PFOS-induced black liver and hepatic aging in mice: A new paradigm in environmental hepatotoxicology

Abstract

Perfluorooctanesulfonic acid (PFOS), a persistent environmental contaminant still imported and widely used in industrial and consumer products, poses poorly defined risks under chronic occupational exposure, with limited research on its role in environmental liver aging. Here, we identify previously unrecognized hepatotoxicity of PFOS in mice exposed to an occupationally relevant dose (10 mg/kg) over subacute (25-day) and chronic (105-day) periods. While both exposures induced mild metabolic-associated steatotic liver disease (MASLD) without fibrosis, only chronic exposure produced a distinct “black liver” phenotype marked by pigmentation, hepatic atrophy, and aging. The black liver discoloration was associated with hepatic congestion and the consequential blood stasis likely contributed to the darkened liver appearance. PFOS also disrupted bile canaliculi by downregulating ZO-1 and mislocalizing BSEP and MRP2, resulting in cholestasis mimicking the Dubin–Johnson Syndrome which displays black liver appearance. Most notably, PFOS activated aging hallmarks, including oxidative stress (superoxide, 4-HNE), genomic instability (γH2AX, 8-oxo-dG), cellular senescence (SA-β-Gal, p21^WAF1/Cip1), and sirtuin reduction (SIRT1 and SIRT7). Mechanistically, chronic PFOS downregulated autophagy complexes (ULK1-FIP200-ATG13 and ATG14L-Beclin1-Vps34) and reduced LC3B-I/II conversion, indicating impaired autophagosome formation and blocked flux. Pharmacological restoration of autophagy in PFOS-treated AML12 cells ameliorated intracellular lipid droplets, canalicular disruption, oxidative stress and aging pigment buildup, suggesting PFOS as a silent driver of irreversible liver aging at least through autophagy suppression. Finally, transcriptomics of PFOS-exposed human liver spheroids recapitulated aging and cholestasis, underscoring translational relevance. These findings redefine the occupational toxicological profile of PFOS, supporting the inclusion of congestion, pigment, and aging biomarkers in occupational PFAS assessments.

1. Introduction

Per- and polyfluoroalkyl substances (PFAS) are a large class of synthetic chemicals defined by their highly stable carbon–fluorine bonds, which make them exceptionally resistant to environmental degradation, earning them the nickname “forever chemicals.” Although legacy perfluorooctanesulfonic acid (PFOS) and perfluorooctanoic acid (PFOA) have been phased out of production in the US, they are still imported and remain widely present in many everyday consumer products such as non-stick cookware, water-repellent fabrics, and firefighting foams; PFAS compounds are now pervasive in water, soil, food, and household environments (Panieri et al., 2022; Manoochehri et al., 2025; Perez et al., 2023). Contaminated groundwater in some U.S. regions may take over 40 years to be fully cleared of PFAS contamination (Jensen et al., 2024).

Among the thousands of PFAS compounds, PFOS and PFOA are the most extensively studied long-chain variants (8 carbon, see Fig. 1A for PFOS chemical structure) due to their environmental persistence, bioaccumulation, and established toxicity. These compounds are detectable in the blood of nearly all individuals worldwide (Ho et al., 2022). However, occupational groups, including chemical plant workers, firefighters, and fishery employees, face substantially higher exposures, with serum PFOS concentrations reaching up to 62,898 ng/mL (Zhou et al., 2014; Gao et al., 2015), compared to 4.3 up to 30.4 ng/mL in the general population (Sonnenberg et al., 2023). Despite this stark exposure disparity, the health effects of chronic, high-level, occupational PFOS exposure remain poorly characterized, particularly with respect to long-term hepatic injury. The liver, as the central organ for xenobiotic metabolism, is a primary target of PFAS toxicity. Epidemiological and preclinical studies have linked PFOS and PFOA to elevated serum alanine aminotransferase (ALT), and nonalcoholic fatty liver disease (NAFLD), now termed metabolic dysfunction-associated steatotic liver disease (MASLD). PFOS in particular exhibits potent hepatotoxicity, contributing to MASLD by disrupting lipid metabolism and homeostasis, and is strongly associated with liver cancer risk. Its strong binding to serum albumin and lipoproteins facilitates hepatic accumulation with minimal metabolism, reinforcing its relevance to human health (Wang et al., 2014; Wan et al., 1820; Jin et al., 2020; Goodrich et al., 2022; Maerten et al., 2024).

Fig. 1. Subacute PFOS exposure drives MASLD, hepatomegaly and hepatic proliferation.

(A-B) Chemical structure of PFOS (A) and in vivo experimental design (B) for the subacute PFOS exposure model (10 mg/kg body weight, i.p.). (C-G) Representative liver images from mice subjected to subacute PFOS exposure or left untreated. Scale bar = 1 cm (C), H&E staining showing normal liver histology (D), body weight (E), liver weight (F), and liver-to-body weight ratio (G). (H-J) Liver tissue sections stained with Oil Red O (ORO) (H), Sirius Red (I), and Ki67 (J); quantifications shown below each image. Scale bar = 50 μm. n.s.: not significant; ***p < 0.001 by student’s t test. Untreated mice (n = 4); subacute PFOS-treated mice (n = 3); representative images quantified (n = 5–20). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

In addition to MASLD, mechanistic studies have shown PFOS disrupts bile acid homeostasis by interfering with nuclear receptor pathways and bile-metabolizing enzymes (Sinal et al., 2000; Behr et al., 2020; Vujic et al., 2024). Bile is synthesized by hepatocytes and excreted into bile canaliculi, specialized apical membrane structures enriched in key hepatobiliary transporters such as BSEP and MRP2 (Stapelbroek et al., 2010; Wang et al., 2022). Dysfunction of these transporters, as seen in genetic disorders like Dubin-Johnson Syndrome (DJS; OMIM #237500), results in impaired bile excretion and characteristic black liver discoloration (Oude Elferink et al., 2006; Strassburg, 2010). However, whether PFOS disrupts canalicular structure and transporter localization independent of gene regulation remains unclear, especially at occupational exposure levels.

In addition to metabolic disruption, PFOS may modulate autophagy, which is a key cellular recycling process essential for degrading damaged organelles, denatured proteins and lipid overload (Weng et al., 2020; Yao et al., 2016; Tang et al., 2018; Rahman et al., 2023; Noda and Inagaki, 2015). Autophagy initiates with the formation of a double membrane autophagosome, which engulfs cellular waste before fusing with lysosomes and the waste in the fused autolysosomes is degraded by lysosomal enzymes (Mizushima, 2005; Mizushima, 2009). While PFOS has shown both inhibitory and stimulatory effects on autophagy in context-dependent studies, its impact on autophagy function in occupational models is not well understood.

To address these mechanistic gaps in response to occupational exposure levels, we utilized a controlled mouse model simulating occupational PFOS exposure to compare subacute and chronic effects on liver. Our study reveals a previously unrecognized hepatotoxic profile unique to chronic exposure to PFOS, marked by black liver discoloration associated with hepatic congestion, and by liver aging primarily driven by autophagy suppression. These findings provide important insights for occupational health surveillance and underscore the need to incorporate liver aging, vascular, and pigment biomarkers into future PFAS risk assessments in the population of occupational workers.

