Four-week-old virgin female C57BL/6 mice were obtained from the local animal facility (Orient Bio Co., Seongnam, Gyunggi-do, Korea) and were maintained in a temperature (23 ± 2℃) and humidity (55 ± 5%)-controlled room with a 12 h dark/light cycle. The experimental procedures used in the current study were approved by Seoul National University Institutional Animal Care and Use Committee (SNU-140807-1-3). After a one-week acclimation period, mice were randomly assigned to one of two groups. The LFD group was fed an LFD (10% calories from fat, 20% from protein, and 70% from carbohydrate, 3.85 kcal/g; #D12450B, Research Diets, Inc., New Brunswick, NJ, USA) and the HFD group was fed an HFD (45% calories from fat, 20% from protein, and 35% from carbohydrate, 4.73 kcal/g; #D12451, Research Diets, Inc., New Brunswick, NJ, USA). Experimental diets were provided for 3 weeks before mating and throughout pregnancy and lactation. Diets and water were provided ad libitum. Females were allowed to mate with mature males (2:1) and the litter size and pup body weights were recorded on postnatal day 3 (PD 3). Out of the 20 females in each group, 14 females in the LFD group and 9 females in the HFD group became pregnant. Lower pregnancy rate in dams fed an HFD was previously reported [1516]. Litters were adjusted to approximately 6 to normalize the growth. Dams (LFD and HFD groups; n = 7 and 5, respectively) and their male offspring at PD 21 (LFD-O and HFD-O groups; n = 10 and 7, respectively) were sacrificed after an overnight fast. Some dams were excluded due to the different sacrifice time point (postpartum day 2), lack of male offspring, or cannibalization of offspring. Some offspring used in the separate long-term intervention study were not sacrificed. Blood samples were obtained rapidly by cardiac puncture. Tissues, including liver, brain, and adipose tissues in dams and liver, brain, spleen, kidney, and adipose tissues in offspring were removed, washed with phosphate-buffered saline (PBS), and weighed. Livers were snap-frozen immediately in liquid nitrogen and stored at -80℃ for further analysis.

Blood was centrifuged at 3,000 rpm for 20 min at 4℃ and stored at -80℃ until analyzed. Serum glucose, triglyceride, total cholesterol and HDL-cholesterol levels were determined using commercial kits (Asan Pharmaceutical Co., Seoul, Korea). Alanine aminotransferase (ALT) activity was measured using the commercial kit (Asan Pharmaceutical Co., Seoul, Korea) and monocyte chemoattractant protein 1 (MCP-1) levels were analyzed using the Quantikine® mouse MCP-1 Immunoassy kit (R&D Systems, Minneapolis, MN, USA). Serum thiobarbituric acid reactive substances (TBARS) levels were measured according to the method of Ohkawa et al. [17]. The absorbance of the butanol layer was measured at 532 nm using 1,1,3,3-tetraethoxypropane as a standard. Serum total protein levels were measured using a Bradford protein assay kit (Bio-Rad, Hercules, CA, USA).

Total lipids were extracted according to the method of Folch et al. [18]. Hepatic triglyceride and cholesterol concentrations were determined by enzymatic colorimetric methods using commercial kits (Asan Pharmaceutical Co., Seoul, Korea). Hepatic TBARS was measured using the same method for serum TBARS. The hepatic lipid peroxide level was expressed as malondialdehyde equivalents per milligram of protein and the protein content of the homogenate was measured using a protein assay kit (Bio-Rad, Hercules, CA, USA). DNA was extracted using a DNeasy blood & tissue kit (Qiagen, Valencia, CA, USA).

Liver tissues were homogenized in 10 volumes (w:v) of ice-cold protein lysis buffer using the Tissue Lyser system (Qiagen, Valencia, CA, USA) with 5 mm sterile stainless steel beads. Thirty micrograms of protein were loaded into the lanes of a SDS-PAGE gel, separated, and blotted onto a PVDF membrane. After blocking with 5% nonfat milk or bovine serum albumin in TTBS, membranes were probed with a specific primary antibody, including proliferating cell nuclear antigen (PCNA; Santa Cruz Biotechnology, Santa Cruz, CA, USA), β-catenin (Santa Cruz Biotechnology, Santa Cruz, CA, USA), cleaved form of caspase-3 (Cell signaling Technology, Danvers, MA, USA) or anti-70-kDa heat shock cognate protein (HSC70; Cell signaling Technology, Danvers, MA, USA). The membranes were then incubated with an IgG-peroxidase-conjugated secondary antibody for chemiluminescent detection. The band intensities were quantified using Quantity One software (Bio-Rad, Hercules, CA, USA).

