Large Litter Rearing Improves Leptin Sensitivity and Hypothalamic Appetite Markers in Offspring of Rat Dams Fed High-Fat Diet During Pregnancy and Lactation

Abstract

Maternal high-fat (HF) diet has long-term consequences on the offspring's metabolic phenotype. Here, we determined the effects of large litter (LL) rearing in offspring of rat dams fed HF diet during gestation and lactation. Pregnant Sprague-Dawley rats were maintained on standard chow (CHOW) or HF diet throughout gestation and lactation. Pups were raised in normal litters (NLs) (10 pups/dam) or LLs (16 pups/dam) during lactation, resulting in 4 groups: CHOW-NL, CHOW-LL, HF-NL, and HF-LL. The offspring were weaned onto to either CHOW or HF diet on postnatal day 21. Male and female pups with maternal HF diet (HF-NL) had greater body weight and adiposity, higher plasma leptin levels, impaired glucose tolerance, abnormal hypothalamic leptin signaling pathways (lower leptin receptor-b [OB-Rb] and signal transducer and activator of transcription 3, higher suppressor of cytokine signaling 3 mRNA expression) and appetite markers (lower neuropeptide Y and Agouti-related peptide mRNA expression), and reduced phospho-signal transducer and activator of transcription 3 level in response to leptin in the arcuate nucleus at weaning, whereas LL rearing normalized these differences. When weaned onto CHOW diet, adult male offspring from HF diet-fed dams continued to have greater adiposity, higher leptin levels, and lower hypothalamic OB-Rb, and LL rearing improved them. When weaned onto HF diet, both adult male and female offspring with maternal HF diet had greater body weight and adiposity, higher leptin levels, impaired glucose tolerance, lower OB-Rb, and higher suppressor of cytokine signaling 3 in hypothalamus compared with those of CHOW dams, whereas LL rearing improved most of them except male OB-Rb expression. Our data suggest that LL rearing improves hypothalamic leptin signaling pathways and appetite markers in an age- and sex-specific manner in this model.

It is widely accepted that obesity has become a worldwide health problem that often leads to many associated comorbidities, such as cardiovascular disease, hypertension, type 2 diabetes, some cancers, sleep apnea, and arthritis (1–4). Genetic and environmental factors both affect the development of obesity (4, 5). Increasing evidence suggests that the early-life environment can influence the development of obesity (6–8). Maternal high-fat (HF) diet throughout gestation and suckling has been shown to have long-term consequences on offspring's metabolic phenotype (9–11).

The hypothalamus is well known to play a key role in energy homeostasis. Neurons in the arcuate nucleus of the hypothalamus (ARC), which express neuropeptide Y (NPY), Agouti-related peptide (AgRP), and proopiomelanocortin (POMC), play an important role in this regulation (12). Leptin, an adipose tissue-secreted hormone, binds to leptin receptor-b (OB-Rb) in ARC and activates the Janus kinase-signal transducer and activator of transcription (STAT) pathway, leading to phosphorylation of STAT3 (13). Phosphorylated STAT3 modulates the expression of neuropeptides that control food intake and energy balance, such as NPY and POMC (14).

In rodents, hypothalamic neurogenesis occurs during midgestation, whereas the neural projections between different nuclei develop during early postnatal life (15, 16). Environmental changes during these critical developmental periods, for example, maternal HF diet consumption during gestation and suckling, may affect hypothalamic neurogenesis or neural projection development and have long-term consequences on offspring's metabolism. Leptin plays a critical neurotrophic role during the development of the hypothalamus, and neural projection pathways from the ARC to other hypothalamic nuclei are disrupted in leptin-deficient mice (17).

A significant issue in Western society is overnutrition as result of the consumption of modern diets that contain high amounts of fat. Because obesity has become a worldwide health problem, weight loss prevention or therapy is particularly important, especially in children. It has been demonstrated that large litter (LL) rearing enhances leptin sensitivity and protects selectively bred obesity-prone rats from becoming obese in diet-induced obesity (DIO) paradigm (8). Although neonates do make small amounts of their own leptin, they also ingest and absorb leptin from maternal milk at least up to postnatal day (PND)12 (18–21). Reduced intake of maternal milk in pups raised in LLs might be expected to reduce their plasma leptin levels.

Previous studies showed that maternal HF diet during gestation and suckling resulted in rat offspring with increased body weight, adiposity, leptin levels, impaired glucose tolerance, and reduced leptin sensitivity at weaning (22–24). Offspring from HF diet-fed dams also were more susceptible to DIO in adulthood (11). In this study, we sought to determine whether LL rearing during the suckling period improves the metabolic phenotype, leptin sensitivity, and hypothalamic gene expression in rat offspring exposed to maternal HF diet.

Materials and Methods

Animals and diet

Pregnant female Sprague-Dawley rats (Medical Experimental Animal Center of Xi'an Jiaotong University) were received on gestation day 2. Animals were individually housed in conventional tub cages with access to food and water ad libitum. The room was maintained on a 12-hour light, 12-hour dark cycle with light onset at 6 am. All animal procedures were approved by the Institutional Animal Care and Use Committee of Xi'an Jiaotong University.

Pregnant rats were divided into 2 groups according to their diet throughout gestation and lactation: standard chow (CHOW) diet (3.00 kcal/g, containing 58.6% carbohydrate, 28.4% protein, and 13.0% fat by calorie; n = 24) or HF diet (4.30 kcal/g, containing 28.9% carbohydrate, 20.6% protein, and 50.5% fat by calorie; n = 24). The 2 diets were formulated in the laboratory. All dams were started on their respective diets upon arrival on gestation day 2. Dams' body weight and food intake were measured daily throughout gestation.

