Maternal Obesity Programs Glucose Intolerance in Pregnant Female Offspring

Abstract

Maternal obesity is a known risk factor for metabolic dysfunction in offspring; however, its effect on metabolism during pregnancy in female offspring remains unclear. This study investigated how maternal obesity, induced by high-fat (HF) feeding in C57BL/6J mice, affects the metabolic adaptation to pregnancy in female offspring. Dams were fed an HF diet (60% fat) or chow for 3 months before and during pregnancy. Offspring of HF diet–fed dams (OF-HFD) exhibited reduced fetal growth, followed by rapid postnatal catch-up and increased adult adiposity, compared with offspring of chow-fed dams (OF-CD), despite having similar baseline glucose and insulin levels. During pregnancy, OF-HFD exhibited diminished increases in maternal body fat, blood triglycerides, and insulin concentrations, accompanied by glucose intolerance. In cultured islets, glucose-stimulated insulin secretion was markedly reduced in pregnant OF-HFD, despite unchanged β-cell mass or proliferation. Hepatic triglyceride secretion was decreased, whereas liver insulin signaling was enhanced, suggesting alterations in lipid and glucose metabolism. Feeding OF-HFD an HF diet before and during pregnancy further impaired fetal growth. These findings indicate that maternal obesity impairs the metabolic adaptation to pregnancy in female offspring, characterized by insulin insufficiency and disrupted lipid homeostasis. This may initiate a transgenerational cycle of metabolic dysfunction, potentially increasing the risk of gestational diabetes in subsequent generations. Our findings underscore the need for more research to explore these mechanisms in humans and develop strategies to reduce the long-term effects of maternal obesity.

Article Highlights

• Maternal high-fat (HF) diet induces adiposity in first-generation (F1) female offspring, impairing metabolic adaptations during pregnancy.

• F1 offspring from HF diet–fed dams show diminished fat gain and reduced serum triglycerides, disrupting nutrient availability for fetal growth.

• Impaired insulin production in F1 pregnancy leads to glucose intolerance, driven by reduced insulin secretion despite normal β-cell mass.

• Unlike male offspring, F1 females exhibit resistance to fat expansion under HF diet challenge, suggesting sex-specific programming.

• These findings underscore a transgenerational cycle of metabolic dysfunction, highlighting the need for interventions against maternal obesity.

Introduction

Fetal programming, also known as the developmental origins of health and disease, suggests that adverse conditions during early development, particularly in utero, predispose individuals to metabolic disorders in adulthood (1,2). This concept has emerged from human epidemiological studies linking poor fetal nutrition to later cardiovascular disease and has since expanded to encompass maternal overnutrition (3). Amid the global obesity epidemic, maternal obesity has become a critical programming factor, heightening the risk of obesity, insulin resistance, and cardiovascular disease in offspring (4–6). In rodent studies, maternal high-fat (HF) feeding restricts fetal growth, followed by rapid postnatal catch-up and increased adult adiposity (5,7–11). Although male offspring often exhibit pronounced insulin resistance and metabolic defects (7–9), the effects on females, particularly during pregnancy, are less well understood.

Pregnancy requires maternal metabolic adaptations, namely, increases in body fat, blood triglycerides (TGs), and insulin levels, to support the fetal demands for nutrients and growth (12,13). Obesity disrupts these adaptations in the mother; however, its effects on metabolism during pregnancy in female offspring remain unclear. Such disruptions could amplify metabolic dysfunction across generations, potentially increasing the risk of gestational diabetes mellitus (GDM), which is a growing public health concern (14). Using C57BL/6J mice fed an HF diet to induce maternal obesity, we show that maternal obesity impairs female offspring’s metabolic adaptation to pregnancy, evidenced by diminished fat gain, reduced TG and insulin levels, and glucose intolerance, suggesting a transgenerational mechanism for metabolic disease (15).

Research Design and Methods

Materials

Glucose, BSA, and DMEM were obtained from Sigma-Aldrich (St. Louis, MO). Antibodies against AKT, phosphorylated AKT (Ser473), hormone-sensitive lipase (HSL), and others were obtained from Cell Signaling, Inc. (Danvers, MA), and insulin and Ki-67 antibodies were sourced from Abcam (Cambridge, MA). HF diet (60% fat; 5.24 kcal/g; cat. no. D12492) and chow (17% fat; 3.1 kcal/g; cat. no. 7912) were obtained from Research Diets, Inc. (New Brunswick, NJ), and Harlan Laboratories (Madison, WI), respectively. TG assay kits were from Wako Diagnostics (Richmond, VA).

