Exposure to PFOS and PFOA through drinking water is associated with greater cigarette smoke-induced COPD-like lung injury and enhanced NF-κB-related inflammatory signaling in mice

Abstract

Chronic obstructive pulmonary disease (COPD) is a major global health burden, and environmental pollutants may modify its progression. Perfluorooctane sulfonate (PFOS) and perfluorooctanoic acid (PFOA) are persistent per- and polyfluoroalkyl substances with known immunotoxic and proinflammatory effects, but their contribution to smoke-related lung injury remains unclear. Here, we investigated whether chronic PFOS or PFOA exposure was associated with greater cigarette smoke (CS)-induced COPD-like lung injury. Mice were exposed to CS for 3 months with concurrent PFOS or PFOA exposure in drinking water at the same nominal concentration (2 mg/L). Lung function, histopathology, transcriptomic profiling, and supportive in vitro assays in RAW264.7 cells were performed. Compared with CS alone, PFOS or PFOA co-exposure with CS was associated with greater pulmonary impairment, including greater alveolar destruction, enhanced airway inflammation, and more pronounced impairment of pulmonary function, as reflected by a lower forced expiratory volume in one second to forced vital capacity ratio (FEV₁/FVC). Transcriptomic and in vitro analyses were consistent with enhanced nuclear factor-κB (NF-κB)-related inflammatory signaling. These findings support a potential role for drinking-water PFAS exposure as a modifier of smoke-induced pulmonary injury.

Graphical abstract

Highlights

• •Drinking-water PFOS/PFOA exposure was associated with greater cigarette smoke-induced lung injury.
• •PFOS/PFOA co-exposure was associated with increased alveolar destruction and pulmonary inflammation.
• •PFOS/PFOA co-exposure was associated with reduced FEV₁/FVC ratio in cigarette smoke-exposed mice.
• •Combined PFOS/PFOA and cigarette smoke exposure was associated with enhanced NF-κB-related inflammatory signaling.
• •Drinking-water PFOS/PFOA exposure may modify smoke-induced pulmonary injury.

1. Introduction

Chronic obstructive pulmonary disease (COPD) is a progressive respiratory disorder characterized by persistent airflow limitation, abnormal inflammatory responses, and irreversible structural damage to the lung parenchyma (Wu et al., 2023; Li et al., 2022; Wan and Chen, 2025). It remains one of the leading causes of morbidity and mortality worldwide, with cigarette smoking recognized as the predominant risk factor (Le Rouzic et al., 2016; Yang et al., 2024). However, epidemiological evidence indicates that a considerable proportion of COPD patients have no history of smoking, suggesting that additional environmental and occupational exposures contribute to disease onset and progression (Prince, 2018). Among these factors, exposure to environmental pollutants such as particulate matter, heavy metals, and perfluoroalkyl substances (PFAS) has been increasingly associated with pulmonary toxicity and chronic respiratory diseases (Protano et al., 2024).

Perfluorooctane sulfonate (PFOS) and perfluorooctanoic acid (PFOA) are two of the most widely used per- and polyfluoroalkyl substances (PFAS). Owing to their chemical stability and surfactant properties, these materials have been extensively applied in industrial and consumer products, including firefighting foams, nonstick coatings, and waterproof materials (Jiao et al., 2021). However, their persistence in the environment and resistance to biodegradation have led to global contamination of air, water, and soil (Drago et al., 2024). PFOS and PFOA have been detected in human serum, breast milk, and even lung tissues, raising significant public health concerns (Dubińska-Magiera et al., 2016). Numerous studies have demonstrated that PFAS exposure can disrupt lipid metabolism, impair immune homeostasis, and induce oxidative stress and inflammation (Li et al., 2021). Although PFOS and PFOA have been restricted or phased out in many regions and are increasingly being replaced by alternative PFAS, they remain environmentally and biologically relevant because of their persistence, historical accumulation, and continued detection in contaminated environmental media and human samples. Moreover, PFOS and PFOA are among the most extensively studied PFAS, providing a well-characterized framework for investigating pollutant–smoke interactions in chronic lung injury. Nevertheless, because alternative PFAS and PFAS mixtures may differ in chemical structure, persistence, bioaccumulation, and toxicokinetic behavior, the present findings should be extrapolated to other PFAS or PFAS mixtures with caution. However, their potential role in chronic lung diseases, particularly COPD, remains largely unexplored.