The day of parturition is PND0. On PND1, offspring were randomized by weight and fostered to dams with the same diet in normal litter (NL) size (n = 10 pups/dam, 5 males and 5 females) or LLs (n = 16 pups/dam, 8 males and 8 females), resulting in 4 groups according to dams' diet and litter size: CHOW-NL (n = 12), CHOW-LL (n = 12), HF-NL (n = 12), and HF-LL (n = 12) (Figure 1). Pups and dams were weighed once a week on PND1, PND7, PND14, and PND21. Dams' food intake was measured daily throughout the suckling period. On PND7, PND14, and PND21, 1 male and 1 female pup per litter was killed by decapitation; blood was collected, centrifuged at 4°C to collect plasma, and stored at −80°C for biochemical analysis; sc and retroperitoneal fat was bilaterally dissected and weighed; and brains were collected and stored at −80°C for gene expression analysis. The pups that had been used for these studies from NLs and LLs were replaced by pups from donor NL and LL dams to maintain litter size.

Figure 1.

Experimental groups. Dams were fed CHOW or HF diet throughout gestation and lactation. The pups were raised in NL size (10) or LL size (16) during lactation. On PND21, offspring were weaned onto either CHOW or HF diet. The groups are described as “maternal diet-litter size-postweaning diet.”

On PND21, after measurement of naso-anal (N-A) length, half of the pups from each litter were weaned onto CHOW, whereas the other half were weaned onto the HF diet. This yielded 8 groups: CHOW-NL-CHOW, CHOW-NL-HF, CHOW-LL-CHOW, CHOW-LL-HF, HF-NL-CHOW, HF-NL-HF, HF-LL-CHOW, and HF-LL-HF (n = 12 per group for each sex) (Figure 1). Adult offspring were killed at 12 weeks of age.

Milk collection

On PND21, dams were removed from their home cages and were deeply anesthetized with ketamine and xylazine (4:3 mixture; 0.1 mL/100g body weight, ip) and injected with 4-IU oxytocin, ip (Sigma). Milk was collected 15 minutes later into glass capillary tubes. Samples were then transferred to microcentrifuge tubes, centrifuged at high speed at 4°C. The clear supernatant was collected and stored at −80°C for later analysis. After milk collection, dams were killed by decapitation.

Adipose depot measurement

On PND7, PND14, PND21, and at age 12 weeks, sc (dorsosubcutaneous and inguinal) and retroperitoneal fat pads were dissected at killing from offspring and weighed. The weight of the fat pad as a percentage of body weight was calculated. Because the amount of retroperitoneal fat is very small in PND7 rat pups and difficult to dissect, we only dissected sc fat on PND7.

Glucose tolerance test

On PND23 and at age 10 weeks, rats were food deprived overnight for 16 hours with only water available. Baseline fasted blood glucose was determined via a small tail nick by a handheld glucose meter (Roche). An oral gavage of glucose (2.0 g/kg body weight, 20% glucose in sterile water solution) was administered. Blood glucose was determined at 15, 30, 45, 60, and 120 minutes after glucose gavage using the glucometer.

Biochemical assays

Plasma and milk leptin concentrations were determined by commercially available ELISA kits for leptin for rat (Cusabio Biotech Co, Ltd). Plasma triglyceride and cholesterol were determined by commercially available kits (Biosino Bio-Technology & Science, Inc).

Leptin sensitivity

On PND21, male and female pups (n = 2 per sex per litter) received an ip injection of recombinant rat leptin (3 mg/kg, ip; Sigma) or saline. Pups were killed 3 hours later by decapitation. Brains were removed and immediately frozen on powdered dry ice and stored at −80°C. ARC was isolated from 500-μm-thick frozen coronal sections using a blunted 16-gauge stainless steel needle (inner diameter, 1.65 mm) based on the coordinates for developing rat brains described by Sherwood and Timiras (25). ARC samples were homogenized in lysis buffer (Sigma) with protease inhibitor cocktail (Roche) and phosphatase inhibitor cocktail (Roche). Protein concentration was determined using a protein assay kit (Thermo Scientific). Protein (30 μg) was run on a 3%–8% Tris-acetate gel and transferred onto polyvinylidene fluoride membranes. Blots were blocked with 5% nonfat dry milk for 2 hours. Phospho-(p-) STAT3 and total-(t-) STAT3 were determined using corresponding antibodies from Cell Signaling. Targeted proteins were revealed using SuperSignal West Pico Chemiluminescent Substrate (Thermo Scientific) and exposed to film (GE Healthcare). The intensity of bands was quantified using Scion Image Software (Scion). The ratio of the intensity of p-STAT3 to t-STAT3 was calculated to represent the level of phosphorylation. β-Actin (Sigma) was used as the loading control.

