Investigation of binding and activity of perfluoroalkyl substances to the human peroxisome proliferator-activated receptor β/δ

Chuan-Hai Li ab, Xiao-Min Ren *a, Lin-Ying Cao ab, Wei-Ping Qin ab and Liang-Hong Guo *abc
aState Key Laboratory of Environmental Chemistry and Eco-toxicology, Research Center for Eco-environmental Sciences, Chinese Academy of Sciences, 18 Shuangqing Road, Beijing 100085, China. E-mail: LHGuo@rcees.ac.cn; xmren@rcees.ac.cn; Fax: +86 010 62849685; Tel: +86 010 62849685 Tel: +86 010 62849338
bCollege of Resources and Environment, University of Chinese Academy of Sciences, Beijing 100039, China
cThe Third Affiliated Hospital of Guangzhou Medical University, Guangzhou 510150, China

Received 9th May 2019 , Accepted 2nd July 2019

First published on 10th July 2019


Previously, perfluoroalkyl substances (PFASs) have been found to be associated with many adverse effects mediated by the peroxisome proliferator-activated receptor α (PPARα) and PPARγ. Here, we found another subtype of the peroxisome proliferator-activated receptors (PPARs); the PPARβ/δ mediated pathway might also be a potential adverse outcome pathway for PFASs. We investigated the direct binding and transcriptional activity of PFASs toward human PPARβ/δ, and further revealed the structure-binding and structure–activity relationship between PFASs and PPARβ/δ. The receptor binding experiment showed that their binding potency was dependent on the carbon chain length and the terminal functional group. For twelve perfluoroalkyl carboxylic acids (PFCAs), an inverted U-shaped relationship existed between the PPARβ/δ binding potency and the carbon chain length, with perfluorododecanoc acid (C12) showing the highest binding potency. The three perfluoroalkane sulfonic acids (PFSAs) exhibited a stronger binding potency than their PFCA counterparts. The two fluorotelomer alcohols (FTOHs) showed no binding potency. In receptor transcriptional activity assays, they enhanced the PPARβ/δ transcriptional activity. Their transcriptional activity was also related to the carbon chain length and the terminal functional group. Molecular docking analysis showed the PFASs fitted into the ligand binding pocket of PPARβ/δ with a binding geometry similar to a fatty acid.



Environmental significance

Perfluoroalkyl substances (PFASs) are ubiquitous environmental pollutants. Peroxisome proliferator-activated receptor α (PPARα) and PPARγ have been investigated as cellular targets of PFASs in their mode of action. In the present study, we found another subtype of PPARs; the PPARβ/δ mediated pathway might also be a potential adverse outcome pathway for PFASs. We found that PFASs could bind to the PPARβ/δ receptor directly, and showed an agonistic effect towards the PPARβ/δ signaling pathway. Their binding potency and agonistic effect is also related to the carbon chain length and terminal functional group; perfluoroalkane sulfonic acids (PFSAs) exhibited a stronger potency compared to their PFCAs counterparts.

1. Introduction

Perfluoroalkyl substances (PFASs) have been ubiquitously detected in environmental, wildlife and human samples in the past few decades.1–5 As the most prevalent compounds of PFASs, the levels of perfluorooctanoate (PFOA) and perfluorooctanesulfonate (PFOS) have declined in the serum samples of the Danish,4 Americans6,7 and Swedes8,9 after their phase-out since the early 2000s. However, the burden of these chemicals on the body remains high in some developing countries such as China.10–12 In addition, the levels of some PFASs with longer (C9–C12) or shorter (C4–C6) fluorinated alkyl chain lengths even increased in humans,6–8,13 suggesting the increased potential health risk of these PFASs.

Owing to the structural similarity of PFASs with fatty acids, peroxisome proliferator-activated receptors (PPARs, including PPARα, PPARβ/δ and PPARγ), with fatty acids as their natural ligands, were thought to be the preferred targets of PFASs.14 PFASs have been demonstrated to induce many adverse effects by directly binding to and activating PPARα or PPARγ.15,16 In addition, the structure-binding and structure–activity relationships of PFASs with PPARα and PPARγ have been revealed in detail.17,18 However, the information for PPARβ/δ, another subtype of PPARs, is very limited. Like PPARα and PPARγ, PPARβ/δ also plays essential roles in many biological functions, such as the regulation of lipid metabolism, cell proliferation, cell differentiation and cell inflammation.19–23 To comprehensively reveal the toxicity mechanism of PFASs and assess their potential toxicity effects medicated by the peroxisome proliferator-activated receptor (PPAR) family, study of the binding and activity of PFASs toward PPARβ/δ and further characterization of the structure–binding and structure–activity relationships are needed.

