Bisphenol A (BPA) is considered as an environmental obesogen. The enzyme 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) converts the inactive hormone cortisone to the active hormone cortisol in adipose tissues and promotes adipogenesis.
To examine whether environmentally relevant concentrations of BPA could increase the expression of 11β-HSD1, as well as that of the adipogenesis-related genes peroxisome proliferator-activated receptor-γ (PPAR-γ) and lipoprotein lipase (LPL), in the adipose tissue of children.
Omental fat biopsies were obtained from 17 children (7 boys and 10 girls between 3 and 13 years of age) undergoing abdominal surgery. The effects of BPA (10 nM, 1 μM, and 80 μM) on 11β-HSD1, PPAR-γ and LPL mRNA expression, and 11β-HSD1 enzymatic activity in adipose tissue and adipocytes were assessed in vitro. Moreover, the effects of carbenoxolone (CBX), an 11β-HSD1 inhibitor, or RU486, a glucocorticoid (GC) receptor antagonist, on 11β-HSD1, PPAR-γ and LPL mRNA expression were assessed in human visceral preadipocytes and adipocytes.
BPA, even at the lowest concentration tested (10 nM), increased the mRNA expression and enzymatic activity of 11β-HSD1 in the omental adipose tissue samples and the visceral adipocytes. Similar effects on PPAR-γ and LPL mRNA expression and lipid accumulation were observed in the adipocytes. CBX treatment inhibited the stimulatory effects of BPA (at 10 nM) on PPAR-γ and LPL mRNA expression, whereas RU486 inhibited 11β-HSD1 mRNA expression in the adipocytes.
BPA, at environmentally relevant levels, increased the mRNA expression and enzymatic activity of 11β-HSD1 by acting upon a GC receptor, which may lead to the acceleration of adipogenesis.
The incidence of obesity has risen dramatically over the past decade. Despite a concerted effort to understand the underlying mechanisms, the causes of this epidemic remain unclear. Although most attention has been focused on high-calorie diets and sedentary lifestyles as the root causes, there is increasing interest in the role of environmental factors. The increase in rates of obesity was preceded by an exponential increase in synthetic chemical production.1 This association has led to the ‘the environmental obesogen hypothesis’.
Bisphenol A (BPA) is an environmental endocrine-disrupting chemical that is used commercially in products containing polycarbonate plastics, such as toys and food and beverage containers, and as an additive in other plastics.2 Children are more susceptible to this environmental pollutant than are adults. For example, BPA is leached from baby bottles during dishwashing, boiling and brushing.3 Accumulating evidence indicates widespread human exposure to BPA, which has been detected in human plasma, urine, and breast milk.4, 5, 6, 7 Notably, the prenatal and neonatal periods represent a more vulnerable window of exposure than adolescence and adulthood.6 Previous studies on the impact of BPA have investigated links between BPA and earlier onset of puberty and altered reproductive function.8 These studies have also raised concerns about the impact of BPA on adiposity.9, 10 BPA has been reported to enhance preadipocyte differentiation and lipid accumulation in mature adipocytes. BPA has also been reported to alter several metabolic functions through the expression of transcription factors and adipocyte-specific genes, such as adiponectin in adipose tissue explants and CAAT enhancer-binding protein-α, peroxisome proliferator-activated receptor-γ (PPAR-γ), and lipoprotein lipase (LPL) in 3T3-L1 cells.11, 12, 13 These studies provide evidence that supports a role for BPA as an obesogen. However, whether BPA at environmentally relevant concentrations (the low nanomolar range) also has this activity has not been examined.
