(2-Ethylhexyl)Terephthalate Metabolites on Thyroid Receptors: in Vitro and in Silico Studies

H OH metabolites OH

Article Comparative Effects of Di-(2-ethylhexyl)phthalate and Di-(2-ethylhexyl)terephthalate Metabolites on Thyroid Receptors: In Vitro and In Silico Studies

Nicolas Kambia 1,†, Isabelle Séverin 2,†, Amaury Farce 3, Laurence Dahbi 2, Thierry Dine 1, Emmanuel Moreau 4 , Valérie Sautou 5,* and Marie-Christine Chagnon 2

1 Université de Lille, CHU Lille, ULR 7365 GRITA, F-59000 Lille, France; nicolas.kambia-kpakpaga@univ-lille (N.K.); [email protected] (T.D.) 2 Université Bourgogne Franche-Comté, INSERM U1231, NUTOX, Derttech “Packtox”, 21000 Dijon, France; [email protected] (I.S.); [email protected] (L.D.); [email protected] (M.-C.C.) 3 Université de Lille, CHU Lille, INSERM U1286, INFINITE, F-59000 Lille, France; [email protected] 4 Université Clermont Auvergne, INSERM U1240, IMOST, 63000 Clermont-Ferrand, France; [email protected] 5 Université Clermont Auvergne, CHU Clermont Ferrand, CNRS, SIGMA Clermont, 63000 Clermont-Ferrand, France * Correspondence: [email protected] or [email protected]; Tel.: +33-4731-78021 † Both authors contributed equally.

Abstract: Plasticizers added to polyvinylchloride (PVC) used in medical devices can be released into patients’ biological fluids. Di-(2-ethylhexyl)phthalate (DEHP), a well-known reprotoxic and Citation: Kambia, N.; Séverin, I.; endocrine disruptor, must be replaced by alternative compounds. Di-(2-ethylhexyl) terephthalate Farce, A.; Dahbi, L.; Dine, T.; (DEHT) is an interesting candidate due to its lower migration from PVC and its lack of reprotoxicity. Moreau, E.; Sautou, V.; However, there is still a lack of data to support the safety of its human metabolites with regard to their Chagnon, M.-C. Comparative Effects hormonal properties in the thyroid system. The effects of DEHT metabolites on thyroid/hormone of Di-(2-ethylhexyl)phthalate and receptors (TRs) were compared in vitro and in silico to those of DEHP. The oxidized metabolites Di-(2-ethylhexyl)terephthalate of DEHT had no effect on T3 receptors whereas 5-hydroxy-mono-(ethylhexyl)phthalate (5-OH- Metabolites on Thyroid Receptors: In MEHP) appeared to be primarily an agonist for TRs above 0.2 µg/mL with a synergistic effect Vitro and In Silico Studies. Metabolites 2021, 11, 94. https://doi.org/ on T3. Monoesters (MEHP and mono-(2-ethylhexyl)terephthalate, MEHT) were also active on T3 10.3390/metabo11020094 receptors. In vitro, MEHP was a partial agonist between 10 and 20 µg/mL. MEHT was an antagonist at non-cytotoxic concentrations (2–5 µg/mL) in a concentration-dependent manner. The results Academic Editor: Sandra Coecke obtained with docking were consistent with those of the T-screen and provide additional information Received: 29 November 2020 on the preferential affinity of monoesters and 5-OH-MEHP for TRs. This study highlights a lack of Accepted: 5 February 2021 interactions between oxidized metabolites and TRs, confirming the interest of DEHT. Published: 10 February 2021 Keywords: DEHT; in silico; T-screen assay; hormonal activities; thyroid receptors Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affil- iations. 1. Introduction Polyvinyl chloride (PVC) is a material widely used in medical devices (MDs), including infusion sets or lines, feeding tubes and tubing, umbilical catheters, oxygen masks, endotracheal tubes, blood transfusers and bags or extracorporeal circuits. Plasticizers Copyright: © 2021 by the authors. are added to the polymer to improve the flexibility and softness of the PVC. However, Licensee MDPI, Basel, Switzerland. since they are not covalently bonded to the PVC matrix, they can easily migrate from the This article is an open access article MD and come into contact with patients during medical procedures [1,2]. Neonates in distributed under the terms and intensive care units are known to be exposed to one of these plasticizers, di-(2-ethylhexyl) conditions of the Creative Commons phthalate (DEHP) present in many MDs [3–5]. This phthalate is now classified as a CMR1B Attribution (CC BY) license (https:// (Carcinogenic Mutagenic or Reprotoxic) substance under the Classification Labeling and creativecommons.org/licenses/by/ Packaging (CLP) Regulation [6] due to its effects on reproduction and fertility. The use 4.0/).

Metabolites 2021, 11, 94. https://doi.org/10.3390/metabo11020094 https://www.mdpi.com/journal/metabolites Metabolites 2021, 11, 94 2 of 18

of DEHP in PVC MDs has been called into question by the European authorities and has been restricted for several years. It must now not exceed 0.1% by mass of the plasticized material, as defined by European Regulation n◦2017/745 on MDs [7]. Other plasticizers, such as di-(2-ethylhexyl)terephthalate (DEHT), have been proposed to replace DEHP to soften the PVC in MDs [1]. This additive is particularly interesting because less of it is released from the PVC MDs than DEHP [2], and therefore there is less exposure. DEHT is less active than DEHP in inducing peroxisome proliferation in rats, which can be explained by the small amount of monoester produced during DEHT metabolism [8]. DEHT is principally hydrolyzed to both terephthalic acid (TPA) and 2-ethylhexanol (EH), with both metabolites being rapidly removed in vivo. This extensive hydrolysis of DEHT to TPA and EH allows only a small fraction to be converted into the monoester and then ultimately to the corresponding oxidized metabolites [9]. MEHT exhibits a lower cytotoxicity than MEHP and its cytotoxicity occurs (0.05 mg/mL) at a much higher concentration than those measured in body fluids [10]. In animal studies, DEHT has shown no reprotoxic effects, a low developmental toxicity and no genotoxicity [9,11]. However, toxicity data are not complete as there is a lack of information regarding the hormonal activities of DEHT and/or its metabolites resulting from in vivo hydrolysis and oxidation. It is, however, very important to assess these activities on hormones since they can occur at very low doses and can have a significant impact on the development of children when they are exposed during critical periods of their development. In a previous study, we performed an in vitro investigation on the potential endocrine-disrupting effects of DEHT and its metabolites on estrogen and androgen receptors and on steroid synthesis. This study demonstrated that DEHT and its metabolites exhibit much weaker effects on hormonal activities than DEHP. However, special attention must be paid to the 5-hydroxy metabolite of mono-(ethylhexyl)terephthalate (5-OH-MEHT) due to co-stimulation of estrogen alpha and human androgen receptors and an increase in estrogen synthesis [12]. To date, data are lacking the effects of DEHT and its ultimate metabolites on thyroid hormonal activity. Phthalates such as DEHP may affect target points in the hypothalamic– pituitary–thyroid axis, such as iodine uptake in the thyroid gland, thyroid hormone synthesis, binding to thyroid hormone receptor or protein transport proteins in blood, biotransforma- tion and excretion and hypothalamic–pituitary control of thyroid hormone production. They may have multiple and possibly overlapping target points, sometimes acting as agonist or antagonist [13]. Ghisari et al. showed that DEHP and other phthalates affected the thyroid hormone-dependent GH3 cell growth using a rat pituitary tumor cell line. A concentration- dependent GH3 cell proliferation was observed with DEHP. Co-treatment of GH3 cells with DEHP and the T3-EC50 inhibited the T3-induced cell proliferation compared to T3-EC50 control [14]. Some studies have demonstrated a correlation between the exposure to DEHP metabolites and thyroid function. The meta-analysis conducted by Huang et al. showed that DEHP exposure can decrease the T4 and increase thyroid-stimulating hormone (TSH). The results suggested that DEHP can affect function in children, adults and pregnant women. Huang and al. concluded that early life phthalate exposure was associated with decreased thyroid hormone levels in young children [15]. In another meta-analysis Kim et al. highlighted that urinary MEHP and 5-OH-MEHP concentration were negatively correlated with total T4, and urinary mono-(2-ethyl-5-oxohexyl)phthalate (5-oxo-MEHP) concentration was positively correlated with thyrotropin [16]. Villanger et al. found that DEHP exposure is associated with thyroid function in mid-pregnancy among Norwegian women. High loadings of DEHP metabolites were associated with a decrease of T3 [14]. These effects may have an impact on neurodevelopment in children. Indeed, the development of the nervous system is extremely dependent on thyroid hormones during the in utero period and the first two years of life, with critical windows of vulnerabil- ity [17–20]. Newborns and premature neonates hospitalized in intensive care units are particularly vulnerable. It is therefore very important to assess whether plasticizers from MDs, and the corresponding metabolites found in the body, affect the thyroid and the active concentrations. Metabolites 2021, 11, x FOR PEER REVIEW 3 of 18

[17–20]. Newborns and premature neonates hospitalized in intensive care units are particularly vulnerable. It is therefore very important to assess whether plasticizers from Metabolites 2021, 11, 94 MDs, and the corresponding metabolites found in the body, affect the thyroid and3 ofthe 18 active concentrations. The objective of this study was to use in silico and in vitro methodologies to assess the effects of DEHT metabolites on thyroid hormonal activities and to compare them with The objective of this study was to use in silico and in vitro methodologies to assess those of DEHP, as both plasticizers are highly metabolized. the effects of DEHT metabolites on thyroid hormonal activities and to compare them with those of DEHP, as both plasticizers are highly metabolized. 2. Results 2.1.2. Results Access to the Secondary Metabolites 5-OH-MEHT, 5-oxo-MEHT and 5-cx-MEHT 2.1. AccessMetabolites to the Secondary5-OH-MEHT, Metabolites 5-oxo-MEHT 5-OH-MEHT, and 5-cx-MEHT, 5-oxo-MEHT mono-(2-ethyl-5-carboxy- and 5-cx-MEHT pentyl)terephthalate,Metabolites 5-OH-MEHT, were synthesized 5-oxo-MEHT andfrom 5-cx-MEHT, 2-ethylhex mono-(2-ethyl-5-carboxypentyl)-5-en-1-ol (1) and 4-((benzyloxy)carbonyl)benzoicterephthalate, were synthesized acid (2), from previously 2-ethylhex-5-en-1-ol synthesized (1) and and characterized 4-((benzyloxy)carbonyl) (1H, 13C NMR ben- andzoic acidHRMS) (2), previouslyby Nüti et synthesized al. and our and lab characterized INSERM U1240 (1H, 13 [10,21],C NMR respectively and HRMS) by(SchemeNüti et 1). al. Derivativesand our lab (4) INSERM and (7) U1240 were obtained [10,21], respectively by Wacker (Schemeoxidation1). of Derivatives the vinylic (group4) and at (7ω)-posi- were tionobtained of compound by Wacker (3) oxidation in a mixture of the vinylic of PdCl group2 and at parabenzoquinoneω-position of compound or a hydroboration (3) in a mixture reaction,of PdCl2 andrespectively. parabenzoquinone Compound or a(4) hydroboration was then converted reaction, into respectively. 5-oxo-MEHT Compound (5) by (hydro-4) was genation,then converted with the into ketone 5-oxo-MEHT group then (5) bybeing hydrogenation, reduced with with NaBH the4 to ketone form group5-OH-MEHT then being (6). Finally,reduced compound with NaBH4 (7)to was form successively 5-OH-MEHT oxidiz (6). Finally,compounded (Jones reagent) (7) and was reduced successively (Method oxidized A; or(Jones inverse reagent) for Method and reduced B) to (Methodgive 5-cx-MEHT A; or inverse (9). The for Methodpurity of B) all to synthesized give 5-cx-MEHT metabolites (9). The waspurity over of all95% synthesized (HPLC/MS). metabolites was over 95% (HPLC/MS).