In our recent study using a luciferase reporter assay based on HEK 293 embryonal kidney cells, we found PFOS exhibited agonistic activity toward human PPARβ/δ.15 However, two other studies showed that PFOS and PFOA had agonistic activity to mouse PPARβ/δ but not human PPARβ/δ based on 3T3-L1 preadipocyte cells24 or COS-1 cells.25 In general, the data obtained to date is still scattered, and sometimes contradictory to each other. As only two PFASs were investigated previously, interpretation about the structure-binding and structure–activity relationship of PFASs on PPARβ/δ has been hampered.

In the present study, by using a fluorescence competitive binding assay, we evaluated the binding affinity of 17 PFASs with different carbon chain lengths and terminal functional groups to the human PPARβ/δ-ligand binding domain (LBD) in vitro. We further investigated their activity toward human PPARβ/δ at the cellular level by using a human PPARβ/δ-driven luciferase reporter assay based on HEK 293 cells. We also analyzed the structural characteristics of their binding and activity to PPARβ/δ by using molecular docking in silico.

2. Materials and methods

2.1. Chemicals

Human PPARβ/δ-LBD (with a purity greater than 85%) was prepared by Zhongding Biotechnology Co. Ltd. (Nanjing, China). The human PPARβ/δ-LBD protein sequences (NP_001165289.1, amino acids 165–441) were obtained from the NCBI GenBank (Table S1, ESI). Seventeen PFASs (with a purity greater than 95%), including twelve perfluoroalkyl carboxylic acids (PFCAs), three perfluoroalkane sulfonic acids (PFSAs), and two fluorotelomer alcohols (FTOHs), were used in this study. Their structures and detailed information are shown in Fig. S1 in the ESI.

2.2. Competitive binding assay

The binding affinity of PFASs with human PPARβ/δ-LBD was determined using a fluorescence polarization (FP)-based competitive binding assay using C1-BODIPY-C12 as the fluorescence probe. The IC50 value (the concentration of the ligand required to displace half of the probe from PPARβ/δ-LBD) was obtained based on the competition curves processed using Origin 8.5 (OriginLab, Northampton, MA, USA) using the sigmoidal model. The relative binding potency (RP) was calculated by dividing the IC50 of linoleic acid (LA) by that of the PFASs. The details of the method are the same as that reported in our previously published study.15

2.3. PPARβ/δ mediated luciferase reporter assay

A cell-based human PPARβ/δ-driven luciferase reporter assay was performed by transfecting a pBIND-PPARβ/δ vector, pGL4.35[luc2P/9XGAL4UAS/Hygro] vector and PRL-TK vector into HEK 293 cells (American Type Culture Collection (ATCC), Manassas, VA). The activity of the 17 PFASs was detected at non-cytotoxic concentrations (Fig. S3, ESI). The detailed procedures for vector construction and transient transfection are provided in the ESI.

2.4. Molecular docking

The 3D crystal structure of human PPARβ/δ-LBD (PDB ID: 3OZ0) was obtained from the RCSB Protein Data Bank (http://www.rcsb.org/pdb). The structures of PFASs were prepared using ChemBioDraw Ultra and transformed into the PDB format through the PRODRG server.26 PFASs were docked into the ligand binding pocket of human PPARβ/δ-LBD by using a Lamarckian genetic algorithm provided by Auto Dock 4.2 (Scripps Research Institute, CA, USA). The details of docking were the same as those reported in our previously published study.15,17

2.5. Statistical analysis

All of the experiments were conducted in triplicate and the data are expressed in terms of the mean ± SEM (the standard error of mean). Comparison of the mean values among the experimental groups were performed using SPSS 17.0 software (Chicago, IL, USA) using one-way analysis of variance (ANOVA), with a significance level set at *p < 0.05.

