Prenatal nicotine exposure-induced intrauterine programming alteration increases the susceptibility of high-fat diet-induced non-alcoholic simple fatty liver in female adult offspring rats

Abstract

Previous studies have indicated that the intrauterine growth retardation (IUGR) fetus is faced with a high susceptibility to adult metabolic syndrome (MS). Non-alcoholic simple fatty liver (NAFL) is considered to be the hepatic manifestation of MS. In the present study, we evaluated the susceptibility of high-fat diet-induced NAFL in female adult IUGR offspring rats, induced by prenatal nicotine exposure, and we further explored the underlying intrauterine programming mechanism for this phenomenon. The IUGR rat model was established by prenatal exposure to nicotine (2 mg kg⁻¹ d⁻¹), the liver tissues from female fetuses and female adult offspring fed with normal or high-fat diets were collected. The female adult offspring in the nicotine-exposed group showed low birth weights and postnatal catch-up growth, as well as severe NAFL under high-fat diets. Moreover, increased gene expression involved in the hepatic insulin-like growth factor 1 (IGF1) pathway, gluconeogenesis and lipid synthesis, and decreased gene expression of lipid output accompanied with elevated serum triglyceride levels, was observed. The female fetuses in the nicotine-exposed group showed down-regulated hepatic IGF1 pathways, and also exhibited similar patterns of increased gluconeogenesis, lipid synthesis and decreased lipid output to those in the adults. The present study demonstrates the intrauterine origin of increased susceptibility to high-fat diet-induced NAFL in female offspring rats by prenatal nicotine exposure, which is most likely mediated by “two intrauterine programming”. That is, the first glucocorticoid-IGF1 axis programming induces postnatal catch-up growth, aggravates glucose and lipid metabolic disorders, and leads to an increased susceptibility to adult NAFL, while the second hepatic glucose and lipid metabolic programming enhances hepatic lipogenesis and reduces lipid oxidation and output, promoting NAFL.

---

Introduction

Metabolic syndrome (MS) is a combination of several disorders, which include hypertension, hyperglycemia, dyslipidemia, and obesity. It often leads directly to a series of diseases, such as fatty liver, diabetes, and cardiovascular diseases. Fatty liver refers to a clinical syndrome characterized as liver steatosis and excessive fat (mainly relevant to triglyceride, TG) storage. Fatty liver can be clinically divided into alcoholic fatty liver disease (ALD) and non-alcoholic fatty liver disease (NAFLD). NAFLD can be further divided into non-alcoholic simple fatty liver (NAFL), non-alcoholic steatohepatitis (NASH), and NASH-related hepatic cirrhosis and hepatocellular carcinoma.^1,2 The incidence of NAFLD has presented an increasing trend over the years. Epidemiological evidence has shown that the prevalence of NAFLD is estimated to be 20–30% of the population in Europe and the United States, and 12–24% in the Asia-Pacific region.^3,4

The developmental origin of NAFLD has recently become the topic of global interest in metabolic studies. Shown mainly as low birth weight, intrauterine growth retardation (IUGR) has been reported to be one of the key risk factors for developing NAFLD.⁵ A recent report by Cao et al. demonstrated that susceptibility to NAFLD is increased in IUGR adult sheep, induced by prenatal hypoxia.⁶ Prenatal food restriction-induced IUGR adult rats have been shown to display NAFLD and a high-risk state of inflammation.⁷ Moreover, female IUGR offspring rats affected by prenatal dexamethasone exposure presented much higher accumulation of liver lipids with high-fat diets.⁸ Furthermore, clinical evidence also indicated the correlation/relevance between IUGR and NAFLD in children. A population-based study showed that among 90 Italian children with biopsy-proven NAFLD, the prevalence of IUGR with NAFLD was approximately 4 times higher than that in all children.⁹ All these findings indicate that the IUGR fetus after birth is faced with a high susceptibility to NAFLD, and that this susceptibility will remain into adulthood and present a high prevalence of NAFLD, induced by several acquired factors such as high-fat diet, inflammation and so on.

Cigarette smoke contains a variety of compounds that may be harmful to the developing fetus. Among all smoking products, nicotine is considered one of the adverse components that perturb fetal development.¹⁰ Epidemiological investigations and animal experiments demonstrated that the risk of adult MS is increased in fetuses with IUGR by prenatal smoke or nicotine exposure.^11–13 As one type of evolution of liver lesions, NAFL is the most common NAFLD, often associated with obesity, glucose or lipid metabolic disorders, and is considered as the clinical characteristic or hepatic manifestation of MS.¹⁴ However, an association between prenatal nicotine exposure and increased susceptibility to NAFL, as well as the underlying mechanism for this, remains unclear.

Studies have suggested that hepatic de novo lipogenesis may mediate the generation of original NAFL, mainly referring to the lipogenic transcriptional factor sterol regulatory element-binding protein-1c (SREBP1c).¹⁵ Insulin-like growth factor 1 (IGF1) plays essential roles in IUGR, caused by an unfavorable intrauterine environment and postnatal catch-up growth.¹⁶ In the liver, IGF1 can induce the expression of SREBP1 and mediate hepatic lipid synthesis.¹⁷ In our previous study, we confirmed that prenatal nicotine exposure might result in IUGR and maternal glucocorticoid (GC) over-exposure.^18,19 Furthermore, we proposed an underlying mechanism for “hypothalamic–pituitary–adrenocortical (HPA) axis-associated neuroendocrine metabolic programmed alteration” for susceptibility to MS in these IUGR offspring with and without high-fat diets.^20,21 In brief, with an over-exposure to maternal GC, the development of the fetal HPA axis of the IUGR fetus was inhibited. Meanwhile, the glucose and lipid metabolism were affected, which could most likely be attributed to the increased fetal circulatory GC level.¹⁹ After birth, the IUGR offspring showed low basal activity but enhanced sensitivity of the HPA axis to chronic stress, as well as GC-dependent alterations of glucose and lipid metabolic function, especially catch-up growth and steatosis of multiple organs under high-fat diets.^20,21 Therefore, we speculate that the correlations between GC and hepatic IGF1 pathway might be associated with IUGR formation, and further contribute to the catch-up growth and development of NAFL in adult IUGR offspring rats.