Scheme 1. Synthesis pathways of 5-oxo-MEHT, 5-hydroxy metabolite of mono-(ethylhexyl)terephthalate (5-OH-MEHT) Scheme 1. Synthesis pathways of 5-oxo-MEHT, 5-hydroxy metabolite of mono-(ethylhexyl)terephthalate (5-OH-MEHT) and 5-cx-MEHT. and 5-cx-MEHT. 2.2. Impact on Thyroid-Dependent Cell Growth 2.2. Impact on Thyroid-Dependent Cell Growth The T-screen assay was used as a fast and functional assay to assess interference with The T-screen assay was used as a fast and functional assay to assess interference with T3 receptors (agonistic or antagonistic potency of xenobiotics at cellular level (Figure1)). T3 receptorsA concentration-dependent (agonistic or antagonistic antagonist potency effect of xenobiotics starting at 2atµ cellularg/mL was level observed, (Figure 1)). be- comingA concentration-dependent significant at 5 µg/mL with antagonist MEHT. Aboveeffect starting 5 µg/mL, at MEHT2 µg/mL was was cytotoxic observed, for thebe- comingcell line. significant MEHT metabolites at 5 µg/mL had with no MEHT. effect. Above 5 µg/mL, MEHT was cytotoxic for the cell line.In contrast, MEHT metabolites MEHP was ahad partial no effect. agonist (resulted in a small but significant increase of cell growthIn contrast, (<15%) MEHP between was 10 a partialµg/mL agonist and 20 µ(resg/mL.ulted Derived in a small oxidative but significant metabolites, increase such ofas cell OH growth metabolite, (<15%) were between 2-fold more 10 µg/mL active, and with 20 a significantµg/mL. Derived effect atoxidative lower concentrations metabolites, such(0.2 µ asg/mL, OH metabolite, 2 µg/mL and were 5 µ 2-foldg/mL) more and withactive, a concentrationwith a significant dependency. effect at lower A synergistic concen- trationsresponse (0.2 was µg/mL, also observed 2 µg/mL whenand 5 cellsµg/mL) were and co-treated with a concentration with T3 up to dependency. a concentration A syn- of ergistic10 µg/mL. response 5-oxo-MEHP was also significantly observed when inhibited cells cellwere growth co-treated at 10 withµg/mL T3 up in theto a presenceconcentra- of tionT3 and of 10 at µg/mL. non-cytotoxic 5-oxo-MEHP concentrations. significantly Higher inhibited concentrations cell growth of at 5-oxo-MEHP 10 µg/mL in were the pres- very encecytotoxic of T3 for and the at cells. non-cytotoxic The CX metabolite concentrations. had no effect.Higher concentrations of 5-oxo-MEHP were very cytotoxic for the cells. The CX metabolite had no effect.

Metabolites 2021, 11, x FOR PEER REVIEW 4 of 18 Metabolites 2021, 11, 94 4 of 18

Figure 1. Effects of plasticizer metabolites on the T-screen assay. GH3 cells were exposed for 96 h to Figure 1. Effects of plasticizer metabolites on the T-screen assay. GH3 cells were exposed for 96 h increasing concentrations of the chemicals, either alone or in the presence of T3. Cell proliferation was to increasing concentrations of the chemicals, either alone or in the presence of T3. Cell prolifera- tionexpressed was expressed relative relative to the maximum to the maximum response respon observedse observed at 10 nM at T3 10 (innM agonist T3 (in agonist mode) or mode) 0.25 nMor T3 0.25(in nM antagonist T3 (in antagonist mode). The mode). response The forresponse the solvent for the control solvent was control set at 0%.was Theset at values 0%. The are valuesmean ± SD areof mean at least ± SD two of independent at least two experimentsindependent performed experiments in triplicate.performed * Signiﬁcantlyin triplicate. different* Significantly from controldif- ferent(* p < from 0.05, control ** p < 0.01 (* p and< 0.05, *** **p

Metabolites 2021, 11, 94 5 of 18 Metabolites 2021, 11, x FOR PEER REVIEW 5 of 18

2.3. Docking T3 was docked into the two subtypes to ve verifyrify the docking protocol. A A single single confor- confor- mation was achieved, which was almost superi superimposablemposable with the co-crystallized conformation. It was tightly bound by a strong ionic interaction between the acid and arginine 228 and 262262 atat thethe basebase ofof the the pocket. pocket. At At the the other other extremity, extremity, the the phenol phenol formed formed a hydrogen a hydro- genbond bond with with His381. His381. The onlyThe differenceonly difference was a wa slights a twistslight of twist the acidof the chain acid to chain better to interact better withinteract the with arginines the arginines (Figure2 (Figure). 2).

Figure 2. Docking of T3 (green) vs. its crysta crystallographicllographic position (red) in 4lnw.

DEHP and DEHT barely fit fit into the pocket. It was clearly a misfitmisfit due to the sheer volume of thethe pocket,pocket, whichwhich waswas able able to to accommodate accommodate the the bulk bulk of of the the diesters. diesters. However, However, it wasit was not not possible possible to to find find a single a single pose, pose, as theas the aromatics aromatics were were only only loosely loosely positioned positioned in the in thesame same area area of the of the pocket, pocket, but but absolutely absolutely not no int ain preferred a preferred conformation conformation and and fully fully lacking lack- inginteractions interactions with with the receptor.the receptor. This This was morewas more marked marked for DEHT for DEHT than forthan DEHP, for DEHP, although alt- houghcommon common to the two to the molecules. two molecules. The monoesters fared well. MEHT showed a single conformation, with an excellent conservation of the position of the aromatic ring close to arginines 228, 262 and 266, which interacted withwith itsits freefree acid acid via via strong strong salt salt bridges. bridges. The The remaining remaining ester ester side side chain chain occupied occu- piedthe other the other side ofside the of pocket the pocket and, withand, onlywith aon smallly a small amount amount of fluctuation, of fluctuation, was positioned was posi- tionedabove Hisabove 381. His Compared 381. Compared to T3, theto T3, aromatic the aromatic group ofgroup MEHT of MEHT was at thewas opposite at the opposite end of endthe pocket.of the pocket. MEHP, MEHP, in contrast, in contrast, fitted into fitted the bindinginto the site binding with itssite aromatic with its group aromatic very group close veryto the close position to the of position the distal of phenylthe distal of thephenyl hormone. of the Thehormone. free acid The was free also acid able was to also bind able to toHis bind 381, to although His 381, the although orientation the orientation of the interaction of the interaction was not perfect was not in the perfect crystallographic in the crys- tallographicconformation conformation of the histidine of sidethe histidine chain. The side ester chain. occupied The ester the part occupied of the pocketthe part close of the to pocketthe arginines close to (Figure the arginines.3). (Figure 3) The hydroxylation of the monoester metabolite in 5-OH-MEHT did not modify its placement in the pocket. However, the hydroxyl at the end of the ester chain only formed inconsistent hydrogen bonds with the skeleton of Gly290, which may be deleterious due to the rather hydrophobic nature of the pocket around this residue. 5-OH-MEHP behaved in much the same way, keeping the same position as its parent but with the acid binding to His381 and the hydroxyl group at the other end of the molecule forming hydrogen bonds with the skeleton of Met259 and, occasionally, with that of Ala283 (Figure4).

Metabolites 2021, 11, x FOR PEER REVIEW 5 of 18

2.3. Docking T3 was docked into the two subtypes to verify the docking protocol. A single conformation was achieved, which was almost superimposable with the co-crystallized conformation. It was tightly bound by a strong ionic interaction between the acid and arginine 228 and 262 at the base of the pocket. At the other extremity, the phenol formed a hydrogen bond with His381. The only difference was a slight twist of the acid chain to better interact with the arginines (Figure 2).

Figure 2. Docking of T3 (green) vs. its crystallographic position (red) in 4lnw.

DEHP and DEHT barely fit into the pocket. It was clearly a misfit due to the sheer volume of the pocket, which was able to accommodate the bulk of the diesters. However, it was not possible to find a single pose, as the aromatics were only loosely positioned in the same area of the pocket, but absolutely not in a preferred conformation and fully lacking interactions with the receptor. This was more marked for DEHT than for DEHP, although common to the two molecules. The monoesters fared well. MEHT showed a single conformation, with an excellent conservation of the position of the aromatic ring close to arginines 228, 262 and 266, which interacted with its free acid via strong salt bridges. The remaining ester side chain occupied the other side of the pocket and, with only a small amount of fluctuation, was positioned above His 381. Compared to T3, the aromatic group of MEHT was at the opposite end of the pocket. MEHP, in contrast, fitted into the binding site with its aromatic group very close to the position of the distal phenyl of the hormone. The free acid was also able Metabolites 2021, 11, 94 to bind to His 381, although the orientation of the interaction was not perfect in the 6crys- of 18 tallographic conformation of the histidine side chain. The ester occupied the part of the pocket close to the arginines. (Figure 3)

Metabolites 2021, 11, x FOR PEER REVIEW 6 of 18

Figure 3. Docking of MEHP (left) and MEHT (right) in TRα1 (4lnw).

The hydroxylation of the monoester metabolite in 5-OH-MEHT did not modify its placement in the pocket. However, the hydroxyl at the end of the ester chain only formed inconsistent hydrogen bonds with the skeleton of Gly290, which may be deleterious due to the rather hydrophobic nature of the pocket around this residue. 5-OH-MEHP behaved in much the same way, keeping the same position as its parent but with the acid binding

Figureto His381 3. Docking and the of MEHP hydroxyl (left) group and MEHT at the (right other) in TRendα1 of(4lnw). the molecule forming hydrogen bonds with the skeleton of Met259 and, occasionally, with that of Ala283 (Figure 4)

Figure 4. Docking of 5-OH-MEHP (left) and 5-OH-MEHT (right) in TRα1 (4lnw). Figure 4. Docking of 5-OH-MEHP (left) and 5-OH-MEHT (right) in TRα1 (4lnw).

The docking results were therefore fairly comparablecomparable so the discussion of the docking will focus on TRα, whichwhich beneﬁtsbenefits fromfrom aa betterbetter resolution.resolution. The conclusionsconclusions drawn are exactly the same for TR β1.

3. Discussion Discussion 3.1. Impact Impact on Thyroid Hormones Thyroid hormones have a wide range of biological effects in vertebrates, during both Thyroid hormones have a wide range of biological effects in vertebrates, during both fetal and prenatal development and with regard to the development of sex organs and the fetal and prenatal development and with regard to the development of sex organs and the central nervous system in mammals [22,23]. The T-screen is based on thyroid hormone- central nervous system in mammals [22,23]. The T-screen is based on thyroid hormone- dependent cell growth of a rat pituitary tumor cell line and is used as a model to study basic dependent cell growth of a rat pituitary tumor cell line and is used as a model to study thyroid hormone-dependent cell physiology, and to study the interference of compounds basic thyroid hormone-dependent cell physiology, and to study the interference of com- with thyroid hormones at a cellular level [24,25]. pounds with thyroid hormones at a cellular level [24,25]. For the DEHT metabolites, only MEHT was an antagonist for cell line proliferation For the DEHT metabolites, only MEHT was an antagonist for cell line proliferation at very low concentrations. In contrast, DEHP metabolites were agonists, particularly at very low concentrations. In contrast, DEHP metabolites were agonists, particularly the the hydroxylated metabolite (5-OH-MEHP), demonstrating a synergism when cells were hydroxylated metabolite (5-OH-MEHP), demonstrating a synergism when cells were coco-treated with T3 up to a concentration of 10 µg/mL. 5-oxo-MEHP only inhibits cell treated with T3 up to a concentration of 10 µg/mL. 5-oxo-MEHP only inhibits cell growth growth at 10 µg/mL and was very cytotoxic for the cells above this concentration. Our atdata 10 agreeµg/mL with and datawas very published cytotoxic by Ghisarifor the ce thatlls demonstratedabove this concentration. a low dependent Our data potency agree withactivation data published of DEHP between by Ghisari 10− that6 M demonstrated and 10−5 M. However, a low dependent Ghisari etpotency al. did activation not examine of −6 −5 DEHP between metabolites 10 [M14 ].and In 10 a TR M. reporter However, gene Ghisari assay et using al. did a recombinant not examine XenopusDEHP metab- laevis olites [14]. In a TR reporter gene assay using a recombinant Xenopus laevis cell line, benzyl butyl phthalate (BBP), dibutyl phthalate (DBP) and DEHP were reported to exhibit a T3- antagonistic activity, and to inhibit the expression of the endogenous TRβ gene [26]. DEHP was also shown to interfere with the binding of T3 to TRβ [27], suggesting that the compound may bind to the receptor. Rodent studies have also confirmed the effects of DEHP on the thyroid [28,29], where significant influences on thyroid hormones and metabolism were observed. Specifically, proliferative changes were noted in the thyroid in vivo, raising the concern of a potential thyroid carcinogenicity of DEHP. Recently, Kim et al. investigated whether DEHP could induce proliferative changes and DNA damage in

Metabolites 2021, 11, 94 7 of 18

cell line, benzyl butyl phthalate (BBP), dibutyl phthalate (DBP) and DEHP were reported to exhibit a T3-antagonistic activity, and to inhibit the expression of the endogenous TRβ gene [26]. DEHP was also shown to interfere with the binding of T3 to TRβ [27], suggesting that the compound may bind to the receptor. Rodent studies have also confirmed the effects of DEHP on the thyroid [28,29], where significant influences on thyroid hormones and metabolism were observed. Specifically, proliferative changes were noted in the thyroid in vivo, raising the concern of a potential thyroid carcinogenicity of DEHP. Recently, Kim et al. investigated whether DEHP could induce proliferative changes and DNA damage in 8505C thyroid carcinoma cell lines, both in vitro and in the thyroid tissue of rats treated orally with DEHP for 90 days from juvenile to full maturation in vivo [16]. They showed that DEHP can stimulate thyroid cell proliferation and DNA damage through the activation of the TSHR pathway, as TSHR plays a key role in the proliferation and differentiation of thyroid cells [30]. All in vitro and in vivo data suggested that DEHP is able to influence thyroid tissues at low doses. Although studies on the in vitro effects of plasticizer metabolites on the thyroid hormone (TH) system are scarce, it is well known that diesters are highly metabolized in vivo. Reduced serum thyroid hormone levels have been well documented in human populations, with higher urine phthalate metabolites in various regions around the globe [31–34]. In this study, MEHP and 5-OH-MEHP appear partial agonists on the GH3 proliferation, 5-OH-MEHP being more active. These data are in accordance with Ghisary et al.’s [14] work showing that DEHP can induce the proliferation of the GH3 cells. However, they did not check if the DEHP hydrolysis can occur in the experimental conditions. We observed that 5-OH-MEHP, but not MEHP, also induced a decrease in the presence of T3 at the highest non-cytotoxic concentration, whereas below this concentration we saw a synergism at 5 µg/mL. In this assay, cell proliferation was measured as a consequence of T3 activation, but the cell proliferation is a complex biological phenomenon, not just TR-dependent. For example, GH3 cells also express peroxisome proliferator-activated receptors (PPARs), which exhibit anti-proliferative activity upon ligand binding [35]. In this study, we can assume that the observed response is linked to T3, as induction by T3 was modulated in the presence of 5-OH-MEHP when the cells were co-exposed. Interestingly, human studies have found an inverse association between MEHP metabolites in the urine and concentrations of free T4 and total T3 levels in adult men [33], which suggests that DEHP can disrupt the homeostasis of the thyroid–pituitary axis.