Emerging evidence suggests that PFAS may act as immunotoxic agents capable of modulating both innate and adaptive immune responses (Ehrlich et al., 2023). Macrophages, as key mediators of pulmonary inflammation, play critical roles in COPD pathogenesis by releasing cytokines and chemokines that recruit neutrophils and lymphocytes into the airways (Min et al., 2025). Cigarette smoke exposure alone is known to trigger macrophage activation and induce the production of tumor necrosis factor (TNF), interleukin (IL)-1β, and other proinflammatory mediators via the nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) pathway (Lugg et al., 2022; Murugan and Peck, 2009). NF-κB is a pivotal transcription factor that regulates genes involved in inflammation, apoptosis, and immune activation. Persistent NF-κB activation has been implicated in both the initiation and maintenance of chronic inflammation in COPD patients (Schuliga, 2015). These smoke-induced inflammatory responses are closely linked to epithelial injury, oxidative stress, macrophage activation, and progressive structural remodeling of the lung parenchyma. Therefore, environmental pollutants that disturb immune homeostasis or amplify oxidative and inflammatory stress may influence the severity of smoke-induced COPD-like pathology. However, whether PFOS or PFOA exposure is associated with greater cigarette smoke-induced lung injury in relation to enhanced inflammatory signaling remains unclear.

Recent experimental studies have suggested that PFOS and PFOA can activate inflammatory signaling in macrophages, upregulate cytokine production, and promote oxidative stress in multiple organs (Wang et al., 2021; Yang et al., 2021). These findings provide a biological rationale for investigating whether PFOS/PFOA exposure may intersect with cigarette smoke-induced lung injury. Cigarette smoke can prime the lung microenvironment through epithelial damage, basal inflammatory activation, oxidative stress, and macrophage recruitment, whereas PFAS-related immunotoxicity may further perturb macrophage inflammatory responses and cytokine production. Given that macrophages are central to COPD pathophysiology (Ishikawa et al., 2015), it is plausible that PFOS and PFOA exposure may be associated with enhanced smoke-induced inflammatory responses and more severe COPD-like lung injury under combined exposure conditions.

In this study, we aimed to investigate whether chronic drinking-water exposure to PFOS or PFOA was associated with greater cigarette smoke-induced lung dysfunction and inflammatory injury, and to determine whether PFOS/PFOA exposure was associated with altered smoke-induced inflammatory responses in the lung. Using a mouse model of chronic smoke exposure combined with concurrent PFOS or PFOA exposure in drinking water, we evaluated lung function, histopathology, and transcriptomic alterations. We further assessed NF-κB-related inflammatory signaling in vivo and in vitro by quantitative PCR and Western blot analyses. Our findings suggest that PFOS and PFOA may be associated with enhanced cigarette smoke-induced inflammation and tissue injury, highlighting the potential contribution of drinking-water PFAS exposure as a modifier of smoke-induced pulmonary injury.

2. Materials and methods

2.1. Animals and treatment

Healthy male C57BL/6 wild-type mice (4–6 weeks old, 18–22 g) were purchased from Charles River Laboratory (Beijing, China). All mice were housed under specific pathogen-free (SPF) conditions in a temperature- (22 ± 2 °C) and humidity- (55 ± 10%) controlled environment with a 12/12-h light/dark cycle and provided with standard rodent chow (Shuyu Biotechnology Co., Ltd., Shanghai, China) and water ad libitum. Only male C57BL/6 mice were used in this study. This single-sex design was adopted to reduce variability related to sex-dependent physiological differences and to maintain consistency with the established cigarette smoke-induced COPD-like mouse model used in our laboratory. After one week of acclimation, the mice were randomly assigned to six groups using a random number generator: (1) Control, (2) PFOS, (3) PFOA, (4) cigarette smoke (CS), (5) CS + PFOS, and (6) CS + PFOA (n = 15 mice per group). The group sizes were determined based on preliminary data and feasibility considerations for this exploratory animal study. In each group, 10 mice were used for histological and morphometric analyses, and five mice were used for pulmonary function testing. For RNA-seq analysis, whole-lung tissues from three mice per group were collected after pulmonary function measurement, with each mouse representing one independent biological replicate. No additional animals were exposed beyond the planned sample size, and no mortality occurred during the exposure period.

PFOS and PFOA were provided through drinking water at a concentration of 2 mg/L to model chronic drinking-water PFAS exposure (Fig. 1A) and to maintain a consistent external exposure design for parallel assessment of each legacy PFAS. This design was not intended to establish equivalent internal PFOS and PFOA burdens, as these two compounds differ in toxicokinetic properties, including elimination kinetics and tissue distribution. Based on an average water consumption of approximately 4 mL/day per adult mouse, this dosing strategy resulted in an estimated daily intake of approximately 8 μg of each PFAS. For a mouse with an average body weight of 25 g, this corresponds to an estimated nominal external dose of approximately 0.32 mg/kg/day. Plasma, lung tissue, and cellular concentrations of PFOS or PFOA were not measured in the present study; therefore, the calculated doses represent estimated external exposure levels rather than internally measured PFAS burdens. This dose was selected based on previous murine studies investigating the immunotoxic effects of PFAS (Roth et al., 2021) and was used to examine biological interaction with cigarette smoke under controlled experimental conditions rather than to directly mimic human inhalational or pulmonary exposure.