Experimental Animals

Eight- to 10-week-old nulliparous female C57BL/6J mice (F0) from The Jackson Laboratory (Bar Harbor, ME) were fed an HF diet or chow for 3 months before and during pregnancy (n = 10–12 per group). Chow-fed males were paired with female mice overnight in the male’s cage; pregnancy was confirmed by vaginal plug (embryonic day [E] 0.5). Fetal sex was determined by PCR using a pair of primers for X-linked Rbm31x (primers listed in Table 1) (16). Postdelivery, all mice received chow, with litter sizes standardized to six to eight. First-generation (F1) female offspring of HF diet–fed (OF-HFD) or chow-fed dams (OF-CD) were monitored for body composition via EchoMRI (Houston, TX). Some F1 females were fed an HF diet for 1 month prepregnancy to challenge latent metabolic defects without inducing full obesity, unlike the 3-month F0 HFD. To clarify group designations, we define the nomenclature for F1 offspring groups in Table 2, based on F0 dam and F1 offspring diets. Experiments were conducted in accordance with the guidelines of the University of California San Diego Animal Care and Use Committee, La Jolla, CA.

Table 1.

Sequences for real-time PCR primers

Gene	Forward (5′ to 3′)	Reverse (5′ to 3′)
β-Actin	GGGGTGTTGAAGGTCTCAAA	CTGAACCCTAAGGCCAACC
Angptl4	gggaccttaactgtgccaag	GAATGGCTACAGGTACCAAACC
C/ebpa	TCACTGGTCAACTCCAGCAC	TGGACAAGAACAGCAACGAG
Adrb3	TCC CGA AGA AGG GAA CTG T	CCT TCC GTC GTC TTC TGT GT
Scd1	GTGGGCAGGATGAAGCAC	AGCTGGTGATGTTCCAGAGG
Apob	TTCTTCTCTGGAGGGGACTG	GGCACTGTGGGTCTGGAT
G6pase	TCT GTC CCG GAT CTA CCT TG	GAA AGT TTC AGC CAC AGC AA
Srebp1c	CTGTCTCACCCCCAGCATAG	GAACTGGACACAGCGGTTTT
Pparγ1	CTGTGTCAACCATGGTAATTTCTT	TGCTGTTATGGGTGAAACTCTG
Pparγ2	ATGCACTGCCTATGAGCACT	CAACTGTGGTAAAGGGCTTG
Fasn	ACTCCACAGGTGGGAACAAG	CCCTTGATGAAGAGGGATCA
Gpat3	GTG CTG GGT GTC CTA GTG C	AAG CTG ATC CCA ATG AAA GC
LPL	GAAGTCTGACCAATAAGAAGGTCAA	TGTGTGTAAGACATCTACAAAATCAGC
CD36	TTGAAAAGTCTCGGACATTGAG	TCAGATCCGAACACAGCGTA
Mtp	TGAGAGGCCAGTTGTGTGAC	GGCAGTGCTTTTTCTCTGCT
Pepck	TCGATGCCTTCCCAGTAAAC	CTGGCACCTCAGTGAAGACA
Hmgcr	TTG TAG CCT CAC AGT CCT TGG	CGT AAG CGC AGT TCC TTC C
Mogat1	CCAGCATCACCATAATTCCA	TGCAGTGGGTCCTGTCCT
Lxr-a	TGG AGC CCT GGA CAT TAC C	TGT GCG CTC AGC TCT TGT
Rbm31x/y	CACCTTAAGAACAAGCCAATACA	GGCTTGTCCTGAAAACATTTGG

Table 2.

