Sage Journals: Discover world-class research

Abstract

Per- and polyfluorinated alkyl substances (PFAS) are ubiquitous, persistent, and toxic chemicals that pose public health risks. Recent carcinogenicity concerns have arisen based on epidemiological studies, animal tumor findings, and mechanistic data. Thousands of PFAS exist; however, current understanding of their toxicity is informed by studies of a select few, namely, perfluorooctanoic acid and perfluorooctanesulfonic acid. Hence, the computational, high-throughput screening tool, the US EPA CompTox Chemical Dashboard’s ToxCast, was utilized to explore the carcinogenicity potential of PFAS. Twenty-three major PFAS that had sufficient in vitro ToxCast data and covered a range of structural subclasses were analyzed with the visual analytics software ToxPi, yielding a qualitative and quantitative assessment of PFAS activity in realms closely linked with carcinogenicity. A comprehensive literature search was also conducted to check the consistency of analyses with other mechanistic data streams. The PFAS were found to induce a vast range of biological perturbations, in line with several of the International Agency for Research on Cancer–defined key carcinogen characteristics. Patterns observed varied by length of fluorine-bonded chains and/or functional group within and between each key characteristic, suggesting some structure-based variability in activity. In general, the major conclusions drawn from the analysis, that is, the most notable activities being modulation of receptor-mediated effects and induction of oxidative stress, were supported by literature findings. The study helps enhance understanding of the mechanistic pathways that underlie the potential carcinogenicity of various PFAS and hence could assist in hazard identification and risk assessment for this emerging and relevant class of environmental toxicants.

Keywords

per- and polyfluorinated alkyl substances carcinogens ToxCast high-throughput in vitro screening

Introduction

Per- and polyfluorinated alkyl substances (PFAS) are a series of synthetic compounds, utilized for several commercial and industrial purposes. They are found in a variety of consumer products, ranging from food packaging materials (e.g., microwave popcorn bags) and nonstick cookware to carpets and furniture to cosmetics and personal care products.¹ Per- and polyfluorinated alkyl substances are derived by 2 major processes: (i) electrochemical fluorination and (2) telomerization.² Typically, they consist of a partially or fully fluorinated carbon skeleton and a hydrophilic head group. Perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS) are the 2 most common and well-studied PFAS. However, this class of compounds includes a huge number of structural variants, based on the presence of different functional groups and varying chain lengths. The strength of the C–F bond,³ resulting from its partial ionic character, lends extraordinary stability to these compounds and, hence, they are resistant to the major environmental, chemical, and biological degradation pathways.⁴

Per- and polyfluorinated alkyl substances are distributed globally and have been detected in every trophic level and environmental phase.⁵ Additionally, human exposure can occur through multiple routes, including ingestion via drinking water or food, inhalation, and contact with residual levels in packaging materials.⁶ Furthermore, several health concerns regarding these compounds have arisen.⁷ The PFAS tend to bioaccumulate in humans and wildlife due to their metabolic stability and long half-lives.⁸ Their unique hydrophobic and lipophobic properties⁹ lead them to bind to plasma proteins and concentrate in organs (e.g., liver, kidney), and levels have been detected in human serum and tissue samples from around the world.¹⁰ For instance, a 2018 regional exposure study in women (n = 425) from the Los Angeles county indicated that the mean serum concentrations of PFAS range from 0.0681 ng/mL for lesser distributed PFAS to as much as 2.13 ng/mL for PFOS (https://biomonitoring.ca.gov). Several of the more well studied PFAS have been found to be toxic. Exposure to PFAS has been associated with developmental and reproductive toxicity, immunological effects, metabolic disorders, and organ-specific (e.g., hepatic, renal) toxicity.^7,11

One major health concern regarding PFAS is their potential to act as carcinogens.¹² Epidemiological links between exposure to PFAS and incidences of cancer,¹³ mechanistic studies in line with the International Agency for Research on Cancer (IARC) key carcinogen characteristics, and in vivo tumor data¹⁴ suggest that carcinogenic activity is associated with exposure to PFAS.

One limitation that scientists including toxicologists face today is the lack of data available to evaluate the health concerns posed by a vast number of environmental chemicals. To mitigate this issue, the US Environmental Protection Agency (EPA) has developed a tool known as Computational Toxicology (CompTox) Chemicals Dashboard (https://comptox.epa.gov/dashboard). It integrates information for thousands of chemicals into a comprehensive database and, utilizing high-throughput screening models such as Toxicity Forecaster (ToxCast), generates in vitro toxicity data for chemicals by measuring the perturbations they induce in biological systems. This allows quick and efficient assessments concerning the toxicity of chemical(s) of interest and reduces the need for expensive and time-consuming animal testing.

Several of the ToxCast in vitro bioassay end points can be mapped directly to one of the 10 IARC key characteristics of carcinogens and thus data can be applied toward gauging toxicity in areas related to carcinogenicity.^15,16 The wide range of health effects currently associated with PFAS exposure suggests they might possess multiple IARC-defined carcinogen characteristics. Moreover, structural variability such as different number of C–F bonds in the molecular backbone and/or the presence of unique functional groups could further influence the possible carcinogenic action of PFAS. Hence, the purpose of this study is to utilize data from a high-throughput toxicological tool to examine the potential biochemical modes by which a highly persistent, diverse, and widespread class of environmental chemicals could behave as carcinogens. The results were compared with literature-based findings to validate the conclusions.

Methods

Chemicals Dashboard and PFAS Selection

The PFAS of interest were identified from EPA’s CompTox Chemicals Dashboard (https://comptox.epa.gov/dashboard/chemical_lists). Utilizing EPA’s master list of potential PFAS, thousands of chemicals were initially examined (as of May 25, 2020), of which a set of 23 PFAS was selected for analysis (Figure 1). The chemicals were selected based on 3 main factors: close structural relation to the most relevant PFAS (i.e., PFOA and PFOS), distinct functional groups to cover a range of structural subclasses, and availability of sufficient high-throughput toxicity data. Alongside PFOA, the carboxylic acid–based PFAS series was selected and included both short- (6-7 carbons) and long-chain (9-11 carbons) variants as well as both free acid and salt forms. Together with PFOS, other sulfonic acids (short chains and salts) and sulfonic derivatives (ie, sulfonamides) were also chosen. Additionally, a variety of alternative functional group (eg, alcohols, methacrylate) based PFAS were selected to study moiety-specific differences. Finally, sufficient toxicity data necessary for inclusion were defined as an active or inactive hit in at least 1 ToxCast assay of interest. Hence, chemicals that were not tested in any of the ToxCast assays were excluded from analysis.

Figure 1.

Chemical structures of the 23 per- and polyfluorinated alkyl substances selected from CompTox chemicals dashboard.