Hypothalamic neuropeptide and receptor mRNA assays by real-time PCR

On PND7, PND14, and PND21, pups were separated from dams at 9 am and were killed by decapitation 1 hour later at 10 am. At age 12 weeks, offspring were fasted 4 hours (9 am to 1 pm) and were killed by decapitation at 1 pm. Brains were removed and immediately frozen on powdered dry ice and stored at −80°C. The hypothalamus was collected based on the coordinates for developing rat brains described by Sherwood and Timiras (25) and adult rat brains described by Paxinos and Watson (26). Total RNA was isolated from the hypothalamus using Fast 1000 kit (Pioneer Biotech Co, Ltd). For each individual sample, 500-ng total RNA was used for reverse transcription using a commercial RT-PCR kit according to the manufacturer's instructions (TaKaRa). After first-strand synthesis, 2 μL of cDNA were added to 25 μL of the PCR mixture comprising 2 μL of the RT reaction product, 12.5 μL of 2× SYBR premix Ex Taq, 0.5 μL of each forward and reverse primer, and 9.5-μL dH₂O. The primers and accession numbers of the genes studied are listed in Supplemental Table 1. The thermal cycling parameters were 1 cycle at 95°C for 10 seconds followed by 40–45 cycles at 95°C for 5 seconds, and 58°C–60°C for 30 seconds, and was performed on a thermal cycler (Bio-Rad PCR thermal cycler, iQ5). To determine relative expression values, the −ΔΔCt method was used, where triplicate cross threshold values for each sample were averaged and subtracted from those derived from the housekeeping gene Actb. Only β-actin (Actb) was used as the housekeeping gene. There were no effects of diet or litter size on housekeeper expression.

Statistical analysis

Data were analyzed by ANOVA, repeated measures ANOVA, or Student's t tests for independent samples, as appropriate, using SPSS 13 software (SPSS, Inc). Subsequent comparisons between groups used Newman-Keuls procedures. Data are presented as the mean ± SEM.

Results

Dams

Gestation

There were no significant differences in maternal body weight between the 2 dietary groups during the gestation period (Figure 2A). However, HF dams had higher caloric intake during gestation than CHOW dams (P < .05) (Figure 2B).

Figure 2.

Body weight (BW) (A) and food intake (B) of dams during gestation and suckling and dams' adipose depots (C) and leptin levels (D) on PND21. Fat for dams is expressed as the weight of the sc and retroperitoneal (RP) fat pad as a percentage of BW; n = 12 per group. *, main effect of maternal HF diet on dams' food intake and leptin levels, P < .05 vs maternal CHOW diet; #, main effect of maternal HF diet on RP fat, P < .05 vs maternal CHOW diet; **, main effect of maternal HF diet on sc fat, P < .05 vs maternal CHOW diet; a–c, groups with different superscript letters differ from each other at P < .05 level.

Lactation

There were no significant differences in maternal body weight among groups during the suckling period (Figure 2A). However, HF dams had greater sc (P < .01) and retroperitoneal (P < .01) adiposity than CHOW dams on PND21 (Figure 2C and Supplemental Table 2). Despite their comparable body weight, HF dams (HF-NL) had higher caloric intake than CHOW dams (CHOW-NL) (P < .05) (Figure 2B). LL rearing increased dams' caloric intake in both CHOW (P < .05) and HF (P < .05) groups (Figure 2B). Consistent with increased adiposity, HF dams had higher plasma (P < .05) and milk (P < .05) leptin levels compared with CHOW dams on PND21 (Figure 2D). LL rearing had no effect on plasma or milk leptin levels in both CHOW and HF dams.

Neonatal offspring

Body weight, adipose depots, and peripheral metabolites

There were no significant differences in litter size (CHOW, 12.1 ± 0.7; HF, 12.5 ± 0.9), male to female ratios, or birth weight of males or females between the dietary groups at birth. By PND7, male and female pups with HF dams (HF-NL) had greater body weight (P < .01) (Figure 3, A and B) and adiposity (P < .01), higher plasma leptin levels (P < .01), and higher triglyceride and cholesterol levels (P < .05) (Table 1) than those with CHOW dams (CHOW-NL), and LL rearing reduced these measurements in both CHOW (P < .01) and HF (P < .01) pups. This effect persisted in both sexes through weaning on PND21. Besides, maternal HF diet increased pups' N-A length at weaning (P < .05) (Table 1), whereas LL rearing (HF-LL) normalized it.

Figure 3.

Body weight of male (A) and female (B) pups before weaning and glucose tolerance test (2.0 g/kg, oral gavage) results for male (C) and female (D) pups on PND23. Blood glucose was determined for 2 hours after oral administration of glucose. The integrated AUC (inset) was determined for glucose using the trapezoidal method; n = 12 per group. *, main effect of maternal HF diet, P < .05 vs maternal CHOW diet; #, main effect of LL rearing, P < .05 vs NL rearing; a–c, groups with different superscript letters differ from each other at P < .05 level. C-N, CHOW-NL; C-L, CHOW-LL; H-N, HF-NL; H-L, HF-LL.

Table 1.

Adipose Depots, Plasma Biochemical Measures, and N-A Length in Male and Female Pups Before Weaning

