Application of In Vitro Models for Studying the Mechanisms Underlying the Obesogenic Action of Endocrine-Disrupting Chemicals (EDCs) as Food Contaminants—A Review

Obesogenic endocrine-disrupting chemicals (EDCs) belong to the group of environmental contaminants, which can adversely affect human health. A growing body of evidence supports that chronic exposure to EDCs can contribute to a rapid increase in obesity among adults and children, especially in wealthy industrialized countries with a high production of widely used industrial chemicals such as plasticizers (bisphenols and phthalates), parabens, flame retardants, and pesticides. The main source of human exposure to obesogenic EDCs is through diet, particularly with the consumption of contaminated food such as meat, fish, fruit, vegetables, milk, and dairy products. EDCs can promote obesity by stimulating adipo- and lipogenesis of target cells such as adipocytes and hepatocytes, disrupting glucose metabolism and insulin secretion, and impacting hormonal appetite/satiety regulation. In vitro models still play an essential role in investigating potential environmental obesogens. The review aimed to provide information on currently available two-dimensional (2D) in vitro animal and human cell models applied for studying the mechanisms of obesogenic action of various industrial chemicals such as food contaminants. The advantages and limitations of in vitro models representing the crucial endocrine tissue (adipose tissue) and organs (liver and pancreas) involved in the etiology of obesity and metabolic diseases, which are applied to evaluate the effects of obesogenic EDCs and their disruption activity, were thoroughly and critically discussed.


Introduction
According to the World Health Organization (WHO), obesity is one of the top ten threats to human health [1]. The research of the Statistical Office of the European Union (Eurostat) shows that 52.7% of the adult population of the European Union (EU) was overweight in 2019, of which approximately 17% were obese [2]. Moreover, in the EU, every eighth child aged 7-8 is obese [3]. Nowadays, there are far more people in the world overweight or classified as obese than malnourished and obesity has become a serious health problem in both developed and developing countries [1]. Obesity is associated with comorbidities such as the increased risk of cardiovascular disease, insulin resistance, Type 2 diabetes mellitus (T2DM), hypertension, as well as non-alcoholic fatty liver disease (NAFLD), and hormone-sensitive cancers [4]. Obesity is characterized by an imbalance between energy intake and total energy expenditure resulting in increased The dramatically increasing production of highly processed foods and evidence of health risks linked with food chemical contamination has led to the development of different chemical analogous and alternatives [52][53][54][55][56][57][58]. Nonetheless, results of several studies have suggested that some of the biological activities of chemical analogous are of a similar or even higher magnitude in comparison to initial compounds [58][59][60][61] but mechanistic aspects underlying the obesogenic effects remain largely unrecognized [62]. Therefore, in vitro models still play an essential role in identifying environmental obesogens and understanding obesogenic mechanisms for further insight into the in vivo action of these chemicals linked with the risk of developing various chronic diseases including obesity and metabolic disorders. Moreover, in vitro models provide a rapid approach to investigating hazards of exposure to EDCs and their toxicity potential, while reducing or eliminating the need for animal testing [63].
Hence, this review aimed to provide information from peer-reviewed literature on currently available 2D in vitro animal and human cell models applied for studying the mechanisms of action of the various obesogenic EDCs as food contaminants. We focused mainly on in vitro models representing the main tissues and organs involved in the etiology of obesity and metabolic diseases, such as adipose tissue, liver, and endocrine pancreas. These in vitro models are currently used to evaluate the chemical toxicity of EDCs and their metabolic disruption activity. We believe that this comprehensive insight could help scientists choose the appropriate in vitro model to study the mechanisms of action of environmental obesogens associated with the risk of human obesity and its comorbidities.

Adipogenesis and Obesogenic Action of Endocrine-Disrupting Chemicals (EDCs)
Adipose tissue (AT) is a metabolic active tissue located in the body under the skin (subcutaneous adipose tissue (SAT)) as well as around the internal organs (visceral adipose tissue (VAT)) [64]. AT contains different types of cells, such as preadipocytes, adipocytes, immune cells, fibroblasts, and endothelial cells [65]. In mammals, there are two types of AT, white (white adipose tissue (WAT)) and brown (brown adipose tissue (BAT)), which differ in anatomical location, morphology, functions, biochemical features, and gene expression patterns [66]. WAT accounts for over 95% of adipose mass, while BAT represents about 1-2% [66]. WAT is responsible for energy storage in triglycerides, formed from the esterification of glycerol-3-phosphate and free fatty acids (FFAs) [64]. The most dominant characteristic of BAT is non-shivering thermogenesis, where energy derived from fatty acid (FA) oxidation generates heat by mitochondrial uncoupling to maintain body temperature [29,66]. The magnitude of adipose tissue mass is determined by the enlargement of existing adipocytes (hypertrophy) and by an increase in preadipocyte number (hyperplasia) [64,67,68]. During childhood and adolescence, the number of adipocytes is determined and remains constant in the adult body, regardless of whether the individual is obese or lean [65]. In adulthood, the mass of WAT increases through hypertrophy [65]. In overnutrition, there is an increased accumulation of fat in the adipocytes and cells undergo cellular hypertrophy. In contrast, lipolysis occurs in the adipocytes during calorie restriction [68].
Adipogenesis is a multi-step process leading to the conversion of mesenchymal stem cells (MSCs) and preadipocytes into mature adipocytes and consists of three stages: commitment of MSCs to the adipocyte lineage, mitotic clonal expansion (replication of DNA and duplication of cells intensively takes place), and terminal differentiation [66]. Peroxisome proliferator-activated receptor γ (PPARγ) and CCAAT/enhancer-binding proteins (C/EBP), C/EBPδ, and C/EBPβ, are key transcription factors during the early stages of differentiation [66]. PPARγ and C/EBPα cooperatively promote differentiation and the induction of adipocyte-specific genes including, inter alia, adipocyte protein 2 (AP2) and glucose transporter type 4 (GLUT4). At the terminal differentiation stage, the preadipocytes acquire the features of mature adipocytes, such as insulin sensitivity, lipid synthesis and transport, and secretion of adipocyte-specific proteins [68].
Recent evidence showed that chronic human exposure to obesogenic EDCs can be associated with inducing preadipocyte differentiation, increasing oxidative stress, and pro-moting a pro-inflammatory state leading to an increase in the risk of obesity and metabolic disorders [7,8,18,69]. Numerous obesogens such as tributyltin (TBT), bisphenol A (BPA) as well as mono-2-ethylhexyl phthalate (MEHP) activate adipogenesis by acting on nuclear receptors (NRs), in particular by activating retinoid X receptor (RXR)/PPARγ-dependent signalling [70][71][72][73]. Activation of RXR/PPARγ plays a crucial role in the regulation of the expression of genes involved in lipid droplet formation, glucose uptake and insulin responsiveness [74]. Furthermore, obesogenic EDCs can promote adipogenesis and lipid storage and fat deposition by interfering with steroid hormone receptors such as glucocorticoid receptors (GRs) and estrogen receptors (ERs) [75,76]. For example, BPA-binding GRs directly increase adipogenesis and lipid accumulation and indirectly via induction of the 11beta-hydroxysteroid dehydrogenase 1 (HSD11B1) mRNA expression involved in cortisone/cortisol conversion [76].
In vitro studies have shown that many obesogenic EDCs not only induce the differentiation of MSCs into adipocytes [55,[77][78][79] but also alter the metabolism of mature adipocytes [8,62,[80][81][82][83][84]. Exposure to EDCs can reduce the sensitivity of adipocytes to insulin, which causes an increase in blood glucose level and, consequently, may lead to insulin resistance in WAT of adulthood, potentially via a reduction in protein (serine/threonine) kinase B (PKB, also known as Akt) and GLUT4 levels [8,85]. Moreover, EDCs affect the expression of genes related to the de novo synthesis of free fatty acids, such as fatty acid synthase (FASN) or sterol regulatory element-binding protein 1c (SREBP1C), as well as the synthesis of triglycerides, such as diacylglycerol acyltransferase 1 and 2 (DGAT1 and DGAT2), leading to a disturbance in the adipose lipid metabolism [86]. Recent evidence showed that the obesogenic action of some EDCs is associated with disruption of appetite/satiety signaling and food preferences [18]. As an endocrine organ, WAT plays an integral role in maintaining global energy homeostasis by secretion of adipokines (leptin, adiponectin), which regulate global insulin sensitivity, satiety and inflammation [66,87]. Several EDCs have been shown to alter leptin levels in animal models, including dichlorodiphenyldichloroethylene (DDE) and DEHP [80,84]. The in vitro effect of EDCs on other adipokines such as resistin correlated with insulin signaling has also been reported [80]. The potential mechanisms of the obesogenic action of daily exposed EDCs on the development of obesity and metabolic syndrome are presented in Figure 1. synthesis and transport, and secretion of adipocyte-specific proteins [68].
Recent evidence showed that chronic human exposure to obesogenic EDCs can be associated with inducing preadipocyte differentiation, increasing oxidative stress, and promoting a pro-inflammatory state leading to an increase in the risk of obesity and metabolic disorders [7,8,18,69]. Numerous obesogens such as tributyltin (TBT), bisphenol A (BPA) as well as mono-2-ethylhexyl phthalate (MEHP) activate adipogenesis by acting on nuclear receptors (NRs), in particular by activating retinoid X receptor (RXR)/PPARγ-dependent signalling [70][71][72][73]. Activation of RXR/PPARγ plays a crucial role in the regulation of the expression of genes involved in lipid droplet formation, glucose uptake and insulin responsiveness [74]. Furthermore, obesogenic EDCs can promote adipogenesis and lipid storage and fat deposition by interfering with steroid hormone receptors such as glucocorticoid receptors (GRs) and estrogen receptors (ERs) [75,76]. For example, BPA-binding GRs directly increase adipogenesis and lipid accumulation and indirectly via induction of the 11beta-hydroxysteroid dehydrogenase 1 (HSD11B1) mRNA expression involved in cortisone/cortisol conversion [76].
In vitro studies have shown that many obesogenic EDCs not only induce the differentiation of MSCs into adipocytes [55,[77][78][79] but also alter the metabolism of mature adipocytes [8,62,[80][81][82][83][84]. Exposure to EDCs can reduce the sensitivity of adipocytes to insulin, which causes an increase in blood glucose level and, consequently, may lead to insulin resistance in WAT of adulthood, potentially via a reduction in protein (serine/threonine) kinase B (PKB, also known as Akt) and GLUT4 levels [8,85]. Moreover, EDCs affect the expression of genes related to the de novo synthesis of free fatty acids, such as fatty acid synthase (FASN) or sterol regulatory element-binding protein 1c (SREBP1C), as well as the synthesis of triglycerides, such as diacylglycerol acyltransferase 1 and 2 (DGAT1 and DGAT2), leading to a disturbance in the adipose lipid metabolism [86]. Recent evidence showed that the obesogenic action of some EDCs is associated with disruption of appetite/satiety signaling and food preferences [18]. As an endocrine organ, WAT plays an integral role in maintaining global energy homeostasis by secretion of adipokines (leptin, adiponectin), which regulate global insulin sensitivity, satiety and inflammation [66,87]. Several EDCs have been shown to alter leptin levels in animal models, including dichlorodiphenyldichloroethylene (DDE) and DEHP [80,84]. The in vitro effect of EDCs on other adipokines such as resistin correlated with insulin signaling has also been reported [80]. The potential mechanisms of the obesogenic action of daily exposed EDCs on the development of obesity and metabolic syndrome are presented in Figure 1. Figure 1. Summary of the effects of daily exposed obesogenic EDCs on crucial organs (liver and pancreas) and tissues (adipose and brain) and the relationship of these effects with human obesity and diabetes. T2DM, Type 2 diabetes mellitus; NAFLD, non-alcoholic fatty liver disease. The main advantage of animal in vitro models over in vivo research is the speed of the experiments, tight control of the environment, reduced cost, well-established protocols, higher throughput and reduced animal use [88]. Regarding adipogenesis, most of these models utilize mouse 3T3-L1 cells to elucidate the mechanisms of EDCs' action during the preadipocyte differentiation processes. However, it is still unclear if the rodent's in vitro models are suitable for studying adipogenic responses due to the main- The main advantage of animal in vitro models over in vivo research is the speed of the experiments, tight control of the environment, reduced cost, well-established protocols, higher throughput and reduced animal use [88]. Regarding adipogenesis, most of these models utilize mouse 3T3-L1 cells to elucidate the mechanisms of EDCs' action during the preadipocyte differentiation processes. However, it is still unclear if the rodent's in vitro models are suitable for studying adipogenic responses due to the maintained species specificity, which may have varying responses to obesogens and limit the application of results for human-based risk assessments [89]. Nonetheless, numerous human in vitro models including primary cells from different organs and tissues have been developed and are used to identify obesogens and elucidate the mechanisms of their action implicated in the adipogenic differentiation process, adipose function and hepatotoxicity [8,83,84,90]. Regarding the adipogenesis process, in recent years, mesenchymal stem cells (MSCs) derived from adipose tissue have been utilized as an alternative to animal and human preadipocyte cell models to investigate the mechanisms of obesogenic EDC action, particularly related to disrupting the programming of adipogenesis during a prenatal period, in combination with a Western diet, which leads to a higher risk of obesity in early life and adolescence [91].

