Integrating Disease Data and Toxicology Studies to Uncover the Mechanisms of Indium Tin Oxide (ITO) Nanoparticle-Induced Pulmonary Fibrosis and Develop an Adverse Outcome Pathway (AOP) Framework

Abstract

Exposure to indium tin oxide (ITO) nanoparticles (NPs) is strongly correlated with the development of indium lung disease. Preliminary studies have explored mechanisms of ITO NP-induced pulmonary toxicity, but a gap remains in effective methods for risk assessments. To address this issue, we integrated data from population disease databases with traditional toxicology and RNA sequencing to conduct mechanistic studies and establish an adverse outcome pathway (AOP) for ITO NP-induced lung injury. Our findings demonstrate that exposure to ITO NPs induces early pulmonary fibrosis, characterized by a persistent inflammatory response in mice. Mechanistic analysis reveals that lung injury is driven by the activation of the NF-κB signaling pathway mediated by IL-17A in macrophages. In the AOP framework for ITO-induced pulmonary fibrosis, IL-17A serves as a molecular initiating event, initiating the activation of the NF-κB signaling pathway in macrophages. This activation results in the production of inflammatory cytokines (IL-1β and TNF-α) and fibrogenic factors (TGF-β1), ultimately triggering a cellular-level inflammatory response. The sustained inflammation further promotes microvascular leakage, which is a key contributor to the progression of pulmonary fibrosis. The qualitative and quantitative evaluations of supportive inconsistent evidence for MIE and KEs show that the confidence of this AOP is moderate.

1. Introduction

Indium Tin Oxide (ITO) is a ceramic semiconductor composed of approximately 90% indium oxide (In₂O₃) and 10% tin oxide (SnO₂). It is widely utilized across various applications, including solar cells, flat panel displays, aerospace components, antifrost glass, and energy-saving building windows. − The production of ITO begins with metal indium, which is processed to create a powdered form that is then transformed into sputtering targets through mixing In₂O₃ and SnO₂ via techniques like isostatic pressing or sintering. Notably, the global demand for ITO, driven largely by its application in solar cells, exceeded 3,000 tons in 2020.

As the demand for ITO in the semiconductor industry continues to rise, concerns about occupational exposure risks during its production and use have emerged. These concerns are particularly relevant regarding the potential health hazards associated with the inhalation of nano-sized ITO particles. Understanding and assessing ITO nanoparticles (NPs), especially ITO NPs, are therefore critical for evaluating their health impacts.

Existing research indicates that exposure to ITO NPs can lead to respiratory toxicity and impaired lung function. − Previous studies have shown that high magnetic property and toxicity of particulate matter generated during welding and cutting processes. Epidemiological studies have identified increased risks of lung injury, cancer, and mortality among semiconductor workers in wafer fabrication plants exposed to ITO NPs. − Consequently, lung injury emerges as a significant consequence of occupational exposure to ITO NPs.

One serious health condition associated with such exposure is pulmonary fibrosis, characterized by scar formation and excessive deposition of connective tissue within the lung parenchyma, leading to impaired respiratory function. Previous studies have demonstrated size-dependent ITO toxicokinetics, with larger particles showing selective pulmonary accumulation and prolonged retention, while smaller counterparts exhibited even slower clearance. Limited extrapulmonary translocation (<5% total dose) occurred despite systemic circulation. Both sizes caused respiratory-specific deposition with potential hepatorenal implications from persistent bioaccumulation. Numerous studies have established that exposure to particulate matter is a risk factor for developing pulmonary fibrosis. − Specifically, subacute and subchronic inhalation studies in rodents have shown that exposure to ITO NPs at doses ranging from 0.01 to 100 mg/m³ can induce pulmonary epithelial proliferation, inflammation, and fibrosis, along with immunological alterations such as pulmonary alveolar proteinosis (PAP) and inflammatory cell infiltration. − Despite these findings, the precise biological mechanisms through which ITO NPs induce pulmonary fibrosis remain unclear. Therefore, elucidating the initial molecular processes involved in occupational lung injury induced by ITO NPsis imperative for developing preventive strategies.

To address this knowledge gap, we propose the use of the Adverse Outcome Pathway (AOP) framework, a novel risk assessment tool in the field of systems toxicology. The AOP framework links molecular initiating events (MIEs) to adverse outcomes (AOs) by connecting a series of key events (KEs) across various biological levels (molecular, cellular, tissue, organ, and individual). , This systematic approach aids in evaluating ecological risks, human health hazards, and regulatory implications. Under the guidance of the Organization for Economic Co-operation and Development (OECD) (OECD, 2018), customized Bradford-Hill considerations can be applied to assess the strength of Key Event Relationships (KERs) using evidence weighting, enhancing mechanistic insights and predictive capabilities for environmental chemical safety assessments, thus providing better mechanistic information and predictive value for environmental chemical safety assessments. ,

We believe that constructing an AOP framework will lay the groundwork for predicting the onset and progression of early stage lung fibrosis due to ITO exposure. In this study, we integrate DisGeNET, GeneCards, and transcriptome sequencing results for in vivo RNA interaction analysis to establish a preliminary AOP framework. Through both in vivo and in vitro experiments, we aim to validate the reliability of this framework and explore the relationship between ITO exposure, pulmonary fibrosis, and the underlying mechanisms involved.

2. Methods

2.1. Physicochemical Characterization of ITO Particles

To prepare ITO NPs for analysis, 10 mg of ITO NP powder was weighed and subjected to dry heat sterilization to eliminate endotoxins, followed by high-pressure sterilization. The sterilized powder was then mixed with 1 mL of ultrapure water, phosphate-buffered saline (PBS), or fresh culture medium to create stock solutions at a concentration of 10 mg/mL. This stock solution was diluted with ultrapure water to a final concentration of 50 μg/mL. A drop of the diluted solution was placed on a copper grid and air-dried with nitrogen gas. Transmission electron microscopy (TEM) was employed to capture images of the ITO NPs. Particle sizes were analyzed using ImageJ software, and the size distribution was plotted as a histogram with ORIGIN 8.5.

2.2. Hydrated Particle Size and Zeta Potential Determination of ITO NPs in Different Dispersion Systems

Prepare working solutions of ITO NPs by adding the corresponding amounts of ultrapure water, PBS, and DMEM complete medium to the stock solution, to achieve a concentration of 25 μg/mL. After ultrasonic dispersion using an ultrasonic cleaner for 20 min, measure the hydrated particle size and zeta potential of ITO NPs using a Malvern laser particle size analyzer. Repeat the measurement three times for each working solution.

2.3. Animals and Treatment

The Specific Pathogen Free (SPF)-grade male C57BL/6J mice (8 weeks old) were purchased from Vital River (Beijing, China, SCXK Certification No.: SCXK (Zhe) 2019–0001). The animal experiments were conducted in compliance with the Regulations of the People’s Republic of China on the Administration of Laboratory Animals (2017). The animal protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of Southeast University.

The mice with body weights 18–22 g and were housed in the SPF Laboratory at Southeast University Animal Center. The laboratory maintained a temperature range of 18–24 °C, a relative humidity of 60%, and a 12/12-h light/dark cycle. The mice were provided with ample food and distilled water and had ad libitum access to food and water. After a 7-day acclimation period, the mice were numbered and weighed. Mice with excessively high or low body weights were excluded from the study. The mice were then randomly divided into five groups: n-ITO (3.6 mg/kg b.w.), m-ITO (3.6 mg/kg b.w.), n-ITO (36 mg/kg b.w.), m-ITO (36 mg/kg b.w.), and PBS vehicle control group. n-ITO means small-size ITO nanoparticles, and m-ITO means large-size ITO nanoparticles. Following anesthesia with pentobarbital sodium (50 mg/kg b.w.), nanoparticle suspensions were administered to mice via oropharyngeal aspiration using a calibrated microsyringe. Each group of mice was exposed to 2 times a week, a total of 28 days. Mice were sacrificed by injection of an overdose of barbiturate, and tissue samples were collected.