	Male				Female
	C-N	C-L	H-N	H-L	C-N	C-L	H-N	H-L
n	12	12	12	12	12	12	12	12
PND7
sc fat (‰ BW)	11.4 ± 1.6a	3.5 ± 1.1b	32.5 ± 3.2c	14.7 ± 1.9a	11.7 ± 1.1a	4.9 ± 1.5b	32.8 ± 3.8c	17.8 ± 1.6a
Leptin (ng/mL)	3.6 ± 0.7a	1.6 ± 0.4b	14.7 ± 2.3c	4.8 ± 1.1a	4.3 ± 0.5a	2.0 ± 0.7b	10.5 ± 1.2c	5.0 ± 0.9a
Triglyceride (mg/dL)	155 ± 15a	102 ± 13b	270 ± 29c	178 ± 27a	143 ± 10a	98 ± 12b	212 ± 18c	157 ± 17a
Cholesterol (mg/dL)	130 ± 3a	106 ± 4b	169 ± 9c	135 ± 8a	121 ± 7a	92 ± 11b	175 ± 15c	129 ± 10a
PND14
sc fat (‰ BW)	14.1 ± 2.2a	6.2 ± 0.9b	37.5 ± 4.3c	23.6 ± 1.7d	17.1 ± 1.9a	6.9 ± 0.5b	39.1 ± 3.9c	24.3 ± 2.4a
RP fat (‰ BW)	1.5 ± 0.1a	0.9 ± 0.1b	6.7 ± 1.2c	3.0 ± 0.5d	2.1 ± 0.2a	1.1 ± 0.1b	5.4 ± 1.0c	2.8 ± 0.2a
Leptin (ng/mL)	4.3 ± 0.8a	2.1 ± 0.5b	12.8 ± 1.6c	8.9 ± 1.0d	4.8 ± 0.5a	2.3 ± 0.6b	9.3 ± 1.6c	5.5 ± 0.8a
Triglyceride (mg/dL)	139 ± 12a	90 ± 15b	323 ± 46c	171 ± 19a	129 ± 6a	92 ± 8b	284 ± 21c	137 ± 15a
Cholesterol (mg/dL)	142 ± 7a	112 ± 9b	189 ± 12c	147 ± 9a	149 ± 11a	108 ± 6b	195 ± 10c	145 ± 7a
PND21
sc fat (‰ BW)	13.2 ± 0.9a	8.3 ± 1.4b	30.6 ± 3.6c	17.5 ± 1.0a	15.5 ± 1.3a	8.7 ± 0.4b	35.0 ± 3.1c	20.0 ± 2.3a
RP fat (‰ BW)	1.0 ± 0.2a	0.6 ± 0.1b	5.1 ± 1.1c	2.0 ± 0.3d	1.8 ± 0.1a	0.8 ± 0.1b	6.8 ± 0.8c	2.1 ± 0.2a
Leptin (ng/mL)	3.6 ± 0.4a	1.8 ± 0.5b	11.8 ± 1.3c	4.9 ± 0.9a	4.1 ± 0.5a	2.1 ± 0.8b	8.8 ± 1.3c	4.5 ± 0.9a
Triglyceride (mg/dL)	150 ± 17a	98 ± 18b	259 ± 19c	161 ± 21a	159 ± 12a	105 ± 11b	232 ± 15c	152 ± 10a
Cholesterol (mg/dL)	120 ± 5a	95 ± 9b	159 ± 12c	127 ± 8a	125 ± 7a	98 ± 5b	162 ± 14c	121 ± 11a
N-A length (cm)	12.3 ± 0.1a	10.2 ± 0.2b	13.2 ± 0.2c	12.2 ± 0.1a	12.1 ± 0.2a	10.3 ± 0.3b	13.1 ± 0.2c	12.3 ± 0.1a

Groups with different superscript letters differ from each other at P < 0.05 level. Data in male and female pups were analyzed separately. RP, retroperitoneal; BW, body weight; C-N, CHOW-NL; C-L, CHOW-LL; H-N, HF-NL; H-L, HF-LL.

Glucose tolerance

At weaning, there were significant effects of both maternal diet (P < .05) and litter size (P < .05) on blood glucose and overall glucose area under the curve (AUC) of male pups (Figure 3C). HF-NL males had significantly higher glucose levels at 30, 45, and 60 minutes (P < .05) and higher glucose AUC (P < .05) (Figure 3C, inset) compared with CHOW-NL group, whereas LL rearing (HF-LL group) normalized them.

Glucose tolerance data for female pups are presented in Figure 3D. HF female pups had higher baseline blood glucose compared with CHOW females (P < .05). Female pups with HF dams (HF-NL) had higher glucose levels at 45, 60, and 120 minutes (P < .05) and higher glucose AUC (P < .05) (Figure 3D, inset) compared with female pups with CHOW dams (CHOW-NL). LL rearing lowered glucose levels at the above time points and the glucose AUC in both CHOW (P < .05) and HF (P < .05) groups.

Hypothalamic leptin signaling pathways mRNA expression

We measured leptin signaling pathways (OB-Rb, STAT3, and suppressor of cytokine signaling 3 [SOCS3]) mRNA expression in hypothalamus of pups on PND7, PND14, and PND21. By PND7, male and female pups with HF dams (HF-NL group) had decreased OB-Rb (P < .05) (Figure 4, A and B) and STAT3 (P < .05) (Figure 4, E and F) mRNA expression and increased SOCS3 (P < .05) (Figure 4, I and J) mRNA expression compared with CHOW-NL group. This effect persisted in both sexes through weaning on PND21.

Figure 4.

mRNA expression of hypothalamic leptin signaling pathways (A, B, E, F, I, and J) and neuropeptides (C, D, G, H, K, and L) in male (A, C, E, G, I, and K) and female (B, D, F, H, J, and L) pups before weaning. On PND7, PND14, and PND21, pups were killed by decapitation (n = 6–8 per group). Hypothalamus was collected and assayed by quantitative RT-PCR. a–c, groups with different superscript letters differ from each other at P < .05 level.

In both male and female pups, compared with HF-NL group, LL rearing (HF-LL) increased OB-Rb mRNA expression by PND7 (P < .05) (Figure 4, A and B) and increased STAT3 (P < .05) (Figure 4, E and F) and decreased SOCS3 (P < .05) (Figure 4, I and J) mRNA expression by PND14, which normalized all these expression to CHOW-NL group before weaning. LL rearing also increased OB-Rb mRNA expression in CHOW pups by PND14 (P < .05) (Figure 4, A and B).