2. Materials and methods

2.1. Animal studies

All animal procedures were performed in accordance with the Guide for the Care and Use of Laboratory Animals (NIH) and approved by the institutional guidelines for animal use and care. Male C57B6 mice (~6-weeks-old; The Jackson Lab #000664) were used for all experiments. Mice were housed in standard cages under controlled conditions (20–21 °C, 12 hr light/dark cycle) in the animal facility room at the University of Cincinnati College of Medicine, with ad libitum access to chow diet and water.

WT male mice were randomly assigned to either the subacute exposure model, receiving no treatment or PFOS (10 mg/kg, Fisher, 50-749-36), or a chronic exposure model, receiving vehicle (2 % Tween-80 in PBS) or PFOS (10 mg/kg). All injections were performed by trained personnel to prevent pain and discomfort. Mice were monitored regularly for physiological signs of toxicity throughout the study. At the study endpoint, mice were anesthetized with a ketamine/xylazine cocktail and euthanized by exsanguination. Blood serum, heart, lungs, and liver were collected for downstream analysis.

An a priori power analysis was calculated in SigmaPlot, based on published literature and preliminary data with Oil Red O (ORO) as the quantitative indicator of MASLD, as increased risk and its positive association with PFOS in humans is known (Sen et al., 2022; Xue et al., 2025). With our expected difference in means of ORO being 0.6, expected standard deviation being 0.25, desired power of 0.8 and α ₌0.05, the minimum sample size required is n = 3 per group to see a difference in MASLD by ORO following PFOS exposure. Sample size was also estimated by G*Power using ALT as the marker of liver injury and toxicity, based on published literature of a positive association of PFOS with serum ALT in humans (Lin et al., 2010; Yun and Kwon, 2023; Gallo et al., 2012). This analysis indicated a sample size of at least n = 4 per group would give a difference in serum ALT with desired power of 0.8. Therefore, we included 3–4 mice per group for this whole study and further calculated the power at the completion of each experiment based on our sample size (3–4/group).

2.2. Tissue histopathological analysis, immunohistochemistry and immunofluorescence

Hearts, lungs, and livers were rinsed with PBS and fixed in 10 % neutral buffered formalin, followed by paraffin embedding. Tissues were sectioned at 6 μm for hematoxylin and eosin (H&E) and Sirius Red staining (1 hr staining in saturated picric acid containing 0.1 % Sirius Red F3B/Direct Red 80 followed by washing with 0.005 % aqueous glacial acetic acid).

2.2.1. For Ki67 immunohistochemistry, deparaffinized 6 μm liver sections underwent antigen retrieval by EDTA (pH 8.0), followed by quenching of endogenous peroxidase activity with 3 % H₂O₂ in methanol. Non-specific binding was blocked with 5 % normal Goat Serum (Vector Labs, S-1000) in 0.1 % PBS-Triton. Sections were incubated with primary anti-Ki67 antibody overnight at 4 °C, followed by incubation with secondary antibody for 1 hr at room temperature. Signal was developed using Vectastain ABC kit (Vector Laboratories PK-6100) and DAB substrate (Sigma D3939), with hematoxylin counterstaining.

2.2.2. Fontana-Masson staining was performed on 5 μm paraffin sections according to the manufacturer’s instructions (Sigma, HT200-1KT). Sections were incubated in warmed ammoniacal silver solution for 1 hr, toned with gold chloride for 30 sec, treated with sodium thiosulfate for 2 min, and counterstained with Nuclear Fast Red for 5 min, with water rinses between each step.

2.2.3. Schmorl’s staining was performed on 6 μm deparaffinized sections using freshly made Schmorl’s solution (30 mL 1 % aqueous ferric chloride, 4 mL 1 % aqueous potassium ferricyanide, 6 mL deionized water) for 10 min. Sections were washed with water, counterstained with Nuclear Fast Red for 5 min, then dehydrated in ethanol and cleared in xylene.

2.2.4. Prussian Blue iron staining kit was used on 6 μm tissue sections following the manufacturer’s instructions (Abcam, AB150674-1001). Sections were incubated in equal parts potassium ferricyanide and hydrochloric acid for 3 min, rinsed with water, stained with Nuclear Fast Red for 5 min, rinsed and dehydrated.

2.2.5. Rhodanine copper staining kit was used on 6 μm tissue sections following manufacturer’s instructions (Fisher, NC0831263). Sections were incubated in working rhodamine solution for 1 hr at 60 °C, followed by a 10 min incubation in diluted hematoxylin and 3 dips in 0.5 % aqueous sodium borate, and rinsed with water between steps.

2.2.6. Liver tissues embedded in OCT compound were sectioned at 8 μm for ORO, DHE, and Sudan Black B lipofuscin staining. For ORO staining, sections were stained in freshly made and filtered ORO (1.5 parts ORO stock to 1-part distilled water; 2.5 g ORO in 400 mL isopropanol for ORO stock) for 5 min and rinsed with running tap water for 30 min. For DHE superoxide staining, sections were incubated in 2 μM DHE (Invitrogen, D11347) at 37 °C for 30 min. Sudan Black B staining procedure was performed according to the supplier’s instructions (VitroVivo Biotech, VB-3012). Briefly, frozen sections were fixed in 10 % formalin for 20 min, immersed in two changes of pre-stain solution (5 min each), stained with Sudan Black for 7 min with agitation, immersed in differentiation solution for 3 min, and counterstained with Nuclear Fast Red for 3 min following a water wash.

2.2.7. For immunofluorescence (IF) staining, OCT-embedded livers were sectioned at 8 μm, fixed in 4 % PFA for 10 min, blocked in 5 % normal goat serum for 30 min, and incubated with primary antibodies overnight at 4 °C. Sections were then incubated with goat-conjugated Alexa Fluor secondary antibodies for 1 h at room temperature and mounted with DAPI, with PBS washes in between.

Slides were analyzed microscopically with a Zeiss Axioplan 2 microscope for immunofluorescence or Olympus IBX41 microscope for immunohistochemistry, quantified using ImageJ and validated by a pathologist.

2.3. Serum analysis

Snap-frozen mouse serums were analyzed using the VetScan Mammalian Liver Profile and VetScan Chemistry Analyzer by the Schubert Research Clinic at Cincinnati Children’s Hospital Medical Center.

2.4. Hepatic bile acid measurement

Total hepatic bile acids were measured in liver tissue homogenates using the Total Bile Acid Assay Kit (Cell BioLabs, STA-631). 10 mg snap-frozen liver tissues were homogenized in cold isopropanol, centrifuged at 10,000 × g for 10 min at 4 °C and further diluted in deionized water. Quantification of bile acids was determined following the manufacturer’s protocol, recording the primary wavelength 405 nm and secondary wavelength 630 nm using Gen5 (BioTek).

