Introduction

Maternal gestational diabetes mellitus (GDM) is a known risk factor for adverse long-term health outcomes in offspring, including an increased risk of obesity and other metabolic disorders [1], and the induction of neurodevelopmental disorders [2, 3]. In addition to an increased risk of obesity, diabetes, and metabolic disease in offspring [4], a growing body of research suggests that GDM is associated with neurodevelopmental problems in offspring [5,6,7,8]. Neurodevelopmental problems in offspring may be linked to impaired maternal-fetal fatty acid metabolism [9]. The sex-specific effects of maternal GDM on offspring have also been reported [10, 11].

Long-chain polyunsaturated fatty acids (LC-PUFA), particularly docosahexaenoic acid (DHA, C22:6 n-3), play a vital role in neurogenesis processes during early life. GDM has been reported to disrupt fatty acid transport across the placenta and to reduce DHA levels in umbilical cord blood [5], which could lead to impaired neurodevelopment. Recently, the major facilitator super family domain containing 2a (MFSD2a) was found to be a transporter of lysophospholipid fatty acids that help the brain take up DHA [12]. MFSD2a knockout mice have significantly reduced levels of DHA in the brain, accompanied by neuronal cell loss, neurological and behavioral deficits, and reduced brain size [13]. The study found that Wnt3/β-catenin signaling mediates MFSD2a to suppress vascular endothelial transcytosis and maintain the blood-retinal barrier in the mouse retina [14], therefore MFSD2a could be the target protein of the Wnt3/β-catenin signaling pathway in the retina.

In previous studies, researchers found a reduction in maternal-fetal transfer of LC-PUFA in GDM [12, 15]. However, it remains to be seen whether the adverse brain fatty acid profile induced by GDM during pregnancy will persist in the future and whether the offspring’s diet could influence the brain fatty acid profile. In addition, a few studies have found a sex-specific effect on offspring exposed to maternal GDM and obesity [16, 17]. The aim of the present study was to investigate the influence of a HF diet on the brain fatty acid profile of weaning and adult offspring exposed to maternal GDM, and the possible sex differences. Whether the function of the Wnt3/β-catenin/MFSD2a signaling pathway of the blood-brain barrier is affected by maternal GDM and HF.

Methods

Animals and diet

Forty-eight female C57BL/6J mice (6–8 weeks old) were obtained from Skbex Biotechnology (Henan, China) and maintained under specific pathogen-free (SPF) housing conditions (12 h light/dark cycle and controlled room temperature of 22 ± 2 °C). Mice accepted an adaptive diet for one week, and then they were randomly divided into 4 groups (n = 12/group), two groups accessing a basal diet (10% kcal from fat H10010; Beijing HFK Bioscience CO., LTD, Beijing, China) and other two groups accessing a HF diet (45% kcal from fat H10045; Beijing HFK Bioscience CO., LTD, Beijing, China) for four weeks before mating. The HF and HF + GDM groups received the HF diet all the time, and their offspring continued to receive the HF diet after weaning. Feed formulation is shown in Supplementary Table S1, and fatty acid content and proportion of all experimental diets are calculated according to China Food Composition (Table 1). All procedures performed in this study were approved by the Ethics Committee of the School of Public Health, Jilin University (2021-04-05).

Table 1 Fatty acid content and proportion of the experimental diets.
Full size table

Female and male mice were mated in a 2:1 ratio during the female estrus phase. Finally, successfully pregnant mice were included in the study. Successful mating was confirmed by vaginal plug formation and daily body weight measurements for the following 7 days. Animals with plugs were recorded as gestation day (GD) 1. Confirmed pregnant females were maintained on their assigned diets throughout gestation. Body weight, food, and water consumption were evaluated throughout the gestation period.

Treatment

Insulin receptor antagonist (S961) (Synpeptide Biotechnology Co., Ltd, Nanjing, China) dissolved in PBS, and 20nmol/kg/day or an equal volume of PBS was injected subcutaneously daily at the same time of day (5 pm) until delivery to establish the GDM model, as described previously [18, 19]. Two groups of mice received basal diet before and during pregnancy, one group received basal diet and PBS as the control group (CON) and another group received basal diet and S961 as the GDM group. Two groups of mice received a HF diet before and during pregnancy, one group received a HF diet and PBS as a HF group and another group received a HF diet and S961 as a HF and GDM group (HF + GDM). The offspring mice received the same diet as their mothers. And one male and one female mice from each litter were randomly selected and sacrificed at week 4 (weaning period) and week 8 (adult period), approximately 6 pups per sex in each group. If the number of litters was large (more than two males and two females), more than two offspring from that litter were randomly selected and sacrificed at week 4 or 8.