ToxCast Assays

Data from EPA’s ToxCast, a high-throughput screening program, were utilized to assess activity of the selected PFAS in areas related to carcinogenicity. Each chemical of interest had toxicity data for a series of in vitro assays that measured changes in either (1) cellular responses or (2) biochemical pathways, upon exposure to the chemical. The toxicity readout per assay indicated 1 of 3 possible outcomes: active, inactive, or not tested yet. Each active hit corresponded to a particular half-maximal activity concentration (AC₅₀) to indicate potency with respect to that end point. The detailed experimental procedures employed to acquire the in vitro ToxCast data are readily available online (https://comptox.epa.gov/dashboard/assay_endpoints/), as well as previously documented.^17-20 The ToxCast assays of interest were ones that could be mapped to 1 of 10 IARC described key characteristics of carcinogens.¹⁵ They were primarily selected using a previously described 2018 study (Chiu et al) which matched the descriptions and end points of assays to the key carcinogen characteristics based on identified mechanistic similarities.¹⁶ The exceptions include inclusion of a new group of nuclear receptor targeting assays (mapped to key characteristic #8) and exclusion of key characteristic #1 mapped assays since this class of chemicals is not metabolically activated. Names and end point details of the 242 ToxCast assays are described in Tables S1-6 (in Supplementary material).

Toxicological Prioritization Index

Toxicological Prioritization Index (ToxPi, http://toxpi.org) is a visual analytics tool that integrates bioassay data from various sources and outputs the results in a graphical manner (i.e., a pie chart), allowing more efficient interpretation of findings.²¹ Additionally, it provides a quantitative score that assists in ranking and prioritization of chemicals. A numerical chemical-assay matrix was generated, which accounted for active (AC₅₀), inactive (1000000), or untested (NaN) assay results and inputted into ToxPi for holistic visualization of activity. For a chemical, each pie slice represents a key characteristic of carcinogens, with the slice width and length referring to the number of assays tested and ratio of positive to total hits, respectively. To calculate scores (y), a negative log scale (i.e., y = −log₁₀(x) + 6) was applied to each slice, as previously described.²² The cumulative score is an aggregation of slice-wise scores and is proportional to the overall predicted activity in areas related to carcinogenicity.

Results

ToxPi Output

Figure 2 depicts the ToxPi graphical output, overall scores, and legend following integration and analysis of ToxCast data of the 23 selected PFAS. Chemicals that displayed the greatest overall activity in assays linked to carcinogenicity were perfluorooctanesulfonamide (PFOSA), PFOS, N-ethyl PFOSA, and the potassium (K) salt of PFOS (scores = 0.7032-0.8235). Perfluorononanoic acid (PFNA), perfluorodecanoic acid (PFDA), and perfluoroundecanoic acid (PFUnA), carboxylic acids with a fluorine-bonded carbon skeleton longer than PFOA, were more active than PFOA. However, there was no significant difference (scores = 0.4975-0.5353) between the overall activity of the 3 chemicals. Nevertheless, PFOA (score = 0.3502) was more active than perfluoroheptanoic acid (PFHpA; score = 0.2438), which was far more active than perfluorohexanoic acid (PFHxA; score = 0.0428). Similarly, the 8:2 fluorotelomer alcohol (8:2 FtOH; score = 0.4320) ranked higher than the 6:2 fluorotelomer alcohol (6:2 FtOH (score = 0.2121) as did K-PFOS (score = 0.7032) compared to K-perfluorohexanesulfonic acid (K-PFHxS; score = 0.2299). Miscellaneous functional group (eg, methacrylate, urea, amide) based PFAS tended to be generally less active (scores <0.2100), though they had considerably more untested assays. Thus, for certain chemicals, particularly those with aromatic rings or an alkyl iodide, the smaller scores might be more indicative of lesser data availability than lower toxicity and so should be interpreted carefully.

Figure 2.

ToxPi output: Plots and scores after integration of ToxCast data for the 23 per- and polyfluorinated alkyl substances. For a chemical, each pie slice represents a key characteristic (as per the legend), with the slice width and length referring to the number of assays tested and ratio of positive to total hits, respectively. A y = −log₁₀(x) + 6 scale was applied to calculate scores (y) for each slice, and the displayed, cumulative score is an aggregation of slice-wise scores and is proportional to the overall predicted activity in areas related to carcinogenicity.

Genotoxicity

Key characteristic #2 (i.e., Is genotoxic) was generally the most minor carcinogenic activity component for most chemicals, with the notable exception being PFOSA (Figure 2). However, all the ToxCast assays corresponding to genotoxicity measured p53-regulated genotoxic stress, an indirect and constrained end point. Hence, the scope for assessing genotoxic activity was limited in this analysis.

Epigenetic Alterations

With regard to key characteristic #4 (i.e., Induces epigenetic alterations), the fluorotelomer alcohols tended to be the most consistently active when it came to enhancing expression of certain transcription factors (Table 1) that are indicative of changes to the epigenome.^23,24 In a previous study, carboxylic and sulfonic acid–based PFAS did not influence expression of FoxA2.²⁵ This finding supports the ToxCast result that only 8:2 FtOH affected FoxA2. On the other hand, sulfonates (free acid and salt) and short-chain carboxylic acids inhibited activity of key epigenome-linked histone and sirtuin deacetylase enzymes (Table 1).

Table 1.

PFAS Active in ToxCast Assays Correlating With IARC Key Characteristic #4 End Points.

Epigenetic factors	Characteristic	Active chemical(s)	Supporting literature
FoxA2↑	Transcription factor	8:2 FtOH	No changes in FoxA2 gene expression following PFOA, PFNA, or PFHxS exposure.²⁵
Pax6↑	Transcription factor	PFDA, PFUnA, PFOS, 6:2 FtOH, 8:2 FtOH, PFOSA, N-ethyl PFOSA	NA
Sox1↑	Transcription factor	PFDA, PFUnA, 8:2 FtOH, N-ethyl PFOSA	NA
Sp1↑	Transcription factor	PFOA, PFNA, PFOS, 8:2 FtOH, PFOSA, N-ethyl PFOSA	NA
HDAC3 inhibition	Deacetylase enzyme	PFHxA, PFOS	NA
SIRT2 inhibition	Deacetylase enzyme	PFOS, K-PFHxS	NA

Abbreviations: IARC, International Agency for Research on Cancer; K-PFHxS, K-perfluorohexanesulfonic acid; NA, not applicable; PFAS, per- and polyfluorinated alkyl substances; PFDA, perfluorodecanoic acid; PFHxA, perfluorohexanoic acid; PFHxS, perfluorohexanesulfonic acid; PFNA, perfluorononanoic acid, PFOA, perfluorooctanoic acid; PFOS, perfluorooctanesulfonic acid; PFOSA, perfluorooctanesulfonamide; PFUnA, perfluoroundecanoic acid.