3.2. Docking DEHP and DEHT were unable to bind to TRα, only occupying the pocket due to a volume barely large enough to accommodate them. It can be concluded that they are not able to directly interact with TRs. On the contrary, the monoesters MEHP and MEHT displayed a rather strong binding mode in the pocket, with a single coherent conformation found for each. It is noteworthy that MEHP, with its aromatic group in close proximity to His381, closely imitates the binding mode of the natural T3 agonist despite lacking any strong interaction with the arginines. On the other hand, MEHT, presenting its acid at the entry of the pocket in front of the arginines, has a different binding mode in which the aromatic is at the opposite end to His 381 and has no interactions with this part of the binding site. The presence of a hydroxyl group on the monoester metabolite did not alter its placement in the pocket, with 5-OH-MEHP once again being closer to the natural agonist and 5-OH-MEHT positioning its aromatic group the other way around, far from His381. This conformation is also somewhat destabilized by the poor hydrophobic ﬁt between the added hydroxyl group and the rather hydrophobic area around His381. Further oxidation of the hydroxyl to a carbonyl group led to the same positioning of the derivatives. In the case of 5-oxo-MEHP, the same position close to His381 was maintained, and the carbonyl formed some hydrogen bonds, although it did not exhibit a clear preference, binding to Arg228, the Ser277 skeleton or side chain, or to nothing, Metabolites 2021, 11, 94 8 of 18

with roughly the same propensity. It may therefore be a less than perfect fit for the TR binding site and clearly inferior to the hydroxylated metabolite. The carbonyl group of 5-oxo-MEHT formed no interactions and lay in the middle of the pocket with a poor fit to its surroundings. Interestingly, of the 30 solutions determined for 5-cx-MEHP, no correctly superimposable conformation was identified. Although all were placed in the same area, there was a good deal of fuzziness in their position, most probably due to the different relative positions of the acids in T3 and in 5-cx-MEHP, hindering a perfect fit to the binding site. The addition of a second acid at the end of the remaining ester on 5-cx-MEHT gave the same position as the other compounds derived from DEHT. The acid of the ester side chain lay squarely in the middle of the pocket and was unable to form any interactions. We can therefore assume that it is less capable of binding than its congeners. Table1 summarizes the theoretical and experimental affinities toward TR α1 obtained in silico and in vitro.

Table 1. Theoretical and experimental afﬁnities toward TRα1 of DEHP and DEHT metabolites derived from the in silico and in vitro bioassays data.

In Silico Studies 1 In Vitro Studies Agonist/Antagonist Compounds Studied Affinity toward TRα1 Activities on Cell Proliferation DEHP 0 NT DEHT 0 NT MEHT ++ +++ antagonist or cytotoxic MEHP +++ ++ agonist 5-OH-MEHT +/− 0 5-OH-MEHP ++++ ++++ agonist and synergic activities 5-oxo-MEHT 0 + antagonist 5-oxo-MEHP + ++ antagonist or cytotoxic 5-cx-MEHT 0 0 5-cx-MEHP 0 0 0: no affinity or no activity; +/−: inconclusive affinity or activity; +: low to very low affinity or activity at concentration ≥ 20 µg/mL; ++: low to medium affinity or activity at concentration ≥ 10 µg/mL; +++: medium to strong affinity or activity at concentration ≥ 5 µg/mL; ++++: very strong affinity or activity at concentration ≥ 2 µg/mL; NT: not tested; 1Classification is based on three criteria: The superposition of the conformations (n = 30) of the chemical structures in the TRα1 receptor; The number of bindings between groupings in the structure and the TRα1 receptor; The type of bond (hydrogen or ionic). This makes it possible to estimate a relative potential for interaction between the chemical structure and the target (TRα1).

3.3. In Vitro Data versus Biomonitoring Values In neonatal intensive care units (NICUs), the use of many plasticized PVC medical devices overexposes neonates to phthalates [3–5]. Biomonitoring studies have shown high concentrations of oxidized metabolites of DEHP in the urine of neonates. Urinary levels of 5-OH-MEHP may exceed 0.2 µg/mL, a concentration at which we have demonstrated a TRβ agonist effect that was confirmed in vitro as being partial and has a synergistic effect with T3. The maximum values of 5-OH-MEHP measured in the urine of newborns hospitalized in NICUs range from 0.43 µg/mL to 13.1 µg/mL, i.e., 2 to 65 times higher than the concentration activating thyroid receptors [36–40]. Certain medical procedures, such as extracorporeal circulation (extracorporeal membrane oxygenation, cardiopulmonary bypass), respiratory assistance, and intravenous nutrition are recognized as situations with a high risk of exposure, which can explain the high values in multi-exposed patients. In the previously cited studies, the median values of urinary 5-OH-MEHT concentrations are generally less than 0.2 µg/mL. Calafat et al. (2004) [38] and Green et al. (2005) [39] showed higher urinary levels of 2.22 µg/mL and 0.26 µg/mL, respectively. However, these studies were carried out in 2004 and 2005, and it can be assumed that exposure to DEHP via MDs has decreased in the 15 years following the recommendations of the current Scientific Committee on Emerging and Newly-Identified Health Risks, SCENIHR, and the European regulations. However, a recent study carried out specifically in newborns in cardiac surgery showed that extracoporeal circulation medical devices remain highly Metabolites 2021, 11, 94 9 of 18

exposed to DEHP. Indeed, Gaynor et al. (2019) [41] demonstrated that the level of 5-OH- MEHP passes from 0.01 µg/mL to 0.229 µg/mL after cardiopulmonary bypass. Our study showed that MEHP had partial agonist effects in the T-screen bioassay, suggesting an action on TRs at concentrations of 10 µg/mL to 20 µg/mL. The data in the literature show that the concentrations of MEHP in biological media are significantly lower than these values, including the study by Eckert et al., who directly measured MEHP in blood in contact with extracorporeal circulation lines during heart surgery in newborns. The maximum concentration of MEHP found in the blood of these patients was 0.56 µg/mL [42]. With regard to DEHT metabolites, only MEHT showed an effect on cellular growth with an antagonistic effect on T3 at non-cytotoxic concentrations of 2 µg/mL to 5 µg/mL using the T-screen assay. There is little biomonitoring data for DEHT in the literature, even less measuring the exposure of hospitalized newborns. Lessmann et al. studied the exposure of non-hospitalized children over 4 years of age to DEHT [43]. However, only the oxidized metabolites of MEHT were measured in the urine. Pinguet et al. presented results of the exposure of patients hospitalized in the NICU to certain plasticizers, including DEHT [40]. The median of the urinary concentrations of MEHT in this cohort of patients was lower than the limit of quantification (0.018 ng/mL) and the maximum concentration measured was 9.90 ng/mL, 200 times less than the concentration showing an antagonistic effect on T3 identified using the T-screen assay in this study [40]. In the Armed Neo clinical trial, MEHT levels found in the urine of premature babies hospitalized in the NICU were also much lower than this value, with a maximum of 1.32 ng/mL (unpublished data). Therefore, it would appear unlikely that MEHT levels of 2 µg/mL would be reached in the biological media of hospitalized newborns. DEHT is a plasticizer that has a low migration from PVC medical devices [2,44]. In addition, MEHT, a metabolite resulting from the enzymatic hydrolysis of DEHT, is very rapidly transformed into oxidized derivatives, which have no in vitro effect on T3-dependent cell proliferation. The urinary excretion factor is 0.02% and 6% for MEHT and MEHP, respectively [45].

4. Materials and Methods 4.1. Metabolites of DEHP and DEHT Primary and secondary metabolites of DEHP and DEHT were synthesized and characterized by the IMOST team (UMR 1240, INSERM) in Clermont-Ferrand, France. The compounds tested are shown Table2. The purity of all our synthesized metabolites and their corresponding intermediates exceeded 95% ((High performance liquid chromatography with mass spectrometry (HPLC/MS) and Nuclear magnetic resonnance (NMR)). MEHT was synthesized according to the method described by Eljezi et al. [10]. All synthesis processes are described in the supplementary data (AppendixA). Metabolites 2021, 11, x FOR PEER REVIEW 9 of 18 MetabolitesMetabolites 20212021,, 1111 Metabolites,, xx FORFOR PEERPEER 2021 REVIEWREVIEW, 11, x FOR PEER REVIEW 99 ofof 1818 9 of 18