2.5. Measurement of cardiomyocyte injury markers

Cardiac injury was assessed using ELSIA kits by serum measurements of BNP (Novus Biologicals, NBP2-70011) and cTn1 (Novus Biologicals, NBP3-00456) following the manufacturer’s protocols provided. Briefly, standard solution or serum samples were added to ELISA plates and immediately incubated with a biotinylated detection antibody for BNP for 45 min at 37 °C, or incubated for 90 min at 37 °C then incubated with a biotinylated detection antibody for cTn1. The plates were washed, followed by incubation with HRP conjugate for 30 min at 37 °C. After washing, the substrate reagent was added for incubation at 37 °C for 15 min, then stop solution was added. Plates were read immediately at 450 nm using Gen5 (BioTek).

2.6. Western blot

Liver tissues were homogenized in RIPA buffer, resolved by SDS-PAGE, and transferred to a nitrocellulose membrane. After blocking with 3 % milk (PBS-Tween) and incubated with primary antibodies overnight at 4 °C, the membrane was incubated with horseradish peroxidase-conjugated secondary antibodies for 1 hr at room temperature and developed with ECL (Revvity, NEL103001EA) for 1 min and visualized by a ChemiDoc MP Imager (BioRad). Protein expression levels were normalized to loading control GAPDH for quantification.

2.7. Cell culture

AML12 cells (ATCC, CRL-2254) were treated with 10uM PFOS or vehicle (2 % Tween-80 in PBS), with 10uM rapamycin or vehicle (DMSO) for 24 hr. For Sudan Black B staining, cells on coverslips were fixed in 4 % PFA for 15 min, rinsed in 70 % ethanol for 2 min and stained as described above. Cells were fixed in 4 % PFA for Fontana Masson staining followed by the above protocol. For IF staining, cells were seeded on coverslips, treated with 10 μM PFOS, 10 μM rapamycin or their vehicles for 24 hr, then fixed with 4 % PFA for 15 min at 4 °C and permeabilized in 0.1 % Triton-X for 12 min at room temperature, followed by the method described above. For neutral lipid staining, cells on coverslips were washed with PBS, stained with BODIPY 493/503 (Invitrogen, D3922) for 15 min at 37 °C protected from light, fixed with 4 % PFA for 15 min at room temperature, and mounted on slides with DAPI with washes in between. Cells were treated with 10uM PFOS or vehicle for 18 hr, homogenized in RIPA buffer and followed the method above for Western blot.

2.8. Human RNAseq analysis

Expression data from human liver spheroids treated with PFOS were downloaded from GEO (GSE144775 and GSE145239) and uploaded to iDEP2.01 to generate heatmaps from KEGG and GO signatures.

2.9. Antibodies

Abcam: MRP2 (ab172630), p21^WAF1/Cip1 (ab7960), 4-HNE (ab46545). Bio-Techne: BSEP (NBP1-31824), 8oxo-dG (4354-MC-050). Cell Signaling: ATG13 (13468), ATG14 (5504), BECLIN1 (3738), FIP200 (12436), γH2AX (9718), KI67 (12202), LC3B (3868), VPS34 (4263), SIRT1 (8469) SIRT3 (5490), ULK1 (4776), pULK1 (12753). DSHB: CK19 (Clone TROMA-III). Promega: β-Galactosidase (Z378A). Proteintech: ZO-1 (21773-1-AP). Thermo: Albumin (PA5-89332), GAPDH (MA5-15738), SIRT7 (PA5-115905).

2.10. Data analysis

Sample size estimations and power analyses were calculated using SigmaPlot 16.0 and G*Power 3.1.9.7 software. Data are expressed as the means ± SD. Experiments were performed with 3 biological replicates assayed in triplicates. Statistical significance was determined in GraphPad Prism by two-tailed Student’s t-test or 2-way ANOVA with p < 0.05. n.s.: not significant, *p < 0.05, **p < 0.01, ***p < 0.001.

3. Results

3.1. Subacute occupationally relevant PFOS exposure in mice induces hepatomegaly and MASLD

While PFOS serum concentrations in the general population range from 4.3–30.4 ng/mL (Sonnenberg et al., 2023), levels in highly exposed occupational workers (e.g., chemical manufacturers, fishery employees) can reach up to 62,898 ng/mL (Zhou et al., 2014; Gao et al., 2015). Within this range specifically, fishermen have been reported to exhibit serum PFOS levels of approximately 11,400 ng/mL (Zhou et al., 2014), representing the lower range of occupational exposure, whereas manufacturing workers display intermediate levels averaging about 36,625 ng/mL (Gao et al., 2015). To replicate this occupational exposure for assessing PFOS toxicity, and based on established correlations in mice (1 mg/kg PFOS ≈ 43 ng/mL serum concentration (Wang et al., 2021; Yang et al., 2021), we implemented two exposure regimens in 6-week-old male C57/B6 mice and both simulate PFOS serum doses achieved in occupations populations. The first, a subacute model (25 daily injections of PFOS at 10 mg/kg over 25 days; Fig. 1B), was estimated to achieve a serum PFOS concentration of approximately ~ 10,750 ng/mL to mimic the low range of the PFOS occupational exposure reported in fishermen (11,400 ng/mL (Zhou et al., 2014)). The second, a chronic model (57 injections of PFOS at 10 mg/kg over 105 days; Fig. 2A), was estimated to achieve ~ 24,510 ng/mL, mimicking the low-to-intermediate range of PFOS exposure reported for manufacturing workers (36,625 ng/mL (Gao et al., 2015)). We found that the subacute PFOS exposure regime induced significant hepatomegaly (Fig. 1C) but no noticeable vasculature or histological damage of liver tissue (Fig. 1D). The liver hepatomegaly was associated with no increase of mouse body weight (Fig. 1E), but significant increase in both the liver weight (Fig. 1F) and the liver-to-body weight ratio (Fig. 1G), indicating selective liver enlargement. Histological analysis revealed MASLD development by the hallmark of neutral lipid accumulation (Oil Red O [ORO] staining; Fig. 1H), without evidence of fibrosis (Sirius Red-negative; Fig. 1I). Elevated Ki67-positive hepatocytes (Fig. 1J) confirmed compensatory proliferation, reflecting subacute injury repair. These findings establish that subacute PFOS exposure at an occupationally relevant dose induces hepatomegaly and steatosis while triggering regenerative responses, without fibrotic progression, suggesting a steatotic, non-fibrotic form of hepatotoxicity.

Fig. 2. Chronic PFOS exposure promotes MASLD and results in black livers.

