Placental transport dysfunction in omega-3 and omega-6 fatty acid in patients with gestational diabetes mellitus

Abstract

Background

Gestational diabetes mellitus (GDM) affects the transfer of fatty acid (FA) from mother to fetus. However, the specific transport alterations remain unclear. This study investigated FA transport from mother to fetus in GDM.

Methods

Maternal serum, umbilical cord serum, and placental tissues were collected from 20 healthy (normal group) and 20 GDM pregnancies receiving glycemic management (GDM group). Fifty-one FA were quantified by gas chromatography–mass spectrometry. Placenta/mother and fetus/placenta ratios of FA compositions were compared between groups using Student’s t-test or Mann–Whitney U test, as appropriate.

Results

The concentrations and compositions of FA profiles were both changed in GDM. The placenta/mother ratios of cis-11,14,17-eicosatrienoic acid (GDM vs normal: 0.648 vs 0.484, p = 0.001), cis-5,8,11,14,17-eicosapentaenoic acid (0.449 vs 0.338, p = 0.048), cis-13,16-docosadienoic acid (DDA, 1.436 vs 1.187, p = 0.047), palmitelaidic acid (2.882 vs 2.272, p = 0.023), petroselaidic acid (2.177 vs 1.746, p = 0.004) and linoelaidic acid (1.770 vs 1.423, p = 0.014) were significantly higher in GDM, while the ratio of trans-10-nonadecenoic acid (0.537 vs 0.799, p < 0.001) was lower. The fetus/placenta ratio of total omega-6 long chain polyunsaturated FA (LC PUFA, 0.545 vs 0.590, p = 0.002) with DDA (4.031 vs 4.951, p = 0.031) and cis-7,10,13,16-docosatetraenoic acid (0.494 vs 0.778, p < 0.001) was lower in GDM.

Conclusions

FA profiles in mother, placenta, and fetus were affected by GDM despite glycemic management, especially in omega-3 and omega-6 LC PUFAs. This finding indicates that addressing fetal FA deficiency may depend on correcting placental dysfunction rather than solely increasing maternal FA supplementation, offering insights for developing nutritional and therapeutic strategies for GDM.

KEY MESSAGES

1. The first study utilizing samples from mothers, placenta, and fetuses from the same cohort to evaluate the transport characteristics of fatty acid profiles in GDM.

2. Fatty acid profiles of the mother, placenta and fetus were affected by GDM despite glycemic management.

3. GDM disrupts placental transport of omega-3, omega-6 long chain polyunsaturated fatty acids and trans fatty acid.

Graphical Abstract

Introduction

Gestational diabetes mellitus (GDM), a common metabolic disorder of pregnancy, affects an estimated 14.0% of pregnancies globally, with significant variations across populations and diagnostic criteria [1]. The prevalence of GDM screened before 20 weeks of gestation ranges widely between countries (0.7%-36.8%). Notably, studies conducting early screening before 12 weeks report a prevalence range from 0.7% to 14.2% [1]. GDM is associated with an increased risk of adverse perinatal outcomes, such as small for gestational age infants [2], preterm birth [3], and impacts the future health of both mother and child, potentially leading to the development of metabolic and cardiovascular disease, obesity, and neurodevelopmental disorders in later life [4]. Pregnant women who are diagnosed with GDM are treated with dietary advice or insulin. However, achieving satisfactory glycemic control does not guarantee improved pregnancy outcomes [5]. This indicates that other metabolites, such as amino acids and lipids, are altered in GDM, potentially leading to short- or long-term adverse outcomes [4].

Fatty acid (FA) are essential nutrients during pregnancy to meet the needs of maternal and fetal growth and development [6]. Studies have investigated changes in circulating FA in GDM mothers, with a particular focus on polyunsaturated FA (PUFA) [5,7–9]. The fetal brain and central nervous system require a significant amount of long chain PUFAs (LC PUFAs), especially arachidonic acid (AA) and docosahexaenoic acid (DHA), and their deficiency can affect brain development, visual function and fetal growth [10]. A meta-analysis of 21 studies found that mothers with GDM have a higher proportion of DHA and AA in their total circulating FA, suggesting the presence of LC PUFA metabolic disorders in GDM mothers [11]. The placenta, as the maternal-fetal interface, connects the maternal and fetal circulations and plays a key role in regulating the molecular-fetal exchange that is important for fetal nutrition. Studies have demonstrated that GDM is associated with impaired placental function and decreased the placental uptake of LC PUFAs [12,13]. Umbilical cord serum has been used to assess changes in FA in fetus. A meta-analysis of 11 observational studies found that levels of AA, DHA, and omega-6 and omega-3 PUFA were lower in umbilical cord blood of mothers with GDM than in controls [11]. A prospective large cohort study revealed that maternal levels of saturated FA (SFA) and DHA during pregnancy were significantly associated with neurodevelopmental outcomes in offspring at 1 year old with GDM mothers [14]. Another study demonstrated that obese pregnant women with GDM had higher levels of diacylglycerol and triacylglycerol (containing multiple FAs), which were predictors of higher birth weights and larger abdominal sizes in newborns [15]. These findings highlight the importance of clarify the effects of GDM on placental FA transport, which is essential for understanding how GDM influences neurodevelopmental abnormalities or disrupts metabolism.

