Acorus gramineus Solander Preserves Corneal Epithelial Barrier Integrity against PM2.5-Induced EMT-Like Remodeling via Suppression of TGF-β/Smad Signaling Authors: Min Yeong Kim, Su Hyun Hong, Yung Hyun Choi Published: 2026 ABSTRACT AbstractObjectivesTo evaluate the effects of particulate matter ≤ 2.5 µm (PM2.5) on corneal epithelial barrier integrity and to examine the protective potential of Acorus gramineus Solander root extract (AGS).MethodsHuman corneal epithelial cells (hCECs) were cultured as three-dimensional spheroids and exposed to PM2.5 with or without AGS treatment. Barrier function, tight junction (TJ) integrity, epithelial–mesenchymal transition (EMT)-related markers, inflammatory mediators, and transforming growth factor (TGF)-β/Smad signaling were analyzed.ResultsPM2.5 induced structural instability of hCEC spheroids without substantial cytotoxicity, accompanied by decreased transepithelial electrical resistance and disruption of TJ proteins such as zonula occludens (ZO)-1, ZO-3, and occludin. These alterations coincided with EMT-like remodeling, characterized by the upregulation of N-cadherin and vimentin, increased expression of matrix metalloproteinases, and elevated production of inflammatory cytokines. These effects were associated with the TGF-β/Smad signaling pathway activation. AGS restored barrier integrity, suppressed EMT-like remodeling and inflammation, and inhibited TGF-β/Smad pathway activation.ConclusionPM2.5 impairs corneal epithelial barrier function through EMT-related and inflammatory mechanisms. AGS exerts protective effects by preserving TJ integrity and suppressing TGF-β/Smad signaling, suggesting its therapeutic potential against air pollution–induced ocular surface damage.Keywords: Acorus gramineus Solander, corneal epithelial barrier, epithelial–mesenchymal transition, particulate matter 2.5, TGF-β/Smad signaling INTRODUCTION The ocular surface is constantly exposed to environmental stressors such as airborne particulate matter (PM), necessitating a well-organized epithelial barrier to maintain tissue homeostasis and visual function [1, 2]. Fine PM with a diameter of ≤ 2.5 μm (PM2.5) is a major environmental pollutant associated with ocular discomfort, dry eye disease, and corneal epithelial damage [3, 4]. Evidence indicates that PM2.5 induces oxidative stress and inflammatory responses in corneal epithelial cells [4-6], disrupting tight junctions (TJs) and increasing epithelial permeability even in the absence of overt cytotoxicity [7, 8]. Nonetheless, the molecular mechanisms underlying PM2.5-induced epithelial instability and potential protective strategies remain incompletely understood. Recent studies have suggested that epithelial barrier dysfunction induced by environmental pollutants is closely associated with epithelial plasticity and epithelial–mesenchymal transition (EMT)-like remodeling [9, 10]. Partial or early EMT is characterized by the loosening of cell–cell junctions, switching of cadherins, remodeling of extracellular matrix, and activation of inflammatory and profibrotic signaling pathways, such as transforming growth factor (TGF)-β/Small mothers against decapentaplegic (Smad) signaling [11, 12]. In ocular surface tissues, these EMT-like changes have been associated with chronic inflammation, impaired wound healing, and compromised epithelial integrity [13, 14]. However, whether PM2.5 induces EMT-associated barrier remodeling in corneal epithelial spheroids and how this process can be therapeutically modulated remains largely unexplored. Acorus gramineus Solander root, known as Seokchangpo in traditional Korean medicine, has been used to treat inflammatory disorders, neurological conditions, and sensory organ dysfunctions [15, 16]. Phytochemical and pharmacological studies have shown that A. gramineus root extract (AGS) and its bioactive compounds possess strong antioxidant, anti-inflammatory, and cytoprotective properties, including the inhibition of reactive oxygen species production and inflammatory cytokine expression [16-20]. Traditional medical texts and modern experimental studies also have reported that AGS exerts protective effects on mucosal and epithelial tissues [17, 21], suggesting its potential as a barrier-stabilizing herbal medicine. Despite its historical clinical use and established anti-inflammatory effects, the effects of AGS on environmental pollutant–induced epithelial barrier dysfunction, particularly that caused by ocular surface exposure to PM2.5, have not been systematically studied. Considering the rising incidence of air pollution–related ocular diseases and the absence of effective preventive measures, exploring herbal medicines that can preserve epithelial integrity by targeting early molecular events rather than irreversible tissue damage is important. Therefore, this study examined the effects of PM2.5 on the structural stability and barrier function of human corneal epithelial cell (hCEC)-derived spheroids and assess whether an ethanol extract of AGS could alleviate PM2.5-induced epithelial instability. We focused on TJ integrity, EMT-like phenotypic changes, inflammatory responses, matrix metalloproteinase (MMP) activation, and TGF-β/Smad signaling to elucidate the molecular mechanisms underlying the protective effects of AGS. This study provides mechanistic insights into the barrier-protective potential of AGS and underscores its therapeutic relevance. MATERIALS AND METHODS 1. AGS preparation and treatment Dried roots of A. gramineus (Daehan Saengyak Products Co., Ltd., Miryang, Republic of Korea) were extracted using 70% ethanol by ultrasonication. The extracts were filtered, concentrated under reduced pressure using a rotary evaporator (EYELA Co., Ltd., Shanghai, China), and lyophilized to obtain a powdered extract (AGS), which was stored at –20℃. AGS was dissolved in dimethyl sulfoxide (DMSO; Amresco, Solon, OH, USA) to prepare a 100 mg/mL stock solution and subsequently diluted with cell culture medium before application to the cells. 2. PM2.5 treatment The PM2.5 used in this study was diesel exhaust PM (Standard Reference Material 1650b; SRM 1650b) obtained from the National Institute of Standards and Technology (NIST, Gaithersburg, MD, USA) and purchased through Sigma-Aldrich (St. Louis, MO, USA). SRM 1650b is a certified urban diesel PM composed mainly of fine carbonaceous particles containing polycyclic aromatic hydrocarbons (PAHs), nitro-substituted PAHs, trace metals, and other combustion-derived organic compounds. Stock solutions were prepared by suspending PM2.5 in DMSO, followed by vigorous vortexing and sonication to minimize particle aggregation. Working solutions were freshly prepared by dilution in culture medium immediately before each experiment. 3. Two-dimensional cell culture and three-dimensional spheroid formation We procured hCECs from the American Type Culture Collection (Manassas, VA, USA). Two-dimensional (2D) cell cultures were performed using a 1:1 (v/v) mixture of Dulbecco’s modified Eagle’s medium/Nutrient Mixture F-12 and keratinocyte serum-free medium, supplemented with 10% fetal bovine serum, bovine pituitary extract (25 mg), epidermal growth factor (2.5 µg), and 1% penicillin/streptomycin. Cultures were incubated at 37℃ in a 5% CO2 atmosphere. All media and reagents were obtained from WELGENE (Gyeongsan, Korea) and Thermo Fisher Scientific (Carlsbad, CA, USA). To construct three-dimensional (3D) hCEC spheroids, cells were seeded at a constant density onto ultra-low-attachment plates (Corning Inc., Corning, NY, USA) and cultured. After 3 days, the cells self-aggregated into spheroids. Spheroids of uniform shape and size were selected under a phase-contrast microscope (ZEISS, Oberkochen, Germany) for further experiments. Following treatment with AGS and PM2.5, alone or in combination, representative images were captured from each experimental group, and the spheroid area was quantified using ImageJ software (National Institutes of Health, Bethesda, MD, USA). 4. Cell viability analysis In 2D cell culture experiments, hCECs were seeded onto culture dishes and cultured for 24 h until they reached the appropriate density. The cells were then treated with AGS, PM2.5, or both for an additional 24 h. For 3D spheroid experiments, hCECs were initially treated with AGS for 1 h, harvested, and redistributed into spheroid microplates. The resulting spheroids were treated with PM2.5, alone or in combination with AGS, for 3 days. Cell viability was assessed using the Cell Counting Kit-8 (CCK-8; Abcam, Inc., Cambridge, UK). According to the manufacturer’s instructions, CCK-8 reagent was added to the cell culture medium and incubated for 2 h. The culture medium was then transferred to a 96-well plate, and the absorbance was measured at 450 nm using a microplate reader (VERSA Max, Molecular Devices, Sunnyvale, CA, USA). Cell viability was calculated as the change in absorbance compared with that of the control group [22]. 5. Measurement of lactate dehydrogenase release To evaluate the extent of cell membrane damage, culture supernatants were collected, and lactate dehydrogenase (LDH) release was measured using an LDH assay kit (Thermo Fisher Scientific). According to the manufacturer’s instructions, the culture supernatants were reacted with the assay reagent, and absorbance was measured using a microplate reader. LDH activity was calculated from the recorded values [23]. 6. Measurement of transepithelial electrical resistance Barrier function was evaluated by measuring transepithelial electrical resistance (TEER) using an epithelial voltohmmeter (World Precision Instruments, Sarasota, FL, USA). TEER values for 3D spheroids and 2D monolayers cultured on permeable inserts were recorded before and after treatment and normalized to surface area. Results were expressed as Ω·cm2 [23]. 7. Immunofluorescence staining Spheroids formed under treatment with PM2.5 and AGS, alone or in combination, were fixed with paraformaldehyde, permeabilized, and blocked before incubation with primary antibodies against zonula occludens (ZO)-1, ZO-3, occludin, or phosphorylated Smad2 (p-Smad2). After washing, the samples were incubated with appropriate fluorescent secondary antibodies. Nuclei were counterstained with 4’,6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich). Images were acquired using a fluorescence microscope, and representative images were selected for analysis [24]. 8. Western blot analysis Total protein was extracted from cells and spheroids using RIPA buffer (Sigma-Aldrich), which contained protease and phosphatase inhibitors. Equal amounts of protein were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride membranes (Merck Millipore, Darmstadt, Germany). The membranes were incubated with primary antibodies targeting TJ proteins (ZO-1, ZO-3, and occludin), EMT-related markers (N-cadherin and vimentin), MMPs (MMP-1, MMP-2, and MMP-3), inflammatory cytokines (tumor necrosis factor [TNF]-α, interleukin [IL]-6, IL-1β), TGF-β1, p-Smad2, total Smad2, and actin. Following incubation with horseradish peroxidase–conjugated secondary antibodies, signals were detected using an enhanced chemiluminescence system (Thermo Fisher Scientific) [25]. The antibodies used in this study were obtained from Cell Signaling Technology (Beverly, MA, USA), Bioworld Technology, Inc. (St. Louis Park, MN, USA), and Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). 9. Enzyme-linked immunosorbent assay Secreted MMP-2, IL-1β, IL-6, and TGF-β1 levels in culture supernatants were quantified using commercial enzyme-linked immunosorbent assay (ELISA) kits (R&D Systems, Minneapolis, MN, USA) according to the manufacturer’s protocols. 10. Statistical analysis All experiments were conducted in triplicate, and data are presented as mean ± standard deviation. Statistical significance was assessed using one-way analysis of variance, followed by appropriate post hoc tests. A p-value < 0.05 was considered statistically significant. RESULTS 1. PM2.5 induces structural spreading and instability of hCEC-derived spheroids, which is alleviated by AGS Spheroids derived from hCECs were cultured for 3 days and exposed to PM2.5 (50 μg/mL) with or without AGS (0.5 μg/mL) for an additional 48 h to assess spheroid morphology, particulate accumulation, and cytotoxicity. Although PM2.5 did not completely disintegrate the spheroids, it caused significant structural spreading and instability, as evidenced by the increased spheroid area and greater PM2.5 accumulation within the spheroids (Fig. 1A, B). Cotreatment with AGS effectively inhibited PM2.5-induced spheroid spreading and preserved compact spheroid morphology. PM2.5 exposure slightly decreased cell viability but did not significantly increase LDH release, indicating no overt cytotoxicity (Fig. 1C). Furthermore, AGS significantly restored cell viability without causing membrane damage (Fig. 1D). These results suggest that PM2.5 primarily induces structural instability rather than disintegrating hCEC spheroids and that AGS preserves spheroid integrity and cellular viability. Effects of particulate matter ≤ 2.5 µm (PM2.5) and Acorus gramineus Solander root extract (AGS) on the structural stability of human corneal epithelial cell (hCEC)-derived spheroids. hCEC spheroids were formed for 3 days and subsequently treated with PM2.5 (50 μg/mL) in the presence or absence of AGS (0.5 μg/mL) for 48 h. (A) Representative phase-contrast and DAPI-stained images showing spheroid morphology and PM2.5 accumulation. PM2.5 treatment induced spheroid spreading and structural instability rather than complete structural disruption. (B) Quantitative analysis of spheroid area. (C) Lactate dehydrogenase release assay. (D) Cell viability analysis. Data are presented as mean ± standard deviation. *p < 0.05, ***p < 0.001 versus control; ###p < 0.001 versus PM2.5-treated group. 2. PM2.5-induced spheroid instability is associated with TJ dysfunction, which is mitigated by AGS To evaluate the association between PM2.5-induced spheroid instability and TJ dysfunction, we measured TEER in hCEC-derived spheroids. PM2.5 exposure significantly reduced TEER, indicating compromised barrier integrity, whereas cotreatment with AGS markedly restored TEER levels (Fig. 2A). To further examine TJ impairment under 2D culture conditions, hCECs were treated with AGS for 48 h, and cell viability was assessed to rule out cytotoxicity. AGS concentrations up to 0.5 μg/mL did not affect cell viability (Fig. 2B), making this concentration suitable for subsequent experiments. In 2D cultures, PM2.5 significantly decreased cell viability and TEER, whereas AGS cotreatment mitigated PM2.5-induced cytotoxicity and restored barrier function (Fig. 2C, D). These results suggest that PM2.5-induced epithelial instability is closely associated with TJ dysfunction and that AGS preserves epithelial barrier integrity. Effects of particulate matter ≤ 2.5 µm (PM2.5) and Acorus gramineus Solander root extract (AGS) on tight junction (TJ) integrity in human corneal epithelial cell (hCEC)-derived spheroids and 2D hCEC cultures. (A) Transepithelial electrical resistance (TEER) in 3D hCEC spheroids following treatment with PM2.5 (50 μg/mL) in the presence or absence of AGS (0.5 μg/mL), indicating changes in TJ integrity. (B) Cell viability of 2D-cultured hCECs following 48 h of treatment with increasing concentrations of AGS. (C) Cell viability of 2D-cultured hCECs treated with PM2.5 or AGS. (D) TEER in 2D hCEC monolayers following PM2.5 and AGS treatment. Data are presented as mean ± standard deviation. **p < 0.01, ***p < 0.001 versus control; ##p < 0.01, ###p < 0.001 vs. PM2.5-treated group. 3. PM2.5-induced spheroid instability is associated with impaired TJ protein expression, which is restored by AGS Considering the PM2.5-induced spheroid instability and reduced TEER, we examined whether these changes were associated with altered TJ protein expression. Immunofluorescence staining showed that PM2.5 significantly decreased the expression and disrupted the distribution of the TJ proteins ZO-1, ZO-3, and occludin in hCEC-derived spheroids (Fig. 3A-C). In contrast, AGS cotreatment effectively preserved the expression and localization of these proteins and maintained a more compact and organized spheroid structure. Western blot analysis confirmed these findings, revealing that PM2.5 exposure resulted in a notable downregulation of ZO-1, ZO-3, and occludin protein levels, whereas AGS treatment restored their expression (Fig. 3D). In addition, TJ protein disruption was accompanied by a marked increase in the expression of N-cadherin and vimentin, mesenchymal markers associated with epithelial plasticity, indicating a phenotypic shift toward a mesenchymal-like state. Notably, AGS cotreatment effectively inhibited PM2.5-induced N-cadherin and vimentin upregulation, suggesting that AGS prevents abnormal remodeling of PM2.5-induced cell–cell adhesion. Effects of particulate matter ≤ 2.5 µm (PM2.5) and Acorus gramineus Solander root extract (AGS) on tight junction (TJ) protein expression in human corneal epithelial cell (hCEC)-derived spheroids. hCEC spheroids were treated with PM2.5 (50 μg/mL) in the presence or absence of AGS (0.5 μg/mL). (A-C) Representative immunofluorescence images showing the expression and distribution of the TJ proteins ZO-1 (A), ZO-3 (B), and occludin (C). Nuclei were counterstained with DAPI. Phase-contrast and merged images are shown. (D) Western blot analysis of ZO-1, ZO-3, occludin, N-cadherin, and vimentin expression, with actin as a loading control. 4. PM2.5-induced TJ dysfunction and EMT-like changes are associated with MMP upregulation and inflammatory responses suppressed by AGS Following the observed TJ disruption and EMT-like phenotypic changes after PM2.5 exposure, we investigated their association with MMP expression and inflammatory responses. PM2.5 significantly increased MMP-1, MMP-2, and MMP-3 protein levels, whereas AGS cotreatment effectively suppressed this upregulation (Fig. 4A). Consistent with the western blot findings, PM2.5 significantly increased secreted MMP-2 levels, which were reduced by AGS (Fig. 4B). PM2.5 also markedly increased the expression of proinflammatory cytokines such as TNF-α, IL-6, and IL-1β, whereas AGS cotreatment significantly mitigated these inflammatory responses (Fig. 4C). ELISA further showed that AGS significantly suppressed PM2.5-induced IL-1β and IL-6 secretion (Fig. 4D). These results suggest that PM2.5-induced epithelial barrier dysfunction and EMT-like remodeling are associated with increased extracellular matrix degradation and inflammatory signaling. Furthermore, AGS exerts protective effects by concurrently inhibiting MMP activation and inflammatory responses. Effects of particulate matter ≤ 2.5 µm (PM2.5) and Acorus gramineus Solander root extract (AGS) on MMP expression and inflammatory responses in human corneal epithelial cell (hCEC)-derived spheroids. hCEC spheroids were treated with PM2.5 (50 μg/mL) in the presence or absence of AGS (0.5 μg/mL). (A) Western blot analysis of matrix metalloproteinase (MMP)-1, MMP-2, and MMP-3 expression, with actin as a loading control. (B) Quantification of secreted MMP-2 levels by enzyme-linked immunosorbent assay (ELISA). (C) Western blot analysis of the proinflammatory cytokines tumor necrosis factor (TNF)-α, interleukin (IL)-6, and IL-1β. (D) Measurement of IL-1β and IL-6 secretion by ELISA. Data are presented as mean ± standard deviation. ***p < 0.001 versus control; ###p < 0.001 versus PM2.5-treated group. 5. PM2.5 activates TGF-β/Smad signaling associated with epithelial instability and EMT-like remodeling, which is suppressed by AGS Considering that PM2.5 exposure results in TJ dysfunction, EMT-like changes, and inflammatory responses, we investigated their association with the activation of the TGF-β/Smad signaling pathway, a key regulator of epithelial plasticity and EMT. PM2.5 significantly increased TGF-β1 protein expression and Smad2 phosphorylation without altering total Smad2 levels, indicating the activation of canonical TGF-β/Smad signaling (Fig. 5A). To determine whether PM2.5-induced Smad2 activation was accompanied by nuclear translocation, we performed cytoplasmic and nuclear fractionation analyses. PM2.5 markedly increased p-Smad2 accumulation in the nuclear fraction, with a corresponding decrease in the cytoplasmic fraction, confirming both Smad2 activation and nuclear translocation (Fig. 5B). AGS cotreatment significantly suppressed PM2.5-induced nuclear enrichment of p-Smad2, suggesting inhibition of Smad2 nuclear signaling. TGF-β/Smad signaling in particulate matter ≤ 2.5 µm (PM2.5)-induced epithelial instability and its inhibition by Acorus gramineus Solander root extract (AGS) in human corneal epithelial cell (hCEC)-derived spheroids. hCEC spheroids were treated with PM2.5 (50 μg/mL) in the presence or absence of AGS (0.5 μg/mL). (A) Western blot analysis of TGF-β1, phosphorylated Smad2 (p-Smad2), and total Smad2, with actin as a loading control. (B) Cytoplasmic (C.F.) and nuclear (N.F.) fractionation followed by immunoblot analysis of p-Smad2. Lamin B and actin were used as nuclear and cytoplasmic markers, respectively. (C) Representative immunofluorescence images showing p-Smad2 localization (red) in hCEC spheroids. Nuclei were counterstained with DAPI (blue). Phase-contrast and merged images are shown. (D) Quantification of secreted TGF-β1 levels in culture supernatants by enzyme-linked immunosorbent assay (ELISA). Data are presented as mean ± standard deviation. ***p < 0.001 versus control; ##p < 0.01 versus PM2.5-treated group. Consistent with these results, immunofluorescence analysis showed increased nuclear accumulation of p-Smad2 in PM2.5-treated spheroids, indicating transcriptional activation of downstream TGF-β signaling (Fig. 5C). In addition, quantitative analysis of secreted TGF-β1 levels using ELISA revealed that PM2.5 significantly increased extracellular TGF-β1 production compared with that in the control group (Fig. 5D), further confirming the activation of TGF-β signaling at the ligand level. Notably, AGS cotreatment markedly reduced PM2.5-induced TGF-β1 expression and Smad2 phosphorylation, decreased nuclear localization of p-Smad2, and suppressed TGF-β1 secretion. These findings indicate that PM2.5-induced epithelial barrier instability and EMT-like remodeling are mediated, at least in part, through the activation of the TGF-β/Smad signaling and that AGS suppresses this pathway. DISCUSSION Exposure to PM, particularly fine particles such as PM2.5, is increasingly recognized as a major environmental stressor that disrupts epithelial barrier integrity and triggers inflammatory responses in ocular surface tissues. Previous studies have shown that PM2.5 induces oxidative stress and activates inflammatory signaling in corneal epithelial cells, resulting in impaired TJ function and increased epithelial permeability [7, 26]. Consistent with these findings, PM2.5 exposure in this study caused structural instability of hCEC-derived spheroids and decreased TEER, suggesting early barrier dysfunction rather than overt cytotoxic damage. TJ disruption is a critical early event in epithelial barrier breakdown and is often associated with phenotypic plasticity and EMT-like changes [27, 28]. EMT is characterized by the loss of epithelial junctional proteins and cadherin switching, including increased expression of mesenchymal markers such as N-cadherin and vimentin, which facilitate epithelial loosening and tissue remodeling [29, 30]. Consistent with this pattern, PM2.5 significantly downregulated the TJ proteins ZO-1, ZO-3, and occludin while upregulating N-cadherin and vimentin expression. Similar PM-induced EMT-like responses have been documented in airway, skin, and retinal pigment epithelial models [31-33], in which environmental pollutants promote epithelial plasticity without immediate cell death [9, 34]. These findings suggest that PM2.5 induces early, partial EMT-like remodeling in human corneal epithelial cells, contributing to spheroid instability and barrier dysfunction. MMPs and inflammatory cytokines are key downstream effectors of epithelial remodeling and EMT. MMPs contribute to extracellular matrix degradation and junctional protein cleavage, thereby exacerbating epithelial barrier disruption [35, 36]. Concurrently, proinflammatory cytokines such as TNF-α, IL-6, and IL-1β destabilize TJs and promote EMT-associated signaling pathways [35, 37]. In this study, PM2.5 significantly elevated MMP-1, MMP-2, and MMP-3 levels and proinflammatory cytokine expression, supporting the involvement of coordinated extracellular matrix remodeling and inflammatory signaling in PM2.5-induced epithelial instability. Furthermore, TGF-β/Smad signaling serves as a central pathway connecting inflammation, EMT, and epithelial barrier dysfunction. Activation of TGF-β signaling results in Smad2 phosphorylation and nuclear translocation, triggering transcriptional programs that diminish epithelial traits and promote mesenchymal characteristics [38, 39]. Previous studies have indicated that PM2.5 activates TGF-β/Smad signaling in epithelial tissues, contributing to fibrosis and chronic barrier dysfunction [31, 40, 41]. Consistent with these findings, PM2.5 increased TGF-β1 expression and Smad2 phosphorylation along with increased nuclear localization of p-Smad2, indicating the activation of canonical TGF-β/Smad signaling in hCEC-derived spheroids. AGS effectively countered PM2.5-induced epithelial instability at multiple levels. It preserved TJ protein expression, suppressed mesenchymal marker upregulation, reduced MMP and inflammatory cytokine expression, and inhibited TGF-β/Smad pathway activation. These layered protective effects indicate that AGS preserves epithelial barrier integrity while preventing EMT-like remodeling and inflammatory amplification rather than merely reducing cytotoxic stress. Despite these findings, several limitations should be acknowledged. The experiments were conducted using an in vitro 3D spheroid model of human corneal epithelial cells, which—although more physiologically relevant than conventional monolayer cultures—cannot fully replicate the complexity of the in vivo ocular surface environment. In particular, interactions with tear film components, immune cells, and neural regulation are not represented. In addition, the PM2.5 exposure conditions may not precisely reflect real-world exposure scenarios in terms of concentration, duration, and compositional variability. Therefore, caution is warranted when extrapolating these findings to clinical settings. In addition to these limitations, the potential clinical applicability of AGS warrants consideration. Therapeutic use in ocular surface disorders would require appropriate formulation strategies, such as topical delivery systems (e.g., eye drops or gel-based formulations), to ensure stability, bioavailability, and sufficient ocular surface retention. Furthermore, in vivo validation using relevant animal models will be necessary to confirm the protective efficacy of AGS under physiological conditions as well as to evaluate pharmacokinetics, safety, and dosing strategies. Collectively, these efforts will be critical to bridge the gap between in vitro findings and the potential clinical applicability of AGS. CONCLUSIONS Overall, this study showed that PM2.5-induced epithelial barrier instability is associated with TJ disruption, EMT-like phenotypic remodeling, inflammatory responses, and TGF-β/Smad signaling. AGS exerts protective effects by suppressing these interconnected pathways (Fig. 6). These findings highlight the therapeutic potential of AGS as a barrier-protective agent against environmental pollutant–induced ocular surface damage. Proposed mechanism by which Acorus gramineus Solander root extract (AGS) protects corneal epithelial barrier integrity against particulate matter ≤ 2.5 µm (PM2.5)-induced damage. Exposure to PM2.5 induces oxidative stress and disrupts tight junction (TJ) integrity in human corneal epithelial cell (hCEC)-derived spheroids, accompanied by reduced barrier function and epithelial spheroid instability. PM2.5 also promotes epithelial–mesenchymal transition (EMT)-like remodeling, characterized by N-cadherin and vimentin upregulation, increased matrix metalloproteinase (MMP) expression, and enhanced inflammatory cytokine production. These changes are associated with activation of canonical transforming growth factor (TGF)-β/Smad signaling, including Smad2 phosphorylation. In contrast, AGS suppresses PM2.5-induced TGF-β/Smad activation, attenuates EMT-like changes and inflammatory responses, preserves TJ protein expression, and maintains epithelial barrier integrity, thereby preventing spheroid instability. ACKNOWLEDGEMENTS The authors would like to thank Core-Facility Center for Tissue Regeneration, Dong-Eui University (Busan, Republic of Korea), for letting us use fluorescence microscope. Footnotes AUTHOR'S CONTRIBUTIONS Conceptualization: Min Yeong Kim, and Yung Hyun Choi; Methodology: Min Yeong Kim and Su Hyun Hong; Investigation: Min Yeong Kim and Su Hyun Hong; Writing—Original draft preparation: Min Yeong Kim; Writing—review and editing: Yung Hyun Choi; Supervision: Su Hyun Hong and Yung Hyun Choi; Project administration: Yung Hyun Choi; Funding acquisition: Yung Hyun Choi. All authors have read and agreed to the published version of the manuscript. DATA AVAILABILITY Not applicable. CONFLICT OF INTEREST The authors have no conflicts of interest to declare. FUNDING This study was supported by the Basic Science Research Program grant (No. RS-2025-16064487 and RS-2026-25472831) from the National Research Foundation (NRF) of the Republic of Korea. REFERENCES