(A) In vivo experimental design for the chronic PFOS exposure model (10 mg/kg body weight, i.p.) (B-F) Representative liver images from vehicle and chronic PFOS-exposed mice; scale bar = 1 cm (B), H&E staining with hepatic central vein and sinusoid congestion (C, middle, arrow) and sinusoid congestion (C, right, arrows). Scale bar = 100 μm. Body weight (D), liver weight (E), and liver-to-body weight ratio (F). (G-K) Liver tissue sections stained with ORO (G), Sirius Red (H) and Ki67 (I); quantification shown to the right of each image. Scale bar = 50 μm. (J-K) Comparison of ORO (J) and Ki67 (K) quantification between the two exposure models. n.s.: not significant; **p < 0.01; ***p < 0.001 by student’s t test or 2-way ANOVA. Vehicle-treated mice (n = 4), chronic PFOS-treated mice (n = 3); representative images quantified (n = 9–30). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3.2. Chronic occupationally relevant PFOS exposure in mice induces distinct black liver discoloration and MASLD

The chronic PFOS exposure regime (Fig. 2A) resulted in a striking black liver discoloration with green and brown undertones (Fig. 2B), contrasting sharply with the reddish-brown livers observed from subacute PFOS-treated mice and vehicle controls. Histologically, we observed pronounced vasculature congestion, including blood stasis within central veins (Fig. 2C, middle, arrow), and sinusoidal congestion accompanied by sinusoidal dilatation (Fig. 2C, right, arrows). This pattern of blood stasis resembles clinical “liver venous congestion,” in which impaired venous outflow leads to the accumulation of blood within hepatic sinusoids (Haddadin et al., 2024; Hilscher and Sanchez, 2016; Spahic et al., 2024). These findings indicate that PFOS exposure at occupationally relevant levels induce a previously unrecognized form of hepatic vascular congestion, likely serving as a major liver-intrinsic contributor to the observed black discoloration.

Given that hepatic blood supply is tightly linked to cardiopulmonary circulation, we next examined whether PFOS affects blood flow in the heart and lungs. Although subacute PFOS exposure did not cause appreciable structural damage to either the heart (Fig. S1A) or lungs (Fig. S1B), chronic PFOS exposure resulted in pathological changes. In the heart, we observed myocarditis-like focal inflammation with immune cell infiltration (Fig. S1C, black arrows), vascular stasis (Fig. S1C, blue arrows), and cardiomyocyte injury indicated by elevated cardiac injury markers BNP (Fig. S1E) and cTnI (Fig. S1F) in the serum. The lungs showed bronchitis with immune cell infiltration (Fig. S1D, middle, black arrows), distended pulmonary vessels (Fig. S1D, middle, blue arrow), and alveolar blood congestion (Fig. S1D, right, arrows), suggestive of pulmonary hypertension (Lai et al., 2014). Based on these observations, we propose an alternative or complementary mechanism to liver-intrinsic congestion: reduced cardiac output in mice chronically exposed to PFOS may cause blood to back up within the cardiopulmonary vasculature, secondarily impairing hepatic venous outflow and leading to hepatic congestion. This venous stasis and accumulation of deoxygenated blood would impart a dark red-to-black appearance to the liver, explaining the striking discoloration observed under chronic PFOS exposure.

Further examination of the liver pathological changes found that despite increased body weight (Fig. 2D), chronic exposure caused significant liver atrophy, evidenced by reduced liver weight (Fig. 2E) and liver-to-body weight ratio (Fig. 2F). Histological analysis again revealed steatosis (ORO; Fig. 2G) comparable to that seen with subacute exposure (Fig. 2J), and still no fibrosis was detected (Sirius Red; Fig. 2H). Notably, Ki67 immunostaining revealed enhanced hepatocyte proliferation in the chronic group (Fig. 2I), exceeding the levels observed in the subacute group (Fig. 2K), indicating a heightened regenerative response to sustained injury. Collectively, the comparable level of steatosis across both subacute and chronic exposure models, coupled with the absence of fibrosis even after prolonged exposure suggests that liver atrophy in the chronic setting arises through alternative hepatotoxic mechanisms beyond steatosis or fibrotic remodeling. Although PFOS-induced lipid dysregulation and its contribution to MASLD are well documented, the emergence of black liver discoloration, resembling rare cholestatic conditions such as Dubin-Johnson Syndrome, alongside liver atrophy appears to be a unique phenotype of chronic PFOS exposure. These distinctive features prompted further investigation into the underlying hepatotoxic mechanisms.

3.3. Chronic PFOS exposure induces hepatic accumulation of melanin and lipofuscin, but not iron and copper, contributing to the characteristic black liver discoloration

In Dubin-Johnson Syndrome, black liver discoloration results from the accumulation of melanin-like pigments derived from polymerized epinephrine metabolites under chronic oxidative stress (Yao et al., 2016; Tang et al., 2018). We hypothesized a similar mechanism may underlie the darkened livers observed following chronic PFOS exposure. Fontana-Masson staining confirmed significant melanin pigment deposition in chronic PFOS-exposed liver sections, which were absent in subacute exposure and the controls (Fig. 3A). We also assessed lipofuscin, a yellow brown “aging pigment” composed of oxidized lipids, proteins, and carbohydrates that accumulates in long-lived cells under chronic stress (Brunk and Terman, 2002; Tauchi et al., 1980). Given PFOS can induce oxidative stress (Wang et al., 2022), lipofuscin accumulation was expected. Sudan Black B staining showed mild deposition after subacute exposure but marked lipofuscin accumulation following chronic PFOS treatment (Fig. 3B), a pattern confirmed by Schmorl’s staining (Fig. 3C), suggesting lipofuscin contributes to the brown-black hue of the liver.

Fig. 3. Chronic PFOS exposure induces hepatic melanin and lipofuscin accumulation.

(A-C) Liver tissue sections stained with Fontana Masson for melanin (A), and both Sudan Black B (B) and Schmorl’s stain (C) for lipofuscin in subacute and chronic PFOS exposure models; quantification shown to the right. Scale bar = 50 μm. n.s.: not significant; **p < 0.01; ***p < 0.001 by 2-way ANOVA. Subacute model: untreated mice (n = 4), PFOS mice (n = 3). Chronic model: vehicle mice (n = 4), PFOS mice (n = 3). Representative images quantified (n = 5–18).

To examine other common pigment sources, we assessed iron and copper, which can darken liver tissue in hemochromatosis or Wilson’s disease (Collins et al., 2010; Anderson and Shah, 2013). However, neither metal showed significant accumulation in either exposure group (Fig. S2A, B). Importantly, pigment accumulation was liver specific. No detectable melanin or lipofuscin, nor iron or copper was observed in the heart (Fig. S3A–D), or lungs (Fig. S4A–D), underscoring the organ-selective nature of PFOS hepatotoxicity. Together, these findings show that chronic PFOS exposure uniquely promotes hepatic melanin-like pigment and lipofuscin accumulation, driven by sustained oxidative stress and systemic congestion, without involvement of iron or copper overload, contributing to the distinctive black liver discoloration phenotype.

3.4. Chronic PFOS exposure induces intrahepatic cholestasis via structural and functional disruption of bile canaliculi in mice

The melanin-like pigment accumulation observed in chronic PFOS exposure, resembling that in Dubin-Johnson Syndrome, suggests impaired bile canalicular transport and possible cholestasis. In line with this, serum alkaline phosphatase (ALP), a cholestatic liver injury marker (Pollock and Minuk, 2017), was significantly elevated in chronic PFOS-exposed mice (Fig. 4A), along with increased serum bile acids (Fig. 4B) and ALT (Fig. 4C), confirming hepatocellular injury. Total bilirubin remained unchanged (Fig. 4D), a pattern consistent with cholestasis without hyperbilirubinemia, where bile acid transport is impaired while bilirubin clearance remains intact (Shani et al., 1970). This was further supported by elevated total hepatic bile acids (Fig. 4E) and visible bile accumulation (yellow-green fluid) in bile canaliculi on H&E staining of liver tissue sections (Fig. 4F, arrows). Normal BUN and slightly increased serum albumin (Fig. S5A, B) indicate preserved synthetic and renal function, reinforcing a selective bile transport defect.

