Skip to main content
NCBI home page
EPA Author Manuscripts logoLink to EPA Author Manuscripts
. Author manuscript; available in PMC: 2021 May 5.
Published in final edited form as: Environ Sci Technol. 2020 Apr 16;54(9):5676–5686. doi: 10.1021/acs.est.0c00049

Nontarget Screening of Per- and Polyfluoroalkyl Substances Binding to Human Liver Fatty Acid Binding Protein

Diwen Yang a, Jiajun Han a, David Ross Hall a, Jianxian Sun a, Jesse Fu a, Steven Kutarna a, Keith A Houck b, Carlie A LaLone c, Jon A Doering c,d, Carla A Ng e, Hui Peng a,f,*
PMCID: PMC7477755  NIHMSID: NIHMS1618951  PMID: 32249562

Abstract

More than one thousand per- and polyfluoroalkyl substances (PFASs) have been discovered by nontarget analysis (NTA), but their prioritization for health concerns is challenging. We developed a method by incorporating size exclusion column co-elution (SECC) and NTA, to screen PFASs binding to human liver fatty acid binding protein (hL-FABP). Of 74 PFASs assessed, 20 were identified as hL-FABP ligands in which 8 of them have high binding affinities. Increased PFASs binding affinities correlate with stronger responses in electrospray ionization (ESI−) and longer retention times on C18 column. This is well explained by a mechanistic model which revealed that both polar and hydrophobic interactions are crucial for binding affinities. Encouraged by this, we then developed a SECC method to identify hL-FABP ligands, and all 8 high-affinity ligands were selectively captured from 74 PFASs. The method was further applied to an aqueous film-forming foam (AFFF) product in which 31 new hL-FABP ligands were identified. Suspect and nontargeted screening revealed these ligands as analogues of perfluorosulfonic acids, and homologues of alkyl ether sulfates (C8- and C10/EOn, C8H17(C2H4O)nSO4− and C10H21(C2H4O)nSO4−). The SECC method was then applied to AFFF-contaminated surface waters. In addition to perfluorooctanesulfonic acid (PFOS) and perfluorohexanesulfonic acid (PFHxS), 8 other AFFF chemicals were discovered as novel ligands, including four C14- and C15/EOn. This study implemented a high-throughput method to prioritize PFASs and revealed the existence of many previously unknown hL-FABP ligands.

Keywords: Nontarget analysis, Physical interactions, Hydrophobicity, Size exclusion chromatography, Aqueous film-forming foams

TOC

graphic file with name nihms-1618951-f0008.jpg

Introduction

Per- and polyfluoroalkyl substances (PFASs) have been used in a variety of industrial and consumer applications including firefighting foams, insecticides, textile and surface coatings in the past decades.1–3 Of the PFASs studied to date, many are extremely stable and persistent in the environment due to their chemical structures with strong carbon-fluorine bonds. The broad distributions of PFASs in water, wildlife, air and human blood have been reported worldwide.3–6 Since the early 2000s, due to concerns regarding their persistence, bioaccumulation and toxicity, perfluorooctanesulfonic acid (PFOS), perfluorooctanoic acid (PFOA) and related PFASs have been phased out in some parts of the world.7 While these well-known classes of PFASs (e.g., perfluorinated acids) have been intensively studied and regulated in regulatory policies, the Toxic Substances Control Act (TSCA) Inventory lists 602 PFASs which are still actively used in U.S. commerce.8 Identification of the unknown PFASs in the environment and their potential health risks is thus an urgent need.

Recent advances in high-resolution mass spectrometry based nontarget analysis (NTA) provide an opportunity to characterize unknown PFASs in the environment.9–12 The existence of >1,000 PFASs in commercial products (e.g., aqueous film-forming foams, AFFFs)12–14 and the environmental matrices including surface waters,11, 15, 16 wastewaters,9 particulate matter10 and sediments,17 has been revealed by the recent NTA studies. The detections of diverse PFASs in fish,18 polar bears19 and human serum20 provide evidence for the bioaccumulation of these PFASs from the environment to organisms. The bioaccumulation of previously unknown PFASs was also observed in laboratory cultured rainbow trout exposed to the AFFF technical products.21 While these studies raised concerns about previously unknown PFASs, prioritization of these PFASs according to their potential health risks related to their bioaccumulation and toxicities is challenging due to their high number and diverse structures.

Liver fatty acid binding protein (L-FABP) is a critical protein regulating the toxicokinetics of PFASs.22–26 Previous studies have documented that the different binding affinities of classic PFASs to L-FABP contribute to their tissue distributions27, bioconcentration potentials28 and placental transfers26. Physiologically based pharmacokinetic (PBPK) modeling also supports that L-FABP plays an essential role in determining the bioaccumulation of PFASs in organisms.29, 30 L-FABP is also a key transporter to deliver lipophilic ligands to peroxisome proliferator activated receptors (PPARs),31, 32 the nuclear receptors, mediating certain toxicities of PFASs.33-36L-FABP may impact the toxicities of PFASs by directly regulating the signaling pathways of PPARs. For instance, sex-dependent impacts of perfluorononanoic acid (PFNA) on triglycerides in zebrafish liver, were found to be mediated by L-FABP, via opposite regulations of expressions of L-FABP.37 Regarding the critical role of L-FABP in toxicity and bioaccumulation, recent studies proposed to prioritize PFASs according to L-FABP binding affinities.27, 38 To address the challenge of prioritizing a large number of PFASs which is infeasible for traditional bioassay-based toxicity tests, the time- and cost-effective computational methods such as molecular dynamics simulations and molecular docking were proposed.27, 38–40 While in silico prediction is promising when a high-quality training set is available, the accuracy of prediction for unknown PFASs is challenged by their diverse structures. Thus, the implementation of a high-throughput experimental method to identify unknown PFASs binding to L-FABP is of great interest.

In this study, we aim to develop an experimental method for nontargeted screening of PFASs binding to hL-FABP, from the environmental mixtures including an AFFF technical product and AFFF-contaminated waters.41 To achieve this, we implemented a method by employing Size-Exclusion Chromatography-Coelution (SECC) and NTA to screen hL-FABP ligands with high selectivity; and the SECC method was then applied to an AFFF product and AFFF-contaminated river waters in which 39 AFFF chemicals were discovered as new hL-FABP ligands.

Materials and Methods

Chemicals and Reagents.

Seventy-four per- and polyfluoroalkyl substances (PFASs) (detailed information is provided in Table S1, Supporting Information, SI) were supplied by Evotec SE (Branford, CT) under contract to the U. S. Environmental Protection Agency (EPA) through a Material Transfer Agreement. The U. S. EPA’s PFAS screening library was established with a chemical category-based prioritization approach through which the chemicals were selected based on structural diversity, toxicity data availability and quantity, physico-chemical properties and chemical standards availability,42 providing a chance to evaluate structure-related binding affinities of PFASs to hL-FABP. A commercial AFFF technical product is representative AFFFs used in different Air Force Bases in USA, which was provided by Davis-Monthan Air Force base in Tucson (Arizona, USA) and manufactured by 3M as described in a previous study.43 AFFF-contaminated water samples were collected from the west end of Lake Niapenco, Hamilton, Canada (details of sample collection are provided in SI).41 Water samples were collected using polypropylene tubes which were then stored on ice during the transportation and extracted within 10 hours. Other reagents, solvents and protocols used are listed in SI.

