Elsevier

Aquatic Toxicology

Volume 247, June 2022, 106174
Aquatic Toxicology

The toxicokinetics of bisphenol A and its metabolites in fish elucidated by a PBTK model

https://doi.org/10.1016/j.aquatox.2022.106174Get rights and content

Highlights

  • An existing PBTK was modified to predict BPA TK in three-spined stickleback.

  • BPA TK was similar between stickleback, zebrafish, and trout.

  • In fish, the liver may not be the only major site of BPA metabolization.

  • Plasma or gills could be a non-negligible site of BPA metabolization.

  • The PBTK highlighted discrepancies in literature data.

Abstract

Bisphenol A (BPA) is a chemical of major concern due to its endocrine disrupting function, high production volume, and persistence in the aquatic environment. Consequently, organisms such as fish are subject to chronic exposure to BPA. However, physiologically-based toxicokinetic (PBTK) models, which are valuable tools to improve the understanding of a chemical's fate in an organism, have never been specifically adapted to model BPA toxicokinetics (TK) in fish. In our work, an existing PBTK developed for four different fish species was modified to model BPA ADME processes (absorption, distribution, metabolization and excretion). The metabolization of BPA into BPA-monoglucuronide (BPA gluc) and BPA-monosulfate (BPA sulf) and their TK in various organs was taking into account in the model. Experiments were performed to generate BPA TK data in a model species commonly used in ecotoxicology, the stickleback. The model structure had to include two sites of metabolization to simulate BPA TK accurately in stickleback organs. Thus, the fish liver may not be the only site of the metabolization of BPA: plasma or gills could also play a role in BPA metabolization. The PBTK model predictive performance evaluated on literature data in zebrafish and rainbow trout concurs with this conclusion. Finally, a calibration mixing data from the three species was compared to the calibration on stickleback data only.

Introduction

During the first decade of the 21st century, bisphenol A (BPA) has raised serious concern among the scientific community. BPA was first synthesized in 1891 and its use became widespread in the 1950s; its production exceeded six billion pounds in 2000 (Vandenberg et al., 2007). BPA was first developed as a synthetic estrogen because it can bind to estrogen receptors (ERs). However, because its binding capacity is 5 to 6 orders of magnitude smaller than that of estradiol, its use as a synthetic estrogen was abandoned (Dodds et al., 1938). Nevertheless, its use as a plastic monomer and plasticizer in various everyday products has led to ubiquitous environmental contamination. Over the last decade, BPA has been detected in freshwater courses, usually below 1 µg/L, but downstream from wastewater discharge levels reached 100 µg/L (Faheem and Bhandari, 2021). Consequently, Flint et al. (2012) determined the typical environmental BPA contamination level to be 12  µg/L or less.

Several studies have focused on the effects of BPA on aquatic communities and, in particular, its endocrine disrupting effects on reproduction and growth process in fish. Moreover, recent papers have demonstrated that exposure to BPA triggers a variety of effects that are not only related to reproduction but also innate immunity, cardiac response, and oxidative stress (Wu et al., 2011; Little and Seebacher 2015; Qiu et al., 2016; Pandey et al., 2018; Gu et al., 2020). In the context of environmental risk assessment, effects at higher levels of organization are extrapolated from individual-level data. However, extrapolations generally fail because the link between each level is mostly empirical (Forbes et al., 2008). The first step in building a mechanistic framework to assess the environmental risk of BPA is to describe the relationship between environmental levels and internal levels in the organism, i.e. the toxicokinetics. On this basis, physiologically-based toxicokinetic (PBTK) models have been proven to be valuable tools to predict how fish accumulate chemicals based on ADME (Absorption, Distribution, Metabolization, and Excretion) processes (Gerlowski and Jain 1983; Brinkmann et al., 2016; Grech et al., 2017; Tebby et al., 2019; Mit et al., 2021). PBTK models have the advantage of simulating the time-course of toxicant concentrations over time and can also provide a mechanistic framework to improve the understanding of the contribution of each ADME process.

Two PBTK models have already been used to simulate the TK of BPA in fish. Both models were generic, meaning that they were not specific to BPA but initially applicated to various substances. Pery et al. (2014) developed a PBTK for zebrafish (Danio rerio) with eight compartments, and Grech et al. (2019) extended the PBTK to three more species used in ecotoxicology, fathead minnow (Pimephales promelas), rainbow trout (Oncorhynchus mykiss), and three-spined stickleback (Gasterosteus aculeatus). The model proposed by Pery et al. (2014) was successfully evaluated on only one study reporting BPA zebrafish whole-body concentrations (Lindholst et al., 2003). The model presented by Grech et al. (2019) was evaluated on both BPA whole-body and organ concentrations in zebrafish and trout (Lindholst et al., 2000; Lindholst et al., 2001; Lindholst et al., 2003; Fang et al., 2016). However, in those models, metabolization was extremely simplified since it was treated as a way of excretion, many parameters were retrieved from publications related to mammals (Shin et al., 2004; Edginton and Ritter 2009) and, since no calibration was performed on the model parameters, ten-fold overpredictions compared to data from the literature were sometimes observed (Grech et al., 2019).