Fig. 1.

Drinking-water PFOS and PFOA exposure was associated with enhanced cigarette smoke-induced airflow limitation in mice. (a) Schematic illustration of the experimental design and exposure protocol. (b–f) Pulmonary function parameters measured at the end of exposure, including forced expiratory volume in one second to forced vital capacity ratio (FEV₁/FVC%) (b), forced vital capacity (FVC) (c), inspiratory capacity (IC) (d), forced expiratory volume in one second (FEV₁) (e), and forced expiratory flow at 25–75% of FVC (FEF₂₅–₇₅%) (f). Data are presented as mean ± SD (n = 5 per group). Statistical analyses were performed using two-way ANOVA with cigarette smoke exposure and PFAS treatment as independent factors, followed by Šídák's multiple comparisons test. The main effects of cigarette smoke exposure, PFAS treatment, and the cigarette smoke × PFAS interaction were evaluated. Statistical significance is indicated as *p < 0.05; **p < 0.01; ***p < 0.001.

2.2. Murine COPD model

The mice were placed in a whole-body exposure chamber measuring 35 × 35 × 20 cm and exposed to smoke generated from eight commercial cigarettes (Double Happiness; 13 mg tar and 1.2 mg nicotine per cigarette; Guangdong China Tobacco Industrial Co., Ltd., Guangdong, China) twice daily, 5 days per week, for 12 weeks. For each cigarette, smoke generation consisted of a 20-s ignition period followed by a 120-s burning period. The generated smoke was retained in the closed exposure chamber for 15 min, after which the chamber was ventilated for 15 min to remove residual smoke before the next cigarette exposure cycle. Each exposure session consisted of eight sequential cigarette exposure cycles. Control mice were exposed to filtered air under otherwise identical conditions.

2.3. Lung function measurement

After 12 weeks of exposure, pulmonary function was assessed via a FlexiVent small-animal ventilator (SCIREQ, Montreal, Canada). Forced expiratory volume in 1 s (FEV₁), forced vital capacity (FVC), the FEV₁/FVC ratio, inspiratory capacity (IC), and the forced expiratory flow at 25–75% of forced vital capacity (FEF₂₅–₇₅%) were measured to evaluate airflow limitation and lung compliance. Pulmonary function analysis was performed in five mice per group.

2.4. Histological analysis

Lung tissues were fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned at a thickness of 5 μm. Sections were stained with hematoxylin and eosin (H&E) to assess alveolar structure and emphysema. The mean linear intercept (MLI), mean alveolar diameter (MAD), and mean alveolar area (MAA) were quantified using ImageJ software (NIH, Bethesda, MD, USA). Histological and morphometric analyses were performed in 10 mice per group. A representative schematic illustration of the morphometric measurements, including MLI, MAD, and MAA, is provided in Supplementary Fig. S1.

2.5. RNA extraction and transcriptomic analysis

After pulmonary function measurement, mice were euthanized by overdose anesthesia with sodium pentobarbital (150 mg/kg, intraperitoneal injection), and death was confirmed by cessation of breathing and heartbeat before tissue collection. Lung tissues were then collected for histological and molecular analyses. For histological analysis, lung tissues from 10 mice per group were fixed in 4% paraformaldehyde, embedded in paraffin, and used for H&E staining and morphometric analysis. For RNA-seq analysis, whole-lung tissues from three of the five mice used for pulmonary function measurement in each group were collected, snap-frozen in liquid nitrogen, and stored at −80 °C until RNA extraction. Each mouse used for RNA-seq represented one independent biological replicate.

Total RNA was extracted from whole-lung tissues used for RNA-seq analysis using TRIzol reagent (Invitrogen, USA). The RNA quality was verified via a NanoDrop spectrophotometer and an Agilent Bioanalyzer. Only RNA samples with RNA integrity number (RIN) values greater than 7.0 were used for library construction. Libraries were prepared with the Illumina TruSeq RNA Library Prep Kit and sequenced on an Illumina NovaSeq platform. Sequencing generated an average of 25.6 million reads per sample, with Q30 values of 93.5% and an average mapping rate of 96.8%. Clean reads were aligned to the mouse reference genome GRCm39 (mm39) using STAR (v2.7.10a), and gene annotation was performed using GENCODE M33 (Ensembl release 110).

Differentially expressed genes (DEGs) were identified using DESeq2 with a design formula of ∼ group (|log₂FC| > 1, adjusted p < 0.05). No batch correction was applied because all libraries were prepared and sequenced in the same batch. GO and KEGG enrichment analyses were performed via the clusterProfiler R package. Enrichment analyses were performed using all significant DEGs from each comparison unless otherwise specified.