Glucocorticoids (GCs) have an important role in the regulation of adipocyte differentiation, lipid synthesis and metabolism.14, 15 The GC enzyme 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) converts inactive GCs to active corticosterone or cortisol, which combines with the GC receptor (GR) to amplify local GC action.16 The potential importance of pre-receptor GC metabolism in metabolic syndromes is supported by studies using transgenic mice that overexpress 11β-HSD1 in adipose tissue.14 The overexpression of 11β-HSD1 in adipose tissue is also associated with obesity and related metabolic disorders in human adults and children.17, 18, 19 Moreover, the activity of 11β-HSD1 is regulated by many obesity-related factors, such as inflammatory cytokines, insulin-sensitising thiazolidinediones (PPAR-γ agonists) and sex steroids.20, 21 Additionally, GC levels are important regulators of PPAR-γ, CAAT enhancer-binding protein, leptin, 11β-HSD1, adiponectin and LPL expression.22, 23 Taken together, these observations strongly suggest that the overexpression of 11β-HSD1 in adipose tissue has a causal role in visceral obesity and that local GC production is tightly associated with obesity-related biofactors.
The aim of this study was to examine whether exposure to BPA could increase the expression of 11β-HSD1 in adipose tissue and thus accelerate the potential for childhood obesity. Therefore, we assessed the effects of BPA at micromolar to nanomolar levels on the gene expression, and enzymatic activity of 11β-HSD1 in omental adipose tissue samples from children. In addition, we examined the capacity of BPA to promote lipid accumulation and to increase the mRNA levels of PPAR-γ and LPL in human visceral adipocytes as a molecular indicator of adipogenesis progression.
Materials and methods
Preadipocyte medium (PAM) and human visceral preadipocytes (HPA-v) that were isolated from human visceral fat tissue were both obtained from ScienCell Research Laboratories (San Diego, CA, USA). Medium 199 (M199) was obtained from Life Technologies (Invitrogen, Carlsbad, CA, USA). Insulin, penicillin, streptomycin, 3-isobutyl-1-methylxanthine (IBMX), rosiglitazone, BPA, carbenoxolone (CBX), mifepristone (RU486) and dexamethasone (DEX) were obtained from Sigma (Sigma-Aldrich, Saint Louis, MO, USA). Oil-red O powder and TRIzol were obtained from Invitrogen.
The study protocol was approved by the Ethical Committee of Nanjing Medical University in China (NO. 2006-1201). Informed consent was provided by the parents of all of the children involved. None of the children had any type of endocrine disorder, malignancy or severe systemic illness. Furthermore, none of the children were taking any medication or had a family history of diabetes. Omental fat biopsies were obtained from 7 boys (with an age range from 3 to 10 years) and 10 girls (with an age range from 3 to 13 years) undergoing surgery for abdominal disorders. These disorders included four cases of appendicitis, five cases of teratoma, two cases of choledochocyst, two cases of intestinal malrotation and four cases of an abdominal mass. Anthropometrical measurements were carried out before surgery. The BMI values, which are calculated by dividing the weight (kg) by the height squared (m2), were transformed to BMI standard deviation scores (SDSs).24 The use of a BMI SDS corrects for variation in age and gender among the children, and the BMI SDS values of the subjects ranged from −0.80 to 1.02, which does not indicate issues with being overweight or obese.
Adipose tissue preparation and incubation
The biopsies (omental adipose tissue) were obtained from the children within 30 min after the start of surgery and were placed in 10 ml of serum-free sterile M199 media before being immediately transported to the laboratory. All subsequent procedures were conducted under a laminar airflow hood. Visible blood vessels and connective tissue were removed from the collected biopsies. The adipose tissue was washed with sterile 0.9% NaCl and cut into small pieces (3–4 mm3) using sharp scissors. The tissue fragments were placed in six-well dishes (100–150 mg adipose tissue per well) that contained 3 ml of M199 supplemented with 1% fetal bovine serum, 1 nM insulin, 100 U ml−1 penicillin, 100 μg ml−1 streptomycin and 25 μM IBMX. Medium containing a low level of IBMX is more suitable for adipose tissue growth and gene expression.21, 25 After preincubation in a humidified incubator at 37 °C under an atmosphere of 95 O2 and 5% CO2 for 24 h, the medium was removed, and 3 ml of fresh medium without IBMX was dispensed into each well. To evaluate the possible effect of BPA on 11β-HSD1 mRNA expression and enzymatic activity ex vivo, adipose tissue pieces were incubated with BPA at different doses (10 nM, 1 μM, or 80 μM) with respect to control samples for 24 h. At the end of the incubation, the adipose tissue was immediately transferred to tubes and stored at −70 °C until further analysis.