Experimental group nomenclature

Group abbreviation	F0 dam diet	F1 offspring diet	Description
OF-CD (black)	Chow	Chow	Offspring of chow-fed dams, maintained on chow diet prepregnancy and during pregnancy
OF-HFD (orange)	HF diet	Chow	Offspring of HF diet–fed dams, maintained on chow diet prepregnancy and during pregnancy
HF-CD (cyan)	Chow	HF diet	Offspring of chow-fed dams, fed HF diet for 1 month prepregnancy and during pregnancy
HF-HFD (purple)	HF diet	HF diet	Offspring of HF diet–fed dams, fed HF diet for 1 month prepregnancy and during pregnancy

Glucose Tolerance and Metabolic Assays

Glucose tolerance tests (GTTs) were performed before and during pregnancy (E12.5) after overnight fasting with intraperitoneal glucose injection (2 g/kg). Hepatic TG secretion was measured at E16.5 using Poloxamer-407 (1,000 mg/kg; BASF, Mount Olive, NJ) injection, as described previously (17). Postheparin lipoprotein lipase (LPL) activity was assessed 10 min after heparin injection (0.2 units/g) using a fluorometric assay.

Tissue Analysis

Placentas, pancreases, and livers were collected at E18.5 and, in the fed state, were fixed and paraffin or optimal cutting temperature embedded. Immunohistochemistry and immunofluorescence were performed using antibodies against insulin, glucagon, and Ki-67. β-Cell mass was calculated from insulin-stained sections (12). Pancreatic islets were isolated via collagenase digestion, cultured overnight, and assayed for glucose-stimulated insulin secretion (GSIS) with 2 and 20 mmol/L glucose (18).

Molecular Assays

Tissue samples for protein and mRNA extraction were collected at E18.5 in the fed state, with no insulin stimulation unless specified. Proteins were extracted, separated by NuPAGE gels, and blotted with the indicated antibodies. RNA was isolated with TRIzol, reverse transcribed, and analyzed by real-time PCR (QuantStudio 3; Invitrogen, Carlsbad, CA) using primers (Table 1), normalized to β-actin.

Statistical Analysis

Data are reported as mean ± SEM. Comparisons used the Student t test or ANOVA with Bonferroni posttesting (Prism software). Significance was set at P < 0.05.

Data and Resource Availability

All data generated or analyzed during this study are included in this published article and the Supplementary Material. Raw data for body composition, GTTs, hepatic TG assays, Western blots, immunohistochemistry, and real-time PCR are presented as mean ± SEM in the figures and tables. No new critical resources or reagents were generated. Data are available on reasonable request from the corresponding author, subject to institutional guidelines.

Results

Maternal HF Feeding Increases Adiposity in Adult Female Offspring

Our prior studies, as well as those of others, have shown that maternal HF feeding induces intrauterine growth restriction in rodents, in contrast with maternal obesity–induced macrosomia in humans and other models (5,11,18–20). Similarly, in this study, significantly lower body weight was observed in F1 fetuses of HF diet–fed dams compared with that in F1 fetuses of chow-fed dams (Supplementary Fig. 1A and B), a trend also seen in neonates at postnatal day 1 (P1) (Fig. 1A and Supplementary Fig. 1C for males). A rapid catch-up in body weight occurred during the first 2 weeks of life, normalizing by P15 (Fig. 1A and Supplementary Fig. 1C for males). Using EchoMRI, we monitored body composition in F1 female offspring from P30 onward (Fig. 1B–D). Consistent with prior reports (5,8), female OF-HFD at P60 exhibited normal body weight but significantly higher body fat and adiposity levels than OF-CD (Fig. 1B–D ). Despite this increased adiposity, blood glucose and insulin concentrations during GTTs at P60 remained comparable between OF-HFD and OF-CD (Fig. 1E and F), indicating normal glucose metabolism in nonpregnant female OF-HFD.

Figure 1.

Maternal HF feeding increased F1 female offspring adiposity. C57BL/6 female mice (F0) were fed HF diet or chow 3 months before and during pregnancy. Groups are defined in Table 2. A: Significant reduction in body weight (BW) was observed in female offspring (F1) at P1. A and B: BW was restored at P15 and stayed in the normal range before pregnancy (n = 4–6 litters). C and D: Increased body fat and adiposity rates were detected in F1 OF-HFD (orange) at P60 compared with OF-CD (black). E and F: Comparable blood glucose and insulin concentrations were observed during intraperitoneal GTTs. Glucose was measured using glucose oxidase. Insulin levels were determined by an ELISA kit. Data are presented as mean ± SEM. *P < 0.05 vs. OF-CD at same age (n = 6–8) (B–F).