Binding Affinities of 74 PFASs to hL-FABP.

The determination of the binding affinities of PFASs to hL-FABP was achieved by using a fluorescence displacement method. His-tagged hL-FABP was overexpressed in BL 21 (DE3) strain of E. coli. The binding of 74 PFASs to hL-FABP was determined by the displacement of bound 1-anilinonaphthalene- 8-sulfonic acid (1,8-ANS) from hL-FABP as previously described.44 1,8-ANS is nonfluorescent in aqueous buffer, but becomes strongly fluorescent once binding to a hydrophobic environment (i.e., proteins). Thus, the binding affinity of 74 PFASs to hL-FABP was measured by spiking each individual PFASs to the mixture of hL-FABP and 1,8-ANS. The decrease in fluorescence was expected as PFASs replaced 1,8-ANS from the hydrophobic protein binding pocket. More details about protein expression and binding affinity measurement are provided in SI.

SEC-Coelution (SECC) Method.

The SEC-coelution method was performed on an Agilent 1260 Infinity II HPLC (Agilent Technologies) with AdvanceBio SEC column (300 Å / 2.7 μm, Agilent Technologies). 0.5 μM of PFOS, PFOA and 3:1 FTOH were spiked with 1 μM of hL-FABP (n = 2) and incubated on ice for 60 min. 100 μL of the aliquots were loaded onto the SEC column. 10 mM of phosphate-buffered saline (PBS) was used as the mobile phase in the HPLC at a flow rate of 0.2 mL/min. The dual-wavelength signals were set to be 260 nm and 280 nm to monitor proteins. Five fractions were collected by Fraction Collector (Agilent 1260 Infinity II, Agilent Technologies) at a time interval of 2 min from 12 min to 22 min. For chemical analysis, 1.2 mL of methanol was added to fractions to denature proteins and extract PFASs. After centrifugation at 14,000 g for 10 min, the supernatants were collected for PFAS analysis. The pellets of the fractions after centrifugation were subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) or proteomics analysis to determine hL-FABP. For all subsequent experiments, hL-FABP was consistently eluted in Fraction 2 (F2) (14–16 min).

To apply the SECC method to identify hL-FABP ligands from 74 PFASs, a 3M AFFF technical product and AFFF-contaminated surface waters, the mixtures of chemicals were individually incubated with non-hL-FABP and hL-FABP overexpressing E. coli lysates. Specifically, 0.5 μM of 74 PFASs mixture, 25 mg L−1 of an AFFF technical product, and 1% of SPE extracts of AFFF-contaminated water in buffer A (50 mM tri-HCl, 150 mM NaCl, 1 mM dithiothreitol, pH 8.0) were used for the incubation (n = 2) individually. The F2 fractions containing hL-FABP from SEC were extracted with methanol as mentioned above. The chemicals co-eluted with hL-FABP in F2 were subjected to LC-Orbitrap for NTA to determine hL-FABP ligands. Details for the analysis of proteome and small molecules including NTA analysis, are provided in SI.

Data Analysis.

Statistical analyses were performed using GraphPad Prism (v7.0.4, GraphPad Software Inc, San Diego, CA, USA) or R studio (v1.1.456, RStudio, Inc., Boston, MA, USA). MaxQuant algorithms (http://maxquant.org/, version 1.6.2.3) were used to process the proteomic raw files. The FASTA file of the human and E. coli proteome was downloaded from UniProt database. One-way ANOVA test was used to detect the statistical significance of peak intensities or retention times between PFASs.

Results and Discussion

Interactions between PFASs and hL-FABP are Predictable by LC-MS Behaviors.

To understand the structure-related interactions between PFASs and hL-FABP, the binding affinities of 74 structurally diverse PFASs from U. S. Environmental Protection Agency (EPA) PFAS screening library42 to hL-FABP were tested through a fluorescence displacement method in which 1,8-ANS23,24 was used as a fluorescence probe. Initial screening concentration of PFASs was fixed at 300 μM. While 44 of the 74 (59.5%) PFASs showed no fluorescence displacement (defined as ‘inactive’), 9 PFASs (12.2%) resulted in >50% reduction in fluorescence intensity (defined as ‘high’ affinity ligands), and 21 (28.4%) PFASs which induced less (10–50%) fluorescence intensity reduction showed ‘weak’ binding affinities (Figure 1A). The 30 positive hits were selected for the next round of quantification of their dissociation constant (Kd) by obtaining dose-response curves. Consistent with initial screening, low μM Kd values (4.3–18.7 μM) were detected for the 9 high-affinity PFASs (Figure 1B). Among these PFASs, 7 are classic perfluoroalkyl acids (PFAAs) and their salts including PFOS (46), PFOA and PFNA (Figure 1C, Table S2), which have been well documented to bind strongly to L-FABP.22, 24, 25 An ether chemical 34 was also detected with a high binding affinity (Kd = 8.2 μM). This result is supported by a very recent study which determined that two other ether PFASs, 6:2 chlorinated polyfluorinated ether sulfonate (6:2 Cl-PFESA) and hexafluoropropylene oxide trimer acid (HFPO-TA), also have strong binding affinities to hL-FABP.45

Figure 1.

Figure 1.

Open in a new tab

Characterization of the binding affinities of 74 PFASs to hL-FABP. (A) The fluorescence intensity of 1,8-ANS (40 μM) in 1 μM hL-FABP, in the presence of individual PFASs at 300 μM. Inactive and active PFASs are labeled in black and red colors respectively. (B) The dose-dependent competition of fluorescence signals of 1,8-ANS (40 μM) in 1 μM hL-FABP by 4 representative PFASs (34, 46, 28 & 57). Fluorescence intensity was normalized to negative controls (1,8-ANS only). (C) Structures of 4 representative active PFASs (34, 46, 28 & 57, red), and 2 inactive PFASs (7 & 9, black). PFAS 46, 57 and 9 refer to PFOS, PFHxS and NEt-FOSE as described in Table S1.