Few data are available regarding the ADME processes of BPA in fish. As in mammals, BPA undergoes phase II metabolism, which is assumed to occur mainly in the liver (Lindholst et al., 2001; Lindholst et al., 2003). Interestingly, the enzymes involved in metabolite synthesis have been characterized since zebrafish cells are commonly used as human metabolization models. The activity of UDP glucuronosyltransferases (UGTs) and sulfotransferases (STs) in the presence of BPA result in the fast production of BPA monoglucuronide (BPA gluc) and BPA monosulfate (BPA sulf) (Lindholst et al., 2001; Lindholst et al., 2003; Ohkimoto et al., 2003; Wang et al., 2014). BPA gluc was shown to be the main metabolization product in mammals and fish, with BPA sulf production being a minor pathway (Lindholst et al., 2003; Gramec Skledar and Peterlin Mašič, 2016). Thus, BPA gluc levels reached 22 times the steady-state whole body BPA concentration in zebrafish, whereas BPA sulf only represented less than a tenth of the BPA level (Lindholst et al., 2003). Nevertheless, it has been shown that BPA gluc cannot bind to ERs and cannot be considered an EDC (Matthews et al., 2001). Yet, as pointed out in Karrer et al. (2018), some mechanisms of action of BPA metabolites may still be unknown. For example, Boucher et al. (2015) showed that BPA gluc induced adipocyte differentiation when it was supposed to be inactive. In addition, BPA metabolites could be subject to deconjugation that could increase the actual concentration in fish organs, as shown in rats (Kawamoto et al., 2007).

This study aimed to propose a new PBTK explicitly developed for BPA in the three-spined stickleback. To this purpose, we modified the PBTK of Grech et al. (2019) to describe the kinetics of the two primary BPA metabolites, BPA gluc, and BPA sulf, in fish organs. New kinetic data in stickleback was obtained and used to calibrate the model. PBTK predictions were then compared to observed data found in the literature for zebrafish and trout. The inter-species variability of BPA kinetics was then studied using the PBTK model and discussed to improve the mechanistic understanding of BPA TK.

Section snippets

Literature experimental data

Five publications were identified in the literature to provide TK data for the calibration of the model (Lindholst et al., 2000; Lindholst et al., 2001; Lindholst et al., 2003; Fang et al., 2016; Eser et al., 2021). A description of each experimentation can be found in Table 1 and an extended description can be found in SI Section 7. Experiments had been carried out on rainbow trout and zebrafish with various exposure scenarios. BPA levels were measured at 1, 5, 12, and 21 days (

Determination of critical parameters

The sensitivity analysis performed prior to calibration (SI Section 3. Sensitivity analysis) showed a varying influence of parameters depending on the exposure period. In most cases, BPA concentration in organs was mainly driven by BPA-related parameters, PCbw, UF, and the plasmatic clearance describing BPA gluc synthesis, but also by plasma parameter (percentage inverse of the hematocrit). Metabolite concentrations were driven by BPA-related and metabolite-related parameters such as plasmatic

Discussion

Fish have been used as a model species for endocrine-disrupting effects for many reasons, including their ease of use in laboratory experiments, the ubiquity of endocrine chemicals in water, and also because some species, such as three-spined stickleback, have shown specific biomarkers indicative of the presence of EDCs (Tyler et al., 1998; Jolly et al., 2009). In particular, BPA effects, and to a lesser extent, BPA kinetics, were the subject of several publications (Flint et al., 2012;

Conclusion

An expected output of this work is to propose an integrating approach to link internal concentrations to BPA effects in three-spined stickleback in the future. Thus, this work consisted of the first step of a scaling-up process where environmentally relevant concentrations of BPA are linked to bioaccumulation in fish organs. The structure of the model was built to allow prediction of BPA concentrations and its metabolites in various organs. The model was then calibrated based on TK data

Funding information

This work was supported by the French National program EC2CO (Ecosphère Continentale et Côtière) as part of the DERBI project and by the French Ministry of ecological transition (P190). The authors wish to thank Cleo Tebby for its contribution in the calibration process and its careful proofreading.