Maternal HF Feeding Protects F1 Female Offspring From Pregnancy- and HF Diet–Induced Fat Gain

Pregnancy typically induces maternal weight gain, including an expansion of body fat, across mammals. Although female OF-HFD had significantly higher body fat before pregnancy (Fig. 1C and D), their rates of fat gain during pregnancy (E0.5–E18.5) were remarkably lower than those of OF-CD when both were fed chow (Fig. 2A). Because of these reduced rates of fat gain, body fat mass at the end of pregnancy was comparable between OF-HFD and OF-CD (Fig. 2B). We then challenged some F1 offspring with an HF diet 1 month before and during pregnancy. Surprisingly, prepregnancy fat gain rates in HF diet–fed OF-HFD (HF-HFD) were significantly lower than those in HF diet–fed OF-CD (HF-CD), resulting in similar body fat between the two groups before mating (Supplementary Fig. 2A and B). Interestingly, unlike pregnant OF-CD and OF-HFD (Fig. 2A), the HF-CD and HF-HFD mice exhibited similar rates of body fat gain during pregnancy (Fig. 2C). Therefore, there was no significant difference in body fat mass between HF-CD and HF-HFD at the end of pregnancy (Fig. 2D). Consistent with our prior findings (19), HF feeding robustly reduced pregnancy-induced fat gain rates in both groups compared with chow-fed counterparts (Supplementary Fig. 2C).

Figure 2.

Maternal HF feeding impaired pregnancy-induced fat gain in F1 offspring. A and C: Rates of body fat gain were calculated using data from E0.5 and E18.5. B and D: Body fat was measured by EchoMRI. E and F: Gonadal fat was collected at E18.5 in the fed state, with protein levels measured by Western blotting (E) and mRNA levels measured using real-time PCR (F). Data are presented as mean ± SEM. ATGL, adipose TG lipase; AU, arbitrary unit; IRβ, insulin receptor β; ns, not significant; p, phosphorylated; PPARγ, peroxisome proliferator–activated receptor γ.

The expression levels of principal adipogenic transcription factors and lipolytic enzymes were measured in gonadal fat to further study pregnancy-induced fat gain. In gonadal fat at E18.5, peroxisome proliferator–activated receptor γ protein levels showed a slight nonsignificant decrease in chow-fed OF-HFD (Fig. 2E), whereas levels of C/EBPα, AKT, LPL, HSL, and adipose TG lipase were comparable between groups (Fig. 2E and F and Supplementary Fig. 2D and E), suggesting minimal changes in adipogenesis and lipolysis.

Maternal HF Feeding Alters F2 Placental and Fetal Body Weight

Maternal metabolism and placental nutrient supply are critical for fetal growth. In chow-fed F1 pregnancies, no difference was seen in F2 fetal body weight at E18.5 between OF-HFD and OF-CD (Fig. 3A). However, placental weight of F2 male fetuses of OF-HFD was significantly lower than that of F2 male fetuses of OF-CD (Fig. 3B), with cross-sectional areas of the labyrinth, junctional zone, and decidua remaining similar (Fig. 3C), indicating a proportional reduction in placental mass. When F1 offspring were fed an HF diet, F2 fetal body weight of fetuses of HF-HFD was significantly lower than that of fetuses of HF-CD for both males and females (Fig. 3D). In contrast, placental weight was comparable (Fig. 3E), with no noticeable changes in placental cross-sectional areas (data not shown). These results suggest that maternal HF feeding influences F2 placental development, with novel sexual dimorphism. Although the placental effects were moderated after HF feeding, the inhibitory effects on fetal growth were revealed in the overnutrient state.

Figure 3.

Maternal HF feeding altered F2 placental and fetal weight. A–C: F0 dams were fed an HF diet or chow for 3 months and during pregnancy. F1 female offspring were fed with chow. D and E: Some F1 female offspring were fed an HF diet 1 month before and during pregnancy and were mated with chow-fed sires. F2 fetuses and placentas were collected at E18.5 in the fed state. Sex was determined by PCR. Placentas were fixed by paraffin and hematoxylin-eosin stained. Placental cross-sectional areas of the labyrinth (LZ), junctional zone (JZ), and decidua basalis (DB) were measured using ImageJ software (data from n = 5–6 images per placenta were averaged and presented). Data are presented as mean ± SEM. *P < 0.05 vs. OF-CD or HF-CD (n = 4–6 litters).