In parallel, we analyzed the same 74 PFASs with LC-Orbitrap under electrospray ionization negative mode (ESI−), the most common ionization mode to analyze PFASs,6, 46, 47 and found that 39 of them could not be detected due to the lack of ionizable functional groups (representative structures are shown in Figure 2A). Comparing with the results of the fluorescence displacement assay, all of these 39 ‘MS-inactive’ PFASs also showed non-binding to hL-FABP. In contrast with these ‘MS-inactive’ PFASs, strong instrumental signals (indicated by high peak intensities) were observed for 20 hL-FABP ligands, especially for high-affinity ligands (Kd ~4.3–18.7 μM). The average peak intensities of high-affinity PFASs (n = 9) are 3.7×108 at 1 μM, significantly (p = 2.9×10−21) higher than those of low-affinity (intensity = 8.2×107, n = 11) and inactive (intensity = 1.9×105, n = 54) PFASs (Figure 2B). A significant relationship was also observed between hL-FABP binding affinities of PFASs and their retention times on a C18 column. The retention times of high-affinity PFASs (4.47 ± 0.30 min, n = 9), are significantly (p = 8.48×10−5) longer than those of weak (3.26 ± 0.95 min, n = 11) and inactive PFASs (2.75 ± 0.65 min, n = 15) (Figure 2C).

Figure 2.

Figure 2.

Open in a new tab

Binding affinities of 74 PFASs to hL-FABP are predictable by their LC-MS behaviors. (A) The peak intensities (red dots) and Kd (blue columns) values of 9 high-affinity (Kd ~4.3–18 μM), 11 low-affinity (Kd ~116–916 μM) and 54 inactive (Kd >1,000 μM) PFASs. Note that all 39 MS-inactive PFASs do not bind to hL-FABP. Structures of representative MS-inactive PFASs are shown in the box. (B) The peak intensities of high-affinity PFASs are significantly higher than those of weak and inactive PFASs. p-value was determined by one-way ANOVA. (C) The retention times of high-affinity PFASs on C18 columns are significantly longer than those of weak and inactive PFASs. 39 PFASs which were not detected by LC-MS are not shown in (C). p-value was determined by one-way ANOVA.

The association of hL-FABP binding affinities with LC-MS behaviors of PFASs is clearly illustrated by the representative structurally similar analogues in Figure 3. The instrumental responses were increased in the order NEt-FOSE (9) < NMe-FOSA (16) < PFOS (46) (Figure 3A), mainly due to the decrease in their pKa values. Such a trend is consistent with the stronger binding affinities of PFOS (Kd =8.1 μM) relative to NMe-FOSA (Kd ~1,000 μM) and NEt-FOSE (Kd >1,000 μM). Similarly, the increasing retention times of three analogues of ether chemicals 7, 28 and 34 are accordant with their increasing binding affinities to hL-FABP. Moreover, the crystal structure of hL-FABP complexed with palmitic acid shows that hL-FABP is capable of binding palmitic acids at two sites.48 The ionic and hydrogen bonds between the carboxylic acid group of palmitic acid and Arg-122, Ser-39 and Ser-124, together with the hydrophobic interactions between the alkyl chain and interior cavity, were revealed to play vital roles in stabilizing the complex at both binding sites. Although the crystal structure of PFASs binding to hL-FABP is not resolved, molecular docking simulations revealed that both polar and hydrophobic interactions play important roles.22, 25, 38

Figure 3.

Figure 3.

Open in a new tab

Proposed mechanisms to link LC-MS behaviors and hL-FABP binding affinities. (A) The binding affinities of three PFOS analogues are related to their ESI− responses. (B) The binding affinities of three ether analogues are related to their retention times on C18 columns. (C) The proposed physico-chemical mechanisms to link LC-MS behaviors and hL-FABP binding affinities. The polar interactions (red color) between the negatively charged acidic group and positively charged Arg-122 on hL-FABP mimic the process of ionization of PFASs under ESI−; the hydrophobic interactions (blue colors) between alkyl chains and Phe-50 and Ile-52 mimic the partition of PFASs to C18 column. PFAS 9, 16 and 46 refer to NEt-FOSE, NMe-FOSA and PFOS as described in Table S1.

Therefore, together with the experimental and in silico information, we proposed a mechanistic model to link LC-MS behaviors to hL-FABP binding affinities of PFASs (Figure 3C): the polar interaction between PFASs and hL-FABP relies on the negative charge of acidic head groups to maintain ionic and hydrogen bonds with Arg-122, Ser39 and Ser124,22 and the responses of PFASs under ESI− source are related to their pKa-dependent ionization efficiencies; moreover, the binding of the alkyl chain of PFAS to the hydrophobic cavity of hL-FABP (i.e., Phe-50, Ile-52, etc.), and the retention of PFASs on the C18 stationary phase are both determined by their hydrophobicity. To our best knowledge, this is the first study to connect LC-MS behaviors to bioactivities of PFASs. The discovery provides a straightforward strategy to prioritize the PFASs detected by NTA: only PFASs detected under ESI− and eluted after 3.0 min (in current LC conditions referred to SI) are possible hL-FABP ligands.

Note this is a descriptive model to initially narrow down priority PFASs. However, not all PFASs meeting the LC-MS criteria (RT >3.0 min under ESI−) are necessarily hL-FABP ligands. For instance, NMe-FOSA (chemical 16 in Figure 3A) was detected under ESI− with high intensity (3.43×108 at 0.5 μM), however, it is a weak hL-FABP ligand with Kd ~1,000 μM. Since the estimated pKa value of NMe-FOSA is 8.78,49 it is neutral at physiological pH (7.4) and is not expected to form a polar interaction with Arg-122, Ser39 and Ser124 on hL-FABP. Moreover, the instrumental responses of nitrogen-containing chemicals were reported to be insensitive to pKa, and all triazines with pKa <10 are efficiently ionized under ESI−.50 In addition, previous studies have reported the decrease in hL-FABP binding affinities for longer-chain PFCAs when their carbon-chain lengths are greater than 12.22 Therefore, the model may not be applied to the PFASs with extremely long carbon-chains since they cannot fit into the binding pocket of hL-FABP. This highlights the need to develop a selective method to further identify hL-FABP ligands which initially narrowed down by the LC-MS criteria.

Development of a SEC-Coelution Method to Screen Unknown hL-FABP Ligands.

We opted to develop a SEC coelution (SECC) method to identify hL-FABP ligands by incorporating SEC fractionation and nontarget analysis under ESI negative mode (Figure 4A). In brief, the hL-FABP proteins were incubated with PFASs and then loaded onto a SEC column for non-denaturing fractionations. The high-affinity PFASs binding to hL-FABP (eluted earlier) was separated from free unbound inactive PFASs (eluted later) according to their sizes. The fractions from the SEC were collected and analyzed by proteomics and nontarget analysis to determine the ligands co-eluted with hL-FABP. To reduce the impacts of nonspecific binding, we opted to use crude E. coli lysates rather than purified hL-FABP for the SECC, as nonspecific bound PFASs will be removed by higher abundant E. coli proteins.

Figure 4.

Figure 4.