2.6. Preparation of cigarette smoke extract (CSE) and co-treatment with PFAS

Cigarette smoke extract (CSE) was freshly prepared before each experiment as previously described, with minor modifications. Briefly, smoke from one cigarette (Guangdong China Tobacco Industrial Co., Ltd., Guangdong, China) was bubbled through 10 mL of pre-warmed serum-free DMEM using a peristaltic pump at a constant flow rate of 0.4 L/min. The resulting suspension was defined as 100% CSE, filtered through a 0.22 μm membrane to remove particles and bacteria, and used immediately after preparation. The optical density at 320 nm (OD₃₂₀ = 0.74 ± 0.05) was measured to monitor batch-to-batch reproducibility. For in vitro assays, 100% CSE was diluted with complete DMEM to a final working concentration of 5%, a condition that induced moderate inflammatory activation in RAW264.7 cells without overt cytotoxicity. PFOS and PFOA were dissolved in DMSO and added to the culture medium at a final concentration of 10 μM (final DMSO concentration < 0.1%). This concentration was selected based on previous experimental toxicology studies using PFOS or PFOA in cultured cells and was intended for mechanistic exploration of CSE-associated inflammatory responses rather than direct environmental exposure equivalence. Cellular concentrations of PFOS and PFOA were not measured; therefore, this concentration represents a nominal in vitro exposure concentration rather than an internally measured cellular PFAS burden.

RAW264.7 cells were assigned to six treatment groups: Control, PFOA, PFOS, CSE, CSE + PFOA, and CSE + PFOS. Cells were exposed to CSE alone (5%), PFOS or PFOA alone (10 μM), or co-treated with CSE and PFAS (5% CSE + 10 μM PFOS or PFOA) for 24 h as a simplified macrophage-like inflammatory model. Control cells received vehicle treatment only.

After exposure, cells were collected for RNA and protein extraction for subsequent qPCR and Western blot analyses. The in vitro qPCR analysis shown in Fig. 4B–E was performed using RNA extracted from RAW264.7 cells after CSE/PFAS treatment. For qPCR analysis, each treatment condition included three independent biological replicates, and each qPCR reaction was performed in technical triplicate. For Western blot analysis, three independent biological replicates were analyzed for each treatment condition. CSE was freshly prepared for each independent experiment.

Fig. 4.

PFOS and PFOA exposure was associated with enhanced NF-κB-related inflammatory signaling in cigarette smoke-exposed mouse lungs and CSE-treated RAW264.7 cells. (a) Heatmap showing the expression of nuclear factor-κB (NF-κB) pathway-related genes in lung tissues from the cigarette smoke (CS), CS plus perfluorooctane sulfonate (CS + PFOS), and CS plus perfluorooctanoic acid (CS + PFOA) groups, as determined by transcriptomic analysis.(b–e) Quantitative real-time PCR (qPCR) analysis of inflammatory gene expression in RAW264.7 cells treated with cigarette smoke extract (CSE, 5%), PFOS or PFOA (10 μM), or their combination for 24 h. The treatment groups were Control, PFOS, PFOA, CSE, CSE + PFOS, and CSE + PFOA. The analyzed genes included Tnf (b), Il1b (c), Cxcl1 (d), and Tnfaip3 (e). (f, g) Representative Western blot images (f) and quantitative analysis of p-p65/p65 relative expression (g) in RAW264.7 cells under the indicated treatment conditions. Bulk RNA-sequencing analysis included n = 3 mice per group. qPCR and Western blot analyses were performed using n = 3 independent biological replicates per treatment condition. Data are presented as mean ± SD. Statistical analyses of qPCR and Western blot data were performed using two-way ANOVA with CSE exposure and PFAS treatment as independent factors, followed by Šídák's multiple comparisons test. The main effects of CSE exposure, PFAS treatment, and the CSE × PFAS interaction were evaluated. Statistical significance is indicated as *p < 0.05; **p < 0.01; ***p < 0.001.

2.7. Quantitative real-time PCR

Total RNA was extracted from RAW264.7 cells after in vitro CSE/PFAS treatment and from mouse lung tissues used for in vivo validation. RNA was reverse-transcribed into cDNA using the PrimeScript RT reagent kit (Takara, Kusatsu, Japan). qPCR was performed using TB Green Premix Ex Taq (Takara, Kusatsu, Japan) on an ABI 7500 Real-Time PCR System. Gapdh served as an internal reference. Relative expression was calculated using the 2⁻ΔΔCt method. The primer sequences are listed in Table 1. The qPCR targets were selected as representative NF-κB-related inflammatory genes and macrophage inflammatory response markers highlighted by the transcriptomic analysis, rather than as an exhaustive validation of all RNA-seq-derived DEGs.

Table 1.

Sequence of primers used for real-time PCR.

