Mid-gestation low-dose LPS administration results in female-specific excessive weight gain upon a western style diet in mouse offspring

The developmental origins of health and disease (DOHaD) hypothesis posits that early-life environments shape adult metabolic, cardiovascular, and neurologic health. Many gestational complications (e.g., preeclampsia, gestational diabetes, maternal obesity) share a pro-inflammatory profile and are linked to adverse long-term outcomes in offspring. Maternal inflammation can alter placental development and nutrient transport, increase oxidative stress, and expose the fetus to cytokines and immune cells, thereby programming organ development. Lipopolysaccharide (LPS) administration is a well-established model of maternal inflammation, provoking a transient Th1 cytokine response. Prior animal work shows dose- and timing-dependent effects of prenatal LPS on growth, metabolism, and cardiovascular function. The present study tests the hypothesis that a single low-dose LPS exposure at mid-gestation programs offspring toward adverse metabolic phenotypes that are unmasked or exacerbated by a post-weaning western-style diet (WSD). The study further explores potential mechanisms in liver, adipose tissue, hypothalamus, and cortex, and examines sex-specific effects.

Prior studies demonstrate that maternal inflammation and complications (e.g., preeclampsia, gestational diabetes, maternal obesity) are associated with long-term offspring metabolic and cardiovascular risk. In rodents, low-dose late-gestation LPS causes intrauterine growth restriction followed by catch-up growth in females, while mid-gestation LPS can induce myocardial fibrosis, elevated blood pressure, and metabolic dysregulation, especially in males. Mechanistically, inflammation alters placental function, increases oxidative stress, and can directly affect fetal tissues; microglial development and hypothalamic circuits are particularly sensitive to maternal immune activation. Nutritional challenges postnatally (e.g., WSD) often amplify programmed phenotypes. Epigenetic regulation of metabolic genes (e.g., Lxra) and hypothalamic pathways involved in leptin signaling and energy balance has been implicated in fetal programming. These data support investigating how mid-gestational low-dose LPS, combined with WSD, impacts sex-specific metabolic and neurobiological outcomes.

Design and animals: Pregnant C57BL/6J mice were generated by timed mating (GD 0.5 = day of plug). Housing was under 12 h light/dark with ad libitum chow and water. At GD 8.5, an sFlt-1 adenovirus (or empty control) was administered per a preeclampsia model protocol; this lot proved inactive, so the study models maternal inflammation only. At GD 10.5, dams in the treatment group received 25 µg/kg LPS (E. coli O111:B4) intraperitoneally; controls received saline. At GD 15.5, maternal tail blood was collected for metabolomics and sFlt-1 ELISA. Dams were allowed to deliver (resorptions/non-deliveries recorded). Postnatal handling: Litters were standardized to 5–6 pups (P1–P3), with at least one male from another dam to normalize rearing. Pups were weaned at P21. Where possible, two males and two females per dam were retained and later split between diet groups, pair-housed with non-littermates when feasible. Diet intervention: From weaning to 12 weeks (WK12), all offspring received a semi-synthetic control diet (CTRD). From WK12–WK24, half of each sex-by-treatment group was switched to a western-style diet (WSD; D12079B), the remainder continued CTRD. Outcomes and timelines: - Body weight weekly from 4–24 weeks. - Body composition (Bruker minispec LF90II) at WK12 and WK24. - Food intake measured at WK22–WK24 and expressed as kcal/animal/day. - Fasting blood (6 h fast) at WK10 and WK22. - Insulin tolerance tests (ITT) after 4 h fast at WK11 (0.5 U/kg insulin in females and 12-week males) and WK23 (0.5 U/kg in females; 0.75 U/kg in 23-week males); tail blood glucose at 0–120 min. - Oral glucose tolerance tests (OGTT) after 10 h overnight fast at WK12 (1 g/kg) and WK24 (2 g/kg); tail glucose at 0–120 min. AUC calculated by trapezoidal rule. - Terminal measures at WK24: anesthesia, aortic blood pressure (catheter), cardiac puncture blood collection, tissue harvest (liver, heart, left kidney, gonadal and inguinal white adipose tissues, whole hypothalamus, whole cortex). Plasma assays: In dams, sFlt-1 by ELISA; metabolomics (MxP Quant500, Biocrates; 630 metabolites) on 10 µL plasma using FIA/LC–MS/MS with internal standards and three-level QCs. In offspring, plasma leptin and insulin (fasted and/or terminal non-fasted), triglycerides (fasted). Molecular analyses: RNA and DNA isolated from liver, gWAT, cortex (TRIzol/back-extraction), and from hypothalamus (AllPrep). qPCR performed (QuantStudio 3) using SYBR Green or TaqMan, relative expression via standard curves normalized to organ-appropriate housekeeping genes (Hprt and Actb for hypothalamus; Gapdh for cortex; 36b4 for liver; Actb for gWAT). Target panels included hepatic and adipose genes in lipid metabolism (e.g., Lxra, Srebf2, Srebf1c, Chrebp, Fasn, Elovl6, Cidea, Cidec, Ppara, Cpt1a, Scd1, Pparg, Acaca), hypothalamic orexigenic/anorexigenic genes (Npy, Agrp, Socs3, Lepr, Pomc, Cartpt, Mc3r, Mc4r) and behavior-related genes (Crh, Fkbp5, Cnr1), and cortical microglia/astrocyte homeostasis/activation markers. DNA methylation: Bisulfite conversion of hepatic DNA; pyrosequencing of promoter regions for Fasn, Lxra, Srebf2. In hypothalamus, selected CpGs in Crh, Pomc, Lepr promoter/exon. Statistics: Three-way ANOVA with factors sex, diet, treatment; mixed ANOVA for repeated measures; Sidak correction for multiple comparisons; Huynh-Feldt for sphericity; transformations applied as needed. Significance at p<0.05. Metabolomics processed in MetaboAnalyst 4.0 with missing value imputation (half-minimum), log-transform, and FDR-adjusted t-tests (q<0.05). Ethical approval: Approved by Dutch national and University of Groningen institutional committees, in accordance with European Convention for the Protection of Vertebrate Animals.

• Maternal outcomes: Low-dose mid-gestation LPS (25 µg/kg, GD10.5) reduced pregnancy success: 6/14 LPS-treated dams did not deliver; some delivered smaller litters, suggesting partial resorption. Maternal metabolomics at GD15.5 showed sustained oxidative stress signatures: xanthine significantly increased versus controls (p<0.01); hypoxanthine was detectable in LPS dams but below detection in controls. - Growth and adiposity: From weaning to WK12 on CTRD, body weights were similar; at WK12, LPS-exposed females had higher body fat mass (p<0.05). From WK12–WK24, WSD unmasked a robust phenotype: LPS+WSD females exhibited hyperphagia (higher average daily caloric intake at WK22–24; p<0.05), greater body weight gain (p<0.01 to p<0.001 across weeks), and higher fat mass at WK24 (p<0.01) versus WSD controls. LPS-CTRD males showed increased fat mass at WK24 (p<0.05) without higher body weight. After weight correction, gWAT and inguinal WAT weights remained higher in LPS-WSD females (p<0.05 and p<0.01). No LPS effect on systolic/diastolic blood pressure. - Glucose–insulin homeostasis: ITT at WK11 showed no differences. At WK23, LPS tended to worsen insulin tolerance in males (AUC p=0.055). OGTT AUC at WK24 was increased by male sex (p<0.01), WSD (p<0.001), and LPS (p<0.05). Fasting insulin at WK22 was unchanged; non-fasted insulin was increased by WSD (p<0.01). Fasting leptin (WK22) was elevated by LPS in WSD offspring only (p<0.01), while non-fasted leptin was increased by diet in both sexes (males p<0.001; females p<0.05). Fasting triglycerides increased with LPS at WK10 (p<0.01). - Hepatic gene expression: LPS increased Lxra (p<0.001) and Srebf2 (p<0.01) expression; trends for Chrebp (p=0.058) and Srebf1c (p=0.063). Sub-analyses indicated LPS effects on Fasn in CTRD males and trends for Elovl6. WSD markedly upregulated Cidea and Cidec (steatosis markers) in males (p<0.001) and to a lesser extent in females, irrespective of LPS. WSD also increased Ppara, Cpt1a, Scd1, and Pparg; Acaca unchanged. - Adipose tissue gene expression: In gWAT, Lxra was higher with LPS (p<0.001). A three-way interaction showed higher Leptin expression in LPS versus controls in CTRD males and WSD females, matching groups with elevated fat mass. - DNA methylation (liver): No LPS-associated differences in promoter methylation of Fasn, Lxra, or Srebf2; Fasn methylation correlated inversely with expression (r=-0.287, p=0.019). - Hypothalamus: LPS decreased Cartpt (anorexigenic) expression (p<0.05) and showed a trend to increase Lepr (p=0.079), most evident in females. Behavior-related Crh trended higher with LPS (p=0.094), especially in females. No consistent LPS effects on average DNA methylation of Crh, Pomc, or Lepr; isolated CpG differences were inconsistent between sexes. - Cortex glial markers: No consistent activation pattern. LPS generally decreased microglia/astrocyte homeostasis and activation markers in subgroup- and sex-dependent ways (e.g., decreased P2ry12 and Tmem119 in male CTRD and female WSD; decreased Aqp4, Gfap, Axl in male CTRD; selective changes in females on WSD). - Overall: A single mid-gestation low-dose LPS exposure programs female offspring to hyperphagia, increased adiposity, and impaired glucose handling when challenged with WSD, with accompanying alterations in hypothalamic, hepatic, and adipose gene expression.

The study demonstrates that transient maternal inflammation at mid-gestation predisposes offspring—most notably females—to hyperphagia and excessive weight and fat mass gain when exposed to a western-style diet in adulthood. The metabolic phenotype in LPS-WSD females (greater caloric intake, adiposity, impaired glucose tolerance, elevated fasting leptin) suggests altered central regulation of energy balance. Hypothalamic gene expression was consistent with reduced anorexigenic signaling (decreased Cartpt) and a trend toward enhanced leptin receptor expression, potentially indicating altered leptin sensitivity or signaling integration. Given that hypothalami were sampled in the fed state and during the light phase, observed differences may underestimate true regulatory changes. Hepatic and adipose gene expression differences (e.g., increased Lxra and Srebf2; elevated gWAT Leptin) likely reflect or modulate the increased adiposity rather than primary programming of hepatic lipogenesis, as hepatic promoter methylation of key lipogenic genes was unaltered. Maternal metabolomics indicated sustained oxidative stress (elevated xanthine/hypoxanthine) days after LPS, suggesting that prolonged oxidative milieu alongside the initial cytokine surge could mediate fetal programming. The pronounced female-specific phenotype may involve sex-dependent placental transport and glucocorticoid sensitivity prenatally, and/or interactions with sex hormones and reward/hedonic feeding pathways postnatally. Cortical microglia/astrocyte transcriptional changes lacked a coherent activation signature, potentially influenced by timing, LPS dose, or tissue heterogeneity, but indicate that prenatal inflammation can leave lasting, sex- and diet-dependent imprints on glial homeostasis. Collectively, the findings align with the hypothesis that maternal inflammation programs central circuits governing energy intake, which are unmasked by a palatable diet, leading to adverse metabolic outcomes primarily in females.

A single mid-gestation low-dose LPS exposure in mice programs female offspring to develop hyperphagia, increased body weight and fat mass, and impaired glucose handling when later challenged with a western-style diet. These outcomes are accompanied by altered expression of genes in the hypothalamus (reduced Cartpt, trends in Lepr and Crh) and changes in liver and adipose tissue gene expression, while hepatic promoter DNA methylation of key lipogenic genes was unchanged. Maternal metabolomics indicated persistent oxidative stress after LPS, a potential mediator of programming. The results underscore inflammation as a common mechanistic contributor to long-term offspring metabolic risk in gestational complications and highlight sex-specific vulnerability. Future research should include detailed behavioral phenotyping (including feeding microstructure and reward paradigms), circadian-phase specific sampling, leptin sensitivity tests, cell type– and nucleus–specific neuroanatomical and transcriptomic analyses, longitudinal profiling of maternal inflammatory/oxidative responses, and exploration of interventions that mitigate inflammatory/oxidative programming.

• The intended preeclampsia double-hit model failed due to inactive sFlt-1 adenovirus; thus, only maternal inflammation was modeled. - No neonatal or behavioral assessments were performed, limiting insight into early mechanisms and feeding behavior. - Hypothalamus and cortex were analyzed as whole regions rather than specific nuclei or isolated cell populations, potentially diluting localized effects. - Hypothalami were collected in the light phase and fed state; circadian and postprandial effects may have blunted differences. - Limited sampling prevented precise temporal characterization of the maternal inflammatory response post-LPS. - Some tissue-specific analyses had relatively low sample numbers. - The duration of WSD exposure and specific LPS dosing/timing may influence the generalizability of microglial/astrocyte findings.