3. Results and discussion

3.1. Binding potency of PFASs with human PPARβ/δ-LBD

Linoleic acid (LA), as a positive control, inhibited the binding of the C1-BODIPY-C12 probe to human PPARβ/δ-LBD in a dose-dependent manner with an IC50 value of 31.1 μM (Fig. S2, ESI), suggesting the accuracy of the method. For the 12 PFCAs, three PFCAs (PFBA, PFHxA, PFHpA) could not displace the C1-BODIPY-C12 probe from the human PPARβ/δ-LBD, even at the highest tested concentration, suggesting that they could not bind to PPARβ/δ-LBD. For the other nine PFCAs, they could bind to human PPARβ/δ-LBD with IC50 values ranging from not obtained to 32.6 μM (Table 1). However, unlike PFCAs with carbon chain lengths of C10–C13, the binding of PFCAs with carbon chain lengths of C9, C14, C16 and C18 did not reach full displacement even at the highest concentration. A higher concentration, which might lead to full probe displacement, was not available owing to their solubility. As shown in Fig. 1 and Table 1, the binding potency of PFCAs with human PPARβ/δ-LBD increased when the carbon chain length increased from C7 (no binding) to C12 (with a RP of 95% compared to LA). However, when the carbon chain length was longer than C12, there was no further increase but a decrease in the binding potency was observed. For the three PFSAs, PFOS (with an IC50 value of 76.9 μM) and PFHxS (with an IC50 value of not obtained) bound to human PPARβ/δ-LBD, while PFBS showed no binding (Table 1 and Fig. 1). For the two FTOHs, no PPARβ/δ-LBD binding was found.
Table 1 IC50, RP (compared with LA) values and hydrogen bond interactions of fatty acids and PFASs for human PPARβ/δ using the competitive binding assay and molecular docking analysisa
Chemicals Molecular Formula Carbon Number IC50 (μM) RP (fold) Hydrogen Bonds
a Note: — indicates no information was collected at that particular examination point; NA indicates not available (a compound could displace the fluorescence probe from the PPARβ/δ LBD but could not reveal the IC50 value). IC50 values followed by different letters were determined using ANOVA, with a significance level set at p < 0.05.
AA CH3(CH2)4(CH = CH–CH2)4, (CH2)2COOH 20 Thr289, His323, His449, Tyr473
LA CH3(CH2)4(CH = CH–CH2)2, (CH2)6COOH 18 31.1 1
PFBA CF3(CF2)2COOH 4 NA NA Thr289, His449, Tyr473
PFHXA CF3(CF2)4COOH 6 NA NA Thr289, His323, His449, Tyr473
PFHpA CF3(CF2)5COOH 7 NA NA Thr289, His449, Tyr473
PFOA CF3(CF2)6COOH 8 NA NA Thr289, His449, Tyr473
PFNA CF3(CF2)7COOH 9 127.9 ± 7.0b 0.24 Thr289, His449, Tyr473
PFDA CF3(CF2)8COOH 10 56.6 ± 14.3d 0.55 Thr289, His323, His449, Tyr473
PFUnA CF3(CF2)9COOH 11 47.7 ± 6.3d 0.65 Thr289, His449, Tyr473
PFDoA CF3(CF2)10COOH 12 32.6 ± 6.1e 0.95 Thr289, His323, Tyr473
PFTrDA CF3(CF2)11COOH 13 52.2 ± 5.8d 0.60 His449, Tyr473
PFTeDA CF3(CF2)12COOH 14 110.8 ± 10.5c 0.28 Thr289, His449
PFHxDA CF3(CF2)14COOH 16 159.6 ± 11.5a 0.19 Thr289, His449, Tyr473
PFOcDA CF3(CF2)16COOH 18 NA NA His323, Tyr473
PFBS CF3(CF2)3SO3H 4 NA NA Thr289, His323, His449
PFHxS CF3(CF2)5SO3H 6 NA NA Thr289, His449, Tyr473
PFOS CF3(CF2)7SO3H 8 76.9 ± 9.2 0.40 Thr289, His449, Tyr473
6[thin space (1/6-em)]:[thin space (1/6-em)]2 FTOH CF3(CF2)5CH2CH2OH 8 NA NA His449
8[thin space (1/6-em)]:[thin space (1/6-em)]2 FTOH CF3(CF2)7CH2CH2OH 10 NA NA His449



image file: c9em00218a-f1.tif
Fig. 1 Competitive binding curves of PFASs with human PPARβ/δ-LBD. (A) Competitive binding curves of PFCAs with carbon chain lengths shorter than C11 to human PPARβ/δ-LBD. (B) Competitive binding curves of PFCAs with carbon chain lengths longer than C11 to human PPARβ/δ-LBD. (C) Competitive binding curves of three PFSAs (PFBS, PFHxS and PFOS) to human PPARβ/δ-LBD. (D) Competitive binding curves of two FTOHs (6[thin space (1/6-em)]:[thin space (1/6-em)]2 FTOH and 8[thin space (1/6-em)]:[thin space (1/6-em)]2 FTOH) to human PPARβ/δ-LBD. The error bars represent the standard deviation of three measurements.

As shown in Fig. S2c in the ESI, a typical inverted U-shaped relationship existed between the PPARβ/δ binding potency (showed with the value of 1/IC50) and the carbon chain length, with perfluorododecanoc acid (PFDoA, C12) showing the highest binding potency. In previous studies, the inverted U-shaped relationship between the protein binding potency and the carbon chain length has also been observed in some other nuclear receptors, such as PPARγ (with C11 showing the highest binding potency)17 and the thyroid hormone receptor (TR, with C10 showing the highest binding potency),27 or transport proteins such as the human liver fatty acid-binding protein (hl-FABP, with C11 showing the highest binding potency)28 and transthyretin (TTR, with C8 showing the highest binding potency).29 We speculated that the difference in the optimal carbon chain length might be due to the different size of the ligand binding pockets of these proteins. For the PFCAs with short carbon chain lengths, their hydrophobicity increased with the increasing carbon chain length, which might lead to a stronger hydrophobic interaction with PPARβ/δ. However, the cavity of PPARβ/δ might not be big enough to accommodate some longer chain PFCAs, which might lead to a decrease in the binding potency. Furthermore, we found that the PPARβ/δ binding potency was also dependent on the terminal functional groups of the same carbon chain length in the order of sulfonate > carboxylate > alcoholic hydroxyl. This order was the same as that observed for the binding potency of PFASs with PPARγ,17 TR,27 TTR,29 hl-FABP28 and the estrogen receptor (ER).30