concentrations of 10 µg/mL to 20 µg/mL. The data in the literature show that the concen- concentrationsconcentrations concentrationsofof 1010 µg/mLµg/mL toto of2020 10 µg/mL.µg/mL. µg/mL TheThe to 20dada tataµg/mL. inin thethe The literatureliterature data in showshow the literature thatthat thethe concen-concen- show that the concentrations of MEHP in biological media are significantly lower than these values, including trationstrations ofof MEHPMEHPtrations inin biologicalbiological of MEHP mediamedia in biological areare significantlysignificantly media are lowerlower significantly thanthan thesethese lower values,values, than includingincluding these values, including the study by Eckert et al., who directly measured MEHP in blood in contact with extra- thethe studystudy byby EckertEckertthe study etet al.,al., by whowho Eckert directlydirectly et al., measuredmeasured who directly MEHPMEHP measured inin bloodblood MEHP inin contactcontact in blood withwith in extra-extra- contact with extracorporeal circulation lines during heart surgery in newborns. The maximum concentra- corporealcorporeal circulationcirculationcorporeal lineslines circulation duringduring heartheart lines susu duringrgeryrgery in inheart newborns.newborns. surgery TheThe in newborns. maximummaximum Theconcentra-concentra- maximum concentration of MEHP found in the blood of these patients was 0.56 µg/mL [42]. tiontion ofof MEHPMEHP foundfoundtion of inin MEHP thethe bloodblood found ofof theseinthese the patients patientsblood of waswas these 0.560.56 patients µg/mLµg/mL was [42].[42]. 0.56 µg/mL [42]. With regard to DEHT metabolites, only MEHT showed an effect on cellular growth WithWith regardregard toto WithDEHTDEHT regard metabolites,metabolites, to DEHT onlyonly metabolites, MEHTMEHT showedshowed only MEHT anan effecteffect showed onon cellularcellular an effect growthgrowth on cellular growth with an antagonistic effect on T3 at non-cytotoxic concentrations of 2 µg/mL to 5 µg/mL withwith anan antagonisticantagonisticwith an effecteffect antagonistic onon T3T3 atat non-cytotoxiceffectnon-cytotoxic on T3 at concentrationsconcentrations non-cytotoxic concentrationsofof 22 µg/mLµg/mL toto 5 5of µg/mLµg/mL 2 µg/mL to 5 µg/mL using the T-screen assay. There is little biomonitoring data for DEHT in the literature, usingusing thethe T-screenT-screenusing assay.assay. the T-screen ThereThere isis assay. littlelittle biomonitoring biomonitoringThere is little biomonitoringdatadata forfor DEHTDEHT data inin thethe for literature,literature, DEHT in the literature, even less measuring the exposure of hospitalized newborns. Lessmann et al. studied the eveneven lessless measuringmeasuringeven lessthethe exposuremeasuringexposure ofof the hospithospit exposurealizedalized ofnewborns.newborns. hospitalized LessmaLessma newborns.nnnn etet al.al. Lessma studiedstudiednn thethe et al. studied the exposure of non-hospitalized children over 4 years of age to DEHT [43]. However, only exposureexposure ofof non-hospitalizednon-hospitalizedexposure of non-hospitalized childrenchildren overover 44 children yearsyears ofof overageage toto4 yearsDEHTDEHT of [43].[43]. age However,However,to DEHT [43].onlyonly However, only the oxidized metabolites of MEHT were measured in the urine. Pinguet et al. presented thethe oxidizedoxidized metabolitesmetabolitesthe oxidized ofof MEHTMEHTmetabolites werewere of measurmeasur MEHTeded were inin thethe measur urine.urine.ed PinguetPinguet in the eturine.et al.al. presentedpresented Pinguet et al. presented results of the exposure of patients hospitalized in the NICU to certain plasticizers, includ- resultsresults ofof thethe exposureexposureresults of ofof the patientspatients exposure hospitalizedhospitalized of patients inin hospitalizedthethe NICUNICU toto certainincertain the NICU plasticizers,plasticizers, to certain includ-includ- plasticizers, including DEHT [40]. The median of the urinary concentrations of MEHT in this cohort of pa- inging DEHTDEHT [40].[40].ing TheThe DEHT medianmedian [40]. ofof Thethethe urinaryurinarymedian concentrationsconcentrationsof the urinary concentrationsofof MEHTMEHT inin thisthis of cohort cohortMEHT ofof in pa-pa- this cohort of patients was lower than the limit of quantification (0.018 ng/mL) and the maximum concen- tientstients waswas lowerlowertients thanthan was thethe limitlowerlimit ofof than quantificationquantification the limit of (0.018 (0.018quantification ng/mL)ng/mL) andand (0.018 thethe ng/mL) maximummaximum and concen- concen-the maximum concentration measured was 9.90 ng/mL, 200 times less than the concentration showing an an- trationtration measuredmeasuredtration waswas measured9.909.90 ng/mL,ng/mL, was 200200 9.90 timestimes ng/mL, lessless thanthan200 timesthethe concentrationconcentration less than the showing showingconcentration anan an-an- showing an antagonistic effect on T3 identified using the T-screen assay in this study [40]. In the Armed tagonistictagonistic effecteffecttagonistic onon T3T3 identifiedidentified effect on usingusing T3 identified thethe T-screenT-screen using assayassay the inT-screenin thisthis studystudy assay [40].[40]. in thisInIn thethe study ArmedArmed [40]. In the Armed Neo clinical trial, MEHT levels found in the urine of premature babies hospitalized in the NeoNeo clinicalclinical trial,trial,Neo MEHTMEHT clinical levelslevels trial, foundfound MEHT inin levelsthethe urineurine found ofof premature prematurein the urine babiesbabies of premature hospitalizedhospitalized babies inin hospitalizedthethe in the NICU were also much lower than this value, with a maximum of 1.32 ng/mL (unpublished NICUNICU werewere alsoalsoNICU muchmuch were lowerlower also thanthan much thisthis lowervalue,value, than withwith this aa maximummaximum value, with ofof 1.32 1.32a maximum ng/mLng/mL (unpublished(unpublished of 1.32 ng/mL (unpublished data). Therefore, it would appear unlikely that MEHT levels of 2 µg/mL would be reached data).data). Therefore,Therefore,data). itit wouldwould Therefore, appearappear it unlikelyunlikely would appear thatthat MEHTMEHT unlikely levelslevels that ofof MEHT 22 µg/mLµg/mL levels wouldwould of 2 bebe µg/mL reachedreached would be reached in the biological media of hospitalized newborns. DEHT is a plasticizer that has a low inin thethe biologicalbiologicalin mediamediathe biological ofof hospitalizedhospitalized media of newbnewb hospitalizedorns.orns. DEHTDEHT newb isis orns.aa plasticizerplasticizer DEHT isthatthat a plasticizer hashas aa lowlow that has a low migration from PVC medical devices [2,44]. In addition, MEHT, a metabolite resulting migrationmigration fromfrommigration PVCPVC medicalmedical from devices devicesPVC medical [2,44].[2,44]. IndevicesIn addition,addition, [2,44]. MEHT,MEHT, In addition, aa metabolitemetabolite MEHT, resultingresulting a metabolite resulting from the enzymatic hydrolysis of DEHT, is very rapidly transformed into oxidized deriv- fromfrom thethe enzymaticenzymaticfrom hydrolysis hydrolysisthe enzymatic ofof DEHT,DEHT, hydrolysis isis vevery ryof rapidlyrapidlyDEHT, transformedistransformed very rapidly intointo transformed oxidizedoxidized deriv-deriv- into oxidized derivatives, which have no in vitro effect on T3-dependent cell proliferation. The urinary ex- atives,atives, whichwhich havehaveatives, nono whichinin vitrovitro have effecteffect no on onin T3-dependentT3-dependentvitro effect on cellcellT3-dependent proliferation.proliferation. cell TheThe proliferation. urinaryurinary ex-ex- The urinary excretion factor is 0.02% and 6% for MEHT and MEHP, respectively [45]. cretioncretion factorfactor isiscretion 0.02%0.02% andandfactor 6%6% is forfor 0.02% MEHTMEHT and andand 6% MEHPforMEHP MEHT,, respectivelyrespectively and MEHP [45].[45]., respectively [45]. 4. Materials and Methods 4.4. MaterialsMaterials andand4. MethodsMethodsMaterials and Methods 4.1. Metabolites of DEHP and DEHT 4.1.4.1. MetabolitesMetabolites of4.1.of DEHPDEHP Metabolites andand DEHTDEHT of DEHP and DEHT Primary and secondary metabolites of DEHP and DEHT were synthesized and char- PrimaryPrimary andand secondarysecondaryPrimary metabolitesandmetabolites secondary ofof DEHP DEHPmetabolites andand DEHTDEHT of DEHP werewere and synthesizedsynthesized DEHT were andand synthesized char-char- and characterized by the IMOST team (UMR 1240, INSERM) in Clermont-Ferrand, France. The acterizedacterized byby thetheacterized IMOSTIMOST byteamteam the (UMR(UMR IMOST 1240,1240, team INSERM)INSERM) (UMR 1240, inin Clermont-Ferrand,Clermont-Ferrand, INSERM) in Clermont-Ferrand, France.France. TheThe France. The compounds tested are shown Table 2. The purity of all our synthesized metabolites and compoundscompounds testedtestedcompounds areare shownshown tested TableTable are 2.2. shown TheThe puripuri Tabletyty ofof 2. allall The ourour puri synthesizedsynthesizedty of all our metabolitesmetabolites synthesized andand metabolites and their corresponding intermediates exceeded 95% ((High performance liquid chromatog- theirtheir correspondingcorrespondingtheir correspondingintermediatesintermediates exceeded exceededintermediates 95%95% ((High((Highexceeded performaperforma 95% ((Highncence liquidliquid performa chromatog-chromatog-nce liquid chromatog- Metabolites 2021, 11, 94 raphy with mass spectrometry (HPLC/MS) and Nuclear magnetic10 of 18 resonnance (NMR)). raphyraphy withwith massmassraphy spectrometryspectrometry with mass (HPLC/MS)(HPLC/MS) spectrometry andand (HPLC/MS) NuclearNuclear magneticmagnetic and Nuclear resonnanceresonnance magnetic (NMR)).(NMR)). resonnance (NMR)). MEHT was synthesized according to the method described by Eljezi et al. [10]. All syn- MEHTMEHT waswas synthesizedsynthesizedMEHT was accordingaccording synthesized toto thethe according methodmethod describedtodescribed the method byby EljeziEljezi described etet al.al. [10].[10].by Eljezi AllAll syn- syn-et al. [10]. All synthesis processes are described in the supplementary data (Appendix A). thesisthesis processesprocessesthesis areare describeddescribed processes inin are thth eedescribed supplementarysupplementary in the supplementary datadata (Appendix(Appendix data A).A). (Appendix A). Table 2. Structures and denominations of DEHP and DEHT metabolites. Table 2. Structures and denominations of DEHP and DEHT metabolites. TableTable 2.2. StructuresStructuresTable andand 2. denominationsdenominations Structures and ofof denominations DEHPDEHP andand DEHTDEHT of DEHP metabolites.metabolites. and DEHT metabolites. DEHP Metabolites DEHT Metabolites DEHP MetabolitesDEHPDEHP MetabolitesMetabolites DEHT Metabolites DEHT DEHT MetabolitesMetabolites DEHP Metabolites DEHTMEHT Metabolites or MEHTP: MEHP:MEHP: mono-(2-ethy mono-(2-ethylhexyl)phthalatelhexyl)phthalate MEHT or MEHTP: mono-(2-ethylhexyl)terephthalate MEHP:MEHP: mono-(2-ethymono-(2-ethyMEHP:lhexyl)phthalatelhexyl)phthalate mono-(2-ethylhexyl)phthalate MEHTMEHT oror MEHTP:MEHTP:mono-(2-ethylhexyl)terephthalateMEHT mono-(2-ethylhexyl)terephthalatemono-(2-ethylhexyl)terephthalate or MEHTP: mono-(2-ethylhexyl)terephthalate

5-oxo-MEHP or MEOHP: mono-(2-ethyl-5- 5-oxo-MEHT or MEOHTP: mono-(2-ethyl-5-oxohexyl)ter- 5-oxo-MEHP5-oxo-MEHP oror5-oxo-MEHP MEOHP:MEOHP: mono-(2-ethyl-5-mono-(2-ethyl-5- or MEOHP: mono-(2-ethyl-5-5-oxo-MEHT5-oxo-MEHT oror5-oxo-MEHT MEOHTP:MEOHTP:5-oxo-MEHT mono-(2-ethyl-5-oxohexyl)ter-mono-(2-ethyl-5-oxohexyl)ter- or or MEOHTP: mono-(2-ethyl-5-oxohexyl)ter- 5-oxo-MEHPoxohexyl)phthalate or MEOHP: ephthalate oxohexyl)phthalateoxohexyl)phthalatemono-(2-ethyl-5-oxohexyl)phthalateoxohexyl)phthalate mono-(2-ethyl-5-oxohexyl)terephthalateephthalateephthalate ephthalate

Metabolites 2021,, 11Metabolites,, xx FORFOR PEERPEER 2021 REVIEWREVIEW, 11, x FOR PEER REVIEW 10 of 18 10 of 18

5-OH-MEHP or MEHHP: mono-(2-ethyl-5-hydroxy- 5-OH-MEHT or MEHHTP: mono-(2-ethyl-5-hydroxy- 5-OH-MEHP5-OH-MEHP oror5-OH-MEHP MEHHP:MEHHP: mono-(2-ethyl-5-hydroxy-mono-(2-ethyl-5-hydroxy- or MEHHP: mono-(2-ethyl-5-hydroxy-5-OH-MEHT5-OH-MEHT oror5-OH-MEHT MEHHTP:MEHHTP: ormono-(2-ethyl-5-hydroxy-mono-(2-ethyl-5-hydroxy- or MEHHTP: MEHHTP: mono-(2-ethyl-5-hydroxy- 5-OH-MEHPhexyl)phthalate or MEHHP: hexyl)terephthalate hexyl)phthalatehexyl)phthalatemono-(2-ethyl-5-hydroxyhexyl)phthalatehexyl)phthalate mono-(2-ethyl-5-hydroxyhexyl)terephthalatehexyl)terephthalatehexyl)terephthalatehexyl)terephthalate

5-cx-MEHP or MECPP:5-cx-MEHP mono-(2-ethyl-5-carboxypen- or MECPP: mono-(2-ethyl-5-carboxypen- 5-cx-MEHP or MECPP: 5-cx-MEHT or MECPTP:5-cx-MEHT5-cx-MEHT mono-(2-ethyl-5-carboxypen- or or MECPTP: mono-(2-ethyl-5-carboxypentyl)phthalatetyl)phthalatemono-(2-ethyl-5-carboxypentyl)phthalate tyl)phthalate mono-(2-ethyl-5-carboxypentyl)terephthalatetyl)terephthalatetyl)terephthalatetyl)terephthalate

4.2. Preparation4.2. Preparation of Samples 4.2.of Samples Preparation of Samples All compoundsAll compounds were dissolvedAll were compounds dissolved in 100% were in ethanol 100% dissolved andethanol tested in and 100% in tested a concentrationethanol in a andconcentration tested range in of a range concentration range 0.2 µg/mLof 0.2 to µg/mL 20 µg/mL. toof 20 0.2 Theµg/mL. µg/mL maximum The to maximum20 µg/mL. concentration Theconc maximumentration of ethanol of conc ethanol inentration the culture in the of culture mediumethanol medium in the culture medium was 0.1%was (vol/vol) 0.1% (vol/vol) in orderwas in 0.1% toorder avoid (vol/vol) to anyavoid cytotoxicin anyorder cytotoxic to effect avoid of effect any the cytotoxicvehicle.of the vehicle. effect of the vehicle.