Previous studies have explored the effects of GDM on maternal or fetal outcomes separately, there is a noticeable gap in the concurrent analysis of maternal, placental, and fetal samples to understand the integrated pathways of FA transport. This uncertainty is mainly due to the absence of research simultaneously analyzing FA levels in an individual’s maternal blood, placental tissue, and fetus. In the present study, we analyzed the profiles of FA in maternal peripheral blood, placenta and fetal umbilical vein blood, aiming to analyze the differences in FA profiles between GDM and normal pregnancy. Additionally, we sought to evaluate the impact of GDM on the placental transport of exogenous FA, including essential FA (EFA), LC PUFA and trans FA (TFA), which are derived from maternal dietary intake and transported across the placenta [16].

Materials and methods

Patients

We conduct a study and randomly selected 20 pregnant women diagnosed with GDM who underwent glycemic control under medical supervision at Changzhou Maternal and Child Health Care Hospital between January and March 2023. Additionally, we selected 20 healthy pregnant women during the same period as the normal group. All participants met the following criteria: (a) pregnant women with complete medical records; (b) underwent a cesarean section between 37 and 41 weeks of gestation. Exclusion criteria were as follows: (a) preexisting heart, liver, or kidney disease, hypertension, diabetes, hypothyroidism, hyperthyroidism or other metabolic diseases before pregnancy; (b) hypertension, hypothyroidism, hyperthyroidism or other diseases specific to pregnancy; (c) twin or multiple pregnancies.

The diagnosis of GDM was made according to the International Association of Diabetes and Pregnancy Study Group criteria (showed in Supplementary files) [17]. The GDM group performed self-monitored blood glucose (SMBG) managements combined with dietary modifications and/or insulin therapies. Maternal peripheral serum, umbilical cord serum and placental tissue were collected from these participants. This study was performed in accordance with the Declaration of Helsinki principles and approved by the Institutional Ethics Committee of Changzhou Maternal and Child Health Care Hospital (approval number: 2022[75]). Written informed consent was obtained from all study subjects.

Sample preparation

Maternal peripheral blood samples (3.5 mL) were collected after an overnight fast within one week before delivery. Umbilical cord bloods (3.5 mL) were collected from umbilical vein immediately after cesarean delivery. The blood samples were centrifuged at 3,500 rpm for 5 min at 4 °C to separate the serum [5,9]. Placental tissues were collected at a similar site on placenta within 30 min of cesarean section and washed thoroughly with sterile phosphate-buffered saline immediately to avoid contamination with red blood cells [18]. The decidua was removed to avoid contamination with maternal tissue. Only villous tissue was dissected from the remaining tissue. All samples were stored at −80 °C for testing.

FA detection

For serum sample, 50 μL was mixed with 2 mL of a 1% sulfuric acid methanol solution and vortexed thoroughly. The mixture was esterified in water bath at 80 °C for 30 min. 1 mL n-hexane was added, vortexed and stand for 5 min. 5 mL H₂O (4 °C) was added for washing and centrifuged at 3500 rpm for 10 min and 700 μL of supernatant was transferred to a 2 mL centrifuge tube. 100 mg anhydrous sodium sulfate was added to remove excess water and vortexing. 300 μL supernatant was transferred to a new centrifuge tube. 15 µL methyl salicylate was added as an internal standard. The mixture was transferred to 200 μL supernatant for GC-MS (gaschromatography-mass spectrometry) analysis. For placenta tissue, 50 mg tissue was accurately weighed and 1 mL chloroform methanol (v:v = 2:1) solution and 100 mg glass beads were added. Then been put into a high-throughput tissue grinder and shaken twice at 55 Hz for 1 min. The sample was placed in an ultrasonic instrument for 30 min and centrifuged at 12,000 rpm for 5 min at 4 °C. The supernatant was taken into a 10 mL glass centrifuge tube. The rest of the detection procedure was same as the serum sample [19].