Fasting glucose and oral glucose tolerance test

Baseline fasting glucose was assessed after adaptive feeding and fasting glucose was assessed after high fat and basal chow feeding for 4 weeks. Fasting glucose was assessed on GD 7. And an oral glucose tolerance test (OGTT) was performed on GD 14, as previously described [18, 19]. Pregnant mice fasted for 5 h, while non-pregnant female mice fasted for 12 h prior to glucose measurements. On GD 14, fasting tail vein glucose was measured in pregnant mice, then pregnant mice received an intragastric administration of glucose (1.2 g/kg) and tail vein glucose measurements were performed at 30, 60, 120, 180 min using an Abbott Optium Xceed blood glucose meter.

Sample collection and fatty acids analysis

Offspring mice were sacrificed at weeks 4 and 8 using isoflurane. Fetal brain tissue samples were collected. Tissue samples were snap frozen in liquid nitrogen and stored at −80 °C until further analysis. Fifty mg of brain tissue was homogenized in 300 μL saline, and then 1.2 mL chloroform: methanol (2:1, v-v) solvent was mixed for total lipid extraction at 4 °C overnight. The mixture was centrifuged at 900 g for 5 min, the substrate liquid was transferred to a new centrifuge tube, and 50 μL heptadecanoic acid (C17:0) internal standard solution and 100 μL sodium methoxide solution were mixed for 5 min at room temperature. Three hundred microlitres of methanol hydrochloride solution was added to complete the reaction. Six hundred microlitres of N-hexane was added and the mixed solution was mixed for 30 s to extract fatty acid methyl esters (FAMEs), the upper N-hexane phase was transferred to a new glass centrifuge tube. The N-hexane was nitrogen-blown to dryness and 50 μL n-hexane containing 2 g/L butylated hydroxytoluene (BHT) was added to dissolve the residue for gas chromatographic analysis. Brain fatty acid results were expressed as a percentage of total fatty acid.

Western blotting

Brain tissue samples were lysed in ice-cold RIPA lysis buffer containing 1 mM protease inhibitor PMSF (Beyotime Biotechnology, China) for 30 min on ice. The tissue homogenate was centrifuged at 12,000 g for 5 min at 4 °C. Protein concentrations were measured using a BCA assay kit. Equal amounts of protein (50 μg) were then separated by SDS-polyacrylamide gel electrophoresis and transferred to PVDF membranes. After blocking with 5% skimmed milk for 2 h at room temperature, the membranes were incubated with primary antibodies overnight at 4 °C, washed with TBST and incubated with horseradish peroxidase-conjugated secondary antibodies for 45 min at room temperature. The membranes were then washed again with TBST and the blots were visualized using an enhanced chemiluminescence reagent. Primary antibodies included Wnt3 (1:3000, Abcam, ab172612), β-catenin (1:5000, Proteintech, 66379-1-Ig), MFSD2a (1:1000, Thermo Fisher Scientific, PA5-21049). The intensity of each band was quantified using Image J.

Statistical analysis

Data were expressed as mean ± SEM or median (interquartile range, IQR). Data from the animal model, OGTT at each time point and area under the curve (AUC) of OGTT were analyzed using one-way ANOVA followed by LSD post hoc test for normally distributed data and Kruskal-Wallis test for abnormally distributed data. The AUC of the OGTT were calculated using GraphPad Prism 9.0. The levels of brain fatty acids and protein expressions were analyzed by factorial analysis to determine the main effects and interactions between GDM, HF diet and sex. The HF diet groups (HF and HF + GDM groups) were compared with the basal diet groups (CON and GDM groups). The gestational diabetes mellitus groups (GDM and HF + GDM groups) were compared with the non-gestational diabetes mellitus groups (CON and HF groups). The male offspring were compared with female offspring. The individual effects of three factors (HF diet, GDM and gender) were analyzed using BONC DSS 25.0. Then the statistical difference between the four groups (CON, GDM, HF and HF + GDM) was then analyzed using one-way ANOVA followed by LSD post hoc test for normally distributed data and Kruskal-Wallis test for abnormally distributed data. Two-sided test results were considered statistically significant if P < 0.05. The figures were plotted using GraphPad Prism 9.0.

