%, HET: 48–61%, Pdiet < 0.05). MTHFR protein expression decreased 25% in PtdCho maternal liver (Pdiet < 0.01).

Conclusions

Phosphatidylcholine and betaine supplementation had contrasting effects on reproductive outcomes. PtdCho increased developmental delays. BTN reduced delays and defects in the Mthfd1S+/- mouse model, but may impair the establishment of pregnancy. The timing, form, and dosage of choline supplementation may be important and require further investigation.

Keywords: MTHFD1, choline, betaine, phosphatidylcholine, developmental delay, developmental defects, embryo, neural tube defects

Introduction

Folate and choline are interdependent nutrients that are essential during embryonic development due to their roles in methylation and in the synthesis of nucleotides and phospholipids [1,2]. These nutrients are both involved in homocysteine remethylation in the liver (Supplemental Figure 1); consequently, genetic and nutritional disturbances in either folate or choline metabolism can affect the other [3]. Folate is well-known for its role in the prevention of neural tube defects (NTD), whereas the importance of choline in the prevention of NTD and adverse pregnancy outcomes is becoming more recognized [4,5]. Recent studies suggest that choline supplementation may help prevent NTD even when folate intakes meet recommended levels [4,6].

Common genetic variants in folate metabolism, such as the 5,10-methylenetetrahydrofolate (methyleneTHF) dehydrogenase–methenyltetrahydrofolate cyclohydrolase–10-formyltetrahydrofolate (formylTHF) synthetase (MTHFD1) R653Q (c.1958 G > A, rs2236225) substitution in the trifunctional enzyme MTHFD1, may alter the dietary requirements for choline [2,7]. The prevalence of the 653QQ genotype varies geographically: European populations are ∼20% 653QQ, whereas African, Asian, and South Asian populations are ∼4%, 7%, and 30% 653QQ, respectively [8]. The MTHFD1 R653Q variant has been linked to increased risk of NTD, heart defects, and other pregnancy complications [[9], [10], [11], [12], [13]]. This variant disrupts the synthetase activity of MTHFD1, which generates formylTHF for purine synthesis; formylTHF can be converted to methyleneTHF through the action of the other 2 MTHFD1 activities. MethyleneTHF is required for thymidylate synthesis or to support methylation reactions via reduction to 5-methyltetrahydrofolate (methylTHF) by 5,10-methylenetetrahydrofolate reductase (MTHFR). Several human studies have shown that MTHFD1 R653Q alters choline metabolism via the methylation cycle, and may increase dietary choline requirements [[14], [15], [16], [17]].

Choline is found in several different forms in food, most notably as choline salts, the phospholipid phosphatidylcholine, or the choline derivative betaine. The form of choline used as a supplement may be important, as the uptake and metabolism of the water and lipid-soluble forms differ [2,18]. Water-soluble choline salts, such as choline chloride or bitartrate, can be used as supplements but may not be ideal because they can be converted to trimethylamine N-oxide (TMAO), which may have toxic effects [2,19]. Phosphatidylcholine is lipid soluble and is broken down to produce choline, as well as contributing to phospholipid metabolism (Supplemental Figure 1). Phosphatidylcholine supplements may be preferable to choline salts, as they do not increase TMAO [19,20]. Betaine is the water-soluble choline derivative used as the carbon donor for folate-independent homocysteine remethylation (Supplemental Figure 1). Betaine supplements can maintain choline pools by replacing the choline that would otherwise be used for conversion to betaine for homocysteine remethylation, and are less likely to be converted to TMAO [21,22].

The MTHFD1-synthetase-deficient (Mthfd1S) mouse replicates the metabolic effects of the 653QQ genotype, and is prone to developmental defects such as malformation of the neural tube and heart [23,24]. The incidence of developmental delays and/or defects is sensitive to low choline, and to both low and high folate diets in this model [[25], [26], [27]]. In this study, our objective was to evaluate the effects of supplementation with phosphatidylcholine or betaine on embryonic development and maternal choline metabolism in our mouse model for the MTHFD1 653QQ genotype. We hypothesized that these supplements would improve reproductive outcomes. We also hypothesized that the effects of phosphatidylcholine would differ from those of betaine because of their differing metabolic functions.

Methods

Mice

All experiments were performed in compliance with Canadian Council on Animal Care guidelines and approved by the Animal Care Committee of the Research Institute of the McGill University Health Centre (protocol 5585). Mthfd1S BALB/cAnN mice [25] were bred in-house and genotyped as in [23]. Mice were group-housed in randomly distributed cages in the same room of a specific pathogen-free facility (18–24°C, 12-h light-dark cycle). Mice were fed standard unpurified diet (Teklad 2918; Inotiv) unless otherwise noted, and provided food and water ad libitum. Nulliparous Mthfd1S+/+ [wild type (WT)] and Mthfd1S+/- [heterozygous (HET)] females were randomly assigned to experimental diets at 4 wk-old and maintained on that diet throughout the experiments. In experiment 1, the control diet (Ctrl) and the phosphatidylcholine-supplemented diet (PtdCho) were compared. In experiment 2, Ctrl and the betaine-supplemented diet (BTN) were compared. Mice were weighed when assigned to diets, after 4 wk, at mating, and at embryo collection. The food was weighed weekly to monitor diet consumption.