Oxidative Stress

Key characteristic #5 (i.e., Induces oxidative stress) is assessed in ToxCast by examining the upregulation of several oxidative stress markers and antioxidative responses (Table 2). Phosphorylation of histone protein H2A and mitogen-activated protein kinase JNK acts as a marker for reactive oxygen species (ROS) triggered DNA and cellular damage^26,27 and hence higher levels of phosphorylated variants following exposure to various PFAS suggested induction of oxidative stress. Relevant PFAS included PFHpA, which has been previously shown to mediate hepatic and cardiac toxicity via activation of the phosphorylated JNK pathway,²⁸ thus supporting the ToxCast finding. Upregulation of the collagenase MMP-1 is another oxidative stress marker²⁹ and was observed to be increased following PFOA and PFOSA exposure. Hypoxic conditions can accumulate ROS,³⁰ and, accordingly, PFOSA, PFUnA, and a methacrylate fluorotelomer increased expression of H1F1α, a hypoxia-inducible factor. NRF2-ARE signaling is the central antioxidative pathway that attempts to restore homeostatic balance. It was triggered in response to exposure to a variety of PFAS, including PFOA, with enhanced NRF2 gene expression. Increased NRF2-ARE signaling in rat testes and liver of frogs exposed to PFOA, with greater expression of downstream NRF2 antioxidative targets (e.g., superoxide dismutase, catalase, heme oxygenase-1), has been documented.^31,32 Other oxidative stress protective pathways include the upregulation of metallothionein proteins (MTs) as well as recruitment of heat shock proteins (HSPs).^33,34 Activation of the metal response element promotes transcription of MTs and was observed in response to several PFAS. Heat shock factor 1 (HSF1) transcribes HSPs, and PFOSA and flubendiamide enhanced expression of the HSF1 gene.

Table 2.

PFAS Active in ToxCast Assays Correlating With IARC Key Characteristic #5 End Points.

Oxidative stress indicators	Characteristic	Active chemical(s)	Supporting literature
H2A phosphorylation↑	Histone protein	PFNA, PFDA, PFUnA, PFOS, K-PFOS, 8:2 FtOH, PFOSA, N-ethyl PFOSA, PFHpA	NA
JNK phosphorylation↑	Mitogen-activated stress kinase	PFHpA, PFDA, PFUnA, 8:2 FtOH, N-ethyl PFOSA, K-PFOS	PFHpA has adverse hepatic and cardiac effects via induction of c-Jun N-terminal kinase (JNK) phosphorylation.²⁸
H1F1α↑	Hypoxia-inducible factor	PFUnA, PFOSA, 6:2 Ft methacrylate	NA
MMP-1↑	Collagenase	PFOA, PFOSA	NA
NRF2↑	Master antioxidative transcription factor	PFOA, PFNA, PFUnA, PFOS, 6:2 FtOH, 8:2 FtOH, PFOSA, K-PFOS, NH₄-PFOA, lufenuron, K-PFHxS	Enhanced NRF2 expression and elevated antioxidative enzyme function in testes of mice exposed to PFOA.³¹
ARE ↑	Antioxidant response element	PFHpA, PFOA, NH₄-PFOA, PFDA, PFUnA, PFOS, PFOSA, N-ethyl PFOSA, Li-PFOS, lufenuron, flubendiamide	Increased ARE signaling in the livers of frogs exposed to PFOA.³²
MRE↑	Metal response element	PFNA, PFDA, PFUnA, PFOS, K-PFOS, 6:2 FtOH, 8:2 FtOH, PFOSA, N-ethyl PFOSA	NA
HSF1↑	Heat shock factor	PFOSA, flubendiamide	NA

Abbreviations: IARC, International Agency for Research on Cancer;K-PFHxS, K-perfluorohexanesulfonic acid; NA, not applicable; PFAS, per- and polyfluorinated alkyl substances; PFDA, perfluorodecanoic acid; PFHpA, perfluoroheptanoic acid; PFNA, perfluorononanoic acid, PFOA, perfluorooctanoic acid; PFOS, perfluorooctanesulfonic acid; PFOSA, perfluorooctanesulfonamide; PFUnA, perfluoroundecanoic acid.

Inflammation

In line with key characteristic #6 (i.e., Induces chronic inflammation), the increase in cell adhesion molecules (CAMs) and pro-inflammatory signaling factors (Table 3) is used to evaluate upregulation of inflammation pathways in ToxCast. Both PFOA and the potassium salt of PFOS were predicted to elevate levels of CAMs, specifically E-selectin and P-selectin. Prolonged exposure to K-PFOS has previously increased P-selectin in human umbilical vascular endothelial cells.³⁵ Nuclear factor (NF)-κB is a protein complex that acts as the master transcription factor in the activation of a systemic pro-inflammatory response. Macrophage generated cytokine tumor necrosis factor α (TNFα) activates NF-κB and results in the downstream production of pro-inflammatory cytokine interleukin-6.³⁶ Increased expression of the 3 proteins in this axis was only observed with exposure to alcohol fluorotelomers and sulfonamides in ToxCast. This observation was reinforced by a study,³⁷ which indicated PFAS fluorotelomers and sulfonamides did not interfere with the degradation of I-κB, inhibitory proteins that block NF-κB transcriptional activity. This contrasted with PFOS, which inhibited the degradation of I-κB, facilitating the sequestration of NF-κB in the cytosol, and possibly even decreasing downstream pro-inflammatory cytokine activity.

Table 3.

PFAS Active in ToxCast Assays Correlating With IARC Key Characteristic #6 End Points.

Inflammatory markers	Characteristic	Active chemical(s)	Supporting literature
NF-κB↑	Master pro-inflammatory transcription factor	6:2 FtOH, 8:2 FtOH, PFOSA, N-ethyl PFOSA	Sulfonic acids inhibit degradation of I-κB, hence sequestering NF-κB in the cytosol.³⁷ However, sulfonamides and fluorotelomers do not interfere with I-κB degradation.³⁷
P-selectin↑	Cell adhesion molecule	K-PFOS	Exposure to PFOS for 48-hour increases P-selectin expression.³⁵
IL-6↑	Pro-inflammatory cytokine	6:2 FtOH	NA
E-selectin↑	Cell adhesion molecule	PFOA, K-PFOS	NA
IL-8↑	Chemokine	PFHpA, PFNA, PFOS, K-PFOS, NH₄-PFOA	NA
TNFα↑	Pro-inflammatory cytokine	N-ethyl PFOSA, hexaflumuron	NA

Abbreviations: IARC, International Agency for Research on Cancer; K-PFHxS, K-perfluorohexanesulfonic acid; NA, not applicable; NF-κB, nuclear factor κB; PFAS, per- and polyfluorinated alkyl substances; PFHpA, perfluoroheptanoic acid; PFNA, perfluorononanoic acid; PFOS, perfluorooctanesulfonic acid; PFOSA, perfluorooctanesulfonamide.

