Maternal Plasma Choline Levels Are Positively Correlated with Maternal and Placental Phospholipid-DHA Content in Females with Obesity Who Receive DHA Supplementation

Abstract

Background

Pregnancies complicated by maternal obesity are characterized by metabolic differences affecting placental nutrient transport and fetal development. Docosahexaenoic acid (DHA) is critical for fetal brain development and is primarily incorporated into phosphatidylcholine (PC). Recent evidence suggests that choline may enhance PC-DHA synthesis; however, data on the impact of maternal plasma choline on placental phospholipid DHA content in females with obesity are limited.

Methods

We conducted a secondary analysis of a DHA supplementation trial (800 mg/d) in 38 pregnant females with obesity (body mass index ≥30 kg/m²). Blood samples at 36 wk gestation and term placentas were analyzed for phospholipids using mass spectrometry. Choline transporter-like (CTL) proteins in the syncytiotrophoblast microvillous (MVM) and basal plasma membranes were quantified by Western blot.

Results

Daily DHA supplementation from 25 wk gestation was associated with higher maternal plasma and placental PC- and lysophosphatidylcholine (LPC)-DHA. A significant interaction (P interaction <0.05) between DHA supplementation and choline indicated that higher choline enhanced the incorporation of DHA into plasma PC. MVM CTL-1 expression was correlated with placental total PC-DHA and LPC-DHA content, suggesting that CTL-1 has a predominate role in placental choline uptake and phospholipid synthesis.

Conclusions

These findings suggest that choline may influence maternal PC- and LPC-DHA synthesis and plasma levels, as well as the expression of placental choline transporters and the resulting PC- and LPC-DHA content in females with obesity. These relationships may have implications for DHA transport to the fetus and overall fetal development.

Introduction

Almost two-thirds of females entering pregnancy are either overweight or obese [1]. Pregnancies complicated by obesity are associated with elevated circulating plasma lipids levels and increased placental triglyceride content [[2], [3], [4]]. Furthermore, females with obesity during pregnancy experience endocrine and metabolic disturbances that can affect placental function and the transport of nutrients that are essential for fetal development [2,5,6]. DHA (22:6 n-3) is an omega-3 long-chain PUFA required for fetal growth and the development of the brain, vision, and cognitive function. Because of low fetal and placental delta-5 and detla-6 desaturase enzyme activity, placental transfer of DHA from maternal circulation is necessary to meet fetal requirements [7,8]. Maternal dietary DHA deficiencies can impair fetal brain growth, neurogenesis, cognitive function, and vision, as demonstrated in animal models [[9], [10], [11]]. Perinatal DHA status has been associated with neurological outcomes in children [[10], [11], [12], [13], [14]]. Recent studies have shown that obesity is associated with lower DHA levels in both nonpregnant [15] and pregnant females [16,17]. Perinatal DHA status has also been reported to be associated with neurological function in children [[10], [11], [12], [13], [14]]. Thus, factors that impair placental function in pregnant females with obesity may also impact DHA metabolism and transport to the fetus.

Maternal DHA is primarily incorporated into phosphatidylcholine (PC) by 2 main pathways: 1) de novo from glycerol-3-phosphate, choline, and fatty acids through the cytidine diphosphate-choline (CDP-choline) pathway; or 2) conversion from phosphatidylethanolamine via successive methylation steps involving S-adenosylmethionine (SAM) in the phosphatidylethanolamine N-methyltransferase (PEMT) pathway. In pregnancy, choline supports maternal PC synthesis through the CDP-choline pathway or indirectly as a methyl donor for SAM production in the PEMT pathway [18]. Prenatal choline supplementation (550 mg/d) has been shown to enhance PC synthesis through the PEMT pathway, increase hepatic DHA export, and elevate maternal plasma PC-DHA levels, suggesting that maternal choline may be a limiting factor in PC-DHA synthesis [19,20]. However, phospholipid synthesis and export by the placenta remains poorly understood.