Diets

The experimental diets (Inotiv) (Supplemental Table 1) were a modification of previously described amino acid-defined diets [25], omitting succinylsulfathiazole. The diets have vitamin, mineral, and nutrient contents as recommended for AIN-93G [28], with the exception of folic acid. The folic acid content was lowered to 1 mg folic acid/kg diet, twice the minimum amount recommended in the National Research Council (NRC) guidelines for mice [29], to avoid masking the effects of the phosphatidylcholine and betaine supplements [30]. Ctrl contains 2.5 g choline bitartrate/kg as recommended for AIN-93G. PtdCho and BTN are the same as Ctrl, with the exception of the added supplement; these diets contain the molar equivalent of 3 times the free choline in Ctrl. PtdCho was supplemented with phosphatidylcholine from soy lecithin by adding 17.266 g Phospholipon 90 G/kg diet (a generous gift from American Lecithin Company). The soybean oil content of PtdCho was adjusted to compensate for the lipid content of the phosphatidylcholine. BTN was supplemented with 2.581 g anhydrous betaine.

Females were mated with Mthfd1S+/- males after 4–6 wk on diets (mean: experiment 1: 4.9 wk, experiment 2: 4.7 wk). The morning that a vaginal plug was discovered was considered embryonic day (E)0.5. The females were killed by carbon dioxide asphyxiation under isoflurane anesthesia at E10.5. Blood was collected in LiHep tubes via cardiac puncture; plasma and tissues were collected and frozen as described [27]. Embryos were collected and developmental stage scored using established morphological markers as in [27]. Embryos were considered delayed if they were ≥1 d behind their most developed littermate. Embryo genotype and sex were determined for embryos with recoverable tissue using DNA extracted from the yolk sac, or if necessary, the entire embryo [27,31]. Scoring of embryos for developmental delays and defects was performed by an individual blinded to diet and genotype.

Metabolite measurement

Plasma total homocysteine (tHcy) was measured by LC-MS/MS as in [32]. Liver methylation metabolites (methionine, cystathionine, choline, betaine, S-adenosylmethionine [SAM], S-adenosylhomocysteine [SAH]) were measured by LC-MS/MS as in [33].

Western blots

Protein extracts were prepared as in [25] using RIPA extraction buffer containing protease (Pierce) and phosphatase (ThermoFisher) inhibitors. Immunoblotting was performed as in [34] using primary antibodies specific for MTHFD1 [35], MTHFR [36], methionine synthase (MTR) (25896-1-AP; Proteintech), betaine-homocysteine methyltransferase (BHMT) [37], phosphatidylethanolamine N-methyltransferase (PEMT) (PA5-42383; Invitrogen), glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (2118; Cell Signaling Technology), VINCULIN (13901; Cell Signaling Technology), and ACTIN (A2066; Sigma-Aldrich). Normalization by ACTIN, VINCULIN, or GAPDH was validated by comparing to normalization by amido black protein staining on the membrane.

Statistics

Data from pregnant females were analyzed by 2-factor analysis of variance (ANOVA) followed by Tukey post hoc analysis. Data analyzed by ANOVA were checked for normality using the Shapiro-Wilk test. Embryonic genotype and sex distributions were compared using χ2 and Fisher’s exact analysis. Categorical data (e.g., delays) were analyzed by binary logistic regression with general linear mixed models (package lme4 [38]) followed by post hoc tests comparing each variable for each combination of the other factors in the analysis, adjusted for multiple testing using the multivariate t method (mvt, package emmeans [39]) using R 4.4.2 [40] in RStudio (version 2024.12.0 + 467, [41]). Diet, maternal and embryonic genotypes, and selected diet-genotype interactions, as determined by the lowest Akaike information criterion (AIC), were specified as fixed effects, and litter was included as a random effect. Embryonic sex distributions were not significantly altered by diet, maternal genotype, or embryonic genotype, and including sex did not improve regression models, so it was not included in the analysis. Null (Mthfd1S-/-), necrotic, and damaged embryos were excluded from analysis of delays and defects. Individual females and embryos were used as the unit of analysis for calculations. Sample sizes were determined using previous experiments with these mice to estimate expected effect sizes and variances [[25], [26], [27]]. Analyses were performed using GraphPad Prism 8.0.1, unless otherwise noted. For all analyses, P ≤ 0.05 was considered significant; P ≤ 0.07 was considered a trend. Significant P values are indicated at the top of the figure panels. Values are presented as mean ± SEM.

Results

Experiment 1: Effects of phosphatidylcholine supplementation during pregnancy

Effect of PtdCho on maternal body and organ weights, and on fertility

There were no significant differences in diet consumption between the diet/genotype groups before mating or during gestation (Supplemental Figure 2A and B). There were no significant differences in weight gain between groups prior to mating or during gestation (Supplemental Figure 3A and B). There were also no significant effects of diet or genotype on maternal body weight or organ weights (Supplemental Figure 3C–F). Total pregnancy losses were observed in a small number of females with vaginal plugs, but were not associated with genotype or diet (Supplemental Figure 4A). There was no necrosis in these cases. The deciduae were small, pale, and were not attached to the uterus. In the normal pregnancies, there was a small but significant increase in the number of implantation sites in the PtdCho mice (P = 0.0301), whereas the number of eggs released, as measured by counting corpus lutea, did not differ between groups (Supplemental Figure 4B and C). As expected, the number of viable embryos per litter decreased and the resorption rate increased in the HET mice due to the lethality of the Mthfd1S-/- genotype [25] (Supplemental Figure 4D). The resorption rate was not affected by diet. Similarly, there was no significant deviation of the genotype distributions of embryos from WT or HET mothers from expected ratios due to diet (Supplemental Table 2).

PtdCho increased developmental delays but had no effect on the incidence of developmental defects

The effects of maternal genotype and diet were analyzed both alone and including embryonic genotype (Figure 1, Supplemental Table 3). In the analysis of maternal factors, the incidence of delay was significantly higher in the PtdCho litters (Figure 1A). When embryonic genotype was included in the analysis, both diet and embryonic genotype were associated with significant increases in delays (Figure 1B).