Gene	Sequence of primer(5′ - > 3′)
	Forward	Reverse
Tnf	CTGGATGTCAATCAACAATGGGA	ACTAGGGTGTGAGTGTTTTCTGT
Il1b	GCAACTGTTCCTGAACTCAACT	ATCTTTTGGGGTCCGTCAACT
Cxcl1	CAAGGCTGGTCCATGCTCC	TGCTATCACTTCCTTTCTGTTGC
Tnfaip3	CCTTGCCATTCCCCTACTTGG	ACTTCGTATTTGGATGCAGAACA
Gapdh	TGGATTTGGACGCATTGGTC	TTTGCACTGGTACGTGTTGAT

2.8. Western blot

Protein was extracted from RAW264.7 cells after in vitro CSE/PFAS treatment and from mouse lung tissues used for in vivo validation using RIPA buffer supplemented with protease and phosphatase inhibitors. Equal amounts of protein were separated by SDS–PAGE and transferred to polyvinylidene difluoride (PVDF) membranes. After blocking with 5% skim milk, the membranes were incubated overnight at 4 °C with primary antibodies against NF-κB p65, phospho-p65, and GAPDH, followed by incubation with HRP-conjugated secondary antibodies. The protein bands were visualized using an enhanced chemiluminescence (ECL) system.

The same primary antibodies were used for Western blot analyses of RAW264.7 cell lysates and mouse lung tissue lysates. The primary antibodies used were as follows: anti-NF-κB p65 (Proteintech Group, Wuhan, China; catalog no. 80979–1-RR; dilution 1:2000), anti-phospho-p65 (Proteintech Group, Wuhan, China; catalog no. 82335–1-RR; dilution 1:2000), and anti-GAPDH (Proteintech Group, Wuhan, China; catalog no. 60004–1-Ig; dilution 1:5000). The horseradish peroxidase (HRP)-conjugated secondary antibodies used were Multi-rAb® HRP-Goat Anti-Rabbit Recombinant Secondary Antibody (H + L) (Proteintech Group, Wuhan, China; catalog no. RGAR001; dilution 1:5000) and Multi-rAb® HRP-Goat Anti-Mouse Recombinant Secondary Antibody (H + L) (Proteintech Group, Wuhan, China; catalog no. RGAM001; dilution 1:5000).

2.9. Statistical analysis

All data are presented as the mean ± SD. Statistical analyses were performed using GraphPad Prism 9.0. For in vivo experiments, data were analyzed using two-way analysis of variance (ANOVA), with cigarette smoke exposure (absence or presence of CS) and PFAS exposure (None, PFOS, or PFOA) as the two fixed factors. The main effects of CS exposure and PFAS exposure, as well as the CS × PFAS interaction term, were evaluated. For in vitro experiments, data were analyzed using two-way ANOVA, with CSE exposure and PFAS treatment as fixed factors. The main effects of CSE exposure and PFAS treatment, as well as the CSE × PFAS interaction term, were evaluated. When appropriate, Šídák's multiple comparisons test was performed for pairwise comparisons among treatment groups following two-way ANOVA. A value of p < 0.05 was considered statistically significant.

Sample sizes were reported in the relevant Methods sections and figure legends. Pulmonary function was assessed in five mice per group, histological and morphometric analyses were performed in 10 mice per group, and RNA-seq analysis was performed using whole-lung tissues from three mice per group. For in vitro qPCR and Western blot analyses, three independent biological replicates were included for each treatment condition.

3. Results

3.1. PFOS and PFOA co-exposure exacerbated cigarette smoke-induced airflow limitation in mice

The experimental mouse exposure protocol is shown in Fig. 1A. Lung function was assessed in six groups of mice: Control, PFOA, PFOS, CS, CS + PFOS, and CS + PFOA. As shown in Fig. 1B-F, CS exposure significantly impaired pulmonary function, characterized by a reduction in the FEV₁/FVC ratio. Compared with CS exposure alone, combined exposure with either PFOA or PFOS further reduced FEV₁/FVC%, indicating enhanced airflow limitation under co-exposure conditions.

PFOS or PFOA exposure alone did not significantly alter pulmonary function compared with controls, suggesting that the adverse effects of PFAS exposure were primarily manifested under cigarette smoke conditions in this model. Although FVC, IC, FEV₁, and FEF₂₅–₇₅% did not show significant differences between CS and CS + PFAS groups after multiple-comparison correction, the consistent trend toward reduced values suggests potential alterations in airflow-related parameters under combined exposure conditions. Detailed statistical comparisons were performed using Šídák-adjusted multiple comparisons following two-way ANOVA.