Since it has been reported that female offspring with IUGR acquire more susceptibility to hepatic steatosis,⁸ in this study, we evaluated the susceptibility of high-fat diet-induced NAFL in female adult offspring rats by prenatal nicotine exposure. We further explored its intrauterine programming mechanism by observing the metabolic phenotypes, gene expression of hepatic glucose and lipid metabolic function, as well as hepatic IGF1 pathways before and after birth. This work may help to elucidate the developmental toxicity of nicotine and explain the susceptibility to adult NAFL for IUGR offspring.

Results

Adult offspring rats

Body weight and food intake. As shown in Table 1, the female offspring rats in the nicotine group showed significantly lower body weights than those in the control group at postnatal week (PW) 1 (P < 0.01). When fed a normal diet, the mean body weight of female rats was lower than that of the control at PW24 (P < 0.01), while the corresponding body weight growth rate was significantly higher than that of the control at PW24 (P < 0.01). However, when fed a high-fat diet, the mean body weight in the nicotine group was close to that of the control group and the corresponding mean body weight growth rate was significantly higher than that of the control at PW24 (P < 0.01). The food intake of the nicotine group exhibited no significant difference compared with the control either under normal or high-fat diets (data not shown).

Table 1 Effects of prenatal nicotine exposure on body weights in female adult offspring rats with normal or high-fat diets^a

Parameters	Normal diet		High-fat diet
	Control	Nicotine	Control	Nicotine
a Mean ± S.E.M., n = 8 offspring from 8 pregnant rats. **P < 0.01 vs. respective controls.
Weight (g, W1)	12.7 ± 1.2	9.2 ± 0.5**	12.9 ± 0.5	9.7 ± 0.4**
Weight (g, W24)	248.8 ± 18.1	226.7 ± 10.2**	451.1 ± 23.2	482.3 ± 12.8
Growth rate (%)	1883.1 ± 285.3	2529.3 ± 242.4**	2115.1 ± 280.3	2918.4 ± 431.7**

---

Serum glucose and TG levels. When fed with a normal diet, the serum glucose and TG levels were close to those of their respective controls for the female adult offspring rats in the nicotine group (Fig. 1A and 1B). However, both the serum glucose and TG levels in the nicotine group were much higher than their respective controls when fed with a high-fat diet (P < 0.05, P < 0.01, Fig. 1A and 1B).

Fig. 1 Effects of prenatal nicotine exposure on serum glucose (A) and triglyceride (B) levels in female adult offspring rats with normal or high-fat diets. Mean ± S.E.M., n = 8 offspring from 8 pregnant rats. *P < 0.05, **P < 0.01 vs. respective controls.

Liver histological analysis. Hematoxylin and eosin (H&E) staining showed that with a normal diet, the hepatocytes of female adult rats in the nicotine group (Fig. 2B) had no obvious pathological changes when compared to those in the control group (Fig. 2A), accompanied with a similar Kleiner score (<1) in these two groups (Fig. 2E). However, when fed with the high-fat diet, the control group displayed scattered hepatocyte microvesicular steatosis (Fig. 2C), while the nicotine group showed prevalent macrovesicular steatosis (Fig. 2D), with a significantly increased Kleiner score (>3) compared to the control (P < 0.01, Fig. 2E), indicating the formation of NAFL in female adult offspring in the nicotine group.

Fig. 2 Effects of prenatal nicotine exposure on liver histology and Klenier score in female adult offspring rats with the normal or high-fat diets (H&E, ×400). A: liver histology of female adult rats in the control group with a normal diet; B: liver histology in the nicotine group with a normal diet; C: liver histology in control group with a high-fat diet; D: liver histology in female nicotine group with a high-fat diet; E: Klenier score comparison between the control and nicotine groups with normal or high-fat diets. Mean ± S.E.M., n = 5 offspring from 8 pregnant rats. **P < 0.01 vs. respective control.

Gene expression of hepatic IGF1 signal pathway, metabolic nuclear factor and key enzymes. With a normal diet, only the expression of glucose transporter 2 (GLUT2) in the nicotine group was significantly increased, as compared with the control (P < 0.05, Fig. 3). However, under the high-fat diet, the expression of all key genes involved in the hepatic IGF1 signal pathway, including IGF1, insulin-like growth factor 1 receptor (IGF1R), insulin receptor substrate 2 (IRS2) and GLUT2, was significantly increased in the nicotine group (P < 0.05, P < 0.01, Fig. 3).

