**7. Bisphenol A**

BPA is used to harden plastics and to manufacture polycarbonate plastics and epoxy resins. It has high abundance in the environment and considered as estrogen-like EDC or as a xenoes‐ trogen. It was shown that BPA acts similarly to 17-β estradiol [114, 115]. BPA induces ERs (weakly to ERα and ERβ and strongly to ERRγ) but in the concentrations approximately 1000 higher (10-6–10-4 M) in comparison to estradiol [116]. BPA was classified as a weak environ‐ mental estrogen before. However, newer studies determined that, even in very low concen‐ trations (picomolar and nanomolar), BPA exerted divergent effects on the physiology of different cells and tissues and can bind to both nuclear and nonnuclear receptors [117]. Furthermore, most of BPA metabolites were shown to exert stronger estrogenic activities than the main compound [118]. Apart from the estrogenic effects, BPA was shown to bind to ARs, which can be related to excessive stimulation prostate and can lead to cancer. BPA was suggested to modulate androgen-dependent prostate cancer cell proliferation even in the concentrations corresponding to its level determined in human blood [119].

Metal food and beverage cans have a thin coating of BPA on the interior surface, which is essential to prevent corrosion of the can and contamination of food. Moreover, fetuses and young infants are commonly exposed to BPA by transplacental transfer of maternal BPA and through ingestion of maternal milk or formula in BPA containing plastic bottles [120].

BPA exhibits moderate acute toxicity to vertebrates. Through the oral, i.p., and i.v. routes, LD50 doses of BPA in rats were found to be 3250, 841, and 35.26 mg/kg body weight, respectively [121, 122]. On the contrary, LD50 doses in mice were found to be 2400 and 150 mg/kg via oral and i.p. routes, respectively. Moreover, the U.S. EPA estimated the reference dose as 50 µg BPA/kg body weight/day [123]. Studies concerning the toxic effects of BPA suggested that humans are more susceptible than rodents to the deleterious action of this substance [124]. Recently, intensive investigations are being realized that refer to toxic, teratogenic, carcino‐ genic, and particularly estrogenic mechanisms of BPA action.

#### **7.1. Biotransformation of Bisphenol A**

BPA can be biotransformed in vertebrates, invertebrates, and plants and it is also biodegraded by microorganisms including bacteria, fungi, and algae. There are differences of species or strain in the metabolism of BPA. It was reported that, in primates, BPA can more easily be absorbed orally or subcutaneously (s.c.) compared to rats, and primates need a longer period of time to eliminate BPA from serum than rats [125]. Moreover, human liver microsomes do not have the ability to glucuronidate BPA as extensively as rat liver microsomes [126]. The metabolism of BPA is faster in female rats than in male rats [127]. It was suggested that gender differences in serum BPA concentrations of adult humans may be caused by differences in the androgen-related metabolism of BPA [128]. In addition, Kim et al. reported higher levels of BPA glucuronide in men than in women [129]; however, the levels of BPA sulfate were higher in women than in men [130].

BPA glucuronide is the major metabolite of BPA. Other metabolites (BPA sulfate conjugate, BPA diglucuronide, 5-hydroxy BPA, and the corresponding sulfate conjugate) were also reported. BPA glucuronide has lower estrogenic effect compared to the main compound [131– 134]. 5-Hydroxy BPA is also less estrogenic than BPA [132, 133]. The sulfate metabolite of BPA does not show an estrogenic effect up to 1 mM [134].