Animal Preadipocytes
The 3T3-L1 cell line is a well-established in vitro system of white preadipocytes from murine Swiss 3T3 cells [92] and consists of unipotent preadipocytes, which can differentiate only into mature adipocytes [93]. Initiation of the adipogenesis in 3T3-L1 cells requires the treatment of several pro-differentiation agents after cell growth arrest, such as dexamethasone (DEX), insulin, and phosphodiesterase inhibitor 1-methyl-3-isobutyl xanthine (IBMX) [92]. The 3T3-L1 cells are easy to culture and tolerate a very large amount of passages [94]. An important feature of 3T3-L1 cells is their ability to differentiate into both white and brown adipocytes [95]. Unfortunately, 3T3-L1 cells differ between batches from different vendors, which makes them impossible to define as a universal test system [39]. Nonetheless, the 3T3-L1 cell line has been used widely to investigate the effects of various EDCs to establish the molecular mechanisms of adipogenesis and evaluate the potential effects on the risk of obesity [93]. 3T3-L1 cells have been used by Sun et al. [93] to evaluate the molecular mechanisms of a widely used surfactant, 4-hexylphenol (4-HP), as a potential EDC-impaired adipogenesis process. Results showed that 4-HP induced adipogenic differentiation via increasing the mRNA level of PPARγ and its target genes such as fatty acid-binding protein 4 (FABP4) (also known as adipocyte protein 2 (AP2)), fatty acid translocase (CD36), perilipin, and adiponectin, but did not disturb C/EBPα expression. Moreover, in 3T3-L1 cells exposed to 4-HP, a significant increase in lipid accumulation was observed [93]. Choi et al. [7] proved that exposure of 3T3-L1 cells to BPA and its analogous such as bisphenol S (BPS), and bisphenol F (BPF) resulted in increased both mRNA and protein levels of PPARγ, C/EBPα, and AP2. De Filippis et al. [96] reported that BPA had no impact on the PPARγ, FABP4, and FASN expression and adipocyte differentiation but increased mRNA levels of pro-inflammatory markers, tumor necrosis factor-α (TNFα) and interleukin 6 (IL-6) in 3T3-L1 cells. Moreover, Sargis et al. [97] showed that induction of the adipocyte differentiation in the 3T3-L1 cells was promoted by exposure to BPA, dicyclohexyl phthalate (DCHP), and tolylfluanid (TF) through stimulating glucocorticoid receptors (GR), without any significant activation of PPARγ expression. Meruvu et al. [98] evaluated the potential of benzyl butyl phthalate (BBP) on the epigenetic modification of genes involved in adipogenesis. BBP exposure induced miR-34a-5p expression and significantly decreased the expression level of its target genes, including nicotinamide phosphoribosyltransferase (NAMPT), sirtuin 1 (SIRT1), and sirtuin 3 (SIRT3), leading to an impairment in 3T3-L1 preadipocyte differentiation and an increase in adipogenesis [98]. Numerous studies using the 3T3-L1 model have confirmed the obesogenic potential of extensively used different classes of pesticides including quizalofop-p-ethyl 7 of 64 (QpE) [99], dichlorodiphenyltrichloroethane (DDT) and dichlorodiphenyldichloroethylene (DDE) [100], diazinon [101], chlorpyrifos (CPF) [102], and tributyltin (TBT) [103], as well as zoxamide, spirodiclofen, flusilazole and acetamiprid [104]. For example, Mangum et al. [50] showed that exposure to 1,1-dichloro-2,2-bis(4-chlorophenyl)ethane (p,p'-DDE) at both the 10 and 20 µM concentrations increased intracellular lipid accumulation by 42% and 58% respectively, compared to the control. The induction of 3T3-L1 preadipocyte differentiation into mature, lipid-storing adipocytes after exposure to these agrochemicals was primarily regulated via PPARγ activation; nonetheless, multiple other obesogenic mechanisms including mitochondrial dysfunction or altered intracellular calcium levels have been also reported [29,[105][106][107].