4.3. T-Screen4.3. T-Screen Assay Assay4.3. T-Screen Assay 4.3.1. Cell4.3.1. Culture Cell Culture and4.3.1. Treatment and Cell Treatment Culture and Treatment The assay was based onThe thyroid assay hormone-dependentwas based on thyroid cell hormone-depe growth of thendent rat pituitarycell growth of the rat pituitary The assay was basedThe assay on thyroid was based hormone-depe on thyroid ndenthormone-depe cell growth◦ndent of the cell rat growth pituitary of the rat pituitary tumor celltumortumor line cellcell GH3 lineline (ATCC, GH3tumorGH3 (ATCC,(ATCC, CCL-82.1).cell line CCL-82.1).CCL-82.1). GH3 The (ATCC, GH3 TheThe cells CCL-82.1). GH3GH3 were cellscells cultured werewereThe GH3 culturedcultured at 37 cellsC atatwere in 3737 a humidi-°C°C cultured inin aa humid-humid- at 37 °C in a humid- fied atmosphere with 5% CO2 in phenol-red-free DMEM/F12 (Gibco-Invitrogen, Fisher ifiedified atmosphereatmosphereified withwith atmosphere 5%5% COCO22 inin with phenol-red-freephenol-red-free 5% CO2 in phenol-red-free DMEM/F12DMEM/F12 (Gibco-Invitrogen,(Gibco-Invitrogen, DMEM/F12 (Gibco-Invitrogen, FisherFisher Fisher Scientific, Illkrich, France) supplemented with 10% (vol/vol) fetal bovine serum (PAN, Scientific, Illkrich,Scientific, France) Illkrich, supplemented France) withsupplemented 10% (vol/vol) with fetal 10% bovine (vol/vol) serum fetal (PAN, bovine Bi- serum (PAN, Bi- Biotech, Dutscher, Brumath, France). Passaging was carried out in 75 cm2 tissue culture otech, Dutscher,otech, Brumath, Dutscher, France). Brumath, Passaging France). was Passagingcarried out was in carried75 cm22 tissuetissueout in culture culture75 cm 2 tissue culture flasks (Falcon) every four days by releasing the cells from the substrate using 0.25% (w/v) flasksflasks (Falcon)(Falcon) everyeveryflasks fourfour(Falcon) daysdays every byby releasingreleasing four days thethe by cellscells releasing fromfrom thethe the substratesubstrate cells from usingusing the substrate 0.25%0.25% (w/v)(w/v) using 0.25% (w/v) trypsin/1 mM EDTA solution (PAN, Biotech, Dutscher, Brumath, France). The T-screen trypsin/1trypsin/1 mMmM EDTAEDTAtrypsin/1 solutionsolution mM EDTA (PAN,(PAN, solution Biotech,Biotech, (PAN, Dutscher, Biotech, Brumath, Dutscher, France). Brumath, The T-screen France). The T-screen was performed as previously described [46]. GH3 cells at 80% confluence were incubated was performedwas as previously performed described as previously [46]. describedGH3 cells at[46]. 80% GH3 confluence cells at 80% were confluence incubated were incubated for 48 h in serum-free PCM medium (which was changed once after 24 h), as originally forfor 4848 hh inin serum-freeserum-freefor 48 h PCM PCMin serum-free mediummedium (whichPCM(which medium waswas changedchanged (which onceonce was afterafterchanged 2424 h),h), once asas originallyoriginallyafter 24 h), as originally described by Sirbasku et al. [47]. PCM consisted of phenol-red-free DMEM/F12 with described by Sirbaskudescribed et byal. [47].Sirbasku PCM et cons al. [47].istedisted PCM ofof phenol-red-freephenol-red-free consisted of phenol-red-free DMEM/F12DMEM/F12 withwith DMEM/F12 1515 with 15 15 mM HEPES (Gibco-Invitrogen, Fisher Scientific, Illkrich, France), 10 µg/mL bovine mM HEPES (Gibco-Invitrogen,mM HEPES (Gibco-Invitrogen, Fisher Scientific, Fisher Illkrich, Scientific, France), Illkrich, 10 µg/mL France), bovine 10 insulin µg/mL bovine insulin insulin (Sigma- Aldrich, St Quentin Fallavier, France, T6634), 10 µM ethanolamine (Sigma- (Sigma-(Sigma- Aldrich,Aldrich,(Sigma- StSt QuentinQuentin Aldrich, Fallavier,Fallavier, St Quentin FranFran ce,Fallavier, T6634), Fran 10 µMce, T6634),ethanolamine 10 µM (Sigma-ethanolamine Al- (Sigma- Al- Aldrich, St Quentin Fallavier, France, E0135), 10 ng/mL sodium selenite (Sigma- Aldrich, drich, St Quentindrich, Fallavier, St Quentin France,, Fallavier, E0135), France,, 10 ng/mL E0135), sodium 10 ng/mLselenite sodium (Sigma- selenite Aldrich, (Sigma- St Aldrich, St St Quentin Fallavier, France, S5261), 10 µg/mL apo-transferrin (Sigma- Aldrich, St Quentin Quentin Fallavier,Quentin France,, Fallavier, S5261), France,, 10 µg/mL S5261), apo-transferrin 10 µg/mL apo-transferrin (Sigma- Aldrich, (Sigma- St Quentin Aldrich, St Quentin Fallavier,Quentin France, Fallavier, T1147),Quentin France,, and Fallavier, 500 S5261),µg/mL France,, 10 bovineµg/mL S5261), serumapo-transferrin 10 albuminµg/mL apo-transferrin(Sigma- (Sigma- Aldrich, Aldrich, (Sigma- St St Quentin Aldrich, St Quentin Fallavier, France,, T1147), and 500 µg/mL bovine serum albumin (Sigma- Aldrich, St QuentinFallavier, Fallavier, France,, France,Fallavier, T1147), A7906). France,, and The 500 cells T1147), µg/mL were and then bovine 500 harvested µg/mLserum inbovinealbumin PCM mediumserum (Sigma- albumin using Aldrich, (Sigma- St Aldrich, St Quentin Fallavier, France,, A7906). The cells were then harvested in PCM medium using a cell scraperQuentin and Fallavier, platedQuentin atFrance,, a density Fallavier, A7906). of France,, 2500 The cells/wellcells A7906). were (100thenThe cellsµharvestedl) on were a 96-well thenin PCM harvested plate medium (Fal- in usingPCM medium using a cell scraper anda cell plated scraper at a anddensity plated of 2500at a densitycells/well of (1002500 µl)cells/well on a 96-well (100 µl) plate on a(Falcon, 96-well plate (Falcon, Dutscher, Brumath,Dutscher, France). Brumath, Following France). an attachment Following periodan attachment of 2 h to period 3 h, the of cells 2 h towere 3 h, the cells were exposed in triplicateexposed and in for triplicate 96 h to andvarious for 96 co ncentrationsh to various coof ncentrationsthe chemicals of to the be chemicals tested to be tested (100(100 µL,µL, 2×2× dosingdosing(100 exposureexposure µL, 2× dosing concentrationconcentration exposure inin concentration PCMPCM medium),medium), in PCM eithereither medium), alonealone oror eitherinin combina-combina- alone or in combina- tiontion withwith aa medianmediantion effectiveeffectivewith a median concentrationconcentration effective ofof concentration 0.250.25 nMnM T3T3 (CAS:(CAS: of 0.25 6893-02-3,6893-02-3, nM T3 (CAS: Sigma-Aldrich,Sigma-Aldrich, 6893-02-3, Sigma-Aldrich, St Quentin Fallavier,St Quentin France Fallavier, T2877). FranceNegative T2877). controlcontrol Negative wellswells containedcontained control wells cellscells contained andand testtest me-me- cells and test medium with the diumsame amountwith the of same ethanol amount (0.1%) of ethanolas the exposed (0.1%) ascells. the Positiveexposed controls cells. Positive in- controls included were T3cluded (agonist were control) T3 (agonist and amioda control)rone and (antagonist amiodarone control, (antagonist CAS: 19774-82-4, control, CAS: 19774-82-4, Sigma-Aldrich,Sigma-Aldrich, St Quentin Fallavier, St Quentin France Fallavier, A8423). FranceThey were A8423). used They as a weredose usedresponse as a dose response control in the assaycontrol (from in the 0.001 assay to 10 (from nM for0.001 T3 to and 10 fromnM for 0.25 T3 to and 500 from nM for0.25 amiodarone) to 500 nM for amiodarone) toto validatevalidate itit forforto detectiondetection validate itofof for agonisticagonistic detection andand of antagonisticantagonistic agonistic and activitiesactivities antagonistic (Appendix(Appendix activities B).B). (Appendix B).

4.3.2. Cytotoxicity/Viability/Proliferation4.3.2. Cytotoxicity/Viability/Proliferation Following a 4 hFollowing incubation a 4period h incubation with 10 periodµL/well with of 0.1 10 mg/mLµL/well resazurin of 0.1 mg/mL (Sigma- resazurin (Sigma- Aldrich, St QuentinAldrich, Fallavier, St Quentin France) Fallavier, in PBS, France) cell proliferation in PBS, cell was proliferation measured wasas relative measured as relative fluorescencefluorescence unitsunitsfluorescence (RFUs)(RFUs) resultingresulting units (RFUs) fromfrom thetheresulting reduction from of the non-fluorescent reduction of non-fluorescent resazurin to the resazurin to the fluorescentfluorescent productproductfluorescent resorufin.resorufin. product Fluorescence,Fluorescence, resorufin. aa Fluorescence, measure of the a measureamount ofof viablethe amount cells pre- of viable cells present, was recordedsent, at was λexex recorded== 530530 nmnm at andand λex λ =emem 530 == 590590 nm nmnm and onon λ emaa microplatemicroplate= 590 nm on readerreader a microplate (Chameleon(Chameleon reader (Chameleon

Metabolites 2021, 11, 94 11 of 18

con, Dutscher, Brumath, France). Following an attachment period of 2 h to 3 h, the cells were exposed in triplicate and for 96 h to various concentrations of the chemicals to be tested (100 µL, 2× dosing exposure concentration in PCM medium), either alone or in combination with a median effective concentration of 0.25 nM T3 (CAS: 6893-02-3, Sigma- Aldrich, St Quentin Fallavier, France, T2877). Negative control wells contained cells and test medium with the same amount of ethanol (0.1%) as the exposed cells. Positive controls included were T3 (agonist control) and amiodarone (antagonist control, CAS: 19774-82-4, Sigma-Aldrich, St Quentin Fallavier, France, A8423). They were used as a dose response control in the assay (from 0.001 to 10 nM for T3 and from 0.25 to 500 nM for amiodarone) to validate it for detection of agonistic and antagonistic activities (AppendixB).

4.3.2. Cytotoxicity/Viability/Proliferation Following a 4 h incubation period with 10 µL/well of 0.1 mg/mL resazurin (Sigma- Aldrich, St Quentin Fallavier, France) in PBS, cell proliferation was measured as relative fluorescence units (RFUs) resulting from the reduction of non-fluorescent resazurin to the fluorescent product resorufin. Fluorescence, a measure of the amount of viable cells present, was recorded at λex = 530 nm and λem = 590 nm on a microplate reader (Chameleon from Hidex, Finland). The assay can detect antagonist effects (inhibition of cell proliferation) measured by a fluorescence decrease. However, it is difficult to distinguish an inhibition of cell proliferation from a cytotoxicity as both effects are expressed by a decrease in fluorescence. So concentrations of compounds that inhibited basal metabolic activity of GH3 in the agonist assay were excluded from statistical analysis. A chemical was considered cytotoxic if the fluorescence was less than the fluorescence of the vehicle control minus 3-fold the standard deviation in the agonist assay.

4.3.3. Data Analysis Cell proliferation was expressed as a function of the maximum response observed at 10 nM T3 (in agonist mode) or 0.25 nM T3 (in antagonist mode), which was set at 100% induction. The response for the solvent control was set at 0%.

4.3.4. Statistical Analysis Obtained data were statistically analyzed using the PC program GraphPad Prism 6.00 (GraphPad Software Incorporated, San Diego, CA, USA). Descriptive statistical char- acteristics (arithmetic mean, minimum, maximum, standard deviation and coefﬁcient of variation) were evaluated. One-way analysis of variance (ANOVA) and the Dunnett’s multiple comparison test were used for statistical evaluations. The level of signiﬁcance was set at *** p < 0.001; ** p < 0.01 and * p < 0.05.

4.4. Docking DEHP, DEHT and their respective metabolites were docked into TRα1, the only TRα subtype to bind T3, and TRβ1, the only β subtype crystallized. The coordinates of the receptor subtypes were taken from the RCSB ProteinDatabank under the entry 4lnw [48] crystallized with T3, and 1n46 [49], bound to an agonist, respectively. The ligands were extracted manually and both molecules were assigned using the Gasteiger–Hückel method. The ligands were subjected to an energy minimization using the maximin2 protocol of the Sybyl 6.9.1 molecular modeling package. The co-crystallized T3 was cross-docked with very good precision, indicating that the docking protocol was sound. It consisted of a 30- solution GOLD run in a binding site deﬁned as a sphere of 10 Å around the co-crystallized ligand with 100,000 operations and the ChemPLP scoring function. The 30 poses were manually inspected to deﬁne the most representative conformations, chosen as the best scored solution from the largest cluster of poses. In a few cases, a high score, but not the best, was selected as it was more representative of all the poses. The two receptor subtypes only differed in 15 residues, all of which were remote from the binding site. The superimposition of the structures was also fairly good, with a Root Mean Square over the Metabolites 2021, 11, 94 12 of 18

heavy atoms of 2.42 Å, which is only very slightly higher than the worst resolution (2.2 Å). The docking results were therefore fairly comparable, so the discussion of the docking focused on TRα, which beneﬁted from a better resolution.

5. Conclusions We can conclude that under these experimental conditions, and regard to the use of alternative methods (in vitro, in silico), that DEHT-oxidized metabolites have negligible effect on T3 hormonal activities when compared to DEHP metabolites. Taking all these data into account, along with human biological enzymatic data, there appears to be no safety concern with DEHT compared to DEHP tested in the same conditions and when based on a mode of action.

Author Contributions: Conception and design: M.-C.C., T.D., V.S.; Synthesis of metabolites: E.M.; In vitro experiments and analysis: L.D., I.S., M.-C.C.; In silico experiments and analysis: A.F., N.K.; writing—original draft preparation: N.K., I.S., E.M., M.-C.C., V.S.; writing—review: all authors; formatting and editing: I.S.; supervision: T.D., V.S., M.-C.C.; project administration: V.S.; funding acquisition: V.S. All authors have read and agreed to the published version of the manuscript. Funding: This study is a part of the ARMED NEO project and received financial support from the French National Agency for the Safety of Medicines and Health Products (ANSM) Grant/award number: AAPR-2015-027. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Data is contained within the article. Conflicts of Interest: The authors declare no conflict of interest.