GC analysis was performed using a trace 1300 gas chromatograph (Thermo Fisher Scientific, USA), with a Thermo TG-FAME capillary column (50 m*0.25 mm ID*0.20 μm). Helium was the carrier gas, with a flow velocity of 0.63 mL/min. The split ratio was 8:1 and the injection volume was 1 μL. The injector temperature was 250 °C, the temperature of the ion source and transfer line were 300 °C and 280 °C, respectively. Oven temperature programming consisted of an initial temperature of 80 °C held for 1 min, an increase in temperature to 160 °C at 20 °C/min, held for 1.5 min; then increased to 196 °C at 3 °C/min, held for 8.5 min; and finally increased to 250 °C at 20 °C/min, held for 3 min. MS analysis was performed on an TSQ 9000 (Thermo Fisher Scientific, USA) for serum sample or an ISQ 7000 (Thermo Fisher Scientific, USA) for tissue sample using the electron impact ionization mode. Single ion monitoring mode was used with the electron energy of 70 eV [19,20].

To determine the concentrations of FA in the samples, a mixed standard stock solution of 51 FAs (Supplementary Table 1) was prepared and diluted to 10 different concentrations, and a standard curve (4000 μg/mL) was plotted from 1 to 2000 μg/mL (1, 5, 10, 25, 50, 100, 250, 500, 1000 and 2000 μg/mL). Stock solutions were stored at −20 °C before use and working solutions were prepared when using. The concentration of each standard was plotted on the x-axis and the ratio of the peak area between the standard and the internal target was plotted on the y-axis to derive a linear regression equation for each substance to calculate its concentration.

Statistical analyses

Characteristic data on maternal and neonatal were obtained from medical records, including age at delivery (years), gestational age (weeks), parity (primipara, multipara), fasting glucose, 1 and 2-hour-post-oral glucose tolerance test (OGTT) glucose, infant sex (male, female), neonatal birth height (cm) and weight (g). Maternal body mass index (BMI) in late pregnancy was calculated by dividing the weight before delivery by the square of the height (kg/m²). Categorical data were summarised as numbers and percentages. For quantitative variables, normally distributed data were presented as mean with standard deviation (SD) and non-normally distributed data were presented as median with interquartile range (IQR). FA concentration was shown as median with IQR and the composition of each FA was expressed as a percentage of the total FA identified. The concentration or composition of total FAs was calculated as a sum of individual FAs. Differences of quantitative variables between case and control groups were tested by the Student’s t-test or Mann-Whitney test, when appropriate. Specifically, Student’s t-test was applied to normally distributed continuous variables, whereas the Mann-Whitney test was used for variables that did not follow a normal distribution. Categorical data were analysed using chi-square test. All analyses were performed using R (version 3.6.3, http://www.R-project.org). Two-sided P values less than 0.05 were considered statistically significant.

Results

Clinical data of the normal and the GDM groups

Baseline clinical characteristics of mothers and their newborns in the GDM and normal groups are presented in Table 1. There were no differences in mother age at delivery, gestational weeks, parity, maternal BMI, weight gain during pregnancy, infant sex, infant birth length, and birth weight between the GDM group and the normal group (p > 0.05). Fasting glucose, 1-hour-post-OGTT glucose, and 2-hour-post-OGTT glucose between 24 and 28 weeks of gestation in the GDM group were higher than the normal group (p < 0.05).

Table 1.

Clinical data of pregnant women and their newborns in the normal and GDM groups.

	Normal (n = 20)	GDM (n = 20)	P
Maternal characteristics
Age at delivery (years)	30.60 ± 4.84	32.80 ± 3.29	0.101
Gestational age (weeks)	38.26 ± 2.97	38.87 ± 0.84	0.384
Parity
Primipara	10 (50%)	7 (35%)	0.337
Multipara	10 (50%)	13 (65%)
BMI (kg/m2)	27.01 ± 1.42	28.04 ± 3.62	0.244
Weight gain during pregnancy	14.55 ± 3.30	14.74 ± 4.33	0.880
Fasting glucose (mmol/L)	4.57 ± 0.33	5.04 ± 0.61	0.005
1-hour-post-OGTT glucose (mmol/L)	7.84 ± 0.89	9.84 ± 1.48	<0.001
2-hour-post-OGTT glucose (mmol/L)	6.91 ± 0.92	8.17 ± 0.86	<0.001
Neonatal characteristics
Infant sex
Male	10 (50%)	9 (45%)	0.752
Female	10 (50%)	11 (55%)
Birth height (cm)	50.40 ± 2.30	50.05 ± 0.22	0.507
Birth weight (g)	3470 ± 316.96	3514 ± 363.11	0.685

Data are shown as mean ± SD or n (%).