NIH3T3-L1 Cell Line
The NIH3T3-L1 cell line was derived from desegregated NIH Swiss mouse embryo fibroblasts [108]. These cells are adherent, exhibit many physiological similarities to primary adipocytes, and are relatively easy to culture, therefore they are a good model to study adipogenesis and adipocyte function [80,108]. Regarding the EDC exposure, Riu et al. [109] evaluated the effect of tetrabromobisphenol (TBBPA) and tetrachlorobisphenol A (TCBPA), the brominated analogues of BPA, on preadipocyte differentiation and showed that TBBPA and TCBPA, via PPARγ activation, promoted triglyceride accumulation in the NIH3T3-L1 cells. In turn, Howell et al. [80] showed that exposure to DDE did not affect adipogenesis in NIH3T3-L1 cells, but significantly increased the level of adipokines such as resistin, adiponectin, and leptin in mature adipocytes.

3T3-F442A Cell Line
The 3T3-F442A cell line contains murine preadipocytes of WAT [81,110] which does not require stimulation with DEX and IBMX to differentiate into mature adipocytes [111]. The 3T3-F442A cell line was derived from fibroblasts isolated from disaggregated Swiss mouse embryos and is used to investigate the mechanism of adipogenesis [110,111]. This cell line was applied to investigate the adipogenic potential of DDT, belonging to the group of organochlorine (OC) insecticides that were used heavily during the 1950s and 1960s [112][113][114]. Several epidemiological studies have linked DDT exposure to T2D and obesity [115][116][117]. An in vitro study by Moreno-Aliaga and Matsumura [111] showed that 1,1,1-trichloro-2,2-bis (p-chlorophenyl)-ethane (p,p'-DDT) at a concentration of 20 µM caused 3T3-F442A cells to obtain the adipocyte-like morphology at day 2 of the differentiation process, but treated cells did not fully differentiate. The mouse 3T3-F442A cell line was also used to investigate the impact of BPA exposure on glucose transport [81] and demonstrated that BPA enhanced basal and insulin-stimulated glucose uptake partially via increased GLUT4 protein levels.

OP9 Cell Line
The murine OP9 is an adipocyte cell culture model established from the calvaria of newborn mice genetically deficient in functional macrophage colony-stimulating factor (M-CSF) [94,118]. OP9 cells are bone marrow-derived stromal preadipocytes of WAT characterized by fast adipogenic differentiation and the rapid accumulation of triacylglycerides in lipid droplets after only 72 h of adipogenic stimuli [39,94]. Moreover, OP9 cells can differentiate into adipocytes after reaching confluence and can be passaged for long periods in culture [39]. In comparison with 3T3-L1 cells, OP9 cells are more sensitive to the induction of adipogenesis by chemicals with the ability to activate PPARγ and RXR [19]. OP9 differentiation is a PPARγ-dependent process, and differentiating preadipocytes express C/EBPA, C/EBPβ, as well as perilipin-1 (PLIN1) and perilipin-4 (PLIN4), similar to other adipocyte models [119]. In the last years, the OP9 cell line has been used to evaluate the effects of various compounds on adipogenesis [39,120]. This cell line was used by Kassotis et al. [39] to evaluate the effect of TBT exposure on preadipocyte differentiation. Authors reported that TBT stimulated preadipocyte differentiation, and significantly enhanced triglyceride accumulation in cells on day 7 of differentiation at 100 nM concentration [39]. In turn, 36 potentially adipogenic chemicals identified by the Toxicological Priority Index (ToxPi) on preadipocyte differentiation and adipocyte metabolism in the OP9 cells were also reported by Andrews et al. [120]. Results showed that TBBPA significantly enhanced adipocyte differentiation and had high efficiency in inducing lipid accumulation in OP9 cells at 20 µM [120].

ST-13 Cell Line
The ST-13 cell line was derived from newborn mouse skin and consists of preadipocytes of WAT [82]. Differentiation of the ST-13 preadipocytes is stimulated via ciglitazone and during cell differentiation, the expression of the adipogenic markers such as PPARγ, C/EBPβ and AP2 is induced [121]. The ST-13 cell line has been used by Yamasaki et al. [82] to evaluate the effect of TBBPA exposure on the expression level of genes related to lipid metabolism in differentiated and undifferentiated adipocytes. Exposure to TBBPA at concentrations of 0.5 µM and 1 µM did not have any effects on lipid accumulation and mRNA level of the acetoacetyl-CoA synthetase (AACS) and succinyl-CoA-3-oxoacid CoA-transferase (SCOT) in mature adipocyte cell culture. However, a significantly increased gene expression of lipid and ketone body-utilizing factors such as AACS, PLIN1 and fatty acid synthase (FAS) in both 0.5 µM and 1 µM of TBBPA in ST-13 preadipocytes was noticed. Surprisingly, data showed that the BAT-related factors, uncoupling protein-1 and -3 (UCP-1 and UCP-3), PR domain-containing 16 (PRDM16), lysine-specific demethylase-1 (LSD-1), as well as cell death-inducing DNA fragmentation factor-alpha-like effector A (CIDEA), were overexpressed in ST-13 preadipocytes upon TBBPA treatment [82].

UCP-1 Cell Line
Immortalized UCP-1 reporter brown preadipocytes were generated from Ucp1-luciferase reporter mice and were used to determine the promoter activity of the UCP1 gene [29]. Only one study investigated the effect of EDC exposure on the expression of UCP-1 in immortalized brown adipocytes [29]. Wang et al. [29] investigated the effects of 34 chemicals commonly found in food due to food processing, packaging, and agriculture practices, and showed that only the organophosphate insecticide CPF suppressed UCP1 expression and mitochondrial respiration in BAT at a concentration of 1 pM. Moreover, RNA sequencing showed that at 1 pM CPF, after 4 h exposure, the mRNA levels of the carnitine palmitoyltransferase I A (CPT1A), carnitine palmitoyltransferase I B (CPT1B), and acetyl-coenzyme A acetyltransferase 3 (ACAT3) genes, important for regulating fatty acid oxidation, as well as the cytochrome C oxidase assembly factor (COX16) gene, were reduced [29].

Primary Human Preadipocytes
Primary human preadipocytes are often applied as in vitro models for the study of preadipocyte differentiation and adipocyte metabolism [122]. They are isolated from adipocyte tissue from different anatomical sites and different donors. Therefore, they reflect donor-and depot-specific characteristics which may lead to some unpredictable differences during experimental studies [94,122,123]. Even though they reflect the characteristics of the donor, they are useful in studies assessing differences between individuals (e.g., obesity, weight loss, age) [94].
Regarding the EDCs, primary subcutaneous human preadipocytes from healthy donors with body mass indices BMI ≤ 24.99 kg m −2 were used to investigate the mechanism of BPA-induced adipogenesis [75]. The study showed that BPA exposition increased the expression of C/EBPα and β, PPARγ as well as preadipocyte lipid accumulation in the absence of GR and GR agonist [75]. In another study [70], in vitro microarray analysis showed that human subcutaneous preadipocytes from donors with BMI ≤ 24.99 kg m −2 exposed to 50 µM BPA revealed 373 differentially expressed genes, with 235 of those upregulated and 138 genes downregulated. Several genes involved in triglyceride (TG) and lipid metabolism were upregulated after BPA exposure, including acetyl-CoA carboxylase α (ACACA), apolipoprotein A1-binding protein (APOA1BP), perilipin 2 (PLIN2), fatty acid desaturase 1 (FADS1), Niemann-Pick 2 (NPC2), and phosphatidic acid phosphatase type 2A (PPAP2A). In addition, for BPA-treated cells, an increase in mRNA levels was noticed for genes related to lipid metabolism such as sterol regulatory element-binding transcription factor 1 (SREBF1), low-density lipoprotein receptor (LDLR), lipoprotein lipase (LPL), and insulin-induced gene 1 (INSIG1), as well as for those related to adipogenesis, such as growth differentiation factor 15 (GDF15). Moreover, network interaction analysis identified the mammalian target of rapamycin (mTOR) signaling and the thyroid-receptor/retinoid X receptor (TR/RXR) activation as potentially involved in BPA-mediated adipogenesis [70]. Wang et al. [76] analyzed the effect of BPA on human visceral preadipocytes and adipocytes and showed that BPA in the lowest concentration tested (10 nM) increased the mRNA level of 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) gene (encoding an enzyme essential in adipogenesis and lipid synthesis). In addition, an increase in PPARγ and lipoprotein lipase (LPL) mRNA levels in preadipocytes and adipocytes was also observed [76]. Currently, scientists are very interested in the molecular effects through which EDCs can disrupt adipose tissue function in children, leading to a risk of childhood obesity and related metabolic syndrome [124]. Therefore, the detrimental effect of BPA on adipogenesis in primary preadipocytes derived from children and modulating endocrine functions has been reported [124]. Menale et al. [124] investigated the molecular mechanisms by which environmentally relevant doses of BPA affect adipogenesis in preadipocytes derived from subcutaneous adipose tissue of non-obese children between 7 and 10 years old. BPA increased the expression of FABP4 and CD36, which are important for lipid metabolism, as well as the expression of proinflammatory cytokines, such as interleukin 1 beta (IL1B), interleukin 18 (IL18) and chemokine (C-C motif) ligand 20 (CCL20) [124].