Fig. 4. Chronic PFOS exposure alters liver function and bile acid metabolism, inducing cholestasis by impairing bile canalicular structure, integrity and bile flow.

(A-D) Measurement of mouse serum alkaline phosphatase (ALP; A), bile acids (BA; B), alanine transaminase (ALT; C), and total bilirubin (TBIL; D) in the chronic PFOS exposure model. (E-F) Measurement of total hepatic bile acid contents in liver homogenates (E) and H&E staining of liver sections showing canalicular bile accumulation (F, arrows); quantification shown to the right. Scale bar = 50 μm. (G-J) Immunofluorescence (IF) staining of liver sections from the chronic PFOS exposure model for ZO-1 (G), BSEP (H), MRP2 (I), and albumin/CK19 (J); quantification of each IF shown below. Scale bar = 10 μm. n.s.: not significant; *p < 0.05; **p < 0.01; ***p < 0.001 by student’s t test. Serum analysis: vehicle (n = 3), chronic PFOS (n = 3). Serum and liver bile acid measurement: vehicle (n = 3), chronic PFOS (n = 2) with technical replicates (n = 3). IF: vehicle (n = 4), chronic PFOS (n = 3) with representative images quantified (n = 3–10).

To investigate the mechanism, we examined the bile canalicular architecture and key transporters. ZO-1, a tight junction protein crucial for maintaining hepatocyte polarity and canalicular integrity (Itoh et al., 2021), showed shortened and disrupted localization by immunofluorescence (IF) staining following chronic PFOS exposure (Fig. 4G), indicating canalicular collapse. We then analyzed two major canalicular transporters: BSEP, which exports bile acids, and MRP2, which excretes conjugated bilirubin. Despite unchanged total protein levels (Fig. S6), BSEP showed fragmented and reduced canalicular localization (Fig. 4H), pointing to defective trafficking rather than altered expression. Similarly, MRP2 was mislocalized to the cytoplasm (Fig. 4I), echoing patterns seen in Dubin-Johnson Syndrome and indicating impaired bile salt–independent excretion. Together, these defects reflect loss of ZO-1–mediated structural polarity and transporter mislocalization. Finally, elevated CK19 expression (Fig. 4J) marking biliary epithelial cells, indicated a ductular reaction, suggesting compensatory proliferation in response to impaired bile flow. In summary, chronic PFOS exposure induces cholestasis by disrupting ZO-1-mediated bile canalicular architecture, mislocalizing BSEP and MRP2, and impairing bile excretion. The resultant bile acid buildup likely contributes to the green-yellow hue in the blackened liver phenotype, revealing a novel mechanism of PFOS-induced hepatotoxicity.

3.5. Chronic PFOS exposure accelerates hepatic aging via oxidative stress, cellular senescence, and nuclear sirtuin suppression in mice

Lipofuscin, a hallmark of oxidative stress and cellular aging, accumulates in long-lived, post-mitotic cells like hepatocytes due to impaired lysosomal degradation and lack of cell division (Brunk and Terman, 2002; Tauchi et al., 1980). Its substantial buildup in PFOS-exposed livers suggests chronic PFOS promotes liver aging, likely through sustained oxidative stress and impaired autophagy, which are key aging hallmarks (Kaushik et al., 2021; Fraser et al., 2022). To explore this further, we assessed a panel of canonical and expanded markers of hepatic aging including oxidative stress, genomic instability, cellular senescence, and longevity related NAD⁺-dependent deacetylases (Schmauck-Medina et al., 2022; Lopez-Otin et al., 2023). Chronic PFOS exposure significantly elevated the aging hallmarks of oxidative stress as evidenced by increased hepatic superoxide levels (DHE; Fig. 5A) and by lipid peroxidation (4-Hydroxynonenal or 4-HNE; Fig. 5B); genomic instability shown by elevated DNA double-strand breaks (γH2AX; Fig. 5C) and oxidative DNA damage (8-oxo-dG; Fig. 5D); and cellular senescence indicated by increased senescence-associated β-galactosidase activity (SA-β-Gal; Fig. 5E) and p21^WAF1/Cip1 (Fig. 5F, G). In addition, we examined sirtuins (SIRTs), NAD⁺-dependent deacetylases that regulate stress responses, metabolism, and longevity and play critical role in aging and disease (Imai and Guarente, 2014). Western blot analysis revealed selective downregulation of nuclear SIRT1 and nucleolar SIRT7, while mitochondrial SIRT3 remained unchanged (Fig. 5F, H–J), suggesting preferential disruption of nuclear sirtuin-mediated longevity pathways, which have been reported as a critical aging mechanism (Gamez-Garcia and Vazquez, 2021). In summary, compared to control mice exposed to vehicle, chronic PFOS exposure accelerates hepatic aging through a multifaceted mechanism involving oxidative damage, senescence, and selective loss of nuclear sirtuin activity, further linking environmental toxicants to age-related liver dysfunction.

Fig. 5. Oxidative stress and senescence accelerate aging in chronic PFOS-exposed livers.

(A-E) DHE staining of superoxide (A), IF staining for 4-HNE (B), γH2AX (C), 8-oxo-dG (D), and SA-β-Gal (E) in liver sections from both the subacute and chronic PFOS exposure models, with quantification shown to the right of each panel. Scale bar = 10 μm. (F-J) WB analysis of aging and senescence markers (p21, SIRT1, SIRT3, SIRT7) (F). Quantification of p21 (G), SIRT1 (H), SIRT7 (I), and SIRT3 (J) expression normalized to GAPDH. n.s.: not significant; **p < 0.01; ***p < 0.001 by 2-way ANOVA or student’s t test. Subacute model: untreated mice (n = 4), PFOS-treated mice (n = 3). Chronic model: vehicle-treated mice (n = 4), PFOS-treated mice (n = 3). Representative images quantified (n = 3–10).

3.6. Chronic PFOS exposure accelerates hepatic aging by inhibiting autophagy and promoting age related pigment accumulation, oxidative stress, and expression of aging markers in mice

Autophagy is a critical cellular quality-control mechanism that degrades damaged organelles and proteins, maintaining proteostasis and mitigating age-related stress (Yamamoto et al., 2023). Its decline is a well-established hallmark of aging (Kaushik et al., 2021; Lopez-Otin et al., 2013) and may contribute to the accumulation of intracellular debris such as aging pigment lipofuscin, indicative of lysosomal dysfunction. Given the pronounced lipofuscin buildup observed in PFOS-exposed livers, we investigated whether PFOS impairs autophagy. Western blot analysis of liver tissue lysates revealed that chronic PFOS exposure disrupts two key autophagy-initiating complexes: firstly, the ULK1 complex (ULK1-FIP200-ATG13) to which PFOS reduced ULK1 activating phosphorylation at Ser317, while leaving ATG13 unaffected and FIP200 slightly increased (Fig. 6A–D). As ULK1 activation is essential for phagophore formation, its suppression effectively blocks autophagy initiation despite intact accessory components (Ganley et al., 2009). Another autophagy complex inhibited by PFOS was the Beclin1 complex (ATG14L-Beclin1-Vps34), to which PFOS selectively depleted ATG14L (Fig. 6A, E), a core scaffold protein required for this complex membrane localization and PI3P production at autophagosome initiation sites (phagophore) for expansion of phagophore into autophagosomes (Diao et al., 2015). The conversion from LC3B-I (un-lipidated) to LC3B-II (lipidated) occurs upon autophagosome formation and a decrease in LC3-II levels suggests a reduction in autophagic flux (i.e., autophagy activity) (Runwal et al., 2019). Consistent with the compromised autophagy-promoting complexes, the conversion from LC3B-I to LC3B-II was reduced (Fig. 6A, H), indicating impaired autophagosome formation and blocked autophagic flux. Together, these findings demonstrate that PFOS impairs both autophagy initiation and progression through targeted inhibition of ULK1 and Beclin1 complexes, contributing to pigment accumulation and accelerated hepatic aging.