2.4. Pathological Observation of Lung Tissue

Lung tissues from each mouse were collected, fixed in 4% paraformaldehyde for 24 h, embedded in paraffin, cut into thin sections (approximately 3 μm thick), and subjected to hematoxylin and eosin (H&E), Masson-Goldner trichrome staining, and Periodic Acid-Schiff stain (PAS), Histological observations were performed under an orthotopic Zeiss microscope. The severity of peribronchial inflammation in HE-stained lung sections was assessed histologically via a six-grade system: grade 0 defined as no inflammatory cell infiltration; grade 1 shows scattered focal cells; grade 2 features a single cell layer around the bronchial wall; grade 3 displays two concentric cell layers; grade 4 has 3–4 continuous cell layers forming an infiltrative band; and grade 5 presents with over 4 cell layers. Using Masson-Goldner stained sections, the collagen volume fraction (CVF) was quantified using ImageJ software to measure pulmonary fibrosis severity. CVF is defined as the ratio of collagen deposition area to total tissue area (CVF = collagen area/tissue area). AB-PAS-stained sections, goblet cell metaplasia in central and peripheral airways was evaluated and classified into five grades based on the percentage of positively stained airway epithelium: grade 0 (no staining); grade 1 (≤25% positive area); grade 2 (26% - 50%); grade 3 (51% - 75%); and grade 4 (>75%).

2.5. Immunohistochemistry

Lung tissue sections were orderly deparaffinized and dehydrated with xylene and a graded ethanol series (70%–100%). Subsequently, the sections were processed for antigen repair with 0.01 mol/L citric acid buffer (pH 6.0) and endogenous blockade (with 3% hydrogen peroxide). Then, incubated with primary antibodies (TGF-β1, IL-17A, α-SMA, FN, IL-1β, MMP13) overnight at 4 °C after being blocked. Next, incubated with an HRP-labeled secondary antibody for 1 h. Finally, the DAB reagent was used to observe binding sites. Nuclei were counterstained using Modified Lillie-Mayer’s Hematoxylin Solution (G1080, Solarbio).

2.6. Immunofluorescence

For sections of lung tissues in strict conformity with the immunofluorescence experiment process, orderly fixed with 4% PFA) and permeabilized with 0.3% Triton-X-100. Tissue sections were incubated with primary antibodies mixed with anti-IL-6 and anti-CD11b at 4 °C, and cells were incubated with anti-NF-κB overnight after blocking the sections with 3% bovine serum albumin (BSA). Next, the slice and cell with the second labeled antibody incubation for 1 h, away from light. Next, the slice with the second labeled antibody and cell incubation for 1 h, away from light. Antibody information is provided in Table S1.

2.7. Cell Culture

Mouse leukemic monocyte/macrophage cell line RAW264.7 cells were obtained from the Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences (Shanghai, China). Cells were cultivated in DMEM high-glucose medium fetal bovine serum and maintained in 5% CO₂ at 37 °C. Cells in the logarithmic growth phase were taken for experiments. According to the results of the preliminary experiment, the cytotoxicity assessment was divided into the following groups: blank Control group (DMEM), n-ITO group (6.25 μg/mL, 12.5 μg/mL, and 25 μg/mL), and m-ITO group (6.25 μg/mL, 12.5 μg/mL and 25 μg/mL).

2.8. Cytotoxicity Assessment

The cytotoxicity assessment comprised two experiments: cell survival rate assay and lactate dehydrogenase (LDH) assay. RAW264.7 cells in the logarithmic growth phase (2 × 10⁵ cells) were seeded in 6-well plates. After overnight culture, the original medium was replaced with a complete medium containing different concentrations of n-ITO and m-ITO for exposure. The plates were then placed in a cell incubator. The survival rate of RAW264.7 cells after 24 h of exposure was determined using the Calcein AM/PI Double Stain Kit (Maokang, china). Calcein AM-positive cells were considered viable, while cells positive for both Calcein AM and PI or PI alone were classified as dead. The cell survival rate was calculated as the ratio of Calcein AM-positive cells to the total number of cells.

In addition, RAW264.7 cells in the logarithmic growth phase (1 × 10⁴ cells) were seeded in 96-well plates. After overnight culture, the original medium was replaced with a complete medium containing various concentrations of n-ITO and m-ITO. Each concentration was prepared in triplicate wells. After 24 h of exposure, LDH content was measured using the LDH cytotoxicity assay kit (Beyotime Biotechnology, China). The optical density (OD) of the solution was measured using a microplate reader at a wavelength of 490 nm. The LDH relative release of cells was calculated as (OD of an experimental group - OD of the blank group)/(OD of the control group - OD of the blank group) × 100%.

2.9. Measurement of Intracellular ROS

RAW264.7 cells in the logarithmic growth phase (2 × 10⁵ cells) were seeded in 12-well plates. After 24 h of exposure, the working solution was added according to the requirements of the ROS kit (Beyotime Biotechnology, China). After mixing, the 12-well plates were placed in the cell incubator and incubated for 30 min. Probes that did not enter the cell were removed. After that, 500 μL PBS was added to each well, and the intracellular ROS level was observed at an excitation wavelength of 488 nm and an emission wavelength of 525 nm using a Zeiss microscope.

2.10. ATP Analysis

RAW264.7 cells in the logarithmic growth phase (2 × 10⁵ cells) were seeded in 6-well plates. After exposure to the above methods. The working solution was added using the ATP assay kit (Beyotime Biotechnology, China) instructions. The intracellular ATP content was detected by a microplate luminescence detector after mixing.

2.11. The Intracellular Glutathione (GSH) Content Was Detected

RAW264.7 cells in the logarithmic growth phase (2 × 10⁵ cells) were seeded in 12-well plates. After 24 h of exposure, the working solution was added according to the requirements of the GSH kit (Beyotime Biotechnology, China). After 24 h of exposure, the cells were pretreated and collected in a 96-well plate according to the requirements of the kit, and the working solution was added. After thorough mixing, the 96-well plate was placed in a microplate reader. Absorbance value (A1) was read at 405 nm wavelength at 30 s, and absorbance value (A2) was read at 10 min and 30 s.

GSH content (μmol/L) = GSH measured value ΔA value (A2-A1)/GSH standard value ΔA value (A2-A1) × standard concentration (50 μmol/L).

2.12. Western Blot Analysis

RAW264.7 cells were seeded onto plates of 10 mm at a density of 1 × 10⁵ cells per well. After removing the supernatant, cells were washed with ice-cold PBS. The cytoplasmic proteins were collected according to the in- instructions of the whole protein extraction kit (KGP150/KGP1100, Jiangsu, China). The cells were mechanically scraped off using a rubber scraper and centrifuged at 12,000 × g for 10 min. Protein concentrations were determined using the BCA kit (KGPBCA, Jiangsu, China) according to the supplier’s instructions. Fifty micrograms of cell proteins were loaded per lane and separated by SDS-PAGE (10%). After electrophoresis, proteins were transferred to PVDF membranes using electrophoretic transfer (Bio-Rad, Berkeley, CA, USA). Membranes were blocked with 5% defat milk in TBST for 2 h at room temperature (RT) and incubated with the following primary antibodies: IL-17A, NF-κB, IκBα, IL-1β, β-actin, TGF-β1 at 4 °C overnight. Thermo32106 ECL luminescence (Thermo Fisher Scientific, USA) measurements were performed after coincubation with the corresponding secondary antibody (V: V = 1:10,000) at room temperature for 1 h. The Tanon MP imaging System (Tanon5200, Shanghai, China) and ImageJ 1.53a were used for image acquisition and strip grayscale quantification.

2.13. Protein Molecular Docking

Rigid protein–protein docking (ZDOCK) was performed between IL-17A (PDB ID: 4 h9) and Tnfsf11 (PDB ID: 3urf) to study the relationships. The PDB format of the protein structural domain was downloaded from the Protein Data Bank PDB database (http://wwwrcsb.org/. The ZDOCk module was run to identify the docking sites and calculate the ZDOCK scores.

2.14. Quantification of Key Event Relationship Weights in the AOP Framework

AOP-score = (Σ­[Assigned Weight × actual Expert Rating] for all Key Events (KEs))/(Σ­[Assigned Weight × Maximum Rating (+++)] for all KEs)

Here, ″+″ symbols represent expert consensus ratings (1+ to 3+), where

Numerator: Sum of each KE’s weight multiplied by its assigned rating

Denominator: Sum of each KE’s weight multiplied by the maximum possible rating (3+)

For example, a KE with weight 2.0 rated as ++ (2+) contributes 4.0 (2.0 × 2) to the numerator and 6.0 (2.0 × 3) to the denominator.

2.15. Statistical Analysis

Data are presented as mean ± standard error of the mean (SEM) of at least three independent experiments. Statistical differences between the experimental groups were assessed using a one-way analysis of variance (ANOVA) followed by the Tukey posthoc test.

3. Results

3.1. Physical Characterization of ITO Particles in Different Sizes

The ultrastructure and distribution of n-ITO and m-ITO were examined using transmission electron microscopy (TEM). Electron microscopy results suggest that the ITO nanoparticles exhibited a near-spherical shape (Figure A, B). Quantitative analysis of the particle size distribution (Figure C, D) indicated that the particle size of n-ITO predominantly ranged between 10 and 50 nm, while that of m-ITO ranged between 40 and 200 nm, with average sizes of 25.21 and 102.89 nm, respectively. Moreover, according to Table , the hydrodynamic particle sizes of n-ITO in ultrapure water, PBS, and complete culture media (DMEM) were found to be 207.1, 837.6, and 48.9 nm, respectively. The corresponding Zeta potentials were measured as −3.98 mV, −4.21 mV, and −9.78 mV, respectively. For m-ITO, the hydrodynamic particle sizes in ultrapure water, PBS, and DMEM complete medium were determined as 356.4, 401.8, and 126.4 nm, respectively, with Zeta potentials of −1.96 mV, −3.92 mV, and −9.72 mV. Notably, we observed that the absolute value of the Zeta potential for ITO NPs in complete culture media was higher compared to that in ultrapure water and PBS.

1.

Physicochemical characterization of ITO. TEM images of m-ITO (A) and n-ITO (B). Size distribution measurements of m-ITO (C) and n-ITO (D).

1. Hydrated Particle Size and Zeta Potential ITO NPs In Different Dispersions.

	n-ITO (15–25 nm)		m-ITO (60–90 nm)
	Hydrated particle size (nm)	Zeta potential (mV)	Hydrated particle size (nm)	Zeta potential (mV)
Uitrapure water	207.1	–3.98	356.4	–1.96
PBS	837.6	–4.21	401.8	–3.92
DMEM	48.9	–9.78	126.4	–9.72

3.2. Exposure to ITO NPs for 28 Days Induces Lung Injury in Mice

Total protein and LDH levels in bronchoalveolar lavage fluid (BALF) were measured to understand the mechanism of lung injury caused by ITO-NPs. Compared to the control group, repeated exposure to low-dose n-ITO (3.6 mg/kg) in mice led to a significant increase in total protein (Figure A) and LDH levels (Figure B) in BALF, while the same dose of m-ITO (3.6 mg/kg) did not cause such statistical difference. Importantly, the inflammatory cytokines and fibrosis factors in BALF demonstrated higher sensitivity to the stimulation of ITO-NPs. As shown in the Figures, the expression levels of IL-6, IL-1β, and TGF-β1 in BALF were significantly elevated in response to low-dose ITO-NPs (3.6 mg/kg) compared to the control group (Figure C–F). When the exposure dose of ITO-NPs was increased to 36 mg/kg, a dose-dependent change in the expression levels of these factors was observed. It is noteworthy that compared to the n-ITO group, the expression levels of cytokines in the BALF of the m-ITO group were lower. The expression of TNF-α, IL-1β, IL-6, TGF-β1 in the high-dose group showed a particle size-dependent change, n-ITO exposure significantly upregulated inflammatory markers compared to m-ITO (Figure C–F).

2.

ITO NPs exposure induced inflammation and fibrosis in the lungs of mice. (A) BCA analysis of Total protein in BALF. (B–F) Elisa analysis of LDH and Inflammatory factors (IL-6, IL-1β, TNF-α, TGF-β1) in BALF. (G, I, K) Representative images of lungs of mice stained with HE, PAS, and Masson’s trichrome. (H, J, L) According to the standard method, lung injury indices were measured. (Results are presented mean ± standard error of the mean (SEM), n = 3, compared with the control group, ^* P < 0.05; comparison between groups of ITO NPs with different sizes, ^# P < 0.05.)

By examining key molecules in BALF, we discovered that ITO-NPs may disrupt cellular structures and induce acute lung injury in mice through the promotion of inflammatory reactions. Furthermore, histological examination using HE staining revealed several pathological changes in the lung tissue of mice repeatedly exposed to ITO-NPs (3.6 mg/kg), including alveolar space narrowing, increased lung interstitial tissue, increased alveolar septal cells, thickening of alveolar septa, neutrophil infiltration, and inflammatory exudate (Figure G-H). Increasing the exposure concentration of ITO-NPs exacerbated the lung injury response, as indicated by more severe effects observed in mice exposed to n-ITO (36 mg/kg). PAS staining indicated that after ITO-NPs treatment, a significant amount of PAS-positive glycoproteins was present in bronchi, terminal bronchioles, and alveolar spaces. Under the same dose of toxin exposure, the expression and distribution of starch or neutral mucin were more obvious in the n-ITO group compared to the m-ITO group. Additionally, at the same particle size, an elevated dose of ITO-NPs resulted in further diffusion of starch or neutral mucin throughout the alveolar space. The PAS staining-positive area increased in mice exposed to ITO-NPs (36 mg/kg), suggesting a higher number of mucous cells occupying the bronchi or alveoli (Figure I-J). Moreover, Masson staining was used to label collagen fibers and muscle fibers. Collagen fibers stained blue, and we observed the deposition of collagen surrounding blood vessels or bronchioles in mice repeatedly exposed to 3.6 mg/kg m-ITO, showing a concentration-dependent collagen deposition. For the 3.6 mg/kg n-ITO group, the distribution of collagen fibers was similar to the m-ITO group, while at higher doses, collagen fibers appeared to be distributed throughout the lung tissue (Figure K-L).

In summary, the results indicate that ITO-NPs can disrupt cellular structures and induce pulmonary fibrosis in mice by promoting inflammatory reactions. Notably, the effects were more pronounced in the n-ITO exposure group, which had smaller particle sizes, and the severity of pulmonary fibrosis was found to be dose-dependent.

3.3. Integration of Transcriptomic Analysis and Disease Gene Database to Identify Pulmonary Fibrosis-Related Genes Following ITO NPs Exposure in Mice

To explore the potential molecular changes in the lung tissue of mice after ITO NPs exposure, through transcriptomic analysis we found significant activation of the IL-17 signaling pathway across ITO NPs of different particle sizes exposure (Figure A). Notably, smaller-sized ITO NPs (n-ITO) exhibited a pronounced effect on the activation of the NF-κB signaling pathway. To delve deeper into the significance of our observations, we have compiled a comprehensive list of 944 distinct genes derived from the differential expression profiles that correlate with exposure to varying sizes of ITO NPs (Figure B). To enhance the biological relevance of these results and establish a clearer link between the molecular mechanisms observed in animal models and human disease, we also compiled a comprehensive set of genes related to pulmonary fibrosis from established disease gene databases. Specifically, we identified 7,260 genes from the GeneCards database and 924 from the DisGeNET database, each with gene-disease association scores exceeding 0.01. After eliminating duplicates, we refined this list to 7,395 unique genes. This integrated analysis not only deepens our understanding of the molecular pathways implicated in pulmonary fibrosis but also bridges the gap between experimental findings in animal models and human disease data.

3.

Gene perturbations associated with ITO NP-induced lung injury were analyzed using multiple databases. (A) KEGG enrichment of differential genes in lung tissue of mice exposed to ITO NPs was analyzed based on RNA sequencing. (B) Based on RNA sequencing, the number of differentially expressed genes in the lung tissue of mice exposed to ITO NPs was analyzed. (C) Genes related to pulmonary fibrosis in the Human Gene Disease database (Genecards and Disgenet) and RNA sequencing of mice after ITO NPs exposure (Control vs m-ITO, Control vs n-ITO) were intersected to obtain the differential genes with clinical research value. (D) The 305 genes obtained were subjected to KEGG analysis. (E) IL-17A signaling pathway-related genes and NF-κB signaling pathway-related gene network interaction analysis (orange lines indicate that IL-17A and Tnfsf11 can interact). (F) Rigid protein–protein docking was performed between IL-17A and TNFSF11 to determine their interaction.

By performing an interaction analysis between these fibrosis-related genes and the differentially expressed genes identified in our RNA sequencing results, we successfully identified 305 genes specifically associated with ITO NP exposure and pulmonary fibrosis (Figure C). These genes were subsequently analyzed using the STRING database for protein–protein interactions, revealing 15 pathways with significant activation effects as determined by the Kyoto Encyclopedia of Genes and Genomes (KEGG) (Figure D). Importantly, our results underscore the critical roles of both the IL-17 and NF-κB signaling pathways in the pathogenesis of pulmonary fibrosis. Further investigation into the regulatory relationship between these pathways, through a protein–protein interaction network (Figure E), revealed a direct interaction between IL-17Aa pivotal protein in the IL-17 signaling pathwayand TNFSF11, a ligand for the receptor activator of NF-κB (Figure F). This interaction suggests a mechanism whereby IL-17A may facilitate the activation of NF-κB, thereby promoting inflammatory processes and fibrosis. To substantiate this hypothesis, we conducted molecular docking analyses of IL-17A and TNFSF11, which indicated that their interaction is stabilized by hydrogen bonds between amino acid residues. The calculated binding energy of −13.1 kcal/mol further supports the stability of this protein complex.

In conclusion, our study provides preliminary evidence that exposure to ITO NPs induces pulmonary fibrosis at the molecular level, emphasizing the role of IL-17A in modulating the NF-κB signaling pathway. This interplay mediates inflammatory responses that contribute to the development of pulmonary fibrosis, highlighting potential therapeutic targets for intervention.

3.4. AOP was Established Based on the Human Disease Database and Gene Perturbations from In Vivo Experimental Transcriptome Sequencing of ITO NPs Exposure

Based on the above research, apart from differentially expressed genes (DEGs) and enrichment analysis, we constructed a heatmap (Figure A) to display specific gene expression in various groups containing the following five categories: Inflammatory Response, fibrosis, oxidative stress, cell proliferation and transforming growth factor-related genes (Table S2). Concurrently, 305 DEGs derived from the interaction analysis between the disease Gene database and transcriptome sequencing were mapped to the corresponding terms in the Gene Ontology (GO) database. The meaningful and abundant DEGS terms were then calculated by hypergeometric distribution. The results of the GO enrichment analysis of DEGS were classified according to biological process (BP), cellular component (CC), and molecular function (MF) (Figure. B). The results suggested that BP related to Immune response was the most significant, such as Defense response, immune response, and Inflammatory response. Related terms in MF correspond to our previous KEGG enrichment results, such as signaling receptor binding, receptor–ligand activity, signaling receptor regulator activity, Receptor ligand activity, and highlighting signaling pathway regulation between the signal path. Therefore, based on the above multivariate big data analysis, we classified these key events at the molecular, cellular, tissue, and organismal levels and constructed a preliminary AOP framework for ITO NPs induced occupational lung injury (Figure C). Then, we first validated the key molecular events in the AOP framework at the genetic level, the results showed that n-ITO activated more genes than m-ITO and IL-17A was significantly increased after low dose n-ITO exposure reminding us that IL-17A may act an important role in n-ITO triggered lung injury. The RT-PCR results suggested that the levels of Tnfrsf9, IL-2RA, IL-17A, IL-7R, and Tnfrsf11 have changed after ITO NPs exposure. IL-17A and IL-17R were increased in both m-ITO and n-ITO groups. However, Tnfsf8, IL-2RA, IL-7R, and Tnfrsf11 were merely upregulated in the n-ITO group (Figure D-H). Therefore, it seems that n-ITO affected more gene expression but the two sized materials shared similar pathways in which IL-17A was involved to induce lung damage and pathologies.

4.

Transcriptomic sequencing of mouse lung tissue after ITO NPs exposure. (A) The heat map shows the different functions of the differentially expressed genes. (B) the GO enrichment analysis of the DEGs on biological processes. (C) The AOP framework was established based on the disease-gene database and RNA sequencing of mouse lung tissue. (D-H) Representative differentially expressed genes were selected for qPCR validation. (Results were presented as mean ± standard error of the mean (SEM), n = 3, Compared with the control group, *P < 0.05; Comparison between groups of ITO NPs with different sizes, ^# P < 0.05.)

3.5. ITO NPs Exposed Mice Induce Macrophage Inflammatory Response and Lung Tissue Fibrosis-Associated Protein Abnormal increase

To further confirm the damaging effect of ITO NPs on mouse lungs, we verified the changes in key molecules at the protein level by immunohistochemistry (IHC). We paid special attention to the expression of IL-17A in tissues, which showed a dose-dependent increase and the smaller the particle size of ITO NPs, the more significant the expression of IL-17A (Figure A, C). TGF-β1 can be used as one of the signs of lung tissue damage as it is an important regulator of many cellular processes such as apoptosis, cell proliferation, differentiation, etc. α-smooth muscle actin (α-SMA) is expressed in activated myofibroblasts and myofibroblasts can generate ECM components like collagen, fibronectin, and laminin. Therefore, to observe that lung injury was caused by different-sized ITO NPs, we chose the three markers containing TGF-β1, α-SMA, and fibronectin (FN). It is not surprising that all three of these molecules were activated following treatment with varying doses and sizes of ITO NPs (Figure A-B, D-E). In particular, it appears that both m-ITO and n-ITO result in greater accumulation of fibrosis-related proteins, with the effect being dependent on the particle size for ITO. Specifically, TGF-β1 is primarily distributed in the cytoplasm, with a smaller amount found in the extracellular matrix. This distribution pattern differs from that observed for α-SMA and FN. As shown in Figure. A, α-SMA is localized along the bronchial and alveolar walls, encompassing the infiltrated nucleus. FN distribution is similar to that of α-SMA, but it fills the alveolar space, indicating that lung fibroblasts may have transformed into myofibroblasts, resulting in increased production of extracellular matrix associated with fibrosis. Interestingly, FN is released at high levels both in response to low and high doses of m-ITO. However, with n-ITO, FN exhibits a dose-dependent effect, being primarily located in the alveolar wall after treatment with smaller-sized ITO NPs, at a low dose, but invading the alveolar space at a high dose. It should be noted that compared to smaller-sized ITO (n-ITO group), exposure to m-ITO appears to promote greater FN production in the lung, suggesting that particle size may contribute to the development of distinct respiratory pathologies caused by ITO NPs (Figure E). It is worth noting that IL-1β levels were significantly elevated and displayed a puncta-like distribution following treatment with various sizes and doses of ITO, except for the high dose of n-ITO (36 mg/kg), where was extracellularly localized in the extracellular area (Figure A, F). Another protein of interest, matrix metallopeptidase-13 (MMP-13), known for its involvement in processes like cell proliferation, migration, and differentiation, acts as a collagen hydrolase. Interestingly, only at a dose of 36 mg/kg did ITO NPs, display a size-dependent increase in MMP-13 distribution, while both sizes of ITO NPs, caused noticeable MMP-13 activation at a dose of 3.6 mg/kg (Figure A, G).

5.

Lung inflammation and fibrosis-associated protein changes in mice lung tissue after ITO NPs exposure. (A) Representative images of IHC staining for TGF-β1, IL-17A, α-SMA, FN, IL-1β, and MMP13 in the lung tissue. (B–G) Positive areas of TGF-β1, IL-17A, α-SMA, FN, IL-1β, and MMP13 were quantified by densitometry (n = 3). (Results are presented as mean ± standard error of the mean (SEM), n = 3, compared with the control group, *P < 0.05; comparison between groups of ITO NPs with different sizes, ^# P < 0.05.) (H) Fluorescence colocalization of CD11b (macrophage marker) and IL-6­(pulmonary fibrosis-related inflammatory factor) in the lung tissue (n = 3); 400× magnification; scale bars = 40 μm; Pr: Pearson correlation.

These results highlight that ITO NPs activate various molecules relevant to pulmonary inflammation and fibrosis to different extents. In addition, our previous findings indicated that macrophages played a crucial role in the development of ITO NP-induced lung injury. Immunofluorescence staining of lung tissue further confirmed the accumulation of a significant number of macrophages around the injured areas accompanied by the release of IL-6 inflammatory factors, it is worth noting that IL-6 is closely related to pulmonary fibrosis (Figure. H). While our in vivo experiments provided valuable insights into the disruption of immune activation and the inflammatory response triggered by ITO NPs in lung tissue of mice, the underlying molecular mechanisms remain unclear. To address this knowledge gap, we transitioned to cellular experiments to investigate the effects at a more specific and detailed level.

3.6. ITO NPs Elicited Proinflammatory and Profibrotic Responses in RAW264.7 Cells

We employed RAW264.7 cells, mouse peritoneal macrophages, as an in vitro model to investigate the correlation between the physicochemical properties of ITO NPs and their cytotoxic effects, to uncover the underlying mechanisms of ITO NP-induced Macrophage dysfunction. Following a 24-h exposure to ITO NPs, RAW264.7 cells exhibited a dose-dependent decrease in cell viability (Figure S1A), a reduction in intracellular ATP content (Figure S1B), and an increase in LDH release (Figure S1C). These findings indicate potential disruptions in cell membrane integrity and cellular metabolism caused by ITO NPs. The toxic effects of exogenous substances on cells can often be visually observed through changes in cellular morphology. Therefore, we used microscopy to examine the morphological alterations of RAW264.7 cells after 24 h of ITO NP exposure. The number of RAW264.7 cells significantly decreased, intercellular connections decreased, and cells exhibited swelling with numerous bubble-like protrusions, indicative of cellular uptake of surrounding ITO NPs (Figure S1D). It is worth noting that the intracellular indium content was higher at the same end point of exposure in the n-ITO group compared to the m-ITO group (Figure S1E), suggesting greater cellular uptake of n-ITO. Furthermore, fluorescence staining (Figure S1F) and flow cytometry analysis (Figure S1G) revealed an elevated level of reactive oxygen species (ROS) within the macrophages. Not only that, but we also observed ITO NPs led to a decrease in intracellular GSH levels within the macrophages (Figure S1H), indicating an imbalance between oxidation and antioxidant processes and inducing oxidative stress in cells.

To further confirm that RAW264.7 cells could be used as a cell model to complement the mechanistic study in vivo, we then validated the RNA-seq results using RAW264.7 cells. As expected, the release of inflammatory and pro-fibrotic factors was increased in RAW264.7 cells exposed to ITO NPs for 12 h, using bleomycin (BLM), induces fibrosis in lung tissue, as a positive control. Inflammatory factors (IL-1β, IL-6, TNF-α, MIP-1α, MCP-1) and pro-fibrotic factors (TGF-β1, PDGF-AA) increased in a dose-dependent manner in the ITO NPs exposure group compared to the control group (P < 0.05). Compared to m-ITO, n-ITO released more inflammatory and pro-fibrotic factors, but only TNF-α at the dose of 6.25 μg/mL showed a statistically significant difference (Figure A-G). In addition, immunofluorescence analysis revealed a dose-dependent alteration in the nuclear translocation of NF-κB in RAW264.7 cells treated with ITO NPs, with a particularly strong change observed in the nucleus of RAW264.7 cells treated with 25 μg/mL of n-ITO NPs (Figure H).

6.

Changes in cytokines released from RAW264.7 cells after treatment with ITO NPs. (A–G) ELISA detection of RAW264.7 cells release of inflammatory factors and fibrosis factors level after ITO NPs treatment. (H) Nuclear translocation of NF-κB was detected by immunofluorescence in RAW264.7 cells after treatment with ITO NPs. (Results are presented as mean ± standard error of the mean (SEM), n = 3, compared with the control group, *P < 0.05; within-group comparisons, ^# P < 0.05.)

In summary, the experimental results using RAW264.7 cells indicate that ITO NPs have proinflammatory and fibrosis-promoting effects. It is noteworthy that, compared to the other exposure groups, the cytotoxicity induced by n-ITO is more pronounced.

3.7. ITO NPs Treatment Induced the Activation of IL-17A and NF-κB Signaling Pathways in RAW264.7 Cells to Mediate Inflammatory Responses

Building upon the observation of elevated inflammatory markers in RAW264.7 cells, RNA-seq results indicated that IL-17A may regulate the NF-κB signaling pathway. The binding of IL-17A to its receptor, IL-17RA, is known to activate the transcription factor NF-κB, suggesting a potential interplay between these signaling pathways. To investigate this further, we examined the expression levels of IL-17A and NF-κB signaling pathway-related proteins in RAW264.7 cells exposed to varying concentrations and sizes of ITO NPs. As illustrated in the accompanying Figure , exposure to ITO NPs for 12 h resulted in a concentration-dependent increase in intracellular IL-17A expression compared to the control group, with a significant difference noted in the n-ITO group at a dose of 25 μg/mL (P < 0.05) (Figure A). Concurrently, NF-κB protein expression also exhibited a concentration-dependent increase, with a notable rise in the n-ITO group at doses of 25 μg/mL (P < 0.05). In contrast, the expression of IκBα, a negative feedback regulator of NF-κB, decreased in a concentration-dependent manner, particularly at 25 μg/mL of m-ITO exposure (P < 0.05). Additionally, IL-1β levels increased with ITO NP treatment, further supporting the pro-inflammatory effects of these particles (Figure B-E). These results underscore the role of ITO NPs in promoting inflammation and immune responses through the activation of the IL-17A and NF-κB pathways. The observed regulatory relationship between these signaling pathways warrants further investigation to elucidate the specific mechanisms by which ITO NPs induce inflammatory signaling in macrophages.

7.

Changes in key molecular events in major pathways in RAW264.7 cells after ITO NPs treatment. (A) Representative images of IL-17A, NF-κB, IκBα, IL-1β protein expression in the prefrontal cortex. (B–E) Protein quantitative analysis of IL-17A, NF-κB, IκBα, IL-1β (n = 3). (Results are presented as mean ± standard error of the mean (SEM), n = 3, compared with the control group, *P < 0.05; comparison between groups of ITO NPs with different sizes, ^# P < 0.05.)

3.8. Inhibition of IL-17A Inhibits the Activation of the NF-κB Pathway, Relieves ITO NPs Induced Inflammation, and Fibrosis Factors Released

To elucidate the relationship between IL-17A and inflammation and fibrosis induced by ITO NPs, we pretreated RAW264.7 cells with secukinumab for 4 h before exposing the cells to ITO NPs for 12 h. Subsequent quantification of IL-17A, NF-κB signaling components, and fibrosis-related proteins revealed that secukinumab treatment led to a significant reduction in the levels of IL-17A, NF-κB, IL-1β, and TGF-β1 compared to the control group, regardless of the ITO NP size (P < 0.05) (Figure A). Additionally, while IκBα levels were elevated in response to ITO NP exposure, secukinumab treatment further enhanced this elevation (P < 0.05) (Figure B–F). These results support the notion that ITO NPs activate the NF-κB signaling pathway via IL-17A, promoting inflammatory responses and fibrosis.

8.

Secukinumab blocks the effect of IL-17A on the expression of key molecular events in RAW264.7 cells after ITO NPs treatment. (A) Representative images of IL-17A, NF-κB, IκBα, IL-1β, and TGF-β1 protein expression in the lung tissue. (B–F) Protein quantitative analysis of of IL-17A, NF-κB, IκBα, IL-1β, and TGF-β1 (n = 3). (Results are presented as mean ± standard error of the mean (SEM), n = 3, compared with the control group, *P < 0.05; comparison between intervention groups, ^# P < 0.05.)

3.9. Inhibition of the NF-κB Signaling Pathway Can Alleviate the Inflammatory Response Induced by ITO NPs

Further supporting the upstream and downstream relationship between IL-17A and NF-κB signaling, we employed DHMEQ, an irreversible NF-κB nuclear translocation inhibitor, to pretreat RAW264.7 cells. This intervention resulted in a marked decrease in NF-κB expression (Figure A, C) and a reversal of the decrease in IκBα expression (Figure A, D). Furthermore, DHMEQ downregulated IL-1β, a downstream target of the NF-κB pathway (Figure A, F) and also reduced the expression of the fibrosis-related factor TGF-β1 (Figure A, E). These findings underscore the NF-κB pathway as a central regulatory hub for inflammatory factors while also highlighting its critical role in promoting tissue fibrosis. Notably, our results indicated that DHMEQ did not diminish IL-17A protein levels (Figure A, B), suggesting that NF-κB does not exert negative feedback regulation on IL-17A expression.

9.

DHMEQ blocks the effect of NF-κB on the expression of key molecular events in RAW264.7 cells after ITO NPs treatment. (A) Representative images of IL-17A, NF-κB, IκBα, and IL-1β protein expression in lung tissue. (B–F) Protein quantitative analysis of IL-17A, NF-κB, IκBα, TGF-1β, IL-1β (n = 3). (Results are presented as mean ± standard error of the mean (SEM), n = 3, compared with the control group, *P < 0.05; comparison between intervention groups, ^# P < 0.05.)

In conclusion, our findings illustrate that ITO NPs contribute to the development of pulmonary fibrosis by promoting IL-17A expression and activating NF-κB inflammatory pathways. Thus, IL-17A emerges as a pivotal factor in ITO NP-induced lung fibrosis in mice, initiating critical molecular events that lead to this pathology.

3.10. Construction and Key Event Relationships (KER) Evaluation of AOP Networks

Based on the above experimental mechanism verification, we further revised the AOP framework of ITO NP-induced occupational lung injury based on Figure to obtain Figure . KERs are pivotal in AOPs, as they establish quantitative, predictive connections between key events (KEs). These relationships are essential for crafting and refining quantitative disease models that enhance the precision of disease outcome forecasting. In our research, we employed the Bradford Hill criteria and the QWOE (Quantitative Weight Of Evidence) method to assign weights to KERs, thereby enhancing the clarity and scientific rigor of the AOP analysis. Previous studies have provided more detailed information on the rationale for weighting factors and scoring criteria. , Here, through previous research, we further review and discuss the logical relationships between key events in the AOP framework (Table ).

10.

AOP framework of ITO-induced pulmonary fibrosis was established based on experimental research.

2. Qualitative Evaluation and Quantitative Rating in the AOP for ITO NPs-Mediated Pulmonary Fibrosis.

Qualitative evaluation of AOP
MIE: Oxidative stress (KE 1392)
1) Supporting data:
Silica nanoparticles promote the transformation of proinflammatory macrophages and foam cells through ROS/PPARgamma/NF-kappaB signaling.
Silica nanoparticles induce pulmonary inflammation in mice through ROS/PARP/TRPM2 signaling mediated lysosomal damage and autophagy dysfunction.
Diesel exhaust particles(DEP) increased IL-17A expression via ROS/NF-κB in airway epithelium. Pretreatment with a ROS scavenger (NAC) significantly inhibited DEP-induced IL-17A mRNA expression.
In human normal bronchial epithelial cells (BEAS-2B), titanium dioxide nanoparticles induced a dose-dependent increase in intracellular ROS levels.
ZIF-8 NPs concentration led to a significant increase in ROS levels and the activities of CAT, SOD, and MDA in zebrafish, indicating a dose - dependent oxidative stress response to ZIF-8. Astaxanthin was found to specifically inhibit ROS - mediated oxidative stress and thus could mitigate ZIF-8 - induced oxidative damage.
Human lung cancer A549 cells treated with aminated polystyrene nanoparticles exhibit time- and dose-dependent lipid deposition, ROS formation, and biochemical damage related to oxidative stress.
2) Potentially inconsistent data: N/A.
KE1:Increase IL-17A
1) Supporting data:
Six-week-old C57BL/6J male mice were exposed to PM or filtered air for 16 weeks in a real-ambient PM exposure system, found that the IL-17 signaling pathway mediated immune dysregulation in PM-induced chronic lung injuries, knockdown of IL-17A significantly alleviated the damage.
IL-17 induces phagocytosis in mouse macrophages and induces an increase in a soluble form of the phagocytic receptor, for example, lectin-like oxidized low-density lipoprotein receptor-1 as well. It can reduce inflammatory responses by suppressing STAT3/IL-17/NF-κB in a mouse model of bacterial acute pneumonia pathway.
The combined administration of peimine, peiminine, and forsythoside A ameliorated inflammatory response in acute lung injury mice synergistically induced by LPS, the mechanism is related to the dampening of the TLR4/MAPK/NF-κB signaling pathway and IL-17 activation.
2) Potentially inconsistent data: N/A.
KE2: Increased TNF-α (KE 1577)
1) Supporting data:
In the workers exposed to indium (at least 6 h per day for one year), the serum indium level was positively correlated with the serum levels of SP-A, IL-1β, and IL-6.
Exposure of Sprague–Dawley (SD)rat models to four different forms of indium compounds (ITO group, indium oxide (In2O3), indium sulfate (In2(SO4)3), indium chloride (InCl3) significantly increased serum indium and lung indium levels, as well as BALF levels of TNF-α with increasing doses.
The elevated macrophage-derived exosomal TNF-α mediates the transmission of inflammatory information between alveolar macrophages and lung epithelial cells, exacerbating the development of PM2.5-induced acute lung injury. In addition, TNF-α-containing exosomes increased surfactant protein (SP) expression in MLE-12 cells.
2) Potentially inconsistent data: N/A.
KE3: Increased activation NF-kB (KE 1172)
1) Supporting data:
Instillation of nano-ITO in SD rats induces proteinosis (PAP) by activating NF-κB signaling pathway.
In2O3, In2(SO4), InCl3 increase the expression of NF-κB in lung tissue.
2) Potentially inconsistent data: N/A.
KE4: Decreased IL-1 production (KE1571)
1) Supporting data:
Sintered ITO (SITO) induced robust cytokine production (IL-1β, IL-6, TNFα, and IL-8) within 24 h in both RAW 264.7 mouse macrophages.
Desert dust induced the release of pro-inflammatory cytokines including IL-1β and TNF-α in the coculture system of alveolar epithelial A549 cells and THP-1 macrophages.
Timed pregnant Sprague–Dawley rats were treated with PM2.5, the concentrations of IL-1, IL-6, and TNF-α in the lung tissue of offspring rats were significantly increased by 2.36-fold, 3.91-fold, and 4.36-fold, respectively.
2) Potentially inconsistent data: N/A.
KE5: Increased TGF-1β (KE 276)
1) Supporting data:
Enhanced fibrosis was observed in BEAS-2B human bronchial epithelial cells treated with silica particles. TGF-β inhibitor can down-regulate TGF-β expression and inhibit the pathogenesis of epithelial-mesenchymal transition (EMsT). Another study found that HMGB1 induces fibroblast to myofibroblast differentiation of lung fibroblasts via NF-κB predominate - mediated TGF-β1 release.
2) Potentially inconsistent data: N/A.
KE6: Increased inflammatory immune responses (KE750)
1) Supporting data:
In rats exposed to different doses of ITO powder, inflammatory factors present a dose-dependent increase in alveolar lavage fluid and promote inflammation of the lungs. ,
Inflammatory cell infiltration induced by ITO nanoparticles into the lungs of rats after intratracheal administration. −
Other studies have found that Sintered indium and tin oxide particles induce pro-inflammatory responses in Raw 264.7 cells through inflammasome activation.
2) Potentially inconsistent data: N/A.
KE7: Microvascular leakage
1) Supporting data:
A single inhalation exposure to In2O3 NPs rats resulted in the development of PAP. ,
2) Potentially inconsistent data: N/A.
KE8: Pulmonary fibrosis (KE 1458)
1) Supporting data:
Instillation of nano-ITO in SD rats induces Pulmonary fibrosis. ,,
2) Potentially inconsistent data: N/A.

A considerable body of research conducted among occupational populations has consistently reported that occupational dust exposure is associated with the development of various lung diseases, including inflammation, fibrosis, impaired lung function, and even lung cancer. In the specific case of indium lung disease, it is well established that exposure to ITO NPs plays a crucial role as the primary etiological factor. Numerous studies have shown that ROS-mediated oxidative stress is a critical initial molecular event in particulate-induced lung injury. Upon entering the lungs, these particles trigger an oxidative stress response, leading to excessive ROS production. − ROS not only directly damage cellular structures but also activate multiple signaling pathways, initiating complex cascades. For example, ROS activate the IL-17A/NF-κB pathway, promoting pro-inflammatory cytokine release and inflammation. They also induce apoptosis-related pathways, causing cell death. The interplay between oxidative stress and inflammation further exacerbates lung tissue damage.

Concerning the relationship between IL-17A and pulmonary fibrosis, studies have consistently demonstrated that IL-17A-mediated immune dysregulation represents a significant mechanism underlying the pathogenesis of chronic lung injury after exposure to real environmental particulate matter. The interaction of IL-17A pathway with inflammatory signaling pathways, including NF-κB signaling, was also demonstrated. − Meanwhile, IL-17A has been shown to exacerbate airway disease in animal models. Therefore, combining with our experimental data IL-17A in the AOP framework of pulmonary fibrosis induced by ITO NPs is a well-documented MIE.

At the cellular level, after ITO NPs exposure, we can observe the activation of alveolar macrophages, which is an early defense mechanism of the body against an adverse environment. There were reports that ITO NPs can stimulate macrophages to release IL-1βinduce epithelial-mesenchymal transition in A549 cells and further participate in the process of pulmonary fibrosis. During this period, the occurrence of inflammatory response is an important link to aggravate lung injury. , Based on the above review, we conducted a quantitative evaluation of KER (Table ). At a score of 69.63%, this confidence of AOP was moderate (Table ).

3. Quantitative Rating of MIE and KEs in the AOP .

		Qualitative rating
Evolved Bradford Hill causal consideration	Assigned weight	MIE	KE1	KE2	KE3	KE4	KE5	KE6	KE7	KE8
Biological plausibility		Some in vivo and in vitro evidence suggests that the ITO NPs can cause pulmonary fibrosis.
Essentiality empirical support	0.4	(+++)1.2	0	(+++)1.2	(+++) 1.2	(++) 0.8	(+) 0.4	(+++) 1.2	(++) 0.8	(+++) 1.2
Dose and incidence concordance	0.2	(++) 0.4	(+) 0.2	(++) 0.4	(++) 0.4	(+++) 0.6	(++) 0.4	(+++) 0.6	(+) 0.2	(++) 0.4
Empirical support temporal concordance	0.2	(++) 0.4	(++) 0.4	(+++) 0.6	(++) 0.4	0	(+++) 0.6	(+++) 0.6	(+++) 0.6	(+) 0.2
Consistency across test systems	0.1	(++) 0.2	(++) 0.2	(+++) 0.3	(+) 0.1	(++) 0.2	(++) 0.2	(+) 0.1	(+) 0.1	(+) 0.1
Analogy multiple studies support KE and KER	0.1	(++) 0.2	(+++) 0.3	(+++) 0.3	(+++) 0.3	(++) 0.2	(++) 0.2	(++) 0.2	(+) 0.1	(+) 0.1
Score		2.4	1.1	2.8	2.4	1.8	1.8	2.7	1.8	2

^aAOP-score = (actual KE scores)/(total possible scores) = (18.8/27) = 69.63%. Conclusion: The confidence of this AOP to induce pulmonary fibrosis was moderate. N/A: not available.

4. Discussion

The process of establishing the AOP framework necessitates the determination of damage effects at the individual level, as this is crucial for the prediction of the systemic toxicity of chemicals. According to epidemiological data on the population, the earliest manifestations of indium lung disease in occupational populations exposed to ITO NPs were pulmonary inflammation and subsequent pulmonary fibrosis. ,− Our experimental design was grounded in workplace exposure data demonstrating that airborne indium-containing particles in manufacturing settings predominantly exhibit diameters below 100 nm, with a substantial fraction measuring under 20 nm. This is corroborated by Jin et al., whose TEM analysis of occupational ITO samples identified primary particles within the 20–50 nm range, consistent with the 30–50 nm diameter fibrous structures observed during membrane synthesis processes in related studies. These findings collectively confirm that our tested size ranges (10–100 nm) appropriately reflect occupationally relevant nanoparticle dimensions. In this study, we chose the persistent subacute exposure model which exerts twice-a-week oropharyngeal aspiration exposure in mice for 4 weeks. First, we validated the pulmonary pathology caused by ITO, which is the same as our previous research and similar to several other works. Apart from neutrophil infiltration, and thickened bronchi, ITO also triggers the lung tissue to secret much mucus and collagens to invade the solid lung tissue and gradually proceed to lung fibrosis. Such pathological changes have been validated by us and other groups. ,,, Then, LDH, total proteins, and cytokines of the BALF were used to assess the lung injury caused by ITO. Increasing total proteins and LDH after treatment indicated that a high dose of both sized ITO can induce the lung tissue to produce plenty of substances to protect against the ITO invasion and smaller size can boost more LDH, and total protein to protect the lung against the injury. So, it is obvious that 28-day repeated exposure challenged the lung tissue constantly and inflammatory response was in progress during the exposure time. Other metal oxide NPs like copper oxide NPs can cause lung inflammation response featuring higher levels of neutrophils and white blood cells. After 28-day exposure to the well-known occupational chemical, silica NPs also mainly cause increased neutrophils in the BALF and the incremental cell number would reduce compared with 24h after exposure to silica NPs. , Therefore, we finally identified ITO-induced pathological damage at the individual level as pulmonary fibrosis, AO in the AOP framework.

To gain insight into the key molecular events of ITO NPs inducing pulmonary fibrosis. For pulmonary fibrosis from the microscopic level, we conducted preliminary confirmation through two distinct avenues. First, pulmonary fibrosis-associated genes were collated from the GeneCards and DisGeNET databases, and an interaction analysis was conducted with the differential genes identified in the RNA sequencing of lung tissue from mice exposed to ITO NPs. Overlapping genes were subjected to KEGG and GO analysis. The annotation of these genes indicated that the IL-17A, TNF-α, and NF-κB signaling pathways played a significant role in the process of lung inflammation and fibrosis. It is worth noting that the protein interaction network hints that IL-17 and TNFSF11 have interaction relations. TNFSF11 is a type II transmembrane protein that functions as a receptor ligand for NF-κB (RANK). RANKL, an activator of NF-κB, binds to NF-κB and induces the differentiation of monocyte/macrophage-lineage cells into osteoclasts, as well as the maturation of osteoclast precursors. Furthermore, we employed molecular docking to predict the interaction between IL-17A and TNFSF11. Our findings indicated that the amino acid residues of IL-17A and TNFSF11 could interact through hydrogen bonds to form a stable complex, which could activate downstream reactions. The data presented above led to the establishment of the AOP framework for ITO NPs inducing pulmonary fibrosis. This finding provides a molecular-level clue for the construction of our AOP framework, which posits that ITO NPs induce pulmonary fibrosis. Subsequently, a cluster analysis of DEGs was performed using a heat map and GO analysis to further complement the key molecular events at the cellular and tissue levels. This framework was subsequently validated through in vivo and in vitro studies.

The pathological features of pulmonary fibrosis can be broadly categorized into three hallmarks: the accumulation of myofibroblasts, excessive deposition of collagen, and ECM remodeling. − The transformation of lung fibroblasts into myofibroblasts plays an important role in the process of pulmonary fibrosis. TGF-β1 can induce the epithelial-mesenchymal transition (EMT) process of alveolar epithelial cells, activate the activity of myofibroblasts, and express collagen and fibronectin (FN), leading to abnormal deposition of extracellular matrix and drive the occurrence of pulmonary fibrosis. α-SMA is a specific marker of myofibroblasts, while increased FN indicates excessive deposition of collagen. − The present study demonstrated that ITO NPs could induce the elevation of TGF-β1, α-SMA, and FN in lung tissue, thereby leading to pulmonary fibrosis. Furthermore, we employed a method of detecting IL-17A and IL-1β in lung tissues to ascertain the extent of lung inflammation caused by ITO NPs. Once bound to its receptor, IL-17A is activated and recruits a variety of inflammatory cells, including macrophages and neutrophils, to exacerbate the local inflammatory response. − IL-1β is a pivotal cytokine in the regulation of the inflammatory response in vivo. The results demonstrated that IL-17A and IL-1β were increased in a dose-dependent and size-dependent manner following ITO NPs exposure. Notably, IL-6 secreted by macrophages can promote the occurrence of pulmonary fibrosis in fibrotic diseases through the IL-6/STAT3/Smad3 axis. By immunofluorescence experiments, we found that the distribution in the surrounding lung cells and macrophages secrete large amounts of IL-6 inflammatory factor, and in the small size of ITO NPs in the exposed group macrophage release effect is more significant. These phenomena have become our further refined ITO NPs fibrosis associated with the establishment of the AOP framework.

Through the above the KEGG enrichment results and immunohistochemistry results indicated that the IL-17 signaling pathway was affected by ITO NPs exposure and that lung injury caused by ITO NPs was characterized by inflammation and fibrosis. First, we utilized the RAW264.7 cell model to investigate the regulatory mechanism of IL-17 signaling in vitro. In our study, we found that the exposure of ITO NPs induced macrophages to oxidative stress as well as general toxic manifestations such as decreased survival rate.

Previous studies have shown that ITO NPs increased the level of IL-17A. Pretreatment with Secukinumab was found to reduce the increase in inflammatory factors (IL-17A, IL-6, TNF-α, MCP-1, IL-1β) and pro-fibrosis factors (TGF-β1) caused by ITO NPs. These results indicated that the production of intracellular inflammatory factors induced by ITO NPs was closely related to the activation of IL-17A in RAW264.7 cells. Following the binding of IL-17A to its receptor IL-17RA, the intracellular SEFIR domain and Toll/IL-1R-like loop of IL-17RA bind to adaptor protein ACT-1, which recruits tumor necrosis factor receptor-associated factor 6 (TRAF6). , The results in the triggering of a downstream signaling cascade. It induces the activation of NF-κB. Second, NF-κB is a key component of several common respiratory diseases, including asthma, lung cancer, pulmonary fibrosis, COPD, pneumonia, and tuberculosis. − Researchers demonstrated that airborne particles can recruit inflammatory cells and release cytokines and reactive oxygen species (ROS) during their passage through the airway, subsequently activating different pathways, including NF-κB, MAP kinase, and Stat-1, to mediate respiratory diseases such as asthma, chronic obstructive pulmonary disease, pulmonary fibrosis, and even cancer. Furthermore, it has been demonstrated that the inhibition of NF-κB signaling pathway activation can reduce pulmonary inflammation and fibrosis. − Finally, in our experiments, we through DHMEQ inhibited the NF-κB signaling pathways, significantly reduced the ITO NPs after treatment, the macrophages of the expression of inflammatory factor and coarse fibrosis factor, and the expression of IL-17A was not affected. Therefore, it can be concluded that ITO NPs activate the NF-κB inflammatory pathway by inducing IL-17A expression and further contribute to the development of pulmonary fibrosis.

In our study, the hierarchical role of particle size in the AOP framework necessitates particular emphasis, as our data establish nanoscale dimensions as a primary driver of oxidative stress cascades. As quantitatively demonstrated in Figure S1, ROS generation caused by n-ITO NPs was higher than that caused by m-ITO NPs at equivalent mass doses. Another similar study showed that 22 nm SiO₂ NPs increased superoxide dismutase and inhibited glutathione S-transferase activity in Ceriodaphnia reticulata, while the larger SiNP had no significant effect on the organism at the concentration tested. , Other studies showed that increased ROS production further upregulated IL-17A expression. Within the AOP architecture, we therefore position particle size as a KE preceding oxidative stress ROS. The physical dimension critically controls the ″hierarchical triggering″ of molecular initiation events. Our experiments show that, at the same dose, smaller ITO NPs are more toxic than larger ones. Thus, ITO NP size may be a key upstream node, not just a material property descriptor. Finally, we constructed a detailed AOP framework for ITO NPs inducing pulmonary fibrosis. And, we improved the transparency and scientific interpretation of the AOP by weighing the evidence based on the quantitative weight of evidence (QWOE) approach. Then the Bradford Hill causal considerations are weighted based on the numerical order of importance and the MIE, KEs are scored. Composite scores and confidence scores for the overall AOP. Finally, the confidence in the AOP framework we have established was found to be moderate.

5. Conclusions

Based on the above, we proposed the AOP framework of ITO NPs induced pulmonary fibrosis. In our study, ITO NPs could cause lung inflammation and further develop into pulmonary fibrosis in mice, which was closely related to the phagocytosis and pro-inflammatory function of pulmonary macrophages. IL-17A was involved in the process of ITO NP-induced lung tissue fibrosis by regulating the NF-κB signaling pathway in vitro. In addition, we validated the critical role of these key molecular events in the AOP framework using dual inhibitors. Notably, IL-17A is the molecular initiating event in this AOP. These findings can help predict the molecules of ITO NPs induced lung injury induced by early initiation and key events, and use the IL-17A in lung activation as the basis, a development based on the mechanism and predictive markers, to assess the nanoparticles due to occupational exposure cause respiratory disease. In conclusion, our study provides evidence for the threat and risk assessment of ITO NPs exposure to respiratory disease, which can inform decision-making to manage this disease.

Glossary

ABBREVIATIONS

AOP

Adverse Outcome Pathway

AO

adverse outcome

BALF

bronchoalveolar lavage fluid

BLM

bleomycin

BP

biological process

BSA

bovine serum albumin

CC

cellular component

DEGs

differentially expressed genes

FN

fibronectin

GSH

glutathione

IHC

immunohistochemistry

ITO

Indium Tin Oxide

KE

key events

KERs

Key Event Relationships

LDH

lactate dehydrogenase

MF

molecular function

MIE

molecular initiating event

NPs

Nanoparticles

PAP

pulmonary alveolar proteinosis

PAS

Periodic Acid-Schiff stain

PBS

phosphate-buffered saline

pSTAT3

phosphorylated STAT3

ROS

reactive oxygen species

RT

glutathione; room temperature

SPF

Specific Pathogen Free

TEM

transmission electron microscopy

α-SMA

α-smooth muscle actin

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/envhealth.5c00053.

• Tables detailing antibody specifications utilized for experimental analyses, associated quantitative data sets, and cluster analysis (heatmap visualization) of differentially expressed genes identified through transcriptomic profiling of murine pulmonary tissues following ITO NP exposure; figure illustrating comparative cytotoxic responses elicited by size-variant ITO NPs in macrophage-derived RAW264.7 cell cultures (PDF)

#.

C.Z., Y.C., and J.Q. are considered as co-first authors.

The authors declare no competing financial interest.

Published as part of Environment & Health special issue “Grand Environmental Challenge: Indoor Air Pollution, Health Effects, and Mitigation”.