PCS-210-010
PCS-210-010 cells are commercially available cells containing normal primary subcutaneous human preadipocytes derived from WAT after liposuction surgery. PCS-210-010 cells can proliferate in an undifferentiated state and possess a higher efficiency of adipogenesis than mesenchymal stem cells. Interestingly, PCS-210-010 preadipocytes can be differentiated down osteogenic and chondrogenic lineages [125]. Regarding the EDCs, only one study reported by El-Atta et al. [83] showed that BPA exposure induced the PPARγ, AP2, and peptidylprolyl isomerase A (PPIA) expression as well as increased adiponectin levels in PCS-210-010 cells.

SGBS
SGBS cells were isolated from the stromal vascular fraction of adipose tissue of a male infant with Simpson-Golabi-Behemel syndrome (SGBS) by Wabitsch et al. [122,126]. The SGBS cells exhibit a high capacity for adipose differentiation and were applied in a study of human adipocyte development and metabolism [126]. SGBS cells, unlike primary preadipocytes from healthy donors, retain the ability to differentiate over at least 50 generations [122].
Recently, several studies applied SGBS as a model to investigate the action of a variety of plasticizers such as BPA and phthalates on human adipocyte metabolism [8,84]. For example, Schaffert et al. [8] reported that BPA and its analogues such as BPS, BPF, bisphenol B (BPB) and bisphenol AF (BPAF) displayed significant binding to PPARγ during adipocyte differentiation, but not activated PPARγ at concentrations of 10 nM, 100 nM, 1 µM and 10 µM in SGBS cells. Interestingly, during the differentiation of SGBS preadipocytes, all bisphenols decreased lipid accumulation and BPS, BPB, BPF and BPAF decreased adiponectin levels. Moreover, 1 µM BPA, BPB and BPS considerably reduced insulin sensitivity in SGBS cells upon insulin stimulation [8]. On the other hand, Schaedlich et al. [84] reported that DEHP downregulated the FABP4, adipose triglyceride lipase (ATGL), LPL, lipase E (LIPE), and CD36 mRNA levels, as well as reduced TGs accumulation in lipid droplets of the SGBS cells. Moreover, decreased adiponectin levels and increased leptin secretion in SGBS cells were also observed [84]. Recently, the effects of 20 alternative plasticizers and their metabolites on SGBS preadipocyte differentiation, induction of adipogenic markers and lipid accumulation in mature adipocytes have been reported by Schaffert et al. [62]. The molecules 1,2-cyclohexanedicarboxylic acid mono 4-methyloctyl ester (MINCH), monohydroxy isononyl phthalate (MHINP) and 6-hydroxy monopropylheptyl phthalate (OH-MPHP), which are the metabolites of bis(7-methyloctyl) cyclohexane-1,2-dicarboxylate (DINCH), diisononyl phthalate (DINP), and bis(2-propylheptyl) phthalate (DPHP), respectively, exhibited the highest adipogenic potential by induction of the SGBS preadipocyte differentiation mediated by PPARγ binding and activation. In mature adipocytes, DINCH, DINP and DPHP as well as their metabolites induced oxidative stress and mitochondrial dysfunction, and disturbed lipid storage and adipokine secretion, which was linked to inflammation and insulin resistance [62].

SW 872 Cell Line
SW 872 is a human liposarcoma cell line that differentiates without the differentiation cocktail and they constitutively express PPARγ and C/EBPα, which are crucial to adipocyte development [127,128]. Campioli et al. [129] showed that exposure to MEHP at a concentration of 10 µM activated the SW 872 preadipocyte differentiation and increased the expression of the glucose transporter type 1 and 4 (GLUT1 and GLUT4), calcium-binding protein B (S100B), as well as adenosine triphosphate citrate lyase (ACLY) and ACACA, involved in de novo lipogenesis. Additionally, MEHP temporarily increased the translocator protein (TSPO) mRNA levels during SW 872 adipocyte differentiation [129].
The obesogenic activity of selected EDCs confirmed on adipose tissue cell models is summarized in Table 1.                   Legend: ↑ increase; ↓ decrease; (−) no observed effects; * concentration (s) at which obesogenic effects were observed.

Animal Adipose-Derived Stem Cells (ADSCs)
The C3H10T1/2 cell line was established in 1973 and derived from C3H mouse embryos that were 14-17 days old [133,134]. C3H10T1/2 cells have a fibroblastic morphology and the capacity to differentiate into adipocytes, chondrocytes and osteocytes [133]. In recent years, C3H10T1/2 cells have been used to investigate the impact of various compounds on preadipocyte differentiation to investigate the molecular mechanisms associated with obesity [94,135,136]. Regarding the EDCs, the C3H10T1/2 cell line was applied to study the impact of parabens such as butylparaben on the disruption of the adipogenesis process. Results showed that exposure to butylparaben stimulated adipogenic differentiation via increased expression of PPARγ, C/EBPα and FABP4, as well as via decreased runt-related transcription factor 2 (RUNX2) mRNA levels, which plays an inhibitory role during adipocyte differentiation [130]. Other studies revealed that BBP induced adipocyte differentiation in C3H10T1/2 stem cells [137,138]. In turn, Zhang and Choudhury [138] showed that expression of the PPARγ and aP2 were significantly increased in C3H10T1/2 stem cells exposed to 50 µM BBP after 8 days of incubation. In addition, decreased SIRT1 mRNA levels, as well as increased β-catenin and forkhead box protein O1 (FoxO1) acetyla-tion under BBP exposure, were also associated with increased adipogenesis [138]. Moreover, the same authors [137] reported that 50 µM BBP significantly downregulated the expression of long non-coding H19 RNA and increased the expression of miR-103/107/let-7 (a, b, c, d, f and g) on day 2 of C3H10T1/2 cell differentiation, which probably stimulated adipogenesis. An adipogenic effect of DEHP, BPA, and TBT as a single compound was also observed in C3H10T1/2 cells by Biemann et al. [139]. Moreover, Biemann, Fisher and Navarrete Santos [140] studied the effects of the EDC mixture at high concentrations (10 µM BPA, 100 µM DEHP, 100 nM TBT), and at environmentally relevant concentrations (10 nM BPA, 100 nM DEHP, 1 nM TBT) and demonstrated that the EDC mixture affects adipogenic differentiation of the C3H10T1/2 cells, but its impact on adipogenesis was dose-dependent. Moreover, in a previous study, Kirchner et al. [78] provided evidence that TBT induced PPARγ2, FABP4 and leptin (LEP) expression in mouse ADSCs while the mRNA level of the adipocyte differentiation-associated protein (PREF-1), an inhibitor of adipocyte differentiation, was decreased.
In recent years ADSCs from domestic animals have gained increased attention because they are a much better model for understanding adipogenesis in vitro and obesityrelated diseases compared to rodent cell models [141]. For example, ADSCs isolated from porcine adipose tissue have been applied to study the obesogenic activity of EDCs. Gigante et al. [142] demonstrated that glyphosate (GLY) at the concentration of 4 µg/mL significantly decreased the viability of ADSCs and inhibited their adipogenic differentiation. Similar results were also observed by Berni et al. [143] who noticed a significantly decreased cell viability of proliferating porcine ADSCs after 48 and 72 h of 1 µM BPS exposure. Moreover, similar to GLY, BPS did not increase the PPARγ and LEP expression during the differentiation nor fat droplet formation in porcine ADSCs [143].

