Role of Nuclear Factor E2 Related Factor 2 (Nrf2) in Environmental Chemical Induced Steatosis and Adipogenesis

Prevalence of non-alcoholic fatty liver disease (NAFLD) and obesity both in the young and adult population is increasing at an alarming rate globally. Along with traditional risk factors, such as diet and sedentary life style, exposure of environmental chemicals is suspected to be a risk factor for metabolic disruption. Moreover, the ‘fetal basis of adult disease’ hypothesis implicates that early life exposure to environmental chemicals is more impactful in causing lipid homeostasis disturbances than exposure in maturity. Nuclear factor E2 related factor 2 (Nrf2) is a cytoprotective transcription factor known to combat oxidative stress. The contribution of Nrf2 to other cellular functions, such as lipid homeostasis relevant to liver and adipose tissue is relatively new emerging area. The work herein assessed the role of Nrf2 in augmenting environmental chemical induced NAFLD and adipogenesis. CD-1 mice were exposed to bisphenol A (BPA) perinatally and peripubertally until age of 5 week at 25 g/kg BW/day and effect of this exposure on hepatic lipid disturbances was evaluated at pubertal (week 5) and adult age (week 39). Developmental exposure of BPA increased hepatic lipid accumulation in both young, as well as adult mice. BPA also increased Nrf2 expression in these animals at both ages. Nrf2 binding to predicted anti-oxidant response elements on the Srebp-1c promoter, a master regulator of denovo lipogenesis, elucidated a potential regulatory role of Nrf2 in lipogenesis. BPA exposure also increased Nrf2 binding to this new putative ARE by about 2 fold. Epigenetic reprogramming of chromatin upon developmental exposure of environmental xenobiotics is a potential mechanism for increasing susceptibly to diseases, such as like NALFD. BPA exposure from gestation day 8 through purberty caused hypomethylation in select regions of promoters that encode for proteins involved in lipogenesis (e.g. Srebp-1c), as well as the antioxidant response (e.g. Nrf2). The work further depicts an undescribed role for Nrf2 in the regulation of lipogenesis via regulation of Srebp-1c. In connection to the findings above, as well as other findings in our laboratory that pointed to Nrf2 having a role in promoting hepatic lipid accumulation, it was decided to evaluate whether chemicals that induce lipid accumulation can do so in association with Nrf2 activation. Next, the role of Nrf2 in lipid accumulation was further characterized using adipose tissue in presence of perfluorooctane sulfonic acid (PFOS). In rodent and human preadipocytes, PFOS exposure elevated adipogenic process by increasing expression of key regulators of adipose differentiation. PFOS-induced adipogenesis was associated with increase in Nrf2 expression and its transcriptional activity. In conclusion, the environmental contaminants, BPA and PFOS alter lipid homeostasis in liver and adipose tissue respectively. Concurrent increase in Nrf2 expression, and its transcription regulation on lipogenic mediators was demonstrated.


ABSTRACT:
Prevalence of obesity and its associated metabolic disorders including nonalcoholic fatty liver disease (NAFLD), are rising at an alarming rate, and hence need a careful search for all possible pathogenic factors contributing this epidemic. Although imbalance of energy intake and energy expenditure is considered most traditional risk factors for obesity and NAFLD, recently "environmental obesogens" hypothesis is the gaining interest in research community, which links exposure of environmental chemicals and obesity.
Terms like toxicants associated fatty liver diseases (TAFLD) are being increasingly used. The current paper has reviewed some the most recent findings of these obesogens and TAFLD contributing chemicals which are studied in rodent, in-vitro model as well as human correlation studies.
Chemical exposures during vulnerable windows in development play a critical role in etiology of diseases. This paper reviewed few developmental exposure studies, which correlated positive correlation between concentration of exposure of these xenobiotic and obesity phenotype both in human and rodent model to understand fetal basis of adult diseases. However, most of studies are only demonstrating link of environment and metabolic effect without elucidating possible mechanism of pathogenesis. Further work is required to better characterize the molecular targets responsible for environmental origin of metabolic dysfunction and disturbances of lipid homeostasis.