FIGURE 1.

FIGURE 1

Open in a new tab

Delay and defect incidence in Ctrl and PtdCho embryos. (A) Effects of PtdCho and maternal genotype on risk of developmental delay, with embryonic genotypes grouped. (B) Effects of PtdCho, maternal and embryonic genotype on risk of developmental delay. (C) Effects of PtdCho and maternal genotype on defect incidence, with the embryonic genotypes grouped. (D) Effects of PtdCho, maternal and embryonic genotype on defect incidence. Values are percentage of affected embryos/total. n = 16–19 litters/group, 96–156 embryos/diet-maternal genotype group, 25–80 embryos/diet-maternal genotype-embryonic genotype group. Analyzed by binary logistic regression. Post hoc by multivariate t: ∗P < 0.05. Ctrl, control diet; e, embryo; HET, heterozygous; m, maternal; PtdCho, phosphatidylcholine diet; WT: wild type.

Embryos were evaluated for gross morphological defects (Supplemental Figure 5). The defects included open neural tubes, improper or misaligned neural tube closure, turning defects (failure to turn, reversed turning), reversed heart looping, lack of development of the telencephalic vesicle, craniofacial malformation, and embryos that were not nulls but had the typical Mthfd1S-/- phenotype (not turned, open head folds, and abnormal heart and tail bud development, as described in [23]). The same types of defects were observed in all groups. 78% of embryos with defects were also delayed. In contrast to delay, there were no significant effects of diet or maternal/embryonic genotype on the incidence of defects (Figure 1C and D).

PtdCho lowered plasma homocysteine and altered maternal hepatic methylation metabolites

Liver is an important organ in folate metabolism, and the site of overlap between folate and choline metabolism due to BHMT and PEMT expression (Supplemental Figure 1). Plasma tHcy is used clinically as an indicator of folate deficiency or disrupted folate metabolism. We therefore measured plasma tHcy as well as critical methylation metabolites in the liver (Figure 2). Plasma tHcy concentration was halved by PtdCho (Figure 2A). Hepatic choline increased ∼44% in HET mice, but was not affected by PtdCho (Figure 2B). In contrast, betaine increased 290% due to PtdCho and was unchanged by genotype (Figure 2C). Methionine concentration increased ∼52% in HET mice, but was not affected by PtdCho (Figure 2D), similar to the results for choline. There was no indication of increased homocysteine detoxification via transsulfuration, as cystathionine concentrations were not associated with diet or genotype, although there was a significant diet-genotype interaction (Figure 2E). SAM decreased with PtdCho and genotype, and SAH increased in PtdCho mice, resulting in significantly lowered methylation potential due to both genotype and PtdCho (Figure 2F–H).

FIGURE 2.

FIGURE 2

Open in a new tab

Plasma tHcy and liver methylation metabolites in WT or HET dams fed PtdCho or Ctrl. (A) Plasma tHcy; (B) Liver choline; (C) Liver betaine; (D) Liver methionine; (E) Liver cystathionine; (F) Liver SAM; (G) Liver SAH; (H) Liver methylation potential (SAM/SAH). Values are mean ± SEM, n = 5–6/group, analyzed by 2-way ANOVA. Tukey post hoc: ∗P < 0.05, ∗∗P < 0.01. ANOVA, analysis of variance; Ctrl, control diet; HET, heterozygous; PtdCho, phosphatidylcholine diet; tHcy, total homocysteine; SAH, S-adenosylhomocysteine; SAM, S-adenosylmethionine; SEM, standard error of the mean; WT, wild type.

MTHFR protein expression in maternal liver was lowered by PtdCho

The expression of the key folate enzyme MTHFR is known to be sensitive to disruptions in folate metabolism [25,26,34]. Total MTHFR expression decreased ∼25% due to PtdCho in the liver of both WT and HET mice (Figure 3A). Three MTHFR isoforms are detected by immunoblotting: 70 kDa, phosphorylated 70 kDa, and 77 kDa [42,43]. The unphosphorylated 70 kDa isoform may be more active and/or less sensitive to SAM regulation than the phosphorylated form [42,44,45]. The percentage of total MTHFR in the unphosphorylated 70 kDa form decreased almost 40% in the PtdCho liver, which may attenuate MTHFR activity (Figure 3B). Neither MTHFR protein nor isoform expression was affected by genotype. The protein expression of MTHFD1, MTR, BHMT, and PEMT was also evaluated; their expression was not affected by PtdCho or genotype (Supplemental Figure 6).

FIGURE 3.

FIGURE 3

Open in a new tab

MTHFR protein expression in the liver of Ctrl- and PtdCho-fed WT or HET dams. (A) MTHFR expression; (B) Percentage of MTHFR in the unphosphorylated isoform; (C) Representative blot. Values are mean ± SEM, n = 6/group, analyzed by 2-way ANOVA. Tukey post hoc: ∗P < 0.05, ∗∗P < 0.01, ∗∗∗∗P < 0.0001. ANOVA, analysis of variance; Ctrl, control diet; HET, heterozygous; MTHFR, 5,10-methylenetetrahydrofolate reductase; PtdCho, phosphatidylcholine diet; SEM, standard error of the mean; WT, wild type.