In a real situation, short and long chain PFASs might co-exist in the organisms. A previous study reported that the properties of bioconcentration and tissue distribution were different between short chain PFASs and long chain PFASs in zebrafish.31 Moreover, long chain PFASs could inhibit the bioconcentration of short chain PFASs in zebrafish, which might be due to the competitive effect of long chain PFASs on the binding potency of shorter chain PFASs with important proteins in zebrafish.32 According to our results, the binding potencies of PFASs with different carbon chain lengths were different. This indicated that the co-exposure of PFASs with different chain lengths may cause competitive effects between longer chain PFASs and shorter chain PFASs, which may lead to different environmental effects compared with the exposure of a single compound.

By comparing the relative binding potency (RP, compared with the natural ligand of the receptor) of the same PFAS (PFOS for example) with PPARβ/δ (with an RP of 40% compared with LA) to others nuclear receptors, such as PPARγ (with an RP of 5% compared with OA),17 TR (with an RP of 2% compared with T3)27 and ER (with an RP of 0.7% compared with E2),30 we found that PFOS showed a higher PPARβ/δ relative binding potency than the other receptors. These results indicated that PFOS might have a higher disruption effect toward PPARβ/δ than the other receptors in vivo.

3.2. Activity of PFASs toward the PPARβ/δ signaling pathway

GW501516 and arachidonic acid (AA), a known PPARβ/δ agonist, enhanced the PPARβ/δ transcriptional activity in a dose-dependent manner (Fig. 2), confirming the accuracy of the method. As shown in Fig. 3, with the exception of 8[thin space (1/6-em)]:[thin space (1/6-em)]2 FTOH, all PFASs enhanced the transcriptional activity in a dose-dependent manner, suggesting they had agonistic activity towards the PPARβ/δ signaling pathway. Among the 12 PFCAs, perfluorotridecanoic acid (PFTrDA, C13) exerted the highest transcriptional activity, with the lowest effective concentration (LOEC) of 10 μM and the highest transcriptional activity of 2.5-fold at 50 μM (Fig. 3a). For the three PFSAs, they had the highest transcriptional activity of 1.4 to 1.7-fold at 50 μM (Fig. 3b). 6[thin space (1/6-em)]:[thin space (1/6-em)]2 FTOH had the highest transcriptional activity of 1.2-fold at 50 μM (Fig. 3c). However, our observed results for PFOA and PFOS towards human PPARβ/δ disagreed with the results of previous studies which showed that these two PFASs had no agonistic activity to human PPARβ/δ.24,25,34 For example, Vanden Heuvel et al. and Takacs et al. found that both PFOA and PFOS could not activate human PPARβ/δ in 3T3-L1 fibroblast cells or COS-1 cell based transient transfection assays.24,25 Buhrke et al. also showed that PFOA could not activate human PPARβ/δ by using a HEK 293 cell based transient transfection assay.34 By comparing the experimental details, we found the cell type and reporter system used in the previous studies were different from ours, which might lead to different results.
image file: c9em00218a-f2.tif
Fig. 2 Effects of GW501516 (A) and AA (B) on human PPARβ/δ mediated luciferase reporter gene transcription activity in HEK 293 cells. The relative luciferase activity was determined by setting 0.1% DMSO (Veh) treated cells as one. *p < 0.05 compared with Veh.

image file: c9em00218a-f3.tif
Fig. 3 Effects of 17 PFASs on the human PPARβ/δ-mediated luciferase reporter gene transcription activity in HEK 293 cells and the relationship between the transcription activity and carbon chain lengths of PFCAs. (A) Effects of 12 PFCAs on luciferase reporter gene transcription activity. (B) Effects of three PFSAs on luciferase reporter gene transcription activity. (C) Effects of two FTOHs on luciferase reporter gene transcription activity. (D) The relationship between PPARβ/δ mediated luciferase reporter gene transcription activity (luciferase transcriptional activity determined at 50 μM) and the carbon chain length of PFCAs. The relative luciferase activity was determined by setting 0.1% DMSO (Veh) treated cells as one. *p < 0.05 compared with Veh.