Maternal PC-DHA can be cleaved to produce lysophosphatidylcholine (LPC)-DHA, which is preferentially transferred across the blood-brain barrier, making LPC-DHA critical for brain development [21]. Although maternal nonesterified DHA is transported across the placenta into the fetal circulation, recent data suggest that LPC-DHA is also transported into fetal circulation, resulting in 4-fold higher concentrations in umbilical venous blood compared with maternal levels [16]. Compared with pregnancies with weight in a normal range, pregnant females with obesity have reduced maternal and umbilical cord LPC-DHA concentrations [16,17], potentially impacting fetal brain DHA availability and neurodevelopment negatively. Interestingly, prenatal choline supplementation has been associated with increased maternal PEMT-derived LPC-DHA concentrations [22], although its effect on fetal DHA levels remains unclear. These findings suggest that infants born to females with obesity and low DHA intake may benefit from optimizing maternal plasma choline concentrations alongside DHA supplementation to enhance maternal and placental LPC-DHA content.

In our previously conducted randomized controlled trial, we demonstrated that DHA supplementation in pregnant mothers with obesity significantly increased placental membrane DHA content, decreased placental inflammation, modified amino acid transporter expression, and increased fatty acid transporting protein 4 expression [5]. However, it is unknown whether maternal DHA supplementation influences maternal and/or placental PC- and LPC-DHA content and how maternal choline status affects these outcomes. Therefore, we conducted a secondary analysis of samples from our previously published randomized control trial to assess the relationship between maternal plasma choline and maternal plasma and placental PC- and LPC-DHA content in placebo and DHA supplemented pregnant females with prepregnancy obesity. We hypothesized that maternal DHA supplementation increases placental PC- and LPC-DHA content and that these levels positively correlate with maternal plasma choline.

Methods

Ethics

Informed, written consent was obtained from all participating subjects after careful explanation of the study. The study protocol was approved by the Institutional Review Board at the University of Texas Health Science Center, San Antonio (IRB HCS20090506H).

Study participants

This secondary analysis used samples from a randomized trial of DHA supplementation during pregnancy in subjects with prepregnancy obesity (BMI ≥30 kg/m²). The study details have been previously published [5,23]. In brief, 38 pregnant females were enrolled at 25–29 wk gestation and were randomly assigned to receive either a placebo (corn/soy oil) or DHA supplementation (800 mg/d; Dutch State Mines (DSM) Nutritional Products) for the remainder of their pregnancy. Supplementation began at enrollment, with an average duration of 15 wk. A blood sample was collected at 24 wk gestation to measure total DHA levels and at 36 wk to confirm adherence to randomization. A modified food frequency questionnaire was used to assess dietary omega-3 intake, and women with high reported consumption were excluded from participation. Among those enrolled and randomly divided, there were no significant differences in dietary omega-3 intake between the treatment and control groups [23]. To ensure consistency, all participants were provided with the same prenatal vitamin, which did not contain choline or DHA. Before enrollment, participants were asked about additional supplement use and were requested to discontinue any use if they agreed to participate. We measured plasma phospholipids and choline using the maternal blood samples collected at 36 wk gestation, and we measured placental phospholipids and choline in placental tissue collected from 35 term deliveries (≥37 wk gestation). Maternally facing microvillous membrane (MVM) and fetal-facing basal plasma membrane (BM) were isolated from all placentas collected.

Sample collection

Placenta samples were collected immediately after delivery, dissected, rinsed, and then homogenized with a Polytron (15,000 x g, 2 min) in ice-cold buffer D (250 mM sucrose 10 mM Hepes, pH 7.4, with protease and phosphatase inhibitors). Maternal plasma was collected from blood samples using standard methods. Placental homogenate and plasma samples were aliquoted and stored at −80°C until further processing and analysis.

Quantification of maternal plasma and placental phosphatidylcholine species

Phospholipids were extracted from maternal plasma and placental homogenate and quantified by liquid chromatography-tandem mass spectrometry in the Nutrition and Obesity Research Center Lipidomics Core facility at the University of Colorado Anschutz Medical Campus [24]. A total of 50 μL of maternal plasma and the volume of placenta homogenate equivalent to 1 mg protein was used for extraction, brought up to a total volume of 750 μL in water, and methanol (900 μL) was added. An internal standard cocktail containing 19:0_19:0 PC(2000 pmol), and d7-18:1 LPC (200 pmol) was added and lipid extraction was performed by the addition of methyl-tert-butyl ether (3 mL) according to Matyash et al. [24].