Receptor-Mediated Effects

Key characteristic #8 (i.e., Modulates receptor-mediated effects) is studied in ToxCast as per changes in the activity of a variety of receptor targets (Table 4 and 5). PFOA, PFOS, and certain studied PFAS are known endocrine disruptors as well as peroxisome proliferators.^38-41 Major targets include the estrogen receptor α (ERα) and peroxisome proliferator–activated receptor γ (PPARγ), respectively. They are also considered mixed ERα agonists/antagonists due to their ability to bind as both. Predictably, PFAS-mediated ERα agonistic and antagonistic effects as well as PPARγ agonism in ToxCast were observed and, in general, the potency of action correlated with the length of the fluorine-bonded carbon chain (Table 4). Additionally, general differences in functional group activity were also seen, including stronger ERα antagonistic action by carboxylic acids relative to sulfonic acids, weaker ERα agonism by carboxylic acids compared to sulfonamides, and stronger PPARγ agonism by carboxylic acids relative to sulfonamides (Table 4).

Table 4.

Potency (1/μM) of PFAS-Modulated ERα and PPARγ Effects in ToxCast Assays.

PFAS	ATG_ERa_TRANS_up(ERα agonism)	TOX21_ERa_BLA_antagonist_ratio(ERα antagonism)	NVS_NR_hPPARg(PPARγ agonism)	ATG_PPARg_TRANS_up(PPARγ agonism)
PFHpA	NA	NA	NA	0.009
NH₄-PFOA	0.022	0.021	NA	0.046
PFOA	0.021	0.019	0.043	0.020
PFNA	0.047	0.043	NA	0.023
PFDA	0.045	0.044	0.073	0.017
PFUnA	0.020	0.526	0.128	0.024
K-PFHxS	NA	NA	0.036	0.033
K-PFOS	0.028	NA	0.049	0.045
PFOS	0.026	0.012	0.168	0.037
PFOSA	0.067	0.021	NA	0.012
N-ethyl PFOSA	0.057	0.704	NA	0.010
8:2 FtOH	0.037	NA	NA	NA
6:2 FtOH methacrylate	0.021	NA	NA	NA
Flubendiamide	NA	0.033	NA	NA

Abbreviations: ERα, estrogen receptor α; NA, not applicable; PFAS, per- and polyfluorinated alkyl substances; PFUnA, perfluoroundecanoic acid; PPARγ, peroxisome proliferator–activated receptor γ.

Table 5.

PFAS Active in ToxCast Assays Correlating With IARC Key Characteristic #8 Endpoints.

Receptor^a	Description	Active chemical(s)	Supporting literature
AhR	Aryl hydrocarbon receptor	6:2 FtOH, hexaflumuron	NA
AR	Androgen receptor	PFNA, PFDA, PFOS, 8:2 FtOH, PFOSA, N-ethyl PFOSA, K-PFOS, NH₄-PFOA, hexaflumuron, lufernuron	NA
AR^b	Androgen receptor	PFHpA, NH₄-PFOA, flubendiamide	NA
CAR	Constitutive androstane receptor	K-PFOS	PFOS activates CAR in a PPARα-independent manner.⁴⁵
CAR^b	Constitutive androstane receptor	PFOA, PFNA, PFOS, N-ethyl PFOSA	NA
DR4	TRAIL receptor 1	Lufenuron	NA
DR5	TRAIL receptor 2	PFUnA, N-ethyl PFOSA	NA
ERβ	β-Form of estrogen receptor	PFOSA	Chemicals with sulfonamide functional groups are estrogen receptor β agonists.⁴²
FXR	Bile acid receptor	PFNA, PFDA, PFOSA, lufenuron	PFNA increases mRNA expression of FXR > 2-fold in wild-type mice livers.⁴⁷
GR	Growth receptor	6:2 FtOH, PFUnA, PFOS, K-PFOS	NA
LXRα	α-Isoform of Liver X receptor	PFNA, PFUnA, N-ethyl PFOSA	NA
LXRβ	β-Isoform of Liver X receptor	Lufenuron	NA
PBREM	CAR heterodimer-binding enhancer module	PFDA, PFUnA, 8:2 FtOH, N-ethyl PFOSA	NA
PPARα	α-Type of peroxisome proliferator activated receptor	PFHxA, PFHpA, PFOA, NH₄-PFOA, PFNA, PFDA, PFOS, K-PFHxS, K-PFOS, PFOSA, N-ethyl PFOSA	PFOA and PFDA reduce activity of hepatic uptake transporters via PPARα activation.⁴⁴ PFNA disrupts cholesterol homeostasis and bile acid metabolism in mice wild-type mice livers via PPARα activation.⁴⁶
PPARδ	δ-Type of peroxisome proliferator activated receptor (PPAR)	PFOA	PFOA activates mouse and human PPARδ in vitro.⁴³
PPARδ^b	δelta-Type of PPAR	NH₄-PFOA, PFNA	NA
PXR	Pregnane X receptor	NH₄-PFOA, PFOA, PFNA, PFDA, K-PFOS, PFOS, N-ethyl PFOSA, K-PFHxS, flubendiamide, tetracosafluorotetradecahydrophenanthrene,	PFOA and PFOS induce cell toxicity pathway via PXR activation.⁴⁷
RARγ	γ-Type of Retinoic acid receptor	PFDA, K-PFHxS	NA
RXRα	α-Type of Retinoid X receptor	PFNA	NA
VDR	Calcitriol receptor	PFDA, PFUnA, 6:2 FtOH, 8:2 FtOH, PFOSA, N-ethyl PFOSA, hexaflumuron, flubendiamide	NA

Abbreviations: IARC, International Agency for Research on Cancer; mRNA, messenger RNA; NA, not applicable; PFAS, per- and polyfluorinated alkyl substances; PFHpA, perfluoroheptanoic acid; PFDA, perfluorodecanoic acid; PFNA, perfluorononanoic acid; PFOS, perfluorooctanesulfonic acid; PFOSA, perfluorooctanesulfonamide.

^a Receptor effects are agonistic, unless stated otherwise. Does not include ERα and PPARγ.

^b Antagonism.

ToxCast data also demonstrated PFAS to be ligands of other estrogen and PPAR receptor types, as well as of the androgen receptor (AR; Table 5). The ERβ was activated only by PFOSA, an observation in line with the literature finding that sulfonamide-based chemicals are selective ERβ agonists.⁴² A number of carboxylic and sulfonic PFAS were PPARα and PPARδ agonists, a finding consistent with the activation of human and mouse α and δ PPAR isoforms by PFOA and PFOS.⁴³ Moreover, PPARα activation is considered to drive the hepatotoxicity of PFOA and other long-chain carboxylic PFAS.⁴⁴ The activity toward the AR was mixed. The shorter chain PFHpA was an antagonist, longer chain PFNA and PFDA were agonists while the ammonium salt of PFOA was both.