Obesity.
Globally the rising prevalence of obesity and its associated diseases is major public health concern in adult, children as well as adolescent population. According to results from the 2011-2012 National Health and Nutrition Examination Survey (NHANES), more than one-third (34.9%) of adults were obese in United States (1). Childhood obesity prevalence is also increasing at an alarming rate. It is estimated that 20.5% of adolescents (12-19 years of age) and 17.7% children (6-11 years age) are obese (2). Obesity is characterized by structural and functional changes to and distribution of white adipose tissue (WAT) in body. Adipose tissue plays a prominent role of balancing energy status of the body by storing excess of fat in form of triglyceride. Initially, excess fat is stored in mature adipocyte, and because of its plasticity it can expand by undergoing hypertrophy. Along with hypertrophy, hyperplasia of adipocytes, which means increasing number of fat cells through differentiation of stem cells to adipocyte lineage, can also contribute for disturbances of normal lipogenic and lipolysis process of adipose tissue (3). Apart from considering WAT as fuel storage depot, is also recognized as an active endocrine organ because of secretion of hormones and cytokines that play crucial role in maintaining of metabolic homeostasis like appetite, inflammation and insulin sensitivity (4). Adipocyte biology is significant to pathogenesis of obesity and hence maintenance of normal adipogenesis and mature adipocyte is very significant. Mature adipocytes form from its precursor mesenchymal stem cells through adipogenesis process, which involves very complex transcriptional cascade regulating programs. Multiple transcription regulators, including CCAAT/ enhancer-binding proteins (Cebp , Cebp ), and peroxisome proliferator-activated receptor (Ppar-), are involved in inducing adipogenic programming (5).