Changes in profile of FAs concentration in GDM

We first analyzed the differences in FA concentrations between the GDM and normal groups (Supplementary Table 2), and used a standard Z score based on the normal group to standardize the data of FAs and visualized in Figure 1. FAs were categorised into five types according to their properties: SFA, monounsaturated FA (MUFA), omega-3 PUFA, omega-6 PUFA and TFA. From this classification perspective, no significant differences were observed between the two groups, except that the total MUFA in maternal serum was elevated in GDM. However, there were significant differences in individual FA between two groups, and these differences were tissue specific. In maternal serum, one SFA (tridecanoic acid), seven MUFAs (myristoleic acid [MOA], cis-10-pentadecenoic acid, oleic acid [OA], petroselinic acid, cis-vaccenic acid, cis-11-eicosenoic acid [11-EA] and cis-15-tetracosenoic acid), one omega-3 PUFA (alpha-Linolenic acid), three omega-6 PUFAs (cis-11,14-eicosadienoic acid [EDA], cis-13,16-docosadienoic acid [DDA], and cis-7,10,13,16-docosatetraenoic acid [DTA]), and six maternal TFAs (myristelaidic acid, trans-10-pentadecenoic acid, trans-10-heptadecenoic acid, trans-10-nonadecenoic acid, trans-11-eicosenoic acid, and brassidic acid) were elevated in GDM, while four SFAs (heneicosanoic acid, behenic acid, tricosanoic acid, and tetracosanoic acid), and one TFA (trans-7-nonadecenoic acid) were reduced (Figure 1A). In the placenta, five SFAs (hexanoic acid, decanoic acid, lauric acid, pentadecanoic acid, and heptadecanoic acid), two MUFAs (MOA and cis-10-heptadecenoic acid), one omega-6 PUFA (gamma-linolenic acid [GLA]), and one TFA (trans-10-nonadecenoic acid) were reduced in GDM (Figure 1B). In umbilical cord serum, only one MUFA (petroselinic acid) was elevated in GDM, while four SFAs (arachidic acid, heneicosanoic acid, tricosanoic acid, and tetracosanoic acid), three MUFAs (cis-10-heptadecenoic acid, cis-13-docosenoic acid, and cis-15-tetracosenoic acid), four omega-6 PUFAs (GLA, EDA, DDA, and DTA), and three TFAs (trans-10-heptadecenoic acid, trans-vaccenic acid, and trans-10-nonadecenoic acid) were reduced (Figure 1C).

Figure 1.

Z score of FA concentrations between the normal and GDM groups. (A) Z score of FA in maternal serum between normal and GDM pregnancy. (B) Z score of FA in placenta between normal and GDM pregnancy. (C) Z score of FA in umbilical cord serum between normal and GDM pregnancy. Red letters: The concentration of individual FA was higher in the GDM than normal groups, p < 0.05; green letters: The concentration of individual FA was lower in the GDM than normal groups, p < 0.05.

Profile of FAs composition in the GDM and normal groups

We analysed the composition of five FA types in maternal serum, placenta and umbilical cord serum (Figure 2). In the normal group, SFA was found to be the most abundant type of FA, with similar compositions in maternal serum and umbilical cord serum, slightly lower than that in the placenta. The composition of MUFA was the third most abundant type of FA, similar in maternal serum and umbilical cord serum, but higher compared to the placenta. The compositions of omega-3 PUFA in the placenta and umbilical cord serum were approximately 2 times higher than in maternal serum. Furthermore, omega-6 PUFA composition was found to be the second most abundant type of FA in the placenta, maternal serum and in umbilical cord serum. TFA was the least abundant type of FA, which showed gradual increasing from maternal serum to placenta and umbilical cord serum. In GDM, the composition of the five FAs in the three tissues was similar to that of the normal group. With the exception of increased total SFA and decreased total omega-6 PUFA in umbilical cord serum, there were no significant differences in any of the other types of FA between the GDM and the normal groups.

Figure 2.

The composition of the five types of FAs in the mother (maternal serum), placenta, and fetus (umbilical cord serum). (A) The composition of the five FA types in healthy pregnancy. (B) The composition of the five FA types in GDM pregnancy. FAs were categorized into five types based on their property: SFA, MUFA, omega-3 PUFA, omega-6 PUFA, and TFA. Data expressed as a percentage of the total FAs, *: p < 0.05 for FA between the mother and placenta or between the mother and fetus, or between the placenta and fetus. #: p < 0.05 for FA between the normal and GDM groups.