In addition, several miscellaneous nuclear receptors were considered targets for PFAS binding (Table 5). Examples include activation of CAR by the potassium salt of PFOS, a notable observation since a study implicated CAR agonism by PFOS as a mode of PPARα-independent toxicity.⁴⁵ Both PFNA and PFDA activated FXR, while increased expression of FXR genes by PFNA exposure in mouse livers has been found in literature.⁴⁶ The PXR was also activated by a number of PFAS, and a study considers agonism of PXR by PFOA and PFOS as yet another mode of toxicity.⁴⁷ Non-nuclear receptor targets included the cell-surface TRAIL receptors (DR4 and DR5), but they were activated by only a few PFAS.

Alterations in Cell Proliferation

With regard to key characteristic #10 (i.e., Alters cell proliferation, cell death, or nutrient supply), many mapped ToxCast assays measured changes in cell growth, viability, and cytotoxicity by utilizing tools such as the SRB assay and mitochondrial toxicity end points. Certain cell-cycle markers were also assessed (Table 6), and these were consistently altered by long chain (≥8) carboxylic acids, alongside various functional group–based PFAS. Higher levels of transforming growth factor β1 (TGFβ1) messenger RNA were expressed, and this observation is bolstered by the findings that PFOA and PFOS stimulate epithelial–mesenchymal transition and subsequent renal fibrosis through upregulation of TGFβ1.^48,49 Increased expression of proto-oncogenes transcribed by Myb and Myc was also seen. HFPO-TA, an ether-based PFAS recently developed as an PFOA alternative, has demonstrated the ability to elevate Myc protein levels,⁵⁰ further validating the involvement of this pathway. Finally, greater expression of Xbp1-transcribed genes by PFAS was supported by higher levels of the active, spliced version of Xbp1 (sXbp1), being observed following PFOA exposure in liver cells.⁵¹ This would suggest PFAS mediate induction of prosurvival early-stage endoplasmic reticulum stress.⁵²

Table 6.

PFAS Active in ToxCast Assays Correlating With IARC Key Characteristic #10 End Points.

Cell-cycle markers	Characteristic	Active chemical(s)	Supporting literature
TGFβ1↑	Transforming growth factor	PFNA, PFDA, PFUnA, PFOS, K-PFOS, N-ethyl PFOSA, lufenuron	PFOS causes epithelial–mesenchymal transition and renal fibrosis via upregulation of TGFβ1.⁴⁸ PFOA exposure induces increased levels of TGFβ1 mRNA and activates renal fibroblasts.⁴⁹
Myb↑	Proto-oncogene	PFUnA, 6:2 FtOH	NA
Myc↑	Proto-oncogene	PFDA, PFUnA, N-ethyl PFOSA	HFPO-TA, a PFOA alternative, increases c-Myc protein levels in mice livers.⁵⁰
Xbp1↑	Transcription factor	PFDA, PFUnA, 6:2 FtOH, 8:2 FtOH, PFOSA, N-ethyl PFOSA	PFOA increases levels of the spliced variant of XBP1 in mice livers.⁵¹

Discussion

In recent years, there have been rising health concerns regarding the carcinogenic potential of PFAS.¹² As a result, studying the mechanistic modes of PFAS toxicity, in relation to key carcinogen characteristics, could play a key role in understanding how they might act as carcinogens.

Analysis of in vitro ToxCast data revealed that PFAS induce a huge number of diverse biological perturbations in several carcinogenicity-implicated molecular pathways (Table 7), many of which were validated by literature findings. The relatively higher overall scores for longer chain (≥8) sulfonamide and sulfonate PFAS (Figure 2) indicate they might be particularly toxic chemicals in mechanistic domains associated with carcinogenicity. This result would be in line with certain toxicity studies that found PFAS-containing sulfonic acid derivative moieties to be more potent than their carboxylic equivalents.^53,54

Table 7.

Summary of Major Biological Perturbations Induced by PFAS in ToxCast Assays.

Key characteristic #	Major pathways affected
4	Augmentation of epigenome-linked transcription factors (FoxA2, Pax6, Sox1, and Sp1) and inhibition of deacetylases (HDAC3 and SIRT2)
5	Upregulation of oxidative stress markers (phosphorylated H2A and JNK, H1F1α, and MMP-1) and elevation of antioxidative responses (NRF2, ARE, MRE, and HSF1)
6	Recruitment of cell adhesion molecules (E- and P-selectin) and functional group-selective influence on TNFα–NF-κB–IL-6 pro-inflammatory axis
8	Promiscuous, nonselective activation of several nuclear receptors, including peroxisome proliferation (PPAR α/δ/γ) and endocrine disruption (ER and AR)
10	Enhancement of transformation (TGFβ1), pro-proliferation (Myb and Myc), and antiapoptosis (Xbp1) markers

Abbreviations: AR, androgen receptor; ER, estrogen receptor; IL-6, interleukin-6; NF-κB, nuclear factor κB; PFAS, per- and polyfluorinated alkyl substances; TNFα, tumor necrosis factor α.

While the potential of PFAS to act as genotoxins (key characteristic #2) was relatively minor (Figure 2), their influence on gene expression through other modes of action was found to be of considerable interest from a standpoint of epigenetic and nuclear receptor-mediated toxicity. In accordance with key characteristic #4, epigenetic alterations involve nonsequence-based changes to the DNA-regulatory protein structure, often leading to different heritable phenotypic traits. For instance, blocking the deacetylation of histone proteins loosens the chromatin complex and is typically linked with transcriptional activation.⁵⁵ Changes in epigenomic markers and inhibition of deacetylase activity (Table 1) in response to PFAS exposure signifies a possible pathway for certain PFAS to influence gene expression. Key characteristic #8 covers the ability of chemicals to mediate effects by acting on receptors and was the most significant area of activity for PFAS. The vast majority of active ToxCast receptor-based assay hits were for nuclear receptors (Tables 4 and 5). This, in conjunction with the recognized endocrine disrupting and peroxisome proliferating action of PFAS, highlighted the capacity of PFAS to mimic endogenous hydrophobic signaling molecules and influence transcription of genes⁵⁶ as a major concern. The PFAS induce generation of ROS in both in vitro and in vivo toxicity models.^57-59 The upregulation of multiple oxidative stress-involved pathways and protein markers as well as elevated antioxidative responses was observed following PFAS exposure in ToxCast assays (Table 2). This implies an exogenously triggered systemic imbalance between ROS and antioxidants and hence further solidifies key characteristic #5 as another major component of PFAS toxicity. Persistent and unregulated inflammatory factor activity results in conditions of chronic inflammation. Correspondingly, the enhancement of CAMs or cytokines by particular PFAS (Table 3) suggests an ability of some of these chemicals to promote cell recruitment, aggregation, and sustained pro-inflammatory signaling (i.e., key characteristic #6). Chemicals that induce changes in cellular proliferation or death can be considered carcinogens as per key characteristic #10. The generally observed increase in cell growth, prosurvival, and transformation markers (Table 6) as well as a decrease in apoptosis suggest PFAS could also potentially exert pro-proliferative effects and inhibit cell death.