Human Adipose-Derived Stem Cells (hADSCs)
Human adipose-derived stem cells (hADSCs) are isolated from biopsies and liposuction specimens [144,145]. hADSCs are available from normal donors and patients with obesity (BMI > 30), Type 1 diabetes or Type 2 diabetes. Moreover, hADSCs have functional and phenotypic characteristics similar to bone marrow-derived mesenchymal stem cells (BMMSC) [145]. Normal hADSCs can differentiate into many different lineages including adipogenic, neural, osteogenic, and chondrogenic cells and can be used in research, including in stem cell differentiation [145]. An important feature of hADSCs is that they can be cultivated up to passage eight with no sign of decline [141]. hADSCs can be used as an alternative to human preadipocytes which have reduced proliferative ability and can exhibit physiological differences related to the fat depot of origin within the body [77]. A main advantage of hADSCs is the commitment of stem cells to preadipocytes and their differentiation to mature adipocytes [146]. hADSCs have been used to assess possible metabolic disruptors in vitro [77,78,147].
Regarding obesogenic EDCs, Valentino et al. [147] showed that the adipose tissuederived stromal vascular fraction (SVF) exposed to 1 nM BPA exposure decreased insulinstimulated glucose utilization and increased cytokine secretion such as IL6 and interferongamma (IFN-γ). However, no changes in mRNA levels of the adipogenic markers such as GLUT4 and PPARγ were found [147]. Ohlstein et al. [148] showed that BPA enhanced adipogenesis in human ADSCs obtained from subcutaneous abdominal tissue of three female donors with a BMI less than 25. BPA increased the expression of PPARγ, C/EBPα, LPL, insulin-like growth factor-1 (IGF1) and dual leucine zipper-bearing kinase (DLK). It has been reported that the adipogenic effect of BPA was compounded via ER activation, and this effect can be blocked by the ER antagonist ICI 182,780 [148]. De Filippis et al. [96] showed that exposure to 1 nM and 3 nM BPA did not affect the cellular commitment of hADSCs to the adipose lineage, nor did it affect PPARγ, C/EBPα and FABP4 expression or lipid accumulation. Recently, Cohen et al. [55] compared the effects of BPA and BPA replacements such as BPAF and tetramethyl bisphenol F (TMBPF) on adipogenesis and lipid accumulation in female hADSCs. BPA at 0.1 µM and BPAF at 0.1 nM increased adipogenesis and lipid accumulation but higher amounts of BPA (1 µM) and BPAF (10 nM) significantly decreased adipogenesis. Moreover, higher doses of BPA and BPAF were more toxic than lower doses, leading to an increase in cell apoptosis, thus contributing to a decreased level of adipogenesis and fat accumulation. In addition, TMBPF at a concentration of 0.01 µM and 0.1 µM also significantly lowered adipogenesis. This compound exhibited cytotoxic and anti-adipogenic effects on hADSCs, which resulted in increased levels of apoptosis and reduced lipid production [55]. Similar results were also obtained in the study of Harnett et al. [56] who demonstrated that BPA (1, 10 µM), BPAF (0.0003 µM, 0.003 µM, 0.03 µM, 0.3 µM) and TMBPF (0.01 µM, 0.1 µM, 1 µM, 10 µM, 50 µM) had high cytotoxicities and significantly decreased cell viability, leading to massive apoptosis in hADSCs. Reina-Pérez et al. [79] examined the impact of BPF and BPS on lipid accumulation and adipogenesis in hADSCs and showed that these BPA analogues at concentrations of 10 µM or 25 µM enhanced their capacity to differentiate into adipocytes and accumulate lipid droplets in a dose-dependent manner. Another study [77] showed that exposure to DDE maintained the undifferentiated state of hADSCs. DDE influenced the expression of genes involved in maintaining the pluripotent state of cells and differentiation (SOX2, OCT4, NANOG, peroxisome proliferator-activated receptor gamma, coactivator 1 beta (PPARγC1B)), lipid metabolism (FASN, sterol regulatory element-binding protein 1 (SREBP1), UCP3) and members of an insulin signaling pathway (homo sapiens v-akt murine thymoma viral oncogene homolog 2 (AKT2), insulin receptor (INSR)). In turn, hADSCs exposed to TBT showed increased cell differentiation via activation of PPARγ and downregulation of the PREF-1 expression, as an inhibitor of adipocyte differentiation [78].
The obesogenic activity of selected EDCs confirmed on mesenchymal stem cell models is summarized in Table 2.     Pesta et al. [77] Legend: ↑ increase; ↓ decrease; (−) no observed effects; * concentration (s) at which biological effects were observed.

Other In Vitro Models of Adipose Tissue and Mesenchymal Stem Cells That Can Be Used for Studying the Mechanism of Obesogenic Action of EDCs
Several immortalized human white adipocyte cell models such as the telomerase reverse transcriptase white preadipocyte cell line (TERT-hWA) [149], LiSa-2 [150], LS14 [151] and Chub-S7 [152], and brown adipocyte cell models such as TERT-hBA [149] and PAZ6 [153], have been generated. These cells, compared to primary cultures, maintain adipogenic potential over time and passages, and therefore have been used over the past few years to study adipocyte function [154].
TERT-hWA and TERT-hBA were isolated by Markussen et al. [149] upon the immortalization of white and brown stromal-vascular cell fractions from superficial and deep neck adipose tissue from a single donor. These cells maintain a fibroblast-like morphology during propagation and exhibit high proliferation and differentiation up to at least passage 20 [149]. The LiSa-2 cell line was isolated by Wabitsch et al. [150] from a poorly differentiated human pleomorphic liposarcoma and displays a high capacity for terminal adipose differentiation [150]. The LiSa-2 cells accumulate lipids and express adipocyte gene markers such as PPARγ, LPL, FASN, hormone-sensitive lipase (HSL), adipocyte most abundant gene transcript-1 (APM1), glycerol-3-phosphate-dehydrogenase (GPD1) and GLUT4 [150]. LS14 is an adipocyte cell line that was derived from a metastatic liposarcoma and shares many of the characteristics of primary preadipocytes that undergo terminal differentiation, with the expression of many adipose-associated genes [151]. The Chub-S7 cell line was derived from human subcutaneous primary preadipocytes which were transfected with human papillomavirus E7 oncoprotein and the human telomerase reverse transcriptase (hTERT) [152]. Chub-S7 adipocytes display expression of the adipocyte markers and the capacity to accumulate triglycerides [152]. Chub-S7 has been applied in the study of adipocyte differentiation, adipogenic miRNA regulation [154,155], as well as cellular metabolism [154,156].
PAZ6 is the first available immortalized human BAT cell line isolated from infant brown adipose tissue [153,154] and has been used both in white and brown preadipocyte models [154]. Differentiated PAZ6 adipocytes accumulate lipids and express brown/white markers including UCP1, β1, β2, and β3 adrenergic receptors (β-AR), adrenergic receptor α2A (α2A-AR), LPL, GLUT1 and GLUT4, as well as LEP [153,154]. Moreover, PAZ6 cells can be passaged for several months without losing their molecular markers and morphological characteristics [153]. Therefore, the above-mentioned cell lines can be potential in vitro models to study the obesogenic action of various EDCs classified as food contaminants [149].

Hepatic Cellular Models
The liver is an important detoxification organ in the body and is responsible for the biotransformation and storage of toxic compounds such as exogenous xenobiotics [160]. More than 90% of orally exposed pollutants absorbed via the stomach and intestine are transported to the liver and removed from the body via biotransformation, catalyzed by UDP-glucuronosyltransferases (UGTs) and cytochromes P450 (CYPs) [161]. The liver plays a crucial role in lipid metabolism such as the synthesis and regulation of blood lipids, de novo lipogenesis, fatty acid oxidation, fatty acid uptake, and triacylglycerol export [68,162]. The excessive lipid storage in the liver may lead to lipotoxicity and NAFLD, which very often accompanies obesity [14,93]. Scientific evidence indicates that EDCs affect liver function and lipid accumulation and induce several metabolic syndromes, including hepatic steatosis and hyperlipidemia, but the mechanisms of their action still need to be explained [28,86,163,164]. Nonetheless, several studies have shown that bisphenols, including BPA, BPS and fluorene-9-bisphenol (BHPF), induce liver toxicity and hepatocyte necrosis at low dosages [165][166][167][168].