3.2. PFOS and PFOA exposure was associated with greater structural lung injury in cigarette smoke-exposed mice

Histopathological examination of lung tissues from the six groups via H&E staining revealed distinct structural alterations. The CS group exhibited alveolar dilation and wall thinning, which were further enhanced in the CS + PFOS and CS + PFOA groups, accompanied by more pronounced alveolar enlargement and structural alterations (Fig. 2A, C–E). In contrast, no obvious structural abnormalities were observed in the control, PFOS, or PFOA groups, suggesting that PFOS and PFOA exposure alone did not cause apparent lung structural alterations under the current experimental conditions. Moreover, co-exposure to cigarette smoke and PFOS or PFOA was accompanied by increased inflammatory cell infiltration within the lung parenchyma, as observed histologically.

Fig. 2.

PFOS and PFOA co-exposure was associated with enhanced cigarette smoke-induced structural lung alterations and inflammatory changes in mice. (a, b) Representative H&E-stained lung sections from the Control, PFOA, PFOS, CS, CS + PFOA, and CS + PFOS groups. Cigarette smoke exposure induced alveolar wall thinning and enlargement, whereas co-exposure to PFOS or PFOA was associated with more pronounced alveolar structural alterations and inflammatory cell infiltration (red arrows). Scale bars: 100 μm (a) and 50 μm (b). (c–e) Quantitative analysis of emphysema-related morphometric parameters, including mean linear intercept (MLI) (c), mean alveolar diameter (MAD) (d), and mean alveolar area (MAA) (e). Statistical analyses were performed using two-way ANOVA with cigarette smoke exposure and PFAS exposure as independent factors, followed by Šídák's multiple comparisons test. Statistical significance is indicated as *p < 0.05; **p < 0.01; ***p < 0.001.

3.3. PFOS and PFOA exposure was associated with altered immune-related transcriptional signatures compared with cigarette smoke exposure alone

Bulk lung transcriptomic analysis was performed to investigate molecular changes associated with PFAS co-exposure under cigarette smoke exposure conditions. Compared with the CS group, the CS + PFOA group showed 2232 differentially expressed genes (DEGs), including 1730 upregulated and 502 downregulated genes. The CS + PFOS group showed 4153 DEGs compared with the CS group, including 2435 upregulated and 1718 downregulated genes.

Heatmap visualization demonstrated increased expression of genes associated with antigen presentation, lysosomal processing, inflammatory activation, and adaptive immune responses in the CS + PFOS and CS + PFOA groups compared with the CS group (Fig. 3A, B). KEGG pathway enrichment analysis further identified enrichment of immune- and inflammation-related pathways in both comparisons (Fig. 3C, D). Overlap analysis identified 1543 shared DEGs between the CS + PFOA versus CS and CS + PFOS versus CS comparisons, suggesting partially overlapping transcriptional responses associated with PFAS co-exposure under cigarette smoke exposure conditions.

Fig. 3.

PFOS and PFOA exposure was associated with altered immune-related transcriptional signatures and enrichment of inflammatory pathways in cigarette smoke-exposed mouse lungs. (a, b) Heatmap visualization of selected differentially expressed genes (DEGs) in lung tissues from the cigarette smoke plus perfluorooctanoic acid (CS + PFOA) versus cigarette smoke (CS) comparison (a) and the cigarette smoke plus perfluorooctane sulfonate (CS + PFOS) versus CS comparison (b). (c, d) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of DEGs from the CS + PFOA versus CS comparison (c) and the CS + PFOS versus CS comparison (d).

3.4. PFOS and PFOA exposure was associated with enhanced NF-κB-related inflammatory signaling in vivo and in vitro

We first analyzed whole-lung transcriptomic data from mice after in vivo exposure. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis comparing the CS + PFOS and CS + PFOA groups with the CS group identified significant enrichment of the NF-κB signaling pathway (Fig. 3C, D). Heatmap visualization of the transcriptomic dataset (Fig. 4A) showed increased expression of multiple NF-κB-related genes, including Tnf, Il1b, Cxcl1, and Tnfaip3 (Meibers et al., 2023), in the CS + PFOS and CS + PFOA groups compared with the CS group, suggesting altered inflammatory transcriptional responses in vivo. These genes were selected for subsequent qPCR validation because they represent NF-κB-related inflammatory mediators highlighted by the transcriptomic analysis, rather than because they were intended to represent all RNA-seq-derived DEGs.

To further validate the transcriptomic findings in mouse lung tissues, we performed additional in vivo validation experiments. Western blot analysis showed increased p-p65/p65 ratios in the CS + PFOS and CS + PFOA groups compared with the CS group after multiple-comparison correction (Supplementary Fig. S2A, B). Consistently, qPCR analysis of lung tissues showed altered expression of representative NF-κB-related inflammatory genes, with significant increases observed for Il1b and Cxcl1, whereas changes in Tnf and Tnfaip3 were also detected under specific exposure conditions (Supplementary Fig. S2C–F). These in vivo validation results were consistent with the RNA-seq-based observation that drinking-water PFOS/PFOA exposure was associated with altered NF-κB-related inflammatory responses in cigarette smoke-exposed mouse lungs.