Hypothalamic neuropeptide and receptor mRNA expression

We measured neuropeptide and receptor (NPY, AgRP, POMC, melanin concentrating hormone, orexin, corticotropin-releasing hormone [CRH], melanocortin-4 receptor, and brain-derived neurotropic factor) mRNA expression in hypothalamus of pups on PND7, PND14, and PND21. On PND21, compared with CHOW-NL group, maternal HF diet (HF-NL) reduced NPY (P < .05) (Figure 4, C and D) and AgRP (P < .05) (Figure 4, G and H) mRNA expression in both male and female pups. There were no significant effects of maternal HF diet on other genes assessed among these groups (Figure 4, K and L, and Supplemental Table 3).

In male pups, LL rearing increased NPY mRNA expression in both CHOW (P < .05) and HF (P < .05) groups by PND14 (Figure 4C) and persisted through weaning on PND21. However, in female pups, LL rearing increased NPY expression by PND7 (P < .05) (Figure 4D). Male (Figure 4G) and female (Figure 4H) pups raised in LLs had increased AgRP mRNA expression in both CHOW (P < .05) and HF (P < .05) groups compared with those raised in NLs by PND7 and persisted through weaning on PND21. In male pups, LL rearing reduced CRH mRNA expression on PND7 (P < .05) (Figure 4K), whereas there was no effect of LL rearing on PND14 and PND21. Female pups raised in LLs had decreased CRH mRNA expression on PND7 (P < .05) and PND14 (P < .05) (Figure 4L). There were no significant effects of LL rearing on other genes assessed among these groups (Supplemental Table 3).

Leptin sensitivity

At weaning, baseline p-STAT3 level was indistinguishable among the 4 groups in both male and female pups (Figure 5). Leptin significantly increased p-STAT3 level in ARC in males and females in all 4 groups (P < .01) (Figure 5). However, leptin induced significantly less p-STAT3 in ARC of HF-NL pups compared with that induced in CHOW-NL pups (P < .05) (Figure 5, C and D), indicating that maternal HF diet reduces leptin-induced activation of STAT3 in pups at weaning. There was no significant difference in leptin-induced p-STAT3 between CHOW-NL and CHOW-LL groups, whereas pups in HF-LL group had higher p-STAT3 level after leptin challenge compared with HF-NL group (P < .05) (Figure 5, C and D), demonstrating that LL rearing improves leptin-induced activation of STAT3 in pups with maternal HF diet.

Figure 5.

Western blotting of p-STAT3 and t-STAT3 in ARC in male (A) and female (B) pups on PND21. The p-STAT3-to-t-STAT3 ratio (compared with leptin-injected pups in CHOW-NL group) was calculated in male (C) and female (D) pups. The pups were injected with saline (n = 6) or leptin (n = 6) in each group. a and b, groups with different superscript letters differ from each other at P < .05 level.

Adult offspring

Body weight, food intake, adipose depots, and leptin concentration

CHOW wean.

When male pups were weaned onto CHOW diet, there were no longer any significant effects of maternal diet on body weight at age 12 weeks, whereas offspring raised in LLs continued to have lower body weight compared with those raised in NLs (P < .05) (Table 2). When female pups were weaned onto CHOW diet, there were no differences in body weight among the groups at age 12 weeks (Table 2). There were no significant differences in food intake among the groups in either males or females (Table 2). HF-NL males had higher percentage of sc (P < .05) and retroperitoneal (P < .05) fat and higher plasma leptin levels (P < .05) compared with CHOW-NL group, whereas LL rearing (HF-LL) normalized them (Table 2). There was no difference in sc fat and plasma leptin levels among adult females (Table 2), whereas HF-NL females had higher percentage of retroperitoneal fat compared with other 3 groups (P < .05) (Table 2).

Table 2.

Body Weight, Food Intake, Adipose Depots, and Plasma Leptin Concentration in Male and Female Offspring at Age 12 Weeks

	Chow Wean				HF Wean
	C-N	C-L	H-N	H-L	C-N	C-L	H-N	H-L
n	12	12	12	12	12	12	12	12
Male
Body weight (g)	433.4 ± 10.0a	398.9 ± 9.0b	437.2 ± 7.3a	399.3 ± 10.7b	437.2 ± 11.8a	403.3 ± 11.0b	493.2 ± 7.6c	434.3 ± 9.2a
Food intake (kcal)	97.1 ± 2.9a	96.8 ± 4.2a	98.2 ± 2.9a	95.5 ± 4.8a	101.9 ± 5.0a	104.6 ± 3.1a	123.2 ± 2.8b	103.4 ± 3.5a
sc fat (‰ BW)	15.8 ± 0.7a	14.3 ± 1.7a	20.5 ± 1.1b	16.3 ± 1.0a	21.6 ± 1.6b	21.4 ± 1.1b	26.7 ± 1.3c	20.3 ± 1.7b
RP fat (‰ BW)	8.5 ± 0.9a	5.3 ± 0.5b	12.0 ± 0.7c	9.1 ± 0.8a	12.3 ± 0.2c	13.1 ± 1.1c	18.5 ± 1.4d	12.8 ± 1.0c
Leptin (ng/mL)	8.3 ± 1.3a	6.9 ± 1.8a	13.8 ± 1.7b	7.9 ± 1.2a	9.0 ± 1.5a	8.2 ± 1.2a	19.3 ± 1.8c	9.2 ± 1.5a
Female
Body weight (g)	258.8 ± 7.4a	264.2 ± 7.9a	261.3 ± 5.0a	263.9 ± 7.8a	255.8 ± 6.6a	261.0 ± 8.3a	299.7 ± 7.8b	264.5 ± 4.9a
Food intake (kcal)	61.9 ± 2.1a	63.6 ± 1.8a	61.5 ± 1.9a	63.2 ± 2.7a	64.8 ± 2.2a	65.3 ± 2.0a	78.0 ± 4.6b	63.9 ± 2.2a
sc fat (‰ BW)	19.9 ± 1.8	18.3 ± 1.9	19.3 ± 1.5	19.5 ± 2.5	22.2 ± 2.0	20.3 ± 1.7	23.0 ± 3.6	20.5 ± 1.7
RP fat (‰ BW)	16.0 ± 1.8a	11.2 ± 0.7b	22.2 ± 1.2c	18.0 ± 1.4a	15.9 ± 1.7a	15.8 ± 1.9a	37.9 ± 3.8d	24.6 ± 2.6c
Leptin (ng/mL)	5.9 ± 1.5a	6.7 ± 0.9a	8.0 ± 1.3a	7.9 ± 1.2a	6.7 ± 1.9a	7.5 ± 1.7a	13.8 ± 1.8b	7.3 ± 1.5a