HPA-v cells culture and differentiation
HPA-v cells were grown to confluence in a six-well plate containing standard medium supplemented with 5% fetal bovine serum, 1% preadipocyte growth supplement and 1% penicillin/streptomycin solution in PAM. One day after reaching confluence, the cells were treated for 2 days with preadipocyte differentiation medium, which contained 1 μM DEX, 100 nM insulin, 100 U ml−1 penicillin and streptomycin, and 0.5 mM IBMX in PAM. Subsequently, the medium was removed, and the cells were cultured for 2 additional days in preadipocyte differentiation medium containing 1 μM rosiglitazone. This medium was replaced every other day with PAM supplemented with 100 nM insulin. The effect of BPA on the expression of 11β-HSD1 mRNA was determined by the addition of 10 nM, 1 μM, and 80 μM BPA into the medium at different stages: (1) proliferation (day 0) for 24 h, and (2) differentiation (days 4–18) for 14 days following the hormonal induction of differentiation. The cells were harvested in 1 ml of TRIzol to measure mRNA expression and stored at −70 °C until further analysis.
To evaluate the mechanism of BPA action, CBX, the 11β-HSD1 inhibitor, and mifepristone (RU486), the GR antagonist, were used. At the proliferation stage, 100 nM RU486 alone or combination with 10 nM BPA was added in the medium for 24 h, and the cells were harvested. DEX (100 nM) was included as a positive control in the experiment. On day 6, after initiation of differentiation, the medium was replaced with medium containing 2 μM CBX, and the cells were incubated for 1 h. Subsequently, the plates were replenished with 2 ml of the medium containing 2 μM CBX or containing a combination of 2 μM CBX with 10 nM BPA for 14 days, and the cells were harvested as described above.
Assessment of lipid accumulation in differentiated adipocytes
The lipid accumulation of differentiated adipocytes was determined using Oil-red O staining. The cells were washed with phosphate-buffered saline (PBS) and fixed with 3.7% (w/v) paraformaldehyde in PBS for 2 min as previously described.26 The cells were incubated for 30 min with 0.5% (w/v) Oil-red O in an isopropyl alcohol/water (60/40, v/v) solution and washed twice with PBS. The percentage of lipid-positive cells was determined using a Nikon Instruments TS300 inverted microscope (JEOL, Tokyo, Japan) as previously described.12 Three different microscopic fields ( × 400 magnification) per culture were photographed. The percentage of lipid-positive cells, which was calculated by dividing the number of lipid-positive cells by the total number of cells (at least 150 cells), in each photograph was determined by two investigators. The results were expressed as the average value of the duplicate cultures.
RNA preparation and quantification
Total RNA was isolated from cultured tissue using TRIzol according to the manufacturer’s protocol. The purity and the quantity of the RNA were spectrophotometrically determined according to optical densities at 260 nM and 280 nm. The integrity of the RNA was verified through the ethidium bromide staining of the rRNA bands, which were separated on a 1% agarose gel. The RNA was stored at −70 °C until further use.