We next performed in vitro experiments using RAW264.7 cells exposed to cigarette smoke extract (CSE) and PFAS. Quantitative real-time PCR (qPCR) analysis showed that PFOS or PFOA alone produced limited changes in the expression of NF-κB-related inflammatory genes. However, combined exposure with CSE and PFAS was associated with altered expression of selected NF-κB-related inflammatory genes compared with CSE alone (Fig. 4B–E). Western blot analysis further showed increased p-p65/p65 ratios in RAW264.7 cells co-treated with CSE and PFOS or PFOA (Fig. 4F, G).

Together, these results suggest that PFOS and PFOA were not strong independent activators of NF-κB-related inflammatory responses under the present experimental conditions, but were associated with enhanced CSE-induced inflammatory responses in this macrophage-like cell model. These in vitro findings were consistent with the inflammatory transcriptional signatures observed in vivo in the CS + PFAS groups. (Hong et al., 2021).

4. Discussion

This study shows that chronic drinking-water exposure to PFOS or PFOA was associated with greater cigarette smoke-induced in mice, supporting the possibility that external PFAS exposure through drinking water may influence smoke-related pulmonary inflammatory responses under controlled experimental conditions. In our model, PFOS/PFOA co-exposure under cigarette smoke conditions was associated with additional alterations in selected lung function parameters, alveolar structural features, and airway inflammatory responses compared with cigarette smoke exposure alone. In parallel, lung transcriptomic analysis and in vitro macrophage experiments were consistent with enhanced NF-κB-related inflammatory responses under combined exposure conditions, suggesting a potential association between macrophage-related inflammatory responses and the observed lung injury phenotype.

The NF-κB pathway is a well-established master regulator of inflammation in COPD pathogenesis (Schuliga, 2015; Barnes, 2016). Our transcriptomic data revealed an inflammatory transcriptional signature in the lungs of CS + PFAS mice, characterized by the upregulation of inflammatory genes including Tnf, Il1b, and Cxcl1 (Barnes, 2006). In vitro, PFOS or PFOA alone caused only modest changes in NF-κB-related signaling in RAW264.7 cells, whereas both compounds were associated with increased p-p65/p65 ratios and higher expression of selected inflammatory genes under CSE exposure conditions. These data suggest that PFOS and PFOA may be more appropriately interpreted as modifiers of cigarette smoke-triggered inflammatory signaling rather than as strong independent initiators under the present experimental conditions. Notably, PFOS or PFOA alone did not cause marked pathology, which further supports the interpretation that their effects were more evident in the context of cigarette smoke co-exposure. One plausible interpretation is that cigarette smoke may establish a pre-existing inflammatory environment, within which PFOS and PFOA could further influence inflammatory responses and contribute to more pronounced smoke-related pathological changes.

Changes in Tnfaip3 (A20), a classical NF-κB feedback regulator, were observed in the transcriptomic and molecular analyses; however, whether these alterations represent compensatory feedback responses or contribute to inflammatory persistence during PFAS co-exposure requires further investigation (Sohn et al., 2014; Galatola et al., 2013). Sustained inflammatory responses may contribute to macrophage-associated tissue injury responses and impaired resolution of inflammation, thereby promoting COPD-like pathological changes. These observations are consistent with previous studies reporting that PFOS and PFOA exposure can induce oxidative stress and NF-κB-related responses in hepatocytes and endothelial cells (Connolly et al., 2025; Xu et al., 2019). Moreover, an epidemiological analysis based on the NHANES database indicated that serum PFAS levels, including PFOS and PFOA, were associated with higher prevalence of COPD (Pan et al., 2024). Our findings provide experimental support consistent with these epidemiological observations by showing that drinking-water PFOS/PFOA exposure was associated with altered pulmonary inflammatory responses in a smoke-exposed setting. Notably, altered expression of immune-related mediators such as Ccl19 may indicate changes in immune-related signaling within the lung microenvironment, although this possibility requires further validation using cell-specific approaches (Cheng et al., 2018).

These findings support the possibility that PFOS and PFOA may represent environmental factors influencing smoke-related lung injury within the scope of the present experimental model. Considering the widespread exposure to PFAS and their long biological half-life (Goodrich et al., 2022), PFAS exposure may represent a relevant environmental factor for further investigation in populations exposed to cigarette smoke. However, the present findings should not be interpreted as direct evidence for human risk thresholds or specific regulatory actions, because the study was conducted using a single drinking-water exposure concentration in male mice and relied primarily on associative mechanistic evidence. Rather, these results highlight the need for further studies evaluating pollutant–smoke co-exposure in chronic lung disease.

This study has certain limitations. PFOS and PFOA were administered via drinking water, which models systemic exposure rather than direct pulmonary exposure. Therefore, the present study is more appropriately interpreted as assessing the potential modifying effect of drinking-water PFOS/PFOA exposure on smoke-induced lung injury rather than directly mimicking real-world inhalational exposure scenarios. In addition, plasma and lung tissue concentrations of PFOS and PFOA were not measured; therefore, the administered concentration represents external exposure rather than internally measured PFAS burden. Given that PFOS and PFOA exhibit distinct toxicokinetic properties, including differences in absorption, distribution, and elimination, the present study cannot determine whether the same nominal drinking-water concentration resulted in equivalent internal exposure levels. Therefore, comparisons between PFOS- and PFOA-associated effects should be interpreted cautiously. The exposure duration was relatively short, and long-term effects require further investigation. Moreover, only male mice were included in this study to maintain consistency with commonly used cigarette smoke-induced COPD-like mouse models and to minimize biological variability during the initial evaluation of PFAS–smoke interactions. However, given the known sex-dependent differences in PFAS absorption, distribution, metabolism, and elimination, the present findings cannot determine whether similar responses would occur in female mice. Future studies including both sexes are warranted to clarify potential sex-specific effects of PFOS and PFOA exposure. Furthermore, the in vitro experiments were performed in RAW264.7 cells as a simplified macrophage-like inflammatory model and therefore do not fully recapitulate the pulmonary macrophage populations involved in COPD. In addition, only a single PFOS/PFOA concentration was tested in the in vitro experiments; therefore, future concentration-response studies are needed to determine whether the observed inflammatory responses occur in a dose-dependent manner. The 10 μM PFOS/PFOA concentration used in vitro was selected for mechanistic exploration based on previous experimental toxicology studies and should not be interpreted as directly equivalent to physiological human serum PFAS exposure levels. Although our data are consistent with the involvement of NF-κB-related inflammatory signaling, the mechanistic evidence remains associative, and other pathways, such as JAK-STAT or inflammasome signaling, may also contribute. Because functional NF-κB inhibition, p65 knockdown, or genetic manipulation experiments were not performed, NF-κB activation should be interpreted as an associated inflammatory signaling feature rather than a proven causal mediator of PFAS-modified smoke-induced lung injury. Future single-cell and functional studies are warranted to delineate cell-specific mechanisms. Another limitation is that biochemical confirmation of cigarette smoke exposure, such as serum cotinine or carboxyhemoglobin measurement, was not performed. Blood or other biological samples specifically collected for these measurements were not available from the completed animal experiment, and therefore retrospective confirmation of internal smoke exposure biomarkers could not be performed. In addition, real-time monitoring of chamber airflow, total particulate matter, and carbon monoxide levels was not available; therefore, the smoke exposure intensity was defined based on the standardized cigarette number, exposure frequency, chamber size, and exposure duration rather than measured internal exposure biomarkers.

Despite these limitations, the study provides experimental evidence supporting associations between pollutant–smoke co-exposure and COPD-like lung injury. In summary, chronic drinking-water PFOS and PFOA exposure was associated with greater cigarette smoke-induced and was consistent with enhanced NF-κB-related inflammatory responses. These findings support a potential role for drinking-water PFOS/PFOA exposure as a factor influencing smoke-associated pulmonary injury, while the underlying mechanisms remain to be functionally established. These findings provide a basis for further investigation into how persistent environmental pollutants may influence chronic airway disease. Future studies should explore the long-term and cell-specific mechanisms by which PFOS and PFOA influence smoke-induced pulmonary injury.

CRediT authorship contribution statement

Manyi Pan: Investigation, Methodology. Yuxin Zou: Investigation, Methodology. Yuqing Lv: Supervision, Resources, Project administration. Hongbin Wang: Supervision, Resources, Project administration. Chengcheng Hou: Formal analysis. Haiyang Zhang: Validation. Tianyu Zhou: Writing – review & editing, Visualization. Lifeng Yan: Writing – review & editing, Visualization. Weining Xiong: Supervision, Resources, Project administration. Xiahui Ge: Supervision, Resources, Project administration. Hui Yin: Funding acquisition. Huaqi Guo: Supervision, Resources, Project administration.

Informed consent statement

Not applicable.

Institutional review board statement

All procedures were approved by the Animal Care Committee of Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine (No. SH9H-2020-A734–1).

Funding

This study was supported by the National Natural Science Foundation of China (grant numbers 82400035, 82470031, and 82090015), the National Science and Technology Major Project for Non-communicable Chronic Diseases (grant number 2024ZD0528500), and the Shanghai Municipal Health Commission Health Industry Research Special Young Project (grant number 20234Y0091).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

Supplementary Material 1

Supplementary Material 2