Fig. 3 Effects of prenatal nicotine exposure on the expression of hepatic insulin-like growth factor 1 (IGF1) pathway, metabolic nuclear factor and key enzymes in female adult offspring rats with the normal or high-fat diet. IGF1R: IGF1 receptor; IRS2: insulin receptor substrate 2; GLUT2: glucose transporter 2; G6Pase: glucose-6-phosphatase; SREBP1c: sterol regulatory element binding protein 1c; FoxOl: fork-head box transcription factor O1; FASN: fatty acid synthase; ACCα: acetyl-CoA carboxylase α; AMPKα, AMP-activated protein kinaseα; CPT1α: carnitine palmitoyl acyl-CoA transferase 1α; MTTP: microsomal triglyceride transfer protein. Mean ± S.E.M., n = 8 offspring from 8 pregnant rats. *P < 0.05, **P < 0.01 vs. respective controls.

We further detected the expression of hepatic metabolic nuclear factors, SREBP1c and fork-head box transcription factor O1 (FoxO1) and key enzymes related to glucose metabolism and lipid metabolism, including glucose-6-phosphatase (G6Pase), fatty acid synthase (FASN), acetyl-CoA carboxylase α (ACCα), AMP-activated protein kinase α (AMPKα), carnitine palmitoyltransferase 1 α (CPT1α) and microsomal triglyceride transfer protein (MTTP). The results indicated that the expression levels of G6Pase and FASN (involved in gluconeogenesis and lipogenesis, respectively), as well as CPT1α and MTTP (involved in lipid oxidation and output, respectively), were all significantly increased in the nicotine group, as compared with the control (P < 0.05, P < 0.01, Fig. 3). While for the nicotine group with the high-fat diet, the expression levels of G6Pase, SREBP1c, FoxO1, FASN and ACCα (involved in lipogenesis) were all significantly increased, accompanied with significant decreases in AMPKα and CPT1α expression (involved in lipid oxidation) as well as decreased MTTP expression (involved in lipid output) (P < 0.05, P < 0.01, Fig. 3).

Fetal rats

Body weight, serum glucose and TG levels. The female fetal rats in the nicotine group showed much lower body weights and higher IUGR rates than those in the control group (P < 0.01, Fig. 4A and 4B). Furthermore, the serum glucose levels were significantly increased while the TG levels were significantly decreased in the nicotine group (P < 0.01, Fig. 4C and 4D), compared to the control.

Fig. 4 Effects of prenatal nicotine exposure on body weight (A), intrauterine growth retardation (IUGR) rate (B), serum glucose (C) and triglyceride (D) levels in female fetal rats on gestational day 20. Mean ± S.E.M., n = 8 litters from 8 pregnant rats. **P < 0.01 vs. control.

Histological and hepatocellular ultrastructure analysis. H&E staining revealed that female fetal rats in the nicotine group exhibited a reduced cellularity of parenchyma cells and an increase in both multinucleated giant cells and vacuolar parenchyma cells (Fig. 5B), as compared with the control (Fig. 5A). Further ultrastructural observation demonstrated that, in the parenchyma cells from the nicotine group (Fig. 5D), not only was the number of mitochondria decreased, but also the mitochondrial structure was damaged due to swelling deformation, and the mitochondrial cristae were destroyed, as compared with the control (Fig. 5C). In addition, some degranulation of the rough endoplasmic reticula had occurred and a mass of large glycogen granules had accumulated in the cytoplasms of the hepatic parenchyma cells in the nicotine group (Fig. 5D).

Fig. 5 Effects of prenatal nicotine exposure on liver histology and ultrastructure in female fetal rats on gestational day 20. A: H&E staining of liver tissue from the control group (H&E, ×400); B: H&E staining of liver tissue from the nicotine group (H&E, ×400); C: ultrastructural observation of parenchyma cells from the control group (TEM, ×10 000); D: ultrastructural observation of parenchyma cells from the nicotine group (TEM, ×10 000). SM: swollen mitochondria; AGG: accumulated glycogen granule.

Gene expression of hepatic IGF1 signal pathway and glucose and lipid metabolic pathways. As compared with the control, the expression of key genes involved in the hepatic IGF1 signal pathway, including IGF1, IGF1R and insulin receptor (INSR), was significantly decreased in the female fetal rats (P < 0.05, Fig. 6). On the other hand, the expression levels of genes involved in gluconeogenesis or lipogenesis (G6pase, SREBP1c, FoxO1, FASN and ACCα), were significantly increased, while the expression of AMPKα, CPT1α and MTTP, which are involved in lipid oxidation or output, was significantly decreased in the nicotine group (P < 0.05, Fig. 6), compared with the control. The genes involved in adiponectin/leptin pathways were unchanged in the nicotine group.

Fig. 6 Effects of prenatal nicotine exposure on the expression of genes in the hepatic insulin-like growth factor 1 (IGF1) pathway, metabolic nuclear factor and key enzymes in fetal rats. IGFBP3: insulin-like growth factor binding protein 3; IGF1R: IGF1 receptor; INSR: insulin receptor; IRS1/2: insulin receptor substrate 1/2; GSK3β: glycogen synthase kinase 3β; G6Pase: glucose-6-phosphatase; SREBP1: sterol regulatory element binding protein 1; FoxOl: fork-head box transcription factor O1; PPARα: peroxisome proliferator-activated receptor α; HNF4: hepatocyte nuclear factor 4; FASN: fatty acid synthase; ACCα: acetyl-CoA carboxylase α; CPT1α: carnitine palmitoyl acyl-CoA transferase 1; AMPKα, AMP-activated protein kinase α; MTTP: microsomal triglyceride transfer protein; APOB: apolipoprotein B; HMGCR: HMG-CoA reductase; AdipoR2: adiponectin receptor 2; LepR: leptin receptor; JAK2: Janus kinase 2; mTOR2: mammalian target of rapamycin complex 2. Mean ± S.E.M., n = 8 litters from 8 pregnant rats. *P < 0.05, **P < 0.01 vs. control.