Quantitative real-time PCR was used to determine the relative mRNA levels of 11β-HSD1, PPAR-γ and LPL. The complementary DNA was synthesised using M-MLV reverse transcriptase (Promega, Southampton, UK) with 1.0 μg of the RNA sample as described by the manufacturer. The PCR amplification on a subset of the complementary DNA samples using glyceraldehyde-3-phosphate dehydrogenase primers confirmed successful reverse transcription. The sequences of the primers are listed in Table 1. Real-time PCR was performed using the SYBR GREEN ABI Prism 7500 (Foster City, CA, USA) sequence detector with the following cycling parameters: 50 °C for 2 min, 95 °C for 10 min, 40 cycles of 95 °C for 15 s, and 60 °C for 1 min. The mRNA levels were normalised to the corresponding glyceraldehyde-3-phosphate dehydrogenase mRNA levels. The data were analysed using the 2−ΔΔ Ct method.27
11β-HSD1 enzymatic activity measurements
The enzymatic activity of 11β-HSD1 in the homogenised samples was measured in the dehydrogenase direction, which is more stable than the reductase direction.28 The enzymatic activity was measured as described previously.21, 29 Briefly, adipose tissue (40–50 mg) was homogenised in homogenisation buffer (10% glycerol, 300 mM NaCl, 1 mM EDTA, 50 mM Tris (pH 7.4)) that contained dithiothreitol (1 mM), and samples were centrifuged at 4 °C. Protein concentration was determined using a Pierce bicinchoninic acid protein assay kit with BSA as a standard (Pierce Biotechnology, Thermo Fisher Scientific, Rockford, IL, USA). The assay tubes contained supernatant of the homogenised samples (0.987 mg ml−1 protein), NADP (2 mM) and (3H)cortisol (50 nM). After incubation in a shaking water bath at 37 °C for 12 h, the reaction was interrupted. The steroids were extracted with ethylacetate, dried, dissolved in ethanol, separated by thin-layer chromatography (mobile-phase chloroform to ethanol, 92:8) and exposed to a Phosphorimager tritium screen (GE Healthcare, Europe GmbH, Freiburg, Germany). The thin-layer chromatography plates were then scanned and quantified using a Typhoon scanner (GE Healthcare, Europe). The results are expressed as the percentage of the substrate (cortisol) converted into the product (cortisone).
The Statistical Package for Social Sciences software (version 13.0; SPSS Inc., Chicago, IL, USA) was used. All values are presented as the mean±s.e. Significant differences among different treatments were analysed using a one-way analysis of variance followed by a post-hoc Fisher’s least significance difference test. The differences in expression of 11β-HSD1 between pre- and mature adipocytes were assessed using Student’s t-test for the same treatments. A value of P<0.05 was considered to be significant.
There were no significant differences in the expression level of 11β-HSD1 after BPA stimulation between samples collected from children of differing BMI SDSs or gender. Therefore, the BMI SDSs and gender were not considered in the subsequent analyses.
Effects of BPA on 11β-HSD1, PPAR-γ and LPL mRNA expression, and 11β-HSD1 dehydrogenase activity in the adipose tissue of children
The omental biopsies from children were incubated with BPA at concentrations of 10 nM, 1 μM and 80 μM. The expression of 11β-HSD1 increased to a similar level in the biopsies treated with all BPA concentrations examined (Figure 1). The expression of PPAR-γ and LPL mRNA increased with BPA stimulation, and the highest expression was observed at 80 μM BPA (Figure 1). The enzymatic activity of 11β-HSD1 in the omental adipose tissue exhibited a U-shaped curve in response to BPA; higher activity was observed at 10 nM and 80 μM, and no change was observed at 1 μM when compared with the control without BPA treatment (Figure 2).
Effects of BPA on 11β-HSD1, PPAR-γ and LPL mRNA expression and lipid accumulation in HPA-v cells
11β-HSD1 mRNA expression also exhibited a similar response to BPA; higher expression was observed at 10 nM and 80 μM than at 1 μM in both HPA-v preadipocytes and adipocytes. Moreover, 11β-HSD1 mRNA expression was higher in adipocytes than in preadipocytes, when exposed to BPA at the same concentration (Figure 3).