Generally, within each functional group class, the activity of PFAS was found to be directly proportional to the length of their fluorine-bonded carbon skeleton. This trend is consistent with literature findings, which deem long-chain PFAS more potent due to the presence of additional C–F bonds.^60-62 Hence, the analysis provides more evidence to support the phasing out of longer chain PFAS, in favor of short-chain alternatives. Some functional group-selective activities were also observed, including fluorotelomer alcohol-predominated epigenetic effects (Table 1) and discrepancy in inflammatory effects exerted by PFAS (Table 3), the latter due to variable influence on the nuclear translocation of NF-κB. The mixed activity toward the AR (Table 5) also indicates that chain length might influence the orientation of interaction with receptor-binding sites. Though, despite the prediction of some specific ligand–receptor interactions (e.g., sulfonamides and ERβ), the sheer variety of receptor targets (Table 5) points to the generally nonselective and promiscuous nature of PFAS as ligands.

Limitations

Certain limitations of the ToxCast data sets have been identified and need to be considered. First, the ToxCast assays do not account for all IARC-defined key characteristics, namely, #3 (Alters DNA repair or causes genomic instability), #7 (Is immunosuppressive), and #9 (Causes immortalization). In particular, the scarcity of assays that can be mapped to key characteristic #7 restricts the extent of the analysis since there is strong epidemiological and toxicological evidence connecting immunosuppression with exposure to PFOA and PFOS.⁶³ The chemical purity of the test substance PFAS utilized in assays has also not been determined and quality control analyses are still underway. Recently, a few assay-associated curve fitting flags that raise concerns of false positives have been identified⁶⁴ and subsequently addressed.⁶⁵ Finally, the surfactant-like properties of PFAS introduce some uncertainty about chemical disposition and partitioning, factors that can influence results of in vitro testing systems. Thus, in general, caution should be exercised before any conclusions or generalizations are made based on the findings of the study.

Conclusion

Overall, the inferences drawn here could act as initial steps in helping uncover the potential carcinogenic nature of PFAS. They might be applied toward further studying the hazards posed by PFAS by examining biomarkers of effect following PFAS exposure, thus providing potential tools to identify possible cancer risk.

Supplemental Material

Supplemental Material, sj-docx-1-ijt-10.1177_10915818211010490 - Exploring Potential Carcinogenic Activity of Per- and Polyfluorinated Alkyl Substances Utilizing High-Throughput Toxicity Screening Data

Supplemental Material, sj-docx-1-ijt-10.1177_10915818211010490 for Exploring Potential Carcinogenic Activity of Per- and Polyfluorinated Alkyl Substances Utilizing High-Throughput Toxicity Screening Data by Nalin Singh and Ching Yi Jennifer Hsieh in International Journal of Toxicology

Footnotes

Authors’ Contribution

Singh, N. contributed to conception and design; acquisition, analysis, and interpretation; drafted manuscript; and critically revised manuscript. Hsieh, C. J. contributed to conception and design; acquisition, analysis, and interpretation; and critically revised manuscript. Both authors gave final approval and agree to be accountable for all aspects of work ensuring integrity and accuracy. The content of this manuscript is solely the responsibility of the authors and does not represent the official views of the Office of Environmental Health Hazard Assessment, California Environmental Protection Agency, or the State of California.

Declaration of Conflicting Interests

The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Nalin Singh

Ching Yi Jennifer Hsieh

Supplemental Material

Supplemental material for this article is available online.

References

Ding

Harlow

Randolph

Jr Loch-Caruso

Park

. Perfluoroalkyl and polyfluoroalkyl substances (PFAS) and their effects on the ovary. Hum Reprod Update. 2020;26(5):724–752. doi:10.1093/humupd/dmaa018

Prevedouros

Cousins

Buck

. Sources, fate and transport of perfluorocarboxylates. Environ Sci Technol. 2006;40(1):32–44. doi:10.1021/es0512475

O’Hagan

. Understanding organofluorine chemistry. an introduction to the C–F bond. Chem Soc Rev. 2008;37(2):308–319. doi:10.1039/b711844a

Kudo

Kawashima

. Toxicity and toxicokinetics of perfluorooctanoic acid in humans and animals. J Toxicol Sci. 2003;28(2):49–57. doi:10.2131/jts.28.49

Betts

. Perfluoroalkyl acids: what is the evidence telling us? Environ Health Perspect. 2007;115(5):A250–A256. doi:10.1289/ehp.115-a250

Sunderland

Dassuncao

, et al. A review of the pathways of human exposure to poly- and perfluoroalkyl substances (PFASs) and present understanding of health effects. J Expo Sci Environ Epidemiol. 2019;29(2):131–147. doi:10.1038/s41370-018-0094-1

Fenton

Ducatman

Boobis

, et al. Per- and polyfluoroalkyl substance toxicity and human health review: current state of knowledge and strategies for informing future research. Environ Toxicol Chem. 2021; 40(3):606–630. doi:10.1002/etc.4890

Houde

Martin

Letcher

Solomon

Muir

. Biological monitoring of polyfluoroalkyl substances: a review. Environ Sci Technol. 2006;40(11):3463–3473. doi:10.1021/es052580b

Benninghoff

Orner

Buchner

, et al. Promotion of hepatocarcinogenesis by perfluoroalkyl acids in rainbow trout. Toxicol Sci. 2012; 125(1):69–78. doi:10.1093/toxsci/kfr267

10.

Lindh

Rylander

Toft

, et al. Blood serum concentrations of perfluorinated compounds in men from Greenlandic Inuit and European populations. Chemosphere. 2012; 88(11):1269–1275. doi:10.1016/j.chemosphere.2012.03.049

11.

Steenland

Fletcher

Savitz

. Epidemiologic evidence on the health effects of perfluorooctanoic acid (PFOA). Environ. Health Perspect. 2010;118(8):1100–1108. doi:10.1289/ehp.0901827

12.

Temkin

Hocevar

Andrews

Naidenko

Kamendulis

. Application of the key characteristics of carcinogens to per and polyfluoroalkyl substances. Int J Environ Res Public Health. 2020;17(5):1668. doi:10.3390/ijerph17051668

13.