Groups with different superscript letters differ from each other at P < 0.05 level. Data in male and female offspring were analyzed separately. RP, retroperitoneal; BW, body weight; C-N, CHOW-NL; C-L, CHOW-LL; H-N, HF-NL; H-L, HF-LL.

HF wean.

When male and female pups were weaned onto a HF diet on PND21, HF-NL offspring had greater body weight compared with CHOW-NL group (P < .05), whereas LL rearing (HF-LL) lowered it (P < .05) (Table 2). The greater body weight in HF-NL group could be attributed to greater food intake in this group of both male and female offspring (P < .05) (Table 2). In males, postweaning HF diet increased sc (P < .05) and retroperitoneal (P < .05) fat in all groups compared with their littermates weaned on CHOW diet, and HF-NL group continued to have greater adiposity (P < .05) than other 3 groups (Table 2). However, in females, postweaning HF diet only increased retroperitoneal fat in CHOW-LL, HF-NL, and HF-LL groups compared with CHOW weaned littermates (P < .05) (Table 2). HF-NL group had higher plasma leptin levels than other 3 groups in both adult male (P < .05) and female (P < .05) offspring (Table 2).

Glucose tolerance

There were no differences in glucose clearance at 10 weeks of age among male or female pups weaned on CHOW diet (Supplemental Figure 1, A and C).

In 10-week-old male offspring weaned onto a HF diet, HF-NL group cleared glucose more slowly relative to other 3 groups and the CHOW weaned males (P < .05) (Supplemental Figure 1B). Among female offspring that were weaned onto HF diet, offspring with maternal HF diet (HF-NL and HF-LL) cleared glucose slower than those with maternal CHOW diet (CHOW-NL and CHOW-LL) and the CHOW weaned females (P < .05) (Supplemental Figure 1D).

Hypothalamic leptin signaling pathways mRNA expression

We measured mRNA expression of components of the leptin signaling pathway (OB-Rb, STAT3, and SOCS3) in hypothalamus of offspring at age 12 weeks.

CHOW wean.

On CHOW diet, male offspring with maternal HF diet (HF-NL group) had decreased OB-Rb mRNA expression compared with CHOW-NL group (P < .05) (Figure 6A), whereas LL rearing (HF-LL group) normalized it (P < .05). There were no differences in mRNA expression of genes assessed in female pups weaned on CHOW diet (Figure 6, B, F, and J).

Figure 6.

mRNA expression of hypothalamic leptin signaling pathways (A, B, E, F, I, and J) and neuropeptides (C, D, G, H, K, and L) in male (A, C, E, G, I, and K) and female (B, D, F, H, J, and L) offspring at age 12 weeks. Male and female offspring were killed by decapitation (n = 8–10 per group). Hypothalamus was collected and assayed by quantitative RT-PCR. a and b, groups with different superscript letters differ from each other at P < .05 level.

HF wean.

Both male and female offspring with maternal HF diet (HF-NL group) had decreased OB-Rb mRNA expression compared with CHOW-NL group (P < .05) (Figure 6, A and B). LL rearing (HF-LL group) did not normalize the OB-Rb mRNA expression in adult male offspring (Figure 6A), whereas it normalized the OB-Rb expression in adult female offspring (Figure 6B). Both male and female offspring in HF-NL group had higher SOCS3 mRNA expression than other 3 groups (P < .05) (Figure 6, I and J).

Hypothalamic neuropeptide and receptor mRNA expression

We measured neuropeptide and receptor mRNA expression in hypothalamus of offspring at age 12 weeks.

CHOW wean.

When male pups were weaned on CHOW diet, CHOW-LL group had higher NPY (P < .05) (Figure 6C) and AgRP (P < .05) (Figure 6G) mRNA expression than other 3 groups. However, when female pups were weaned on CHOW diet, there were no significant differences in NPY (Figure 6D) and AgRP (Figure 6H) mRNA expression. CHOW weaned males with maternal HF diet or LL rearing or combination had lower mRNA expression of CRH than CHOW-NL group (P < .05) (Figure 6K). However, only females raised in LLs had reduced CRH mRNA expression (P < .05) (Figure 6L).

HF wean.

When weaned onto HF diet, the mRNA expression of CRH was similar with the CHOW weaned offspring in both sexes (Figure 6, K and L). There were no significant differences in NPY and AgRP mRNA expression in both male (Figure 6, C and G) and female (Figure 6, D and H) offspring.

There were no significant effects of maternal diet, LL rearing, or postweaning diet on other genes assessed among these groups (Supplemental Table 4).

Discussion

Maternal HF diet during gestation and suckling results in rat offspring with increased body weight, adiposity, leptin levels, impaired glucose tolerance, reduced leptin sensitivity at weaning (22–24), and greater susceptibility to DIO in adulthood (11). A study by Patterson et al (8) showed that LL rearing enhanced leptin sensitivity and protected selectively bred obesity-prone rats from becoming obese in a DIO paradigm. In this study, we used a different model, offspring of rat dams fed HF diet during gestation and lactation, to determine whether LL rearing can improve the metabolic phenotype, leptin sensitivity, and hypothalamic gene expression in these rats. Our data suggest that LL rearing improves hypothalamic gene expression of components of the leptin signaling pathway and appetite markers in an age- and sex-specific manner in this model.

In this experiment, we cross-fostered all the pups to get “LL size,” which is consistent with previous studies (8, 27, 28). We need to mention that both maternal HF diet and LL rearing may result in altered maternal behavior, which may affect offspring development. Thus, we will pay more attention to maternal behavior in future studies. Previous studies show that there are a number of periods in the life of an animal when it is intrinsically less responsive to stress, including immediately after birth (29, 30) and during pregnancy and lactation (31, 32). According to these findings, the pups and dams may be hyporesponsive to the stress of the manipulations in this study, including “cross-fostering” and “pup removal and replacement on PND7, PND14, and PND21.”

Consistent with our previous studies (11, 22, 23), those dams fed the HF diet consumed more calories during gestation and lactation, but body weight did not differ significantly between the dietary groups throughout the experiment. However, HF dams had more fat mass than CHOW dams. Thus, although we do not have the overall body composition data, we may speculate that these HF dams may have more fat mass and less lean mass. Besides, although the HF dams had an overall higher caloric intake, we did not see a spike in caloric consumption in the dams allocated to the HF diet. This is similar with our previous study (22) and may be due to the overall increased intake in pregnancy.

In rodents, hypothalamic neurogenesis occurs during midgestation, whereas the neural projections between different nuclei develop during early postnatal life (15, 33). Leptin plays a critical neurotrophic role during the development of the hypothalamic projections (16). Abnormal leptin levels during neonatal life, such as premature peak, excess, and deficiency of leptin, can have significant adverse effects on hypothalamic development and metabolic phenotype (17, 34). We previously found that pups with maternal HF diet had higher plasma leptin levels than those with maternal CHOW diet, and this difference begins during the first postnatal week and persists throughout the suckling period (11, 22). We found similar results in this study, again suggesting that hypothalamic development may be altered in offspring raised by HF-fed dams. Although maternal HF diet increased pups' leptin levels on PND7, PND14, and PND21, we did not see any differences in leptin levels within groups. Further studies with more time points (measuring leptin levels daily during PND1–PND21) would be needed to identify the leptin surge in this model. In this experiment, LL rearing reduced leptin levels in pups with maternal HF diet as early as PND7. Other studies also showed that reduction in plasma leptin levels occurred within 1 day of grouping pups in LLs (8, 35). This normalization of leptin levels may contribute to improved hypothalamic development in HF-LL pups.

These data call into question how LL rearing reduces leptin levels in HF pups. First, pups raised in LLs have a lower ingestion of milk due to the competition of the other littermates for the same resource, resulting in decreased growth rate and fat mass (35). The reduction in the amount of adipose tissue induces decrease in plasma leptin levels (27). Second, although neonates do make small amounts of their own leptin, they also ingest and absorb leptin up to at least PND12 from maternal milk (18–21). Our results showed that LL rearing had no effect on milk leptin levels in HF dams on PND21. Thus, reduced intake of maternal milk in pups raised in LLs may reduce their plasma leptin levels. However, the milk leptin was only measured at 1 time point so interpretations are limited. Reductions in other hormones that were not measured in this experiment, such as ghrelin, which has recently been implicated as a trophic factor in hypothalamic development (36), and fat content of maternal milk (37) may also contribute to improved hypothalamic development in HF pups raised in LLs. Dams maintained on HF diet throughout gestation and lactation spend more time nursing their pups than CHOW-fed dams during the first week after parturition (24). Thus, LL rearing may improve hypothalamic development through reduced maternal nursing time in HF-LL pups.

In this experiment, we found that maternal HF diet reduced OB-Rb and STAT3 mRNA expression and increased SOCS3 expression in hypothalamus as early as PND7, suggesting that leptin signaling may be impaired during the early postnatal period. This is consistent with our previous study, which showed that maternal HF diet reduced leptin sensitivity in pups as early as PND10 (22). This study again shows that maternal HF diet reduces pups' leptin sensitivity at weaning. Morris and Chen (38) have also shown that pups with maternal HF diet had lower hypothalamic mRNA expression of leptin receptor and STAT3 on PND1. However, Gupta et al (39) reported that maternal HF diet increased OB-Rb and STAT3 mRNA expression in fetuses on gestation day 21. The possible explanation is that Gupta's study used a HF diet period before conception to induce maternal obesity, whereas our study only used HF diet during gestation and suckling period. Maternal obesity alone could have adverse effects on offspring independent of diet during gestation and suckling, because it has significant effects on oocyte development, maturation, and embryo development (40, 41). The increased leptin levels and impaired leptin signaling pathways during early postnatal period may both contribute to altered hypothalamus development in HF pups.