Appendix A. Supplementary Data—Synthesis of DEHP and DEHT Metabolites Materials for Chemical Syntheses Unless otherwise mentioned, all experiments were performed under argon; all reagents were purchased from the following commercial suppliers: Sigma-Aldrich, Acros Organics, Carlo Erba, TCI Europa, Alpha Aesar. Anhydrous dimethylformamide, DMF, anhydrous trimethylamine, anhydrous pyridine were purchased from Acros Organics. Tetrahydrofu- ran, THF was dried over a Pure Solv™ Micro Solvent Puriﬁcation System (Sigma-Aldrich) with an alumina column. Dichloromethane was distilled over hydride calcium. Nuclear magnetic resonance (NMR) spectra were acquired on a Bruker AC-200 or 500 operating at 200 or 500 for 1H NMR and 50 or 125 MHz for 13C NMR, respectively. All 1H and 13C NMR spectra are reported in δ units, parts per million (ppm). Coupling constants are indicated in Hertz (Hz). The following abbreviations are used for spin multiplicity: s = singlet, d = doublet, t = triplet, q = quadruplet, m = multiplet, and brs = broad singlet. Thin-layer chromatography, TLC, was performed on pre-coated silica gel sheets (POLYGRAM® 60F254 plates, Macherey-Nagel, Hoerdt, France) and visualized under UV light (254 nm). Revelators used were KMnO4 (1.5 g KMnO4, 10 g K2CO3, and 1.25 mL 10% NaOH in 200 mL water) and ninhydrin (1.5 g ninhydrin in 100 mL of n-butanol with 3 mL AcOH). Column chromatography was performed using silica gel normal phase (35–70 µm). Uncorrected melting points (Mp) were measured on an IA9100 Digital Melting Point Apparatus. Infrared spectra (IR) were recorded on a Bruker FT Vector 22. The HRMS analysis was performed using a Thermo Exactive benchtop Orbitrap® instrument (UCA PARTNER, Clermont-Ferrand, France). DEHP, MEHP, 5-OH-MEHP, 5-oxo-MEHP and 5-cx-MEHP were synthesized using the procedures previously described by Nüti et al. [21]. Metabolites 2021, 11, 94 13 of 18 Metabolites 2021, 11, x FOR PEER REVIEW 13 of 18

HOH2C BnO (1) O (2)

(i)

O O O (iii) OH (v) OH O O O HO BnO BnO O O O (3) (7) (10)

(ii) (vi) (vi)

O O O O OH OH (iii) O O O HO O BnO BnO O O O O (4) (8) (9) 5-cx-MEHT

(iii)

O O O OH (iv) O O HO HO

O O (5) (6) 5-oxo-MEHT 5-OH-MEHT

Conditions and reactants: (i) DMAP, DCC, dichloromenthane, rt, overnight; (ii) PdCl2;DMF/H2O(17/3,v/v),rt,overnight;(iii)H2, Pd/C 10%,absolute ethanol, 4 hours; (iv) NaBH4, rt, overnight; (v) (a) B2H6, anhydrous THF, rt, 1 h; (b) NaOH (3M), then H202 30%, 50°C, 2 h; (vi) CrO3/H2SO4,acetone;rt;30min Figure A1. Synthesis of 5-OH-MEHT, 5-oxo-MEHT and 5-cx-MEHT.

Benzyl (2-ethylhex-5-en-1-yl)terephthalate (3) 4-((benzyloxy)carbonyl)benzoic acid acid(2) 2(2)(22 g, (2 7.80 g, mmoles),7.80 mmoles), DMAP DMAP (1.37 g, (1.37 11.24 mmoles)g, 11.24 ◦ mmoles)dissolved dissolved in anhydrous in dichloromethaneanhydrous dichloromethane (75 mL), cooled (75 to 0mL),C, undercooled argon to 0 atmosphere,°C, under argon DCC atmosphere,(2.32 g, 11.24 mmoles)DCC (2.32 was g, added11.24 mmoles) to a solution was of added 2-ethylhex-5-en-1-ol to a solution (1)of 12-ethylhex-5-en-1-ol(1.2 g, 9.30 mmoles). (1)The1 (1.2 reaction g, 9.30 mixture mmoles). was allowedThe reaction to warm mixture to room was temperature allowed to and warm was to then room stirred temperature overnight. ® andThe was reaction then mixture stirred overnight. was filtered The through reaction Celite mixture545 was and filtered washed through with dichloromethhane Celite® 545 and washed(2 × 20 mL with). Thedichloromethhane filtrate was evaporated (2 × 20 mL). under The reduced filtrate pressurewas evaporated and the resultingunder reduced crude pressureproduct wasand purifiedthe resulting on silica crude gel product eluted with was dichloromethane/cyclohexanepurified on silica gel eluted with (9/1, dichloro-v/v) to −1 1 methane/cyclohexaneproduce the expected compound (9/1, v/v) (3)to (1.91produce g, 66%). the IRexpected (cm ) compoundν 1246, 1263, (3) 1716, (1.91 2117, g, 66%). 2929; IRH δ (cmNMR−1) (500ν 1246, MHz, 1263, CDCl 1716,3) 0.95 2117, (t, 2929; J = 7.5 1H Hz, NMR 3H), (500 1.17–1.51 MHz, (m, CDCl 4H, 1.71—1.763) δ 0.95 (t, (m, J = 1H), 7.5 Hz, 2.10–2.15 3H), 1.17–1.51(m, 2H), 4.27 (m, (d, 4H, J = 1.71—1.76 5.6 Hz), 4.94–4.96 (m, 1H), (m, 2.10–2.15 1H), 5.00–5.04 (m, 2H), (m, 4.27 1H), (d, 5.38 J (s,= 5.6 2H), Hz), 5.76–5.94 4.94–4.96 (m, 1H),(m, 13 1H),7.33–7.39 5.00–5.04 (m, 3H), (m, 7.43–7.451H), 5.38 (m,(s, 2H), 2H), 5.76–5.94 8.08 (m, 2H), (m, 1H), (8.12 7.33–7.39 (m, 2H); (m,CNMR 3H), (1257.43–7.45 MHz, (m, CDCl 2H),3) HRMS for C H O m/z [M+ H]+ calc.: 367.18311; found: 367.19023. 8.08 (m, 2H),23 (8.1226 4(m, 2H); 13CNMR (125 MHz, CDCl3) HRMS for C23H26O4 m/z [M+ H]+ calc.:Benzyl 367.18311; (2-ethyl-5-oxohexyl)terephthalate found: 367.19023. (4) BenzylA solution (2-ethyl-5-oxohexyl)terephthalate of compound (3) (0.39 g, 1.06 mmoles) (4) dissolved in 2 mL of DMF/H2O (17/3, v/v) was added dropwise to a solution of PdCl2 (0.197 g, 1.11 moles) and para- A solution of compound (3) (0.39 g, 1.06 mmoles) dissolved in 2 mL of DMF/H2O benzoquinone (0.132 g, 1.22 mmoles) dissolved in 10 mL of DMF/H2O (17/3, v/v). The (17/3, v/v) was added dropwise to a solution of PdCl2 (0.197 g, 1.11 moles) and para-ben- resulting mixture was stirred overnight in the dark. A solution of HCl (3N) (20 mL) was zoquinone (0.132 g, 1.22 mmoles) dissolved in 10 mL of DMF/H2O (17/3, v/v). The resulting added dropwise to the reaction mixture. The reaction solution was extracted with diethyl mixture was stirred overnight in the dark. A solution of HCl (3N) (20 mL) was added ether (3 × 30 mL). The combined organic layers were washed with brine (30 mL), dried dropwise to the reaction mixture. The reaction solution was extracted with diethyl ether over MgSO , ﬁltered and evaporated under reduced pressure. The crude product was (3 × 30 mL).4 The combined organic layers were washed with brine (30 mL), dried over puriﬁed on silica gel eluted with a gradient dichloromethane/methanol (100 to 98/2) to MgSO4, filtered and evaporated under reduced pressure. The crude product was purified produce the expected compound (4) (0.247 g, 61%) as an oil. IR (cm−1) ν 1246, 1263, 1716, on silica gel eluted with a gradient dichloromethane/methanol (100 to 98/2) to produce the 2931; 1H NMR (200 MHz, CDCl ) δ 0.99 (t, J = 7.4 Hz), 1.45–1.51 (m, 2H), 1.70–1.76 (m, 3H), expected compound (4) (0.247 g,3 61%) as an oil. IR (cm−1) ν 1246, 1263, 1716, 2931; 1H NMR 2.17 (s, 3H), 2.51–2.56 (m, 2H), 4.28 (d, J = 4.4 Hz, 2H), 5.40 (s, 2H), 7.38–7.49 (m, 5H), (200 MHz, CDCl3) δ 0.99 (t, J = 7.4 Hz), 1.45–1.51 13(m, 2H), 1.70–1.76 (m, 3H), 2.17 (s, 3H), 8.09 (d, J = 8.6 Hz, 2H), –8.17 (d, J = 8.6 Hz, 2H); C NMR (125 MHz, CDCl3) HRMS for 2.51–2.56 (m, 2H), 4.28 (d,+ J = 4.4 Hz, 2H), 5.40 (s, 2H), 7.38–7.49 (m, 5H), 8.09 (d, J = 8.6 Hz, C23H26O5 m/z [M+ H] calc.: 383.17802; found: 383.18549.

Metabolites 2021, 11, 94 14 of 18

4-(((2-Ethyl-5-oxohexyl)oxy)carbonyl)benzoic acid (5) (5-oxo-MEHT) Pd/C 10% (24.7 mg) was added to a solution of compound (4) (0.247 g, 0.64 mmoles) dissolved in absolute ethanol (30 mL). The resulting mixture was degassed three times and stirred under hydrogen atmosphere for 4 h. The reaction solvent was filtered on a Celite® 545 pad and rinsed with 50 mL of absolute ethanol. The filtrate was evaporated under reduced pressure. The crude product was purified on silica gel with ethyl acetate to produce the expected compound (5) (0.148 g, 79%) as a solid. Mp: 87–89 ◦C; IR (cm−1) 1 ν 1259, 1313, 1678, 1708, 2552, 2960; H NMR (200 MHz, CDCl3) δ 1.00 (t, J = 7.4 Hz, 3H), 1.39–1.60 (m, 2H), 1.72–1.82 (m, 3H), 2.20 (s, 3H), 2.58 (t, J = 7.1 Hz, 2H), 4.30 (d, J = 4.1 Hz, 2H), 8.14 (d, J = 8.6 Hz, 2H), −8.19 (d, J = 8.6 Hz, 2H), 10.2 (brs, 1H); 13C NMR (125 MHz, CDCl3) δ 11.0, 24.0, 24.9, 38.5, 40.6, 67.6, 129.5, 130.2, 132.9, 135.1, 165.9, 170.6, 209.0; HRMS + for C16H20O5 m/z [M+ H] calc.: 292.13107; found: 291.12415. 4-(((2-Ethyl-5-hydroxyhexyl)oxy)carbonyl)benzoic acid (6) (5-OH-MEHT) NaBH4 (0.031 g, 0.06 mmoles) was added portion-wise (very exothermic reaction) to a solution of ketone compound (5) (0.079 g, 0.02 mmoles) dissolved in absolute ethanol, cooled to 0 ◦C with an ice bath. After stirring overnight at room temperature, water (20 mL) was added with caution, followed by a solution of HCl (1N) (pH = 1). The aqueous solution was extracted with ethyl acetate (3 × 20 mL). The combined organic layers were dried over MgSO4, filtered and evaporated under reduced pressure. The crude product was purified on silica gel eluted with ethyl acetate/ethanol (75/25; v/v) to provide the expected compound (6) (0.069 g, 87%) as a solid. Mp: 74–76 ◦C; IR (cm−1) ν 1271, 1311, 1 1686, 1709, 2550, 2851, 2929, 2961, 3211; H NMR (500 MHz, CDCl3) δ 0.95 (t, J = 7.4 Hz, 3H), 1.22 (d, J = 6.1 Hz, 3H), 1.45–1.54 (m, 4H), 1.73–1.76 (m, 2H), 3.82–3.86 (m, 1H), 4.28 13 (d, J = 5.6 Hz, 2H), 8.07–8.13 (m, 4H); C NMR (125 MHz, CDCl3) HRMS for C16H22O5 m/z [M+ H]+ calc.: 294.14672; found: 293.13979. Benzyl (2-ethyl-6-hydroxyhexyl)terephthalate (7) B2H6 (1M) (1.96 mL, 1.96 mmoles) was added dropwise to a solution of compound (3) (0.54 g, 1.47 mmoles) dissolved in anhydrous THF (10 mL), cooled to 0 ◦C. The resulting mixture was stirred to room temperature for 1 h. NaOH (3M) (182 µL), followed by 33% ◦ H2O2 (182 µL), were then added dropwise. The resulting mixture was heated to 50 C for 2 h. After cooling to room temperature, water (10 mL) was added. The resulting mixture was extracted with diethyl ether (3 × 10 mL). The combined organic layers were washed with brine (30 mL), dried over MgSO4, filtered and evaporated under reduced pressure. The crude product was purified on silica gel and eluted with dichloromethane/cyclohexane (8/2; v/v) to produce the expected compound (7) (0.262 g, 79%) as an oil. IR (cm−1) ν 1264, 1 1716, 2932; H NMR (200 MHz, CDCl3) δ 0.98 (t, J = 7.4 Hz), 1.46–1.58 (m, 7H), 1.63–1.80 (m, 2H), 3.67 (d, J = 6.3 Hz, 2H), 4.29 (d, J = 5.6 Hz, 2H), 5.41 (s, 2H), 7.39–748 (m, 5H), 8.09–8.19 13 (m, 4H); C NMR δ (125 MHz, CDCl3) 11.1, 23.0, 23.9, 33.0, 39.0, 62.8, 67.1, 67.6, 128.3, 128.3, 128.4, 128.7, 129.5, 129.7, 129.7, 133.9, 134.4, 135.7, 165.7, 165.9; HRMS for C23H28O5 m/z [M+ H]+ calc.: 384.19367; found: 385.20062. 5-(((4-((benzyloxy)carbonyl)benzoyl)oxy)methyl)heptanoic acid (8) 11.31 mL of Jones’ reagent (prepared from 670 mg CrO3, 600 µLH2SO4 and 5 mL of water) was added dropwise to a solution of compound (7) (0.43 g, 1.12 mmoles) dissolved in acetone (3 mL), cooled to 0 ◦C. After stirring for 30 min at room temperature, water (10 mL) was carefully added. The mixture was then extracted with diethyl ether (3 × 30 mL). The combined organic layers were dried over MgSO4, filtered and evaporated under reduced pressure to produce an oil. The crude product was purified on silica gel and eluted with ethyl acetate/cyclohexane (2/8, v/v) to produce the expected compound (8) (0.113 g, 26%) −1 1 as an oil. IR (cm ) ν 1264, 1715, 2959; H NMR (500 MHz, CDCl3) δ 0.95 (t, J = 7.5 Hz, 3H), 1.45–1.48 (m, 4H), 1.70–1.77 (m, 3H), 2.37 (t, J = 7.3 Hz, 2H), 2.27 (d, J= 5.4 Hz, 2H), 5.38 (s, 2H), 7.25–7.38 (m, 3H), 7.40–7.45 (m, 2H), 8.07–8.09 (m, 2H), 8.12–8.14 (m, 2H); 13C NMR (125 MHz, CDCl3) δ11.0, 21.5, 23.8, 30.3, 34.1, 38.8, 67.0, 67.4, 127.9, 128.4, 128.8, 129.5, 129.7, + 134.0, 134.4, 136.0, 165.5, 166.0, 179.3; HRMS for C23H26O6 m/z [M− H] calc.: 397.17294; found: 397.16547. Metabolites 2021, 11, x FOR PEER REVIEW 15 of 18