Although the current study did not provide any mechanical information, these results suggest that F0 maternal obesity alters the placental development and intrauterine growth of F2 offspring, especially when F1 offspring are overnourished. These results indirectly indicate that maternal obesity impairs F1 offspring’s metabolic adaptation to pregnancy. Sexual dimorphism is widely observed in programmed obesity and insulin resistance of F1 offspring (7–10,20–23). However, our demonstration of sexual differences in F2 placental development and fetal growth is novel.

Maternal HF Feeding Diminishes Pregnancy-Induced Increase in Maternal Blood TG Concentrations in F1 Offspring

Maternal blood TG concentrations steadily increase in normal pregnancies (19,24). Despite uncertainty regarding the underlying mechanisms, maternal blood TG levels are associated with birth weight (25,26). Our results showed that blood TG concentrations in nonpregnant OF-HFD were slightly but not significantly higher than those in controls (P > 0.05) (Fig. 4A, basal). When these chow-fed F1 offspring were pregnant, there was a significant increase in blood TG concentrations in OF-CD (Fig. 4A). However, blood TG concentrations in OF-HFD were significantly lower than those in OF-CD (Fig. 4A). As expected, HF feeding robustly increased blood TG levels in nonpregnant F1 HF-CD compared with OF-CD (127.2 ± 14.6 vs. 69.41 ± 8.4 mg/dL; P < 0.05). In contrast, blood TG concentrations in HF-HFD were remarkably lower than those in HF-CD before pregnancy (Fig. 4B, basal). Consistent with our previous reports (19,27), a pregnancy-induced increase in maternal blood TG concentrations was abolished in both HF-CD and HF-HFD (Fig. 4B, basal vs. E18.5). Most importantly, blood TG concentrations in HF-HFD were significantly lower than those in HF-CD (Fig. 4B, E18.5). Together, these results demonstrate that maternal blood TG concentrations in F1 OF-HFD were reduced considerably during pregnancy.

Figure 4.

Maternal HF feeding diminished the pregnancy-induced increase in maternal blood TG concentrations in F1 offspring. A and B: Blood TG concentrations were measured using fed blood samples collected before mating or at E18.5. Basal was the day before mating. C, D, F, and G: Blood and tissue samples were collected at E18.5 in the fed state. Postheparin blood (PHB) LPL activity was determined using samples 10 min after intravenous heparin injection (C), with mRNA levels measured by real-time PCR (D) and liver TG content determined using the lipid extraction approach (F) and oil red O staining (G). E: Hepatic TG release assay was performed at E16.5 with overnight fasting and endogenous LPL inhibition. Data are presented as mean ± SEM. *P < 0.05 vs. OF-CD or HF-CD, #P < 0.05 vs. same group at basal (A) or OF-CD and OF-HFD (C).

The balance between hepatic TG production and LPL-directed TG clearance in peripheral tissues determines blood TG concentrations. In addition to elevated hepatic TG secretion during pregnancy, progressively increased ANGPTL4 inhibits LPL activity, increasing maternal blood TG concentrations (19). Although LPL activity was significantly increased in HF diet–fed F1 offspring, there was no significant difference in LPL activity between OF-HFD and OF-CD or HF-HFD and HF-CD (Fig. 4C). Similarly, Angptl4 expression levels were not altered in fat or livers (Figs. 2F and 4D). In contrast, hepatic TG release rates were significantly reduced in both OF-HFD and HF-HFD (Fig. 4E). Together, these results indicate that the diminished increase in maternal blood TG levels in OF-HFD and HF-HFD might be caused by reduced hepatic TG secretion. The significantly higher levels of hepatic TGs in OF-HFD and HF-HFD provide additional support for this conclusion (Fig. 4F and G).