Discussion

An increased susceptibility to NAFL in female adult offspring rats with IUGR, induced by prenatal nicotine exposure

A large number of studies have pointed out an increased risk of adult MS in prenatal nicotine exposure-induced IUGR offspring, however, the potential role of prenatal nicotine exposure in the susceptibility to adult NAFL has not been reported. Histological examination and Kleiner score are the “gold standard” for NAFLD diagnosis.²² In the present study, we found that when fed with a normal diet, the female adult offspring in the nicotine group exhibited light steatosis, but large areas of macrovesicular steatosis were found under the high-fat diet, accompanied with a significantly higher Kleiner score (>3) than that of the control. These results indicated that prenatal nicotine exposure had an effect on stimulating the hepatic lipid accumulation and increasing the susceptibility to high-fat diet-induced NAFL in female offspring. This was consistent with the finding that liver lipid accumulated much more easily in female IUGR offspring rats with the high-fat diet.⁸

Increased susceptibility to NAFL is mediated by intrauterine programming of hepatic glucose and lipid metabolism

Hepatic lipid dysregulation and lipid accumulation are the major mechanisms of NAFL, including increased de novo lipid synthesis and lipid input, as well as reduced fatty acid oxidation and lipid output.¹⁷ In this study, severe NAFL was observed in prenatal nicotine-exposed female offspring rats with the high-fat diet. These female offspring showed enhanced hepatic lipid synthesis (SREBP1c, FoxO1, FASN and ACCα), as well as decreased lipid oxidation (AMPKα and CPT1α) and decreased output (MTTP), all of which may contribute to the formation of NAFL. We also observed up-regulated gluconeogenesis (G6Pase), which may contribute to increased blood glucose level, and the latter may further prompt hepatic lipid synthesis by inducing SREBP1c and FASN expression.²³

Hepatic ultrastructure changes and lipid accumulations have been reported in a variety of IUGR fetal models.^6,24 In this study, the hepatic ultrastructure analysis of fetal rats in the nicotine group also indicated increased vacuolar parenchyma cells, a decreased number of mitochondria accompanied with their destroyed structures, and accumulation of glycogen granules. Hepatic multiplex gene expression analysis further detected increased gluconeogenesis and lipogenesis (G6pase, SREBP1c, FoxO1, FASN and ACC), as well as decreased lipid oxidation and output (AMPKα, CPT1 and MTTP). A decreased TG level in fetal blood was also detected, which may presumably be due to impaired hepatic lipid output, while MTTP as a key molecule for very low density lipoprotein output was significantly decreased.²⁵ All these results implied that prenatal nicotine exposure could impair the structure and function of the fetal liver, resulting in enhanced hepatic gluconeogenesis and lipid synthesis, and reduced lipid oxidation and output. This was consistent with the alterations in adult offspring rats under the high-fat diet.

Intrauterine programming is the process by which the structure and function of tissues are altered permanently by negative influences acting during the intrauterine period.²⁶ The consistent alteration of hepatic glucose and lipid metabolism before and after birth revealed the possibility that this alteration originates from intrauterine programming. Programmed glucose and lipid metabolic function alteration induced by prenatal nicotine exposure could promote the formation of high-fat diet-induced NAFL in adult offspring.

Intrauterine GC-IGF1 axis programming contributes to catch-up growth and aggravates the metabolic disorders

GC is an important regulatory hormone that regulates fetal growth, development and maturity in utero, and plays a vital role in multiple metabolic processes. IGF1 plays an insulin-like growth-promoting role and is crucial for pre- and post-natal growth. In utero, the blood IGF1 level in IUGR fetuses may be modified by GC alterations, but not growth hormone.^27–29 Our previous studies showed that a prenatal nicotine exposure-induced IUGR fetus suffered maternal GC overexposure but had decreased blood GC levels when fed with a normal diet or a high-fat diet after birth.^19–21 The present study further indicates the significantly reduced body weight accompanied with a down-regulated hepatic IGF1 signal pathway in utero. Moreover, catch-up growth appeared and was more obvious under the high-fat diet after birth, accompanied with a significantly enhanced hepatic IGF1 signal pathway in the female adult offspring. The negative correlation between serum GC-IGF1 and body weight before and after birth suggested an adaptive alteration of the “intrauterine GC-IGF1 axis programming” in IUGR offspring by prenatal nicotine exposure.

IGF1 is considered as the main factor for adverse intrauterine environment-induced IUGR and postnatal catch-up growth.^16,30 Catch-up growth induced by IGF1 has been observed in IUGR offspring in many studies, which is closely related to the high risk of developing adult MS.³¹ We speculate that the IUGR and postnatal catch-up growth of offspring with prenatal nicotine exposure might be induced by the adaptive alteration of “intrauterine GC-IGF1 axis programming”. Moreover, elevated serum glucose and TG levels were also observed in adult offspring of the nicotine group, suggesting that the up-regulated hepatic IGF1 pathway could aggravate the metabolic disorders under the high-fat diet, which could promote the occurrence of NAFL.