For phospholipid analysis, samples were injected onto an HPLC system connected to a triple-quadrupole mass spectrometer (Sciex 3200) and normal phase chromatography was performed using a silica column (150 × 2 mm, Luna Silica 5 μm, Phenomenex). The mobile-phase system consisted of solvent A (isopropanol: hexane:water [58:40:2, v/v/v]) and solvent B (hexane:isopropanol:water [300:400:84, v/v/v]) both containing 10 mM ammonium acetate. Mass spectrometric analysis was performed in the negative-ion mode using multiple reaction monitoring. The precursor ions monitored were the acetate adducts [M+CH3COO]⁻ for PC and LPC. The product ions analyzed after collision-induced decomposition were the carboxylate anions corresponding to the acyl chains esterified to the glycerol backbone. Quantitative results were generated using stable isotope dilution with standard curves for saturated and unsaturated PC compounds and results were normalized to protein content.

Quantification of maternal plasma total choline

Maternal plasma total choline was assayed by the Colorado Translational Research Center Metabolomics Core Laboratory, University of Colorado, Denver. We performed a single-phase extraction using a lysis buffer composed of methanol, acetonitrile, and water (5:3:2) and agitated for 30 min at 4°C. After centrifugation at 10,000 g, the supernatant was collected and stored at −80°C until analysis with an ultra-performance liquid chromatography-tandem mass spectrometer. Metabolites were identified using the Maven Metabolomic Analysis and Visualization Engine. The single-phase extraction provided a measurement of total choline in the sample; this method does not distinguish between the aqueous and organic phases of choline.

Isolation of syncytiotrophoblast microvillous and basal plasma membranes

Syncytiotrophoblast MVM and BM were isolated from placental homogenates (n = 18/placebo, n = 17/DHA) by differential ultracentrifugation, Mg²⁺ precipitation, followed by sucrose density gradient centrifugation as previously described [25,26]. MVM enrichment was determined by the MVM/homogenate ratio of alkaline phosphatase activity and has been previously reported [27,28]. Alkaline phosphatase activity was, on average, ∼18-fold greater in isolated MVM vesicles, and ferroportin expression, a BM marker, was 30-fold higher in BM vesicles compared with activity/expression in placental homogenates, confirming high enrichment of both membrane fractions. There were no differences in membrane enrichment in DHA supplemented compared with placebo groups.

Immunoblotting

Protein abundance of choline transporter-like protein (CTL)-1 was measured in MVM and BM using the ProteinSimple Jess capillary immunoblotting system (SM-PN01-1, Protein Simple) as previously described [29]. Briefly, 0.5 μg/μL of isolated MVM and BM vesicles were loaded into individual capillaries for analysis. Proteins were probed using primary antibodies for CTL-1 (BS-3667R, Bioss, 1:100). Protein abundance was normalized to total capillary protein measured using the total protein detection module (AM-PN01, Protein Simple). Analysis was performed using Compass for Simple Western software [30].

Data analysis and statistics

Before conducting the main analysis, we assessed whether covariate adjustment was necessary by examining bivariate associations between participant sociodemographic and clinical characteristics with erythrocyte DHA content (expressed as a percent of total fatty acids) and plasma choline levels using t-tests. Age and prepregnancy weight were categorized as greater than compared with less than or equal to the median value within the study. We compared mean ± SE of erythrocyte DHA content and the concentrations of plasma and placental PC- and LPC-DHA lipids between supplementation groups to determine whether DHA supplementation increased PC- and LPC-DHA levels. Given that choline has been shown to influence hepatic PC-DHA synthesis and export, we hypothesized a priori that maternal plasma choline modifies the association of DHA supplementation in both maternal plasma and placental PC- and LPC-DHA incorporation. We tested for effect modification by plasma choline using general linear models to evaluate multiplicative interaction with supplementation group. To illustrate this effect modification, we stratified choline levels as higher compared with lower or equal to the median (>16.5 μM compared with ≤16.5 μM) and plotted differences in plasma and placental 16:0/22:6 PC concentrations in placebo and DHA supplementation groups. A Dunnett post hoc correction was applied to account for multiple comparison against the reference group, which in this analysis was DHA supplemented subjects with choline ≤16.5 μM. Pearson correlation coefficients were used to measure continuous linear associations between maternal and placental outcomes. Statistical analyses were performed using SAS software version 9.4. A P < 0.05 from 2-sided tests was considered statistically significant.