Non-alcoholic Fatty Liver Disease (NAFLD).
Obesity associated disturbance of adipogenesis process has deleterious consequences on insulin resistance, which can contribute to pathogenesis of metabolic syndrome comprising type-2 diabetes, dyslipidemia, atherosclerosis, hypertension and NAFLD. Prevalence of suspected NAFLD increased from 3.9% in 1988-1994 to 10.7% in 2007-2010. Globally, 10-39% and in United States 20% of general adult population is suspected to have NAFLD (6). About 3-17% of children in the United States are detected with fatty liver diseases. As expected, nearly 75% adults and 38% children from overweight/ obese BMI have NAFLD (7,8). NAFLD is a broad spectrum liver disease ranging from preliminary hepatic steatosis to more aggressive form of nonalcoholic steatohepatitis (NASH), which in turn may lead to cirrhosis and hepatocellular carcinoma (HCC). Although pathogenesis of NAFLD is not well understood, globally accepted "two-hit hypothesis" explain the progression of disease.
"First hit" is associated accumulation of lipids, especially triglycerides, in the hepatocytes cytoplasm, contributing to 5% or more of the liver weight. Although steatosis is asymptomatic and reversible, this chronic fat accumulation in liver triggers "second hit" covering various inflammatory cytokines and oxidative stress. Steatotic liver is more susceptible for lowgrade inflammation and disease progression like steatohepatitis. Dyslipidemia and lipid deposition in liver are not only involved in first hit of NAFLD. Non-esterified free fatty acids not only contribute to the steatosis, but also in hastening the progression to NASH, challenging two-hit hypothesis (9).
Recent research also points towards "multi-hit theory" as opposed to two hits for pathogenesis of NAFLD-reviewed by (10). Considering the complexity of steatogenesis, necroinflammation of the hepatocytes, oxidative stress and fibrogenesis, multi-hit theory is being considered more thorough approach towards pathogenesis of NAFLD.
Because of this increasing prevalence, numerous studies are focusing on early diagnosis and treatment for NAFLD. There is no specific biochemical marker or serological test for the diagnosis of NAFLD. Liver biopsy is the gold standard for diagnosis and staging of NAFLD, particularly for the diagnosis of NASH. Due to the invasive nature of liver biopsy and risk associated with procedure, it cannot be extensively used for large-scale population. Different imaging techniques like ultrasonography (US), computed tomography (CT), magnetic resonance imaging and magnetic resonance spectroscopy have been approved as noninvasive alternative methods to detect hepatic steatosis (11,12). Association of NAFLD with cardiovascular disease, type 2 diabetes, cirrhosis, and hepatocellular carcinoma along with difficulty in diagnosis and lack of exact therapeutic intervention for these conditions make NAFLD a serious public health concern.
There are multiple sources that contribute to the hepatic fatty acid pool. 1) Dietary TG that reach the liver as chylomicron particles from the intestine; 2) Denovo synthesis in the liver under a transcriptionally regulated process; 3) Fatty acid influx into the liver from lipolysis of adipose tissue in obese and insulin-resistant state ; 4) diminished export of lipids from the liver in very-lowdensity lipoproteins; and 5) reduced oxidation of fatty acids. The accumulated fatty acids are stored in liver in the form of TG, eventually contributing to steatosis.
NAFLD pathogenesis may involve disturbances of proliferator-activated receptor alpha (Ppar ) and/ or sterol regulatory element binding protein-1c (Srebp-1c), transcription factors that control enzymes responsible for oxidation and synthesis of fatty acids respectively. The regulation of lipid metabolism in liver is usually integrated with adipose tissue. Both of these organs play a very critical role in trafficking and handling of lipid based on energy status of the body (13). Hepatic triglyceride (TG) can transport to adipose tissues in\ very low-density lipoprotein where TG can store. Under hormonal regulation, fatty acids are released from adipose tissue and are transported to liver for oxidation. The control of whole body lipid homeostasis is mainly depending on efficient regulation of this cycle (14). Therefore, disturbances at one organ lipid homeostasis can have an impact on the other organ. This is why NAFLD prevalence is high in obese population, and it remains major cause of mortality and morbidity in obese people.