At terminal differentiation, the cells displayed a rounded shape, which is characteristic of mature adipocytes, and most of the cells in the treatment wells contained several cytoplasmic lipid droplets (Figure 4a). Oil-red O staining revealed lipid droplets in ∼45% of cells in the untreated cultures. Lipid droplets were found in 67, 49 and 89% of cells treated with BPA at concentrations of 10 nM, 1 μM and 80 μM , respectively. BPA at the low and high concentrations caused a significant increase in the percentage of lipid-positive cells compared with the control. Similar to the expression of 11β-HSD1 in adipocytes, BPA at 10 nM and 80 μM also increased PPAR-γ and LPL mRNA expression (Figure 4b).
Effects of CBX and BPA treatment on 11β-HSD1, PPAR-γ and LPL mRNA expression in HPA-v adipocytes
CBX (inhibitor of 11β-HSD1) was added on day 6, and the effect was examined on day 18. Oil-red O staining showed that the percentage of lipid-positive cells was lower in cultures stimulated with BPA and CBX than in those stimulated with BPA alone (46 vs 72%; Figure 5a). The BPA-induced mRNA expression of 11β-HSD1 was inhibited by adding CBX (Figure 5b). The BPA-induced mRNA expression of PPAR-γ and LPL was also completely blocked with CBX treatment.
Effect of BPA and GR agonist treatment on 11β-HSD1 mRNA expression in HPA-v preadipocytes
The GR has an important role in local GC action, and it can be targeted by BPA.30 The effect of the GR agonist RU486 on 11β-HSD1 mRNA expression was examined during the culture of HPA-v preadipocytes. RU486 alone did not have any effect on 11β-HSD1 mRNA expression. DEX alone and BPA alone increased 11β-HSD1 mRNA expression, and this increase was completely inhibited by the addition of RU486 (Figure 6).
This study showed that BPA, even at environmentally relevant (very low) concentrations, could increase 11β-HSD1 mRNA expression and enzymatic activity in adipose tissue samples isolated from children and in HPA-v cells (preadipocytes and adipocytes). Similar effects were found on PPAR-γ and LPL mRNA expression. These data support the hypothesis that BPA-mediated regulation of 11β-HSD1 could promote preadipocyte differentiation and adipogenesis, accelerating obesity during childhood.
Previously, the effects of BPA were examined at high concentrations (micromolar doses). For instance, BPA at 100 μM stimulated insulin-dependent glucose uptake and increased the expression of the glucose transporter 4 in 3T3-F442A murine adipocytes.31 BPA at 80 μM accelerated the terminal differentiation of 3T3-L1 adipocytes and increased LPL activity.12 However, BPA is found at much lower concentrations in the environment. BPA has been detected at concentrations between 0.2 to 20.6 ng ml−1 in humans, and the geometric mean was 2.6 ng ml−1,7 which is equivalent to ∼10 nM. Recently, BPA has been found to increase body weight and early adipogenesis in rats,9, 10 and BPA (at 1 nM) was found to reduce adiponectin release from human adipose tissue.13 In the present work, two doses of BPA, 80 μM and 1 μM, were selected as ‘positive controls’ based on previous studies.12, 30 These two doses most likely exceed the in vivo levels of BPA. Therefore, an additional dose of 10 nM was included in the present study as an environmentally relevant concentration. Our results showed that BPA at both low and high concentrations (10 nM and 80 μM) increased 11β-HSD1 enzymatic activity in adipose tissue samples isolated from children and mRNA expression in HPA-v preadipocytes and adipocytes. It has been estimated that increased 11β-HSD1 activity is associated with visceral adipose tissue accumulation in humans.32 Therefore, we suggest that BPA exposure at environmentally relevant concentrations has the capacity to stimulate the expression and enzymatic activity of 11β-HSD1 in the adipose tissue of children, which may be a contributing factor to childhood obesity.