Bonefeld-Jorgensen

Long

Bossi

, et al. Perfluorinated compounds are related to breast cancer risk in Greenlandic Inuit: a case control study. Environ Health. 2011;10. doi:10.1186/1476-069X-10-88

14.

Klaunig

Hocevar

Kamendulis

. Mode of action analysis of perfluorooctanoic acid (PFOA) tumorigenicity and human relevance. Reprod. Toxicol. 2012; 33(4):410–418. doi:10.1016/j.reprotox.2011.10.014

15.

Smith

Guyton

Gibbons

, et al. Key characteristics of carcinogens as a basis for organizing data on mechanisms of carcinogenesis. Environ. Health Perspect. 2016;124(6):713–721. doi:10.1289/ehp.1509912

16.

Chiu

Guyton

Martin

Reif

Rusyn

. Use of high-throughput in vitro toxicity screening data in cancer hazard evaluations by IARC monograph working groups. Altex 2018;35(1):51–64. doi:10.14573/altex.1703231

17.

Sipes

Martin

Kothiya

, et al. Profiling 976 ToxCast chemicals across 331 enzymatic and receptor signaling assays. Chem Res Toxicol. 2013;26(6):878–895. doi:10.1021/tx400021f

18.

Judson

Houck

Martin

, et al. Editor’s highlight: analysis of the effects of cell stress and cytotoxicity on in vitro assay activity across a diverse chemical and assay space. Toxicol Sci. 2016;152(2):323–339. doi:10.1093/toxsci/kfw092

19.

Martin

Dix

Judson

, et al. Impact of environmental chemicals on key transcription regulators and correlation to toxicity end points within EPA’s ToxCast program. Chem Res Toxicol. 2010;23(3):578–590. doi:10.1021/tx900325g

20.

Attene-Ramos

Miller

Huang

, et al. The Tox21 robotic platform for the assessment of environmental chemicals – from vision to reality. Drug Discov Today. 2013;18(15-16):716–723. doi:10.1016/j.drudis.2013.05.015

21.

Reif

Sypa

Lock

, et al. ToxPi GUI: an interactive visualization tool for transparent integration of data from diverse sources of evidence. Bioinformatics 2013;29(3):402–403. doi:10.1093/bioinformatics/bts686

22.

Fry

Reif

. Characterizing the effects of missing data and evaluating imputation methods for chemical prioritization applications using ToxPi. BioData Min. 2018;11:10. doi:10.1186/s13040-018-0169-5

23.

Honda

Pazin

Wernyj

Morin

. Crucial roles of Sp1 and epigenetic modifications in the regulation of the CLDN4 promoter in ovarian cancer cells. J Biol Chem. 2006;281(30):21433–21444. doi:10.1074/jbc.M603767200

24.

Ooki

Dinalankara

Marchionni

, et al. Epigenetically regulated PAX6 drives cancer cells toward a stem-like state via GLI-SOX2 signaling axis in lung adenocarcinoma. Oncogene. 2018;37(45):5967–5981. doi:10.1038/s41388-018-0373-2

25.

Das

Wood

Lin

, et al. Perfluoroalkyl acids-induced liver steatosis: effects on genes controlling lipid homeostasis. Toxicology. 2017;378:37–52. doi:10.1016/j.tox.2016.12.007

26.

Jayakumar

Panickar

Murthy Ch

Norenberg

. Oxidative stress and mitogen-activated protein kinase phosphorylation mediate ammonia-induced cell swelling and glutamate uptake inhibition in cultured astrocytes. J Neurosci. 2006;26(18):4774–4784. doi:10.1523/jneurosci.0120-06.2006

27.

Yang

Huang

. Oxidative stress induces H2AX phosphorylation in human spermatozoa. FEBS Lett. 2006;580(26):6161–6168. doi:10.1016/j.febslet.2006.10.016

28.

Kim

Park

Son

, et al. Perfluoroheptanoic acid affects amphibian embryogenesis by inducing the phosphorylation of ERK and JNK. Int J Mol Med. 2015; 36(6):1693–1700. doi:10.3892/ijmm.2015.2370

29.

Alge-Priglinger

Kreutzer

Obholzer

, et al. Oxidative stress-mediated induction of MMP-1 and MMP-3 in human RPE cells. Invest Ophthalmol Vis Sci. 2009;50(11):5495–5503. doi:10.1167/iovs.08-3193

30.

Pialoux

Mounier

Brown

, et al. Relationship between oxidative stress and HIF-1 alpha mRNA during sustained hypoxia in humans. Free Radic Biol Med. 2009;46(2):321–326. doi:10.1016/j.freeradbiomed.2008.10.047

31.

Yuan

, et al. Attenuation of perfluorooctanoic acid-induced testicular oxidative stress and apoptosis by quercetin in mice. RSC Adv. 2017;7(71):45045. doi:10.1039/c7ra09036f

32.

Tang

Jia

Gao

, et al. Role of the Nrf2-ARE pathway in perfluorooctanoic acid (PFOA)-induced hepatotoxicity in Rana nigromaculata. Environ Pollut. 2018;238:1035–1043. doi:10.1016/j.envpol.2018.02.037

33.

Dalton

Bittel

Liang

Oxidative stress activates metal-responsive transcription factor-1 binding activity. occupancy in vivo of metal response elements in the metallothionein-I gene promoter. J Biol Chem. 1996;271(42):26233–26241. doi:10.1074/jbc.271.42.26233

34.

Liu

Wang

. The role of heat shock proteins in oxidative stress damage induced by Se deficiency in chicken livers. Biometals 2015;28(1):163–173. doi:10.1007/s10534-014-9812-x

35.

Liao

Wang

Huang

, et al. Evaluation of cellular response to perfluorooctane sulfonate in human umbilical vein endothelial cells. Toxicol In Vitro 2012;26(3):421–428. doi:10.1016/j.tiv.2012.01.011

36.

Brasier

. The nuclear factor-kappaB-interleukin-6 signalling pathway mediating vascular inflammation. Cardiovasc Res. 2010;86(2):211–218. doi:10.1093/cvr/cvq076

37.

Corsini

Luebke

Germolec

DeWitt

. Perfluorinated compounds: emerging POPs with potential immunotoxicity. Toxicol Lett. 2014;230(2):263–270. doi:10.1016/j.toxlet.2014.01.038

38.

Cimini

Cristiano

Bernardo

, et al. Presence and inducibility of peroxisomes in a human glioblastoma cell line. Biochim Biophys Acta. 2000;1474(3):397–409. doi:10.1016/S0304-4165(00)00036-2

39.