Results

Establishment of maternal GDM animal model

Female mice were fed either a basal diet (CON and GDM groups) or a HF diet (HF and HF + GDM groups) for four weeks. After mating, pregnant mice were included in follow-up experiment (nCON = 10, nGDM = 11, nHF = 10, nHF+GDM = 9). Body weights of female mice in the HF and HF + GDM groups were higher than in the CON and GDM groups (P < 0.01) (Fig. 1A). Gestational weight gain was higher in the HF group than in the GDM group (P = 0.039) (Fig. 1B).

Fig. 1: Characteristics of the animal model (mean ± SEM).
figure 1

A The pre-pregnancy weight of female mice (nCON = 10, nGDM = 11, nHF = 10, nHF+GDM = 9); B Gestational weight gain of pregnant mice (n = 9/group); C Area under the curve (AUC) of OGTT (n = 9/group); *P < 0.05, **P < 0.01; D OGTT of pregnant mice on GD 14, P < 0.05 $ GDM vs. CON, # HF + GDM vs. CON, £ GDM vs. HF, ￥HF + GDM vs. HF.

Full size image

Before HF diet feeding, there was no difference among the four groups in the baseline fasting glucose of female mice. After four weeks of feeding HF and a basal diet, there was no difference between the four groups, and fasting glucose was not different at GD7 (Table S2).

After seven days of S961 or PBS treatment (GD 14), pregnant mice were subjected to an OGTT, and S961-treated pregnant mice with abnormal glucose tolerance and maternal mice that gave birth to live pups were included in the study. Finally, 36 mother mice were included in this study (n = 9/group). The AUC of the OGTT was higher in the GDM and HF + GDM groups (P < 0.001), with the HF + GDM group having a higher AUC than the GDM group (P = 0.043) (Fig. 1C). The OGTT showed that the S961-treatment mice displayed significantly higher glucose at 30 min, 60 min, 120 min and 180 min (P < 0.05), reproducing the main features of gestational diabetes, but the glucose intolerance was not found in HF group (Fig. 1D).

The effect of HF on the growth of offspring exposed to maternal GDM

Birth weights of live offspring were weighed on the first day of birth. The birth weight was lower in the GDM and HF + GDM groups than in the CON group (P = 0.019, P = 0.024) (Fig. 2A). In contrast to birth weight, offspring body weight from week 1 to week 4 was higher in the HF and HF + GDM groups than in the CON group (P < 0.01). In addition, the body weight of the GDM group was significantly higher than the CON group at week 4 (P < 0.001), and the HF + GDM group was significantly higher than the CON, GDM, HF groups from week 1 to week 4 (P < 0.05) (Fig. 2B).

Fig. 2: The effect of HF on the growth of offspring exposed to maternal GDM (mean ± SEM).
figure 2

A The birth weight of pups (nCON = 42, nGDM = 49, nHF = 52, nHF+GDM = 49), *P < 0.05; B The body weight of offspring from week 1 to week 4 (Week 1: nCON = 36, nGDM = 37, nHF = 44, nHF+GDM = 44; Week 2: nCON = 36, nGDM = 34, nHF = 44, nHF+GDM = 44; Week 3: nCON = 34, nGDM = 32, nHF = 41, nHF+GDM = 42; Week 4: nCON = 22, nGDM = 18, nHF = 17, nHF+GDM = 16), *P < 0.05; C The body weight of male offspring from week 4 to week 8 (nCON = 7, nGDM = 7, nHF = 8, nHF+GDM = 7); D The body weight of female offspring from week 4 to week 8 (nCON = 8, nGDM = 7, nHF = 6, nHF+GDM = 7); P < 0.05 $ GDM vs. CON, * HF vs. CON, # HF+GDM vs. CON, £ HF vs. GDM, ￥HF vs. HF + GDM, ^GDM vs. HF + GDM.

Full size image

Male offspring in the HF and HF + GDM groups had significantly higher body weights than the CON group at weeks 4, 5, 6, 7, 8, and the HF + GDM group had higher body weights than the GDM group at weeks 4, 5, 6, 7, 8 (P < 0.01). Maternal GDM also increased the body weight of male offspring. The body weight of the offspring in the GDM group was significantly higher than that of the CON group at weeks 4 and 8 (P < 0.05), and the tendency to increase was also seen in the offspring of the HF + GDM group compared with the HF group at week 8 (Fig. 2C).

