Organophosphate ester flame retardants have antiandrogenic potential and affect other endocrine related endpoints in vitro and in silico

Highlights

• Tested OPFRs have endocrine disruptive potential in at least one of the models.

• Androgen and aryl hydrocarbon receptor activity was affected by most tested OPFRs.

• Some OPFRs displaced transthyretin bound ligand and changed estradiol levels.

• Specific chemicals traits may play a role in AR activity and lack of TTR binding.

Abstract

Organophosphate ester flame retardants (OPFRs) are used to prevent ignition and spreading of fire. They are present in various human matrices suggesting adult, fetal, and neonate exposure. Endocrine related effects have been observed in vivo, but information at the molecular level is lacking for some OPFRs. Also, a better understanding of potential contribution from chemical substructures is needed. The aim of this study was to screen OPFRs for endocrine disruptive potential in vitro and in silico. We selected eleven substances to represent some OPFRs with 1) little information on endocrine activity and others to represent 2) varied chemical substructures. We used in vitro assays for androgen receptor (AR), aryl hydrocarbon receptor (AhR), and Nrf2 activity, effects on steroidogenesis, and transthyretin (TTR) binding, as well as in silico models covering estrogen, thyroid, and CYP3A4 induction related endpoints. Ten OPFRs affected AR and AhR activity, seven affected TTR binding, and five affected 17β-estradiol levels. Several substances had IC₅₀-values below 10 μM and exhibited efficacious effects. These included TPHP, CDP, TMPP, TIPPP, and EHDPP for AR antagonism, suggesting that the degree of arylation and the size of the substance can play a role for the activity. Chlorinated OPFRs had low/no effect on TTR binding. No clear trend was observed for AhR and steroidogenesis, but all arylated OPFRs were predicted to have alert for estrogen receptor binding in an in silico model with metabolism simulator included. Collectively, our data suggest that OPFRs have endocrine disruptive potential warranting further studies to enable human risk assessment.

Introduction

Flame retardants (FRs) are substances used in industrial applications and consumer products to prevent materials to catch fire and to slow spreading of fire (EFRA, 2007). The manufacture and use of FRs has been ongoing for many decades, for instance polychlorinated biphenyls have been used as FRs since the late 1920s (Bergman et al., 2012). Over the last 15 years, the production and use of some organophosphate ester flame retardants (OPFRs) has increased (Greaves and Letcher, 2017). Chemically they belong to the group of organophosphate esters (OPEs), which are characterized by the presence of a phosphate group that is bound to alkyl or aryl groups. The degree of arylation, halogenation, as well as the length and branching of side-chains vary, and therefore as such the group is chemically diverse.

OPFRs are additives not chemically bound to materials (van der Veen and de Boer, 2012) and can therefore transfer to the surrounding environment. They have been measured in matrices such as dust (Brommer and Harrad, 2015; He et al., 2018c; Larsson et al., 2018; Zeng et al., 2018), indoor air (He et al., 2018a; Persson et al., 2018; Takeuchi et al., 2018; Vykoukalová et al., 2017), foods (Poma et al., 2017), and water (Kim and Kannan, 2018). In concert, human biomonitoring studies from the US, Australia, EU, or China have found OPFR or metabolites thereof in urine (He et al., 2018b; Ospina et al., 2018; Sun et al., 2018; Van den Eede et al., 2013), placenta (Ding et al., 2016), and/or breast milk (He et al., 2018b; Sundkvist et al., 2010) with some urinary metabolites present in >60% of samples (Ospina et al., 2018). Collectively, this suggest that human exposure to OPFR is widespread and that the fetus and neonate could be exposed through placental transfer or breast feeding. Consequently, it is vital that sufficient knowledge on potential hazards is available.

There are studies indicating that exposure to OPFRs may lead to adverse effects originating from disruption of the endocrine system such as reproductive and neurodevelopmental effects. OPFR levels in humans and matrices of human relevance are suggested to be associated with preterm delivery of daughters (Hoffman et al., 2018), affected sperm parameters, and increased blood prolactin levels in adult men (Meeker and Stapleton, 2010). Further, OPFRs are associated with reduced IQ and working memory in children exposed during pregnancy (Castorina et al., 2017), changed social behavior in children (Lipscomb et al., 2017), and affected thyroid hormone levels in adults (Meeker and Stapleton, 2010; Preston et al., 2017). These findings are corroborated by rat studies with observations of increased thyroid hormone levels in dams, advanced puberty onset in female offspring, as well as behavioral changes caused by exposure during gestation and lactation (Baldwin et al., 2017; Patisaul et al., 2013). Similarly, in fish, reproductive parameters were affected (Li et al., 2019, 2018; Liu et al., 2013, 2012), as well as sex hormone levels, and expression of vital genes of the steroid hormone system (Liu et al., 2013, 2012). Further, hypo- and hyper-activity, altered exploration patterns (Glazer et al., 2018; Jarema et al., 2015; Noyes et al., 2015; Oliveri et al., 2015), and altered thyroid hormone levels have been reported in zebrafish (Wang et al., 2013). Collectively, these findings suggest that OPFRs induce reproductive and neurodevelopmental effects by an endocrine mechanism of action. However, knowledge is still lacking with respect to the molecular endocrine disruptive potential of some OPFRs, as well as whether specific chemical substructures of OPFRs characterize the endocrine activity.

The aim of this study was to screen OPFRs for endocrine activity in vitro and in silico. Eleven OPFRs were selected including compounds with known endocrine disruptive activity in vitro and compounds with little in vitro hazard information. In addition, the compounds were also selected to represent diverse chemical structural traits. We tested for effects on 1) androgen receptor activity linked with effects on male health (Luccio-Camelo and Prins, 2011), 2) transthyretin binding linked with thyroid hormone transport (Zoeller et al., 2007), 3) aryl hydrocarbon receptor activity linked with induction of liver metabolizing enzymes (Ma, 2008), 4) sex steroid hormone synthesis linked with hormone level imbalances and potential developmental and reproductive effects, and 5) an indirect measure of oxidative stress – Nrf2 activity (Hayes and Dinkova-Kostova, 2014) - linked to e.g. effects on sperm (Dutta et al., 2019; Gharagozloo and Aitken, 2011). This study will provide new information on endocrine disrupting activity of some members of this chemical class, allow for directly comparing effects between members of the class, as well as examine whether specific chemical traits can explain the observed activity.