We then compared the composition of individual FA between the normal and GDM groups (Supplementary Table 3, Figure 3). For individual SFAs, maternal serum in GDM had lower compositions of five SFAs (arachidic acid, heneicosanoic acid, behenic acid, tricosanoic acid, and tetracosanoic acid) than normal. In the placenta, two SFAs (pentadecanoic acid and heptadecanoic acid) had lower compositions in GDM, while one SFA (tridecanoic acid) had a higher composition. In umbilical cord serum, palmitic acid (PA), an SFA with a high percentage of total SFAs, had a higher composition in GDM, while another SFA (tetracosanoic acid) had lower compositions. For individual MUFA, placental OA showed higher composition in GDM, along with cis-vaccenic acid and 11-EA. The changes in MUFAs in umbilical cord serum were more diverse, with the composition of three MUFAs (cis-10-heptadecenoic acid, cis-vaccenic acid, and cis-13-docosenoic acid) showing lower levels, while petroselinic acid showed higher composition in GDM. There were no significant differences in the compositions of omega-3 PUFAs in the maternal serum, placenta, and umbilical cord serum between normal and GDM. For omega-6 PUFA, only two omega-6 PUFAs in umbilical cord serum showed significantly lower compositions, including GLA and DTA. Despite constituting a very small percentage of total FA, individual TFA exhibited some interesting changes. Compared to normal, two TFAs (elaidic acid and trans-7-nonadecenoic acid) in maternal serum and one TFA (trans-10-nonadecenoic acid) in placenta and cord serum showed a lower composition in GDM.

Figure 3.

Z score of FA compositions between the normal and GDM groups. (A) Z score of FA in maternal serum between normal and GDM pregnancy. (B) Z score of FA in placenta between normal and GDM pregnancy. (C) Z score of FA in umbilical cord serum between normal and GDM pregnancy. Red letters: The composition of individual FA was higher in the GDM than normal groups, p < 0.05; green letters: The composition of individual FA was lower in the GDM than normal groups, p < 0.05.

Effect of GDM on FA transport

We compared the placenta/mother ratios using FA percentages between the GDM and normal groups to assess the impact of GDM on the transport of exogenous FA from the mother to the placenta, including EFA, LC PUFA and TFA (Table 2). There was no difference in EFA between the GDM and normal. The ratios of two omega-3 LC PUFAs: cis-11,14,17-eicosatrienoic acid and cis-5,8,11,14,17-eicosapentaenoic acid; one omega-6 LC PUFA (DDA), three TFAs (palmitelaidic acid, petroselaidic acid, and linoelaidic acid), were significantly higher in the GDM group. Conversely, the ratios of trans-10-nonadecenoic acid were lower.

Table 2.

Comparison of the placenta/mother and fetus/placenta ratios of EFA, omega-3 LC PUFA, omega-6 LC PUFA and TFA compositions between GDM and normal pregnancy.