Possible epigenetic mechanism for intrauterine programming alteration

Numerous studies have suggested that epigenetics provides a molecular link between the prenatal environment effects on genes and subsequent susceptibility to adult diseases.^32,33 In Lane and Fu's studies, the reduced expression of CPT1 in IUGR rats persisted from the fetal to the adult period, induced by site-specific changes in histone H3 acetylation, and the rats were predisposed to develop diabetes.^34,35 The expression of histone deacetylase silence signal regulating factor 1 (SIRT1) in IUGR rats consistently decreased after birth, which may cause liver steatosis by affecting the expression of several transcription factors.³⁶ Hepatic G6Pase expression was increased in newborn offspring with prenatal protein restriction, which was also mediated by epigenetic modification.²³ In addition, a growing number of studies suggested that abnormal epigenetic modifications play an important role in intrauterine programming.^37,38 For example, an epigenetic modification was considered to have participated in perinatal stress response programming of the HPA axis.^39,40 Our previous studies also indicated the possible roles of epigenetic modifications of several fetal genes, such as hippocampal 11β-hydroxysteroid dehydrogenase-2 (11β-HSD-2), adrenal steroid acute regulatory protein (StAR) and liver IGF1, in HPA axis-associated neuroendocrine metabolic programming, induced by prenatal caffeine or nicotine exposure.^41–43 Therefore, in the present study, the possible epigenetic mechanisms, such as epigenetic modifications of liver IGF1, G6Pase and SIRT1, may also involve in intrauterine programming of the GC-IGF1 axis as well as hepatic glucose and lipid metabolism, which still needs further investigation.

Dosage range of nicotine exposure and its relevance with human's daily life

Smoking can cause many health problems, however, more than one third of people in the world smoke, and approximately 20%–50% of women smoke during pregnancy, while 25%–29% keep smoking until the end of their pregnancy.^44,45 Additionally, 50% of non-smoking mothers are exposed to passive smoking during pregnancy.⁴⁵ In the present study, 2.0 mg kg⁻¹ d⁻¹ nicotine was used to establish a stable IUGR rat model, based on our previous studies. This dosage is much less than other labs’ research, where 6.0 mg kg⁻¹ d⁻¹ nicotine was infused continuously via osmotic mini pumps to pregnant rats, and this procedure mimicked maternal cigarette smoke exposure.^46,47 Therefore, by using the dose conversion between humans and rats (human : rats = 1 : 6.17),⁴⁸ 2.0 mg kg⁻¹ d⁻¹ nicotine exposure in pregnant rats is equivalent to nicotine exposure in a pregnant woman of about 70 kg who smokes 2.3 cigarettes daily (a regular cigarette contains about 10 mg of nicotine).⁴⁹ Slightly more than two cigarettes a day is a very conservative estimation of a dosage for a regular smoker. Under this dosage, the alterations of intrauterine programming of the GC-IGF1 axis as well as glucose and lipid metabolic function are obvious in female offspring rats. Therefore, the effects exerted by this dosage of nicotine can provide reference, at least partly, to estimating risks in population-based studies.

Materials and methods

Chemicals and reagents

Nicotine (CAS no 54-11-5, with a purity of 98%) was provided by Sanqiang Co, Ltd (Weifang, China). Isoflurane was purchased from Baxter Healthcare Co. (Deerfield, IL, USA). A glucose oxidase assay kit was provided by Mind Bioengineering Co., Ltd (Shanghai, China). A TG assay kit was from Sangon Biotech Co., Ltd (Shanghai, China). Reverse transcription and real-time reverse-transcription PCR (RT-PCR) kits were purchased from Takara Biotechnology Co., Ltd (Dalian, China). GeXP multiplex gene expression analysis kits were purchased from Beckman-Coulter Inc. (Fullerton, CA, USA). The oligonucleotide primers for rat RT-PCR genes (PAGE purification) and GeXP multiplex gene expression analysis (HPLC purification) were synthesized by Sangon Biotech Co., Ltd (Shanghai, China). All other chemicals and agents were of analytical grade.

Animals and treatments

Specific pathogen-free Wistar rats (weighing 200–240 g for females and 260–300 g for males) were provided by the Experimental Center of Hubei Medical Scientific Academy (no. 2009-0004, Hubei, China). Animal experiments were performed at the Center for Animal Experiment of Wuhan University (Wuhan, Hubei, P.R. China), which has been accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC International). All experimental procedures involving animals were approved by and performed in accordance with the Guidelines for the Care and Use of Laboratory Animals of the Chinese Animal Welfare Committee.

Animals were held under temperature-controlled conditions on a 12 h light-dark cycle and had ad libitum access to standard chow and tap water. The present study comprised three independent experiments: fetal rats, and adult offspring rats fed with the normal or high-fat diets. Pregnancies and nicotine treatments were the same for these three experiments, i.e., after one week of acclimation, two female rats were mated with one male rat for one night. Then the day was tagged as gestational day 0 (GD0) when sperm was observed in the vaginal smears of the two female rats and mating was thus confirmed successful. Pregnant females were then transferred to individual cages. The rat IUGR model was induced by prenatal nicotine exposure as described in previous studies.^19,20 Pregnant rats were randomly divided into two groups: the control and the nicotine group. Starting from GD11 until term delivery (GD21–22), the nicotine group was subcutaneously administered with 1.0 mg kg⁻¹ nicotine twice per day. The control group was given the same volume of saline.

In the experiment for fetal rats, 8 pregnant rats from each group were euthanized on GD20 under isoflurane anesthesia. Pregnant rats with litter sizes between 8 and 14 were considered qualified. The female fetuses were quickly removed and assessed for body weight, and IUGR was diagnosed when the body weight of a fetus was two standard deviations lower than that of the control group.⁵⁰ Fetal blood samples were collected and serum was isolated. Fetal livers were separated and collected. Samples collected from each littermate were pooled together and immediately frozen in liquid nitrogen, followed by storage at −80 °C for subsequent analyses. One fetal liver from each group was randomly selected and routinely fixed in 4% paraformaldehyde for light microscopic and transmission electron microscopic (TEM) analysis.