Fig. 6. Chronic PFOS exposure impairs autophagy in mouse liver.

(A-H) WB analysis of autophagy-related proteins (pULK1, ULK1, ATG13, ATG14, FIP200, Vps34, Beclin1, LC3B) (A). Quantification of GAPDH normalized expression of pULK1 (B, left), ULK1 (B, right), ATG13 (C), FIP200 (D), ATG14L (E), Beclin1 (F), Vps34 (G), and LC3B-II/LC3B-I ratio (H). n.s.: not significant; *p < 0.05; **p < 0.01; ***p < 0.001 by student’s t test. Vehicle mice (n = 3), chronic PFOS mice (n = 3).

3.7. Autophagy suppression is a primary mechanism by which PFOS induces steatosis, canalicular structure defects, oxidative stress and aging pigmentation in AML12 cells

To further investigate the functional role of autophagy in chronic PFOS induced liver phenotypes, we switched to an in vitro approach using the AML12 hepatocyte line, in which PFOS induction of autophagy suppression was recapitulated (Fig. 7A–C). Next, we examined whether impaired autophagy contributes to the PFOS-induced liver effects in this in vitro cell model. We restored autophagy function through treating PFOS-exposed AML12 cells with rapamycin, an autophagy activator (via inhibiting mTORC1), known to activate ULK1 to initiate autophagy (Hosokawa et al., 2009). Co-treatment of AML12 cells with rapamycin and PFOS decreased cellular steatosis levels (Fig. 7D), corrected ZO-1 localization to bile canaliculi (Fig. 7E), restored BSEP expression (Fig. 7F), reduced lipid peroxidation (4-HNE) (Fig. 7G), and attenuated accumulation of aging pigments lipofuscin and melanin (Fig. 7H, I), supporting the hypothesis that PFOS-induced inhibition of autophagy in the liver is a major driver of most of the liver pathological phenotypes observed in mice.

Fig. 7. Impaired autophagy by PFOS treatment of AML12 cells promotes cellular injury recapitulating PFOS effects on mouse livers: steatosis, canalicular structure, oxidative stress, and aging pigment accumulation.

(A-C) WB analysis of autophagy-related proteins (pULK1, ULK1, LC3B) in AML12 cells treated with vehicle (2 % Tween 80 in PBS) or 10uM PFOS for 18 hr (A). Quantification of GAPDH normalized expression of pULK1 (B, left), ULK1 (B, right), LC3B-I (C, left), and LC3B-II (C, right). (D-G) AML12 cells treated with 2 % Tween 80 in PBS or 10uM PFOS +/− DMSO or 10uM rapamycin for 24 hr and stained with BODIPY 493/503 for neutral lipids (D), IF staining of ZO-1 (E), BSEP (F), and 4-HNE (lipid peroxidation) (G). Scale bar = 10 μm. (H-I) AML12 cells treated with 2 % Tween 80 in PBS or 10uM PFOS +/− DMSO or 10uM rapamycin for 24 hr and stained with Sudan Black B for lipofuscin (H) or Fontana Masson for melanin (I), with quantification shown to the right. Scale bar = 50 μm. n.s.: not significant; *p < 0.05; ***p < 0.001 by student’s t test or 2-way ANOVA.

3.8. Clinical relevance of chronic PFOS exposure in accelerating hepatic aging and cholestasis: evidence from human liver organoid studies

Analysis of publicly available GEO datasets (GSE144775 and GSE145239) from human liver spheroids exposed to PFOS revealed transcriptomic alterations that align with our findings in mouse models. Spheroids from both datasets were comprised of isolated primary hepatocytes and Kupffer cells, pooled from 10 different human liver donors, and exposed to a wide range of PFOS from 0.02 to 100 μM (GSE144775), and from 0.2 to 100 μM (GSE145239) or DMSO, for 1, 4, 10, or 14 days. Specifically, PFOS exposure at two concentrations (2 μM and 10 μM) led to a time-dependent downregulation of genes involved in bile secretion (Fig. 8A, B), supporting the clinical relevance of PFOS in inducing bile accumulation and cholestasis. Furthermore, exposure to 2 μM PFOS resulted in a time-dependent downregulation of genes involved in cell cycle regulation (Fig. 8C), along with a 20 μM PFOS suppression of genes associated with cellular senescence (Fig. 8D). While not confirmatory, the human liver spheroid findings provide supportive evidence that chronic PFOS exposure may contribute to accelerated hepatic aging in humans.

Fig. 8. Clinical relevance and a model of the synergistic hepatotoxic effects of chronic PFOS exposure.

(A-D) Transcriptomic analysis of human liver spheroids treated with PFOS for bile secretion genes (A, B), cell cycle transcripts (C), and senescence pathway genes (D). Graphic illustrating a working model of previously unrecognized toxicities associated with chronic occupational PFOS exposure (E). Data from this study establish chronic occupational levels of PFOS exposure as a driver of “black liver” pathology and hepatic aging. In this working model, PFOS-induced hepatic venous congestion leading to blood stasis in the liver contributes to the liver’s dark discoloration. Meanwhile, our data support cardiac injury by PFOS which could induce cardiopulmonary congestion (dashed arrows), contributing to hepatic venous congestion and the liver’s dark discoloration. Concurrently, intrahepatic vascular pressure causes cholestasis by impairing bile flow, which contributes to the greenish hue of the liver. Intrahepatic cholestasis in turn worsens congestion, intensifying liver injury. Hepatic pigmentation, characterized by melanin-like (M) and lipofuscin (L) deposits due to suppression of autophagy by sustained liver injury, adds a brown tone to the liver, and the combined effects produce the distinctive black liver phenotype. These stressors elevate ROS and hepatocellular injury (magenta arrows). Given the established effect of autophagy suppression on promoting cholestasis, pigmentation and liver aging due to lack of autophagic degradation of oxidative byproducts and other cellular wastes, suppression of autophagy by chronic occupational levels of PFOS exposure perpetuates cellular senescence, ultimately culminating in hepatic aging. Notably, the hepatotoxicity underlying the black liver pathology and hepatic aging appear to reinforce one another, establishing a vicious cycle of persistent liver injury, atrophy, and degeneration. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