#### **7.2. Genotoxic effects of Bisphenol A**

BPA did not show any mutagenic effect in Ames test even after metabolic activation [135]. However, BPA was shown to cause DNA damage in eukaryotic cells, that is, BPA induced DNA strand breaks in L5178Y mouse lymphoma cells [136] and induced aneuploidy and structural chromosomal aberrations in ER-positive MCF-7 cells and in CHO-K1 cells [132, 133, 137]. The genotoxic potential of BPA was also evaluated by Lee et al. [138]. In mutant chicken DT40 cell lines (deficient in DNA repair pathways), researchers determined chromosomal aberrations and double-strand breaks [138]. BPA was shown to induce ROS generation, which in turn caused an induction of the production of DNA oxidative bases [139]. However, Audebert et al. [140] did not observe any genotoxicity when BPA was administered to three human cell lines [human intestinal cell line (LS174T), hepatoma cell line (HepG2), and renal cell line (ACHN)]. On the contrary, BPA metabolites (BPA-quinone) were shown to induce DNA damage by forming covalent adducts with DNA and adducts with deoxyguanosine [131, 141].

BPA was also suggested to alter gene and protein expression. Recently, Fernandez et al. [142] showed that BPA increased the expression of some genes (i.e., BRCA1, BRCA2, BRCC3, and BCL2L11) that are involved in DNA repair and apoptosis in human breast epithelial cells. The authors suggested that women who have BRCA1 or BRCA2 mutations may be more suscep‐ tible to such effects of BPA [142].

### **7.3. Epigenetic effects of Bisphenol A**

*In utero* or neonatal exposure to low doses of BPA may cause alterations in DNA methylation, modifications in histones, and changes in the expression patterns of noncoding RNAs. These changes can up-regulate or down-regulate different gene expressions, which in turn may result in permanent health effects such as neural and immune disorders, infertility, and late onset of complex diseases. BPA induced permanent alterations in DNA methylation patterns of different genes that are responsible for cellular signaling [64, 143].

The epigenetic effect of BPA was clearly demonstrated in viable yellow mice [144]. By decreasing CpG methylation in the IAP retrotransposable sequence inserted upstream of the Agouti gene, the maternal exposure to BPA shifted the coat color distribution of viable yellow mouse offspring toward yellow. This effect was completely prevented by maternal supple‐ mentation with folic acid or the phytoestrogen genistein, both of which are sources of methyl group [145].

BPA was suggested to induce mainly breast and prostate cancer in both animals and humans. Keri et al. [146] suggested that BPA may increase the risk of such cancers by affecting various cell processes such as DNA methylation and chromatin remodeling during development. BPA exposure was shown to cause epigenetic alterations in rodent prostate and have been postu‐ lated to be the underlying cause of neoplastic development in later life [147]. Neonatal exposure of rats to BPA resulted in an increased incidence of prostate intraepithelial neoplasia, and the prostate tissues showed consistent methylation changes. For example, the phospho‐ diesterase type 4 variant 4 (Pde4d4) gene was found to be hypomethylated at the regulatory CpG island and started to have an elevated expression in the adult rat prostate [148, 149]. On the contrary, neonatal BPA exposure was also reported to alter the promoter methylation and expression of nucleosome binding protein-1 (Nsbp1) and hippocalcin-like 1 (Hpcal1) genes in rats [150]. Also, the neonatal exposure to BPA was shown to induce hypermethylation of ER promoter regions in rat testis. This phenomenon mediated epigenetic changes that in turn induced adverse effects on spermatogenesis and fertility [151].

Concerning miRNAs, BPA exposure of human placental cell lines has been shown to alter miRNA expression levels; particularly, miR-146a was strongly induced after BPA application. Induction of miR-146a caused slower cellular proliferation rates and higher sensitivity to bleomycin, which strongly induces DNA strand breaks [152]. In mouse Sertoli cell line (TM4), 24 h BPA exposure leads to up-regulation or more often to down-regulation of 37 miRNAs [153].

#### **7.4. Carcinogenicity of Bisphenol A**

Currently, there is no evaluation of the IARC for BPA. However, "bisphenol A diglycidyl ether" is classified as a Group III carcinogen (not classifiable as to its carcinogenicity to humans) by the IARC [154].

At low doses, BPA was shown to alter mammary gland development and increased incidence of tumors in Sprague-Dawley rats [155]. Besides, at the comparable amounts to a reference dose, BPA caused development of breast, prostate, and nipple cancers in both mice and rats [146]. Acevedo et al. [156] recently evaluated the malignant potential of BPA in rats and showed that suggested that BPA acted as a mammary gland carcinogen at doses comparable to those present amounts in human urine and blood determined by epidemiological studies. BPA was also suggested to cause the development of hematopoietic cancers and induced testicular cancer in mice and rats [157].

It was also shown that BPA promoted proliferation of human epithelial ovarian cancer cells (OVCAR-3), increased the susceptibility of endometrial cells to the effects of estradiol, and inhibited the 17-β-estradiol-induced genomic activity of ERβ in DLD-1 colon cells [158–161].

#### **7.5. Reproductive toxicity of Bisphenol A**

Perinatal exposure to environmentally relevant BPA doses may predispose the tissue to earlier onset of disease, reductive fertility, and mammary and prostate cancers, as BPA may cause morphological and functional alterations in both male and female genital tracts and mammary glands. The estrogenicity of BPA has been shown by many rodent studies. When pregnant mice were exposed to environmentally relevant doses of BPA, earlier vaginal opening and earlier first estrous cyclicity in their offspring were observed. BPA elevated prostate weight in mice. An advanced reproductive aging was also observed [162]. BPA was also shown to disturb the development of reproductive organs, testosterone excretion, and sperm production in mice [163].

Some studies have suggested that BPA may not alter estrogenic function in rodents. Ryan et al. [164] showed that BPA exposure of male and female rats (2–200 µg/kg body weight/day) did not affect maternal pregnancy or weight gain or F1 female birth weight as well as repro‐ ductive morphology, fertility, fecundity, or sexual dimorphic behaviors. Recently, Ziv-Gal et al. [165] suggested that BPA at low doses (110–438 µM) decreased estradiol levels and inhibited growth of follicles isolated from wild-type and AhR knockout mice and that AhR signaling pathways may be significantly involved in the modulatory effect of BPA on follicular growth.

### **7.6. Hepatotoxic effects of Bisphenol A**

Atkinson and Roy [130] found that the BPA metabolite, bisphenol-*o*-quinone, could bind DNA *in vitro* and *in vivo*, leading to the induction of hepatotoxicity. Besides, BPA was also shown to cause damage to hepatocytes by oxidative stress. BPA markedly decreased antioxidant enzymes and glutathione *S*-transferase activities as well as depleted reduced glutathione levels in rats. Moreover, BPA (50 mg/kg) significantly increased the biochemical levels of liver enzymes and reduced the expression of hepatic antioxidant genes. The authors concluded that BPA generated ROS and reduced the antioxidant gene expression causing hepatotoxicity.

Huc et al. [166] observed that low BPA doses of BPA led to hepatic (HepG2) cell damage and this effect might be due to significant mitochondrial dysfunction including ROS production, mitochondrial transmembrane hyperpolarization, lipid peroxidation, and release of proin‐ flammatory cytokines. Moon et al. [167] also observed that BPA exposure caused oxidative and proinflammatory damage in rat hepatocytes. The researchers observed a decrease in glutathione peroxidase activity, an increase in lipid peroxidation, and decreases in the proinflammatory cytokines [i.e., interleukin (IL)-6 and tumor necrosis factor-α]. Moreover, the researchers showed that, at low doses, BPA decreased the oxygen rate and ATP production and caused mitochondrial dysfunction. Based on these studies, they concluded that BPA induced hepatic mitochondrial dysfunction below the no observed effect (NOEL) value (5 mg BPA/kg body weight/day) [167].

#### **7.7. Effects of Bisphenol A on other organs/systems**

BPA was proven to affect not only estrogenic system but also functions of androgens, prolactin, insulin, and thyroid hormones [117, 168]. Gentilcore et al. [169] observed that BPA at low doses (10−9 M) affected the expression of the genes involved in thyroid hormone synthesis, thyroid follicular cells, and altered thyroid-specific transcriptional factors in zebrafish.