CRediT authorship contribution statement

Corentin Mit: Investigation, Methodology, Software, Formal analysis, Writing – original draft, Writing – review & editing. Anne Bado-Nilles: Supervision, Writing – review & editing. Gaëlle Daniele: Formal analysis, Investigation, Writing – review & editing. Barbara Giroud: Formal analysis, Investigation, Resources, Writing – review & editing. Emmanuelle Vulliet: Formal analysis, Investigation, Resources, Writing – review & editing. Rémy Beaudouin: Formal analysis, Software, Supervision, Writing

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References (63)

  • C. Jolly et al.

    Detection of the anti-androgenic effect of endocrine disrupting environmental contaminants using in vivo and in vitro assays in the three-spined stickleback

    Aquat. Toxicol.

    (2009)
  • Y. Kawamoto et al.

    Development of a physiologically based pharmacokinetic model for bisphenol A in pregnant mice

    Toxicol. Appl. Pharmacol.

    (2007)
  • R. Lavado et al.

    Microsomal biotransformation of chlorpyrifos, parathion and fenthion in rainbow trout (Oncorhynchus mykiss) and coho salmon (Oncorhynchus kisutch): mechanistic insights into interspecific differences in toxicity

    Aquat. Toxicol.

    (2011)
  • M.J. Leaver et al.

    Piscine UDP-glucuronosyltransferase 1B

    Aquat. Toxicol.

    (2007)
  • C. Lindholst et al.

    Estrogenic response of bisphenol A in rainbow trout (Oncorhynchus mykiss)

    Aquat. Toxicol.

    (2000)
  • C. Lindholst et al.

    Uptake, metabolism and excretion of bisphenol A in the rainbow trout (Oncorhynchus mykiss)

    Aquat. Toxicol.

    (2001)
  • C. Lindholst et al.

    Metabolism of bisphenol A in zebrafish (Danio rerio) and rainbow trout (Oncorhynchus mykiss) in relation to estrogenic response

    Comp. Biochem. Physiol. C Toxicol. Pharmacol.

    (2003)
  • A.G. Little et al.

    Temperature determines toxicity: bisphenol A reduces thermal tolerance in fish

    Environ. Pollut.

    (2015)
  • C. Mit et al.

    Modeling acetylcholine esterase inhibition resulting from exposure to a mixture of atrazine and chlorpyrifos using a physiologically-based kinetic model in fish

    Sci. Total Environ.

    (2021)
  • J.W. Nichols et al.

    A physiologically based toxicokinetic model for the uptake and disposition of waterborne organic chemicals in fish

    Toxicol. Appl. Pharmacol.

    (1990)
  • J.W. Nichols et al.

    In vitro-in vivo extrapolation of quantitative hepatic biotransformation data for fish. I. A review of methods, and strategies for incorporating intrinsic clearance estimates into chemical kinetic models

    Aquat. Toxicol.

    (2006)
  • E.S. Noël et al.

    Analysis of the Albumin/alpha-Fetoprotein/Afamin/group specific component gene family in the context of zebrafish liver differentiation

    Gene Expr. Patterns

    (2010)
  • K. Ohkimoto et al.

    Sulfonation of environmental estrogens by zebrafish cytosolic sulfotransferases

    Biochem. Biophys. Res. Commun.

    (2003)
  • E. Ougier et al.

    Human biomonitoring initiative (HBM4EU): Human biomonitoring guidance values (HBM-GVs) derived for bisphenol A

    Environ. Int.

    (2021)
  • W.H. Qiu et al.

    The potential immune modulatory effect of chronic bisphenol A exposure on gene regulation in male medaka (Oryzias latipes) liver

    Ecotoxicol. Environ. Saf.

    (2016)
  • I.M. Sobol et al.

    Estimating the approximation errors when fixing unessential factors in global sensitivity analysis

    Reliab. Eng. Syst. Saf.

    (2007)
  • C. Tebby et al.

    Investigating the interaction between melamine and cyanuric acid using a Physiologically-Based Toxicokinetic model in rainbow trout

    Toxicol. Appl. Pharmacol.

    (2019)
  • L.N. Vandenberg et al.

    Human exposure to bisphenol A (BPA)

    Reprod. Toxicol.

    (2007)
  • A. Vidal et al.

    Elucidating the fate of perfluorooctanoate sulfonate using a rainbow trout (Oncorhynchus mykiss) physiologically-based toxicokinetic model

    Sci. Total Environ.

    (2019)
  • X. Yang et al.

    Development of a physiologically based pharmacokinetic model for assessment of human exposure to bisphenol A

    Toxicol. Appl. Pharmacol.

    (2015)
  • S.L. Bertelsen et al.

    Evaluation of log K-ow and tissue lipid content as predictors of chemical partitioning to fish tissues

    Environ. Toxicol. Chem.

    (1998)
  • Cited by (0)

    View full text