FaO Cell Line
The rat hepatoma (FaO) cell line is a well-characterized liver cell line used as an in vitro model of steatosis [171]. FaO cells characterize a low mRNA level of the estrogen receptor β (ERβ) and lack the estrogen receptor α (Erα) expression compared to rat liver [171]. A study by Grasselli et al. [171] showed that BPA at non-cytotoxic concentrations, 30 and 300 ng/mL, induced lipid droplet accumulation and triglyceride content in FaO cells. Additionally, decreased expression of PPARα, γ, β and δ, as well as decreased expression acyl-CoA oxidase (AOX) and CPT1 genes involved in lipid oxidation, was observed. BPA had no effect on the expression of lipogenic genes (FAS, glycerol-3-phosphate acyltransferase (GPAT). Moreover, it lowered the level of mRNA transcripts of apolipoprotein B (APOB) and extracellular triglycerides, which suggests that it may cause changes in lipid secretion [171].

BRL-3A Cell Line
The BRL-3A cell line is an epithelial cell line derived from buffalo rat liver [172]. Cells of this line are capable of division in the absence of serum [172]. Using BRL-3A, Zhang et al. [173] showed that MEHP induced cellular lipid accumulation and fatty acid synthesis through inhibition of the JAK2/STAT5 signaling.

Hepa1-6 Cell Line
Hepa1-6 is a murine hepatoma obtained from the BW7756 hepatoma tumor that emerged spontaneously in C57L/J mice used as an in vitro clinical model for preclinical immunotherapy studies [177]. Regarding EDCs, Ke et al. [178] using Hepa1-6 hepatocytes investigated the effects of BPA at 0.001 µM, and 0.01 µM concentrations on the mRNA level of DNA methyltransferases and genes involved in lipid metabolism. BPA decreased the expression of DNA methyltransferase 1, 3-α and 3-β (DNMT1, DNMT3a, DNMT3b) but increased the FASN, 3-hydroxy-3-methylglutaryl CoA reductase (HMGCR), SREBF1 and SREBF2 genes [178]. 7.1.6. FL83B Cell Line FL83B is a hepatocyte cell line derived from a liver of a 15-17 day-old fetal mouse [179,180]. FL83B hepatocytes actively synthesize cholesterol and store glycogen [181]. The FL83B cells have been used to study the effects of various compounds on hepatocytes' function, including fucoxanthin [182], herbal tea extracts [183] and heavy metals such as cadmium (Cd) [184]. FL83B cells were used by Lo et al. [185] who showed that DEHP at different concentrations (125 µM, 250 µM, 500 µM, 1000 µM) injured liver FL83B cells by reducing cell viability, increasing LDH and alanine aminotransferase (ALT) release, as well as increasing cell populations of sub-G1 and S phase in a dose-dependent manner.

RTL-W1 Cell Line
RTL-W1 is the epithelial cell line derived from the liver of 4 year-old male rainbow trout [186,187]. This in vitro model was applied to study the ability of EDCs, namely TBT, terpyridine platinum(II) chloride (TPT), 4-nonylphenol (4-NP), BPA and DEHP, to alter the expression of markers of cellular lipid metabolism leading to steatosis in fish [186]. Results presented by Dimastrogiovanni et al. [186] showed that DEHP and BPA significantly increased the accumulation of lipids in RTL-W1 cells, whereas TBT, 4-NP, BPA and DEHP altered membrane lipids such as phosphatidylcholines (PCs) and plasmalogen PCs. Furthermore, RTL-W1 cells exposed to BPA, TBT, TPT, DEHP and 4-NP altered mRNA levels of ATP-binding cassette transporter A1 (ABCA1), CD36, fatty acid transport protein 1 (FATP1), FAS, LPL, PPARα and PPARβ [186].

PLHC-1 and ZFL Cell Line
The fish hepatoma PLHC-1 cell line has been derived from topminnow (Poeciliopsis lucida) [188,189] whereas the zebrafish liver cell line (ZFL) has been isolated from zebrafish (Danio rerio) [188]. These cells maintain several differentiated cell functions of hepatocytes and have been extensively used to assess the cytotoxicity and changes in gene transcription associated with xenobiotic exposure [188,189]. Regarding the effects of EDCs, Marqueño et al. [190] reported that exposure to BPA, BPF and bisphenol A bis(3-chloro-2hydroxypropyl) ether (BADGE·2HCl) induced the accumulation of ether-triacylglycerides (ether-TGs) and dihydroceramides in hepatic ZFL cells. Moreover, BPA and BADGE·2HCl increased the level of saturated TGs and lowered the levels of unsaturated TGs. Concentrations of 20 µM BPA and 20 µM BPF led to an increase in the expression of the lipogenic genes such as SCD and ELOVL fatty acid elongase 6 (ELOVL6), while the PPARα mRNA level was down-regulated by 20 µM BPF and 5 µM BADGE·2HCl [190]. In the PLHC-1 cells, exposure to BADGE·2HCl induced a strong decrease of triacylglycerides (TGs), while DEHP and dibutyl phthalate (DBP) stimulated the accumulation of TGs [191]. Furthermore, the effect of TBT on the dysregulation of lipid metabolism in PLHC-1 and ZFL cells, as well as the alteration of the FASN, SCD, and ELOVL6 expression in ZFL cells, was also reported [188].

Human Hepatocytes
Since the liver is the main organ involved in the metabolism and the toxicity of xenobiotics, isolated primary human hepatocytes (PHHs) have been increasingly used as a model in pharmaco-toxicological studies for the detection of toxic chemicals and evaluating their mechanism of toxicity [192,193]. While PHHs represent a valuable tool for studying liver function, the main limitation of their utilization is the restricted accessibility, heterogeneity, phenotypic instability and limited time for cell proliferation in in vitro culture [193,194]. Therefore, alternative hepatocyte models have been explored and used, including cells from human liver tumors or immortalized adult or fetal human hepatic cells [193]. The advantage of these cell lines is their unlimited availability and rapid growth, but they are dedifferentiated, and compared to normal adult hepatic cells, they show less liver-specific metabolism [195].

Human Primary Hepatocytes
Primary hepatocytes isolated from cancer-free portions of the liver after resection were used to assess the level of FA accumulation upon 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) exposure [90]. TCDD increased total FAs in hepatocytes, including stearate, palmitate, oleic and linoleic acids after 48 h exposure at a 10 nM concentration [90].

HepG2 Cell Line
HepG2 is the best-characterized human hepatoma cell line [194]. This cell-based model is cost-effective, easy to handle, and ensures the repeatability of the obtained results [194]. Regarding the differences between the HepG2 cell line and normal hepatocytes, HepG2 cells do not possess the complete set of xenobiotic-metabolizing enzymes (XMEs), especially UDP-glucuronosyltransferases (UGTs) and some cytochromes P450 (CYPs) such as CYP2A6, CYP2D6, CYP3A4, CYP2C9, CYP2C19, etc., that are involved during phase I of xenobiotic oxidation in the liver [164,196,197]. Despite this, HepG2 cells retain most of the metabolic functions performed by normal hepatocytes, which allows them to be used in studies of the toxic effects of drugs, nanoparticles, and heavy metals in vitro [198]. The HepG2 cell line has been applied in hepatotoxicity assessments [194] and has been used to evaluate the link between EDC exposure and fatty liver disease associated with an increased risk of obesity [93]. Regarding the surfactants, Sun et al. [93] reported that exposure to 4-HP increased lipid accumulation in oleic acid (OA)-treated HepG2 cells and inhibited de novo lipogenesis by decreasing the acetyl-CoA carboxylase (ACC) and SREBP1c expression as well as the fatty acid oxidation by decreasing the PPARα and CPT1A mRNA levels. Moreover, 4-HP accelerated the uptake process of OA in hepatocytes by an increase of the CD36 mRNA level [93]. In turn, Lu et al. [199] reported that 1,3-dichloro-2propanol (1,3-DCP) at a 0.5 to 2 µg/mL concentration increased lipid droplet accumulation as well as total cholesterol (TC) and TGs content in HepG2 cells. The molecule 1,3-DCP considerably increased the mRNA level of LDLR, SREBP2 and HMGCR, associated with lipid metabolism [199]. Furthermore, an increase in the lipid accumulation in HepG2 cells exposed to pentabromotoluene (PBT), hexabromocyclododecane (HBCD), and tetrabromobenzoate (TBB) was also reported by Maia et al. [200]. Recently, Vasconcelosa, Silva and Louro [57] showed that DINCH as a non-phthalate plasticizer induced oxidative DNA damage in HepG2 cells, which can be correlated to numerous human diseases including diabetes and cardiovascular disease.

HepaRG Cell Line
HepaRG is an immortalized hepatic cell line that has a similar expression of nuclear receptors, key metabolic enzymes (XMEs), and drug transporters as primary human hepatocytes [201]. HepaRG is a good model used in the field of toxicology because HepaRG cells can enter into a differentiation program toward hepatocyte-like and biliary-like cells [164]. Regarding pesticides, Stossi et al. [202] showed that TBT induced lipid accumulation in HepaRG cells via increased mRNA levels of SREBF1 and FASN involved in de novo lipoge-nesis. On the other hand, HepaRG cells exposed to BPA had significantly greater cellular triglyceride and neutral lipid accumulation at a 2 nM concentration [164]. BPA induced hepatic lipid accumulation by increasing the apolipoprotein A4 (APOA4) mRNA level, whereas no effect on perilipin 3 (TIP47) and perilipin 2 (PLIN2) gene expression, involved in lipid droplets accumulation, nor for genes associated with carbohydrate homeostasis, was observed [164].

HPR116 Cell Line
HPR116 cells, which are differentiated HepaRG cells, are used to save time in the experiment because they are ready-to-use and easy to use. Cells from the same batch show a repeated differentiation level and have the same behaviour. In addition, HPR116 cells are long living, remaining viable and usable for at least two weeks. They exhibit responses and functions similar to those of primary hepatocytes [203].

Huh-7 Cell Line
Huh-7 (human hepatoma) is an immortalized cell line consisting of tumorigenic cells [204]. Wada et al. [205] showed that BPA and 4-NP exposure stimulated lipid accumulation in Huh-7 cells. Similarly, Lee et al. [206] confirmed that BPA increased intracellular lipid accumulation and fatty acid uptake in Huh-7 cells.

Huh-6 Cell Line
Fetal HuH6 hepatocytes from a hepatoblastoma of a 1 year-old male donor are also used in in vitro tests of the endocrine-disrupting effects of different substances [207,208]. The advantages of this cell line include availability, unlimited growth, high reproducibility of results, and the expression of enzymes involved in the metabolism of xenobiotics [209].

L02 Cell Line
The human normal liver cell line L02 is used to study the effects of various compounds on lipogenesis [211]. The L02 cell line was derived from primary normal human hepatocytes and immortalized in 1980 [212]. This cell line has found applications in the research of human hepatocellular functions such as drug hepatotoxicity, hepatic steatosis, and chemical carcinogenesis [212]. Zhang et al. [213] showed that triclosan (TCS), widely used as an antibacterial and antifungal agent, promoted the perturbation of intracellular lipids in L02 cells.

Pancreatic Cellular Models
The α and β cells of the pancreas play an important role in blood glucose control through the secretion of glucagon and insulin [14]. Glucagon secreted by α-cells is involved in the synthesis and mobilization of glucose in the liver [214]. The insulin secreted by β-cells reduces blood glucose levels via increasing glucose uptake by insulin-sensitive tissues, such as the liver, adipose tissue and skeletal muscle, and inhibiting hepatic glucose production [68]. Type 2 diabetes mellitus (T2DM) is characterized by hyperglucagonemia and hypoinsulinemia, which results in increased blood glucose levels [14]. T2DM is comprised of a series of interrelated abnormalities such as insulin resistance (IR) and metabolic syndrome [28]. IR in WAT, skeletal muscle and liver combined with inappropriate insulin secretion from pancreatic β cells is the major cause of human T2DM [68]. Recently, an increasingly significant role in the development of T2DM is attributed to EDCs [215]. Moreover, EDCs may be of great importance in the pathogenesis of Type 1 diabetes, especially during the developmental period [23].

Rat Pancreatic Islets
Pancreatic islets isolated from male Wistar rats at 8 weeks of age were used to investigate the effects of acute and long-term exposure to BPA and NP on insulin secretion [216]. Adachi et al. [216] showed that acute exposure (60 min) to BPA and NP at 0.1, 1, 10 and 100 µg/L did not affect insulin secretion in pancreatic cells with glucose stimulation. In turn, 24 h exposure to BPA (10 and 100 µg/L) or NP (0.1, 1, 10 and 100 µg/L) with 16.7 mM glucose significantly increased insulin secretion via cytosolic/nuclear estrogen receptors [216]. Ghaemmaleki et al. [217] reported that 10 µM TBT reduced the viability of the pancreatic islets of Langerhans isolated from 2-3 month-old male Wistar rats by 50%. In addition, increased insulin secretion at both basal (2.8 mM) and stimulatory (16.7 mM) concentrations of glucose after 10 µM TBT exposure was observed [217].

INS-1 Cell Line
The INS-1 cells were isolated from a rat insulinoma induced by X-ray irradiation and are applied in the studies of insulin secretory mechanisms. INS-1 cells are bi-hormonal and are capable of expressing both insulin and proglucagon proteins [218]. There is strong evidence that BPA in the concentration range from 0.002 to 2 µM lowers the viability of INS-1 cells and increases apoptosis via a mitochondria-mediated pathway in a dose-dependent manner [219]. Interestingly, BPA at higher concentrations (0.2 and 2 µM) significantly decreased insulin secretion in response to glucose stimulation, but at 0.002 µM slightly increased insulin secretion [219].

INS-1E Cell Line
The INS-1E cell line is a stable rat insulinoma pancreatic β-cell line cloned from the INS-1 cells, characterized as less heterogeneous than the INS-1 cell line [23]. INS-1E cells are widely used in studies of β cell function [23]. It has been reported that BPA (1 µM) and TBT (200 nM) decreased INS-1E cell viability by inducing apoptosis [23]. In addition, TBT caused a reduction in the expression of the MAFA, which is a transcription factor regulating the expression of genes involved in the biosynthesis and secretion of insulin (such as pancreatic and duodenal homeobox 1 (PDX1) and glucokinase (GCK)), as well as Pdx1. Moreover, triphenylphosphate (TPP), PFOA (perfluorooctanoic acid), TCS and DDE at a concentration of up to 1 µM did not affect the viability of INS-1E cells, nor the expression level of genes involved in insulin biosynthesis and secretion [23]. Another study showed that p,p'-DDT (10 µM) and its metabolite p,p'-DDE (10 µM) reduced the intracellular level of proinsulin (precursor of insulin) and insulin monomer (the active form of insulin), as well as decreased insulin 1 (INS1) and 2 (INS2) mRNA levels in INS-1E cells [220]. Additionally, p,p'-DDT decreased the intracellular level of hexameric insulin, which is a final form of insulin secreted by pancreatic β cells. p,p'-DDT decreased the expression of actin and mortalin/GRP75 and increased the expression of tubulin beta-5 chain, annexin A4, and vitamin D-binding protein (VDBP). In turn, p,p'-DDE also increased VDBP expression but decreased glucosidase 2 subunit beta (GLU2β) precursor expression [220].

RIN-m5F Cell Line
The RIN-m5F cell line is a rat pancreatic β-cell line that can produce and secrete insulin [106,221]. Chen et al. [106] demonstrated that TBT at the concentration between 0.05 and 0.2 µM did not affect the viability of RIN-m5F cells after 24 h incubation, but increased glucose-stimulated insulin secretion (GSIS) in β cells after 0.1 and 0.2 µM TBT treatment. In a study reported by Huang et al. [221], TBT at a 0.5 µM dose increased the number of apoptotic RIN-m5F cells after 24 h incubation, which was associated with the phosphorylation of mitogen-activated protein kinases (MAPKs)-c-Jun N-terminal kinase (JNK), as well as extracellular signal-regulated protein kinase (ERK1/2) and poly (ADPribose) polymerase (PARP) cleavage. Furthermore, TBT (0.5 µM, 1 µM) was found to significantly decrease GSIS after 24 h treatment [221]. In turn, Suh et al. [222] reported that PFOA at the 100-500 µM concentration range significantly decreased the viability of RIN-m5F cells and increased their apoptosis. It turned out that this EDC caused oxidative stress and mitochondrial dysfunction via the reduction of adenosine triphosphate (ATP) level, as well as induction of cardiolipin peroxidation, mitochondrial membrane potential collapse as well as cytochrome c release [222].

Mouse Pancreatic Islets
The mouse islets derived from adult male mice (12-14 weeks old) have been used by Dos Santos et al. [23] who demonstrated that BPA (1 nM, 1 µM) and TBT (20 nM, 200 nM) promoted apoptosis in dispersed mice islets. Carchia et al. [105] showed that BPA at a low dose (0.001 µM) changed the functioning of primary murine pancreatic islets and glucose homeostasis. BPA led to the dysfunction of the mitochondria and their destruction by inhibiting the expression of genes important in mitochondrial activity. This caused a decrease in insulin secretion by β cells after 1 h glucose stimulation (16 mM), as well as a decrease in the viability of these cells [105]. Soriano et al. [223] reported that BPA at a concentration of 1 nM increased insulin secretion and decreased ATP-sensitive K + (K ATP ) channel activity in β cells from wild-type (WT) mice (C57), which was not recorded in cells from estrogen receptor β (ERβ) knockout (ERβ -/-) mice (BERKO mice). Recently, Marroqui et al. [224] showed that BPS and BPF at 1 nM and 1 µM concentrations increased insulin secretion and lowered K ATP channel activity (1 nM BPS, 10 nM BPF) in pancreatic β cells from WT mice (C57BL/6J). In turn, Chen et al. [106] reported that 0.1 µM and 0.2 µM TBT increased GSIS in isolated mouse islets.

MIN-6 Cell Line
The MIN6 cell line is derived from a mouse insulinoma. This transformed β-cell line retains GSIS and is used to study insulin secretion [225,226]. Nonetheless, it should be noted that the long-term culture of MIN6 cells results in the loss of their insulin secretory capacity in response to glucose [227]. This is probably because β-cells dominating in the culture respond poorly to glucose or due to the increased expression of genes responsible for GSIS changes over time [228]. Al-Abdulla et al. [229] showed that 100 nM BPA increased GSIS. Moreover, upregulation of, e.g., MAFA, hepatocyte nuclear factor 4 alpha (HNF4α) and PDX1, which are important for insulin secretion and normal glucose sensing, was observed in cells treated with BPA. In turn, BPS, DEHP, perfluorooctanesulfonic acid (PFOS) and DDE decreased insulin release, while cadmium chloride (CdCl 2 ) had no effect on GSIS in MIN-6 cells [229].

β-TC-6 Cell Line
The beta-tumor cell-6 (β-TC-6) cell line is a mouse islet β-cancer cell line derived from a transgenic mouse expressing genes encoding insulin, glucagon and somatostatin; β-TC-6 cells are capable of secreting insulin in response to glucose [230]. The β-TC-6 cell line was applied in the study reported by Qin et al. [231] who showed that exposure to 50 µM and 100 µM PFOS stimulated GSIS in β-TC-6 cells and increased intracellular calcium levels via G protein-coupled receptor 40 (GPR40) activation. In another study [232], 10 µM p,p'-DDE significantly increased basal and GSIS in β-TC-6 cells. Scientists speculated that this EDC does not affect insulin transcription because it does not increase the levels of PDX1 that regulates insulin gene transcription. Probably, DDE alters insulin translation by increasing the level of prohormone convertase (PC), which is involved in the cleavage of insulin to its mature form [232].

Human Pancreatic Islets
Studies conducted on β-cells from the islets of Langerhans from different human donors demonstrated that 1 nM BPA decreased K ATP channel activity (closure of K ATP channels), which contributed to an increase in GSIS [223]. Moreover, Chen et al. [106] reported that 0.1 µM TBT significantly increased GSIS in human islets from patients with benign pancreatic tumors.

EndoC-βH1 Cell Line
The EndoC-βH1 is a human cell line widely used in diabetes and islet biology research [23]. These cells express all genes that determine the primary β-cells phenotype. However, unlike primary cells, EndoC-βH1 cells may show a different expression of disallowed genes and some β-cell markers and contain approximately 5-10% of the insulin present in native β-cells [23]. Moreover, EndoC-βH1 cells have a greater ability to proliferate than adult human β-cells and present similar insulin secretion in response to glucose in human islets [23].
Using EndoC-βH1cells with BPA and TBT as positive controls, Dos Santos et al. [23] evaluated the adverse effects of PFOA, TPP, TCS, and DDE exposure on β-cell viability and GSIS. Results showed that 1 µM of the DDE, TCS, TPP, and PFOA did not affect the viability, whereas higher concentrations of PFOA (20 to 200 µM) induced apoptosis in the β-cells upon 24 h treatment. In contrast, 1 µM BPA and 200 nM TBT reduced the cell viability and induced the apoptosis of the β-cells. BPA and TCS did not affect GSIS whereas TPP, DDE, and TBT increased insulin secretion. Interestingly, PFOA decreased insulin secretion both at high and low glucose concentrations. All tested compounds, except TBT, did not modify the insulin content. PFOA, BPA, TCS, TPP, and DDE did not affect the expression of genes related to insulin biosynthesis and secretion in comparison to TBT, which increased the glucose transporter type 2 (GLUT2) expression in EndoC-βH1 cells [23]. Al-Abdulla et al. [229] investigated the effects of BPA, BPS, BPF, PFOS, DEHP, CdCl 2 and DDE exposure at different concentrations ranging from 100 pM to 10 µM on human pancreatic β-cell function. BPA, PFOS and CdCl 2 treatment resulted in a marked increase in GSIS, whereas a decrease in insulin secretion in EndoC-βH1 cells upon BPS and DEHP exposure was observed. BPF and DDE had no effect on insulin release. Regarding BPS, a significant decrease of the GLUT1, MAFA, MAFB, synaptosome-associated protein 25 (SNAP25) and KIR6.2 mRNA levels in pancreatic β-cells was also noticed [229].

NES2Y Cell Line
The NES2Y is a human pancreatic β-cell line characterized by a constitutive insulin release and possesses an insulin promoter unresponsive to changes in glucose levels [234]. NES2Y cells are proliferative, lack functional ATP-sensitive potassium channels (K ATP ), and also carry a defect in the insulin gene-regulatory transcription factor (PDX1) [234].

PANC-1 Cell Line
The PANC-1 cell line is an epithelioid carcinoma cell line derived from the human pancreas [235] and is widely used as a human model of pancreatic cells because of the cells' ability to secrete insulin in response to high amounts of glucose in the culture medium [236]. Using the PANC-1 line, Menale et al. [124] reported that 10 nM BPA effectively impaired insulin secretion in the exposed cells via downregulation of the proprotein convertase subtilisin/kexin type 1 (PCSK1) expression gene involved in insulin production. Other human pancreatic beta-cell lines such as CM, TRM-1, and Blox5 have also been used in adipogenic differentiation studies [227]. However, to the best of our knowledge, they have not yet been applied to study the impact of EDCs on the adipogenesis process.
The obesogenic effects of selected EDCs on pancreatic cellular models are summarized in Table 4.                      Legend: ↑ increase; ↓ decrease; (−) no observed effects; * concentration(s) at which biological effects were observed.

Conclusions
In this review, we have made extensive insight into the available literature regarding animal and human 2D in vitro cell models applied to evaluate the obesogenic action of various environmental EDCs. According to the available literature, animal in vitro cell models especially from rodents are extensively used for assessing adipogenesis and applied to screen environmental obesogens. The main advantage of animal in vitro cell models is their commercial availability and well-established protocols for their cultivation. Nonetheless, the translation from animal models is limited by their metabolic heterogeneity, especially between murine and human types. To reduce this risk, human primary cells isolated from adipocyte tissue and crucial endocrine organs such as the liver and pancreas are increasingly used. The main advantage of these cells is comparable identity and functions to native tissue, but their maintenance under in vitro conditions still remains a challenge, because such cells are characterized by a limited number of divisions, after which they lose their biochemical functionality and undergo programmed cell death. Recently, there has been a growing interest in the derivation of hADSCs that does not require cell transformation. The main advantage of hADSCs is the possibility of studying the exposure of chemical compounds on adipo-and lipo-genesis mechanisms by omitting the extrapolation step.
It should be noted that in vitro models are a relatively simplified system compared to the complexity of a living organism and its response to exogenous factors. It is difficult to estimate what concentration of a given compound in vitro corresponds to an in vivo dose, as well as to analyze interactions between different cell types and simulate the effects of long-term exposure to a given compound in cellular models. In addition, it is also worth noting that in vitro tests can provide unreliable results, e.g., in the case of examining the obesogenic effect of phthalates, as they are present in some laboratory plastics. Moreover, in studies with the use of cell models, not only should individual EDCs be tested, but also EDC mixtures, which reflect the "real-life" obesogenic effect much better.
In summary, in vitro studies carried out mainly on cell cultures or isolated tissue samples are used extensively to investigate the mode of action of possible industrial obesogens. While in vitro models have limitations that must be resolved, they are generally simpler, more cost-effective and can be performed in a large series of experiments under the same conditions. Nonetheless, for a better understanding of the mechanisms of the obesogenic EDCs, all information from in vitro and in vivo models should be combined.