*In vitro* experiments showed that BPA may cause toxicity in nervous system cells. BPA in high concentrations (>100 µM) was shown to induce apoptosis of hippocampal neuronal cells by increasing calcium and ROS levels and then by activating caspase-3 and mitogen-activated protein kinases (MAPK) [170]. In rat embryonic midbrain cells, at relatively low concentrations (10−12 to 10−4 M), BPA caused S- and G2/M-phase arrests and elevated the percentage of apoptotic cells. BPA also lowered the phosphorylation of c-Jun N-terminal kinase and increased the mRNA expression level of proapoptotic proteins (i.e., Bax and p53) [171].

BPA did not cause morphological and neurobehavioral changes in F1 offspring of rats treated with different doses (0.15–2250 ppm) [172]. BPA (administrated 2–200 µg/kg/body weight/day by gavage) did not have any impact on the sensory system and neurobehavioral activity in Long-Evans rats [163]. However, some studies showed that BPA may have an effect on the neurotransmitter levels of rodents. Nakamura et al. (2010) observed that BPA (20 µg/kg s.c.) had increased brain levels of dopamine and its metabolite 3,4-dihydroxyphenylacetic acid (DOPAC), whereas the levels of serotonin and its derivative, 5-hydroxyindoleacetic acid (5- HIAA), decreased in pregnant mice [173].

BPA was suggested to both stimulate and inhibit the activity of immune system cells. It is postulated that BPA can modulate the immune activity by affecting ERs, AhR, and probably peroxisome proliferator-activated receptor (PPAR) [174]. Youn et al. [175] showed that BPA (in drinking water) caused increases in interferon-γ and decreases in IL-4 production in T lymphocytes of mice, whereas Lee and Lim [176] observed that BPA elevated IL-4 and IL-8 levels in mouse T lymphocytes. Exposure of mouse splenic lymphocytes to low BPA concen‐ trations (1 µM) could inhibit mitogenesis, particularly the mitogenesis of B lymphocytes [177]. Goto et al. [178] observed that BPA produced lymphocytes with higher amounts of immuno‐ globulin A (IgA) and IgG2a in mice. Sugita-Konishi et al. [179] reported that BPA caused depletion in neutrophil activity and inhibited IL-6 formation in mice infected with nonpatho‐ genic *Escherichia coli*.

Roy et al. [180] showed that offspring of female mice exposed to BPA were more susceptible to infection by influenza A virus because of the modulation of their innate immunity by BPA. However, the researchers did not observe impairment in antiviral adaptive immune response, which is a crucial response for virus clearance and survival.

#### **7.8. Epidemiological studies concerning the effects of Bisphenol A on reproductive/ endocrine systems**

Most of the epidemiological studies on BPA mainly focus on its effects on endocrine and reproductive systems. Meeker et al. [181] analyzed urinary BPA concentrations of men (*n*=167) who were admitted to an infertility clinic. The researchers reported that their urinary BPA levels were inversely correlated with their estradiol/testosterone ratio [181]. In a cross-sectional study, Melzer et al. [182] determined the urinary BPA levels of subjects from Chianti, Italy (n=1453, age=20–102 years), and observed positive associations between higher urinary BPA concentrations and higher expression of two estrogen-responsive genes, encoding ERβ and estrogen-related receptor α (ERRα). Li et al. [183] examined the urinary BPA levels of 218 men with and without BPA exposure in the workplace. Increasing urinary BPA levels were statistically significantly associated with decreased sperm concentration, total sperm count, sperm vitality and sperm motility after adjustment for potential confounders using linear regression. Men with detectable urinary BPA concentrations had more than risk of decreased sperm concentration, sperm vitality, count, and motility compared to men who had undetect‐ able urinary BPA levels. However, urinary BPA levels were not associated with semen volume or abnormal sperm morphology [183].

Takeuchi and Tsutsumi [127] investigated serum BPA concentrations of men (*n*=11), women (*n*=14), and women with polycystic ovary syndrome (PCOS; *n*=16). Serum BPA levels were significantly higher in normal men and in women with PCOS. There were significant positive correlations between serum BPA and total testosterone and free testosterone concentrations in all subjects and between serum BPA and total testosterone (*r* 5 0.559; *P*<0.01) and free testosterone concentrations in all female subjects; however, there was no association between serum BPA and other sex-related hormone concentrations in any group. The researchers determined that there were gender differences in serum BPA concentrations, where men had higher serum BPA levels (1.49 ± 0.11 ng/mL in men, 1.04 ± 0.10 ng/mL in women with PCOS, and 0.646; 0.10 ng/mL in women) possibly due to the differences in the androgen-related metabolism of BPA [27].

Miao et al. [184] conducted a cross-sectional study among female workers from BPA-exposed (n=106) and unexposed (*n*=250) factories in China. They observed a significant positive association between increased urine BPA concentration and higher prolactin and progesterone levels. Among exposed workers, a positive association between urine BPA and estradiol was also determined. In addition, a statistically significant inverse correlation between urinary BPA concentration and FSH was found among unexposed group. The researchers suggested that BPA exposure may lead to alterations in female reproductive hormone levels [184]. Mínguez-Alarcón et al. [185] conducted a prospective cohort study at the Massachusetts General Hospital Fertility Center, which included 256 women (*n*=375 *in vitro* fertilization cycles) who provided up to two urine samples before oocyte retrieval (total *N*=673) between 2004 and 2012. Urinary BPA concentrations were not associated with endometrial wall thickness, peak estradiol levels, proportion of high-quality embryos, or fertilization rates. The researchers did not observe any correlation between urinary BPA concentrations and implantation, clinical pregnancy, or live birth rates per initiated cycle or per embryo transfer. Women older than 37 years had thinner endometrial thickness across increasing quartiles of urinary BPA concen‐ trations, whereas women younger than 37 years had thicker endometrial thickness across increasing quartiles of urinary BPA concentrations [185]. A cross-sectional study was carried out by Liu et al. [186] to determine the associations between urinary BPA levels serum reproductive hormone levels among male Chinese adults (*n*=592). A multiple linear regression and log-binomial model was used to examine the associations between urinary BPA level and hormone levels after controlling for age and smoking status. Increased urinary BPA levels were positively and significantly correlated with prolactin, estradiol, and SHBG levels and were negatively and significantly associated with androstenedione and free androgen index levels. The researchers suggested that high urinary BPA levels were associated with increased prolactin, estradiol, and SHBG level in males and these associations may contribute to male infertility [186]. Liu et al. [187] conducted a cross-sectional study to investigate the correlations between maternal phenolic exposure and cord sex steroid hormones and AGD in male newborns. Mother-infant pairs from each of two hospitals [one in a polluted town (Guiyu; *n*=77) and the other in a cleaner town (Haojiang; *n*=60)] were recruited in the study. Maternal urinary BPA concentrations (log2 transformed) were negatively correlated with testosterone levels and testosterone/estradiol ratio in male fetal cord blood samples (after adjustment for potential confounders in linear regression models). No significant associations between AGD or anogenital index (AGI) and BPA or cord hormone levels were found [187].

Another study by our group recruited nonobese girls newly diagnosed with idiopathic central precocious puberty (ICPP; *n*=28; ages 4–8 years) and 25 healthy age-matched girls with no history of ICPP or any other endocrine disorder. Urinary BPA levels were significantly higher in ICPP group compared to the control group. There were no significant associations between urinary BPA levels and serum LH, FSH, and estradiol levels [188].