The body weight of the female offspring in the HF and HF + GDM groups was higher than that of the CON and GDM groups from week 4 to week 7 (P < 0.01). However, the body weight gain of the female offspring in the HF and HF + GDM groups tended to be slow at week 8. The HF + GDM female offspring had the higher body weight than the CON group (P < 0.05), but there was no difference between the other groups at week 8 (Fig. 2D).

The effect of HF on the brain fatty acid profile of weaning male offspring exposed to maternal GDM

The brain fatty acid profile of male and female offspring at weaning is shown in Table 2. Male offspring in the GDM, HF and HF + GDM groups had lower C14:0 than those in the CON group, and the HF + GDM group had the lower C14:0 than those in GDM and HF groups (P < 0.05). Male offspring in the GDM and HF groups had a significantly higher brain C18:0 compared with the CON group (P = 0.016, P = 0.017). The level of C20:0 in the brain of weaning male offspring was higher in the HF + GDM group than in the HF group (P = 0.043). At weaning, the male offspring in the HF group had significantly higher levels of total saturated fatty acids (SFA) in the brain than those in the CON and HF + GDM groups (P = 0.001, P = 0.016).

Table 2 The effect of HF diet on the fatty acid profile of weaning offspring brain exposed to maternal GDM.
Full size table

The level of C18:1 in the brain of the male offspring at weaning was significantly higher in the GDM and HF + GDM groups than in the CON group (P = 0.030, P = 0.022). The level of C20:1 in the brain of the male offspring was higher in the HF + GDM group than in the CON and HF groups (P < 0.05). The level of total monounsaturated fatty acids (MUFA) was higher in the HF + GDM group than in the CON group (P = 0.022), and the trend was for an increase in the HF + GDM group compared with the HF group (P = 0.055).

The HF and HF + GDM groups had higher levels of linoleic acid (C18:2n-6, LA) and C20:2n-6 in the brain compared with the CON and GDM groups (P < 0.01), and the increased trend of C20:2n-6 can also be observed in the HF + GDM group than in the GDM group (P = 0.055). The GDM and HF + GDM groups had lower levels of arachidonic acid (C20:4n-6, AA) and total n-6 PUFA than the CON group, and the levels of AA and total n-6 PUFA were lower in the HF + GDM group compared with the HF group (P < 0.01).

The brains of male offspring in the HF and HF + GDM groups had lower levels of EPA than those in the CON group (P < 0.001). The HF + GDM group had significantly lower DHA and total n-3 PUFA than the CON group (P = 0.033, P = 0.030), and a decreasing trend of DHA and total n-3 PUFA can be seen in the GDM group compared to the CON group (P = 0.099, P = 0.095).

The effect of HF on the brain fatty acid profile of weaning female offspring exposed to maternal GDM

The fatty acid profile of the brain of the weaning female offspring is shown in Table 2. Female offspring in the HF + GDM group had significantly higher brain C18:0 than those in the CON, GDM and HF groups (P < 0.01).

The increased trend was observed in the level of C18:1 in the GDM and HF + GDM groups than in the CON group (P = 0.073, P = 0.065). The level of C20:1 in the brains of female weaning offspring was higher in the GDM group than in the CON group, and the increased trend was also observed in the HF + GDM group compared with the CON group (P = 0.022, P = 0.065). MUFA levels were higher in the GDM and HF + GDM groups than in the CON group (P = 0.035, P = 0.002).

Levels of LA and C20:2n-6 were higher in the HF and HF + GDM groups than in the CON and GDM groups (P < 0.05), and the HF + GDM group had significantly higher LA than the HF group (P = 0.041). The female offspring of the GDM group had significantly lower AA than the CON group (P = 0.022), and the decreasing trend was also observed in the HF + GDM group compared with the CON group (P = 0.065). The female offspring of the HF group had significantly higher total n-6 PUFA than the GDM and HF + GDM groups (P = 0.026, P = 0.014).

The HF + GDM group had significantly lower brain EPA, DHA and total n-3 PUFA than the CON, GDM, and HF groups (P < 0.001), and the HF group had the lower EPA than that of the CON and GDM groups (P < 0.05).

The effect of GDM, HF and offspring sex on brain fatty acid profile of weaning offspring

The results of the factorial analysis of GDM, HF and offspring sex on the brain fatty acid profile are shown in Table 2. Maternal GDM significantly increased C18:0, C20:0, C18:1, C20:1 and total MUFA in weaning offspring (P < 0.05). Maternal GDM significantly decreased AA, total n-6 PUFA, EPA, DHA and total n-3 PUFA in weaning offspring (P < 0.01). The HF diet significantly increased C18:0, SFA, LA and C20:2n-6 (P < 0.05) and decreased EPA, DHA and total n-3 PUFA (P < 0.01). In addition, the brains of the female offspring had lower levels of DHA and total n-3 PUFA than those of the male offspring (P < 0.01).

The effect of HF on the brain Wnt3/β-catenin/MFSD2a pathway in weaning offspring exposed to maternal GDM

There was no difference in the protein expression of Wnt3 in the brain of the offspring in the male and female groups (P > 0.05) (Fig. 3A). Male weaning offspring had lower expression of β-catenin in the HF + GDM group compared to the CON and GDM groups (P = 0.045, P = 0.027). Female weaning offspring had lower β-catenin expression in the HF + GDM group compared to the CON group (P = 0.035) (Fig. 3B). Female weaning offspring had lower expression of MFSD2a in the HF and HF + GDM groups than in the CON group (P = 0.041, P = 0.018) (P > 0.05) (Fig. 3C).

Fig. 3: The effect of HF on the protein expression of Wnt3/β-catenin/MFSD2a in the brain of weaning offspring exposed to maternal GDM (mean ± SEM, n = 6/group).
figure 3

A: Wnt3; B: β-catenin; C: MFSD2a; The P value above the bar chart represents the result of the factorial analysis. *P < 0.05 represent the different among the four groups.

Full size image

Factorial analysis showed that HF significantly decreased the brain β-catenin in male weaning offspring (P = 0.015) (Fig. 3B). In addition, HF significantly decreased the brain MFSD2a in female weaning offspring (P = 0.006) (Fig. 3C).

The effect of HF on the brain fatty acid profile of male adult offspring exposed to maternal GDM

The fatty acid profile of the male adult offspring brain is shown in Table 3. Male adult offspring in the HF and HF + GDM groups had a lower brain C14:0 than those in the CON and GDM groups (P < 0.05). The HF and HF + GDM groups had significantly higher C18:0 than the CON group (P = 0.017, P = 0.007).

Table 3 The effect of HF diet on the fatty acid profile of adult offspring brain exposed to maternal GDM.
Full size table

The level of LA was significantly higher in the HF and HF + GDM groups than in the CON group (P = 0.002, P = 0.001). The HF and HF + GDM groups had higher C20:2n-6 than that of the CON and GDM groups (P < 0.05). The HF + GDM group had higher AA than in the HF group (P = 0.038). The HF + GDM group had higher n-6 PUFA than that of the CON group (P = 0.023).

The HF + GDM group had lower EPA than that of the GDM group (P = 0.026). The HF + GDM group had significantly lower DHA than the CON group (P = 0.038).

The effect of HF on the brain fatty acid profile of female adult offspring exposed to maternal GDM

The fatty acid profile of female adult offspring brain is shown in Table 3. The HF + GDM group had lower brain C16:0, SFA, and higher C20:0 than that of the CON and GDM groups (P < 0.05).

The HF + GDM group had higher levels of brain C18:1, C20:1, MUFA than the CON and GDM groups (P < 0.01). In addition, the HF group had a higher brain C18:1 than the CON group (P = 0.048).

The HF and HF + GDM groups had significantly higher LA and C20:2n-6 than the CON and GDM groups (P < 0.05). The HF + GDM group had lower AA than that of the CON group (P = 0.035).

The HF + GDM group had significantly lower DHA and n-3 PUFA than the CON and GDM groups (P < 0.05). The decreased trend of DHA and n-3 PUFA were also observed in HF group than that of CON group (P = 0.070, P = 0.063).

The effect of GDM, HF and offspring sex on the brain fatty acid profile of adult offspring

Factorial analysis showed that the HF diet decreased C14:0, C16:0, DHA and total n-3 PUFA in the brain, but increased C18:0 and C20:0, C18:1, C20:1, MUFA and LA (P < 0.05) (Table 3). In addition, the sex difference was more pronounced in the adult offspring. The brains of the female offspring had lower levels of C16:0 and total SFA (P < 0.01), but the higher levels of C14:0, C20:0 than those of the male offspring (P < 0.05). The brains of the female offspring had higher levels of C18:1, C20:1 and total MUFA than those of the male offspring (P < 0.01). The brain of the female offspring brain had higher levels of C20:2n-6, but lower levels of AA and total n-6 PUFA than that of the male offspring (P < 0.05).

The effect of HF on the brain Wnt3/β-catenin/MFSD2a pathway of adult offspring exposed to maternal GDM

There was no difference in brain Wnt3 protein expression in the offspring (P > 0.05). Female adult offspring had lower expression of β-catenin in the HF + GDM group compared to CON (P = 0.036), while no difference was observed in male offspring (P > 0.05). In addition, the expression of MFSD2a in male adult offspring was higher in the HF + GDM group than in the GDM group (P = 0.020), but the expression of MFSD2a in female adult offspring was lower in the HF + GDM group than in the CON group (P = 0.046).

Factorial analysis showed that the trend of HF decreased the brain β-catenin in female adult offspring (P = 0.076) (Fig. 4B). In addition, HF significantly decreased the brain MFSD2a in female adult offspring (P = 0.049), however, HF increased the brain MFSD2a in male adult offspring (P = 0.034) (Fig. 4C).

Fig. 4: The effect of GDM on the protein expression of the Wnt3/β-catenin/MFSD2a signaling pathway in the brain of adult male offspring (mean ± SEM, n = 4/group).
figure 4

A Wnt3; B β-catenin; C MFSD2a; The P value above the bar chart represents the result of the factorial analysis. *P < 0.05 represent the different among the four groups.

Full size image

Discussion

The prevalence of childhood obesity is increasing worldwide, and childhood obesity can affect children’s growth and development, increasing the risk of adult obesity, type 2 diabetes mellitus, and so on [20, 21]. Exposure to GDM in the womb is an important risk factor for obesity in childhood [22,23,24]. In this study, the insulin receptor antagonist (S961) and HF diet were used to establish the GDM animal model. The AUC of OGTT in the HF + GDM group was higher than that in the GDM group. Hyperinsulinemia is reported to be an underlying cause of obesity-associated insulin resistance [25, 26]. And administration of S961 can increase insulin levels in mice, therefore, when combined with a HF diet, HF + GDM may increase elevated blood glucose and exacerbate its harmful effects. A large population study found that gestational glucose intolerance was associated with increased odds of offspring being overweight/obese in late adolescence [27]. The offspring of GDM mothers had a higher BMI at 17 years of age, and even after adjusting for birth weight, the difference remained [28]. In this study, the maternal mice in the GDM and HF + GDM groups had abnormal glucose tolerance, while the maternal mice in the HF diet group had normal glucose tolerance. Therefore, maternal GDM was associated with an increase in offspring body weight during the lactation period and the offspring of HF group had lower body weight than that in HF + GDM group. And after weaning, maternal GDM increased the body weight of male offspring in both the HF + GDM and GDM groups, suggesting that maternal GDM increases the susceptibility of male offspring to obesity. Maternal GDM has been reported to have a sex-specific effect on skinfold thickness and glucose metabolism [29]. A population study has also suggested that maternal GDM increases the amount of fat in preschool boys [16]. A large number of studies have shown that maternal GDM is a risk factor for adult male obesity, but not for female offspring [11, 30]. A cohort study of 15009 individuals showed that male offspring exposed to GDM had higher BMI and risk of obesity in childhood and early adulthood, but no difference in female offspring [31], in agreement with our findings. In this study, we found the sex-specific effect of the HF diet on the growth and development of offspring exposed to GDM, with male offspring having higher body weight and susceptibility to obesity. In female offspring, the increase in body weight also be found before week 8 but tends to be slow at week 8 exposed to HF diet, which could be explained by the obesity resistance of C57BL/6 J female mice exposed to HF diet [32, 33], female mice obesity resistance could be related with the circulating ovarian hormones protect the integrity of daily rhythms during high-fat feeding [33]. However, in the HF + GDM group, the increase of body weight was also observed until week 8, but the female offspring body weight of HF group has no differ with CON group. The weight gain of the female offspring in the HF + GDM group may be caused by the mothers’ abnormal glucose tolerance during intrauterine development [34], whereas the mothers in the HF group did not have abnormal glucose tolerance during pregnancy. Therefore, the weight of the female offspring in the HF group alone tends to increase but slowdown in adulthood.

Increasing evidence links maternal GDM with autism, attention deficit hyperactivity disorder and schizophrenia in offspring [35, 36]. In addition, animal studies also found a link between maternal GDM and impaired neurodevelopment and cognitive function in offspring [37, 38], but the mechanism was uncertain. It has been reported that GDM induces a disruption in maternal fatty acid metabolism and placental fatty acid transport dysfunction, resulting in a decrease in cord blood DHA. Therefore, it is reasonable to hypothesize that maternal GDM caused the alteration of the brain fatty acid profile, especially the reduced DHA, followed by the neurodevelopmental disorder in the offspring. While umbilical cord blood DHA is reduced in fetuses exposed to GDM [12, 39], brain fatty acid cannot be measured in humans. In our previous study, we observed that maternal GDM decreased fetal brain DHA during uterine development [18]. In the present study, we investigated the effect of the HF diet on the brain fatty acid profile of weaning and adult offspring exposed to maternal GDM. Maternal GDM decreased brain AA and DHA levels and increased brain MUFA levels in the offspring during the weaning period. However, HF has no effect on the AA in the weaning offspring. These may be due to maternal GDM and HF affecting breast milk fatty acid profile [40]. Metabolomics study of GDM breast milk suggests that GDM affects the levels of amino acids and fatty acids in colostrum, transitional and mature milk [41]. Therefore, the alteration in the fatty acid profile of the brain of weaning offspring, which could be due to intrauterine impairment or alteration of breast milk constituents, should be further investigated. Moreover, it was found that there was more endogenous synthesis of AA in milk and more transfer of dietary LA to the mammary gland in low-fat fed rats compared to HF (high LA) fed rats [42]. Maternal adaptations may compensate for dietary deficiencies in PUFA, including increased endogenous AA synthesis and increased uptake of PUFA. This idea is supported by the results of another population-based study [43].

As offspring age, dietary fatty acids play a more important role in the fatty acid profile of the brain. Higher dietary SFA, MUFA, and LA increased the proportion of SFA, MUFA, and n-6 PUFA in the offspring’s brain, with a corresponding decrease in the proportion of n-3 PUFA in the brain (Approximately 99% of brain n-3PUFA is DHA, with very little other n-3 PUFA). In weaning female offspring, the HF + GDM group has lower β-catenin and MFSD2a expression than the CON group, but there is no difference in brain MFSD2a expression in male offspring. Correspondingly, the HF + GDM group has lower DHA than the CON group in both male and female offspring brain, in which gender affect the level of brain DHA, female offspring has lower DHA than male offspring. MFSD2a is an important protein for DHA uptake by the brain, so the reduced MFSD2a may be the possible reason for the reduced DHA in the brains of HF + GDM female offspring. However, we also observed reduced brain DHA, but no difference in MFSD2a expression, in the male offspring. There are other ways in which DHA in the brain can be affected. This could be due, for example, to the fact that the HF diet has a lower ALA content than the basal diet group. ALA is a synthetic precursor of n-3 PUFA, and the reduced intake of the precursor led to a decrease in the levels of DHA produced by endogenous metabolism and then affected the levels of DHA in the brain. The specific mechanism of this effect will be the subject of further investigation.

HF had sex-specific effects on offspring brain Wnt3/β-catenin/MFSD2a pathway. With the duration of HF intake (high level of SFA and MUFA), HF induced the decrease of MFSD2a in female adult offspring brain, but increase in male adult offspring brain. In addition, other PUFAs could be taken up by MFSD2a in addition to DHA [13]. Accordingly, we observed the decrease of total PUFA in female adult offspring induced by HF, and the proportion of PUFA in female adult offspring was lower than in male. Therefore, HF may have long-term effects on the brain development of female offspring.

In conclusion, maternal GDM increases susceptibility to HF diet-induced obesity in male offspring. Maternal GDM induced an adverse brain fatty acid profile in both male and female weaning offspring. In addition, the HF diet exacerbated the adverse fatty acid profile. As the offspring grew, the effect of maternal GDM on offspring brain fatty acids decreased or disappeared, but the effect of the HF diet dominated. There are sex differences in the effects of HF on offspring exposed to maternal GDM.