4. Discussion

This study uncovers previously unrecognized toxicities associated with chronic occupational-level PFOS exposure. While subacute exposure induces a reversible MASLD phenotype, chronic exposure leads to irreversible “black liver” pathology and hepatic aging. As depicted in Fig. 8E, our findings suggest a working model in which occupationally relevant PFOS exposure induces a previously unrecognized form of hepatic vascular congestion and hence hepatic stasis, likely serving as a major liver-intrinsic contributor to the observed black discoloration. We also propose an alternative or complementary mechanism to liver-intrinsic congestion whereby reduced cardiac output and blood backup within the cardiopulmonary vasculature impede hepatic venous outflow, thereby promoting secondary hepatic congestion. Such venous stasis and accumulation of deoxygenated blood in liver would confer the dark red–to–black appearance characteristic of livers from chronically PFOS-exposed mice. Concurrently, blood stasis elevates intrahepatic vascular pressure and hence impairs bile flow, causing intrahepatic cholestasis and the retained bile confers a greenish hue for the liver appearance. Intrahepatic cholestasis, in turn, aggravates blood congestion and stasis, thereby intensifying the vicious cycle of liver injury. The combination of venous stasis and cholestasis amplifies oxidative stress and elevates reactive oxygen species (ROS) levels, promoting chronic hepatocellular injury. Under normal conditions, autophagy alleviates such damage by removing oxidative byproducts. However, PFOS-induced chronic injury suppresses autophagy in hepatocytes, leading to the accumulation of lipids, disruption of canalicular structure, oxidative products and other cellular waste, among which, lipofuscin (wear-and-tear” pigment), together with melanin-like polymerized pigments, produce the characteristic brown-to-black discoloration of the liver, and meanwhile, these pigments further elevates ROS. Prolonged autophagy suppression further exacerbates reactive oxidative stress, buildup of oxidized byproducts, and perpetuates sustained hepatic injury, driving hepatocyte senescence and liver aging. Notably, the hepatocellular toxicity underlying the “black liver” pathology and hepatic aging could reinforce one another, establishing a vicious cycle of persistent liver injury, atrophy, and degeneration (Fig. 8E).

This study included a mouse number of n = 3–4/group. A post hoc power analysis revealed that with the actual sample sizes, the achieved power to detect the observed effect size was 0.758 for Oil Red O analysis of hepatic lipid accumulation and 0.893 for serum ALT between vehicle and chronic PFOS-exposed groups. Further, the power achieved for SA-β-Gal was 0.962, compared to the desired power of 0.80. This post hoc power analysis confirmed that the observed hepatotoxic effects related to mouse size reached the desired level of statistical confidence, and retrospectively demonstrated that the study was appropriately powered, neither underpowered (too small to detect real effects) nor overpowered (wasting unnecessary resources) and is in accordance with the new NIH guidelines aimed at minimizing animal use where possible. Nonetheless, future studies can include larger sample sizes to strengthen toxicological analyses.

The occupationally relevant PFOS doses, critical epidemiological support of our models, and the environmental implications.

Our two PFOS exposure regimens in mouse models of subacute (~10,750 ng/mL) and chronic (~24,510 ng/mL) are comparable to serum PFOS levels observed in fishermen (11,400 ng/mL (Zhou et al., 2014; Gao et al., 2015)) and chemical manufacturing workers (36,625 ng/mL (Zhou et al., 2014; Gao et al., 2015)). We acknowledge that we were unable to measure actual serum PFOS levels in our mice due to the lack of on-site facilities capable of performing these assays. These measurements will be carried out once a suitable commercial laboratory is identified. Nonetheless, the simulated serum PFOS concentrations used in our study fall within ranges observed in human populations, thereby supporting the translational relevance of our models. Further, epidemiological evidence links higher PFAS exposure to accelerated biological aging. A large U.S. study of 14,865 adults from eight NHANES cycles (2003–2018) using multiple aging markers and the ENABL Age algorithm found that elevated PFAS levels, including PFOS, were associated with faster biological aging (Zhao et al., 2024). Another NHANES-based study (1999–2000, n = 262) similarly reported that PFOS exposure was linked to increased epigenetic age acceleration among males (Khodasevich et al., 2025).

Regarding cholestasis, epidemiological and mechanistic studies suggest an association of high PFOS exposure to cholestasis: elevated markers of liver function, like bilirubin and gamma-glutamyl transferase (GGT), which can indicate liver damage or cholestasis. Mechanistic studies also show that PFOS can impair bile acid metabolism, and these combined findings support the hypothesis that PFOS could contribute to cholestasis through its effects on liver enzymes and transporters (Vujic et al., 2024). Together, these epidemiological findings further support the translational relevance of our models and the environmental implication of the findings from this study.

Beyond aging and cholestasis, PFAS exposure has been associated with elevated cardiovascular risk factors and potentially clinical cardiovascular disease (CVD) (Biggeri et al., 2024). The World Health Organization estimates that at least one-quarter of global cardiovascular disease cases are attributable to environmental exposures. Epidemiological data further suggest that PFAS exposure correlates with increased risks of hypertension, hypercholesterolemia, and hyperuricemia and these conditions are closely linked to vascular congestion (Meneguzzi et al., 2021; Schlezinger and Gokce, 2024). However, epidemiological evidence remains limited, particularly regarding occupational PFAS exposure and its associations with hepatic-intrinsic or cardiogenic liver congestion. Further clinical and mechanistic studies are needed to clarify causal pathways and deepen understanding of PFAS toxicology in humans. Notably, the observed links between PFOS, congestion, and accelerated aging from the current study highlight an important new direction for epidemiological research in occupational settings. Within the context of human exposure-relevant doses and the epidemiological support, this study substantially expands our understanding of the toxicological profile of PFOS beyond lipid accumulation to include black liver, congestion, and accelerated hepatic aging which redefines PFOS as a far more hazardous compound. The findings warrant urgent reevaluation of current PFAS safety standards and call attention to the silent, irreversible damage that chronic exposure can inflict on vital organs.

Black liver: a novel and distinct pathological outcome of chronic PFOS exposure resembling Dubin-Johnson Syndrome.

A major novel finding is the identification of the “black liver” phenotype, an irreversible pathology characterized by hepatic congestion, cholestasis, and pigment accumulation. Unlike liver steatosis or fibrosis, this black liver pathology reflects PFOS-induced multifactorial disruption of liver structure and function. We hypothesize that both liver-intrinsic congestion and cardiogenic liver congestion may contribute to the subsequent black discoloration. In the latter case, PFOS-induced cardiotoxicity could cause passive hepatic venous congestion, implicating the cardiopulmonary-hepatic axis in environmentally mediated liver damage. This mechanism is supported by clinical evidence linking systemic congestion and cardiac dysfunction to hepatic hypoxia, highlighting its serious pathological consequences (Haddadin et al., 2024; Hilscher and Sanchez, 2016; Spahic et al., 2024). These liver-intrinsic and cardiogenic mechanisms will be investigated in the future. Regarding any clinical cases of black liver/pigmentation, patients with Dubin-Johnson Syndrome exhibit black and pigmented livers, and the appearance of our mouse livers closely resembles this phenotype, suggesting translational relevance. Similarly, Dubin-Johnson-like syndromes have been described, sharing the characteristic black pigmentation (Kobayashi et al., 2004). Although PFAS exposure-induced black/pigmented livers have not been reported in humans, there is strong evidence linking human PFAS exposure to altered melanin biosynthesis (Goodrich, 2023); supporting the potential for similar liver changes in exposed individuals. These findings underscore the need for increased clinical attention to this possible outcome.