An inverted U-shaped relationship was also observed between the PPARβ/δ transcriptional activity and the carbon chain length of the PFCAs, with PFTrDA (C13) showing the highest activity (Fig. 3a and d). Moreover, the inverted U-shaped relationship had also been observed when studying the activity of PFCAs toward PPARα (with C8 showing the highest agonistic activity) and PPARγ (with C11 showing the highest agonistic activity).17,18 The transcriptional activity of PFASs toward PPARβ/δ was also dependent on the terminal functional group in the order of sulfonate > carboxylate > alcoholic hydroxyl, which was the same as that observed for PPARα and PPARγ previously.17,18

Generally, a PFAS with a higher PPARβ/δ binding potency showed a higher PPARβ/δ transcriptional activity, suggesting the effect of the binding potency on their transcriptional activity. The only inconformity was that PFTrDA (C13), which has a weaker binding potency than PFDoA (C12), showed a higher transcriptional activity at the cellular level. Once a ligand enters into the cell, it should bind to the nuclear receptor for transactivation. Presumably, more ligand molecules should occupy more receptors for transactivation. Previously, Rosenmai et al. determined the bio-availability of different PFASs in Hep G2 liver cancer cells, showing the amount of PFTrDA that entered into cells was higher than that of PFDoA.18 Therefore, we inferred the higher transcriptional activity of PFTrDA might be due to the higher amount of PFTrDA in cells and we inferred that the higher transcriptional activity of PFTrDA was due to its higher bio-availability. Our results suggested that, in addition to the binding potency, the bio-availability of PFASs might also affect the transcriptional activity toward PPARβ/δ.

Previous studies have demonstrated PFASs exert agonistic activity toward PPARα and PPARγ mediated signaling pathways.16–18 Here, we found that PFASs also exerted agonistic activity towards the PPARβ/δ signaling pathway. In our previous study, we compared the transcriptional activity and adipogenesis activity of PFOS toward three subtypes of PPARs, and found that PFOS showed an even higher activity to PPARβ/δ as compared with PPARα and PPARγ.15 By combining the results of the receptor binding and transcriptional activity assay, we found that PPARβ/δ might be a priority cellular target when PFASs enter into an organism, especially in those cells with abundant PPARβ/δ expression.33 Combining the results of our present study and those of previous studies, we suggest that the toxicity effects mediated through the PPARβ/δ pathway should not be neglected.

3.3. Molecular docking of PFASs interactions with PPARβ/δ

The ligand binding pocket of PPARβ/δ is a “Y”-shaped ligand-binding cavity and consists of a three-arm binding cavity (Fig. 4a).35 Arm I, in which the entrance of the ligand binding pocket is located, is a substantially polar cavity which could form hydrogen bonds with the polar acid end group of ligands. Arm II and arm III, which are located in the inner part of the ligand binding pocket, are two hydrophobic cavities.35 As shown in Fig. 4a, AA was docked into PPARβ/δ with its hydrophobic alkyl chain residing toward the inner part (staying at arm II) and the polar acid end group residing toward the entrance of the ligand binding pocket (staying in arm I). It formed hydrogen bonds with residues Thr289, His323, His449 and Try473 (Fig. 4a and Table 1). The docking results for AA were almost in line with the results obtained from the crystal study previously,36 indicating the feasibility and accuracy of the docking method.
image file: c9em00218a-f4.tif
Fig. 4 Molecular docking results of: (A) AA; (B) PFDA; (C) PFDoA; and (D) PFTrDA with human PPARβ/δ-LBD. The human PPARβ/δ-LBD is represented in blue, and the chemicals are colored by the atom type (carbon is gray, oxygen is red, fluorine is green and sulfur is yellow). The orange part shows the three arms of the binding pocket of human PPARβ/δ-LBD.

As shown in Fig. 4 and S4 (ESI), the PFASs fitted into the ligand binding pocket of PPARβ/δ with a binding geometry similar to AA, with their hydrophobic fluorinated alkyl chain residing towards the inner part (staying at arm II or arm III) and the polar end group residing towards the entrance of the ligand binding pocket (staying in arm I). The PFCAs with different chain lengths bound to PPARβ/δ with a slightly different binding geometry. For the PFCAs with a carbon chain length shorter than C11, they bound to PPARβ/δ with their hydrophobic part located at arm III (Fig. 4 and S4 in the ESI). For the PFCAs with a carbon chain length longer than C11, they all bound to PPARβ/δ with their hydrophobic part located at arm II but not arm III, this binding geometry conformation was very similar to AA, which might lead to long chain PFCAs exhibiting a stronger binding potency than short chain PFCAs. However, for the PFCAs with a carbon chain length longer than C12, their PPARβ/δ binding potency decreased when the carbon chain length increased. According to the results of the molecular docking (Fig. 4 and S4 in the ESI), PFCAs with carbon chain length more than 11 (from PFUnA to PFOcDA) had to bend to fit the cavity. However, owing to the strong electronegativity of fluorine, which makes the PFASs rigid, it seems impossible for these PFASs to bend.28,31 This might be a reason that the inverted U-shaped relationship between the protein binding potency and the carbon chain length has also been observed for PPARβ/δ. Combining the results of the receptor binding and transcriptional activity assays, we found that PFOS showed a higher response than PFOA. As the number of fluorinated carbons of PFOA was one less than PFOS, PFOS might have a higher hydrophobicity interaction with PPARβ/δ than PFOA. Furthermore, the sulfonic acid group of PFOS is more hydrophilic than the carboxylic acid group of PFOA, which might mean that it is easier for PFOS to form hydrogen bonds with PPARβ/δ than PFOA. These two reasons might explain our experimental results that PFOS displayed a stronger potency in the receptor binding and activation than PFOA.