In the experiment for adult offspring rats fed with a normal diet, 8 pregnant rats from each group were kept until normal delivery. On postnatal day 1 (PD1), the numbers of pups were randomly normalized to 8 per litter to ensure adequate and standardized nutrition. After weaning (at PW4), one female pup per litter was randomly selected from each group and fed with a normal diet (providing 22% of its calorie intake as protein, 63% as carbohydrate, and only 5% as fat). The offspring rats were weighed weekly. The rate of body weight growth was calculated as follows: Gain rate in body weight (%) = [(body weight at PW24 − body weight at PW1)]/body weight at PW1 × 100%. At PW24, the offspring rats were anesthetized with isoflurane and decapitated in a room separate from that where the other animals were kept. Serum was prepared from the subjects’ blood and stored at −80 °C until used for measurement. The livers were dissected, and a section of 5 randomly selected livers from each group were fixed in 4% paraformaldehyde solution for light microscopic analysis, while all 8 livers from each group were immediately frozen and stored at −80 °C for gene expression analyses.

In the experiment for adult offspring rats fed with high-fat diets (containing 88.0% corn flour, 11.5% lard, and 0.5% cholesterol, which provided 18.9% kcal from protein, 61.7% kcal from carbohydrate and 19.4% kcal from fat),⁵¹ the animal treatment was the same as the adult rats fed with the normal diet except the diet intake.

Analysis for blood samples

Serum glucose and TG levels were detected by biochemical assay kits following the manufacturer's protocol.

Light microscopic analysis and Kleiner score

Liver tissue stained with H&E was processed by standard procedures in gradient alcohols and xylene, paraffin-embedded. Sections were observed and photographed with an Olympus AH-2 light microscope (Olympus, Tokyo, Japan). Five H&E sections from each group were selected and five random fields of each section were scored under the microscope.

The grading and staging of NAFLD were scored according to the system in the previous study by Kleiner et al.²² The scoring system comprises 3 histological features that are evaluated semi-quantitatively: steatosis (0–3), lobular inflammation (0–3) and hepatocellular ballooning (0–2). A Kleiner score is calculated by the sum of the steatosis, inflammation and ballooning scores. Sections with a score of no less than 5 were diagnosed with NASH, while sections with a score of greater than 3 were diagnosed with NAFL. Steatosis was graded on a 4-point scale: grade 0 for no or negligible, grade 1 for mild, grade 2 for moderate and grade 3 for severe.

TEM analysis

1 mm³ tissue blocks of liver samples were placed in a 3% glutaraldehyde–1.5% paraformaldehyde solution with 0.1 mol L⁻¹ PBS. Samples were postfixed for 1.5 h in 1% osmium tetroxide–1.5% potassium ferrocyanide solution and washed in 0.1 mol L⁻¹ PBS, dehydrated in gradient concentrations of ethanol, and embedded in Epon 618. Epoxy blocks were sliced on an ultratome (LKB-V, LKB, Stockholm, Sweden, 70 nm), stained with uranyl acetate and lead citrate, and examined with a Hitachi Hu-12A transmission electron microscope (Hitachi, Co., Tokyo, Japan). Digital images were acquired directly by a computer.

Real-time quantitative RT-PCR

Total RNA extraction, reverse transcription as well as real-time quantitative PCR analysis were performed following the procedures in our previous report.¹⁹ The target genes include: IGF1, IGF1R, IRS2, GLUT2, G6Pase, SREBP1c, FoxOl, FASN, ACCα, AMPKα, CPT1α, MTTP and housekeeping gene glyceraldehyde phosphate dehydrogenase (GAPDH). Primer sequences and annealing conditions for each gene are listed in Table S1 (ESI †). RT-PCR products were routinely gel-resolved to exclude the formation of splicing variants.

Multiple gene expression analysis

Multiplex analysis containing 23 target genes and 3 housekeeping genes was completed using a multiplex primer designed by using the GenomeLabTM eXpress Profiler software (Beckman-Coulter, Fullerton, CA). Multiplex optimization (e.g., primer validation and attenuation) was performed according to the manufacturer's instructions. Briefly, a primer pair was considered valid if only one PCR product of less than one nucleotide differed from its predicted size after being run on the GenomeLab GeXP Genetic Analysis System (Beckman-Coulter, Fullerton, CA). The 23 target genes are: insulin/IGF1 pathways (IGF1, IGFBP3, IGF1R, INSR, IRS1, IRS2, GSK3β and G6Pase), metabolic nuclear factors (SREBP1, FoxO1, PPARα and HNF4), key enzymes related to lipid metabolism (FASN, ACCα, CPT1α, AMPKα, MTTP, APOB and HMGCR) and adiponectin/leptin pathways (AdipoR2, LepR, JAK2 and mTOR2). Three housekeeping genes in this study are: GAPDH, ACTB and HPRT. The primer pairs of these genes are given in Table S2 (ESI †).

To account for the different scale of expression of each gene, the proportion of each reverse primer in the multiplex reverse transcription reaction was adjusted to obtain similar peak signals for each gene. Reverse transcription was performed on 100 ng of RNA as a template according to the GenomeLab GeXP Genetic Analysis System protocol. Real-time PCR amplification was performed by using a mixture of the forward primers, and the resulting reactions were analyzed by capillary electrophoresis on the GenomeLab GeXP with the GeXP Start kit reagents. Relative RNA expression levels were calculated against a pooled RNA standard using the GeXP Quant Tool software and normalized to GAPDH, β-actin, and HPRT expression levels. Typical spectra of fetal liver multiplex gene expression of the control and nicotine groups are shown in Fig. S1A and S1B (ESI †), respectively.