	Ratio of placenta/mother			Ratio of fetus/placenta
	Normal	GDM	P	Normal	GDM	P
EFA
Alpha-Linolenic acid (C18:3n3)	0.201 ± 0.060	0.207 ± 0.067	0.757	NA	NA	NA
Linoleic acid (C18:2n6)	0.315 ± 0.036	0.337 ± 0.052	0.138	1.050 ± 0.163	1.127 ± 0.167	0.149
Omega-3 LC PUFA
Cis-11,14,17-Eicosatrienoic acid (C20:3n3)	0.484 ± 0.101	0.648 ± 0.188	0.001	4.316 ± 1.339	3.615 ± 1.168	0.085
Cis-5,8,11,14,17-Eicosapentaenoic acid (C20:5n3)	0.338 ± 0.109	0.449 ± 0.219	0.048	4.636 ± 1.186	4.107 ± 1.007	0.137
Cis-7,10,13,16,19-Docosapentaenoic acid (C22:5n3)	3.544 ± 0.695	3.615 ± 0.574	0.727	0.426 ± 0.057	0.457 ± 0.104	0.246
Docosahexaenoic acid (C22:6n3)	3.068 ± 0.729	2.992 ± 0.483	0.700	0.706 ± 0.147	0.718 ± 0.113	0.775
Total omega-3 LC PUFA	2.880 ± 0.655	2.863 ± 0.427	0.924	0.705 ± 0.130	0.717 ± 0.112	0.762
Omega-6 LC PUFA
Cis-11,14-Eicosadienoic acid (C20:2)	1.559 ± 0.166	1.605 ± 0.263	0.515	1.205 ± 0.204	1.090 ± 0.149	0.05
Cis-8,11,14-Eicosatrienoic acid (C20:3n6)	3.877 ± 0.588	4.173 ± 0.975	0.252	0.614 ± 0.120	0.604 ± 0.096	0.759
Arachidonic acid (C20:4n6)	4.843 ± 1.096	5.146 ± 1.268	0.424	0.532 ± 0.051	0.507 ± 0.062	0.172
Cis-13,16-Docosadienoic acid (C22:2)	1.187 ± 0.338	1.436 ± 0.424	0.047	4.951 ± 1.627	4.031 ± 0.865	0.031
Cis-7,10,13,16-Docosatetraenoic acid (C22:4)	7.720 ± 2.109	6.881 ± 1.665	0.171	0.778 ± 0.297	0.494 ± 0.159	<0.001
Cis-4,7,10,13,16-Docosapentaenoic acid (C22:5n6)	3.258 ± 0.615	3.308 ± 0.662	0.805	0.927 ± 0.192	0.884 ± 0.149	0.427
Total omega-6 LC PUFA	4.519 ± 0.817	4.765 ± 0.986	0.395	0.590 ± 0.035	0.545 ± 0.050	0.002
TFA
Myristelaidic acid (C14:1t)	4.460 ± 3.690	4.594 ± 2.000	0.888	1.790 ± 0.596	1.606 ± 0.327	0.234
Trans-10-Pentadecenoic acid (C15:1t)	2.433 ± 0.714	2.735 ± 1.266	0.358	2.609 ± 0.663	2.670 ± 0.584	0.762
Palmitelaidic acid (C16:1t)	2.272 ± 0.692	2.882 ± 0.915	0.023	2.392 ± 0.689	2.088 ± 0.418	0.1
Trans-10-Heptadecenoic acid (C17:1t)	3.070 ± 1.020	2.981 ± 0.608	0.739	1.841 ± 0.518	1.718 ± 0.354	0.386
Elaidic acid (C18:1n9t)	1.760 ± 0.489	2.003 ± 0.392	0.091	2.455 ± 0.663	2.335 ± 0.476	0.515
Petroselaidic acid (C18:1n12t)	1.746 ± 0.439	2.177 ± 0.445	0.004	2.595 ± 0.604	2.323 ± 0.479	0.122
Trans-Vaccenic acid (C18:1n7t)	2.137 ± 0.670	2.611 ± 0.862	0.06	1.880 ± 0.638	1.630 ± 0.343	0.131
Trans-7-Nonadecenoic acid (C19:1n12t)	NA	NA	NA	NA	NA	NA
Trans-10-Nonadecenoic acid (C19:1n9t)	0.799 ± 0.260	0.537 ± 0.143	<0.001	3.901 ± 1.459	4.700 ± 1.209	0.067
Trans-11- Eicosenoic acid (C20:1t)	2.378 ± 0.586	2.658 ± 0.844	0.231	2.356 ± 0.504	2.213 ± 0.478	0.362
Brassidic acid (C22:1n9t)	2.309 ± 0.635	2.762 ± 0.944	0.083	2.704 ± 0.675	2.366 ± 0.559	0.093
Linoelaidic acid (C18:2n6t)	1.423 ± 0.380	1.770 ± 0.467	0.014	3.575 ± 1.018	3.126 ± 0.636	0.103
Total TFA	2.354 ± 0.838	2.620 ± 0.685	0.280	2.263 ± 0.552	2.129 ± 0.425	0.395

Data were showed as mean ± SD. NA: Not available, indicates that the ratio cannot be calculated.

Blood circulating in the umbilical vein is derived from placental capillaries, so its composition depends on placental transfer activity. To analyse whether exogenous FA in the placenta supply to the fetus is affected by GDM, we compared the fetus (umbilical cord serum)/placenta ratios of individual FA composition between the GDM and normal groups to assess the transfer ability of the placenta. Only total omega-6 LC PUFA were altered in the GDM group compared to the normal group, with two individual omega-6 LC PUFA (DDA and DTA) significantly reduced.

Discussion

This is the first study utilizing samples from mothers, placenta, and fetuses from the same cohort to evaluate the transport characteristics of FA profiles in GDM. Our results showed that both the concentration and composition of FA profiles of the mother, placenta and fetus were affected by GDM. GDM disrupts placental transport of omega-3, omega-6 LC PUFAs and TFAs, with an elevated placental uptake ratio of these FAs, but not a concomitant elevation in the placental transfer ratios; or with decreased placental transfer ratio while the placental uptake ratio remained unchanged.