Experiment 2: effects of betaine supplementation during pregnancy

Effect of BTN on maternal body and organ weights, and on fertility

There were no significant differences in diet consumption between the diet/genotype groups before mating or during gestation (Supplemental Figure 2C and D). There were no significant differences in weight gain between groups prior to mating or during gestation (Supplemental Figure 7A and B). There were also no significant effects of diet or genotype on maternal body weight or organ weights (Supplemental Figure 7C–F). The same type of total pregnancy losses observed in experiment 1 were observed in this experiment, but unlike PtdCho, there was a significant increase associated with BTN (Supplemental Figure 8A). The number of implantation sites and the number of corpus lutea did not differ between groups (Supplemental Figure 8B, C). As expected, the number of viable embryos per litter decreased and the resorption rate increased in HET mice due to the lethality of the Mthfd1S-/- genotype (Supplemental Figure 8D). Genotype distributions of embryos from WT or HET mothers did not vary significantly from expected ratios (Supplemental Table 4).

BTN reduced the incidence of delays in HET mothers and tended to decrease the incidence of developmental defects

There was a significant interaction between diet and maternal genotype in the analysis of maternal factors and incidence of developmental delay, such that delays were significantly lower in the litters of BTN HET mothers as compared to Ctrl HET (Figure 4A, Supplemental Table 5). There was no additional effect when the embryonic genotype was included in the analysis (Figure 4B).

FIGURE 4.

FIGURE 4

Open in a new tab

Delay and defect incidence in Ctrl and BTN embryos. (A) Effects of BTN and maternal genotype on risk of developmental delay, with embryonic genotypes grouped. (B) Effects of BTN, maternal and embryonic genotype on risk of delay. (C) Effects on BTN and maternal genotype on incidence of defects, with embryonic genotypes grouped. (D) Effects of BTN, maternal and embryonic genotype on defect incidence. Values are the percentage of affected embryos/total. n = 17–22 litters/group, 105–171 embryos/diet-maternal genotype group, 22–87 embryos/diet-maternal genotype-embryonic genotype group. Analyzed by binary logistic regression. Post hoc by multivariate t: ∗P < 0.05. BTN, betaine diet; Ctrl, control diet; e, embryo; HET, heterozygous; m, maternal; WT, wild type.

The defects observed in experiment 2 were the same types as those observed in experiment 1 (Supplemental Figure 5). 72% of embryos with defects were also delayed. In contrast to experiment 1, BTN tended to decrease the incidence of defects in the litters of HET mothers (Figure 4C). Embryonic genotype had no effect on defect incidence (Figure 4D). If Ctrl and BTN are analyzed separately, there is a significant effect of maternal genotype on defect incidence in the BTN group (P = 0.022), but not the Ctrl group (P = 0.31).

BTN lowered plasma homocysteine and altered maternal hepatic methylation metabolites

We measured plasma tHcy and critical methylation metabolites in the liver (Figure 5). Similar to the observations with PtdCho, plasma tHcy was lowered ∼30% by BTN in both WT and HET mice (Figure 5A). Hepatic choline tended to decrease in BTN mice, but was not affected by genotype (Figure 5B). In contrast, betaine increased 220% due to BTN and was unchanged by genotype (Figure 5C), similar to the results with PtdCho (Figure 2C). Methionine concentration was not affected by BTN or genotype (Figure 5D). There was no indication of increased homocysteine detoxification via transsulfuration, as cystathionine concentrations were not associated with diet or genotype (Figure 5E). SAM concentration decreased in HET mice, particularly in the HET mice fed BTN, and SAH increased in BTN mice, resulting in significantly lowered methylation potential due to both HET genotype and BTN (Figure 5F–H).

FIGURE 5.

FIGURE 5

Open in a new tab

Plasma tHcy and liver methylation metabolites in WT or HET dams fed BTN or Ctrl. (A) Plasma tHcy; (B) Liver choline; (C) Liver betaine; (D) Liver methionine; (E) Liver cystathionine; (F) Liver SAM; (G) Liver SAH; (H) Liver methylation potential (SAM/SAH). Values are mean ± SEM, n = 5–6/group, analyzed by 2-way ANOVA. Tukey post hoc: ∗P < 0.05, ∗∗P < 0.01, ∗∗∗∗P < 0.0001. ANOVA, analysis of variance; BTN, betaine diet; Ctrl, control diet; HET, heterozygous; tHcy, total homocysteine; SAH, S-adenosylhomocysteine; SAM, S-adenosylmethionine; SEM, standard error of the mean; WT, wild type.

MTHFR protein expression in the liver was altered by BTN

There was a very small but statistically significant increase in total MTHFR expression due to the HET genotype in the maternal liver (Figure 6A). The percentage of total MTHFR in the unphosphorylated 70 kDa form decreased by almost 25% in BTN liver (Figure 6B). Neither MTHFR protein nor isoform expression was affected by genotype. These changes in MTHFR expression were more subtle than the changes observed with PtdCho in experiment 1, which may have contributed to the differences in outcomes. MTR, BHMT, and PEMT expression was not affected by BTN or genotype, whereas MTHFD1 expression increased by a very small but statistically significant amount in HET mice (Supplemental Figure 9).

FIGURE 6.

FIGURE 6

Open in a new tab

MTHFR protein expression in the liver of Ctrl- and BTN-fed WT or HET dams. (A) MTHFR expression; (B) Percentage of MTHFR in the unphosphorylated isoform; (C) Representative blot. Values are mean ± SEM, n = 6/group, analyzed by 2-way ANOVA. Tukey post hoc: ∗P < 0.05. ANOVA, analysis of variance; BTN, betaine diet; Ctrl, control diet; HET, heterozygous; MTHFR, 5,10-methylenetetrahydrofolate reductase; SEM, standard error of the mean; WT, wild type.

Discussion

The maternal liver is an important organ during pregnancy, producing folate- and choline-derived metabolites that are transferred to embryos to support their development [46]. As the major site of expression of MTHFD1, BHMT, and PEMT, the liver is also the site of the interaction of folate and choline metabolism [3]. Folate-dependent MTR and choline-dependent BHMT both participate in the remethylation of homocysteine in the liver, and phosphatidylcholine production by PEMT is a major consumer of the SAM produced by the methylation cycle [46]. Consequently, genetic variants in folate metabolism, such as MTHFD1 R653Q, may increase dietary choline requirements [[14], [15], [16], [17],47]. Phosphatidylcholine and betaine supplements may increase choline availability to embryos by conversion to choline or by maintaining choline that might otherwise be used in betaine synthesis for methylation, respectively.

In this study, we performed separate experiments to evaluate the effects of phosphatidylcholine and betaine supplements on the incidence of developmental delays and defects in the Mthfd1S mouse model. These supplemented and control diets (PtdCho, BTN, and Ctrl) contained 1 mg folic acid/kg diet, which is half the amount in AIN-93G, but twice the NRC-recommended minimum. This lower concentration of folic acid would avoid obscuring the potential impact of the supplements, and would also model human intakes in early pregnancy when the neural tube is closing [48].

Both PtdCho and BTN lowered plasma tHcy as compared to Ctrl, indicating that both supplements successfully contributed to homocysteine remethylation. Both supplements dramatically increased betaine in the liver. Hepatic choline was unaffected by PtdCho, which suggests that choline, which is derived from phosphatidylcholine, was converted to betaine in the liver and did not accumulate in this tissue. Similar effects were observed in a human trial: equimolar doses of phosphatidylcholine and betaine increased serum betaine equally, showing that phosphatidylcholine was broken down to choline and converted to betaine [20]. In that study, phosphatidylcholine increased serum choline, suggesting that excess hepatic choline is exported from the liver [20]. The increase in hepatic betaine with both supplements likely contributed to the decrease in tHcy. Likewise, both supplements reduced SAM/SAH in HET mice. As the effects of the diets on liver metabolites are so similar, they do not explain the differences in reproductive outcomes between the PtdCho and BTN diets. It is possible that the effects of PtdCho and BTN are tissue-specific; supplemental phosphatidylcholine and betaine may have metabolic effects outside of the liver that influence embryonic development.

MTHFR protein expression is known to change in response to dietary folate [25,26,34]. As is the case with folate, MTHFR protein expression appears to be sensitive to the amount and form of dietary intakes of choline metabolites. Total MTHFR expression decreased in PtdCho-fed mice, but not BTN-fed mice, although the percentage of the unphosphorylated 70 kDa isoform decreased in both diets. These changes in MTHFR expression do not seem to be affecting the methylation cycle, as tHcy is not increased, and the difference in MTHFR expression between PtdCho and BTN is not reflected in the SAM/SAH ratio. MTHFR is an irreversible switch point between the methylation cycle and nucleotide production. The decreased MTHFR in the PtdCho mothers would impair methyleneTHF reduction to methylTHF, altering the normal flow of carbon through folate-dependent pathways. It is possible that these disruptions could cause imbalances in methyleneTHF-related pathways, including nucleotide synthesis, serine/glycine interconversion, or other crucial metabolites of the folate pathways. Although the increase in developmental delays due to PtdCho was unexpected, it echoes the report that the development of bovine embryos in vitro was impaired by supplemental phosphatidylcholine [49]. In contrast, PtdCho had no effect on the incidence of defects or total pregnancy loss in our study.

The effects of BTN on pregnancy outcomes were not the same as those of PtdCho. BTN appears to decrease developmental delays and defects, specifically in the litters of HET mothers. BTN was associated with increased total pregnancy losses in both WT and HET mice, which suggests that BTN may, under certain circumstances, negatively affect the uterine environment [50]. Phosphatidylchoine and betaine have different metabolic roles outside of homocysteine remethylation. Betaine is an important osmolyte, and so may influence pregnancy and embryonic development through other nonmetabolic mechanisms [51,52]. It is possible that excess betaine could contribute to a uterine environment that is inhospitable to the events following conception. Phosphatidylcholine, on the contrary, is an important component of plasma membranes and phospholipid metabolism, which are also essential for embryonic development.

Choline and phosphatidylcholine supplements during gestation have been demonstrated to benefit brain development and immune function at later time points [20,46,53]. This suggests that both the timing and the amount of supplementation may be critical. Supplemental phosphatidylcholine may be beneficial later in gestation, but detrimental at early time points. Similarly, betaine supplements could be beneficial during pregnancy, particularly for people with the MTHFD1 653QQ genotype, but may not be beneficial in the early postconception phase. It is also possible that supplementation with a mix of sources would have a greater benefit than any 1 choline metabolite form.

Conclusion

Current choline intakes worldwide are known to be lower than recommended. This finding has generated interest in supplementation to prevent pregnancy complications and birth defects [4,5]. Our results suggest that betaine supplementation could be beneficial, particularly for some individuals, but further research is required to establish the appropriate timing of supplementation, as well as the form and dosage.

Author contributions

The authors’ responsibilities were as follows – RR, KEC: designed research; KEC, M-LF, VK, ATG, TB: conducted research; KEC: analyzed data; KEC, RR: wrote the paper; RR: had primary responsibility for final content; and all authors: read and approved the final manuscript.

Data availability

Data presented and summarized in the manuscript will be made available upon request, pending approval.

Funding

This study was funded by the Canadian Institutes of Health Research (CIHR), grant number PJT-156389 (KEC and RR). CIHR has no involvement or restrictions regarding this publication.

Conflict of interest

The authors report no conflicts of interest.

Acknowledgments

We thank Loydie Jerome-Majewska for her expert opinion.

Footnotes

Appendix A

Supplementary data to this article can be found online at https://doi.org/10.1016/j.tjnut.2025.07.013.

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Multimedia component 1
mmc1.pdf (2MB, pdf)

References

  • 1.Bailey L.B., Stover P.J., McNulty H., Fenech M.F., Gregory J.F., Mills J.L., et al. Biomarkers of nutrition for development-folate review. J. Nutr. 2015;145(7):1636S–1680S. doi: 10.3945/jn.114.206599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Wallace T.C., Blusztajn J.K., Caudill M.A., Klatt K.C., Natker E., Zeisel S.H., et al. Choline: the underconsumed and underappreciated essential nutrient. Nutr. Today. 2018;53(6):240–253. doi: 10.1097/NT.0000000000000302. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Chen V., Schwartz J.L., Cho C.E. Folate and choline: does it take two to tango in early programming of disease? Lifestyle Genom. 2023;16(1):177–191. doi: 10.1159/000534452. [DOI] [PubMed] [Google Scholar]
  • 4.Obeid R., Derbyshire E., Schön C. Association between maternal choline, fetal brain development, and child neurocognition: systematic review and meta-analysis of human studies. Adv. Nutr. 2022;13(6):2445–2457. doi: 10.1093/advances/nmac082. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Nguyen H.T., Oktayani P.P., Lee S.D., Huang L.C. Choline in pregnant women: a systematic review and meta-analysis. Nutr. Rev. 2025;83(2):e273–e289. doi: 10.1093/nutrit/nuae026. [DOI] [PubMed] [Google Scholar]
  • 6.Petersen J.M., Smith-Webb R.S., Shaw G.M., Carmichael S.L., Desrosiers T.A., Nestoridi E., et al. Periconceptional intakes of methyl donors and other micronutrients involved in one-carbon metabolism may further reduce the risk of neural tube defects in offspring: a United States population-based case-control study of women meeting the folic acid recommendations. Am. J. Clin. Nutr. 2023;118(3):720–728. doi: 10.1016/j.ajcnut.2023.05.034. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Ganz A.B., Klatt K.C., Caudill M.A. Common genetic variants alter metabolism and influence dietary choline requirements. Nutrients. 2017;9(8):837. doi: 10.3390/nu9080837. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.dbSNP Reference SNP (rs) Report, rs2236225. https://www.ncbi.nlm.nih.gov/snp/rs2236225#frequency_tab [cited June 2025]. Available from:
  • 9.Christensen K.E., Rohlicek C.V., Andelfinger G.U., Michaud J., Bigras J.L., Richter A., et al. The MTHFD1 p.Arg653Gln variant alters enzyme function and increases risk for congenital heart defects. Hum. Mutat. 2009;30(2):212–220. doi: 10.1002/humu.20830. [DOI] [PubMed] [Google Scholar]
  • 10.Parle-McDermott A., Pangilinan F., Mills J.L., Signore C.C., Molloy A.M., Cotter A., et al. A polymorphism in the MTHFD1 gene increases a mother’s risk of having an unexplained second trimester pregnancy loss. Mol. Hum. Reprod. 2005;11(7):477–480. doi: 10.1093/molehr/gah204. [DOI] [PubMed] [Google Scholar]
  • 11.Furness D.L., Fenech M.F., Khong Y.T., Romero R., Dekker G.A. One-carbon metabolism enzyme polymorphisms and uteroplacental insufficiency. Am. J. Obstet. Gynecol. 2008;199(3):276.e1. doi: 10.1016/j.ajog.2008.06.020. 278.e1. [DOI] [PubMed] [Google Scholar]
  • 12.Zheng J., Lu X., Liu H., Zhao P., Li K., Li L. MTHFD1 polymorphism as maternal risk for neural tube defects: a meta-analysis. Neurol. Sci. 2015;36(4):607–616. doi: 10.1007/s10072-014-2035-7. [DOI] [PubMed] [Google Scholar]
  • 13.Yi K., He S.E., Guo T., Wang Z.Q., Zhang X., Xu J.G., et al. Association of MTHFD1 G1958A (rs2236225) gene polymorphism with the risk of congenital heart disease: a systematic review and meta-analysis. BMC Med. Genomics. 2025;18(1):20. doi: 10.1186/s12920-024-02052-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Kohlmeier M., da Costa K.A., Fischer L.M., Zeisel S.H. Genetic variation of folate-mediated one-carbon transfer pathway predicts susceptibility to choline deficiency in humans. Proc. Natl. Acad. Sci. U S A. 2005;102(44):16025–16030. doi: 10.1073/pnas.0504285102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Ivanov A., Nash-Barboza S., Hinkis S., Caudill M.A. Genetic variants in phosphatidylethanolamine N-methyltransferase and methylenetetrahydrofolate dehydrogenase influence biomarkers of choline metabolism when folate intake is restricted. J. Am. Diet Assoc. 2009;109(2):313–318. doi: 10.1016/j.jada.2008.10.046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Ganz A.B., Shields K., Fomin V.G., Lopez Y.S., Mohan S., Lovesky J., et al. Genetic impairments in folate enzymes increase dependence on dietary choline for phosphatidylcholine production at the expense of betaine synthesis. FASEB J. 2016;30(10):3321–3333. doi: 10.1096/fj.201500138RR. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Ilozumba M.N., Cheng T.D., Neuhouser M.L., Miller J.W., Beresford S.A., Duggan D.J., et al. Associations between plasma choline metabolites and genetic polymorphisms in one-carbon metabolism in postmenopausal women: the Women’s Health Initiative observational study. J. Nutr. 2020;150(11):2874–2881. doi: 10.1093/jn/nxaa266. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Zeisel S.H., Mar M.H., Howe J.C., Holden J.M. Concentrations of choline-containing compounds and betaine in common foods. J. Nutr. 2003;133(5):1302–1307. doi: 10.1093/jn/133.5.1302. [DOI] [PubMed] [Google Scholar]
  • 19.Böckmann K.A., Franz A.R., Minarski M., Shunova A., Maiwald C.A., Schwarz J., et al. Differential metabolism of choline supplements in adult volunteers. Eur. J. Nutr. 2022;61(1):219–230. doi: 10.1007/s00394-021-02637-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Hoffman M.C., Olincy A., D’Alessandro A., Reisz J.A., Hansen K.C., Hunter S.K., et al. Effects of phosphatidylcholine and betaine supplements on women’s serum choline. J. Nutr. Intermed. Metab. 2019;16 doi: 10.1016/j.jnim.2019.100094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Dilger R.N., Garrow T.A., Baker D.H. Betaine can partially spare choline in chicks but only when added to diets containing a minimal level of choline. J. Nutr. 2007;137(10):2224–2228. doi: 10.1093/jn/137.10.2224. [DOI] [PubMed] [Google Scholar]
  • 22.Rath S., Heidrich B., Pieper D.H., Vital M. Uncovering the trimethylamine-producing bacteria of the human gut microbiota. Microbiome. 2017;5(1):54. doi: 10.1186/s40168-017-0271-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Christensen K.E., Deng L., Leung K.Y., Arning E., Bottiglieri T., Malysheva O.V., et al. A novel mouse model for genetic variation in 10-formyltetrahydrofolate synthetase exhibits disturbed purine synthesis with impacts on pregnancy and embryonic development. Hum. Mol. Genet. 2013;22(18):3705–3719. doi: 10.1093/hmg/ddt223. [DOI] [PubMed] [Google Scholar]
  • 24.Christensen K.E., Deng L., Bahous R.H., Jerome-Majewska L.A., Rozen R. MTHFD1 formyltetrahydrofolate synthetase deficiency, a model for the MTHFD1 R653Q variant, leads to congenital heart defects in mice, Birth Defects Res. A Clin. Mol. Teratol. 2015;103(12):1031–1038. doi: 10.1002/bdra.23451. [DOI] [PubMed] [Google Scholar]
  • 25.Christensen K.E., Hou W., Bahous R.H., Deng L., Malysheva O.V., Arning E., et al. Moderate folic acid supplementation and MTHFD1-synthetase deficiency in mice, a model for the R653Q variant, result in embryonic defects and abnormal placental development. Am. J. Clin. Nutr. 2016;104(5):1459–1469. doi: 10.3945/ajcn.116.139519. [DOI] [PubMed] [Google Scholar]
  • 26.Christensen K.E., Bahous R.H., Hou W., Deng L., Malysheva O.V., Arning E., et al. Low dietary folate interacts with MTHFD1 synthetase deficiency in mice, a model for the R653Q variant, to increase incidence of developmental delays and defects. J. Nutr. 2018;148(4):501–509. doi: 10.1093/jn/nxy013. [DOI] [PubMed] [Google Scholar]
  • 27.Christensen K.E., Malysheva O.V., Carlin S., Matias F., MacFarlane A.J., Jacobs R.L., et al. Mild choline deficiency and MTHFD1 synthetase deficiency interact to increase incidence of developmental delays and defects in mice. Nutrients. 2021;14(1):127. doi: 10.3390/nu14010127. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Reeves P.G. Components of the AIN-93 diets as improvements in the AIN-76A diet. J. Nutr. 1997;127(5):838S–841S. doi: 10.1093/jn/127.5.838S. [DOI] [PubMed] [Google Scholar]
  • 29.National Research Council (US) Subcommittee on Laboratory Animal Nutrition . Fourth Revised Edition. National Academies Press (US); Washington (DC): 1995. Nutrient Requirements of Laboratory Animals; pp. 80–102. [PubMed] [Google Scholar]
  • 30.Fernàndez-Roig S., Cavallé-Busquets P., Fernandez-Ballart J.D., Ballesteros M., Berrocal-Zaragoza M.I., Salat-Batlle J., et al. Low folate status enhances pregnancy changes in plasma betaine and dimethylglycine concentrations and the association between betaine and homocysteine. Am. J. Clin. Nutr. 2013;97(6):1252–1259. doi: 10.3945/ajcn.112.054189. [DOI] [PubMed] [Google Scholar]
  • 31.Tunster S.J. Genetic sex determination of mice by simplex PCR. Biol. Sex Differ. 2017;8(1):31. doi: 10.1186/s13293-017-0154-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Ducros V., Belva-Besnet H., Casetta B., Favier A. A robust liquid chromatography tandem mass spectrometry method for total plasma homocysteine determination in clinical practice. Clin. Chem. Lab. Med. 2006;44(8):987–990. doi: 10.1515/CCLM.2006.178. [DOI] [PubMed] [Google Scholar]
  • 33.Kalecký K., Ashcraft P., Bottiglieri T. One-carbon metabolism in Alzheimer’s disease and Parkinson’s disease brain tissue. Nutrients. 2022;14(3):599. doi: 10.3390/nu14030599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Christensen K.E., Faquette M.L., Leclerc D., Keser V., Luan Y., Bennett-Firmin J.L., et al. Folic acid and methyltetrahydrofolate supplementation in the Mthfr677C>T mouse model with hepatic steatosis. Nutrients. 2024;17(1):82. doi: 10.3390/nu17010082. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Gardam M.A., Mejia N.R., MacKenzie R.E. The NADP-dependent trifunctional methylenetetrahydrofolate dehydrogenase purified from mouse liver is immunologically distinct from the mouse NAD-dependent [corrected] bifunctional enzyme. Biochem. Cell Biol. 1988;66(1):66–70. doi: 10.1139/o88-008. [DOI] [PubMed] [Google Scholar]