As shown in Table 1, PFASs formed hydrogen bonds with some or all of the four residues that AA formed hydrogen bonds with. Previous studies have shown that hydrogen bond interaction with the activation function 2 (AF-2) region play important roles in the PPARβ/δ receptor binding and activation.35,37 Hydrogen bond interactions with the residues in the AF-2 region could lead to packing of the AF-2 helix into the binding pocket of PPARβ/δ, resulting in coactivator binding and PPARβ/δ mediated gene transcription.35,37 Therefore, we inferred that the PFASs might activate PPARβ/δ in a similar manner to the fatty acid. Previously, we compared the binding mode of PFOS with PPARα and PPARβ/δ, as well as PPARγ, and found PFOS formed hydrogen bond interactions with the residues in the AF-2 region of all of these three PPARs.15 Therefore, the PFASs might activate PPARs in the same manner.

4. Conclusions

In the present study, we demonstrated PFASs bound to and activated human PPARβ/δ-LBD directly. The PPARβ/δ binding potency and transcriptional activity of PFASs were all related to the carbon chain length and the terminal functional group. Our results suggest that, in addition to the PPARα and PPARγ mediated pathways, the PPARβ/δ mediated pathway should also be taken into consideration when studying the toxicity mechanism of PFASs.

Conflicts of interest

The authors declare that there are no conflicts of interest.

Acknowledgements

This work was supported by the Chinese Academy of Sciences (XDB14040100, QYZDJ-SSW-DQC020), the National Natural Science Foundation of China (91543203, 21621064, and 21777187), and the Royal Society International Collaboration Awards for Research Professors (IC160121).