One of the critical questions to address in future studies is how specific this black liver phenotype is to PFOS and whether this is a broader or universal risk among environmental pollutants. We plan to investigate whether the black-liver phenotype can be induced by other environmental contaminants, including short-chain PFAS alternatives, bisphenol A (BPA), organophosphate flame retardants (OPFRs), heavy metals, and microplastics. These studies will involve exposing mice to a wide and systematically defined dose range (including both low and high doses) for each pollutant to (i) identify dose thresholds associated with liver discoloration, (ii) characterize accompanying liver injury and functional impairment, (iii) define the temporal dynamics of onset and progression, and (iv) determine whether a true ceiling effect exists versus a compound-specific, mechanistically distinct response. Collectively, these planned investigations will help clarify the specificity and broader relevance of the black-liver phenotype and better define the translational and public health implications of our findings.

PFOS disrupts bile canaliculi: a new cholestatic mechanism by environmental toxicant.

While earlier studies of PFOS-associated cholestasis focused on nuclear receptor-mediated transcriptional regulation, our findings reveal a novel structural mechanism. Chronic PFOS exposure downregulates ZO-1, a critical tight junction protein, resulting in structural collapse of bile canaliculi and mislocalization of hepatobiliary transporters (BSEP and MRP2). This cholestasis mirrors the canalicular defects seen in Dubin-Johnson Syndrome, which also features melanin-like pigment accumulation and liver darkening, highlighting the clinical relevance of PFOS-induced injury and establishing a new paradigm for understanding environmental cholestatic liver damage.

Autophagy inhibition by PFOS: a central pro-aging mechanism in PFOS toxicology.

It has been well established for autophagy suppression in promotion of aging in mouse models (Rubinsztein et al., 2011). Our data demonstrate that chronic PFOS exposure inhibits autophagy through dual suppression of ULK1 and Beclin1 complexes, preventing autolysosome formation and the clearance of lipofuscin, melanin-like pigments, and damaged cellular components. This represents the first evidence of a panel of beneficial functions of autophagy in alleviating cellular steatosis, canalicular structure, oxidative stress and aging pigment accumulation in the liver. The resulting pigmentation phenocopies genetic disorders like Dubin-Johnson Syndrome and marks a new paradigm in environmental liver toxicology. Importantly, autophagy activator rapamycin reduces cellular steatosis, restores canalicular structure reduces oxidative stress, and pigment buildup in hepatocytes, demonstrating that autophagy dysfunction is not only central to PFOS hepatotoxicity but also suggests a potential therapeutic target. Future studies can include in vivo validation with rapamycin co-treatment with PFOS exposures in our mouse models. In addition to its effects on autophagy, PFOS exposure may also impair the function of other hepatic cell types, including stellate and Kupffer cells, leading to dysregulation of inflammatory pathways. As PFOS is known to be associated with lipid metabolism (Raza et al., 2025), and mitochondria are a major site of lipid metabolic processes, PFOS exposure is likely to exert its effects through mitochondrial dysregulation. These alterations collectively contribute to increased oxidative stress and an accelerated aging process. It is noteworthy that although rapamycin-mediated autophagy reactivation reduced PFOS-induced aging-associated pigments (lipofuscin and melanin) in AML12 cells, other aging markers were not rescued under our experimental conditions (data not shown). As the conditions required for PFOS-induced aging in AML12 cells remain largely unknown, we plan to establish a dedicated in vitro model of PFOS-induced hepatic aging. Building on prior work showing that sequential H₂O₂–induced oxidative stress can drive aging phenotypes in AML12 cells (Singh et al., 2020), we will adapt a similar sequential stress approach using PFOS, with further optimization as needed.

Chronic PFOS exposure accelerates liver aging independent of steatosis.

The aging population is increasing; however, environmental liver aging is an understudied area (Lopez-Otin et al., 2023; Lopez-Otin et al., 2013; Hunt et al., 2019). This is the first study to demonstrate that chronic PFOS exposure at occupational levels accelerates hepatic aging independently of steatosis and fibrosis. Our data show PFOS exposure activates hallmark pathways of aging: oxidative stress, DNA damage, senescence, autophagy inhibition, and sirtuin downregulation. These pathways can form vicious cycles: congestion-induced blood stasis and hypoxia worsen oxidative stress and senescence; autophagy failure leads to accumulation of pigments, damaged organelles and proteins; and aging-related decline impairs cellular stress responses, perpetuating injury. Remarkably, PFOS exposure promotes atrophic senescence without fibrosis, distinguishing it from classical cirrhosis but potentially increasing susceptibility to fibrosis or carcinogenesis. These findings redefine PFOS as a chemical stressor capable of hijacking core aging pathways and inducing silent, long-term organ deterioration. Additional studies are needed to gain insights into the progression from steatosis to liver aging/atrophy/pigmentation by including earlier timepoints (e.g., 60 days). Moreover, future studies will include lower dose ranges (e.g., 1, 0.3, and 0.1 mg/kg) in both subacute and chronic models to help clarify whether the observed effects are dose-dependent or threshold-specific.

Translational relevance and broader impact on environmental health regulation.

Our transcriptomic analysis of PFOS-treated human liver spheroids corroborates our murine findings: bile acid dysregulation, senescence and aging pathways are similarly perturbed in human spheroids in a dose- or time-dependent manner. This cross-species validation underscores the human relevance of the observed effects and supports human liver spheroids as powerful tools in environmental toxicology. The broader implications are far-reaching. This study challenges existing regulatory frameworks which often focus on acute or subacute endpoints, missing the chronic, degenerative trajectory revealed here. PFOS recapitulates genetic disease phenotypes (e.g., Dubin-Johnson Syndrome), raising concern that it may unmask or worsen subclinical liver conditions in exposed individuals. The persistent pigmentary markers, melanin and lipofuscin, could serve as novel biomarkers for long-term PFOS toxicity. Given the absence of fibrosis despite advanced aging, current liver toxicity screens may underestimate PFOS risk. These findings call for new occupational health strategies, including biomonitoring for senescence and pigment accumulation, potential use of new biomarkers such as serum lipofuscin metabolites and ZO-1 fragments, and development of targeted interventions such as senolytics, autophagy inducers, and lysosomal activators.

In conclusion, this study redefines chronic PFOS hepatotoxicity as a multifactorial synthetic pathology involving vascular disruption, canalicular collapse, autophagy impairment, pigmentation, and accelerated aging. By elucidating mechanisms that mimic genetic diseases and irreversibly compromise liver resilience, our findings provide a compelling case for strengthening public health protections against PFAS exposure and adopting more comprehensive, aging-aware toxicological assessments. Early detection of these pathologies may enable timely intervention to mitigate long-term health risks in high-exposure occupational settings.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.envint.2025.109999.

Data availability

Data will be made available on request.