  • 36.Frosst P., Blom H.J., Milos R., Goyette P., Sheppard C.A., Matthews R.G., et al. A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nat. Genet. 1995;10(1):111–113. doi: 10.1038/ng0595-111. [DOI] [PubMed] [Google Scholar]
  • 37.Rao P.V., Garrow T.A., John F., Garland D., Millian N.S., Zigler J.S., Jr. Betaine-homocysteine methyltransferase is a developmentally regulated enzyme crystallin in rhesus monkey lens. J. Biol. Chem. 1998;273(46):30669–30674. doi: 10.1074/jbc.273.46.30669. [DOI] [PubMed] [Google Scholar]
  • 38.Bates D., Mächler M., Bolker B., Walker S. Fitting linear mixed-effects models using lme4. J. Stat. Soft. 2015;67(1):48. doi: 10.18637/jss.v067.i01. [DOI] [Google Scholar]
  • 39.Lenth R.V. emmeans: Estimated Marginal Means, aka Least-Squares Means; R package version 1.11.1. 2025. [Google Scholar]
  • 40.R Core Team . R Foundation for Statistical Computing; 2024. R: A language and environment for statistical computing. [Google Scholar]
  • 41.RStudio Team . RStudio, PBC; Boston, MA: 2020. RStudio: Integrated development for R. [Google Scholar]
  • 42.Yamada K., Strahler J.R., Andrews P.C., Matthews R.G. Regulation of human methylenetetrahydrofolate reductase by phosphorylation. Proc. Natl. Acad. Sci. U S A. 2005;102(30):10454–10459. doi: 10.1073/pnas.0504786102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Tran P., Leclerc D., Chan M., Pai A., Hiou-Tim F., Wu Q., et al. Multiple transcription start sites and alternative splicing in the methylenetetrahydrofolate reductase gene result in two enzyme isoforms. Mamm. Genome. 2002;13(9):483–492. doi: 10.1007/s00335-002-2167-6. [DOI] [PubMed] [Google Scholar]
  • 44.Zheng Y., Ramsamooj S., Li Q., Johnson J.L., Yaron T.M., Sharra K., et al. Regulation of folate and methionine metabolism by multisite phosphorylation of human methylenetetrahydrofolate reductase. Sci. Rep. 2019;9(1):4190. doi: 10.1038/s41598-019-40950-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Yamada K., Mendoza J., Koutmos M. Structural basis of S-adenosylmethionine-dependent allosteric transition from active to inactive states in methylenetetrahydrofolate reductase. Nat. Commun. 2024;15(1):5167. doi: 10.1038/s41467-024-49327-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Korsmo H.W., Jiang X., Caudill M.A. Choline: exploring the growing science on its benefits for moms and babies. Nutrients. 2019;11(8):1823. doi: 10.3390/nu11081823. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Yan J., Winter L.B., Burns-Whitmore B., Vermeylen F., Caudill M.A. Plasma choline metabolites associate with metabolic stress among young overweight men in a genotype-specific manner. Nutr. Diabetes. 2012;2(10):e49. doi: 10.1038/nutd.2012.23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Tinker S.C., Hamner H.C., Qi Y.P., Crider K.S. U.S. women of childbearing age who are at possible increased risk of a neural tube defect-affected pregnancy due to suboptimal red blood cell folate concentrations, National Health and Nutrition Examination survey 2007 to 2012, Birth Defects Res. A Clin. Mol. Teratol. 2015;103(6):517–526. doi: 10.1002/bdra.23378. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Estrada-Cortés E., Negrón-Peréz V.M., Tríbulo P., Zenobi M.G., Staples C.R., Hansen P.J. Effects of choline on the phenotype of the cultured bovine preimplantation embryo. J. Dairy Sci. 2020;103(11):10784–10796. doi: 10.3168/jds.2020-18598. [DOI] [PubMed] [Google Scholar]
  • 50.Roberts E.R., Bhurke A.V., Ganeshkumar S., Gunewardena S., Arora R., Chennathukuzhi V.M. Loss of PRICKLE1 leads to abnormal endometrial epithelial architecture, decreased embryo implantation, and reduced fertility in mice. PNAS Nexus. 2025;4(2) doi: 10.1093/pnasnexus/pgaf024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Lever M., Slow S. The clinical significance of betaine, an osmolyte with a key role in methyl group metabolism. Clin. Biochem. 2010;43(9):732–744. doi: 10.1016/j.clinbiochem.2010.03.009. [DOI] [PubMed] [Google Scholar]
  • 52.Zhao G., He F., Wu C., Li P., Li N., Deng J., et al. Betaine in inflammation: mechanistic aspects and applications. Front. Immunol. 2018;9:1070. doi: 10.3389/fimmu.2018.01070. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Lewis E.D., Richard C., Goruk S., Dellschaft N.S., Curtis J.M., Jacobs R.L., et al. The form of choline in the maternal diet affects immune development in suckled rat offspring. J. Nutr. 2016;146(4):823–830. doi: 10.3945/jn.115.225888. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Multimedia component 1
mmc1.pdf (2MB, pdf)

Data Availability Statement

Data presented and summarized in the manuscript will be made available upon request, pending approval.

Articles from The Journal of Nutrition are provided here courtesy of American Society for Nutrition

RESOURCES