FAs concentrations are generally expressed in two formats: absolute and relative [21]. We first analysed the absolute concentrations of FA profiles and observed that many FAs exhibited significant differences between the GDM and normal groups. In mothers with GDM, almost all FAs increased, including some high-concentration (>100 μg/mL) and moderate-concentration FAs (10–100 μg/mL), such as PA, OA, petroselinic acid and cis-vaccenic acid. These results were consistent with the reports by Chen and Stirm [7,22], but contrary to the report by Ortega-Senovilla and Zhao [5,23]. This contradiction may be due to the fact that the concentration of maternal circulating FA is affected by many factors, such as genetics and diet [11]. In addition, it is reported that maternal BMI had an effect on maternal [24] and placental FA levels [18,25]. To mitigate the effect of BMI in our study, pregnant women in the normal and GDM groups had similar maternal BMI. This similarity may explain why our results differ from other studies. Many FAs are reduced in the placenta and fetus in GDM, which mainly occur in low (< 10 μg/g or <10 μg/mL) to moderate (10–100 μg/g or μg/mL) concentrations of FAs, differs from the results in the mother.

The composition of FA is considered to be a measure of relative concentration [21]. For normal pregnant women, the composition of SFA were highest in both the mother, placenta and fetus, which is consistent with most reports [5]. In contrast, omega-3 PUFAs in placenta and fetus were more than twice as high as in the mother, which indicates the predominance of omega-3 PUFAs in placental transport since they originate from mother [26]. With regard to omega-6 PUFA, the predominant maternal omega-6 PUFA is linoleic acid (LA), whereas the predominant omega-6 PUFA in the placenta and fetus is a series of LA metabolites. Maternal TFA composition was lower than previously reported [27–29], which may be largely attributed to the inclusion of different types of TFA in each study or to varying dietary habits between countries. But fetal TFA compositions were similar to those reported in some literature [27,30]. Studies have demonstrated that TFA can be transferred across the placenta from mother to fetus and may also accumulate within placental tissue [31–33]. In this study, TFA accounting for the lowest proportion of total FAs. However, there was a typical feature of TFA transport with an ascending trend from mother to placenta to fetus, confirming the progressive preferential transport of TFA in the placenta and fetus. This is of interest because TFA are commonly thought to have adverse effects on the fetus [32,34]. These results seem to suggest controlling the intake of TFA for pregnant women.

There were also many differences in FA composition between the GDM and normal groups. However, most differences were concentrated in FA with low compositions, except for fetal SFA and omega-6 PUFA. These findings cannot be directly compared to individual studies because the composition was considered as a relative concentration, which was linked to the total FA, and the fact that different studies encompass different types of FA. Previous studies have shown that the expression format of FA can lead to inconsistent results [35], which were also observed in this study. Pregnant women with GDM have higher triglyceride levels than women with normal glucose tolerance [36], which can result in higher total FA concentrations than normal pregnant women [37,38]. Therefore, it is not easy to determine whether the change in concentrations of GDM is due to a specific increase in individual FA or to the overall increase in FA concentration caused by hyperlipidemia [21]. In addition, it’s worth noting that all mothers with GDM accepted glycaemic management in our study, which may have improved the metabolism of FA, resulting in a change in the concentration of FA but the composition remained unchanged due to the low variability in composition [21].

Umbilical cord serum has been used to assess changes in FA in fetus with GDM mother [11]. Umbilical arterial blood reaches placental capillaries from fetal tissue and can assess fetal metabolic changes of FA [39]. In contrast, umbilical vein blood is derived from placental capillaries [39], and may reflect the extent to which maternal FAs are absorbed by the placenta and transferred to the fetus. In this study, we tried to assess the changes in placental FA transport in GDM using umbilical vein blood. SFA and MUFA were not included because the placenta and fetus synthesise their own SFA and MUFA [7]. In contrast, placental and fetal omega-3, omega-6 PUFAs, and TFAs are supplied by the mother via placental transport [16,40], therefore, only these FAs were included in the subsequent analyses. Since composition of FA can standardize the FA in different tissues and often helps to better explain metabolic associations [41], we calculated the placenta/mother ratios of FA compositions to assessed the differences in FA placental uptake between the normal and GDM groups, and compared the fetus/placenta ratios to evaluated the difference in FA placental transfer between the two groups. We found that GDM was associated with an elevated placental uptake ratio, but not a concomitant elevation in placental transfer ratio. In some other FAs, placental transfer ratio decreased while placental uptake ratio remained unchanged. Compared to placental absorption, we speculate that GDM has a greater effect on placental FA transport. Consistent with our speculation, Godhamgaonkar et al. suggested that lower fetal LC PUFA and higher placental LC PUFA in GDM indicate an increased placental sequestration, which may result in altered transport of LC PUFA from placenta to fetus [26]. The peak increase in maternal FA and the maximum transfer of FA to the fetus occur in the third trimester of pregnancy. The heightened FA metabolism is essential for improving the availability of specific important FA for fetus. Impaired transfer of omega-3 and omega-6 LC PUFAs can impact the development of fetal brain. However, when PUFA supplementation was administered to mothers with GDM, it only increased PUFA levels in the maternal blood, but not in the cord blood [42,43]. Based on our findings, we hypothesized that primary cause of the decrease in omega-3 and omega-6 LC PUFAs in GDM fetuses may not be inadequate maternal intake, but rather impaired placental transport. Correcting impaired placental transport caused by GDM may be more important than supplementing the mother with FA. In addition, we confirmed that a series of TFAs increased from GDM mother into placenta. Since we have not searched for observational or experimental studies on the effects of TFA on placenta in GDM and the percentage of TFA of total FA was indeed very low, it is not clear whether this increase would impair placental function.