Open in a new tab

Benchmarking the SECC method to identify PFASs binding to hL-FABP. (A) The workflow of the SECC method. hL-FABP ligands (bound) and inactive PFASs (free) were separated by SEC according to their sizes. Proteins and PFASs in each HPLC fraction were analyzed by proteomics and nontarget analysis to assign the interactions. (B) hL-FABP was eluted in F2 (14–16 min) from SEC as evidenced by SDS-PAGE. Red color arrow indicates the band of hL-FABP (~14 kDa). (C) The coelution of PFOA and PFOS with hL-FABP in F2. Note that 3:1 FTOH (inactive) was eluted in F5 (20–22 min) as a free chemical. (D) Benchmarking the SECC method with a mixture of 74 PFASs (0.5 μM for each chemical). Note that only 8 high-affinity PFASs were co-eluted with hL-FABP in F2, compared to non-hL-FABP lysates. (E) SECC method selectively captured the high-affinity PFASs. NEt-FOSA (blue dot) was not captured by the SECC method even though it was identified as a high-affinity hL-FABP ligand by the fluorescence displacement method. Only 6 individual high-affinity PFASs are shown (2 of the hL-FABP ligands are salts of PFOA and PFOS).

For the proof of concept, we used PFOS (Kd = 8.1 μM), PFOA (Kd = 18.8 μM) and 3:1 FTOH (Kd > 1,000 μM) with different binding affinities which were determined by the fluorescence displacement experiment, to verify the SECC method. 0.5 μM of 3 PFASs were individually incubated with 1 μM of E. coli lysates overexpressing hL-FABP, or non-hL-FABP overexpressing E. coli lysates, and then fractionated with SEC column. SDS-PAGE revealed that hL-FABP (~14 kDa) was eluted in Fraction 2 (F2, 14–16 min), while other background E. coli proteins were eluted in earlier fractions (F1, 12–14 min) due to their bigger sizes (Figure 4B). The presence of hL-FABP in F2 was confirmed with proteomics (Figure S5). One of the two hL-FABP ligands, PFOS was co-eluted with hL-FABP in F2 at 0.11 ± 0.028 μM (Figure 4C). The other ligand PFOA was also co-eluted in F2, albeit smaller amount (0.09 ± 0.025 μM), partially due to its weaker affinity relative to PFOS. In contrast, the inactive 3:1 FTOH was eluted later in F5 as a free chemical. For the non-hL-FABP lysates, all three PFASs were eluted in F5 as free chemicals. These results verified the selectivity of the SECC method to identify hL-FABP ligands.

To further benchmark the method to identify hL-FABP ligands from a mixture, we mixed all 74 PFASs at 0.5 μM, and then subjected the mixture for the ligand identification. The majority of weak or inactive PFASs were eluted in F4 or F5 as free chemicals which were observed in both hL-FABP overexpressing and non-hL-FABP lysates (Figure 4D). 8 PFASs were eluted in F2 when incubating with hL-FABP overexpressing lysates, but not in non-hL-FABP lysates, clearly suggesting the binding of these PFASs to hL-FABP. This result was consistent with the fluorescence displacement experiment through which all 8 of these PFASs were defined as strong hL-FABP ligands, such as PFOA (74), PFNA (17), and FOSA (60) (Table S1). The bound amounts of these high-affinity PFASs coeluted in F2 were 0.21–0.46 μM, while <0.01 μM for other low-affinity PFASs (Figure 4E). Interestingly, NEt-FOSA which was also determined as a strong hL-FABP ligand by the fluorescence displacement method, was coeluted in F2 with a minor amount (blue dot in Figure 4E). To verify the inconsistency between two methods, we used LC-MS to analyze the standard of NEt-FOSA, and PFOS was detected as a high percentage (30%) impurity which mainly explained the high binding affinity of NEt-FOSA to hL-FABP in the fluorescence displacement experiment (Figure S6). Therefore, the results indicated that NEt-FOSA is not a strong hL-FABP ligand and its exclusion from the screening assay was caused by the PFOS impurity in the sample. Together with the results, we conclude that SECC is a highly selective method to identify hL-FABP ligands from a mixture. Effect-Directed Analysis (EDA) is the predominant strategy to identify causative chemicals from the environmental mixture.51, 52 However, EDA is prone to false positive rates due to the co-elution of hundreds of chemicals in the same fraction. In contrast, the SECC method is very effective in identifying unknown ligands even if multiple ligands are present at the same time. We proposed that SECC might be a supplementary method to EDA for the identification of unknown ligands in the environment, and the EDA method could be further employed to quantify the contributions of individual ligands.

Sulfonic Acid Analogues were Identified as the Major hL-FABP Ligands in an AFFF Product.

We then applied the SECC method to identify PFASs binding to hL-FABP in a 3M AFFF technical product. 25 mg L−1 (~50 μM) of an AFFF product was incubated with 1 μM of E. coli lysates overexpressing hL-FABP, or non-hL-FABP lysates for the SECC method. Nontarget analysis was employed to determine the chemicals eluted in F2 from hL-FABP overexpressing and non-hL-FABP lysates. Only chemicals showing significantly higher abundances (10-fold, p < 0.05) in hL-FABP overexpressing lysates than non-hL-FABP lysates were considered as hL-FABP ligands. Among 4,056 features detected under ESI−, 33 chemicals were specifically co-eluted with hL-FABP in F2 (Figure 5A) after excluding isotopic peaks and adducts. Three major classes of homologues were clearly observed according to their stepwise increase in m/z and retention times (Figure 5B). Corresponding to the LC-MS mechanistic model proposed above, the retention times of these detected hL-FABP ligands fall into a narrow range at 3.07–4.66 min, while a wide range of retention times was detected for 544 other AFFF chemical features (Figure 5C). This confirms the above results that only PFASs with longer retention times (>3 min) exhibit sufficient binding affinities to hL-FABP; at the same time, large PFASs with extremely long retention times (>5 min) may also exhibit decreased binding affinities due to the limited size of the hL-FABP binding pocket.

Figure 5.

Figure 5.

Open in a new tab

Nontarget screening of PFASs binding to hL-FABP in AFFF technical product. (A) Volcano plot representing the log transformed ratios and corresponding p-values of features detected under ESI−. The ratio of peak abundances of PFASs detected in hL-FABP-overexpressing E. coli lysates to those in non-hL-FABP lysates were calculated as described in the method section in SI. The red dot indicates features significantly co-eluted with hL-FABP in F2. (B) The retention times and m/z values of hL-FABP ligands detected under ESI−. Note that all ligands are grouped into three classes. (C) Comparison of the m/z and retention times of hL-FABP ligands (red color) to all AFFF chemicals (grey color) detected under ESI−. (D) Structures of 6 novel hL-FABP ligands as PFOS analogues, at confidence level 3. Note that multiple isomers may exist but only 1 representative structure from PFAS library53 was shown for each chemical.

To get exact chemical identities for these detected hL-FABP ligands, we searched against the PFAS library established in Liu’s study,53 which summarized 1021 PFASs in more than 130 classes discovered in previous NTA studies (details of suspect screening are provided in SI). 11 PFASs were identified with this strategy, and all of them are analogues to perfluorosulfonic acids (PFSAs). PFOS and perfluorohexanesulfonate (PFHxS), two high-affinity hL-FABP ligands, were identified as the two most abundant ligands according to their peak intensities (Figure 5B). Several other classic odd-chained PFSAs, including perfluoropentanesulfonate (PFPeS), perfluoroheptanesulfonate (PFHpS) and perfluorononanesulfonate (PFNS) were detected with lower abundances, and the identities of PFPeS and PFHxS were confirmed by authentic standards (Figure S7). In addition to these legacy PFSAs, several other PFOS analogues (Figure 5D) including chlorinated (1’), unsaturated (2’), or hydrogenated PFOS (3’) were also reproducibly detected across replicates. Except for PFHxS and PFOS, 9 of the 11 PFSAs detected as hL-FABP ligands by the SECC method have never been reported before. Due to the lack of authentic standards, 7 of the detected ligands were assigned at confidence level 2 or 3 according to the Schymanski et al. scale54, by exact mass, MS2 spectra and isotopic peaks. Further studies are warranted to synthesize standards to determine their binding affinities.

Although PFSAs were identified, two other classes of homologues (Figure 5B) were not identified in the PFAS library. We hypothesized these chemicals might be nonfluorinated AFFF chemicals which comprised the major portion of AFFF technical products (PFASs only comprise 0.9–1.5% w/w).55 A very recent suspect screening study identified nine classes of hydrocarbon surfactants in AFFFs.55 Thus, we searched against the AFFF surfactant library (869 compounds)55 and found two of the hL-FABP ligands, identified as C8H17O4S and C10H21O4S alkyl sulfates. Interestingly, these two alkyl sulfates belonged to two unidentified classes of hL-FABP ligands. The chromatograms of the homologues of C8H17O4S were extracted (Figure 6A), and their formulas were predicted as C8H17(C2H4O)nO4S, the alkyl ether sulfates (C8/EOn). The prediction of structures was also supported by their high resolution MS2 spectra, in which HSO4− and SO3− fragments were observed as characteristic fragments of sulfate (Figures 6C and 6D). Homologues of C10H21O4S were detected at longer retention times than C8/EOn (Figure 6B), due to their longer hydrophobic alkyl chains. MS2 spectra (Figures 6E and 6F) further confirmed C10H21O4S homologues as C10/EOn. The identification of C8- and C10/EOn as the predominant hL-FABP ligands, was consistent with the structure-related activities of the 74 PFASs we established in the fluorescence displacement experiment, by which the ether analogue (34) of perfluorinated carboxylic acids was demonstrated to be a strong hL-FABP ligand.

Figure 6.

Figure 6.

Open in a new tab

Alkyl ether sulfates were identified as predominant classes of hL-FABP ligands in the AFFF technical product. (A) Chromatograms of six identified C8/EOn and their assigned formulas. (B) Chromatograms of six identified C10/EOn and their assigned formulas. (C) MS2 spectra and assigned fragments of C8/EOn 2. (D) MS2 spectra and assigned fragments of C8/EOn 3. (E) MS2 spectra and assigned fragments of C10/EOn 1. (F) MS2 spectra and assigned fragments of C8/EOn 2.

The Predominance of Sulfonic Acid Analogues in AFFF-Contaminated River Waters is Evidenced by SECC Method.

We then applied the SECC method to identify PFASs binding to hL-FABP in AFFF-contaminated surface waters, collected from Welland River next to the Hamilton International Airport.41 Consistent with previous studies,41, 56 high concentrations of PFOS (232 ng/L) were detected in water samples collected from Welland River. The extracts of river waters from C18 SPE cartridges were spiked with 1 μM of E. coli lysates overexpressing hL-FABP, and non-hL-FABP lysates for the SECC method. 207 chemicals were detected as putative ligands of hL-FABP as shown in Figure 7A. Only 10 of them were assigned by searching against AFFF, PFAS and surfactant libraries, including 2 PFASs (PFOS and PFHxS) and 8 alkyl ether sulfates. Since it has been reported that hL-FABP may bind to many other endogenous metabolites,57 this may explain the majority of hL-FABP ligands we detected in the natural waters. The detection of PFOS and PFHxS as the only fluorinated ligands corresponded to the results of the AFFF technical product in which PFOS and PFHxS are the two most abundant fluorinated ligands. Other PFASs such as PFOA, were not detected in the contaminated water which may be due to their 10–100 times lower concentrations than PFOS.41, 56

Figure 7.

Figure 7.

Open in a new tab

Nontarget screening of hL-FABP ligands in AFFF-contaminated river waters. (A) Volcano plot representing the log transformed ratios and corresponding p-values of features detected. The red dot indicates features significantly co-eluted with hL-FABP in F2. The blue dots indicate PFASs or AFFF surfactants detected. (B) Structures of 4 detected C15- and C14-ether sulfates. (C) Chromatograms of PFOS and 4 alkyl ether sulfates. Red and black colors indicate the chromatograms from two replicates of hL-FABP overexpressing lysates. Blue and grey colors indicate the chromatograms from two replicates of non-hL-FABP lysates.

Interestingly, the alkyl ether sulfates were also detected by the SECC method as the ligands of hL-FABP in AFFF-contaminated waters. This was consistent with the results of the AFFF technical product. However, C8- and C10/EOn were detected as the only ligands in the AFFF product, whereas C14- and C15/EOn were detected as the major ligands in river waters (Figures 7B and 7C). This could be due to the components of the AFFF used at the Hamilton Airport differing from the 3M AFFF technical product we used in the current study. Differences in the components of AFFF products has been well documented in previous studies.12, 21, 55 Alternatively, alkyl ether sulfates are increasingly used in North America as surfactants.58 The identification of different alkyl ether sulfates in the river water from the AFFF product may indicate AFFF-independent sources of these compounds. The identification of alkyl ether sulfates as hL-FABP ligands is significant since a large amount of these surfactants are present in the AFFF technical product, and they were reported to occur in groundwater as well.55

Implications.

Current studies on nontarget analysis and toxicities of PFASs are disconnected, due to the challenges posed by the large numbers (>1,000) and diverse structures of PFASs. The SECC method provides a high-throughput experimental way to tackle the challenge of prioritizing PFASs according to key proteins, especially when their authentic standards are not available. While this study is focused on hL-FABP due to its critical role in regulating the toxicokinetics of PFASs, the protein-centric method could also be adapted to screen PFASs binding to other key proteins, such as PPARs. Computational toxicology is the predominant strategy for high-throughput predictions of toxicities of chemical contaminants. However, these models are prone to uncertainties, especially when the training dataset is inappropriate. The SECC method provides an opportunity to identify ligands even if prior information on binding schemes is limited (e.g., alkyl ether sulfates). We believe the SECC method would be valuable, in combination with computational toxicology, to uncover previously unknown PFASs binding to key proteins. However, one of the limitations of the current study is that only a single AFFF technical product and AFFF-contaminated waters were applied. Since previous studies have well documented the variation in chemical components across AFFF products,12 the application of the SECC method to more AFFF-contaminated environmental samples is of ongoing interest.

The present study revealed the existence of many previously unknown, AFFF-derived hL-FABP ligands. Unexpectedly, nonfluorinated AFFF chemicals were identified as an even bigger class of hL-FABP ligands. Indeed, these nonfluorinated ionic surfactants were proposed as environmentally safe chemicals to replace PFASs or organic solvents, but limited data is available regarding their potential toxicities.58–60 Here, we highlighted hL-FABP as a potential critical target for these nonfluorinated ionic surfactants.

Supplementary Material

Supplement1

Acknowledgements.

This research was supported by the National Sciences and Engineering Research Council (NSERC) Discovery Grant, and Environment and Climate Change Canada. The authors acknowledge the support of instrumentation grants from the Canada Foundation for Innovation, the Ontario Research Fund, and the NSERC Research Tools and Instrument Grant.

Footnotes

Disclaimer: The views expressed in this paper are those of the authors and do not necessarily reflect the statements, opinions, views, conclusions, or policies of the U.S. Environmental Protection Agency.

Supporting Information Available

The information is available freely via the Internet at http://pubs.acs.org/.

References

  • 1.Giesy JP; Kannan K. Perfluorochemical surfactants in the environment. Environ. Sci. Technol 2002, 36 (7), 146A-152A. [DOI] [PubMed] [Google Scholar]
  • 2.Lau C; Anitole K; Hodes C; Lai D; Pfahles-Hutchens A; Seed J. Perfluoroalkyl acids: A review of monitoring and toxicological findings. Toxicol. Sci 2007, 99 (2), 366–394. [DOI] [PubMed] [Google Scholar]
  • 3.Houde M; De Silva AO; Muir DCG; Letcher RJ Monitoring of perfluorinated compounds in aquatic biota: an updated review PFCs in aquatic biota. Environ. Sci. Technol 2011, 45 (19), 7962–7973. [DOI] [PubMed] [Google Scholar]
  • 4.Wilkinson J; Hooda PS; Barker J; Barton S; Swinden J. Occurrence, fate and transformation of emerging contaminants in water: An overarching review of the field. Environ. Pollut 2017, 231, 954–970. [DOI] [PubMed] [Google Scholar]
  • 5.Zareitalabad P; Siemens J; Hamer M; Amelung W. Perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS) in surface waters, sediments, soils and wastewater - A review on concentrations and distribution coefficients. Chemosphere 2013, 91 (6), 725–732. [DOI] [PubMed] [Google Scholar]
  • 6.Peng H; Wei QW; Wan Y; Giesy JP; Li LX; Hu JY Tissue Distribution and Maternal Transfer of Poly- and Perfluorinated Compounds in Chinese Sturgeon (Acipenser sinensis): Implications for Reproductive Risk. Environ. Sci. Technol 2010, 44 (5), 1868–1874. [DOI] [PubMed] [Google Scholar]
  • 7.EPA. Risk Management for Per- and Polyfluoroalkyl Substances (PFAS) under TSCA; https://www.epa.gov/assessing-and-managing-chemicals-under-tsca/risk-management-and-polyfluoroalkyl-substances-pfas.
  • 8.Organisation for Economic Co-operation and Development Toward a New Comprehensive Global Database of Per- and Polyfluoroalkyl Substances (PFASs): Summary Report on Updating the OECD 2007 List of Per- and Polyalkyl Substances (PFASs). Series on Risk Management No. 39. 2018. [Google Scholar]
  • 9.Wang Y; Yu NY; Zhu XB; Guo HW; Jiang JG; Wang XB; Shi W; Wu JC; Yu HX; Wei S. Suspect and Nontarget Screening of Per- and Polyfluoroalkyl Substances in Wastewater from a Fluorochemical Manufacturing Park. Environ. Sci. Technol 2018, 52 (19), 11007–11016. [DOI] [PubMed] [Google Scholar]
  • 10.Yu NY; Guo HW; Yang JP; Jin L; Wang XB; Shi W; Zhang XW; Yu HX; Wei S. Non-Target and Suspect Screening of Per- and Polyfluoroalkyl Substances in Airborne Particulate Matter in China. Environ. Sci. Technol 2018, 52 (15), 8205–8214. [DOI] [PubMed] [Google Scholar]
  • 11.Liu YN; Pereira AD; Martin JW Discovery of C-5-C-17 Poly- and Perfluoroalkyl Substances in Water by In-Line SPE-HPLC-Orbitrap with In-Source Fragmentation Flagging. Anal. Chem 2015, 87 (8), 4260–4268. [DOI] [PubMed] [Google Scholar]
  • 12.Barzen-Hanson KA; Roberts SC; Choyke S; Oetjen K; McAlees A; Riddell N; McCrindle R; Ferguson PL; Higgins CP; Field JA Discovery of 40 classes of per- and polyfluoroalkyl substances in historical aqueous film-forming foams (AFFFs) and AFFF-impacted groundwater. Environ. Sci. Technol 2017, 51 (4), 2047–2057. [DOI] [PubMed] [Google Scholar]
  • 13.Place BJ; Field JA Identification of Novel Fluorochemicals in Aqueous Film-Forming Foams Used by the US Military. Environ. Sci. Technol 2012, 46 (13), 7120–7127. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.D’Agostino LA; Mabury SA Identification of Novel Fluorinated Surfactants in Aqueous Film Forming Foams and Commercial Surfactant Concentrates. Environ. Sci. Technol 2014, 48 (1), 121–129. [DOI] [PubMed] [Google Scholar]
  • 15.Newton S; McMahen R; Stoeckel JA; Chislock M; Lindstrom A; Strynar M. Novel Polyfluorinated Compounds Identified Using High Resolution Mass Spectrometry Downstream of Manufacturing Facilities near Decatur, Alabama. Environ. Sci. Technol 2017, 51 (3), 1544–1552. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Gebbink WA; van Asseldonk L; van Leeuwen SPJ Presence of Emerging Per- and Polyfluoroalkyl Substances (PFASs) in River and Drinking Water near a Fluorochemical Production Plant in the Netherlands. Environ. Sci. Technol 2017, 51 (19), 11057–11065. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Lin YF; Ruan T; Liu AF; Jiang GB Identification of Novel Hydrogen-Substituted Polyfluoroalkyl Ether Sulfonates in Environmental Matrices near Metal-Plating Facilities. Environ. Sci. Technol 2017, 51 (20), 11588–11596. [DOI] [PubMed] [Google Scholar]
  • 18.Liu YN; Qian ML; Ma XX; Zhu LY; Martin JW Nontarget Mass Spectrometry Reveals New Perfluoroalkyl Substances in Fish from the Yangtze River and Tangxun Lake, China. Environ. Sci. Technol 2018, 52 (10), 5830–5840. [DOI] [PubMed] [Google Scholar]
  • 19.Liu YN; Richardson ES; Derocher AE; Lunn NJ; Lehmler HJ; Li XS; Zhang YF; Cui JY; Cheng LH; Martin JW Hundreds of Unrecognized Halogenated Contaminants Discovered in Polar Bear Serum. Angew. Chem. Int. Edit 2018, 57 (50), 16401–16406. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Rotander A; Karrman A; Toms LML; Kay M; Mueller JF; Ramos MJG Novel Fluorinated Surfactants Tentatively Identified in Firefighters Using Liquid Chromatography Quadrupole Time-of-Flight Tandem Mass Spectrometry and a Case-Control Approach. Environ. Sci. Technol 2015, 49 (4), 2434–2442. [DOI] [PubMed] [Google Scholar]
  • 21.Yeung LWY; Mabury SA Bioconcentration of Aqueous Film-Forming Foam (AFFF) in Juvenile Rainbow Trout (Oncorhyncus mykiss). Environ. Sci. Technol 2013, 47 (21), 12505–12513. [DOI] [PubMed] [Google Scholar]
  • 22.Zhang LY; Ren XM; Guo LH Structure-based investigation on the interaction of perfluorinated compounds with human liver fatty acid binding protein. Environ. Sci. Technol 2013, 47 (19), 11293–11301. [DOI] [PubMed] [Google Scholar]
  • 23.Luebker DJ; Hansen KJ; Bass NM; Butenhoff JL; Seacat AM Interactions of flurochemicals with rat liver fatty acid-binding protein. Toxicology 2002, 176 (3), 175–185. [DOI] [PubMed] [Google Scholar]
  • 24.Woodcroft MW; Ellis DA; Rafferty SP; Burns DC; March RE; Stock NL; Trumpour KS; Yee J; Munro K. Experimental characterization of the mechanism of perfluorocarboxylic acids’ liver protein bioaccumulation: the key role of the neutral species. Environ. Toxicol. Chem 2010, 29 (8), 1669–1677. [DOI] [PubMed] [Google Scholar]
  • 25.Sheng N; Li J; Liu H; Zhang AQ; Dai JY Interaction of perfluoroalkyl acids with human liver fatty acid-binding protein. Arch. Toxicol 2016, 90 (1), 217–227. [DOI] [PubMed] [Google Scholar]
  • 26.Gao K; Zhuang TF; Liu X; Fu JJ; Zhang JX; Fu J; Wang LG; Zhang AQ; Liang Y; Song MY; Jiang GB Prenatal Exposure to Per- and Polyfluoroalkyl Substances (PFASs) and Association between the Placental Transfer Efficiencies and Dissociation Constant of Serum Proteins-PFAS Complexes. Environ. Sci. Technol 2019, 53 (11), 6529–6538. [DOI] [PubMed] [Google Scholar]
  • 27.Cao HM; Zhou Z; Wang L; Liu GL; Sun YZ; Wang YW; Wang T; Liang Y. Screening of potential PFOS alternatives to decrease liver bioaccumulation: experimental and computational approaches. Environ. Sci. Technol 2019, 53 (5), 2811–2819. [DOI] [PubMed] [Google Scholar]
  • 28.Wen W; Xia XH; Hu DX; Zhou D; Wang HT; Zhai YW; Lin H. 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. [DOI] [PubMed] [Google Scholar]
  • 29.Ng CA; Hungerbuhler K. Bioconcentration of perfluorinated alkyl acids: how important is specific binding? Environ. Sci. Technol 2013, 47 (13), 7214–7223. [DOI] [PubMed] [Google Scholar]
  • 30.Cheng WX; Ng CA A permeability-limited physiologically based pharmacokinetic (PBPK) model for perfluorooctanoic acid (PFOA) in male rats. Environ. Sci. Technol 2017, 51 (17), 9930–9939. [DOI] [PubMed] [Google Scholar]
  • 31.Wolfrum C; Borrmann CM; Borchers T; Spener F. Fatty acids and hypolipidemic drugs regulate peroxisome proliferator-activated receptors alpha- and gamma-mediated gene expression via liver fatty acid binding protein: A signaling path to the nucleus. Proc. Natl. Acad. Sci. U. S. A. 2001, 98 (5), 2323–2328. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Velkov T. Interactions between Human Liver Fatty Acid Binding Protein and Peroxisome Proliferator Activated Receptor Selective Drugs. PPAR Res. 2013, DOI: 10.1155/2013/938401. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Zhang LY; Ren XM; Wan B; Guo LH Structure-dependent binding and activation of perfluorinated compounds on human peroxisome proliferator-activated receptor gamma. Toxicol. Appl. Pharmacol 2014, 279 (3), 275–283. [DOI] [PubMed] [Google Scholar]
  • 34.Ishibashi H; Kim EY; Iwata H. Transactivation Potencies of the Baikal Seal (Pusa sibirica) Peroxisome Proliferator-Activated Receptor alpha by Perfluoroalkyl Carboxylates and Sulfonates: Estimation of PFOA Induction Equivalency Factors. Environ. Sci. Technol 2011, 45 (7), 3123–3130. [DOI] [PubMed] [Google Scholar]
  • 35.Ishibashi H; Hirano M; Kim EY; Iwata H. In vitro and in silico evaluations of binding affinities of perfluoroalkyl substances to Baikal seal and human peroxisome proliferator-activated receptor alpha. Environ. Sci. Technol 2019, 53 (4), 2181–2188. [DOI] [PubMed] [Google Scholar]
  • 36.Takacs ML; Abbott BD Activation of mouse and human peroxisome proliferator-activated receptors (alpha, beta/delta, gamma) by perfluorooctanoic acid and perfluorooctane sulfonate. Toxicol. Sci 2007, 95 (1), 108–117. [DOI] [PubMed] [Google Scholar]
  • 37.Zhang W; Zhang YT; Zhang HX; Wang JS; Cui RN; Dai JY Sex differences in transcriptional expression of FABPs in zebrafish liver after chronic perfluorononanoic acid exposure. Environ. Sci. Technol 2012, 46 (9), 5175–5182. [DOI] [PubMed] [Google Scholar]
  • 38.Cheng WX; Ng CA Predicting Relative Protein Affinity of Novel Per- and Polyfluoroalkyl Substances (PFASs) by An Efficient Molecular Dynamics Approach. Environ. Sci. Technol 2018, 52 (14), 7972–7980. [DOI] [PubMed] [Google Scholar]
  • 39.Cao HM; Wang L; Liang Y; Li ZJ; Feng HR; Sun YZ; Zhang AQ; Fu JJ Protonation state effects of estrogen receptor alpha on the recognition mechanisms by perfluorooctanoic acid and perfluorooctane sulfonate: A computational study. Ecotox. Environ. Safe 2019, 171, 647–656. [DOI] [PubMed] [Google Scholar]
  • 40.Ng CA; Hungerbuehler K. Exploring the use of molecular docking to identify bioaccumulative perfluorinated alkyl acids (PFAAs). Environ. Sci. Technol 2015, 49 (20), 12306–12314. [DOI] [PubMed] [Google Scholar]
  • 41.de Solla SR; De Silva AO; Letcher RJ Highly elevated levels of perfluorooctane sulfonate and other perfluorinated acids found in biota and surface water downstream of an international airport, Hamilton, Ontario, Canada. Environ. Int 2012, 39 (1), 19–26. [DOI] [PubMed] [Google Scholar]
  • 42.Patlewicz G; Richard AM; Williams AJ; Grulke CM; Sams R; Lambert J; Noyes PD; DeVito MJ; Hines RN; Strynar M; Guiseppi-Elie A; Thomas RS A chemical category-based prioritization approach for selecting 75 per- and polyfluoroalkyl substances (PFAS) for tiered toxicity and toxicokinetic testing. Environ. Health Perspect 2019, 127 (1), DOI: 10.1289/EHP4555. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Rodriguez-Freire L; Abad-Fernandez N; Sierra-Alvarez R; Hoppe-Jones C; Peng H; Giesy JP; Snyder S; Keswani M. Sonochemical degradation of perfluorinated chemicals in aqueous film-forming foams. J. Hazard. Mater 2016, 317, 275–283. [DOI] [PubMed] [Google Scholar]
  • 44.Velkov T; Chuang S; Wielens J; Sakellaris H; Charman WN; Porter CJH; Scanlon MJ The interaction of lipophilic drugs with intestinal fatty acid-binding protein. J. Biol. Chem 2005, 280 (18), 17769–17776. [DOI] [PubMed] [Google Scholar]
  • 45.Sheng N; Cui RN; Wang JH; Guo Y; Wang JS; Dai JY Cytotoxicity of novel fluorinated alternatives to long-chain perfluoroalkyl substances to human liver cell line and their binding capacity to human liver fatty acid binding protein. Arch. Toxicol 2018, 92 (1), 359–369. [DOI] [PubMed] [Google Scholar]
  • 46.Peng H; Zhang SY; Sun JX; Zhang Z; Giesy JP; Hu JY Isomer-specific accumulation of perfluorooctanesulfonate from (N-ethyl perfluorooctanesulfonamido)ethanol-based phosphate diester in Japanese Medaka (Oryzias latipes). Environ. Sci. Technol 2014, 48 (2), 1058–1066. [DOI] [PubMed] [Google Scholar]
  • 47.Schultz MM; Barofsky DF; Field JA Quantitative determination of fluorinated alkyl substances by large-volume-injection liquid chromatography tandem mass spectrometry - Characterization of municipal wastewaters. Environ. Sci. Technol 2006, 40 (1), 289–295. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Sharma A. Fatty acid induced remodeling within the human liver fatty acid-binding protein. J. Biol. Chem 2011, 286 (36), 31924–31928. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Rayne S; Forest K. A new class of perfluorinated acid contaminants: Primary and secondary substituted perfluoroalkyl sulfonamides are acidic at environmentally and toxicologically relevant pH values. J. Environ. Sci. Health Part A-Toxic/Hazard. Subst. Environ. Eng 2009, 44 (13), 1388–1399. [DOI] [PubMed] [Google Scholar]
  • 50.Henriksen T; Juhler RK; Svensmark B; Cech NB The relative influences of acidity and polarity on responsiveness of small organic molecules to analysis with negative ion electrospray ionization mass spectrometry (ESI-MS). J. Am. Soc. Mass Spectrom 2005, 16 (4), 446–455. [DOI] [PubMed] [Google Scholar]
  • 51.Weiss JM; Simon E; Stroomberg GJ; de Boer R; de Boer J; van der Linden SC; Leonards PEG; Lamoree MH Identification strategy for unknown pollutants using high-resolution mass spectrometry: Androgen-disrupting compounds identified through effect-directed analysis. Anal. Bioanal. Chem 2011, 400 (9), 3141–3149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Sun JX; Peng H; Alharbi HA; Jones PD; Giesy JP; Wiseman SB Identification of Chemicals that Cause Oxidative Stress in Oil Sands Process-Affected Water. Environ. Sci. Technol 2017, 51 (15), 8773–8781. [DOI] [PubMed] [Google Scholar]
  • 53.Liu YN; Dagostino LA; Qu GB; Jiang GB; Martin JW High-resolution mass spectrometry (HRMS) methods for nontarget discovery and characterization of poly- and per-fluoroalkyl substances (PFASs) in environmental and human samples. Trend. Anal. Chem 2019, 121, 115420. [Google Scholar]
  • 54.Schymanski EL; Jeon J; Gulde R; Fenner K; Ruff M; Singer HP; Hollender J. Identifying small molecules via high resolution mass spectrometry: communicating confidence. Environ. Sci. Technol 2014, 48 (4), 2097–2098. [DOI] [PubMed] [Google Scholar]
  • 55.Garcia RA; Chiaia-Hernandez AC; Lara-Martin PA; Loos M; Hollender J; Oetjen K; Higgins CP; Field JA Suspect Screening of Hydrocarbon Surfactants in AFFFs and AFFF-Contaminated Groundwater by High-Resolution Mass Spectrometry. Environ. Sci. Technol 2019, 53 (14), 8068–8077. [DOI] [PubMed] [Google Scholar]
  • 56.Gewurtz SB; Bhavsar SP; Petro S; Mahon CG; Zhao XM; Morse D; Reiner EJ; Tittlemier SA; Braekevelt E; Drouillard K. High levels of perfluoroalkyl acids in sport fish species downstream of a firefighting training facility at Hamilton International Airport, Ontario, Canada. Environ. Int 2014, 67, 1–11. [DOI] [PubMed] [Google Scholar]
  • 57.Furuhashi M; Hotamisligil GS Fatty acid-binding proteins: role in metabolic diseases and potential as drug targets. Nat. Rev. Drug Discov 2008, 7 (6), 489–503. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Jackson M; Eadsforth C; Schowanek D; Delfosse T; Riddle A; Budgen N. Comprehensive Review of Several Surfactants in Marine Environments: Fate and Ecotoxicity. Environ. Toxicol. Chem 2016, 35 (5), 1077–1086. [DOI] [PubMed] [Google Scholar]
  • 59.Oskarsson A; Wright MC Ionic Liquids: New Emerging Pollutants, Similarities with Perfluorinated Alkyl Substances (PFASs). Environ. Sci. Technol 2019, 53 (18), 10539–10541. [DOI] [PubMed] [Google Scholar]
  • 60.Wang P. Application of green surfactants developing environment friendly foam extinguishing agent. Fire Technol. 2015, 51 (3), 503–511. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement1
NIHMS1618951-supplement-Supplement1.pdf (1.1MB, pdf)

RESOURCES