Maternal HF Feeding Impairs F1 Female Offspring Glucose Tolerance During Pregnancy

Our prior work has shown that prolonged maternal HF feeding reduces fetal and neonatal blood glucose levels (18). Here, F1 OF-HFD glucose levels normalized postnatally (Fig. 5A) and remained comparable to those in OF-CD before pregnancy (Fig. 1E). However, during pregnancy, OF-HFD exhibited significantly elevated blood glucose levels but decreased insulin concentrations during GTTs (Fig. 5B and C). At the end of pregnancy (E18.5), significantly higher random-fed glucose concentrations were observed in OF-HFD (Fig. 5D). Pregnant HF-HFD also showed elevated glucose compared with HF-CD (Fig. 5D). The random-fed blood insulin concentrations at E18.5 were noticeably lower in both OF-HFD and HF-HFD compared with their controls (Fig. 5E), indicating insulin insufficiency.

Figure 5.

Impaired glucose metabolism during F1 female offspring pregnancy. A: Comparable fed blood glucose levels were observed in F1 female offspring at P20. B and C: GTT was performed at E12.5 after overnight fasting with intraperitoneal glucose injection. D and E: Fed blood samples (E18.5) were used for glucose and insulin concentration measurement. Glucose concentrations were determined using glucose oxidase. Insulin level was measured using an ELISA kit. Data are presented as mean ± SEM. *P < 0.05 vs. OF-CD at same time point (n = 6) (B and C). AUC, area under the curve.

Maternal HF Feeding Impairs F1 Offspring β-Cell Adaptation and Insulin Secretion During Pregnancy

Increasing insulin production is a hallmark of maternal metabolic adaptation during pregnancy, which controls maternal glucose and lipid metabolism (28–31). Our previous study showed that HF diet–induced maternal obesity promotes F1 fetal β-cell development and neonatal insulin production (18). In contrast, blood insulin concentrations were lower in pregnant F1 offspring (Fig. 5C and E). Therefore, we studied β-cell maturation and insulin secretion in F1 offspring before their own pregnancy. Results showed that blood insulin concentrations and islet structure in OF-HFD were quickly restored to levels seen in OF-CD after birth (Fig. 6A and B). During the pregnancies of F1 offspring, the proliferation marker Ki-67 protein–positive rates of β-cells (Fig. 6C and Supplementary Fig. 3) and β-cell mass (Fig. 6D) in OF-HFD and HF-HFD were still comparable to those in OF-CD or HF-CD, respectively. However, the low blood insulin concentrations in OF-HFD and HF-HFD (Fig. 5C and E) suggest a compromised functional adaptation of islets to pregnancy. Indeed, GSIS rates were remarkably reduced in islets from OF-HFD and HF-HFD during pregnancy (Fig. 6E), consistent with prior models showing the effect of maternal obesity on islet function (32).

Figure 6.

Maternal HF feeding reduced F1 offspring insulin secretion during pregnancy. A and B: F1 female offspring samples were collected at P10 and P20 in the fed state, with blood insulin levels measured by ELISA (A) and islet structure studied using immunofluorescence (IF) with anti-insulin and glucagon antibody (scale bar, 50 μm) (B). C: Ki-67 protein–positive rates were calculated using IF images with anti–Ki-67 and anti-insulin antibodies. D: β-Cell mass was determined using anti-insulin antibody–probed IHC images with a series of sections. E: Islets were isolated at E16.5 and cultured overnight; GSIS was performed using size-matched islets with 2 or 20 mmol/L glucose. F: Protein levels were measured by Western blotting using tissues at E18.5 in the fed state. Data are presented as mean ± SEM. AU, arbitrary unit; IRβ, insulin receptor β; ns, not significant; p, phosphorylated.

Although an appropriate blood insulin concentration is essential for maintaining systemic glucose metabolism, insulin signaling in peripheral tissues is also critical to this process. Interestingly, phosphorylation levels of AKT and GSK3 were the same in skeletal muscle (Fig. 6F) and gonadal fat (Fig. 2E) between the two groups of F1 offspring, regardless of whether they were fed chow or an HF diet during pregnancy (Supplementary Figs. 2D and 4A). In contrast, phosphorylation levels of AKT significantly increased in the livers of OF-HFD (Fig. 6F) and HF-HFD (Supplementary Fig. 4B), without changes in protein levels of insulin receptor β or PTEN. Consistent with this increased phosphorylation of AKT in the liver, mRNA levels of the gluconeogenic gene G6pase were significantly reduced in OF-HFD (Fig. 4D). These results suggest that impaired glucose tolerance and elevated blood glucose concentrations in pregnant F1 OF-HFD and HF-HFD are primarily caused by insulin insufficiency. Additional studies are required to elucidate how maternal obesity enhances insulin signaling in the livers of F1 female offspring during pregnancy and the role of increased insulin signaling in hepatocytes in systemic glucose and lipid metabolism in these unique offspring.

Discussion

Abnormal intrauterine metabolism underlies the developmental origins of health and disease, with maternal metabolic adaptation playing a pivotal role in maintaining fetal nutrient supply. This study demonstrates that maternal HF diet–induced obesity in C57BL/6J mice adversely alters the metabolic adaptation of F1 female offspring during pregnancy. Our findings reveal impaired maternal lipid and glucose metabolism, coupled with compromised insulin production, providing compelling evidence that F0 obesity initiates a cascade of metabolic dysfunction in F1 females that manifests during pregnancy.

The concept of fetal programming originated from epidemiological studies linking maternal nutrient deficiency to adult disease (1,2). Subsequent research expanded this paradigm to include maternal obesity and overnutrition (3). Using HF diet–induced obese rodents, particularly C57BL/6 mice, studies have consistently shown programming effects, such as obesity and metabolic abnormalities, in offspring, often more pronounced in males (7–10). Our study confirms increased adiposity in nonpregnant F1 female offspring (Fig. 1C and D), consistent with prior reports (5,32). Unlike male offspring, however, nonpregnant F1 females exhibited no significant changes in blood glucose or TG concentrations (Figs. 1E and 4A), reinforcing the sexual dimorphism in metabolic outcomes. This suggests that although F0 obesity elevates adiposity universally, its metabolic effect in females may remain latent until challenged by pregnancy. Notably, unpublished data and other studies indicate that HF feeding during pregnancy alone, without prepregnant obesity, does not produce an obese phenotype in adult F1 females (8), highlighting the critical role of prolonged F0 obesity in programming F1 metabolic alterations.

A striking finding is the diminished pregnancy-induced fat gain in F1 female OF-HFD, despite their higher prepregnancy adiposity (Figs. 1C and D and 2A and B). In C57BL/6 mice, pregnancy typically doubles maternal fat mass to support fetal growth (12,18,33); however, OF-HFD females exhibited significantly lower fat gain rates (Fig. 2A). Unexpectedly, prepregnancy HF feeding in HF-HFD females also resulted in markedly reduced fat gain (∼148% vs. 310% in HF-CD) (Supplementary Fig. 2A), leading to comparable absolute fat mass before and at the end of pregnancy (Fig. 2D and Supplementary Fig. 2B). This contrasts with prior studies where OF-HFD, particularly males, showed accelerated fat accumulation on HF diet challenge (5,7,8). This female-specific attenuation suggests programmed resistance to fat expansion, potentially mediated by estrogen or altered adipogenic signaling (10,23). However, gonadal fat showed no significant changes in adipogenic (peroxisome proliferator–activated receptor γ and C/EBPα) or lipolytic (HSL and adipose TG lipase) markers (Fig. 2E and F and Supplementary Fig. 2D and E), indicating other mechanisms, such as lipid partitioning or energy expenditure, may be involved. This diminished fat gain highlights disrupted F1 metabolic adaptation, potentially limiting fetal nutrient reserves. Future studies comparing sexes and exploring epigenetic or hormonal drivers could clarify this transgenerational resilience.

The reduction in pregnancy-induced blood TG concentrations further illustrates the adverse programming of F0 obesity. Typically, maternal TG levels rise steadily to support fetal growth (19,24), correlating with birth weight (25). In OF-HFD, this increase was significantly diminished (Fig. 4A). In HF-HFD, TG levels were even lower than those in HF-CD (Fig. 4B). Our prior work showed that obesity increases TG breakdown via LPL (19). However, here, LPL activity and Angptl4 expression remained unchanged (Fig. 4C, D, and F), whereas hepatic TG secretion decreased (Fig. 4E), with elevated liver TG content (Fig. 4F and G). This suggests F0 obesity programs a reduction in TG release rather than enhanced clearance, impairing lipid availability during F1 pregnancy. Reduced maternal serum TGs in OF-HFD and HF-HFD may limit lipid supplies to F2 fetuses and their growth. Enhanced liver insulin signaling in OF-HFD and HF-HFD (Fig. 6F and Supplementary Fig. 4B) may suppress VLDL TG secretion (34,35), contributing to this effect. This disruption of lipid metabolism highlights a critical failure in F1 metabolic adaptation, potentially compromising fetal nutrient supply.

Glucose metabolism, a cornerstone of maternal adaptation, was also impaired in pregnant F1 offspring. Normal pregnancy reduces maternal glucose levels via increased use by the fetus and hyperinsulinemia (13,36). OF-HFD displayed glucose intolerance and elevated blood glucose (Fig. 5B and D), hallmarks of GDM (4,14). This was exacerbated in HF-HFD (Fig. 5D), with lower insulin levels in both (Fig. 5E), pointing to insulin insufficiency. Pregnancy demands β-cell expansion and increased insulin secretion (37,38); however, although β-cell mass and proliferation were unchanged (Fig. 6C and D), GSIS was markedly reduced in OF-HFD and HF-HFD (Fig. 6E). Although F0 obesity enhances fetal β-cell development (18), this benefit does not persist into F1 adulthood under pregnancy stress, as seen in other models (32,39,40). This compromised β-cell function, unique to pregnancy, highlights how F0 obesity predisposes a failure in insulin adaptation, leading to glucose dysregulation in F1 females. Regulation of maternal β-cell proliferation and insulin production is a systemic process, involving hormonal systems and energy metabolism (37,38,40). A primate study reported that a maternal Western-style diet increases offspring’s β-cell mitochondrial fragmentation and causes excessive insulin production during early life, which may predispose the offspring to β-cell failure later in life (40). Therefore, additional studies are warranted to determine the mechanism by which maternal obesity adversely programs F1 female offspring’s islet adaptation to pregnancy, including β-cell mitochondrial network and intracellular energy metabolism (40).

Interestingly, although peripheral insulin signaling (skeletal muscle and fat) was unaltered (Fig. 6F and Supplementary Figs. 2D and 4A), liver insulin signaling was enhanced in OF-HFD and HF-HFD (Fig. 6F and Supplementary Fig. 4B), with reduced G6pase expression (Fig. 4D). This contrasts with the typical insulin resistance seen in pregnancy (37) and suggests protective feedback to insulin insufficiency, although insufficient to normalize systemic glucose (Fig. 5B and D). PTEN levels were unchanged, leaving the mechanism of this hepatic sensitivity unclear (41). This paradoxical enhancement further illustrates how F0 obesity disrupts F1 metabolic homeostasis, with implications for the regulation of lipids and glucose.

Although the current study does not provide extensive mechanistic information to explain how maternal obesity adversely affects F1 female offspring’s metabolic adaptation to pregnancy, epigenetic modifications, such as altered methylation of adipogenic, β-cell functional genes, may contribute (1,20,22). Catch-up growth after low birth weight likely exacerbates these effects, because it is associated with programmed adiposity and metabolic dysfunction. Maternal HF diet enhances F1 fetal and early life insulin production (18,40); however, normal prepregnancy glucose (Fig. 1E) gives way to pregnancy-induced insulin insufficiency and glucose intolerance (Figs. 5B–D and 6E), suggesting latent mechanisms unmasked by pregnancy stress. Future studies using cross-fostering or epigenetic profiling could elucidate these drivers.

In summary, our findings demonstrate that F0 HF diet–induced obesity adversely alters F1 female metabolic adaptation to pregnancy. Diminished fat gain, reduced TG levels, glucose intolerance, and impaired insulin secretion collectively indicate programmed dysfunction originating from maternal obesity. These effects, if translatable to humans, could explain the transgenerational persistence of GDM and offer a novel mechanism for the developmental origins of metabolic disease. Future studies should investigate the molecular foundations, possibly including epigenetic factors, and verify these findings in human cohorts to guide preventive strategies.

This article contains supplementary material online at https://doi.org/10.2337/figshare.30236443.

Funding Statement

This work was supported by National Institutes of Health grants DK095132 (J.S.) and DK113007 (J.S.).