Vanparys

Maras

Lenjou

, et al. Flow cytometric cell cycle analysis allows for rapid screening of estrogenicity in MCF-7 breast cancer cells. Toxicol In Vitro 2006; 20(7):1238–1248. doi:10.1016/j.tiv.2006.05.002

40.

Jensen

Leffers

. Emerging endocrine disrupters: perfluoroalkylated substances. Int J Androl. 2008;31(2):161–169. doi:10.1111/j.1365-2605.2008.00870.x

41.

Yamamoto

Yamane

Oishi

Kobayashi-Hattori

. Perfluorooctanoic acid binds to peroxisome proliferator-activated receptor gamma and promotes adipocyte differentiation in 3T3-L1 adipocytes. Biosci Biotechnol Biochem. 2015;79(4):636–639. doi:10.1080/09168451.2014.991683

42.

Roberts

Armor

Barker

, et al. Sulfonamides as selective oestrogen receptor beta agonists. Bioorg Med Chem Lett. 2011;21(19):5680–5683. doi:10.1016/j.bmcl.2011.08.041

43.

Takacs

Abbott

. 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:10.1093/toxsci/kfl135

44.

Cheng

Klaassen

. Critical role of PPAR-alpha in perfluorooctanoic acid- and perfluorodecanoic acid-induced downregulation of Oatp uptake transporters in mouse livers. Toxicol Sci. 2008;106(1):37–45. doi:10.1093/toxsci/kfn161

45.

Rosen

Das

Rooney

, et al. PPAR alpha-independent transcriptional targets of perfluoroalkyl acids revealed by transcript profiling. Toxicology 2017;387:95–107. doi:10.1016/j.tox.2017.05.013

46.

Zhang

Klaassen

Cheng

. Alteration of bile acid and cholesterol biosynthesis and transport by perfluorononanoic acid (PFNA) in mice. Toxicol Sci. 2018;162(1):225–233. doi:10.1093/toxsci/kfx237

47.

Gao

Xiang

, et al. Molecular mechanisms of PFOA-induced toxicity in animals and humans: implications for health risks. Environ Int. 2017;99:43–54. doi:10.1016/j.envint.2016.11.014

48.

Chou

Wen

Chang

, et al. From the cover: L-carnitine via PPARγ- and sirt1-dependent mechanisms attenuates epithelial-mesenchymal transition and renal fibrosis caused by perfluorooctanesulfonate. Toxicol Sci. 2017;160(2):217–229. doi:10.1093/toxsci/kfx183

49.

Rashid

Ramakrishnan

Fields

Irudayaraj

. Acute PFOA exposure promotes epigenomic alterations in mouse kidney tissues. Toxicol Rep. 2020;7:125–132. doi:10.1016/j.toxrep.2019.12.010

50.

Sheng

Pan

Guo

Sun

Dai

. Hepatotoxic effects of hexafluoropropylene oxide trimer acid (HFPO-TA), a novel perfluorooctanoic acid (PFOA) alternative, on mice. Environ Sci Technol. 2018;52(14):8005–8015. doi:10.1021/acs.est.8b01714

51.

Yan

Zhang

Wang

Zheng

Dai

. Perfluorooctanoic acid exposure induces endoplasmic reticulum stress in the liver and its effects are ameliorated by 4-phenylbutyrate. Free Radic Biol Med. 2015;87:300–311. doi:10.1016/j.freeradbiomed.2015.06.043

52.

Corazzari

Gagliardi

Fimia

Piacentini

. Endoplasmic reticulum stress, unfolded protein response, and cancer cell fate. Front Oncol. 2017;7:78. doi:10.3389/fonc.2017.00078

53.

Slotkin

MacKillop

Melnick

Thayer

Seidler

. Developmental neurotoxicity of perfluorinated chemicals modeled in vitro. Environ Health Perspect. 2008;116(6):716–722. doi:10.1289/ehp.11253

54.

. Toxicity of perfluorooctane sulfonate and perfluorooctanoic acid to plants and aquatic invertebrates. Environ Toxicol. 2009;24(1):95–101. doi:10.1002/tox.20396

55.

Chen

Zhao

. Histone deacetylases and mechanisms of regulation of gene expression. Crit Rev Oncog. 2015;20(1-2):35–47. doi:10.1615/critrevoncog.2015012997

56.

Sladek

. What are nuclear receptor ligands? Mol Cell Endocrinol. 2011;334(1-2):3–13. doi:10.1016/j.mce.2010.06.018

57.

Wielsøe

Long

Ghisari

Bonefeld-Jørgensen

. Perfluoroalkylated substances (PFAS) affect oxidative stress biomarkers in vitro. Chemosphere. 2015;129:239–245. doi:10.1016/j.chemosphere.2014.10.014

58.

Chen

Zhou

, et al. Maternal exposure to perfluorooctanoic acid inhibits luteal function via oxidative stress and apoptosis in pregnant mice. Reprod Toxicol. 2017;69:159–166. doi:10.1016/j.reprotox.2017.02.010

59.

Sonkar

Kay

Choudhury

. PFOS Modulates interactive epigenetic regulation in first-trimester human trophoblast cell line HTR-8/SVneo. Chem Res Toxicol. 2019;32(10):2016–2027. doi:10.1021/acs.chemrestox.9b00198

60.

Upham

Deocampo

Wurl

Trosko

. Inhibition of gap junctional intercellular communication by perfluorinated fatty acids is dependent on the chain length of the fluorinated tail. Int J Cancer. 1998;78(4):491–495. doi:10.1002/(sici)1097-0215(19981109)78:4<491:: Aid-ijc16>3.0.Co;2-9

61.

Xin

Wan

Yang

, et al. Perfluoroalkyl acid exposure induces protective mitochondrial and endoplasmic reticulum autophagy in lung cells. Arch Toxicol. 2018; 92(10):3131–3147. doi:10.1007/s00204-018-2266-0

62.

Pan

Qin

Liu

Dong

. Effects of carbon chain length on the perfluoroalkyl acids-induced oxidative stress of erythrocytes in vitro. J Agric Food Chem. 2018;66(25):6414–6420. doi:10.1021/acs.jafc.8b02197

63.

DeWitt

Blossom

Schaider

. Exposure to per-fluoroalkyl and polyfluoroalkyl substances leads to immunotoxicity: epidemiological and toxicological evidence. J Expo Sci Environ Epidemiol. 2019;29(2):148–156. doi:0.1038/s41370-018-0097-y

64.

Rose

Felter

Gadhia

, et al. Comment on coumarin hazard identification. Int J Toxicol. 2020; 39(5):491–492. doi:10.1177/1091581820942750

65.

Hsieh

Sun

Osborne

, et al. Response to comment on coumarin hazard identification. Int J Toxicol. 2020;39(5):493–494. doi:10.1177/1091581820945412

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB