Perfluorinated chemicals and adolescent respiratory health: Epidemiological evidence and mechanistic insights

Xinfeng Xu; Xinyao Jiang; Meng Zou; Jinyan Hui; Guang Huang; Qian Wu

doi:10.1371/journal.pone.0336788

Abstract

Perfluorinated compounds (PFCs) are persistent environmental pollutants with near-universal human exposure, yet their respiratory health impacts during adolescence remain insufficiently explored. This investigation evaluated single and combined effects of serum PFCs on pulmonary function and respiratory morbidity in a nationally representative adolescent cohort (n = 976, ages 12–19 years) utilizing 2007–2012 NHANES data. Advanced analytical approaches including multivariable regression, mixture modeling (BKMR and WQS), and mediation analysis were employed to assess associations with spirometric parameters (FEV₁, FVC, FEV₁/FVC) and respiratory symptoms while examining inflammatory and oxidative stress pathways. Computational approaches integrating network toxicology and molecular docking identified key protein targets. Analytical results demonstrated significant associations between specific PFC congeners (PFOA, PFHS, PFOS) and pulmonary function measures, with age-stratified effects observed for wheezing symptoms. Mixture analyses revealed PFOA as the predominant contributor to observed respiratory effects, partially mediated through oxidative stress pathways (6.8–8.2% mediation). Molecular investigations identified critical signaling nodes (INS, AKT1, TP53, TNF, IL6, ALB and PPARγ) potentially linking PFC exposure to respiratory outcomes. These findings provide mechanistic insights into PFC-induced pulmonary effects during adolescence, highlighting the need for continued investigation of these environmentally persistent compounds’ impact on developing respiratory systems. The integrated epidemiological-computational approach demonstrates the utility of combining population-level data with mechanistic modeling to elucidate environmental health effects.

Citation: Xu X, Jiang X, Zou M, Hui J, Huang G, Wu Q (2025) Perfluorinated chemicals and adolescent respiratory health: Epidemiological evidence and mechanistic insights. PLoS One 20(11): e0336788. https://doi.org/10.1371/journal.pone.0336788

Editor: Jiafu Li, University of South Carolina, UNITED STATES OF AMERICA

Received: July 29, 2025; Accepted: October 27, 2025; Published: November 14, 2025

Copyright: © 2025 Xu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the manuscript and its Supporting information files.

Funding: This work was supported by the National Natural Science Foundation of China (82073630, Qian Wu) and the Natural Science Foundation of the Higher Education Institution of Jiangsu Province (2020240510, Qian Wu). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Abbreviations: PFCs, Perﬂuorinated chemicals; FEV₁, forced expiratory volume in 1 second; FVC, forced vital capacity; PFOA, perfluorooctanoic acid; PFHS, perfluorohexane sulfonic acid; PFOS, perfluorooctane sulfonic acid; MPAH, 2-(N-Methyl-perfluorooctane sulfonamido) acetic acid; PFNA, perfluorononanoic acid; INS, insulin; AKT1, v-akt murine thymoma viral oncogene homologue 1; TP53, tumor protein p53; TNF, tumor necrosis factor; IL6, interleukin-6; ALB, albumin; PPARγ, peroxisome proliferator-activated receptor gamma; NHANES, nutrition examination survey; SPE-HPLC-TIS-MS/MS, solid phase extraction coupled to High Performance Liquid Chromatography-Turbo Ion Spray ionization-tandem Mass Spectrometry; PFHP, perfluoroheptanoic acid; PFDE, perfluorodecanoic acid; PFUA, perfluoroundecanoic acid; PFDO, perfluorododecanoic acid; PFBS, perfluorobutane sulfonic acid; PFSA, perfluorooctane sulfonamide; EPAH, 2-(N-ethyl-perfluorooctane sulfonamido) acetic acid; LOD, the limit of detection; COPD, chronic obstructive pulmonary disease; NLR, Neutrophil-to-lymphocyte ratio; SII, immune-inflammation index; SIRI, systemic inflammation response index; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; PPI, protein-protein interaction network; MCC, maximal clique centrality; MNC, maximum neighborhood component; EPC, edge percolated component; RCSB, research collaboratory for structural bioinformatics database; SD, standard deviation; OR, odds ratio; CI, confidence interval; RCS, restricted cubic spline; ML, machine learning; SHAP, SHapley Additive explanation; BKMR, Bayesian Kernel Machine Regression; WQS, Weighted Quantile Sum; IQR, interquartile range; AUC, area under the curve; HDM, House Dust Mite; Der p1, Dermatophagoides pteronyssinus group 1 antigen; LPS, lipopolysaccharide; BMI, body mass index.

1. Introduction

Perﬂuorinated chemicals (PFCs) are synthetic compounds characterized by a fully ﬂuorinated carbon backbone of variable length, terminated by an ionic functional group (typically carboxylate or sulfonate) [1,2]. These compounds possess unique characteristics such as water and oil repellency, thermal stability, and chemical resistance, making them valuable for commercial and industrial applications [3–5]. The extraordinary stability of PFCs’ carbon-fluorine bonds renders them highly persistent environmental contaminants [6]. Global monitoring studies have detected these compounds across diverse environmental matrices [2,7,8]. Key long-chain variants – including perfluorooctanoic acid (PFOA), perfluorooctane sulfonic acid (PFOS), perfluorononanoic acid (PFNA), and perfluorohexane sulfonic acid (PFHS) – exhibit remarkable persistence in humans, with biological half-lives spanning several years [9–11]. Epidemiologic studies detect these compounds in serum samples from >98% of the U.S. population, including children [10,12–15]. PFCs, particularly PFOA and PFOS, are recognized as priority contaminants due to their multiple exposure pathways (ingestion, dermal absorption, and inhalation) and demonstrated toxic effects in animal studies, including hepatic, developmental, immunological, neurobehavioral, endocrine, and metabolic toxicity [10,16–19].

Recent epidemiological studies have shown that serum PFC exposure may contribute to respiratory dysfunction, including asthma. An early analysis of the National Health and Nutrition Examination Survey (NHANES) data from 1999–2000 and 2003–2008 cycles revealed a significant positive association between serum PFOA levels and asthma diagnosis among adolescents aged 12–19 years [20]. Another study found an association, especially for PFUA exposure, with asthma among 3–11-year children from the 2013–2014 [21]. However, a study by using linear regression and weighted quantile sum regression found no associations between PFOA, PFNA, PFHS and PFOS and mixtures and the pulmonary function measures in 12–19 years adolescents (NHANES 2007–2012) [22]. But in the 2009–2010 Genetic and Biomarkers study for Childhood Asthma, PFAS concentrations showed significant associations with impaired pulmonary function in children with asthma [23].

Children and adolescents undergo critical lung development with heightened vulnerability to environmental pollutants. Growing evidence indicates early-life contaminant exposure may impair pulmonary maturation and elevate the risk of respiratory diseases in adult [24]. However, the results of the effect of PFCs on respiration health in children and adolescents is inconsistent. Consequently, there is a critical need to precisely characterize the impact of serum PFCs on respiration outcomes, specially impaired lung function and asthma development in children and adolescents.

In this study, we aimed to explore the association between serum PFC levels and health indicators in 12–19 aged children and adolescents using data from NHANES, 2007–2012. To assess the combined effects of PFC exposure on respiratory health, we applied mixture-exposure models. Additionally, we employed network toxicology analyses to explore potential toxicity pathways and molecular mechanisms underlying PFC-induced lung function impairment. Our findings, derived from a broad range of PFC exposure levels, may enhance the understanding of PFC-related health risks in pediatric populations.

2. Materials and methods

2.1. Study population

This study analyzed the data from a representative adolescent cohort (n = 976, ages 12–19 years) utilizing 2007–2012 NHANES data, which were accessed on April 1, 2025. NHANES is a national survey that assesses the health and nutrition status of adults and children across the United States, which offers open-access data and documentation for public use. A total of 976 eligible participants aged 12–19 years with serum-PFC measurements were included from the 2007–2008, 2009–2010, and 2011–2012 NHANES cycles, and they have provided lung function indicators and responses to the questions regarding respiration health in the questionnaires. NHANES data are publicly available, de-identified data approved by the National Center for Health Statistics (NCHS) Research Ethics Review Board, with study protocol approved by this review board and written consent obtained from all surveyed individuals. In addition, NHANES data are disseminated in compliance with strict confidentiality and privacy protection standards. As such, the authors did not have access to privacy information that could identify individual participants at any stage.

2.2. Exposure assessment

Serum PFCs were measured using standardized NHANES laboratory protocols. Briefly, solid phase extraction coupled to High Performance Liquid Chromatography-Turbo Ion Spray ionization-tandem Mass Spectrometry (online SPE-HPLC-TIS-MS/MS) was employed for the quantitative detection. For quantification, the internal standard method was employed. The following serum PFCs were detected: perfluoroheptanoic acid (PFHP), perfluorooctanoic acid (PFOA), perfluorononanoic acid (PFNA), perfluorodecanoic acid (PFDE), perfluoroundecanoic acid (PFUA), perfluorododecanoic acid (PFDO), perfluorobutane sulfonic acid (PFBS), perfluorohexane sulfonic acid (PFHS), perfluorooctane sulfonic acid (PFOS), perfluorooctane sulfonamide (PFSA), 2-(N-ethyl-perfluorooctane sulfonamido) acetic acid (EPAH), and 2-(N-methyl-perfluorooctane sulfonamido) acetic acid (MPAH). Samples with non-detectable levels were assigned a value of LOD/√2 in accordance with standard practice.

2.3. Health indices

The spirometry data collection is designed to investigate the prevalence of asthma and adult chronic obstructive pulmonary disease (COPD) in the U.S. population. In this study, three key spirometric parameters were analyzed: forced vital capacity (FVC), forced expiratory volume in 1 second (FEV₁), and their percentage ratio (FEV₁/FVC%), with the latter serving as the primary diagnostic criterion for obstructive lung disorders in current NHANES spirometry study.

Data on wheezing manifestations and asthma diagnosis were collected through health questionnaires. Current wheezing was defined based on an affirmative response to experiencing chest wheezing or whistling sounds within the previous year. Asthma cases were identified based on self-reported professional medical diagnoses.

Complete blood cell counts, including lymphocytes, neutrophils, monocytes, and platelets, were quantified using the Beckman Coulter DXH 800 hematology analyzer. Inflammation biomarkers were calculated as follows: neutrophil/lymphocyte ratio (NLR), systemic immune-inflammation index (SII = platelets × neutrophils/lymphocytes), and systemic inflammation response index (SIRI = monocytes × neutrophils/lymphocytes). Serum concentrations of oxidative stress biomarkers-total bilirubin and gamma-glutamyl transferase (GGT)-were quantitatively analyzed [25,26].

2.4. The prediction of PFC-related target genes

The potential molecular target connecting PFC exposure to juvenile respiratory disorders were systematically identified through GeneCards database and CTD database. Subsequent bioinformatics characterization was conducted through DAVID-based Gene Ontology (GO) term classification and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment studies. The Cytoscape (Version 3.9.1) was employed to construct the protein-protein interaction network (PPI). Subsequently, hub targets were systematically identified by evaluating seven topological parameters: maximal clique centrality (MCC), maximum neighborhood component (MNC), edge percolated component (EPC), stress, degree, closeness, and betweenness.

Protein structures for key targets were sourced from the RCSB Protein Data Bank, while the PFC configuration was obtained through PubChem. Subsequently, PFC, designated as the ligand, and the hub proteins, regarded as the receptors, were imported into AutoDockTools version 1.5.6, involving the following steps: (1) removal of crystallographic water molecules, (2) addition of polar hydrogens, and (3) computation of partial atomic charges. The prediction of ligand binding pockets in the protein of interest was carried by Prankweb. Protein-ligand docking simulations was performed using AutoDock Vina, with subsequent visualization and interaction analysis in PyMOL.

2.5. Statistical analysis

The baseline characteristics, e.g., demographic and clinical characteristics in the population were evaluated using descriptive analytical approaches. Continuous variables were described as median with interquartile range (IQR), and categorical variables were presented as numbers (n) and frequency (%). Spearman correlation coefficients was used to assess the associations between eight serum PFCs.

Associations between serum PFC concentrations and respiratory health parameters were evaluated using multiple linear regression, with results presented as regression coefficients (beta) or odds ratios (OR) value and 95% confidence interval (CI). In order to investigate potential disparities attributed to gender and age, the associations between serum PFCs and lung health were separately analyzed across distinct age and gender subgroups. Furthermore, the dose-response patterns linking serum PFC exposure to respiratory outcomes (lung function, wheeze, and asthma) were further explored using restricted cubic spline (RCS) regression.

Generalized Linear Model was used to assess the relationship between oxidative stress and inflammation related index and lung health as well as serum PFCs. A mediation analysis was employed to assess the mediating role of inflammation-related indices in the association between serum PFCs and lung health.

Due to the uncertain functional relationships between the PFC exposure levels and pulmonary function parameters (FEV₁, FVC, FEV₁/FVC), as well as respiratory conditions (wheeze and asthma), we initially applied machine learning (ML) methods for variable selection. We used the “pycaret” to get the training accuracy of different ML classification models and R2 of different ML regression models. After identifying the best-performing model through accuracy and R² metrics, we conducted additional training and utilized SHapley Additive exPlanation (SHAP) methodology to interpret variable contributions, while maintaining 10-fold cross-validation for all model. Through SHAP summary plots, model interpretation is achieved at two scales: overall feature importance (global) and individual prediction explanations (local), with feature importance ranked by descending mean SHAP values.

The associations between multiple PFC exposures and asthma, wheeze, FEV₁, FVC, and FEV₁/FVC were examined using Bayesian Kernel Machine Regression (BKMR) and Weighted Quantile Sum (WQS) regression. BKMR was applied to explore the potential nonlinear and non-additive joint effects of PFC mixtures. BKMR is a Bayesian semiparametric model that flexibly estimates exposure-response relationships and potential interactions among mixture components [27]. In addition, WQS regression was used as a mixture analysis method to construct a weighted index of exposures, ranking them into quantiles and estimating their joint association with the outcomes [28]. WQS assumes directional homogeneity (i.e., all exposures within the index affect the outcome in the same direction) and allows identification of the relative contribution of each chemical to the overall mixture effect. The WQS index was constructed using bootstrap sampling, with 40% of the data allocated for training and 60% for validation. Mediation analysis were analyzed by using R package “mediation”, which implemented causal mediation analysis. Data analysis was carried out in R software (version 4.4.3) and the significance level was set at P < 0.05 for all hypothesis tests.

3. Results

3.1. The characteristics of population and the distribution of PFC levels in the serum

Table 1 showed the characteristics of the 976 participants aged 12−19 years old from NHANES (2007−2012) included in our study. The median (interquartile range, IQR) age of eligible participants was 15 (14−17) years, with 45.18% females and 54.82% males. Clinically, the median level of total bilirubin (0.7 mg/dL) in this study was within the reference physiological range (≤ 1.34 mg/dL) [29], and the median GGT (14 U/L) was also within the reference physiological range (8−40 U/L for males, 6−26 U/L for females) [30], suggesting no widespread, severe hepatic oxidative stress in the cohort. The detection rates of serum PFCs in our studies were 100% for PFOA, 100% for PFNA, 100% for PFOS, 99.3% for PFHS, 83.8% for PFDE, 76.7% for MPAH, and 54.8% for PFUA, respectively. Among the measured PFAS, PFOS exhibited the highest median level of serum (6.495 ng/mL, IQR: 4.0–10.4), followed by PFOA (2.7 ng/mL, IQR: 1.8–3.7), PFHS (1.5 ng/mL, IQR: 0.8–2.9) and PFNA (0.984 ng/mL, IQR:0.71–1.394). In contrast, PFDE and MPAH showed lower concentrations, with medians of 0.2 ng/mL (IQR: 0.14–0.3) and 0.2 ng/mL (IQR: 0.1–0.3625), respectively. Spearman correlations among serum PFCs were presented in S1 Fig.

Download:

Table 1. Demographic characteristics of participants aged 12-19 years with available data in the NHANES 2007-2012 cycles (N = 976).

https://doi.org/10.1371/journal.pone.0336788.t001

3.2. Associations between serum PFCs and pulmonary function and wheeze and asthma

We analyzed the relationship between serum PFCs and pulmonary function indices and wheeze and asthma (Table 2). In serum PFCs, PFOA was positively correlated with FVC (β = 68.85 (40.1,97.59), P < 0.001) and FEV₁ (β = 42.37 (17.32,67.41), P < 0.001), and negatively correlated with FEV₁/FVC (β = −0.31 (−0.58,-0.04), P = 0.025); PFHS was positively correlated with FEV₁ (β = 8.96 (0.31,17.62), P = 0.042) and FVC (β = 14.4 (4.43, 24.37), P = 0.005); PFOS was positively correlated with FEV₁ (β = 6.53 (0.28,12.79), P = 0.041) and FVC (β = 10.15 (2.94,17.36), P = 0.006); MPAH was positively correlated with FVC (β = 114.63 (7.83, 221.43), P = 0.035) and wheeze (β = 0.05 (0, 0.1), P = 0.043) after covariate adjustment for confounding.

Download:

Table 2. Association of serum PFCs with lung health parameters.

https://doi.org/10.1371/journal.pone.0336788.t002

And the population was stratified according to gender to evaluate whether the association between serum PFCs and lung health differed between males and females. Age-stratified analyses revealed that PFOA was positively associated with FEV₁ and FVC in 12–15-year-old adolescents; PFOA was positively associated with wheeze and FVC, and negatively associated with FEV₁/FVC in 16–19-year-old adolescents. Sex-stratified analyses revealed that PFOA, PFHS, PFOS, MPAH were significantly associated with lung function (FEV₁ and FVC) in females and PFOA was significantly associated with FVC in males (S1 Table).

3.3. Restricted cubic spline models examined the dose-response relationship between serum PFC and pulmonary function and wheeze and asthma

Restricted cubic spline models were utilized to examine the dose-response relationships between serum PFC concentrations and pulmonary function (FEV₁, FVC, FEV₁/FVC ratio) as well as respiratory outcomes (wheeze and asthma). As shown in Figs 1 and 2, PFOA, PFHS and PFOS exhibited significant nonlinear dose-response relationships with FEV₁ and FVC. PFOA and PFOS showed an inverse U-shaped association with asthma. And PFHS showed a positive dose-response relationship with FEV₁ and FVC. MPAH demonstrated a positive linear association with FVC. And MPAH demonstrated a linear association with FVC. However, no significant dose-response relationships were observed for FEV₁/FVC ratio, wheeze, or asthma with any serum PFCs (S2–S4 Figs).

Download:

Fig 1. Relationships between serum PFC concentrations and FEV₁ from restricted cubic splines.

Three knots of RCS were located at 10th, 50th, and 90th. The dose-response relationships between PFC and FEV₁ were: PFOA-FEV₁, nonlinear; PFHS-FEV₁, nonlinear; PFOS-FEV₁, nonlinear.

https://doi.org/10.1371/journal.pone.0336788.g001

Download:

Fig 2. Relationships between serum PFC concentrations and FVC from restricted cubic splines.

Three knots of RCS were located at 10th, 50th, and 90th. The dose-response relationships between PFC and FVC were: PFOA-FVC, nonlinear; PFHS-FVC, nonlinear; PFOS-FVC, nonlinear; MPAH-FVC, linear.

https://doi.org/10.1371/journal.pone.0336788.g002

3.4. The joint effectiveness of serum PFC for respiratory health analyzed by Bayesian kernel machine regression and weighted quantile sum regression

We employed Bayesian kernel machine regression and weighted quantile sum regression to assess the joint effects of serum PFC mixtures on pulmonary function (FEV₁, FVC, FEV₁/FVC) and respiratory outcomes (wheeze, asthma). PFC co-exposure was significantly correlated with increased FEV₁ (Fig 3) and FVC (Fig 4). Conversely, the mixture showed a negative association with wheeze prevalence (Fig 5), indicating reduced odds of wheeze at higher levels of PFC. No significant linkages were identified between PFC mixtures and FEV₁/FVC and asthma risk in either model (S5 and S6 Figs). The weights of individual PFCs in the WQS model, indicating their contribution to the overall mixture effect, were shown in S7 Fig.

Download:

Fig 3. The effects of serum PFCs on FEV₁ was estimated by BKMR models.

(A) The joint effect of mixed exposure on FEV₁ compared with that at certain percentiles with those at the 50th percentile; (B) The effect of individual PFC exposures on FEV₁ along with their respective 95% confidence intervals; (C) The univariate risk for individual PFC exposures by quartiles; (D) Bivariate exposure-response at the 10% (red line), 50% (green line), and 90% (blue line) percentiles with all other PFCs set at their respective medians. Model was adjusted for age, gender, race and BMI.

https://doi.org/10.1371/journal.pone.0336788.g003

Download:

Fig 4. The effects of serum PFCs on FVC was estimated by BKMR models.

(A) The joint effect of mixed exposure on FVC compared with that at certain percentiles with those at the 50th percentile; (B) The effect of individual PFC exposures on FVC along with their respective 95% confidence intervals; (C) The univariate risk for individual PFC by quartiles; (D) Bivariate exposure-response at the 10% (red line), 50% (green line), and 90% (blue line) percentiles with all other PFCs set at their respective medians. Model was adjusted for age, gender, race and BMI.

https://doi.org/10.1371/journal.pone.0336788.g004

Download:

Fig 5. The effects of serum PFCs on wheeze risk was estimated by BKMR models.

(A) The joint effect of mixed exposure on wheeze risk compared with that at certain percentiles with those at the 50th percentile; (B) The effect of individual PFC exposures on wheeze risk along with their respective 95% confidence intervals; (C) The univariate risk for individual PFC by quartiles; (D) Bivariate exposure-response at the 10% (red line), 50% (green line), and 90% (blue line) percentiles with all other PFCs set at their respective medians. Model was adjusted for age, gender, race and BMI.

https://doi.org/10.1371/journal.pone.0336788.g005

3.5. Variable selection

S2–S6 Tables showed the performance of the ML models output by the “pycaret”. Combined R² or AUC, Lasso regression had the best regression performance for FEV₁, automatic relevance determination for FVC, support vector regression for FEV₁/FVC, and logistic regression for wheeze and asthma, respectively. To interpret the final model, the SHAP method was applied to assess the predictive contribution of each serum PFC. For example, the model suggested that individuals with higher PFOA levels were more likely to exhibit greater FEV₁ and FVC values while PFDE has the opposite effect (Fig 6). Although logistic regression had the best performance in current methods for wheeze and asthma, the AUC was less than 0.6 (S9 Fig). It is important to note that given this poor performance, these models are for exploratory analysis and aims to identify potential exposure-outcome relationships within a complex mixture.

Download:

Fig 6. Variables importance rankings measuring the association between PFCs exposure and lung functions, wheeze and asthma based on ML models interpreted with the SHAP approach.

(A) Bar plot for FEV₁ based on Lasso Regression model; (B) Dot plot for FEV₁ based on Lasso Regression model; (C) Bar plot for FVC based on Automatic Relevance Determination model; (D) Dot plot for FVC based on Automatic Relevance Determination model.

https://doi.org/10.1371/journal.pone.0336788.g006

3.6. Identification of potential mediators using mediation analysis

S7 Table presented the relationship among serum PFC and immune parameters as well as oxidative stress indicator. PFOA was associated with total bilirubin. Meanwhile, NLR, SIRI, gamma-glutamyl transferase and total bilirubin level were correlated with FVC and FEV₁ (S8 and S9 Tables). The outcomes of the mediation analysis indicated that total bilirubin explained 8.2% of the effect of PFOA on FEV₁ (indirect effect β = 3.55, 95% CI: 0.51, 7.35; total effect β = 42.10). Similarly, it mediated 6.8% of the effect of PFOA on FVC (indirect effect β = 4.73, 95% CI: 0.84, 10.10; total effect β = 69.05) and 9.0% of the effect of PFOS on FVC (indirect effect β = 0.94, 95% CI: 0.01, 2.13; total effect β = 10.15).

3.7. Screening for candidate target genes associated with PFC exposure, pulmonary function impairment and asthma

Total of 12593 PFCs-related genes were filtered from the comparative toxicogenomics database and 133 genes were highly related with inflammation as well as 924 genes were highly associated with lung function impairment in teenagers through the GeneCard database (relevance scores>50). Finally, we screened 97 cross-targets as targets for PAHs-induced pulmonary function impairment in teenagers. Gene ontology and Kyoto encyclopedia of genes and genomes analysis of the 25 targets based on the DAVID databases were performed (Fig 7A–7D). The results of pathway enrichment indicated that these genes were related with pathways in cancer, and Th17 cell differentiation. We constructed the protein-protein interaction network to show the interactions of targets by Cytoscape (Fig 7E). Then the 7 topological information was used to further screen the target genes and seven hub genes (INS, AKT1, TP53, TNF, IL6, ALB and PPARγ) were marked (Fig 7F). We also investigated the interaction between PFOA and seven hub proteins using molecular docking analysis (Fig 8). The binding affinity of PFOA with PPARγ and ALB was estimated to range from −7.5 kcal/mol to −9.0 kcal/mol, and from −8 kcal/mol to −10 kcal/mol, respectively, indicating moderate to strong spontaneous binding.

Download:

Fig 7. Exploration of the molecular mechanism by which serum PFCs metabolites influence lung health.

(A) Bubble diagram of BP in GO enrichment analysis; (B) Bubble diagram of CC in GO enrichment analysis; (C) Bubble diagram of MF in GO enrichment analysis; (D) Bar plot of KEGG enrichment analysis; (E) The PPI network of overlapping targets; (F) UpSet plot of 7 topological algorithms.

https://doi.org/10.1371/journal.pone.0336788.g007

Download:

Fig 8. Molecular docking in each protein with the PFCs.

(A) Molecular docking of PFOA and INS protein; (B) Molecular docking of PFOA and AKT1 protein; (C) Molecular docking of PFOA and TP53 protein; (D) Molecular docking of PFOA and ALB protein; (E) Molecular docking of PFOA and TNF protein; (F) Molecular docking of PFOA and IL-6 protein; (G) Molecular docking of PFOA and PPARγ protein.

https://doi.org/10.1371/journal.pone.0336788.g008

4. Discussion

Our analysis of NHANES data (2007–2012) revealed that serum PFCs, notably PFOA, PFOS, PFHS, and MPAH, were positively associated with lung function parameters (FEV₁ and FVC) in adolescents, particularly in females, and PFOA and MPAH exposure was significantly associated with an increased risk of wheeze specifically in older adolescents aged 16–19 years. Intriguingly, higher PFC concentrations exhibited a nonlinear dose-response relationship with improved FEV₁ and FVC. Conversely, PFOA was negatively associated with FEV₁/FVC, suggesting potential airway obstruction effects. Mediation analysis implicated total bilirubin in partially explaining the PFC-lung function relationship, and gene enrichment studies identified pathways related to inflammation (e.g., Th17 differentiation) and key hub genes (e.g., IL6, TNF, PPARγ) potentially linking PFC exposure to lung health.

These results were inconsistent with a meta-analysis which reported PFC-associated declines in childhood lung function and increased asthma risk [31]. However, our findings aligned with emerging evidence suggesting that PFC effects may vary by age, exposure window, and sex. For instance, the sex-specific associations-stronger in females was observed, a finding aligned with sex-stratified immune responses to PFCs in adolescents reported by Pan et al. [32]. Meanwhile, the stronger associations in females also were found in the COPSAC2010 cohort, where prenatal PFC exposure was linked to altered immune profiles in girls but not boys [33]. Hormonal interactions, particularly estrogen’s role in modulating PPARγ and oxidative stress pathways, may explain this sex disparity. For instance, estrogen enhances PPARγ expression, potentially amplifying anti-inflammatory responses to PFAS in females [32]. Conversely, PFCs are known endocrine disruptors, and their interference with hormone signaling could disproportionately affect females during puberty-a critical window for lung development [34]. Such discrepancies emphasized the need to consider demographic and developmental factors in PFC toxicity assessments.

The lack of significant associations between serum PFCs and asthma aligned with several studies [35,36] but contrasted with others [37,38]. This inconsistency may stem from heterogeneity in asthma phenotypes, exposure timing (prenatal vs. postnatal), or outcome measurement (e.g., self-reported wheeze vs. clinician-diagnosed asthma). Our reliance on NHANES data, which uses self-reported wheeze and asthma, may underestimate true associations due to misclassification bias. Furthermore, the low AUC (<0.6) for asthma models highlighted the limitations of current PFC biomarkers in predicting complex multifactorial diseases. We further explored the association between serum PFC levels and health indicators in 12–19 aged children and adolescents(n = 3352) using data from the 2013–2014, 2015–2016, and 2017–2018 NHANES circles. However, the data from these circles did not included three key spirometric parameters. S9 Table showed demographic characteristics of participants aged 12–19 years with available data in the NHANES 2013–2018 cycles. Age-stratified analyses revealed that PFDE (OR=1.22, P = 0.036) and MPAH (OR=1.152, P = 0.038) was positively associated with asthma in 12–15-year-old adolescents. Sex-stratified analyses revealed that PFDE were significantly associated with asthma in males (OR=1.19, P = 0.046). PFOA exhibited significant nonlinear dose-response relationships with asthma in 12–15-year-old adolescents in restricted cubic spline models (S9 Fig). There were no joint effects of serum PFC mixtures on asthma by Bayesian kernel machine regression and weighted quantile sum regression. And also, there were no mediating effects of immune parameters as well as oxidative stress indicator between serum PFC and asthma. However, studies using murine asthma models demonstrate that PFC exposure induces asthma-related outcomes, including airway hyperresponsiveness and enhanced inflammatory responses [11]. Early-life exposure to PFCs in mice attenuated airway antigen bioactivity, impairing the modulation of pulmonary inflammatory responses and posing risks to lung development. Specifically, PFOS exhibited high-affinity binding to the major HDM allergen Der p1 and the lipid A moiety of lipopolysaccharide, resulting in the functional inactivation of both antigens [39]. These findings suggested that the potential impact of serum PFC exposure on asthma pathogenesis warrants further investigation through a comprehensive approach combining robust epidemiological studies with carefully designed animal experiments.

Our gene enrichment and molecular docking analyses provide a tentative roadmap for mechanistic research. The identification of Th17 differentiation and cancer-related pathways aligns with experimental evidence linking serum PFC to Th17/regulatory T-cell imbalance—a hallmark of allergic inflammation [40]. Similarly, hub genes like IL6 and TNF, central to inflammatory cascades, were prioritized in protein-protein interaction networks. Molecular docking results further suggested that serum PFC may directly interact with proteins such as PPARγ and ALB, potentially disrupting lipid metabolism and inflammatory signaling. These findings resonate with Solan and Park’s [41] call for leveraging route-to-route extrapolation for risk assessment to unravel PFAS toxicity mechanisms.

There was some limitation in this study. (1) The study used cross-sectional data from NHANES (2007–2012), which limits our ability to establish temporal relationships or infer causality. Longitudinal studies are needed to confirm the directionality of these associations. (2) Although we adjusted for key demographic covariates (e.g., age, sex, BMI), unmeasured confounders (e.g., diet, indoor air pollution, genetic susceptibility) may still influence the observed associations. (3) PFC exposure was assessed using single-timepoint serum measurements, which may not accurately reflect long-term exposure patterns, especially for short-chain PFC (e.g., PFBA) with shorter biological half-lives. Intra-individual variability in PFC levels over time could lead to exposure misclassification, potentially biasing results toward the null. Future studies should incorporate repeated exposure measurements to improve accuracy. (4) It must be acknowledged that the network toxicology analyses in this study serve as a supportive component rather than a core mechanistic investigation. While these analyses provide valuable insights into the potential toxicity pathways and molecular mechanisms underlying PFC-induced lung function impairment, they do not constitute novel mechanistic evidence and should be interpreted as hypothesis-generating for future research.

5. Conclusion

Our results demonstrated that serum PFC exposure contributes to impaired respiratory health among vulnerable groups, suggesting that PFC effects are dose-, age-, sex-, and outcome-specific. These findings underscore the importance of addressing environmental exposures during early life stages to mitigate the burden of adult respiratory diseases.

Supporting information

S1 Fig. Spearman correlation coefficients between different serum PFCs.

https://doi.org/10.1371/journal.pone.0336788.s001

(TIF)

S2 Fig. Relationships between serum PFC concentrations and FEV₁/FVC from restricted cubic splines.

Three knots of RCS were located at 10th, 50th, and 90th.

https://doi.org/10.1371/journal.pone.0336788.s002

(TIF)

S3 Fig. Relationships between serum PFC concentrations and wheeze risk from restricted cubic splines.

Three knots of RCS were located at 10th, 50th, and 90th.

https://doi.org/10.1371/journal.pone.0336788.s003

(TIF)

S4 Fig. Relationships between serum PFC concentrations and asthma risk from restricted cubic splines.

Three knots of RCS were located at 10th, 50th, and 90th.

https://doi.org/10.1371/journal.pone.0336788.s004

(TIF)

S5 Fig. The effects of serum PFCs on FEV₁/FVC was estimated by BKMR models.

(A) The joint effect of mixed exposure on FEV₁/FVC compared with that at certain percentiles with those at the 50th percentile; (B) The effect of individual PFC exposures on FEV₁/FVC along with their respective 95% confidence intervals; (C) The univariate risk for individual PFC by quartiles; (D) Bivariate exposure-response at the 10% (red line), 50% (green line), and 90% (blue line) percentiles with all other PFCs set at their respective medians. Model was adjusted for age, gender, race and BMI.

https://doi.org/10.1371/journal.pone.0336788.s005

(TIF)

S6 Fig. The effects of serum PFCs on asthma risk was estimated by BKMR models.

(A) The joint effect of mixed exposure on asthma risk compared with that at certain percentiles with those at the 50th percentile; (B) The effect of individual PFC exposures on asthma risk along with their respective 95% confidence intervals; (C) The univariate risk for individual PFC by quartiles; (D) Bivariate exposure-response at the 10% (red line), 50% (green line), and 90% (blue line) percentiles, with all other PFCs set at their respective medians. Model was adjusted for age, gender, race and BMI.

https://doi.org/10.1371/journal.pone.0336788.s006

(TIF)

S7 Fig. Association between serum PFC levels and FEV₁ (A), FVC (B), FEV₁/FVC (C), wheeze (D) and asthma (E) based on weighted quantile sum (WQS) regression analysis.

https://doi.org/10.1371/journal.pone.0336788.s007

(TIF)

S8 Fig. Variables importance rankings for the association between PFCs exposure and lung functions, wheeze and asthma based on ML models interpreted with the SHAP approach.

(A) Bar plot for FEV₁/FVC based on Support Vector Regression model; (B) Dot plot for FEV₁/FVC based on Support Vector Regression model; (C) Bar plot for wheeze based on Logistic Regression model; (D) Dot plot for wheeze based on Logistic Regression model; (E) Bar plot for asthma based on Logistic Regression model; (F) Dot plot for asthma based on Logistic Regression model.

https://doi.org/10.1371/journal.pone.0336788.s008

(TIF)

S9 Fig. Relationships between PFOA concentrations and asthma risk from restricted cubic splines.

Three knots of RCS were located at 10th, 50th, and 90th.

https://doi.org/10.1371/journal.pone.0336788.s009

(TIF)

S1 Table. Associations of serum PFCs with lung health stratified by age and gender in the NHANES 2007–2012.

https://doi.org/10.1371/journal.pone.0336788.s010

(DOCX)

S2 Table. Performance of the machine learning model for regression of “FEV₁”.

https://doi.org/10.1371/journal.pone.0336788.s011

(DOCX)

S3 Table. Performance of the machine learning model for regression of “FVC”.

https://doi.org/10.1371/journal.pone.0336788.s012

(DOCX)

S4 Table. Performance of the machine learning model for regression of “FEV₁/FVC”.

https://doi.org/10.1371/journal.pone.0336788.s013

(DOCX)

S5 Table. Performance of the machine learning model for classification of “Wheeze”.

https://doi.org/10.1371/journal.pone.0336788.s014

(DOCX)

S6 Table. Performance of the machine learning model for classification of “Asthma”.

https://doi.org/10.1371/journal.pone.0336788.s015

(DOCX)

S7 Table. Association between serum PFCs and immune indices and oxidative stress index.

https://doi.org/10.1371/journal.pone.0336788.s016

(DOCX)

S8 Table. Association between immune indices and oxidative stress index and lung health parameters.

https://doi.org/10.1371/journal.pone.0336788.s017

(DOCX)

S9 Table. Demographic characteristics of participants aged 12‐19 years with available data in the NHANES 2013‐2018 cycles (N = 3352).

https://doi.org/10.1371/journal.pone.0336788.s018

(DOCX)

References

1. Prevedouros K, Cousins IT, Buck RC, Korzeniowski SH. Sources, fate and transport of perfluorocarboxylates. Environ Sci Technol. 2006;40(1):32–44. pmid:16433330
- View Article
- PubMed/NCBI
- 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):366–94. pmid:17519394
- View Article
- PubMed/NCBI
- Google Scholar
3. Ahrens L, Bundschuh M. Fate and effects of poly- and perfluoroalkyl substances in the aquatic environment: a review. Environ Toxicol Chem. 2014;33(9):1921–9. pmid:24924660
- View Article
- PubMed/NCBI
- Google Scholar
4. Černá M, Grafnetterová AP, Dvořáková D, Pulkrabová J, Malý M, Janoš T, et al. Biomonitoring of PFOA, PFOS and PFNA in human milk from Czech Republic, time trends and estimation of infant’s daily intake. Environ Res. 2020;188:109763. pmid:32540571
- View Article
- PubMed/NCBI
- Google Scholar
5. Paul AG, Jones KC, Sweetman AJ. A first global production, emission, and environmental inventory for perfluorooctane sulfonate. Environ Sci Technol. 2009;43(2):386–92. pmid:19238969
- View Article
- PubMed/NCBI
- Google Scholar
6. Wang T, Wang Y, Liao C, Cai Y, Jiang G. Perspectives on the inclusion of perfluorooctane sulfonate into the Stockholm Convention on Persistent Organic Pollutants. Environ Sci Technol. 2009;43(14):5171–5. pmid:19708337
- View Article
- PubMed/NCBI
- Google Scholar
7. Boulanger B, Vargo JD, Schnoor JL, Hornbuckle KC. Evaluation of perfluorooctane surfactants in a wastewater treatment system and in a commercial surface protection product. Environ Sci Technol. 2005;39(15):5524–30. pmid:16124283
- View Article
- PubMed/NCBI
- Google Scholar
8. Higgins CP, Field JA, Criddle CS, Luthy RG. Quantitative determination of perfluorochemicals in sediments and domestic sludge. Environ Sci Technol. 2005;39(11):3946–56. pmid:15984769
- View Article
- PubMed/NCBI
- Google Scholar
9. Olsen GW, Burris JM, Ehresman DJ, Froehlich JW, Seacat AM, Butenhoff JL, et al. Half-life of serum elimination of perfluorooctanesulfonate,perfluorohexanesulfonate, and perfluorooctanoate in retired fluorochemical production workers. Environ Health Perspect. 2007;115(9):1298–305. pmid:17805419
- View Article
- PubMed/NCBI
- Google Scholar
10. Calafat AM, Needham LL, Kuklenyik Z, Reidy JA, Tully JS, Aguilar-Villalobos M, et al. Perfluorinated chemicals in selected residents of the American continent. Chemosphere. 2006;63(3):490–6. pmid:16213555
- View Article
- PubMed/NCBI
- Google Scholar
11. Corsini E, Luebke RW, Germolec DR, DeWitt JC. Perfluorinated compounds: emerging POPs with potential immunotoxicity. Toxicol Lett. 2014;230(2):263–70.
- View Article
- Google Scholar
12. Olsen GW, Church TR, Miller JP, Burris JM, Hansen KJ, Lundberg JK, et al. Perfluorooctanesulfonate and other fluorochemicals in the serum of American Red Cross adult blood donors. Environ Health Perspect. 2003;111(16):1892–901. pmid:14644663
- View Article
- PubMed/NCBI
- Google Scholar
13. Olsen GW, Church TR, Larson EB, van Belle G, Lundberg JK, Hansen KJ, et al. Serum concentrations of perfluorooctanesulfonate and other fluorochemicals in an elderly population from Seattle, Washington. Chemosphere. 2004;54(11):1599–611. pmid:14675839
- View Article
- PubMed/NCBI
- Google Scholar
14. Calafat AM, Kuklenyik Z, Reidy JA, Caudill SP, Tully JS, Needham LL. Serum concentrations of 11 polyfluoroalkyl compounds in the u.s. population: data from the national health and nutrition examination survey (NHANES). Environ Sci Technol. 2007;41(7):2237–42. pmid:17438769
- View Article
- PubMed/NCBI
- Google Scholar
15. Kato K, Calafat AM, Wong L-Y, Wanigatunga AA, Caudill SP, Needham LL. Polyfluoroalkyl compounds in pooled sera from children participating in the National Health and Nutrition Examination Survey 2001-2002. Environ Sci Technol. 2009;43(7):2641–7. pmid:19452929
- View Article
- PubMed/NCBI
- Google Scholar
16. Kennedy GL Jr, Butenhoff JL, Olsen GW, O’Connor JC, Seacat AM, Perkins RG, et al. The toxicology of perfluorooctanoate. Crit Rev Toxicol. 2004;34(4):351–84. pmid:15328768
- View Article
- PubMed/NCBI
- Google Scholar
17. Lau C, Butenhoff JL, Rogers JM. The developmental toxicity of perfluoroalkyl acids and their derivatives. Toxicol Appl Pharmacol. 2004;198(2):231–41. pmid:15236955
- View Article
- PubMed/NCBI
- Google Scholar
18. Cheng J, Lv S, Nie S, Liu J, Tong S, Kang N, et al. Chronic perfluorooctane sulfonate (PFOS) exposure induces hepatic steatosis in zebrafish. Aquat Toxicol. 2016;176:45–52. pmid:27108203
- View Article
- PubMed/NCBI
- Google Scholar
19. Chen N, Li J, Li D, Yang Y, He D. Chronic exposure to perfluorooctane sulfonate induces behavior defects and neurotoxicity through oxidative damages, in vivo and in vitro. PLoS One. 2014;9(11):e113453. pmid:25412474
- View Article
- PubMed/NCBI
- Google Scholar
20. Humblet O, Diaz-Ramirez LG, Balmes JR, Pinney SM, Hiatt RA. Perfluoroalkyl chemicals and asthma among children 12-19 years of age: NHANES (1999-2008). Environ Health Perspect. 2014;122(10):1129–33.
- View Article
- Google Scholar
21. Wang Y-F, Xie B, Zou Y-X. Association between PFAS congeners exposure and asthma among US children in a nationally representative sample. Environ Geochem Health. 2023;45(8):5981–90. pmid:37195568
- View Article
- PubMed/NCBI
- Google Scholar
22. Shi S, Ding Y, Wu B, Hu P, Chen M, Dong N, et al. Association of perfluoroalkyl substances with pulmonary function in adolescents (NHANES 2007-2012). Environ Sci Pollut Res Int. 2023;30(18):53948–61. pmid:36869952
- View Article
- PubMed/NCBI
- Google Scholar
23. Qin X-D, Qian ZM, Dharmage SC, Perret J, Geiger SD, Rigdon SE, et al. Association of perfluoroalkyl substances exposure with impaired lung function in children. Environ Res. 2017;155:15–21. pmid:28171771
- View Article
- PubMed/NCBI
- Google Scholar
24. Agyapong PD, Jack D, Kaali S, Colicino E, Mujtaba MN, Chillrud SN, et al. Household Air Pollution and Child Lung Function: The Ghana Randomized Air Pollution and Health Study. Am J Respir Crit Care Med. 2024;209(6):716–26.
- View Article
- Google Scholar
25. Nocentini A, Bonardi A, Pratesi S, Gratteri P, Dani C, Supuran CT. Pharmaceutical strategies for preventing toxicity and promoting antioxidant and anti-inflammatory actions of bilirubin. J Enzyme Inhib Med Chem. 2022;37(1):487–501. pmid:34986721
- View Article
- PubMed/NCBI
- Google Scholar
26. Zhang T, Yao C, Zhou X, Liu S, Qi L, Zhu S, et al. Glutathione‑degrading enzymes in the complex landscape of tumors (Review). Int J Oncol. 2024;65(1):72. pmid:38847236
- View Article
- PubMed/NCBI
- Google Scholar
27. Bobb JF, Claus Henn B, Valeri L, Coull BA. Statistical software for analyzing the health effects of multiple concurrent exposures via Bayesian kernel machine regression. Environ Health. 2018;17(1):67. pmid:30126431
- View Article
- PubMed/NCBI
- Google Scholar
28. Czarnota J, Gennings C, Wheeler DC. Assessment of weighted quantile sum regression for modeling chemical mixtures and cancer risk. Cancer Inform. 2015;14(Suppl 2):159–71. pmid:26005323
- View Article
- PubMed/NCBI
- Google Scholar
29. Reference intervals for common clinical biochemistry tests-Part 4: Serum total bilirubin and direct bilirubin (in Chinese), WS/T 404.4-2018[S].
30. Reference intervals of clinial biochemistry tests commonly used for children (in Chinese), WS/T 780-2021[S].
31. Rafiee A, Faridi S, Sly PD, Stone L, Kennedy LP, Mahabee-Gittens EM. Asthma and decreased lung function in children exposed to perfluoroalkyl and polyfluoroalkyl substances (PFAS): An updated meta-analysis unveiling research gaps. Environ Res. 2024;262(Pt 1):119827. pmid:39182754
- View Article
- PubMed/NCBI
- Google Scholar
32. Pan Z, Guo Y, Zhou Q, Wang Q, Pan S, Xu S, et al. Perfluoroalkyl substance exposure is associated with asthma and innate immune cell count in US adolescents stratified by sex. Environ Sci Pollut Res Int. 2023;30(18):52535–48. pmid:36840869
- View Article
- PubMed/NCBI
- Google Scholar
33. Sevelsted A, Pedersen C-ET, Gürdeniz G, Rasmussen MA, Schullehner J, Sdougkou K, et al. Exposures to perfluoroalkyl substances and asthma phenotypes in childhood: an investigation of the COPSAC2010 cohort. EBioMedicine. 2023;94:104699. pmid:37429082
- View Article
- PubMed/NCBI
- Google Scholar
34. Sonthithai P, Suriyo T, Thiantanawat A, Watcharasit P, Ruchirawat M, Satayavivad J. Perfluorinated chemicals, PFOS and PFOA, enhance the estrogenic effects of 17β-estradiol in T47D human breast cancer cells. J Appl Toxicol. 2016;36(6):790–801. pmid:26234195
- View Article
- PubMed/NCBI
- Google Scholar
35. Impinen A, Nygaard UC, Lødrup Carlsen KC, Mowinckel P, Carlsen KH, Haug LS, et al. Prenatal exposure to perfluoralkyl substances (PFASs) associated with respiratory tract infections but not allergy- and asthma-related health outcomes in childhood. Environ Res. 2018;160:518–23. pmid:29106950
- View Article
- PubMed/NCBI
- Google Scholar
36. Kvalem HE, Nygaard UC, Lødrup Carlsen KC, Carlsen KH, Haug LS, Granum B. Perfluoroalkyl substances, airways infections, allergy and asthma related health outcomes - implications of gender, exposure period and study design. Environ Int. 2020;134:105259. pmid:31733527
- View Article
- PubMed/NCBI
- Google Scholar
37. Atagi T, Hasegawa K, Motoki N, Inaba Y, Toubou H, Shibazaki T, et al. Associations between prenatal exposure to per- and polyfluoroalkyl substances and wheezing and asthma symptoms in 4-year-old children: The Japan Environment and Children’s Study. Environ Res. 2024;240(Pt 1):117499. pmid:37914018
- View Article
- PubMed/NCBI
- Google Scholar
38. Philippat C, Coiffier O, Lyon-Caen S, Boudier A, Jovanovic N, Quentin J, et al. In utero exposure to poly- and perfluoroalkyl substances and children respiratory health in the three first years of life. Environ Res. 2023;234:116544. pmid:37406719
- View Article
- PubMed/NCBI
- Google Scholar
39. Wang M, Li Q, Hou M, Chan LLY, Liu M, Ter SK. Inactivation of common airborne antigens by perfluoroalkyl chemicals modulates early life allergic asthma. Proc Natl Acad Sci U S A. 2021;118(24).
- View Article
- Google Scholar
40. von Holst H, Nayak P, Dembek Z, Buehler S, Echeverria D, Fallacara D, et al. Perfluoroalkyl substances exposure and immunity, allergic response, infection, and asthma in children: review of epidemiologic studies. Heliyon. 2021;7(10):e08160. pmid:34712855
- View Article
- PubMed/NCBI
- Google Scholar
41. Solan ME, Park J-A. Per- and poly-fluoroalkyl substances (PFAS) effects on lung health: a perspective on the current literature and future recommendations. Front Toxicol. 2024;6:1423449. pmid:39092081
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Prevedouros K, Cousins IT, Buck RC, Korzeniowski SH. Sources, fate and transport of perfluorocarboxylates. Environ Sci Technol. 2006;40(1):32–44. pmid:16433330
View Article
PubMed/NCBI
Google Scholar

[2] View Article

[3] PubMed/NCBI

[4] Google Scholar

[ref2] 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):366–94. pmid:17519394
View Article
PubMed/NCBI
Google Scholar

[6] View Article

[7] PubMed/NCBI

[8] Google Scholar

[ref3] 3. Ahrens L, Bundschuh M. Fate and effects of poly- and perfluoroalkyl substances in the aquatic environment: a review. Environ Toxicol Chem. 2014;33(9):1921–9. pmid:24924660
View Article
PubMed/NCBI
Google Scholar

[10] View Article

[11] PubMed/NCBI

[12] Google Scholar

[ref4] 4. Černá M, Grafnetterová AP, Dvořáková D, Pulkrabová J, Malý M, Janoš T, et al. Biomonitoring of PFOA, PFOS and PFNA in human milk from Czech Republic, time trends and estimation of infant’s daily intake. Environ Res. 2020;188:109763. pmid:32540571
View Article
PubMed/NCBI
Google Scholar

[14] View Article

[15] PubMed/NCBI

[16] Google Scholar

[ref5] 5. Paul AG, Jones KC, Sweetman AJ. A first global production, emission, and environmental inventory for perfluorooctane sulfonate. Environ Sci Technol. 2009;43(2):386–92. pmid:19238969
View Article
PubMed/NCBI
Google Scholar

[18] View Article

[19] PubMed/NCBI

[20] Google Scholar

[ref6] 6. Wang T, Wang Y, Liao C, Cai Y, Jiang G. Perspectives on the inclusion of perfluorooctane sulfonate into the Stockholm Convention on Persistent Organic Pollutants. Environ Sci Technol. 2009;43(14):5171–5. pmid:19708337
View Article
PubMed/NCBI
Google Scholar

[22] View Article

[23] PubMed/NCBI

[24] Google Scholar

[ref7] 7. Boulanger B, Vargo JD, Schnoor JL, Hornbuckle KC. Evaluation of perfluorooctane surfactants in a wastewater treatment system and in a commercial surface protection product. Environ Sci Technol. 2005;39(15):5524–30. pmid:16124283
View Article
PubMed/NCBI
Google Scholar

[26] View Article

[27] PubMed/NCBI

[28] Google Scholar

[ref8] 8. Higgins CP, Field JA, Criddle CS, Luthy RG. Quantitative determination of perfluorochemicals in sediments and domestic sludge. Environ Sci Technol. 2005;39(11):3946–56. pmid:15984769
View Article
PubMed/NCBI
Google Scholar

[30] View Article

[31] PubMed/NCBI

[32] Google Scholar

[ref9] 9. Olsen GW, Burris JM, Ehresman DJ, Froehlich JW, Seacat AM, Butenhoff JL, et al. Half-life of serum elimination of perfluorooctanesulfonate,perfluorohexanesulfonate, and perfluorooctanoate in retired fluorochemical production workers. Environ Health Perspect. 2007;115(9):1298–305. pmid:17805419
View Article
PubMed/NCBI
Google Scholar

[34] View Article

[35] PubMed/NCBI

[36] Google Scholar

[ref10] 10. Calafat AM, Needham LL, Kuklenyik Z, Reidy JA, Tully JS, Aguilar-Villalobos M, et al. Perfluorinated chemicals in selected residents of the American continent. Chemosphere. 2006;63(3):490–6. pmid:16213555
View Article
PubMed/NCBI
Google Scholar

[38] View Article

[39] PubMed/NCBI

[40] Google Scholar

[ref11] 11. Corsini E, Luebke RW, Germolec DR, DeWitt JC. Perfluorinated compounds: emerging POPs with potential immunotoxicity. Toxicol Lett. 2014;230(2):263–70.
View Article
Google Scholar

[42] View Article

[43] Google Scholar

[ref12] 12. Olsen GW, Church TR, Miller JP, Burris JM, Hansen KJ, Lundberg JK, et al. Perfluorooctanesulfonate and other fluorochemicals in the serum of American Red Cross adult blood donors. Environ Health Perspect. 2003;111(16):1892–901. pmid:14644663
View Article
PubMed/NCBI
Google Scholar

[45] View Article

[46] PubMed/NCBI

[47] Google Scholar

[ref13] 13. Olsen GW, Church TR, Larson EB, van Belle G, Lundberg JK, Hansen KJ, et al. Serum concentrations of perfluorooctanesulfonate and other fluorochemicals in an elderly population from Seattle, Washington. Chemosphere. 2004;54(11):1599–611. pmid:14675839
View Article
PubMed/NCBI
Google Scholar

[49] View Article

[50] PubMed/NCBI

[51] Google Scholar

[ref14] 14. Calafat AM, Kuklenyik Z, Reidy JA, Caudill SP, Tully JS, Needham LL. Serum concentrations of 11 polyfluoroalkyl compounds in the u.s. population: data from the national health and nutrition examination survey (NHANES). Environ Sci Technol. 2007;41(7):2237–42. pmid:17438769
View Article
PubMed/NCBI
Google Scholar

[53] View Article

[54] PubMed/NCBI

[55] Google Scholar

[ref15] 15. Kato K, Calafat AM, Wong L-Y, Wanigatunga AA, Caudill SP, Needham LL. Polyfluoroalkyl compounds in pooled sera from children participating in the National Health and Nutrition Examination Survey 2001-2002. Environ Sci Technol. 2009;43(7):2641–7. pmid:19452929
View Article
PubMed/NCBI
Google Scholar

[57] View Article

[58] PubMed/NCBI

[59] Google Scholar

[ref16] 16. Kennedy GL Jr, Butenhoff JL, Olsen GW, O’Connor JC, Seacat AM, Perkins RG, et al. The toxicology of perfluorooctanoate. Crit Rev Toxicol. 2004;34(4):351–84. pmid:15328768
View Article
PubMed/NCBI
Google Scholar

[61] View Article

[62] PubMed/NCBI

[63] Google Scholar

[ref17] 17. Lau C, Butenhoff JL, Rogers JM. The developmental toxicity of perfluoroalkyl acids and their derivatives. Toxicol Appl Pharmacol. 2004;198(2):231–41. pmid:15236955
View Article
PubMed/NCBI
Google Scholar

[65] View Article

[66] PubMed/NCBI

[67] Google Scholar

[ref18] 18. Cheng J, Lv S, Nie S, Liu J, Tong S, Kang N, et al. Chronic perfluorooctane sulfonate (PFOS) exposure induces hepatic steatosis in zebrafish. Aquat Toxicol. 2016;176:45–52. pmid:27108203
View Article
PubMed/NCBI
Google Scholar

[69] View Article

[70] PubMed/NCBI

[71] Google Scholar

[ref19] 19. Chen N, Li J, Li D, Yang Y, He D. Chronic exposure to perfluorooctane sulfonate induces behavior defects and neurotoxicity through oxidative damages, in vivo and in vitro. PLoS One. 2014;9(11):e113453. pmid:25412474
View Article
PubMed/NCBI
Google Scholar

[73] View Article

[74] PubMed/NCBI

[75] Google Scholar

[ref20] 20. Humblet O, Diaz-Ramirez LG, Balmes JR, Pinney SM, Hiatt RA. Perfluoroalkyl chemicals and asthma among children 12-19 years of age: NHANES (1999-2008). Environ Health Perspect. 2014;122(10):1129–33.
View Article
Google Scholar

[77] View Article

[78] Google Scholar

[ref21] 21. Wang Y-F, Xie B, Zou Y-X. Association between PFAS congeners exposure and asthma among US children in a nationally representative sample. Environ Geochem Health. 2023;45(8):5981–90. pmid:37195568
View Article
PubMed/NCBI
Google Scholar

[80] View Article

[81] PubMed/NCBI

[82] Google Scholar

[ref22] 22. Shi S, Ding Y, Wu B, Hu P, Chen M, Dong N, et al. Association of perfluoroalkyl substances with pulmonary function in adolescents (NHANES 2007-2012). Environ Sci Pollut Res Int. 2023;30(18):53948–61. pmid:36869952
View Article
PubMed/NCBI
Google Scholar

[84] View Article

[85] PubMed/NCBI

[86] Google Scholar

[ref23] 23. Qin X-D, Qian ZM, Dharmage SC, Perret J, Geiger SD, Rigdon SE, et al. Association of perfluoroalkyl substances exposure with impaired lung function in children. Environ Res. 2017;155:15–21. pmid:28171771
View Article
PubMed/NCBI
Google Scholar

[88] View Article

[89] PubMed/NCBI

[90] Google Scholar

[ref24] 24. Agyapong PD, Jack D, Kaali S, Colicino E, Mujtaba MN, Chillrud SN, et al. Household Air Pollution and Child Lung Function: The Ghana Randomized Air Pollution and Health Study. Am J Respir Crit Care Med. 2024;209(6):716–26.
View Article
Google Scholar

[92] View Article

[93] Google Scholar

[ref25] 25. Nocentini A, Bonardi A, Pratesi S, Gratteri P, Dani C, Supuran CT. Pharmaceutical strategies for preventing toxicity and promoting antioxidant and anti-inflammatory actions of bilirubin. J Enzyme Inhib Med Chem. 2022;37(1):487–501. pmid:34986721
View Article
PubMed/NCBI
Google Scholar

[95] View Article

[96] PubMed/NCBI

[97] Google Scholar

[ref26] 26. Zhang T, Yao C, Zhou X, Liu S, Qi L, Zhu S, et al. Glutathione‑degrading enzymes in the complex landscape of tumors (Review). Int J Oncol. 2024;65(1):72. pmid:38847236
View Article
PubMed/NCBI
Google Scholar

[99] View Article

[100] PubMed/NCBI

[101] Google Scholar

[ref27] 27. Bobb JF, Claus Henn B, Valeri L, Coull BA. Statistical software for analyzing the health effects of multiple concurrent exposures via Bayesian kernel machine regression. Environ Health. 2018;17(1):67. pmid:30126431
View Article
PubMed/NCBI
Google Scholar

[103] View Article

[104] PubMed/NCBI

[105] Google Scholar

[ref28] 28. Czarnota J, Gennings C, Wheeler DC. Assessment of weighted quantile sum regression for modeling chemical mixtures and cancer risk. Cancer Inform. 2015;14(Suppl 2):159–71. pmid:26005323
View Article
PubMed/NCBI
Google Scholar

[107] View Article

[108] PubMed/NCBI

[109] Google Scholar

[ref29] 29. Reference intervals for common clinical biochemistry tests-Part 4: Serum total bilirubin and direct bilirubin (in Chinese), WS/T 404.4-2018[S].

[ref30] 30. Reference intervals of clinial biochemistry tests commonly used for children (in Chinese), WS/T 780-2021[S].

[ref31] 31. Rafiee A, Faridi S, Sly PD, Stone L, Kennedy LP, Mahabee-Gittens EM. Asthma and decreased lung function in children exposed to perfluoroalkyl and polyfluoroalkyl substances (PFAS): An updated meta-analysis unveiling research gaps. Environ Res. 2024;262(Pt 1):119827. pmid:39182754
View Article
PubMed/NCBI
Google Scholar

[113] View Article

[114] PubMed/NCBI

[115] Google Scholar

[ref32] 32. Pan Z, Guo Y, Zhou Q, Wang Q, Pan S, Xu S, et al. Perfluoroalkyl substance exposure is associated with asthma and innate immune cell count in US adolescents stratified by sex. Environ Sci Pollut Res Int. 2023;30(18):52535–48. pmid:36840869
View Article
PubMed/NCBI
Google Scholar

[117] View Article

[118] PubMed/NCBI

[119] Google Scholar

[ref33] 33. Sevelsted A, Pedersen C-ET, Gürdeniz G, Rasmussen MA, Schullehner J, Sdougkou K, et al. Exposures to perfluoroalkyl substances and asthma phenotypes in childhood: an investigation of the COPSAC2010 cohort. EBioMedicine. 2023;94:104699. pmid:37429082
View Article
PubMed/NCBI
Google Scholar

[121] View Article

[122] PubMed/NCBI

[123] Google Scholar

[ref34] 34. Sonthithai P, Suriyo T, Thiantanawat A, Watcharasit P, Ruchirawat M, Satayavivad J. Perfluorinated chemicals, PFOS and PFOA, enhance the estrogenic effects of 17β-estradiol in T47D human breast cancer cells. J Appl Toxicol. 2016;36(6):790–801. pmid:26234195
View Article
PubMed/NCBI
Google Scholar

[125] View Article

[126] PubMed/NCBI

[127] Google Scholar

[ref35] 35. Impinen A, Nygaard UC, Lødrup Carlsen KC, Mowinckel P, Carlsen KH, Haug LS, et al. Prenatal exposure to perfluoralkyl substances (PFASs) associated with respiratory tract infections but not allergy- and asthma-related health outcomes in childhood. Environ Res. 2018;160:518–23. pmid:29106950
View Article
PubMed/NCBI
Google Scholar

[129] View Article

[130] PubMed/NCBI

[131] Google Scholar

[ref36] 36. Kvalem HE, Nygaard UC, Lødrup Carlsen KC, Carlsen KH, Haug LS, Granum B. Perfluoroalkyl substances, airways infections, allergy and asthma related health outcomes - implications of gender, exposure period and study design. Environ Int. 2020;134:105259. pmid:31733527
View Article
PubMed/NCBI
Google Scholar

[133] View Article

[134] PubMed/NCBI

[135] Google Scholar

[ref37] 37. Atagi T, Hasegawa K, Motoki N, Inaba Y, Toubou H, Shibazaki T, et al. Associations between prenatal exposure to per- and polyfluoroalkyl substances and wheezing and asthma symptoms in 4-year-old children: The Japan Environment and Children’s Study. Environ Res. 2024;240(Pt 1):117499. pmid:37914018
View Article
PubMed/NCBI
Google Scholar

[137] View Article

[138] PubMed/NCBI

[139] Google Scholar

[ref38] 38. Philippat C, Coiffier O, Lyon-Caen S, Boudier A, Jovanovic N, Quentin J, et al. In utero exposure to poly- and perfluoroalkyl substances and children respiratory health in the three first years of life. Environ Res. 2023;234:116544. pmid:37406719
View Article
PubMed/NCBI
Google Scholar

[141] View Article

[142] PubMed/NCBI

[143] Google Scholar

[ref39] 39. Wang M, Li Q, Hou M, Chan LLY, Liu M, Ter SK. Inactivation of common airborne antigens by perfluoroalkyl chemicals modulates early life allergic asthma. Proc Natl Acad Sci U S A. 2021;118(24).
View Article
Google Scholar

[145] View Article

[146] Google Scholar

[ref40] 40. von Holst H, Nayak P, Dembek Z, Buehler S, Echeverria D, Fallacara D, et al. Perfluoroalkyl substances exposure and immunity, allergic response, infection, and asthma in children: review of epidemiologic studies. Heliyon. 2021;7(10):e08160. pmid:34712855
View Article
PubMed/NCBI
Google Scholar

[148] View Article

[149] PubMed/NCBI

[150] Google Scholar

[ref41] 41. Solan ME, Park J-A. Per- and poly-fluoroalkyl substances (PFAS) effects on lung health: a perspective on the current literature and future recommendations. Front Toxicol. 2024;6:1423449. pmid:39092081
View Article
PubMed/NCBI
Google Scholar

[152] View Article

[153] PubMed/NCBI

[154] Google Scholar

Figures

Abstract

1. Introduction

2. Materials and methods

2.1. Study population

2.2. Exposure assessment

2.3. Health indices

2.4. The prediction of PFC-related target genes

2.5. Statistical analysis

3. Results

3.1. The characteristics of population and the distribution of PFC levels in the serum

3.2. Associations between serum PFCs and pulmonary function and wheeze and asthma

3.3. Restricted cubic spline models examined the dose-response relationship between serum PFC and pulmonary function and wheeze and asthma

3.4. The joint effectiveness of serum PFC for respiratory health analyzed by Bayesian kernel machine regression and weighted quantile sum regression

3.5. Variable selection

3.6. Identification of potential mediators using mediation analysis

3.7. Screening for candidate target genes associated with PFC exposure, pulmonary function impairment and asthma

4. Discussion

5. Conclusion

Supporting information

S1 Fig. Spearman correlation coefficients between different serum PFCs.

S2 Fig. Relationships between serum PFC concentrations and FEV1/FVC from restricted cubic splines.

S3 Fig. Relationships between serum PFC concentrations and wheeze risk from restricted cubic splines.

S4 Fig. Relationships between serum PFC concentrations and asthma risk from restricted cubic splines.

S5 Fig. The effects of serum PFCs on FEV1/FVC was estimated by BKMR models.

S6 Fig. The effects of serum PFCs on asthma risk was estimated by BKMR models.

S7 Fig. Association between serum PFC levels and FEV1 (A), FVC (B), FEV1/FVC (C), wheeze (D) and asthma (E) based on weighted quantile sum (WQS) regression analysis.

S8 Fig. Variables importance rankings for the association between PFCs exposure and lung functions, wheeze and asthma based on ML models interpreted with the SHAP approach.

S9 Fig. Relationships between PFOA concentrations and asthma risk from restricted cubic splines.

S1 Table. Associations of serum PFCs with lung health stratified by age and gender in the NHANES 2007–2012.

S2 Table. Performance of the machine learning model for regression of “FEV1”.

S3 Table. Performance of the machine learning model for regression of “FVC”.

S4 Table. Performance of the machine learning model for regression of “FEV1/FVC”.

S5 Table. Performance of the machine learning model for classification of “Wheeze”.

S6 Table. Performance of the machine learning model for classification of “Asthma”.

S7 Table. Association between serum PFCs and immune indices and oxidative stress index.

S8 Table. Association between immune indices and oxidative stress index and lung health parameters.

S9 Table. Demographic characteristics of participants aged 12‐19 years with available data in the NHANES 2013‐2018 cycles (N = 3352).

References

S2 Fig. Relationships between serum PFC concentrations and FEV₁/FVC from restricted cubic splines.

S5 Fig. The effects of serum PFCs on FEV₁/FVC was estimated by BKMR models.

S7 Fig. Association between serum PFC levels and FEV₁ (A), FVC (B), FEV₁/FVC (C), wheeze (D) and asthma (E) based on weighted quantile sum (WQS) regression analysis.

S2 Table. Performance of the machine learning model for regression of “FEV₁”.

S4 Table. Performance of the machine learning model for regression of “FEV₁/FVC”.