Results

Characteristics of the analytical sample

The characteristics of the cohort are presented in Table 1. The mean (SE) participant age at enrollment was 28.2 (0.85) and prepregnancy BMI was 34.6 (0.6). Most subjects were Hispanic and delivered a male infant. The erythrocyte DHA content and plasma choline levels did not differ by participant characteristics at baseline.

TABLE 1.

Characteristics of the study subjects and levels of red blood cell DHA % and choline measured at 36 gestational weeks (N = 38).

Biomarker level	Full sample	Red blood cell DHA	P	Choline	P
	N (%)	%		μM
Characteristic
Age			0.09		0.84
<28 y	18 (47.4)	10.0 ± 0.6		15.6 ± 0.8
≥28 y	20 (52.6)	9.9 ± 0.4		15.8 ± 0.7
Race/ethnicity			0.96		0.44
Non-Hispanic black	4 (10.53)	8.1 ± 1.8		14.7 ± 1.4
Hispanic	34 (89.47)	8.0 ± 0.5		15.9 ± 0.5
Employment status			0.47		0.62
Full-time	19 (51.35)	7.4 ± 0.7		15.3 ± 1.2
Part-time	6 (16.22)	8.5 ± 1.5		16.5 ± 1.0
Unemployed	12 (32.43)	8.7 ± 0.8		15.4 ± 0.7
Prepregnancy weight			0.67		0.44
<185 lbs.	22 (57.9)	8.2 ± 0.7		15.4 ± 0.5
≥185 lbs.	16 (42.1)	7.8 ± 0.7		16.2 ± 1.0
Supplement use in pregnancy			0.33		0.31
None	2 (5.71)	4.9 ± 0.6		11.7 ± 0.4
Prenatal	31 (88.57)	8.2 ± 0.6		15.9 ± 0.5
Multi-vitamin	2 (5.71)	8.3 ± 1.5		14.4 ± 0.3
Seafood allergies			0.62		0.97
No	36 (94.74)	8.1 ± 0.5		15.7 ± 0.5
Yes	2 (5.26)	7.0 ± 2.3		15.7 ± 1.7
Pregnancy complication1			0.79		0.14
No	30 (81.08)	8.0 ± 0.6		15.4 ± 0.6
Yes	7 (18.92)	7.7 ± 1.2		17.3 ± 1.2
Fetal sex			0.84		0.23
Male	26 (68.42)	8.2 ± 0.7		16.1 ± 0.6
Female	12 (31.58)	8.0 ± 0.7		14.8 ± 0.9

Mean ± SE.

¹Includes anxiety, kidney infection, hypertension at term, preeclampsia, cesarean section, miscarriage.

Maternal DHA supplementation increases maternal plasma and placental PC-DHA and LPC-DHA

At 36 wk gestation, erythrocyte DHA content and plasma choline levels were significantly higher in the DHA supplemented group compared with placebo (Table 2). All individual plasma PC-DHA species, including plasma LPC-DHA and total PC-DHA, were higher among the DHA supplementation group compared with placebo treated group (P < 0.05). Similarly, individual placental PC-DHA species, including LPC-DHA and total PC-DHA, were higher among the DHA supplementation group compared with placebo (P < 0.05, Table 2).

TABLE 2.

Choline and biomarkers of lipid classes in plasma and placenta by supplementation group.

	DHA supplement	Placebo	P value for difference between treatment groups	P value for overall effect of interaction term DHA supplement ∗ choline
RBC 22:6 DHA (%)	10.010 ± 0.555	6.084 ± 0.522	<0.001
Choline (μM)	17.082 ± 0.684	14.367 ± 0.585	0.01
Plasma pmol lipid/μL
16:0/22:6PC	41.926 ± 2.663	27.047 ± 1.647	0.01	<.001
16:0e/22:6PC	1.993 ± 0.171	1.165 ± 0.078	<0.001	0.26
16:1e/22:6PC	1.034 ± 0.074	0.599 ± 0.046	<0.001	<.001
18:0/22:6PC	22.362 ± 1.546	14.250 ± 1.086	<0.001	<.001
18:0e/22:6PC	1.870 ± 0.161	1.166 ± 0.066	<0.001	<.001
18:1/22:6PC	4.102 ± 0.271	2.959 ± 0.175	<0.001	0.01
18:1e/22:6PC	2.921 ± 0.228	1.832 ± 0.123	<0.001	0.01
22:6-LPC	0.233 ± 0.025	0.144 ± 0.014	<0.001	0.05
Total PC-DHA	76.441 ± 4.962	49.163 ± 3.044	<0.001	0.01
Placenta pmol lipid/μg protein
16:0/22:6PC	0.363 ± 0.017	0.260 ± 0.013	0.01	0.33
16:0e/22:6PC	0.041 ± 0.003	0.031 ± 0.002	<0.001	0.06
16:1e/22:6PC	0.042 ± 0.003	0.030 ± 0.002	<0.001	0.02
18:0/22:6PC	0.160 ± 0.010	0.111 ± 0.008	<0.001	0.41
18:0e/22:6PC	0.017 ± 0.001	0.013 ± 0.001	<0.001	0.41
18:1/22:6PC	0.105 ± 0.006	0.073 ± 0.005	<0.001	0.09
18:1e/22:6PC	0.049 ± 0.004	0.036 ± 0.003	<0.001	0.10
22:6-LPC	0.0009 ± 0.0001	0.0005 ± 0.0001	0.01	0.61
Total PC-DHA	0.777 ± 0.042	0.552 ± 0.031	0.01	0.16
P2 DHA (%)	9.785 ± 0.814	6.782 ± 0.510	<0.001	0.95

Mean ± SE.

Abbreviations: P2, placental membrane fraction; PC, phosphatidylcholine.

Maternal plasma choline and associations with plasma and placental PC-DHA species

Maternal choline at 36 wk was positively associated with all maternal plasma PC-DHA and LPC-DHA species measured at 36 wk (Figure 1A–H, P < 0.05). Positive correlations were also observed between maternal plasma choline at 36 wk and specific placental PC species (16:0/22:6 PC; 18:0/22:6 PC; 22:6 LPC) and total PC-DHA (Figure 2A–E). Maternal choline was not correlated with placental PC containing vinyl ether linkages (PC 16:0e/22:6, PC-16:1e/22:6, PC-18:0e/22:6, PC-18:1e/22:6 species, P > 0.05, data not shown).

FIGURE 1.

Maternal choline is positively correlated with plasma phosphatidylcholine species containing DHA. Correlations between maternal plasma total choline with maternal plasma phosphatidylcholine (PC) species 16:0_22:6 PC (A), 16:0e_22:6 PC (B), 16:1e_22:6 PC (C), 18:0_22:6 PC (D), 18:0e_22:6 PC , 18:1_22:6 PC (F), 18:1e_22:6 PC (G), and total 22:6 PC (H). Pearson’s correlation coefficient, P < 0.05 was considered statistically significant.

FIGURE 2.

Maternal choline is positively correlated with placenta phosphatidylcholine species containing DHA. Correlations between maternal plasma total choline with placenta phosphatidylcholine (PC) species 16:0_22:6 PC (A), 18:0_22:6 PC (B), 22:6 lysophosphatidylcholine (LPC) (C) and total 22:6 PC (D). Pearson’s correlation coefficient, P < 0.05 was considered statistically significant.

When assessing differences in maternal plasma PC-DHA and LPC-DHA by supplementation group, all species except plasma 16:0e/22:6 PC (P = 0.26) and LPC-DHA (P = 0.05) demonstrated a significant positive interaction between maternal choline levels and DHA supplementation on maternal plasma PC-DHA levels (P for interaction < 0.05, Table 2). Conversely, most tests for interaction between DHA supplementation and maternal plasma choline with placental PC-DHA and LPC-DHA were not significant (Table 2). The only significant interaction observed for placental phosphlipids was for PC-16:1e/22:6, where maternal choline and DHA supplementation had a significant negative interaction effect (P = 0.02). Thus, although several placental PC-DHA species were correlated with maternal choline, a positive interaction between supplementation and maternal choline on placental PC-DHA species was not found.

