Genotoxic activity of bisphenol A and its analogues bisphenol S, bisphenol F and bisphenol AF and their mixtures in human hepatocellular carcinoma (HepG2) cells

Highlights

• BPF and BPAF showed higher cytotoxic/genotoxic potential than BPA in HepG2 cells.

• BPS showed the lowest cytotoxic/genotoxic potential compared to BPA.

• At low environmentally relevant concentrations, BPs did not induce DNA damage.

• At combined exposure, BPs induced additive effects in the expression of specific genes.

• Not all BPA analogues present safer replacement for BPA.

Abstract

The use of bisphenol A (BPA) in manufacturing of plastics is being gradually replaced by presumably safer analogues such as bisphenol S (BPS), bisphenol F (BPF) and bisphenol AF (BPAF). Despite their widespread occurrence in the environment, there is a knowledge gap in their toxicological profiles. We investigated cytotoxic/genotoxic effects as well as changes in the expression of selected genes involved in the xenobiotic metabolism, response to oxidative stress and DNA damage upon exposure to BPs and their mixtures in human hepatocellular carcinoma HepG2 cells.

BPS and BPF slightly decreased the viability of HepG2 cells, while BPAF was the most cytotoxic compound tested. BPA, BPF and BPAF induced the formation of DNA double strand breaks determined with γH2AX assay, while BPS was inactive (5–20 μg/mL). All four BPs up-regulated the expression of CYP1A1 and UGT1A1, while BPS up-regulated and BPAF down-regulated also the expression of GST1A. Only BPA up-regulated oxidative stress responsive gene GCLC, while BPAF up-regulated the expression of CDKN1A and GADD45a.

At concentrations relevant for human exposure (ng/mL range) BPA and its analogues as individual compounds and in mixtures did not exert genotoxic activity, whereas BPA and BPAF as well as the mixtures up-regulated the expressions of CYP1A1 and UGT1A1.

Introduction

Bisphenol A (BPA) is one of the most widely used materials in the production of polycarbonate plastics, epoxy resins and phenolic resin in the manufacture of a variety of consumer products such as plastic food packaging products, can linings, baby feeding bottles, medical tubings, toys, water pipes, dental sealants, eyeglass lenses, several paper products etc. (Park et al., 2018; Vandenberg et al., 2007). The global consumption of BPA is estimated to be around 8 million metric tons in 2016, and the global BPA demand is projected to increase to 10.6 million tons by 2022 (Lehmler et al., 2018). Consequently, it is ubiquitously present in the environment, which causes huge concern as many studies reported that BPA induces adverse environmental and human health effects. It is considered that food is the major contributor to the total exposure to BPA for most population groups (EFSA, 2015), however, recently thermal paper has been identified as important source of exposure through dermal transfer (Björnsdotter et al., 2017; Eckardt and Simat, 2017). In 2017 The European Chemical Agency (ECHA) added BPA to the Candidate List of substances of very high concern. Due to its known endocrine disrupting potential the use of BPA has been banned for the manufacturing of child care products including baby bottles in European Union, the United States and Canada (Lehmler et al., 2018). In the EU, the use of BPA in thermal paper has been limited since 2016 and the ban will come into force in 2020. Since March 2018 products containing BPA must be classified and label as toxic to reproduction. The industry responded to these restrictions by development and gradual replacement of BPA by its chemical analogues such as bisphenol S (BPS), bisphenol F (BPF) and bisphenol AF (BPAF), presumably as safer alternatives to BPA.

Bisphenols (BPs) are a class of chemicals known as diphenylmethanes, which contain two benzene rings separated by one central carbon atom, usually with a 4-OH substituent on both benzene rings. BPS, BPF and BPAF have similar chemical structure as BPA (Table 1). BPS is the most commonly used replacement monomer in BPA-free products (Chen et al., 2016; Qiu et al., 2019). BPS became ubiquitous in various consumer products and environmental compartments although generally its concentrations are lower than concentrations of BPA (Wu et al., 2018; Zhang et al., 2019). BPF has a broad range of industrial applications such as lacquers, varnishes, liners, adhesives plastics, water pipe and is also used in dental sealants, oral prosthetic devices, tissue substitutes and coatings for food packaging (Liao and Kannan, 2013). BPAF is used as a cross-linker in fluoroelastomers, electronics and optical fibers, and as a high-performance monomer specialty polymers including membranes and plastic optical fibers (for review see (Chen et al., 2016)).

Massive production of BPs and broad spectrum of their use results in their presence in the environment. The average concentrations of BPA detected in surface waters are mostly around 100 ng/L but individual measurements up to 44,000 ng/L were recorded (Corrales et al., 2015). There is less data about the concentrations of BPA analogues in the environment but they are in the range between 1 and 100 ng/L; however, even 100-fold higher concentrations were detected down-flow from industrial effluents (Yamazaki et al., 2015). In humans BPA was detected in >90% of tested urine samples with average concentration 2600 ng/L (Calafat et al., 2008), while its uptake is estimated to 50 ng/kg bw/day (Huang et al., 2017). BPA analogues are detected in urine less frequently and in lower concentrations; however, they occur simultaneously in the same samples (Ye et al., 2015).

The information about toxicological properties and adverse health impacts of BPA is quite comprehensive, while such information for BPA analogues is limited. Due to structural similarities and physicochemical properties, it can be hypothesized that analogues might exhibit similar toxicological profile and similar or even stronger toxic potential as BPA. Related to the endocrine disrupting potential the studies demonstrated that BPS and BPF exhibited estrogenic, anti-estrogenic, androgenic, and anti-androgenic activities comparable to those of BPA (Björnsdotter et al., 2017; Cabaton et al., 2009; Chen et al., 2016; Perera et al., 2017), while for BPAF it was reported that it has even stronger estrogenic and anti-androgenic potential than BPA (Chen et al., 2016; Fic et al., 2014; Usman and Ahmad, 2016). Another concern is potential genotoxicity of BPs (for review see Usman and Ahmad, 2016). The mechanisms of BPA genotoxicity have not been fully elucidated, while for its analogues the data on their genotoxic activities are scarce. BPA does not induce gene mutations in bacteria (Fic et al., 2013; Schweikl et al., 1998; Tiwari et al., 2012; Xin et al., 2015) and in mammalian cells in vitro (Schweikl et al., 1998; Tsutsui et al., 1998), while evidence is accumulating that it induces DNA and chromosomal damage. BPA and its metabolites have been shown to form DNA adducts in vitro (Kolšek et al., 2012; Tsutsui et al., 1998; Wu et al., 2017; Xin et al., 2015) and in vivo (Izzotti et al., 2009), to induce DNA strand breaks in vitro (Fic et al., 2013; García-Córcoles et al., 2018; Li et al., 2017; Xin et al., 2015) and in vivo (Tiwari et al., 2012) as well as structural and numerical chromosomal aberrations in vitro (Santovito et al., 2018; Xin et al., 2015) and in vivo (Aghajanpour-Mir et al., 2016; Santovito et al., 2018; Tiwari et al., 2012; Xin et al., 2015).

It is considered that BPA analogues are safer than BPA; however, this may also be due to the lack of sufficient toxicological data to support the risk assessment. Therefore, the aim of our study was to evaluate cytotoxic and genotoxic potential of commonly used analogues BPS, BPF and BPAF in comparison to BPA in the experimental model with metabolically competent human hepatocellular carcinoma (HepG2) cells. Genotoxicity was determined by detection of H2AX phosphorylation, which reflects an early reaction to a genotoxic insult resulting in the formation of DNA double strand breaks. Formation of γ-H2AX foci correlates with the formation of micronuclei (Watters et al., 2009), and the γ-H2AX assay is more precise and sensitive than the comet assay (Kuo and Yang, 2008). The underlying mechanisms of potential toxicity of BPs analogues have been studied by analysing changes in mRNA expression of selected genes involved in the xenobiotic metabolism (CYP1A1, UGT1A1 and GSTA1), response to oxidative stress (GCLC, GPX1, GSR, SOD1A and CAT) and DNA damage response (TP53, MDM2, CDKN1A, GADD45A, CHEK1 and ERCC4).

The number of chemicals and combinations thereof to which humans and the environment are continuously exposed is potentially enormous. Concerns about the current limitations of assessing compounds individually without proper understanding of risks associated with chemical mixtures are arising. In vitro methods are one of potential tools enabling better understanding of the underlying mechanisms of mixture effect (Kienzler et al., 2016). Given that in the environmental compartments, foodstuff and consumer products BPA analogues co-exist with BPA, we also investigated the genotoxicity and mechanisms of action of the mixtures of BPA with its analogues at low concentrations that may be relevant for human exposure.