References

  1. W. Dong, B. Liu, Y. Song, H. Zhang, J. Li and X. Cui, Occurrence and Partition of Perfluorinated Compounds (PFCs) in Water and Sediment from the Songhua River, China, Arch. Environ. Contam. Toxicol., 2018, 74(3), 492–501 CrossRef CAS PubMed.
  2. P. Karásková, M. Venier, L. Melymuk, J. Bečanová, Š. Vojta, R. Prokeš, M. L. Diamond and J. Klánová, Perfluorinated alkyl substances (PFASs) in household dust in Central Europe and North America, Environ. Int., 2016, 94, 315–324 CrossRef PubMed.
  3. M. D. Sedlak, J. P. Benskin, A. Wong, R. Grace and D. J. Greig, Per-and polyfluoroalkyl substances (PFASs) in San Francisco Bay wildlife: Temporal trends, exposure pathways, and notable presence of precursor compounds, Chemosphere, 2017, 185, 1217–1226 CrossRef CAS PubMed.
  4. C. Bjerregaard-Olesen, C. C. Bach, M. Long, M. Ghisari, R. Bossi, B. H. Bech, E. A. Nohr, T. B. Henriksen, J. Olsen and E. C. Bonefeld-Jørgensen, Time trends of perfluorinated alkyl acids in serum from Danish pregnant women 2008–2013, Environ. Int., 2016, 91, 14–21 CrossRef CAS PubMed.
  5. L. Roosens, W. D'Hollander, L. Bervoets, H. Reynders, K. Van Campenhout, C. Cornelis, R. Van Den Heuvel, G. Koppen and A. Covaci, Brominated flame retardants and perfluorinated chemicals, two groups of persistent contaminants in Belgian human blood and milk, Environ. Pollut., 2010, 158(8), 2546–2552 CrossRef CAS PubMed.
  6. G. W. Olsen, C. C. Lange, M. E. Ellefson, D. C. Mair, T. R. Church, C. L. Goldberg, R. M. Herron, Z. Medhdizadehkashi, J. B. Nobiletti and J. A. Rios, Temporal trends of perfluoroalkyl concentrations in American Red Cross adult blood donors, 2000–2010, Environ. Sci. Technol., 2012, 46(11), 6330–6338 CrossRef CAS PubMed.
  7. M. Wang, J.-S. Park and M. Petreas, Temporal changes in the levels of perfluorinated compounds in California women's serum over the past 50 years, Environ. Sci. Technol., 2011, 45(17), 7510–7516 CrossRef CAS PubMed.
  8. A. Axmon, J. Axelsson, K. Jakobsson, C. H. Lindh and B. A. Jönsson, Time trends between 1987 and 2007 for perfluoroalkyl acids in plasma from Swedish women, Chemosphere, 2014, 102, 61–67 CrossRef CAS PubMed.
  9. A. Glynn, U. Berger, A. Bignert, S. Ullah, M. Aune, S. Lignell and P. O. Darnerud, Perfluorinated alkyl acids in blood serum from primiparous women in Sweden: serial sampling during pregnancy and nursing, and temporal trends 1996–2010, Environ. Sci. Technol., 2012, 46(16), 9071–9079 CrossRef CAS PubMed.
  10. Z. Zhou, Y. Shi, R. Vestergren, T. Wang, Y. Liang and Y. Cai, Highly elevated serum concentrations of perfluoroalkyl substances in fishery employees from Tangxun lake, china, Environ. Sci. Technol., 2014, 48(7), 3864–3874 CrossRef CAS PubMed.
  11. Y. Gao, J. Fu, H. Cao, Y. Wang, A. Zhang, Y. Liang, T. Wang, C. Zhao and G. Jiang, Differential accumulation and elimination behavior of perfluoroalkyl acid isomers in occupational workers in a manufactory in China, Environ. Sci. Technol., 2015, 49(11), 6953–6962 CrossRef CAS PubMed.
  12. J. Fu, Y. Gao, L. Cui, T. Wang, Y. Liang, G. Qu, B. Yuan, Y. Wang, A. Zhang and G. Jiang, Occurrence, temporal trends, and half-lives of perfluoroalkyl acids (PFAAs) in occupational workers in China, Sci. Rep., 2016, 6, 38039 CrossRef CAS PubMed.
  13. K. Kato, L.-Y. Wong, L. T. Jia, Z. Kuklenyik and A. M. Calafat, Trends in exposure to polyfluoroalkyl chemicals in the US population: 1999–2008, Environ. Sci. Technol., 2011, 45(19), 8037–8045 CrossRef CAS PubMed.
  14. B. Grygiel-Górniak, Peroxisome proliferator-activated receptors and their ligands: nutritional and clinical implications-a review, Nutr. J., 2014, 13(1), 17 CrossRef PubMed.
  15. C.-H. Li, X.-M. Ren, T. Ruan, L.-Y. Cao, Y. Xin, L.-H. Guo and G. Jiang, Chlorinated Polyfluorinated Ether Sulfonates Exhibit Higher Activity towards Peroxisome Proliferator-Activated Receptors Signaling Pathways than Perfluorooctane Sulfonate, Environ. Sci. Technol., 2018, 52(5), 3232–3239 CrossRef CAS PubMed.
  16. M. A. Cwinn, S. P. Jones and S. W. Kennedy, Exposure to perfluorooctane sulfonate or fenofibrate causes PPAR-α dependent transcriptional responses in chicken embryo hepatocytes, Comp. Biochem. Physiol., Part C: Toxicol. Pharmacol., 2008, 148(2), 165–171 Search PubMed.
  17. L. Zhang, X.-M. Ren, B. Wan and L.-H. Guo, Structure-dependent binding and activation of perfluorinated compounds on human peroxisome proliferator-activated receptor γ, Toxicol. Appl. Pharmacol., 2014, 279(3), 275–283 CrossRef CAS PubMed.
  18. A. K. Rosenmai, L. Ahrens, T. G. Le, J. Lundqvist and A. Oskarsson, Relationship between peroxisome proliferator-activated receptor alpha activity and cellular concentration of 14 perfluoroalkyl substances in HepG2 cells, J. Appl. Toxicol., 2018, 38(2), 219–226 CrossRef CAS PubMed.
  19. K. Tachibana, D. Yamasaki and K. Ishimoto, The role of PPARs in cancer, PPAR Res., 2008, 102737 Search PubMed.