Environmental chemicals and adipose tissue lipid alterations.
Imbalances of energy intake and energy expenditure either because of sedentary lifestyle or some genetic predisposition are well-recognized risk factors for obesity and NAFLD. However, increases in rates of health concerns like metabolic syndrome and co-ordinate increase in synthetic chemical production, toxic burden in our air, food and water since last few of decades, it seems prudent approach to study the effects of the environmental chemicals that are found in the majority of people, in increasing obesity epidemic. It has generated significant interest of studying "nontraditional" risk factors (e.g., environmental chemicals, stress, micronutrients, gut microbiome) to the etiology of these health conditions. Moreover, because of lipophilic property of most environmental chemicals like endocrine disruptors (EDC),, adipocytes are an obvious target organ for their biological effects (15). Many bioactive EDCs predicted to bioaccumulate in lipid droplets of mature adipocyte that can be released into the systemic circulation slowly, which may results into chronic exposure of chemical (16). Human get exposed to these chemicals on routine basis either through occupational or non-occupational means. Food, water, air contamination is the most common route of exposure. Some chemicals get into food by leaching out of products that food and beverages are stored. A prime example is BPA, which is used in polycarbonate reusable food and beverage storage containers and the resin lining of cans, which can lead to substantial levels of human exposure. PFAs are stain repellent, and also used in firefighting foams, whereas PBDEs are primarily used in flame retardant applications for manufacturing household rugs, carpets, non-stick utensils and so on (18)(19)(20)(21). Although actual acute exposure of chemicals may be at nanomolar or lower concentrations, persistent exposure through these routes cannot be ignored.
In order to assess environmental xenobiotic induced obesogenic effects, numerous in-vivo as well as in-vitro approaches have been employed. Many population epidemiological studies have identified intriguing link between environmental chemical levels in biological fluids and metabolic syndrome parameters. Importantly, although these human correlation studies provide valuable information in predicting the risk associated with xenobiotic exposure, precise mechanisms for underlying effects remain largely unknown. Table 1 summarizes representative human correlation studies which associate serum, urinary, or tissue specific levels of xenobiotic with obesity parameters including BMI, waist circumference, body weight, plasma insulin, leptin levels, systemic inflammation and so on (22)(23)(24)(25)(26)(27)(28)(29)(30)(31). Majority of these data are gathered from participants of the National Health and Nutrition Examination Surveys   (44,45).
Although lots of chemicals have been suspected for their ability to induce fatty liver independently of obesity, very few have been examined in-vivo. Table 5 enlists chemicals which have been tested in in-vitro as well as in-vivo, and human correlation studies for their NAFLD inducing abilities.
As mentioned earlier, diagnosis of NAFLD is difficult and non-invasive methods are not as reliable as biopsy, prevalence of NAFLD may be higher Obesity and NAFLD increase risk of cardiometabolic diseases, as evident from overlapping population prevalence. Moreover treatment of NAFLD is also a concern, as currently there are barely any therapies available in market targeting specifically on NAFLD/ NASH in people. Ongoing pharmaceutical research on these therapies is also constantly challenged by difficulty of diagnosis and unreliable non-invasive tools (71).
Based on research from government agencies, and also academic findings, few of the manufacturing chemicals have been banned for use in some countries, however, majority of chemicals listed in this review remain in industrial use. Also, the replacement chemicals being investigated for manufacturing use seem to have similar structural and chemical profiles, indicating real possibility of these new chemicals also not being risk-free.
Another issue with few such environmental chemicals is their prolonged halflife in nature as well as inside the living system. PFAs exemplify bioaccumulating chemicals that are not degraded to inactive form for years in the food chain (72). Overall, rising awareness about environmental chemical exposure contribution to the epidemic of obesity/ NAFLD seems imperative.
More studies are necessary for thorough mechanistic investigation of these environmental origins of the childhood and adulthood diseases, so that preventive or therapeutic approaches can be guided.

23.
Trasande L, Attina TM, Blustein J. Association between urinary bisphenol A concentration and obesity prevalence in children and adolescents. Jama 2012;308:1113-1121.    food containers, metal cans, and thermal receipts has been recently associated with NAFLD (6, 7). BPA exposure is ubiquitous, and it was detected in more than 90% of the urine samples from the US population (8).
BPA has also been detected in human breast milk, amniotic fluid and cord blood samples, which demonstrates the potential for fetal and neonatal exposure (9)(10)(11). A positive correlation between urinary BPA levels and obesity risk has been reported in adults and children, especially in adolescent females (12). Recently, increased markers for NAFLD were positively associated with urine BPA levels in a small cohort of children (13). Therefore, there is growing health concern for early-life BPA exposure that has not been fully addressed (14). In rodents, perinatal BPA exposure increased hepatic lipid content and lipogenic gene expression, along with imbalance in adipokine levels and insulin signaling disturbances in female offspring during adolescence and adulthood (15)(16)(17). The mechanisms by which BPA induces fat accumulation are largely undescribed, but likely involve induction of lipogenesis through both direct (6) and epigenetic mechanisms (18).
Epigenetic mechanisms, such as DNA methylation and histone modifications contribute to NAFLD (19)(20)(21). DNA methylation patterns and lipogenic gene expression has been correlated in liver biopsy tissues from NAFLD patients (22). Induction of fatty liver via a high-fat, high-sucrose diet was associated with hypomethylation at some CpG sites and increased transcripts for lipogenic proteins, such as sterol regulatory element binding protein 1c (Srebp-1c) and glycerol-3-phosphate dehydrogenase (Gpat) (23). Similarly, hypomethylation at specific CpG sites of the Fatty acid synthase (Fas) promoter was induced by short-term or long-term high fat diet feeding and correlated with up-regulation of Fas expression (24,25). The mechanism by which induction of Srebp1-c in rodents with early-life exposure to BPA has not been elucidated, and it is hypothesized that it could also occur through promoter hypomethylation (7).
Nuclear factor E2 related factor 2 (Nrf2) functions primarily as an antioxidant defense system of the cell. Recently, Nrf2 has been linked to adipose differentiation and lipid homeostasis-reviewed by (26 hypomethylation at CpG sites in the promoter regions. Moreover, we described increased Nrf2 recruitment to the Srebp-1c promoter in livers of BPA exposed mice. alone (50% DMSO in water) and BPA (25 µg/kg bw/day). Care was taken to limit BPA contamination through housing or feeding. These exposures to dams were continued until weaning age of pups, which is postnatal day 21 (PND 21) and pups exposed to BPA through lactation. After weaning age, exposure of BPA to pups continued through drinking water until PND 32 (week 5). Livers and serum were collected from female pups at week 5 and at week 39 and stored at -80°C until analysis (n=10 or 5/ group for week 5 and week 39 respectively).

Triglyceride quantification.
Total lipids were extracted from liver tissue by method previously described (28). Briefly, liver tissue (50mg) was homogenized in 1 ml of phosphate buffer saline (PBS) and extracted with chloroform-methanol mixture (2:1). Lipid extracts were vacuum-dried by using a Speedvac (Thermo Scientific) at 45°C. The lipid residue was re-suspended in 1% Triton X-100 in 100% ethanol. Triglyceride quantification was performed using a commercially available kit from Pointe Scientific Inc. (Canton, MI).

Determination of relative protein expression by western blot. 50mg
of liver tissue was homogenized in 1ml of RIPA buffer using a Dounce homogenizer. The lysate was centrifuged a 12,000 rpm for 15 min at 4°C and the resulting supernatant was collected as a total protein extract. Nuclear proteins were extracted from liver tissue using a NE-PER® kit (Thermo The resulting protein bands were visualized by autoradiography films, which were quantified using Quantity One software (Biorad, Hercules, CA). Specific information about the source, dilution, type, and molecular weight of primary and secondary antibodies is detailed in Supplementary Table S2.  S3). Promoter sequences were identified using UCSC genome browser.

Chromatin Immunoprecipitation
Results are represented as agarose gel scans (end-point PCR) as well as fold enrichment (qPCR).

Statistical Analysis. Control and treated groups were analyzed by
Student's t-test. Asterisk * indicates statistically significant difference between the control and BPA treated group (p<0.05).  increased Ppar-γ levels in nuclear fractions by 300% compared to controls.

Effects of perinatal peripubertal bisphenol
BPA increased pSrebp1 protein levels to two-fold compared to controls. At week 39, significant induction in protein expression of Ppar-γ and Srebp1c was noted (Fig 3A, 3B). At week 5, BPA increased Fas, Acc, and pAcc protein expression by more than 135%, 200% and 135% respectively. At week 39, BPA25 increased Fas and pAcc protein expression by 50 and 75%, but did not significantly increase Acc levels (Fig. 4A, 4B).

PNPP BPA exposure enhanced Nrf2 and Nrf2-dependent protein
expression. Messenger RNA and protein expression for Nrf2, and its target enzymes Gclc were determined (Fig. 5). Protein expression of nuclear Nrf2 and total Gclc were increased by 30% or more with BPA treatment in both week 5 and week 39 livers.
To examine whether BPA exposure was associated with increased oxidative stress, reduced GSH was measured in liver. At week 39, hepatic GSH concentration was similar between controls and BPA exposed group (Supplementary figure S2), indicating a lack of significant oxidative stress in liver. The Environmental Protection Agency and US Food and Drug Administration published tolerable daily intake (TDI) value for BPA around 50μg/kg, and the exposure level chosen in current study is lower than TDI. More importantly, published studies suggest that dose of 400μg/kg BW/day in mice resulted in serum levels (0.5ng/ml) of unconjugated BPA much lower than amount noted in human serum after environmental exposure (2 ng/ml) (33). In addition to dose, timing of exposure to potential endocrine disruptors like BPA is also a critical determinant for disease susceptibility (6,34). Effects of fetal or neonatal exposure of BPA are hypothesized to be as critical as adult exposure (6,35). As based on 'fetal plasticity' theory any effects that affect embryo can remain persistent for life. In our model, female mice were exposed to BPA during gestation and lactation. Overall, it was observed in the study herein that BPA induced hypomethylation at some sites in the liver genome that persisted (e.g. Srebp1c -1325 to -1456; Fas -654 to -832, -306 to -472; Nrf2 -1059 to -1168), whereas methylation in other sites in the genome differed at weeks 5 and 39. This data suggests the potential that some sites might be persistently "marked", whereas others are more "plastic" with the ability to be more dynamically regulated by methylation-regulating enzymes or changes in methyl donor content. This observation is highly relevant to the field of epigenetics and supports other observations of plasticity in methylation (36).

PNPP BPA exposure was associated with hypomethylation in
Effects at low dose, as well as exposure time dependent effects support the notion of BPA being a "critical window of exposure" compound.
In the current study, as well as published findings (7), liver weights remained unchanged with BPA exposure irrespective of route and time of administration in both mice and rats. Interestingly, BPA exposed rodents exhibited evident hepatic lipid accumulation among several studies (6,7). A major source of liver lipids besides dietary consumption is de novo synthesis (37). Our observations herein, as well as ones from the latter studies using rodents are supported to some degree by recent finding published associated NAFLD with BPA exposure in children (13). BPA-mediated induction of lipid synthesis enzyme expression has been demonstrated (6), however the upstream mechanism by which BPA imparts this upregulation remains unknown. One potential mechanism could be via upregulation of Nrf2 expression and regulation of lipogenic genes as previously described (28,40). Although most commonly studied for countering oxidative stress, Nrf2 can be upstream regulator of adipogenesis through Ppar-γ and Cebp-β transcriptional regulation in rodent pre-adipocytes (29,30).         12hrs. Luciferase activity was measured using a commercial kit.    Messenger RNA was converted to cDNA and subsequently quantified using real time polymerase chain reaction (RT-PCR) using primers specific for peroxisome proliferator activated receptor alpha (Ppar-α), cytochrome P450 4a10 (Cyp4a10), carnitine palmitoyltransferase 1a (Cpt1a). Raw data was normalized to respective control expression, and statistical differences between the groups were analyzed by Student's t-test. Asterisk (*) represents significant difference in expression between BPA treated and control animals (p≤0.05).

ABSTRACT:
PFOS is a chemical of nearly ubiquitous exposure in humans. Recent studies have associated PFOS exposure to adipose tissue-related effects.
The present study was to determine whether PFOS alters the process of adipogenesis in mouse and human preadipocytes and regulates insulinstimulated glucose uptake. In murine-derived 3T3-L1 preadipocytes, PFOS Adipogenesis is a process involving sequential coordinated gene induction (Rosen and Spiegelman, 2000). CCAAT/enhancer-binding protein (Cebp) and Cebp promoting Ppar-and Cebp-signaling, and induction of adipogenesis that increases lipid accumulation in 3T3-L1 preadipocytes. Moreover, proadipogenic effects were observed in human visceral preadipocytes exposed to PFOS.    triglyceride content in 3T3-L1 adipocytes by more than 20% above control, but this effect was not observed with the relatively lower PFOS concentrations (1-100 nM) ( Figure 2B). The data suggest that PFOS has the potential to potentiate induction of mouse preadipocyte differentiation to mature adipocytes and promote lipid accumulation.
Additionally, the mRNA levels of Nrf2 and two target genes, Nqo1 and Gclc were determined. After 3 days of induction to adipocytes (Day 3), PFOS significantly increased Nrf2, Nqo1 and Gclc mRNA levels in 3T3-L1 adipocytes than vehicle-treated group by more than 15-fold, suggesting that PFOS has the potential to activate Nrf2 signaling in preadipocytes ( Figure 3B).

PFOS promotes insulin-stimulated glucose uptake in 3T3-L1
adipocytes. In order to assess the metabolic consequence of PFOS treatment in 3T3-L1 preadipocytes, insulin-stimulated glucose uptake were monitored in vehicle-and PFOS-treated adipocytes. After 3T3-L1 preadipocytes were differentiated to adipocytes for 10 days, cells were exposed to PFOS for 5 hrs, then 2-NBDG was added, and the fluorescence activity was monitored. There was no significant difference between the low dose PFOS treatments (1 and 5 M) and vehicle-treated group.
However, insulin-stimulated glucose uptake was significantly higher in  (Figure 4C). significantly increased staining in mature adipocytes ( Figure 6A).

Adipogenic gene expression and
Furthermore, the stain was extracted and quantified spectrophotometrically. Figure 6B illustrates that PFOS increased staining in adipocytes by 48% at 5 M and 40% at 50 M, suggesting PFOS may increase adipogenesis in human visceral preadipocytes, contribute to enhance lipid accumulation in PFOS-treated group (Figure 6B). suggesting PFOS increased ARE binding activity ( Figure 7A).

PFOS increases Antioxidant
Furthermore, to determine whether the increased ARE binding was via the increased Nrf2 binding in PFOS treatment, Chip assay was carried out. Figure 7B and 7C illustrate that two days into differentiation, PFOS increased Nrf2 binding to ARE sites in mouse Nqo1 promoter (by 31%), suggesting that Nrf2 binding was increased after PFOS treatment during the differentiation process (Figure 7B, 7C).

DISCUSSION:
PFOS has been manufactured for over 60 years. Epidemiological studies and recent research with PFOS primarily focuses on widespread exposure The activation of Ppar-by PFOS suggested that PFOS has Pparindependent mechanism, which is in agreement with the previous study (Rosen et al., 2010 which the pregnant rats were given with PFOS (0.5 or 1.5 mg/kg BW/day) from gestation day 0 to postnatal day 21. The pups displayed impaired glucose tolerance and enhanced insulin resistance index, suggesting PFOS disrupted insulin signaling in integrate animal study (Lv et al., 2013).
In the current in vitro study, we demonstrated that PFOS increased insulininduced glucose uptake and increased gene expression related to insulin signaling. One possible reason is that PFOS increased adipogenesis in 3T3-L1 preadipocytes and enhanced adipogenesis increases capacity for glucose uptake (Nugent et al., 2001). Also, enhanced Ppar-and Glut4 expression will help to promote glucose uptake, as well as improve insulin signaling and insulin-response activity. Our previous study reported that enhanced Nrf2 activity by Keap1-KD increased glucose tolerance, and increased insulin-stimulated Akt phosphorylation without Glut4 expression change , suggesting activation Nrf2 can promote insulin signaling, contribute to positively regulate glucose uptake in differentiated 3T3-L1 preadipocytes.
The study herein reported that PFOS could induce Nrf2 activation in preadipocytes. PFOS administration has been shown to increase ROS production, which induces oxidative stress and activate Nrf2 signaling (Qian et al., 2010). ARE consensus elements are the typical transcriptional factor binding sites, which are described to induce the antioxidant gene expression for Nrf2 target genes, such as Ho1, Gclc, Nqo1, and Multiple resistance-associated proteins to provide a protective role against oxidative and cytochemical stress (Nguyen et al., 2009 Overall, this study reported that novel effects of PFOS in inducing Ppar-and Cebp-expression and adipogenesis, via enhancing ARE binding activity and Nrf2 signaling in preadipocytes ( Figure 7D).
Additionally, PFOS increased insulin-stimulated glucose uptake and increased gene expression related with insulin signaling. This study points out the potential roles of PFOS promoting adipose tissue differentiation and the related metabolic conditions of obesity consequentially.      Relative lipid content (%) was displayed using differentiated media APPENDIX 1: Objective: Studying the hepatic lipid accumulation with short-term high fat diet feeding in Keap1 knockdown (Keap1 KD) mice Experimental design: Nine week old WT and Keap1 KD mice were fed 10% or 60% kCal fat diet for 5 weeks, and then euthanized to collect organs.
Tissue and serum lipid levels (TG, FFA, LDL/HDL) were detected by commercial kits. For all the figures, differences between the groups were calculated by two-way ANOVA followed by Tukey's multiple comparison tests.
Asterisk * represents significant difference between WT and Keap1 KD mice, whereas # represents significant difference between SD and HFD groups.

Synopsis:
As elaborated in all the manuscripts in this thesis, elucidating role of Nrf2 in lipid accumulation is one of the primary goals of my research. In this project, mice having diminished expression of Keap1 protein, and subsequently constitutive activation of Nrf2 were utilized. After 5 weeks on HFD, these mice appeared to be protected from lipid accumulation in liver.
Keap1 KD mice fed SD as well as HFD contained less hepatic lipids as compared to WT controls (App 1 fig 2). TG content of the white adipose tissue, on the contrary, was higher in Keap1 KD mice relative to controls, on both diets. On both diets, Keap1 KD mice demonstrated increasing trend in LDL/HDL ratio (App 1 fig 2). Lipogenic target mRNA and protein expression was decreased in Keap1 KD mice on SD, and the expression was further diminished in HFD fed groups (App 1 fig 4 and 5). Fatty acid oxidation target expression data was inconclusive. Keap1 KD mice demonstrated high pAmpk 124 APPENDIX 2: Objective: Studying the effect of perinatal peripubertal (PNPP) effect of bisphenol A (BPA) on hepatic lipid accumulation and its mechanism in female mice Experimental design: Pregnant mice were implanted with osmotic pumps containing 250μg/kg BPA or vehicle. The pups were exposed to BPA after birth through milk, and until week 5 of age through drinking water. One set of animals was euthanized at week 5, whereas another set was euthanized at week 39, without any further BPA exposure beyond week 5. The liver tissue was used for lipid accumulation (TG, FFA content), mRNA expression (RT-PCR), protein expression (western blot), promoter methylation (methylated DNA immunoprecipitation), and binding assay (chromatin immunoprecipitation). This data is originally a cohort from studies described in 'manuscript 2', hence all the methods have been described in detail.
Differences between the groups were checked with Student's t-test, asterisk * represents significant difference between BPA treated and vehicle group.

Synopsis:
The treatment group represented here, BPA 250μg/kg is another treatment cohort from the same study that is presented in 'manuscript 2' of this thesis. Simultaneous study with BPA 25μg/kg and 250μg/kg were performed, and data was compared with vehicle group. Hepatic lipid accumulation (liver TG and oil red o staining) was significantly increased with BPA 25μg/kg but not with BPA 250μg/kg. This is the reason study with this exposure level of BPA was not included in original manuscript. Although phenotypic increase in hepatic lipids was not evident in BPA 250μg/kg (App 2 fig 1), lipogenic gene expression (App 2 fig 2), protein expression (App 2 fig 3 and 4), Nrf2 signaling (App 2 fig 5), and lipogenic promoter hypomethylation (App 2 fig 6) was altered similar to BPA 25μg/kg exposure. Nrf2 binding on the Srebp-1c promoter was not enhanced by BPA 250μg/kg exposure (App 2 fig 7), indicating transcriptional mechanism of Nrf2 mediated lipid accumulation in liver is evident only in BPA 25μg/kg exposure, but not in BPA 250μg/kg.