Metabolites 2021, 11, 94 15 of 18

129.7, 134.0, 134.4, 136.0, 165.5, 166.0, 179.3; HRMS for C23H26O6 m/z [M- H]+ calc.: 397.17294; found: 397.16547. 4-(((2-Ethyl-6-hydroxyhexyl)oxy)carbonyl)benzoic4-(((2-Ethyl-6-hydroxyhexyl)oxy)carbonyl)benzoic acid acid (10) (10) 10%10% Pd/CPd/C (0.02 (0.02 mg) mg) was was added added to to a asolution solution of ofcompound compound (8) (8)(0.2(0.2 g, 0.52 g, 0.52 mmoles) mmoles) dissolveddissolved inin absoluteabsolute ethanolethanol (20(20 mL).mL). The The resulting resulting mixture mixture was was degassed degassed three three times times and ® stirredand stirred under under a hydrogen a hydrogen atmosphere atmosphere for 4 for h. The4 h. reactionThe reaction solvent solvent was was filtered filtered on a on Celite a 545Celite pad® 545 and pad rinsed and withrinsed 20 with mL 20 of mL absolute of absolute ethanol. ethanol. The filtrateThe filtrate was was evaporated evaporated under reducedunder reduced pressure pressure to produce to produce the expected the expected compound compound(10) (0.138 (10) (0.138 g, 90%). g, 90%). The The product prod- was ◦ −1 useduct was for used the next for the step next without step without purification. purification. Mp: 64–66 Mp: 64–66C; IR °C; (cm IR (cm) ν−1)1261, ν 1261, 1427, 1427, 1506, 1 1680,1506, 1708,1680, 1708, 2856, 2856, 2932; 2932;H NMR1H NMR (500 (500 MHz, MHz, MeOD) MeOD)δ δ 0.940.94 (t, J J = = 7.5 7.5 Hz, Hz, 3H), 3H), 1.42–1.53 1.42–1.53 (m,(m, 9H 9H),), 1.70–1.731.70–1.73 (m,(m, 1H), 3.54 3.54 (t, (t, J J = = 6.5 6.5 Hz, Hz, 2H), 2H), 4.25 4.25 (d, (d, J = J 5.4 = 5.4 Hz, Hz, 2H), 2H), 8.03 8.03 (d, (d,J = J8.2 = 8.2 Hz,Hz, 2H),2H), 8.078.07 (d,(d, J = 8.2 Hz, 2H); 2H); 1313CC NMR NMR (125 (125 MHz, MHz, MeOD) MeOD) δ 10.2,δ 10.2, 22.4, 22.4, 23.8, 23.8, 30.1, 30.1, 32.1, 32.1, + 39.0,39.0, 61.5,61.5, 66.8, 129.4, 129.4, 129.7, 129.7, 133.3, 133.3, 136.0, 136.0, 165.9, 165.9, 168.2; 168.2; HRMS HRMS for C for16H C22O165H m/z22O [M+5 m/z H]+ [M+calc.: H] calc.:295.1539; 295.1539; found: found: 295.1540. 295.1540. 4-(((5-Carboxy-2-ethylpentyl)oxy)carbonyl)benzoic4-(((5-Carboxy-2-ethylpentyl)oxy)carbonyl)benzoic acid acid (9) (9) (5-cx-MEHT) (5-cx-MEHT) MethodMethod A: 10% 10% Pd/C Pd/C (0.02 (0.02 g) was g) was added added to compound to compound (8) (0.2(8) g,(0.2 0.054 g, mmoles) 0.054 mmoles) dis- dissolvedsolved in inabsolute absolute ethanol ethanol (20 (20 mL). mL). The The reaction reaction mixture mixture was was degassed degassed three three times times and and stirredstirred under under aa hydrogenhydrogen atmosphereatmosphere forfor 7 h. The reaction mixture mixture was was filtered filtered on on a a pad pad of Celiteof Celite® 545.® 545. The The filtrate filtrate was was evaporated evaporated under under reduced reduced pressure. pressure. The The crude crude product product was purifiedwas purified on silica on silica gel andgel and eluted eluted with with ethyl ethyl acetate acetate to to produce produce the the expected expected compound (9) (0.017(9) (0.017 g, 8%) g, 8%) as a as solid. a solid. Method B: 107 µL of Jones’ reagent (prepared from 670 mg CrO3, 600 µL H2SO4 and Method B: 107 µL of Jones’ reagent (prepared from 670 mg CrO3, 600 µLH2SO4 and 55 mL mL ofof water)water) was added dropwise dropwise to to a asolution solution of of compound compound (10)(10) (0.1(0.1 g, 0.34 g, 0.34 mmoles) mmoles) dissolveddissolved inin acetoneacetone (1.5(1.5 mL),mL), cooledcooled toto 00◦ °C.C. AfterAfter stirringstirring forfor 3030 minmin atat roomroom temperature,tempera- waterture, (10water mL) (10 was mL) carefully was carefully added. added. The mixture The mixture was extracted was extracted with diethyl with diethyl ether (3*15 ether mL ). (3*15 mL). The combined organic layers were dried over MgSO4, filtered and evaporated The combined organic layers were dried over MgSO4, filtered and evaporated under reducedunder reduced pressure. pressure. The crude The crude product product was purifiedwas purified on silicaon silica gel gel and and eluted eluted with with ethyl acetateethyl acetate to produce to produce the expected the expected compound compound(9) (0.130 (9) (0.130 g, 96%) g, 96%) in solid in solid form. form. Mp: 65–67 °C; IR (cm−1) ν 1266, 1685, 1712, 2560, 2875, 2924, 2957; 1H NMR (200 MHz, Mp: 65–67 ◦C; IR (cm−1) ν 1266, 1685, 1712, 2560, 2875, 2924, 2957; 1H NMR (200 MHz, MeOD) δ 0.99 (t, J = 7.4 Hz, 3H), 1.38–1.75 (m, 9H), 4.32 (d, J = 5.5 Hz, 2H), 8.08 (m, 4H); MeOD) δ 0.99 (t, J = 7.4 Hz, 3H), 1.38–1.75 (m, 9H), 4.32 (d, J = 5.5 Hz, 2H), 8.08 (m, 4H); 13C NMR (125 MHz, MeOD) δ 10.6, 23.9, 23.8, 30.3, 32.5, 39.0, 61.2, 67.3, 129.0, 129.4, 133.5, 13 δ C NMR (125 MHz, MeOD) 10.6, 23.9, 23.8, 30.3,+ 32.5, 39.0, 61.2, 67.3, 129.0, 129.4, 133.5, 135.8, 165.7, 165.7; HRMS for C16H20O6 m/z [M- H] calc.:+ 307.12599; found: 307.11904. 135.8, 165.7, 165.7; HRMS for C16H20O6 m/z [M− H] calc.: 307.12599; found: 307.11904. Appendix B. Supplementary Data—Effects of Reference Positive Controls in the T- AppendixScreen Assay B. Supplementary Data—Effects of Reference Positive Controls in the T-Screen Assay

Figure A2. GH3 cells were exposed during 96 h to increasing concentrations of the chemicals alone (agonist assay) or in the presence of 0.25 nM T3 (antagonist assay). Cell proliferation was expressed relative to the maximum response observed at 10 nM T3 (in agonist mode) or 0.25 nM T3 (in antagonist mode). The response for the solvent control was set at 0%. The values are mean ± SD of at least two independent experiments in triplicate. Absolute numbers for controls values were: 5455 (solvent control)-15784 (10 nM T3) RFU. Metabolites 2021, 11, 94 16 of 18