Excessive GC regeneration could result in the accumulation of abdominal fat due to acceleration of preadipocyte differentiation and adipogenesis,33 and 11β-HSD1 enzymatic activity amplifies local GC action by converting inactive GCs to active cortisol. We therefore examined PPAR-γ, which is a terminal differentiation factor that is involved in preadipocyte differentiation and the modulation of several other transcription factors related to adipogenesis.34, 35 We also examined the expression of LPL and the percentage of lipid-positive cells in differentiated adipocytes, which are hallmarks of adipogenesis in mature adipocytes.11 Under the same experimental conditions, BPA, even at a low concentration, increased the mRNA expression of PPAR-γ and LPL and increased lipid accumulation, which is consistent with previous reports that DEX-induced 3T3-L1 preadipocyte differentiation transiently increases PPAR-γ expression36 and LPL activity.11 Thus, we suggest that exposure to BPA at environmentally relevant concentrations accelerates the terminal differentiation of adipocytes and adipogenesis. This action was inhibited by CBX, an inhibitor of 11β-HSD1 activity, suggesting that the BPA promotes preadipocyte differentiation and adipogenesis by increasing the expression of 11β-HSD1.
The GR is an important regulator of GC-induced gene expression22 and adipogenesis in adipocytes.15 BPA can affect the endocrine pathway through binding to the GR, oestrogen receptors (α and β), the androgen receptor, or endocrine-related receptors,2 although the binding energy of BPA might be different for each receptor. The relative binding affinity of BPA for both oestrogen receptors is at least 10 000-fold lower than that of oestradiol,37 but BPA and other endocrine-disrupting chemicals are similar to DEX in terms of their affinity for the GR and their effects on GR activation.30, 38, 39 In this study, increasing 11β-HSD1 expression following exposure to 100 nM DEX or 10 nM BPA was blocked by RU486, the antagonist of the GR. This result confirmed that the BPA-stimulated expression of 11β-HSD1 in adipocytes involves the GR pathways.
It should be noted that the mass of adipose tissue is limited in young children, and this study primarily focused on the mRNA levels and enzymatic activity of 11β-HSD1 in the adipose tissue of children. Future studies of obese children and infants are necessary to fully understand the impact of BPA on the development of obesity and/or obesity-related metabolic complications during childhood. In particular, these studies should consider testing the effects of BPA at even lower doses, such as 0.1 nM or 1 nM.
In addition, the dual activity of 11-βHSD1 (as a reductase and dehydrogenase) is apparent in humans. In intact cells, 11-βHSD1 appears to prefer the reduction reaction (that is, cortisone to cortisol, E to F).40 In homogenised tissue, however, 11β-oxidation is more stable than the reductase reaction.28, 41 This striking change in directionality between intact cells and the homogenised samples has yet to be satisfactorily explained. The published enzyme kinetics data for 11β-HSD1 in homogenised adipose tissue have been measured in terms of the dehydrogenase activity (that is, cortisol to cortisone, F to E) rather than reductase activity. This activity is proportional to total 11β-HSD1 protein present in the incubation, and there is no evidence of the conversion of cortisol to other metabolites. Moreover, differences in 11β-HSD1 enzymatic activity that may arise in measuring F to E vs E to F also remain to be studied.
In this study, we provide evidence that BPA at nanomolar concentrations could upregulate 11β-HSD1 mRNA expression and accelerate adipocyte differentiation and adipogenesis in humans. Given the persistence of BPA in the environment, as well as its presence in the human body, and the increase in GC activity in adipose tissue at environmentally relevant BPA concentrations, we suggest that the daily BPA exposure of children may be an important risk factor for obesity and metabolic homoeostasis.
This work was supported through funding from 973 Program of China (2013CB530604), the National Natural Science Foundation of China (81273064), the Blue Project of the Jiangsu Education Department of China (JX10410533), the Scientific Research Foundation for Returned Overseas Scholars of the Ministry of Education of China (DG216G15013), and the Project Funder by the Priority Academic Program Development of Jiangsu Higher Education Institutions. We thank the doctors in the Department of Surgery at Nanjing Children’s Hospital for collecting samples, Associate Professor Baoqing Mo (Department of Public Health, Nanjing Medical University) for help with statistical analysis, and Professor Duan Chen (Department of Cancer Research and Molecular Medicine, Norwegian University of Science and Technology) for valuable discussions.