  20. W. R. Oliver, J. L. Shenk, M. R. Snaith, C. S. Russell, K. D. Plunket, N. L. Bodkin, M. C. Lewis, D. A. Winegar, M. L. Sznaidman and M. H. Lambert, A selective peroxisome proliferator-activated receptor δ agonist promotes reverse cholesterol transport, Proc. Natl. Acad. Sci. U. S. A., 2001, 98(9), 5306–5311 CrossRef CAS PubMed.
  21. D. K. Krämer, L. Al-Khalili, S. Perrini, J. Skogsberg, P. Wretenberg, K. Kannisto, H. Wallberg-Henriksson, E. Ehrenborg, J. R. Zierath and A. Krook, Direct activation of glucose transport in primary human myotubes after activation of peroxisome proliferator-activated receptor δ, Diabetes, 2005, 54(4), 1157–1163 CrossRef PubMed.
  22. T. Nagasawa, Y. Inada, S. Nakano, T. Tamura, T. Takahashi, K. Maruyama, Y. Yamazaki, J. Kuroda and N. Shibata, Effects of bezafibrate, PPAR pan-agonist, and GW501516, PPARδ agonist, on development of steatohepatitis in mice fed a methionine- and choline-deficient diet, Eur. J. Pharmacol., 2006, 536(1–2), 182–191 CrossRef CAS PubMed.
  23. J. M. Peters and F. J. Gonzalez, Sorting out the functional role (s) of peroxisome proliferator-activated receptor-β/δ (PPARβ/δ) in cell proliferation and cancer, Biochim. Biophys. Acta, 2009, 1796(2), 230–241 CAS.
  24. J. P. Vanden Heuvel, J. T. Thompson, S. R. Frame and P. J. Gillies, Differential activation of nuclear receptors by perfluorinated fatty acid analogs and natural fatty acids: a comparison of human, mouse, and rat peroxisome proliferator-activated receptor-α, -β, and -γ, liver X receptor-β, and retinoid X receptor-α, Toxicol. Sci., 2006, 92(2), 476–489 CrossRef CAS PubMed.
  25. M. L. Takacs and B. D. Abbott, Activation of mouse and human peroxisome proliferator-activated receptors (α, β/δ, γ) by perfluorooctanoic acid and perfluorooctane sulfonate, Toxicol. Sci., 2006, 95(1), 108–117 CrossRef PubMed.
  26. A. W. SchuÈttelkopf and D. M. Van Aalten, PRODRG: a tool for high-throughput crystallography of protein–ligand complexes, Acta Crystallogr., Sect. D: Biol. Crystallogr., 2004, 60(8), 1355–1363 CrossRef PubMed.
  27. X.-M. Ren, Y.-F. Zhang, L.-H. Guo, Z.-F. Qin, Q.-Y. Lv and L.-Y. Zhang, Structure–activity relations in binding of perfluoroalkyl compounds to human thyroid hormone T3 receptor, Arch. Toxicol., 2015, 89(2), 233–242 CrossRef CAS PubMed.
  28. L. Zhang, X.-M. Ren and L.-H. Guo, Structure-based investigation on the interaction of perfluorinated compounds with human liver fatty acid binding protein, Environ. Sci. Technol., 2013, 47(19), 11293–11301 CrossRef CAS PubMed.
  29. J. M. Weiss, P. L. Andersson, M. H. Lamoree, P. E. Leonards, S. P. van Leeuwen and T. Hamers, Competitive binding of poly-and perfluorinated compounds to the thyroid hormone transport protein transthyretin, Toxicol. Sci., 2009, 109(2), 206–216 CrossRef CAS PubMed.
  30. A. D. Benninghoff, W. H. Bisson, D. C. Koch, D. J. Ehresman, S. K. Kolluri and D. E. Williams, Estrogen-like activity of perfluoroalkyl acids in vivo and interaction with human and rainbow trout estrogen receptors in vitro, Toxicol. Sci., 2010, 120(1), 42–58 CrossRef PubMed.
  31. W. Wen, X. Xia, D. Zhou, H. Wang, Y. Zhai, H. Lin, J. Chen and D. Hu, Bioconcentration and tissue distribution of shorter and longer chain perfluoroalkyl acids (PFAAs) in zebrafish (Danio rerio): Effects of perfluorinated carbon chain length and zebrafish protein content, Environ. Pollut., 2019, 249, 277–285 CrossRef CAS PubMed.
  32. W. Wen, X. Xia, D. Hu, D. Zhou, H. Wang, Y. Zhai and H. Lin, Long-chain perfluoroalkyl acids (PFAAs) affect the bioconcentration and tissue distribution of short-chain PFAAs in zebrafish (Danio rerio), Environ. Sci. Technol., 2017, 51(21), 12358–12368 CrossRef CAS PubMed.
  33. B. D. Abbott, Review of the expression of peroxisome proliferator-activated receptors alpha (PPAR alpha), beta (PPAR beta), and gamma (PPAR gamma) in rodent and human development, Reprod. Toxicol., 2009, 27(3), 246–257 CrossRef CAS PubMed.
  34. T. Buhrke, A. Kibellus and A. Lampen, In vitro toxicological characterization of perfluorinated carboxylic acids with different carbon chain lengths, Toxicol. Lett., 2013, 218(2), 97–104 CrossRef CAS PubMed.
  35. C.-C. Wu, T. J. Baiga, M. Downes, J. J. La Clair, A. R. Atkins, S. B. Richard, W. Fan, T. A. Stockley-Noel, M. E. Bowman and J. P. Noel, Structural basis for specific ligation of the peroxisome proliferator-activated receptor δ, Proc. Natl. Acad. Sci. U. S. A., 2017, 114(13), E2563–E2570 CrossRef CAS PubMed.
  36. C. A. Luckhurst, L. A. Stein, M. Furber, N. Webb, M. J. Ratcliffe, G. Allenby, S. Botterell, W. Tomlinson, B. Martin and A. Walding, Discovery of isoindoline and tetrahydroisoquinoline derivatives as potent, selective PPARδ agonists, Bioorg. Med. Chem. Lett., 2011, 21(1), 492–496 CrossRef CAS PubMed.
  37. V. Zoete, A. Grosdidier and O. Michielin, Peroxisome proliferator-activated receptor structures: ligand specificity, molecular switch and interactions with regulators, Biochim. Biophys. Acta, 2007, 1771(8), 915–925 CrossRef CAS PubMed.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c9em00218a

This journal is © The Royal Society of Chemistry 2019