References 1. SCENIHR. Scientific Committee on Emerging and Newly-Identified Health Risks. Opinion on the Safety of Medical Devices Containing DEHP-Plasticized PVC or Other Plasticizers on Neonates and Other Groups Possibly at Risk. 2016. Available online: https://ec.europa/health/scientific_committees/emerging/docs/scenihr_o_047.pdf (accessed on 15 October 2020). 2. Bernard, L.; Cueff, R.; Breysse, C.; Décaudin, B.; Sautou, V.; Armed Study Group. Migrability of PVC plasticizers from medical devices into a simulant of infused solutions. Int. J. Pharm. 2015, 485, 341–347. [CrossRef] 3. Mallow, E.B.; Fox, M.A. Phthalates and critically ill neonates: Device-related exposures and non-endocrine toxic risks. J. Perinatol. 2014, 34, 892–897. [CrossRef][PubMed] 4. Fischer Fumeaux, C.J.; Stadelmann Diaw, C.; Palmero, D.; M’Madi, F.; Tolsa, J.F. Phthalate exposure in the neonatal intensive care unit. Arch. Pediatr. 2015, 22, 203–210. [CrossRef] 5. Stroustrup, A.; Bragg, J.B.; Busgang, S.A.; Andra, S.S.; Curtin, P.; Spear, E.A.; Just, A.C.; Arora, M.; Gennings, C. Sources of clinically significant neonatal intensive care unit phthalate exposure. J. Expo. Sci. Environ. Epidemiol. 2020, 30, 137–148. [CrossRef] 6. Regulation (EC) No 1272/2008 of the European Parliament and of the Council of 16 December 2008 on Classification, Labelling and Packaging of Substances and Mixtures, Amending and Repealing Directives 67/548/EEC and 1999/45/EC, and Amending Regulation (EC) No 1907/2006 (Text with EEA Relevance). 2008. Available online: http://data.europa.eu/eli/reg/2008/1272/oj/ eng (accessed on on 15 October 2020). 7. Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on Medical Devices, Amending Directive 2001/83/EC, Regulation (EC) No 178/2002 and Regulation (EC) No 1223/2009 and Repealing Council Directives 90/385/EEC and 93/42/EEC (Text with EEA Relevance). 2017. Available online: http://data.europa.eu/eli/reg/2017/745/oj/eng (accessed on on 15 October 2020). 8. Barber, E.D.; Fox, J.A.; Giordano, C.J. Hydrolysis, absorption and metabolism of di(2-ethylhexyl) terephthalate in the rat. Xenobiotica 1994, 24, 441–450. [CrossRef] 9. CPSC Staff Statement on University of Cincinnati Report “Toxicity Review for Di-2 Ethylhexyl Terephthalate (DEHT)”. 2018. Avail- able online: https://www.cpsc.gov/s3fs-public/Toxicity%20Review%20of%20DEHT.pdf?FObpuBBqgypVtw7gIEGMFXHN5 H7vbeEz (accessed on 15 October 2020). 10. Eljezi, T.; Pinta, P.; Richard, D.; Pinguet, J.; Chezal, J.M.; Chagnon, M.C.; Sautou, V.; Grimandi, G.; Moreau, E. In vitro cytotoxic effects of DEHP-alternative plasticizers and their primary metabolites on a L929 cell line. Chemosphere 2017, 173, 452–459. [CrossRef][PubMed] 11. Faber, W.D.; Deyo, J.A.; Stump, D.G.; Navarro, L.; Ruble, K.; Knapp, J. Developmental toxicity and uterotrophic studies with di-2-ethylhexyl terephthalate. Birth Defects Res. B Dev. Reprod. Toxicol. 2007, 80, 396–405. [CrossRef] 12. Kambia, N.K.; Séverin, I.; Farce, A.; Moreau, E.; Dahbi, L.; Duval, C.; Dine, T.; Sautou, V.; Chagnon, M.C. In vitro and in silico hormonal activity studies of di-(2-ethylhexyl)terephthalate, a di-(2 ethylhexyl) phthalate substitute used in medical devices, and its metabolites. J. Appl. Toxicol. 2019, 39, 1043–1056. [CrossRef][PubMed] 13. Villanger, G.D.; Drover, S.S.M.; Nethery, R.C.; Thomsen, C.; Sakhi, A.K.; Øvergaard, K.R.; Zeiner, P.; Hoppin, J.A.; Reichborn- Kjennerud, T.; Aase, H.; et al. Associations between urine phthalate metabolites and thyroid function in pregnant women and the influence of iodine status. Environ. Int. 2020, 137, 105509. [CrossRef][PubMed] 14. Ghisari, M.; Bonefeld-Jorgensen, E.C. Effects of plasticizers and their mixtures on estrogen receptor and thyroid hormone functions. Toxicol. Lett. 2009, 189, 67–77. [CrossRef][PubMed] 15. Huang, H.-B.; Chuang, C.-J.; Su, P.-H.; Sun, C.-W.; Wang, C.-J.; Wu, M.-T.; Wang, S.-L. Prenatal and Childhood Exposure to Phthalate Diesters and Thyroid Function in a 9-Year Follow-up Birth Cohort Study: Taiwan Maternal and Infant Cohort Study. Epidemiol. Camb. Mass. 2017, 28 (Suppl. 1), S10–S18. [CrossRef][PubMed] 16. Kim, M.J.; Moon, S.; Oh, B.-C.; Jung, D.; Choi, K.; Park, Y.J. Association Between Diethylhexyl Phthalate Exposure and Thyroid Function: A Meta-Analysis. Thyroid 2019, 29, 183–192. [CrossRef][PubMed] 17. Zoeller, R.T.; Rivet, J. Timing of thyroid hormone action in the developing brain: Clinical observations and developmental findings. Neuroendocrinology 2004, 16, 809–818. [CrossRef] 18. Vulsma, T. Impact of exposure to maternal PCBs and dioxins on the neonate’s thyroid hormone status. Epidemiology 2000, 11, 239–241. [CrossRef] 19. De Cock, M.; Maas, Y.G.H.; Van de Bor, M. Does perinatal exposure to endocrine disruptors induce autism spectrum and attention deficit hyperactivity disorders? Rev. Acta Paediatr. 2012, 101, 811–818. [CrossRef][PubMed] 20. Qian, X.; Li, J.; Xu, S.; Wan, Y.; Li, Y.; Jiang, Y.; Zhao, H.; Zhou, Y.; Liao, J.; Liu, H.; et al. Prenatal exposure to phthalates and neurocognitive development in children at two years of age. Environ. Int. 2019, 131, 105023. [CrossRef] 21. Nuti, F.; Hildenbrand, S.; Chelli, M.; Wodarz, R.; Papini, A.M. Synthesis of DEHP metabolites as biomarkers for GC–MS evaluation of phthalates as endocrine disrupters. Bioorg. Med. Chem. 2005, 13, 3461–3465. [CrossRef][PubMed] 22. Porterfield, S.P.; Hendrich, C.E. The role of thyroid hormones in prenatal and neonatal neurological development-current perspectives. Endocr. Rev. 1993, 14, 94–106. [PubMed] 23. Bernal, J.; Nunez, J. Thyroid hormones and brain development. Eur. J. Endocrinol. 1995, 133, 390–398. [CrossRef] Metabolites 2021, 11, 94 17 of 18

24. Hohenwarter, O.; Waltenberger, A.; Katinger, H. An in vitro test system for thyroid hormone action. Anal. Biochem. 1996, 234, 56–59. [CrossRef] 25. Gutleb, A.C.; Meerts, I.A.T.M.; Bergsma, J.H.; Schriks, M.; Murk, A.J. T-Screen as a tool to identify thyroid hormone receptor active compounds. Environ. Toxicol. Pharmacol. 2005, 19, 231–238. [CrossRef][PubMed] 26. Sugiyama, S.; Shimada, N.; Miyoshi, H.; Yamauchi, K. Detection of thyroid system-disrupting chemicals using in vitro and in vivo screening assays in Xenopus laevis. Toxicol. Sci. 2005, 88, 367–374. [CrossRef][PubMed] 27. Ishihara, A.; Sawatsubashi, S.; Yamauchi, K. Endocrine disrupting chemicals: Interference of thyroid hormone binding to transthyretins and to thyroid hormone receptors. Mol. Cell. Endocrinol. 2003, 199, 105–117. [CrossRef] 28. Dong, X.; Dong, J.; Zhao, Y.; Guo, J.; Wang, Z.; Liu, M.; Zhang, Y.; Na, X. Effects of Long-Term In Vivo Exposure to Di-2- Ethylhexylphthalate on Thyroid Hormones and the TSH/TSHR Signaling Pathways in Wistar Rats. Int. J. Environ. Res. Public Health 2017, 14, 44. [CrossRef][PubMed] 29. Liu, C.; Zhao, L.; Wei, L.; Li, L. DEHP reduces thyroid hormones via interacting with hormone synthesis-related proteins, deiodinases, transthyretin, receptors, and hepatic enzymes in rats. Environ. Sci. Pollut. Res. Int. 2015, 22, 12711–12719. [CrossRef] [PubMed] 30. De Felice, M.; Pia Postiglione, M.; Di Lauro, R. Thyrotropin receptor signalling in development and differentiation of the thyroid gland: Insights from mouse models and human diseases. Endocrinology 2004, 145, 4062–4067. [CrossRef] 31. Boas, M.; Feldt-Rasmussen, U.; Skakkebaek, N.E.; Main, K.M. Environmental chemicals and thyroid function. Eur. J. Endocrinol. 2006, 154, 599–611. [CrossRef] 32. Gao, H.; Wu, W.; Xu, Y.; Jin, Z.; Bao, H.; Zhu, P.; Su, P.; Sheng, J.; Hao, J.; Tao, F. Effects of Prenatal Phthalate Exposure on Thyroid Hormone Concentrations Beginning at The Embryonic Stage. Sci. Rep. 2017, 7, 13106. [CrossRef] 33. Meeker, J.D.; Calafat, A.M.; Hauser, R. Di(2-ethylhexyl) phthalate metabolites may alter thyroid hormone levels in men. Environ. Health Perspect. 2007, 115, 1029–1034. [CrossRef] 34. Wu, M.T.; Wu, C.F.; Chen, B.H.; Chen, E.K.; Chen, Y.L.; Shiea, J.; Lee, W.T.; Chao, M.C.; Wu, J.R. Intake of phthalate-tainted foods alters thyroid functions in Taiwanese children. PLoS ONE 2013, 8, e55005. [CrossRef] 35. Chen, F.Y.; Wang, S.S.; Wang, S.L.; Wang, R.M. Troglitazone inhibits proliferation of GH3 cell in vitro. Acad. J. Sec. Mil. Med. Univ. 2008, 29, 1052–1055. 36. Demirel, A.; Çoban, A.; Yıldırım, ¸S.;Do˘gan,C.; Sancı, R.; Ince,˙ Z. Hidden Toxicity in Neonatal Intensive Care Units: Phthalate Exposure in Very Low Birth Weight Infants. J. Clin. Res. Pediatr. Endocrinol. 2016, 8, 298–304. [CrossRef] 37. Strømmen, K.; Lyche, J.L.; Blakstad, E.W.; Moltu, S.J.; Veierød, M.B.; Almaas, A.N.; Sakhi, A.K.; Thomsen, C.; Nakstad, B.; Brække, K.; et al. Increased levels of phthalates in very low birth weight infants with septicemia and bronchopulmonary dysplasia. Environ. Int. 2016, 89–90, 228–234. [CrossRef] 38. Calafat, A.M.; Needham, L.L.; Silva, M.J.; Lambert, G. Exposure to di-(2-ethylhexyl) phthalate among premature neonates in a neonatal intensive care unit. Pediatrics 2004, 113, e429–e434. [CrossRef] 39. Green, R.; Hauser, R.; Calafat, A.M.; Weuve, J.; Schettler, T.; Ringer, S.; Huttner, K.; Hu, H. Use of di(2-ethylhexyl) phthalate- containing medical products and urinary levels of mono(2-ethylhexyl) phthalate in neonatal intensive care unit infants. Environ. Health Perspect. 2005, 113, 1222–1225. [CrossRef] 40. Pinguet, J.; Kerckhove, N.; Eljezi, T.; Lambert, C.; Moreau, E.; Bernard, L.; Boeuf, B.; Decaudin, B.; Genay, S.; Masse, M.; et al. New SPE-LC-MS/MS method for the simultaneous determination in urine of 22 metabolites of DEHP and alternative plasticizers from PVC medical devices. Talanta 2019, 198, 377–389. [CrossRef][PubMed] 41. Gaynor, J.W.; Ittenbach, R.F.; Calafat, A.M.; Burnham, N.B.; Bradman, A.; Bellinger, D.C.; Henretig, F.M.; Wehrung, E.E.; Ward, J.L.; Russell, W.W.; et al. Perioperative Exposure to Suspect Neurotoxicants From Medical Devices in Newborns With Congenital Heart Defects. Ann. Thorac. Surg. 2019, 107, 567–572. [CrossRef] 42. Eckert, E.; Müller, J.; Höllerer, C.; Purbojo, A.; Cesnjevar, R.; Göen, T.; Münch, F. Plasticizer exposure of infants during cardiac surgery. Toxicol. Lett. 2020, 330, 7–13. [CrossRef] 43. Lessmann, F.; Correia-Sá, L.; Calhau, C.; Domingues, V.F.; Weiss, T.; Brüning, T.; Koch, H.M. Exposure to the plasticizer di(2- ethylhexyl) terephthalate (DEHTP) in Portuguese children—Urinary metabolite levels and estimated daily intakes. Environ. Int. 2017, 104, 25–32. [CrossRef][PubMed] 44. Bernard, L.; Eljezi, T.; Clauson, H.; Lambert, C.; Bouattour, Y.; Chennell, P.; Pereira, B.; Sautou, V.; ARMED Study Group. Effects of ﬂow rate on the migration of different plasticizers from PVC infusion medical devices. PLoS ONE 2018, 13, e0192369. [CrossRef] [PubMed] 45. Lessmann, F.; Schütze, A.; Weiss, T.; Langsch, A.; Otter, R.; Brüning, T.; Koch, H.M. Metabolism and urinary excretion kinetics of di(2-ethylhexyl) terephthalate (DEHTP) in three male volunteers after oral dosage. Arch. Toxicol. 2016, 90, 1659–1667. [CrossRef] 46. Schriks, M.; Vrabie, C.M.; Gutleb, A.C.; Faassen, E.J.; Rietjens, I.M.; Murk, A.J. T-screen to quantify functional potentiating, antagonistic and thyroid hormone-like activities of poly halogenated aromatic hydrocarbons (PHAHs). Toxicol. Vitr. 2006, 20, 490–498. [CrossRef][PubMed] 47. Sirbasku, D.A. Thyroid hormone dependent pituitary tumor cell growth in serum-free chemically deﬁned culture. A new regulatory role for apotransferrin. Biochemistry 1991, 30, 7466–7477. [CrossRef] Metabolites 2021, 11, 94 18 of 18

48. Souza, P.C.; Puhl, A.C.; Martinez, L.; Aparicio, R.; Nascimento, A.S.; Figueira, A.C.; Nguyen, P.; Webb, P.; Skaf, M.S.; Polikarpov, I. Identiﬁcation of a new hormone-binding site on the surface of thyroid hormone receptor. Mol. Endocrinol. 2014, 28, 534–545. [CrossRef][PubMed] 49. Dow, R.L.; Schneider, S.R.; Paight, E.S.; Hank, R.F.; Chiang, P.; Cornelius, P.; Lee, E.; Newsome, W.P.; Swick, A.G.; Spitzer, J.; et al. Discovery of a Novel Series of 6-Azauracil-Based Thyroid Hormone Receptor Ligands: Potent, TRbeta Subtype-Selective Thyromimetics. Bioorg. Med. Chem. Lett. 2003, 13, 379–382. [CrossRef]