Limitations and perspectives

There were some limitations in our study. Although significant differences were observed among the FA profiles of three tissue types in GDM and normal, we cannot rule out the possibility that some actual differences might still be underestimated for statistical reasons, given relatively small sample size in this study. Further analysis should expand the study size to improve statistical robustness and ensure our findings. This study was unable to determine the changes in placental and fetal metabolism under pathological conditions such as premature birth and oxygen deficiency caused by GDM, because the participants included only pregnant women who had given birth at full term.

Furthermore, our study lacks dietary data, which precludes comprehensive assessment of how varying nutritional patterns influence circulating FAs between the two groups. Our study to be a preliminary exploration and future investigations with large sample size should integrate standardized dietary assessment tools, such as validated food frequency questionnaires or 24-hour dietary recalls, to control for the influence of short- and long-term dietary habits on circulating FA levels. It is worth noting that accumulating evidence indicates circulating FA showed limited correlation with dietary intake [44], suggesting a significant impact of other factors such as absorption, endogenous synthesis and tissue metabolism on circulating FA levels [44]. And our study had rigorous inclusion/exclusion criteria that excluded cases with preexisting metabolic disorders. All enrolled GDM participants received SMBG management and data reflecting long-term metabolic status and energy balance (e.g. late pregnancy BMI, weight gain during pregnancy, neonatal birth weight and height) did not differ between the two groups, which indirectly mitigated potential dietary confounding effects. In addition, the use of ratios to assess placental transport function can control the interference of maternal dietary fluctuations. For example, even if maternal circulating FA is elevated by a short-term high-fat diet, the ratio can still reliably reflect placental transport efficacy. Therefore, we believe the differences in FA profiles between the two groups are surely related to GDM, and are less influenced by the maternal diet. Notably, the inherent limitations of our cross-sectional study impede a comprehensive understanding of dynamic alterations in FA transport during pregnancy. We recommend that future studies implement longitudinal designs to provide insights into how GDM affect FA transport under different hormonal changes throughout pregnancy.

Our study’s unique design allowed us to collect maternal serum, placental tissue and fetal serum samples from the same subject. This approach is essential for evaluating the placental transport of FA and understanding the effects of GDM on FA metabolism from mother to placenta to fetus. The main idea of this study is that GDM affects the placental FA transport function. FA transport involves multiple molecular mechanisms, including FA binding proteins, transporters, enzyme systems, etc [40]. Understanding these mechanisms helps to develop targeted therapeutic strategies for GDM. For example, optimizing placental FA transport in GDM by regulating the expression of specific transporters (e.g. fatty acid transport protein) can enhance placental FA transport efficiency. Moreover, combining maternal dietary supplements with targeted therapies for placental FA transporters can achieve a synergistic therapeutic effect.

Conclusions

In conclusion, our results indicated that concentration and composition of FA profiles in mother, placenta and fetus were affected by GDM despite glycemic management. Future studies could discuss whether different degrees of glycemic control might influence the extent of FA transport disruptions. We found GDM disrupts placental transport of omega-3, omega-6 LC PUFA and TFA, which suggests that to improve FA deficiency in GDM fetus, we should focus more on placental dysfunction rather than simply supplementing mothers with more FAs. FA transport involves multiple molecular mechanisms, studying the molecular mechanisms of FA transport helps to develop targeted therapies for GDM, such as optimizing placental function through drug interventions, dietary adjustments, or lifestyle changes. Moreover, we must also take the maternal nutritional status into account and achieve a synergistic therapeutic effect by combining maternal dietary supplementation with therapies targeting placental FA transporters. Notably, there are inherent limitations in our cross-sectional study, and future longitudinal studies should be conducted to examine the temporal effects of GDM on FA transport.

Funding Statement

This work was supported by the Grant Project of Changzhou Health Commission (GN2023036), Applied and Basic Research Project of Changzhou Science and Technology Institution (CJ20239027) and Changzhou Key Laboratory of Maternal and Child Health Medicine (Grant No. CM20240013).

Disclosure statement

There are no conflicts of interest to declare.

Data availability statement

The original data of this study will be made available by contacting the corresponding author.