Total RNA of liver tissue was isolated using RNAiso Plus (Takara Bio Inc., Otsu, Shiga, Japan). cDNA was synthesized using 2 mg of total RNA with the Superscript®II Reverse Transcriptase (Invitrogen, Carlsbad, CA, USA) for qRT-PCR. All amplification reactions were performed using a StepOne™ Real Time PCR System (Applied Biosystems, Foster city, CA, USA) according to the manufacturer's protocol. Mouse ribosomal protein L19 (RPL19) was used as a reference gene and relative gene expression levels were analyzed using the 2^-ΔΔCt method. Primers for heme oxygenase 1 (HO-1; F 5'-CCTCACTGGCAGGAA ATCATC-3', R 5'-CCTCGTGGAGACGCTTTACATA-3'), interleukin-1 β (IL-1β; F 5'-CAACCAACAAGTGATATTCTCCATG-3', R 5'-ATCCAC ACTCTCCAGCTGCA-3'), p40phox (F 5'-CCTGCCCACATTGCCAGC CA-3', R 5'-AGACCGGCAGGCTCAGGAGG-3'), and scavenger receptor B1 (SR-B1; F 5'-CCTTCAATGACAACGACACCG-3', R 5'-CCATGCGACTTGTCAGGCT-3') were used for amplification of DNA. Primer sequences for ATP-binding cassette sub-family A member 1 (ABCA1), ABCG5, ABCG8, cluster of differentiation 36 (CD36), carnitine palmitoyltransferase 1A (CPT1a), cholesterol 7 alpha-hydroxylase (CYP7A1), fatty acid synthase (FASN), farnesoid X receptor (FXR), 3-hydroxy-3-methyl-glutaryl-CoA reductase (HMGCR), low density lipoprotein receptor (LDLR), liver X receptor α (LXRα), MCP-1, stearoyl-CoA desaturase-1 (SCD1), SR-A, sterol regulatory element-binding protein 2 (SREBP-2), peroxisome proliferator-activated receptor α (PPARα), PPARγ, tumor necrosis factor α (TNFα), and RPL19 were previously published [19].

Statistical analyses were performed using SPSS version 22 software (IBM SPSS Inc., Chicago, Il, USA). Student t-test was used to determine statistical significance between two groups for all experiments. Data were expressed as mean ± SEM and differences were considered statistically significant at P < 0.05. Correlations between two variables were determined by Pearson's correlation coefficient.

There were no significant differences in maternal body weights measured at baseline, mating, and postpartum day 3 between groups (Fig. 1). In terms of total daily calorie intake, dams under an HFD consumed more calories than dams fed an LFD before mating, and during pregnancy and lactation (data not shown). Maternal consumption of HFD increased body weight and adipose tissue weight of dams at the end of the study (21 days after giving birth to offspring) (Table 1). Interestingly, a higher liver weight was observed in the LFD group compared to the HFD group. Significantly higher serum lipid parameters including triglyceride and total cholesterol were observed in HFD-fed dams (Fig. 2A). No significant difference in glucose, HDL-cholesterol, total protein, and ALT levels in the sera was observed between groups. HFD feeding significantly led to increased hepatic triglyceride and decreased cholesterol levels in dams (Fig. 2B).

Because altered lipid levels were observed in both serum and liver of dams fed an HFD, mRNA levels of hepatic genes involved in lipid metabolism were determined. There were no significant differences in major transcription factors and most enzymes involved in triglyceride metabolism (Fig. 3A). Interestingly, significantly lower mRNA levels of SCD1 were observed in the HFD group compared with the LFD group, suggesting the increased accumulation and VLDL incorporation of saturated fatty acids in the liver of dams fed an HFD as previously reported [2021]. There were significant differences in mRNA levels of major transcription factors involved in cholesterol metabolism, including LXRα, SREBP-2, and FXR, and their target genes including ABCG8 and LDLR (Fig. 3B & C).

Due to the increased metabolic demand during pregnancy, the maternal liver grows by hyperplasia and hypertrophy [10]. Therefore, total liver DNA and protein contents were determined to examine whether dysregulated hepatic proliferation and growth during pregnancy may result in lower liver weights in the HFD group. As shown in Fig. 4A, higher DNA and protein contents were observed in dams fed an LFD. The catch-up in liver growth during the lactation period was not observed in dams fed an HFD based on PCNA and active form of capase-3 protein levels (Fig. 4B).

Hypercholesterolemia has been suggested as an important risk factor for hepatic inflammation in animal models [19]. Therefore, we examined the effects of HFD feeding on oxidative stress and inflammation in dams. Significantly higher hepatic serum and lipid peroxide levels were observed in dams fed an HFD (Fig. 5A). Significantly higher hepatic mRNA levels of HO-1 and p40phox, enzymes involved in cellular antioxidant capacity, were observed in dams fed an LFD (Fig. 5B).

Although serum MCP-1 levels were not significantly different between groups (Fig. 5C), significantly higher expression of hepatic MCP-1 was observed in dams fed an HFD compared to dams fed an LFD. Expression of other inflammatory genes, including TNFα and CD68, was not significantly different between groups (data not shown).

Maternal HFD feeding did not change litter size, and the body weights of offspring were not significantly different between the two groups until weaning (data not shown). Maternal diet had no significant effects on offspring organ weights including liver, adipose tissue, brain, spleen, and kidney (Table 2).

Significantly higher serum total cholesterol levels, but not serum glucose, triglyceride, and ALT levels, were observed in the offspring from HFD-fed dams (HFD-O group) compared to the offspring from LFD-fed dams (LFD-O group) (Fig. 6A). Interestingly, serum total cholesterol levels of offspring were negatively correlated with liver weights of dams and positively correlated with hepatic TBARS levels of dams (Fig. 6B). No significant difference in hepatic triglyceride and cholesterol levels was observed between groups (data not shown).

Previous studies have reported that maternal HFD consumption interferes with liver development in offspring, which may induce development of liver disease later in life [82223]. Although no difference in relative liver weight was observed, additional biomarkers of liver development were determined in order to confirm the effect of maternal HFD on liver development in offspring. As shown in Fig. 7, no significant differences in cell proliferation, growth, and apoptosis in the liver were observed between the two groups.

Because significantly higher serum cholesterol levels were observed in the HFD-O group compared with the LFD-O group, the hepatic expression of genes involved in cholesterol metabolism was determined. Expression levels of most genes were not significantly different, however significant differences in mRNA levels of ABCG8, which promotes excretion of cholesterol into bile, were observed between the two groups (Fig. 8). No significant differences were observed in hepatic expression of genes involved in oxidative stress and inflammation (data not shown).