Statistical analysis

SPSS 17 (SPSS Science Inc., Chicago, Illinois) and Prism (GraphPad Software, La Jolla, CA, USA) were used for data analysis. Quantitative data were expressed as the mean ± S.E.M. and evaluated with the Independent Samples t-test. The mean weights for each litter were calculated and used for statistical analysis. For enumeration data, the body weight growth rate as well as IUGR rate was arcsine square-root transformed before t-test evaluations. The Kleiner score was evaluated with the Mann–Whitney U test. Statistical significance was defined as P < 0.05.

Conclusions

The present study systematically demonstrates the intrauterine origin of increased susceptibility to high-fat diet-induced NAFL in female offspring rats with prenatal nicotine exposure, which is most likely mediated by intrauterine programming of hepatic glucose and lipid metabolic function and the GC-IGF1 axis (Fig. 7). The first GC-IGF1 axis programming may induce postnatal catch-up growth and aggravate glucose and lipid metabolism disorders, leading to an increased susceptibility to adult NAFL, while the second hepatic glucose and lipid metabolic programming may enhance hepatic lipogenesis and reduce lipid oxidation and output, promoting the occurrence of NAFL.

Fig. 7 Intrauterine programming alteration increases the susceptibility of high-fat diet-induced non-alcoholic simple fatty liver (NAFL) in female adult offspring rats with prenatal nicotine exposure. GC: glucocorticoid; IGF1: insulin-like growth factor 1.

Conflicts of interest

The authors declare that there are no conflicts of interest.

Acknowledgements

This work was supported by the Major International Joint Research Project of the National Natural Science Foundation of China (no. 81220108026), Key Programs and General Programs of the National Natural Science Foundation of China (no. 81430089, 30830112, 81072709, 81371483), a Key grant of the Chinese Ministry of Education (no. V200801), and National Science & Technology Pillar Program of China (no. 2013BAI12B01-3).

References

1. N. M. de Alwis and C. P. Day, J. Hepatol., 2008, 48, S104–S112.
2. G. C. Farrell and C. Z. Larter, Hepatology, 2006, 43, S99–S112.
3. G. Bedogni, L. Miglioli, F. Masutti, C. Tiribelli, G. Marchesini and S. Bellentani, Hepatology, 2005, 42, 44–52.
4. S. H. Park, W. K. Jeon, S. H. Kim, H. J. Kim, D. I. Park, Y. K. Cho, I. K. Sung, C. I. Sohn, D. K. Keum and B. I. Kim, J. Gastroenterol. Hepatol., 2006, 21, 138–143.
5. A. Alisi, S. Cianfarani, M. Manco, C. Agostoni and V. Nobili, Ann. Med., 2012, 44, 29–40.
6. L. Cao, C. Mao, S. Li, Y. Zhang, J. Lv, S. Jiang and Z. Xu, Endocrinology, 2012, 153, 4955–4965.
7. T. R. Magee, G. Han, B. Cherian, O. Khorram, M. G. Ross and M. Desai, Am. J. Obstet. Gynecol., 2008, 199, 271e1–271e5.
8. D. L. Carbone, D. G. Zuloaga, R. Hiroi, C. D. Foradori, M. E. Legare and R. J. Handa, Endocrinology, 2012, 153, 295–306.
9. V. Nobili, A. Alisi, N. Panera and C. Agostoni, Pediatr. Endocrinol. Rev., 2008, 6, 241–247.
10. D. Yildiz, Toxicon, 2004, 43, 619–632.
11. H. Chen, M. A. Iglesias, V. Caruso and M. J. Morris, PLoS One, 2011, 6, e27260.
12. A. C. Holloway, D. Q. Cuu, K. M. Morrison, H. C. Gerstein and M. A. Tarnopolsky, Endocrine, 2007, 31, 254–259.
13. M. Oyama, K. Nakamura, Y. Tsuchiya and M. Yamamoto, Tohoku J. Exp. Med., 2009, 217, 67–72.
14. V. Greenfield, O. Cheung and A. J. Sanyal, Curr. Opin. Gastroenterol., 2008, 24, 320–327.
15. J. D. Horton, N. A. Shah, J. A. Warrington, N. N. Anderson, S. W. Park, M. S. Brown and J. L. Goldstein, Proc. Natl. Acad. Sci. U. S. A., 2003, 100, 12027–12032.
16. X. S. Qiu, T. T. Huang, H. Y. Deng, Z. Y. Shen, Z. Y. Ke, K. Y. Mei and F. Lai, Chin. Med. Sci. J., 2004, 19, 189–192.
17. A. Hagiwara, M. Cornu, N. Cybulski, P. Polak, C. Betz, F. Trapani, L. Terracciano, M. H. Heim, M. A. Ruegg and M. N. Hall, Cell Metab., 2012, 15, 725–738.
18. M. Chen, T. Wang, Z. X. Liao, X. L. Pan, Y. H. Feng and H. Wang, Exp. Toxicol. Pathol., 2007, 59, 245–251.
19. D. Xu, G. Liang, Y. E. Yan, W. W. He, Y. S. Liu, L. B. Chen, J. Magdalou and H. Wang, Toxicol. Lett., 2012, 209, 282–290.
20. L. Liu, F. Liu, H. Kou, B. J. Zhang, D. Xu, B. Chen, L. B. Chen, J. Magdalou and H. Wang, Toxicol. Lett., 2012, 214, 307–313.
21. D. Xu, L. P. Xia, L. Shen, Y. Y. Lei, L. Liu, L. Zhang, J. Magdalou and H. Wang, Acta Pharmacol. Sin., 2013, 34, 1526–1534.
22. D. E. Kleiner, E. M. Brunt, M. Van Natta, C. Behling, M. J. Contos, O. W. Cummings, L. D. Ferrell, Y. C. Liu, M. S. Torbenson, A. Unalp-Arida, M. Yeh, A. J. McCullough and A. J. Sanyal, Hepatology, 2005, 41, 1313–1321.
23. Y. Jia, R. Cong, R. Li, X. Yang, Q. Sun, N. Parvizi and R. Zhao, J. Nutr., 2012, 142, 1659–1665.
24. M. Yamada, D. Wolfe, G. Han, S. W. French, M. G. Ross and M. Desai, Congenital Anomalies, 2011, 51, 167–173.
25. N. Shindo, T. Fujisawa, K. Sugimoto, K. Nojima, A. Oze-Fukai, Y. Yoshikawa, X. Wang, O. Yasuda, H. Ikegami and H. Rakugi, J. Hepatol., 2010, 52, 903–912.
26. A. L. Fowden, A. J. Forhead, P. M. Coan and G. J. Burton, J. Neuroendocrinol., 2008, 20, 439–450.
27. A. L. Fowden, Placenta, 2003, 24, 803–812.
28. A. Hellstrom, C. Jansson, M. Boguszewski, R. Olegard, L. Laegreid and K. Albertsson-Wikland, Acta Paediatr., 1996, 85, 1456–1462.
29. K. A. Woods, M. van Helvoirt, K. K. Ong, A. Mohn, J. Levy, F. de Zegher and D. B. Dunger, Pediatr. Res., 2002, 51, 76–80.
30. N. Arends, L. Johnston, A. Hokken-Koelega, C. van Duijn, M. de Ridder, M. Savage and A. Clark, J. Clin. Endocrinol. Metab., 2002, 87, 2720.
31. S. Cianfarani, D. Germani and F. Branca, Arch. Dis. Child. Fetal Neonat. Ed., 1999, 81, F71–F73.
32. R. L. Jirtle and M. K. Skinner, Nat. Rev. Genet., 2007, 8, 253–262.
33. S. E. Pinney and R. A. Simmons, Trends Endocrinol. Metab., 2010, 21, 223–229.
34. R. H. Lane, D. E. Kelley, E. M. Gruetzmacher and S. U. Devaskar, Am. J. Physiol.: Regul., Integr. Comp. Physiol., 2001, 280, R183–R190.
35. Q. Fu, R. A. McKnight, X. Yu, L. Wang, C. W. Callaway and R. H. Lane, Physiol. Genomics, 2004, 20, 108–116.
36. D. Wolfe, M. Gong, G. Han, T. R. Magee, M. G. Ross and M. Desai, Am. J. Obstet. Gynecol., 2012, 308, e301–e306.
37. H. Liang, W. Xiong and Z. Zhang, Zoology, 2007, 110, 118–126.
38. D. C. Dolinoy and R. L. Jirtle, Environ. Mol. Mutagen., 2008, 49, 4–8.
39. M. J. Meaney, M. Szyf and J. R. Seckl, Trends Mol. Med., 2007, 13, 269–277.
40. I. C. Weaver, A. C. D'Alessio, S. E. Brown, I. C. Hellstrom, S. Dymov, S. Sharma, M. Szyf and M. J. Meaney, J. Neurosci., 2007, 27, 1756–1768.
41. T. Wang, M. Chen, L. Liu, H. Cheng, Y. E. Yan, Y. H. Feng and H. Wang, Toxicol. Appl. Pharmacol., 2011, 257, 328–337.
42. D. Xu, B. Zhang, G. Liang, J. Ping, H. Kou, X. Li, J. Xiong, D. Hu, L. Chen, J. Magdalou and H. Wang, PLoS One, 2012, 7, e44497.
43. Y. Tan, J. Liu, Y. Deng, H. Cao, D. Xu, F. Cu, Y. Lei, J. Magdalou, M. Wu, L. Chen and H. Wang, Toxicol. Lett., 2012, 214, 279–287.
44. S. Higgins, Curr. Opin. Obstet. Gynecol., 2002, 14, 145–151.
45. M. Contal, G. Masson, C. Boyer, C. Cazevielle and P. Mares, J. Gynecol. Obstet. Biol. Reprod, 2005, 34(Spec No 1), 3S215–3S222.
46. W. Lichtensteiger, U. Ribary, M. Schlumpf, B. Odermatt and H. R. Widmer, Prog. Brain Res., 1988, 73, 137–157.
47. L. C. Murrin, J. R. Ferrer, W. Y. Zeng and N. J. Haley, Life Sci., 1987, 40, 1699–1708.
48. S. Reagan-Shaw, M. Nihal and N. Ahmad, FASEB J., 2008, 22, 659–661.
49. K. A. Hadidi and F. I. Mohammed, Saudi Med. J., 2004, 25, 912–917.
50. G. Valsamakis, C. Kanaka-Gantenbein, A. Malamitsi-Puchner and G. Mastorakos, Ann. N. Y. Acad. Sci., 2006, 1092, 138–147.
51. L. Zhang, D. Xu, B. Zhang, Y. Liu, F. Chu, Y. Guo, J. Gong, X. Zheng, L. Chen and H. Wang, Arch. Med. Res., 2013, 44, 335–345.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4tx00092g

‡ Contributed equally to this work.

---