We postulate that in HF-LL pups, the early reduction in plasma leptin levels was the proximate cause of the increase in hypothalamic leptin receptor number and that this led to increased leptin sensitivity in association with improved hypothalamic development. Previous studies have shown that leptin receptor expression has an inverse relationship to plasma leptin levels (42, 43). In this study, LL rearing normalized plasma leptin levels and hypothalamic leptin receptor expression as early as PND7 in HF pups. This is during the critical window (PND6–PND14) when the hypothalamic projections are mainly formed. Thus, the hypothalamic projections may be improved in HF-LL pups. Further research will be done to examine this possibility.

Previous studies have shown that hypothalamic NPY system is critical for stimulating the appetitive drive that is important during the rapid growth phase of postnatal development (44–46). NPY and AgRP expression is sensitive to feeding status in developing rats (47, 48). In this study, maternal HF diet reduced NPY and AgRP mRNA expression on PND21, which is consistent with studies by Chen et al (49) and Morris and Chen (38). However, the expression of NPY and AgRP in the whole hypothalamus cannot represent the expression of these neuropeptides in the specific nucleus. Studies with in situ hybridization (27, 50) can clarify the changes of expression of these neuropeptides throughout the ARC, and the differences in the regulation of the expression of NPY in different nuclei (eg, the ARC vs the dorsomedial hypothalamus). Future studies will be focused on this. Similar to results from Plagemann et al (35), we found that LL rearing increased NPY and AgRP expression as early as PND7, suggesting a strong appetitive drive in these pups. Further, work by López et al (27) also shows that pups raised in LLs have increased expression of NPY and AgRP in ARC at weaning. However, we did not see any significant effect of maternal HF diet or LL rearing on POMC or MC4R expression before weaning, which might suggest that POMC system is not the primary regulator of food intake during development, whereas other systems like those stimulating food intake may rather be predominant at this time (50).

The offspring from HF dams in this study were heavier than that of CHOW dams at weaning, but their body weight was not different from CHOW offspring in adulthood when weaned on CHOW diet. Only those pups that were weaned on HF diet were heavier than control pups, suggesting that offspring with maternal HF diet were more susceptible to DIO. This is in contrast with our previous studies (22, 23), in which body weight of the male offspring from HF dams was higher at weaning, and the difference persisted through adulthood after being weaned on a CHOW diet. The chow diet used previously (LabDiet 5001) contains sugars (4.22%), whereas the diet used in this study did not. Offspring with maternal HF diet were hyperphagic on the LabDiet 5001 in our previous studies, whereas those on chow diet were not in this study. Thus, the difference in body weight outcomes in this study compared with our previous study may be due to greater palatability of the CHOW diet used previously.

When weaned onto CHOW diet, adult male offspring with maternal HF diet continued to have lower hypothalamic OB-Rb expression, which is consistent with their higher plasma leptin levels and greater fat depots. However, there were no significant differences in these measurements among adult female offspring, suggesting sex-specific metabolic consequences of maternal HF diet. There are sex differences in the control of energy homeostasis, and it has been suggested that estrogen may have a protective effect against obesity (51, 52). The lower hypothalamic OB-Rb expression in adult male offspring of HF-fed dams is consistent with our previous study (23), in which central leptin sensitivity was impaired in these rats. Interestingly, CHOW weaned adult male rats in CHOW-LL group continued to have higher hypothalamic NPY and AgRP expression, suggesting a continuous strong appetitive drive in case of nutrition restriction. However, changes in mRNA expression are not necessarily representative of parallel changes in peptide release or receptor function. Research at the protein level and in specific hypothalamic nuclei is our future direction.

Finally, it is noted that LL rearing may have an impact on the adrenal/stress axis and the timing of puberty. Bulfin et al (28) demonstrated that male rats raised in LLs had attenuated hypothalamic-pituitary-adrenal axis responses to psychological stress. Castellano et al (53) showed that LL rearing delayed the timing of puberty in female rats. All these changes may affect hypothalamic neuroendocrine circuitry in adult offspring and may partially explain how LL rearing improves metabolic phenotype in offspring with maternal HF diet. The possible underlying mechanisms remain to be determined.

Summary and conclusion

Early life environment can influence the development of obesity, and a growing body of evidence suggests that maternal HF diet during gestation and suckling has long-term consequences on offspring's metabolic phenotype. We demonstrate here that LL rearing improved hypothalamic leptin signaling pathways and appetite markers in an age- and sex-specific manner in offspring with maternal HF diet. Further studies are needed to elucidate these mechanisms, including measurement at the protein level and in specific hypothalamic nuclei, assessment of leptin sensitivity in adulthood, research on peripheral tissues, such as liver and muscle, and so on. These data provide a better understanding of how maternal HF diet affects developing and adult offspring's metabolic phenotype and how LL rearing improves it.

Funding Statement

This study was supported by the National Natural Science Foundation of China Grant 31300966, the National Science Foundation for Postdoctoral Scientists of China Grant 2013M540761, the Science Foundation for Postdoctoral Scientists of Shaanxi Province, China, the Natural Science Foundation of Shaanxi Province Grant 2014JQ4124, and the Fundamental Research Funds for the Central Universities, China, Grant xjj2013053.