To illustrate the interactive effects of DHA supplementation and choline, Figure 3 shows plasma and placenta 16:0/22:6 PC concentrations stratified by choline levels above compared with below or equal to the median. Females with higher choline levels who received DHA supplementation had higher mean plasma 16:0/22:6 PC levels compared with those with lower choline levels (47.4 pmol/μL ± 2.2 compared with 37.2 pmol/μL ± 3.7, P = 0.02, Figure 3A). The P value was attenuated to 0.06 after applying a Dunnett post hoc correction. As previously mentioned, although several placental PC-DHA species were correlated with maternal choline, the effect of DHA supplementation on placental 16:0/22:6 PC did not differ by choline status (Figure 3B).

FIGURE 3.

Effect of DHA supplementation on plasma and placenta 16:0/22 PC by choline levels. Mean and SE derived from linear regression models of the effect of DHA supplementation on plasma 16:0/22 PC (A) and placenta 16:0/22 PC (B), both stratified by choline levels ≤16.5 μM compared with >16.5 μM, which is the median choline level at 36 wk in this study. Data are stratified to show the significant interaction of DHA supplementation ∗ choline on plasma and placenta 16:0/22 PC. P values are for comparisons of mean differences in plasma and placenta 16:0/22 PC among females supplemented with DHA by choline status. Punadj., unadjusted P value; Pdunnett, Dunnett adjusted P value for multiple comparisons setting the DHA and low choline group as the reference.

Association of maternal plasma and placental PC-DHA species

We determined correlations between maternal and placental PC-DHA or LPC-DHA species. Most correlations between maternal plasma and placental PC-DHA species, including total and LPC-DHA, were not significant. The exception was 16:0/22:6 PC, which showed a significant correlation between plasma and placental levels (R² = 0.11, P = 0.03, Supplemental Table 1).

Syncytiotrophoblast choline-like transporter-1 expression and correlation with placental PC-DHA species

Placental CTL-1 expression was significantly higher in the MVM compared with the BM (+55%, P < 0.05), with similar MVM CTL-1 levels between the placebo and DHA supplemented groups (P > 0.05, Figure 4A). MVM CTL-1 expression was positively correlated with placental total PC-DHA (R² = 0.31, P = 0.02, Figure 4B) and LPC-DHA content (R² = 0.23, P = 0.04, Figure 4C).

FIGURE 4.

Placenta choline-like transporter (CTL)-1 expression and correlation with placental total PC-DHA and LPC-DHA content. Representative image of immunoblotting bands and relative abundance of CTL-1 in MVM and BM plasma membrane vesicles (A) from placebo (n = 18) and DHA (n = 17) placentas. Correlations between MVM CTL-1 expression and placenta total PC-DHA (B) and LPC-DHA (C) content. Data are means ± SEM. P < 0.05 was considered statistically significant (Student’s t-test or Pearson’s correlation). BM, basal plasma membrane; LPC, lysophosphatidylcholine; MVM, microvillous plasma membrane; PC, phosphatidylcholine.

Discussion

In this secondary analysis of a randomized DHA supplementation trial (800 mg/d) during pregnancy, we found that daily DHA supplementation starting at 24–26 wk gestation was associated with higher maternal plasma PC- and LPC-DHA levels at 36 wk and greater placental PC- and LPC-DHA content at term. Maternal choline levels were associated with higher maternal plasma PC- and LPC-DHA and some placental PC-DHA species. However, a significant interaction between DHA supplementation and choline status was observed only for maternal plasma PC-DHA species. Given recent recognition of the importance of maternal choline for DHA status in pregnancy [31], these findings support recommendations emphasizing the optimization of both choline and DHA levels to improve maternal DHA status during pregnancy. The interaction between maternal choline and DHA on placental PC-DHA content and fetal transfer, however, remains unclear and warrants further investigation. Our data suggest that the expression of choline transporters in the placenta has a strong influence on placental PC-DHA content.

Our study aimed to determine the influence of maternal DHA supplementation on maternal and placental PC- and LPC-DHA content and to determine how maternal choline status impacts these outcomes. We found that DHA supplementation increased plasma and placental PC-DHA species, with a significant interaction showing that higher choline levels likely enhanced the incorporation of DHA into maternal plasma PCs. Although choline was correlated with some nonvinyl ether placental PC species and total placental PC-DHA, plasma PC-DHA species measured at 36 wk did not generally correlate with corresponding placental PC-DHA species at delivery. This suggests that the placenta may independently synthesize PC-DHA rather than relying on maternal uptake. Furthermore, the lack of correlation between plasma choline and placental PCs containing vinyl ethers may indicate distinct pathways for synthesizing vinyl ether PCs in the placenta compared with nonvinyl ether PC-DHA species.

Our prior work demonstrated that females with obesity have lower plasma levels of several PC-DHA species [32,33]. We recently published findings indicating that the recommended daily dose of DHA (200 mg/d) increased erythrocyte DHA in pregnancies with obesity but did not significantly impact the DHA status of normal-weight pregnant females or placental phospholipid content, regardless of obesity status [33]. A recent study in subjects with prepregnancy BMI <32 kg/m² showed that choline supplementation at 550 mg/d was sufficient to enhance PC synthesis via the PEMT pathway, increase hepatic DHA export, and elevate maternal plasma PC-DHA [19,20]. Collectively, these findings highlight the need for future research to assess whether DHA dosing should be tailored according to maternal weight and choline status, particularly because not all DHA supplementation trials have found significant effects on neonatal or childhood neurological outcomes [34].

The variability in prior study findings regarding the impact of DHA supplementation on fetal growth and offspring development [[9], [10], [11], [12], [13],34] may be related to differences in maternal choline status and DHA dosing. Our previous work showed that females with obesity had lower circulating levels of several PCs containing DHA [32,35], underscoring the need for studies focused on combined choline and DHA supplementation in this high-risk population. Such research could help refine recommendations to improve placental-fetal DHA status and long-term neurological outcomes in children.

Mechanistic insight

We previously reported a preference for phospholipid synthesis in cultured primary trophoblast cells incubated in uniformly labeled ¹³C nonesterified fatty acids, including ¹³C-DHA [36]. After 24 h, ∼40% of all PC-DHA species contained ¹³C-labeled DHA. However, DHA incorporation appeared to occur primarily through the remodeling (Lands cycle) pathway, rather than de novo synthesis. This pathway involves hydrolysis of an acyl chain. This is a known mechanism essential for incorporating long-chain PUFAs, such as DHA into phospholipids to maintain membrane fluidity and protein function, particularly in lipid rafts [37]. The remodeling pathway also produces LPC-DHA, which may be a critical form for DHA delivery to the fetal circulation by the transporter Major Facilitator Superfamily Domain containing 2a (MFSD2a) [17,38]. For instance, high levels of LPC-DHA have been observed in human umbilical circulation [17], and in a mouse model of trophoblast-specific gene targeting to reduce placental expression of MFSD2a, we found that offspring had significantly lower brain weight and phospholipid DHA brain content [39]. These data suggest that placental LPC-DHA transport is likely critical for DHA transfer across the placenta and to the fetus [39].

Given this strong preference for phospholipid synthesis, and a role for MFSD2a and LPC-DHA transport in the placenta, we propose that adequate maternal choline levels are necessary for efficient DHA incorporation into PCs in maternal and placental compartments, facilitating its preferential transport to the fetus as LPC-DHA. Our results suggest that maternal plasma choline is not the rate-limiting factor for placental phospholipid DHA incorporation. Rather, placental choline transporters may have a significant role in this process. Uptake of choline by the human placenta is believed to be mediated by transporters in the MVM, specifically CTL-1 and CTL-2 [40]. CTL-1, a high-affinity choline transporter, is predominantly localized in the MVM, mediating choline uptake into the syncytiotrophoblast from the maternal circulation [41]. In this study, MVM CTL-1 expression was correlated with placental total PC-DHA and LPC-DHA content, underscoring its role in choline uptake from the maternal circulation and subsequent placental phospholipid synthesis. This may explain the lack of interaction between maternal choline, DHA supplementation, and placental PC-DHA levels. The expression of placental choline transporter CTL-1 may be a rate-limiting step in placental PC synthesis by the trophoblast given that the expression of CTL-1 was positively correlated with placental PC-DHA species.

Furthermore, the data reported here suggest that maternal liver and placental phospholipid synthesis are largely independent. Although choline is likely crucial for PC synthesis in both the liver and placenta, there appear to be important differences in processing and utilization. Given that plasma PC-DHA species did not generally correlate with corresponding placental PC-DHA species at delivery, the lack of correlation between maternal plasma and placental PC- and LPC-DHA species points to independent synthesis pathways in each organ, involving both the de novo (Kennedy) and remodeling (Lands cycle) pathways. Our data also support that intact PC species are not directly transported across the placenta and highlight the importance of essential precursors for placental phospholipid synthesis such as glycerol, fatty acids, and head group constituents such as choline.

Limitations and future directions

Our study has several limitations that should be addressed in future research. First, we measured plasma choline at a single time point (36 wk gestation), limiting our ability to assess changes in choline levels across gestation and in relationship to DHA supplementation. Second, we did not collect cord blood, preventing a comprehensive analysis of the mechanism by which maternal PC-DHA is transferred to the fetus. Future studies should incorporate the collection of maternal blood, placental tissue, and cord blood to better elucidate this transfer mechanism. Third, we did not conduct a detailed dietary assessment, which restricted our ability to identify the dietary source of choline. Inclusion of multiple dietary recalls during pregnancy in future studies would help identify intake sources and changes over time. Fourth, there is a lack of consensus on the most robust biomarker for dietary choline intake. Although total choline and choline-containing compounds, such as PCs, are commonly used as surrogate measures of maternal choline supply [42], this remains an area for further investigation [43]. In this study, we quantified total choline using ultra-performance liquid chromatography-tandem mass spectrometer, a sensitive and valid method for detecting choline and its assigned metabolites [44]. Future studies should consider 2-phase extraction methods to measure both lipid-soluble choline forms, which are involved in membrane synthesis and signaling and aqueous forms in conjunction with free choline and betaine that have alternative biological roles in 1-carbon metabolism and methylation pathways. Fifth, larger studies are needed to ensure sufficient power to detect significant subgroup effects, such as those based on fetal sex. Lastly, the dose of DHA used in this study (800 mg/d) was higher than the currently recommended 200–300 mg/d. Therefore, the findings may not reflect the relationship between DHA and choline in pregnant females taking DHA at currently recommended doses.

Conclusions

Maternal DHA supplementation at 800 mg/d during pregnancy in females with obesity significantly increased maternal plasma PC- and LPC-DHA levels. Maternal choline levels had a positive interaction with plasma PC-DHA, indicating that higher choline may enhance DHA incorporation into plasma phospholipids. Although DHA supplementation resulted in greater placental PC- and LPC-DHA content, maternal plasma choline was likely not a rate-limiting factor for placental DHA incorporation, as suggested by the correlation between placental PC-DHA levels and CTL-1 expression in syncytiotrophoblast MVM. This finding underscores the critical role of placental choline transporters, such as CTL-1, in mediating and facilitating choline uptake from the maternal circulation and promoting phospholipid synthesis.

Our data suggest that maternal choline availability could be important for optimizing plasma PC- and LPC-DHA synthesis, with potential downstream effects on placental levels. However, the independence of maternal liver and placental phospholipid synthesis, as highlighted by the lack of correlation between maternal plasma and placental PC-DHA species, indicates that targeted supplementation strategies may need to consider both maternal and placental requirements. Future studies should investigate whether choline and DHA supplementation can better support placental lipid synthesis and transport, ultimately enhancing fetal brain development. Additionally, research should address variations in maternal choline status, standardize DHA dosing, and potential subgroup effects to refine recommendations for pregnant females, especially those with obesity.

Author contributions

The authors’ responsibilities were as follows – JHD, ECF, TJ, TLP: designed research (project conception, development of overall research plan, and study oversight); JHD, KZB: conducted research (hands-on conduct of the experiments and data collection); ECF: performed statistical analysis; JHD, ECF, TLP: wrote the paper and had primary responsibility for final content; and all authors: read and approved the final manuscript.

Data availability

Data described in the manuscript, code book, and analytic code will be made available upon request to TLP after publication.

Funding

The study was supported by grants from NIH (grant numbers HD104644, HD065007, HD068370 and HD007186). ECF is supported by NIH (HD108272). The Lipidomics Core Facility is supported by the Colorado Nutrition Obesity Research Center grant P30DK048520.

Conflict of interest

The authors report no conflicts of interest.

Appendix A. Supplementary data

The following is the Supplementary data to this article: