Endoplasmic reticulum stress-mediated disruption of mitochondrial homeostasis drives airway epithelial barrier fragility and enhances susceptibility to recurrent respiratory tract infections in children Authors: Xiaoyi Chang, Wei Jing, Qingbin Liu, Xueli Wu Published: 2026 ABSTRACT BackgroundRecurrent respiratory tract infections (RRTIs) are among the most common pediatric health problems and impose substantial healthcare burden. Airway epithelial fragility is increasingly recognized as an important contributor to RRTIs in children, although the upstream molecular mechanisms remain incompletely understood. Endoplasmic reticulum (ER) stress and mitochondrial dysfunction have been implicated in epithelial injury in chronic airway diseases, yet their interplay in pediatric RRTIs is unclear. This study aimed to investigate whether ER stress-mediated disruption of mitochondrial homeostasis drives airway epithelial barrier fragility and enhances inflammatory susceptibility in children with RRTIs.MethodsNasal swab samples from healthy children and RRTI patients were analyzed for ER stress-, barrier-, and mitochondrial-related markers. Tunicamycin-induced ER stress and polyinosinic-polycytidylic acid [Poly(I:C)]-stimulated inflammatory amplification models were established in 16HBE cells, followed by rescue treatment with 4-phenylbutyric acid (4-PBA) or MitoTEMPO.ResultsCompared with healthy controls, RRTI children exhibited significantly increased HSPA5 and DDIT3 mRNA expression and elevated IL-6, IL-8, and CXCL10 protein levels (all P<0.001), whereas TJP1, OCLN, and TFAM expression was markedly reduced (all P<0.001). In 16HBE cells, tunicamycin induced ER stress activation, mitochondrial dysfunction, transepithelial electrical resistance (TEER) reduction, and epithelial permeability elevation. Under ER stress preconditioning, Poly(I:C)-induced inflammatory responses, mitochondrial reactive oxygen species (mtROS) accumulation, mitochondrial membrane potential collapse, and barrier leakage were further aggravated (all P<0.001). Both 4-PBA and MitoTEMPO partially reversed these pathological alterations.ConclusionsER stress-mediated mitochondrial homeostasis disruption promotes airway epithelial barrier fragility and enhances inflammatory susceptibility in pediatric RRTIs. Highlight box Key findings • Nasal mucosa from children with recurrent respiratory tract infections (RRTIs) showed enhanced endoplasmic reticulum (ER) stress (HSPA5, DDIT3), reduced barrier- and mitochondria-related gene expression (TJP1, OCLN, TFAM), and elevated interleukin (IL)-6, IL-8, and C-X-C motif chemokine ligand 10 (CXCL10) levels. • In 16HBE cells, tunicamycin-induced ER stress caused mitochondrial dysfunction [mitochondrial reactive oxygen species (mtROS) accumulation and membrane potential collapse] and epithelial barrier impairment, which were further aggravated by polyinosinic-polycytidylic acid [Poly(I:C)] stimulation. • Both 4-phenylbutyric acid (4-PBA) and MitoTEMPO partially reversed these alterations. What is known and what is new? • ER stress and mitochondrial dysfunction are implicated in chronic airway epithelial injury, but their interplay in pediatric RRTIs remains unclear. • This study identifies an ER stress-mitochondrial homeostasis-barrier fragility axis that heightens inflammatory susceptibility in pediatric RRTIs. What is the implication, and what should change now? • Targeting ER stress or mitochondrial oxidative injury may help reduce recurrent airway inflammatory injury in susceptible children. • Future work should validate this axis in vivo and with clinically relevant viral stimuli. Introduction Recurrent respiratory tract infections (RRTIs) remain one of the most common pediatric health problems worldwide and are associated with substantial healthcare burden, repeated antibiotic exposure, impaired quality of life, and increased risk of chronic airway disorders (1). Although immature immune function has long been considered a major contributor to recurrent infections in children, accumulating evidence suggests that structural vulnerability of the airway mucosal barrier may represent an equally important determinant of infection susceptibility (2,3). The airway epithelium is not merely a passive physical interface; rather, it actively coordinates pathogen sensing, inflammatory signaling, and tissue repair (4). Once epithelial integrity is disrupted, pathogens and inflammatory mediators gain easier access to subepithelial tissues, thereby amplifying local immune activation and perpetuating recurrent infection cycles (3). However, the upstream molecular events responsible for airway epithelial fragility in RRTI children remain incompletely understood. Among the multiple intracellular stress pathways involved in epithelial dysfunction, endoplasmic reticulum (ER) stress has attracted increasing attention in recent years. ER stress is activated when unfolded or misfolded proteins accumulate within the ER lumen, subsequently triggering the unfolded protein response (UPR) to restore cellular homeostasis (5). Moderate activation of this pathway may be adaptive, whereas persistent or excessive ER stress often results in inflammatory amplification, oxidative imbalance, and epithelial injury (6,7). Previous studies have shown that ER stress participates in chronic airway diseases including asthma, chronic obstructive pulmonary disease, and viral airway inflammation (8). Meanwhile, several reports demonstrated that ER stress can directly impair epithelial tight junction organization and increase epithelial permeability (9). Nevertheless, most existing studies have focused primarily on inflammatory regulation, while the relationship between ER stress and epithelial barrier vulnerability in pediatric recurrent respiratory infections has not been sufficiently clarified. Mitochondria are essential regulators of epithelial homeostasis because they provide cellular energy and coordinate redox signaling, apoptosis, and innate immune responses (10). Emerging evidence indicates that ER stress and mitochondrial dysfunction are tightly interconnected through ER-mitochondria contact sites and calcium signaling pathways. Sustained ER stress may induce mitochondrial membrane potential collapse, excessive mitochondrial reactive oxygen species (mtROS) generation, and impaired mitochondrial biogenesis, ultimately leading to epithelial dysfunction (11,12). In airway epithelial cells, mitochondrial instability has been increasingly linked to tight junction disassembly, impaired barrier repair, and exaggerated antiviral inflammation (13). Importantly, viral mimetics such as polyinosinic-polycytidylic acid [Poly(I:C)] can further intensify mitochondrial oxidative stress and inflammatory cascades under pre-existing epithelial stress conditions (14). These findings imply that mitochondrial homeostasis may serve as a critical intermediary linking ER stress to epithelial barrier fragility and heightened infection susceptibility. Yet, direct mechanistic evidence supporting this “ER stress-mitochondrial dysfunction-barrier vulnerability” axis in pediatric RRTI remains scarce. Therefore, this study investigated whether ER stress-mediated mitochondrial homeostasis disruption contributes to airway epithelial barrier fragility and enhanced susceptibility to RRTIs in children. Clinical nasal swab analyses and a tunicamycin-induced 16HBE epithelial barrier model were combined to evaluate the relationship among ER stress, mitochondrial dysfunction, barrier impairment, and inflammatory amplification. In addition, rescue experiments using 4-phenylbutyric acid (4-PBA) and MitoTEMPO were performed to further explore the mechanistic involvement of mitochondrial homeostasis in this process. We present this article in accordance with the MDAR reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0508/rc). Methods Clinical sample collection and grouping (Experiment 1) There were 20 children enrolled in this study, including 10 healthy children and 10 children diagnosed with RRTIs. All participants were recruited from the Pediatric Intensive Care Unit and the Department of Pediatric Internal Medicine IV, Affiliated Hospital of Changchun University of Chinese Medicine. The diagnosis of RRTI was established according to the relevant pediatric clinical criteria. Healthy controls had no acute respiratory infection within the previous 4 weeks and had no history of chronic airway diseases. Children in the RRTI group were preferentially sampled during the stable stage of disease. Nasal epithelial specimens were collected using sterile flocked swabs (Copan Diagnostics, USA) from the inferior turbinate or nasopharyngeal region by gentle rotation for 5–10 s. Two swabs were obtained from each participant. One swab was immediately transferred into RNA preservation solution (Invitrogen, USA) and stored at −80 °C for quantitative real-time polymerase chain reaction (qRT-PCR) analysis, while the second swab was immersed in phosphate-buffered saline (PBS; Gibco, USA). After vortex mixing and centrifugation, the supernatant was collected and preserved at −80 °C for subsequent cytokine detection (15-17). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Before initiation, this study was approved by the Medical Ethics Committee of Affiliated Hospital of Changchun University of Chinese Medicine (No. CCZYFYKYLL2026-246). Written informed consent was obtained from the participants’ parents or legal guardians. Cell culture and ER stress induction (Experiment 2) Human bronchial epithelial 16HBE cells (ATCC, USA; Cat. No. CRL-2741) were cultured in Dulbecco’s modified Eagle medium (DMEM; Gibco) supplemented with 10% fetal bovine serum (FBS; Gibco) and 1% penicillin-streptomycin solution (Beyotime, China). The cells were maintained at 37 °C with 5% CO2. To establish the ER stress model, 16HBE cells were seeded into culture plates and allowed to reach approximately 70–80% confluence before treatment. Tunicamycin (TM; MedChemExpress, USA) was dissolved in dimethyl sulfoxide (DMSO; Sigma-Aldrich, USA) and further diluted in complete medium before use. Cells were divided into five groups: a control group and TM treatment groups at concentrations of 0.25, 0.5, 1.0, and 2.0 µg/mL for 12 h (18,19). Control cells received the same volume of vehicle solution. Rescue intervention experiments (Experiment 3) To further investigate the role of mitochondrial homeostasis in ER stress-mediated epithelial barrier dysfunction, rescue experiments were performed using the ER stress inhibitor 4-PBA and the mitochondrial-targeted antioxidant MitoTEMPO. ER stress was induced by treatment with TM (1.0 µg/mL) for 12 h. Before TM stimulation, 16HBE cells were pretreated with either 4-PBA (5 mM; Sigma-Aldrich) for 2 h or MitoTEMPO (10 µM; MedChemExpress) for 1 h. Control cells received the corresponding vehicle solution for the same duration (9,20,21). Following treatment, cells and culture supernatants were collected for subsequent analyses. Poly(I:C)-induced inflammatory amplification model (Experiment 4) To establish a virus-like inflammatory amplification model under ER stress conditions, 16HBE cells were first exposed to TM (1.0 µg/mL) for 12 h to generate ER preconditioning. Poly(I:C) (InvivoGen, USA) was subsequently transfected into cells using Lipofectamine 3000 reagent (Invitrogen) at a final concentration of 1 µg/mL. Transfection was maintained for 6 h, followed by medium replacement and another 6 h of incubation. For the ER + Poly(I:C) group, cells were first treated with TM (1.0 µg/mL) for 12 h and subsequently exposed to Poly(I:C) stimulation. In rescue groups, cells were pretreated with either 4-PBA or MitoTEMPO before TM administration, followed by Poly(I:C) transfection. Control cells and ER-alone cells received blank transfection complexes without Poly(I:C). The total duration of the Poly(I:C) stimulation protocol was 12 h, including 6 h of transfection and 6 h of post-transfection incubation (14,22,23). Cells were then collected for downstream functional analyses. qRT-PCR Using TRIzol reagent (Invitrogen), total RNA was isolated from both nasal swab specimens and treated 16HBE cells following the manufacturer’s instructions. Before RNA extraction, nasal swab samples preserved in RNA preservation solution were thoroughly vortexed, whereas cell samples from Experiments 2 and 3 were directly lysed after the indicated treatments. RNA purity and concentration were determined with a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, USA), and only qualified RNA samples were subjected to subsequent analysis. For complementary DNA synthesis, a reverse transcription kit (Takara, China) was employed according to the standard protocol. Quantitative PCR amplification was then carried out using SYBR Green qPCR Master Mix (Vazyme, China) on a QuantStudio 5 Real-Time PCR system (Applied Biosystems, USA). In Experiment 1, expression levels of HSPA5, DDIT3, TJP1, OCLN, and TFAM were analyzed in pediatric nasal epithelial samples. Meanwhile, HSPA5 and DDIT3 mRNA expression was evaluated in Experiments 2 and 3 to determine ER stress activation in 16HBE cells. Each sample was analyzed in triplicate, and GAPDH served as the endogenous reference gene. Relative transcription levels were calculated by the 2−ΔΔCt method. Detailed primer sequences are provided in Table 1. Table 1Primer sequences used for quantitative real-time polymerase chain reactionGeneDirectionPrimer sequence (5'-3') HSPA5 ForwardGCTGGCCTAAATGTTATGAGGReverseCAGATCCATGTTGAGCTCTTC DDIT3 ForwardTTATCCTGCAGATGTGCTTTTCCReverseCTTGAGCCGTTCATTCTCTTC TJP1 ForwardGATGAAGATGGTCATACTGTGGReverseTCAGCAGTGTTTCACCTTTCTC OCLN ForwardCTGCTCAGTCTTCTGGATCTReverseATTGGACTTGTCATACCTGTCC TFAM ForwardGCAAGTTGTCCAAAGAAACCTGReverseCGAGTTTCGTCCTCTTTAGC GAPDH ForwardAAGATCATCAGCAATGCCTCCReverseAGGTTTTTCTAGACGGCAGG Primer sequences used for quantitative real-time polymerase chain reaction Enzyme-linked immunosorbent assay (ELISA) Levels of IL-6 (70-EK106/2), IL-8 (70-EK108/2), and CXCL10 (70-EK110/2) in nasal swab eluates (Experiment 1) and cell culture supernatants (Experiment 4) were measured using human ELISA kits (MultiSciences Biotech, China). Frozen samples were thawed on ice and centrifuged before analysis. Standards and samples were added into precoated 96-well plates, followed by incubation with detection antibodies and horseradish peroxidase-conjugated reagents according to the manufacturer’s instructions. After tetramethylbenzidine substrate development and reaction termination, absorbance at 450 nm was measured using a microplate reader (Bio-Rad, USA). Cytokine concentrations were calculated based on the corresponding standard curves. Transepithelial electrical resistance (TEER) measurement Barrier integrity of 16HBE monolayers was evaluated by measuring TEER using an epithelial voltohmmeter (EVOM2; World Precision Instruments, USA). Cells were seeded onto Transwell inserts (0.4 µm pore size; Corning, USA) and cultured until a stable epithelial monolayer was formed. After the indicated treatments in Experiments 2–4, TEER values were recorded at 0 h and 12 h. Before measurement, culture plates were equilibrated to room temperature for approximately 20 min to minimize temperature-associated fluctuations. The resistance value of blank inserts without cells was subtracted from each reading, and the final TEER values were normalized to membrane surface area (Ω·cm2). Absolute TEER was recorded in Ω·cm2. For Figures 2B,3B,4B, TEER data are presented as TEER ratio (%) relative to the corresponding 0 h baseline of each insert. Three independent measurements were obtained for each insert, and the average value was used for subsequent statistical analysis. Tunicamycin dose screening established stable ER stress and impaired barrier function in 16HBE cells. (A) qRT-PCR analysis of HSPA5 and DDIT3 mRNA expression in Ctrl, TM-0.25, TM-0.5, TM-1.0, and TM-2.0 groups. (B) TEER measurements of 16HBE monolayers at 0 h and 12 h after TM treatment. (C) Representative flow cytometric apoptosis profiles (Annexin V-FITC/PI staining) in Ctrl, TM-1.0, and TM-2.0 groups after 12 h treatment. (D) Quantitative analysis of apoptotic cell percentages. Data are presented as mean ± SD. **P<0.01, ***P<0.001 vs. Ctrl. Ctrl, control; ER, endoplasmic reticulum; FITC, fluorescein isothiocyanate; PI, propidium iodide; qRT-PCR, quantitative real-time polymerase chain reaction; SD, standard deviation; TEER, transepithelial electrical resistance; TM, tunicamycin. ER stress disrupted mitochondrial homeostasis and impaired epithelial barrier integrity, which was partially reversed by 4-PBA or MitoTEMPO. (A) qRT-PCR analysis of HSPA5 and DDIT3 mRNA expression in Ctrl, ER, ER + 4-PBA, and ER + MitoTEMPO groups. (B) TEER measurements at 0 h and 12 h after treatment. (C) FITC-dextran permeability analysis of epithelial monolayers. (D) Representative MitoSOX fluorescence images and quantitative analysis of mtROS levels. Scale bar =50 µm. (E) Representative JC-1 staining images and quantitative analysis of JC-1 red/green fluorescence ratio. Scale bar =50 µm. Data are presented as mean ± SD. ***P<0.001 vs. Ctrl group; ##P<0.01, ###P<0.001 vs. ER group. 4-PBA, 4-phenylbutyric acid; Ctrl, control; ER, endoplasmic reticulum; FITC, fluorescein isothiocyanate; mtROS, mitochondrial reactive oxygen species; qRT-PCR, quantitative real-time polymerase chain reaction; SD, standard deviation; TEER, transepithelial electrical resistance. Poly(I:C)-induced inflammatory responses were amplified under ER stress preconditioning, accompanied by aggravated barrier leakage and mitochondrial dysfunction. (A) ELISA analysis of IL-6, IL-8, and CXCL10 levels in culture supernatants. (B) TEER measurements at 0 h and 12 h after treatment. (C) FITC-dextran permeability analysis of epithelial monolayers. (D) Representative MitoSOX fluorescence images and quantitative analysis of mtROS levels. Scale bar =50 µm. (E) Representative JC-1 staining images and quantitative analysis of JC-1 red/green fluorescence ratio. Scale bar =50 µm. Data are presented as mean ± SD. ***P<0.001 vs. Ctrl group; ##P<0.01, ###P<0.001 vs. ER group; &&&P<0.001 vs. ER + Poly(I:C) group. 4-PBA, 4-phenylbutyric acid; Ctrl, control; ELISA, enzyme-linked immunosorbent assay; ER, endoplasmic reticulum; FITC, fluorescein isothiocyanate; mtROS, mitochondrial reactive oxygen species; Poly(I:C), polyinosinic-polycytidylic acid; SD, standard deviation; TEER, transepithelial electrical resistance. Flow cytometric detection of cell apoptosis After treatment, 16HBE cells (from Ctrl, TM-1.0, and TM-2.0 groups) were harvested and washed twice with ice-cold PBS. Cell staining was subsequently performed using an Annexin V-FITC/propidium iodide (PI) Apoptosis Detection Kit (Beyotime; Cat. No. C1062M), in which cells resuspended in binding buffer were incubated with Annexin V-FITC and PI for 15 min at room temperature in the dark according to the manufacturer’s instructions. Fluorescence signals were acquired using a CytoFLEX flow cytometer (Beckman Coulter, USA). The proportions of viable, early apoptotic, and late apoptotic cells were subsequently analyzed with FlowJo software (Tree Star, USA), and the overall apoptosis rate was calculated based on Annexin V-positive populations. MitoSOX fluorescence staining Following the indicated treatments in Experiments 3–4, intracellular mtROS levels were assessed using MitoSOX Red mitochondrial superoxide indicator (Invitrogen; Cat. No. M36008), with cells incubated in serum-free medium containing 5 µM MitoSOX reagent for 10 min at 37 °C in the dark. After being washed three times with warm PBS to remove residual dye, cells were immediately observed under a fluorescence microscope (Olympus, Japan), from which representative fluorescence images were captured under identical exposure settings. For quantitative analysis, fluorescence intensity was measured using ImageJ software (National Institutes of Health, USA), and the average fluorescence signal from at least five random microscopic fields was calculated for each sample. JC-1 staining for mitochondrial membrane potential After treatment in Experiments 3–4, mitochondrial membrane potential was evaluated using a JC-1 mitochondrial membrane potential assay kit (Beyotime; Cat. No. C2006), in which cells were incubated with JC-1 working solution for 20 min at 37 °C under light-protected conditions. Following incubation, cells were washed twice with JC-1 staining buffer and immediately subjected to fluorescence imaging using a fluorescence microscope (Olympus), through which red and green fluorescence signals were acquired under identical acquisition parameters. The ratio of red to green fluorescence intensity, which reflected mitochondrial membrane potential alterations, was quantified using ImageJ software based on at least five randomly selected microscopic fields from each sample. FITC-dextran permeability assay Following treatment in Experiments 3–4, epithelial permeability was evaluated in Transwell-cultured 16HBE monolayers by adding fluorescein isothiocyanate-dextran (FITC-dextran, 4 kDa; Sigma-Aldrich; Cat. No. FD4) to the apical chamber at a final concentration of 1 mg/mL, after which cells were incubated for 2 h at 37 °C. Subsequently, medium from the basolateral chamber was collected, and fluorescence intensity was measured using a multifunctional microplate reader (BioTek, USA) at excitation/emission wavelengths of 485/528 nm. Permeability changes were quantified according to fluorescence intensity values obtained from the lower chamber, with all measurements performed in triplicate under identical experimental settings. Statistical analysis All experiments were independently repeated at least three times, and data are presented as mean ± standard deviation (SD). Statistical analyses were conducted using SPSS 23.0 software (IBM Corp., USA), whereas graph construction was performed with GraphPad Prism 9.0 software (GraphPad, USA). Comparisons between two groups were analyzed using the unpaired Student’s t-test, while differences among multiple groups were evaluated by one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. A value of P<0.05 was considered statistically significant. Results RRTI children exhibited enhanced ER stress, reduced barrier-associated gene expression, and elevated inflammatory responses in nasal swab samples To explore whether airway mucosal stress and barrier abnormalities were present in children with RRTIs, nasal swab specimens collected from Healthy and RRTI groups were subjected to molecular analysis. As presented in Figure 1A, significantly higher mRNA expression levels of HSPA5 and DDIT3 were detected in the RRTI group than in healthy controls (both P<0.001), whereas markedly reduced expression of TJP1, OCLN, and TFAM was simultaneously observed (P<0.001). Along with the increase in ER stress-related transcripts, genes associated with epithelial tight junction integrity and mitochondrial homeostasis showed a clear downward trend. RRTI children exhibited enhanced ER stress, reduced barrier-associated gene expression, and elevated inflammatory cytokine levels in nasal swab samples. (A) qRT-PCR analysis of HSPA5, DDIT3, TJP1, OCLN, and TFAM mRNA expression in Healthy and RRTI groups. (B) ELISA analysis of IL-6, IL-8, and CXCL10 levels in nasal swab eluates. Data are presented as mean ± SD. P<0.001 vs. Healthy. ***, P<0.001 vs. Healthy. ELISA, enzyme-linked immunosorbent assay; ER, endoplasmic reticulum; qRT-PCR, quantitative real-time polymerase chain reaction; RRTI, recurrent respiratory tract infection; SD, standard deviation. Analysis of inflammatory mediators further revealed that, relative to the Healthy group, concentrations of IL-6, IL-8, and CXCL10 in nasal swab eluates from RRTI children were all significantly elevated (all P<0.001; Figure 1B). Increased cytokine production was accompanied by the altered molecular signatures associated with epithelial stress and barrier dysfunction. Taken together, these results indicated that nasal mucosa from RRTI children exhibited enhanced ER stress activation together with impaired barrier-associated molecular profiles and exaggerated local inflammatory responses. Tunicamycin dose screening established stable ER stress and impaired barrier function in 16HBE cells To identify a suitable ER stress induction condition for subsequent experiments, increasing concentrations of tunicamycin were applied to 16HBE cells, followed by evaluation of ER stress activation, epithelial barrier integrity, and apoptosis. As illustrated in Figure 2A, expression levels of HSPA5 and DDIT3 were gradually elevated after TM exposure, and the increase became particularly evident in TM-1.0 and TM-2.0 groups relative to the Ctrl group (both P<0.001). In contrast, only mild upregulation was observed under lower-dose conditions. Barrier function changes were subsequently assessed by TEER analysis. At baseline (0 h), comparable TEER values were detected among all groups, whereas obvious reductions appeared after 12 h treatment (Figure 2B). With increasing TM concentration, TEER values progressively declined, indicating gradual impairment of epithelial barrier integrity. Compared with the Ctrl group, significantly lower TEER values were observed in TM-1.0 and TM-2.0 groups (P<0.05 or P<0.01), among which the TM-2.0 group exhibited the greatest decrease. The influence of TM on cell viability was further examined by flow cytometry. Representative apoptosis profiles are shown in Figure 2C, while quantitative results are summarized in Figure 2D. After 12 h stimulation, the apoptotic rate in TM-1.0-treated cells increased moderately compared with Ctrl cells (P<0.001), whereas a much higher proportion of apoptotic cells was detected in the TM-2.0 group (P<0.001). Notably, although TM-1.0 effectively activated ER stress signaling and reduced TEER values, overall apoptosis remained within a relatively acceptable range. Overall, these results demonstrated that tunicamycin induced ER stress and barrier dysfunction in a concentration-dependent manner in 16HBE cells, and TM at 1.0 µg/mL was selected as the optimal condition for subsequent mechanistic studies. ER stress disrupted mitochondrial homeostasis and impaired epithelial barrier integrity, which was partially reversed by 4-PBA or MitoTEMPO Since TM at 1.0 µg/mL successfully established stable ER stress without excessive cytotoxicity, subsequent rescue experiments were performed to further determine whether mitochondrial dysfunction contributed to ER stress-associated epithelial barrier impairment. After ER stress induction in 16HBE cells, significantly elevated HSPA5 and DDIT3 mRNA levels were detected in the ER group relative to Ctrl cells (both P<0.001), whereas substantial reductions were observed following treatment with either 4-PBA or MitoTEMPO (both P<0.001; Figure 3A), indicating that pharmacological intervention effectively attenuated TM-induced ER stress activation. Although comparable TEER values were present among groups at baseline, marked barrier impairment became evident after 12 h stimulation, during which the ER group exhibited dramatically decreased TEER values compared with Ctrl cells (P<0.001; Figure 3B). This reduction was partially reversed in cells receiving 4-PBA or MitoTEMPO treatment (both P<0.001), which was further supported by FITC-dextran permeability analysis showing significantly increased epithelial permeability under ER stress conditions that was subsequently alleviated by both interventions (P<0.001 or P<0.01; Figure 3C). As demonstrated by MitoSOX staining, robust mtROS accumulation was induced following TM exposure, with fluorescence intensity in the ER group markedly exceeding that in Ctrl cells (P<0.001; Figure 3D), whereas obvious suppression of mtROS signals was detected after 4-PBA or MitoTEMPO administration (both P<0.001). In parallel, JC-1 staining revealed pronounced mitochondrial membrane potential collapse in ER-stressed cells (P<0.001), as reflected by the decreased JC-1 red/green fluorescence ratio, while restoration of mitochondrial membrane potential was observed after rescue treatment (both P<0.001; Figure 3E). Collectively, these findings demonstrated that ER stress-induced epithelial barrier dysfunction was accompanied by mitochondrial homeostasis disruption, which could be partially reversed through inhibition of ER stress or mitochondrial oxidative injury. Poly(I:C)-induced inflammatory responses were amplified under ER stress preconditioning, accompanied by aggravated barrier leakage and mitochondrial dysfunction To further determine whether pre-existing ER stress increased epithelial susceptibility to virus-like stimulation, Poly(I:C) was introduced into TM-preconditioned 16HBE cells, after which inflammatory activation, barrier integrity, and mitochondrial function were comprehensively evaluated. Following Poly(I:C) stimulation in TM-preconditioned 16HBE cells, markedly enhanced inflammatory responses were observed, with IL-6, IL-8, and CXCL10 levels in the ER + Poly(I:C) group significantly exceeding those detected in Poly(I:C)-alone cells (all P<0.001; Figure 4A), whereas only limited cytokine elevation was induced by ER stress alone. Although Poly(I:C) itself increased inflammatory mediator production compared with Ctrl cells (all P<0.001), substantially lower cytokine levels were detected after administration of either 4-PBA or MitoTEMPO than in the ER + Poly(I:C) group (all P<0.001). Barrier integrity analysis further demonstrated that, while baseline TEER values were already moderately decreased in ER-related groups, a much more pronounced reduction developed after 12 h stimulation in ER + Poly(I:C)-treated cells, whose TEER values became significantly lower than those of Poly(I:C)-alone cells (both P<0.001; Figure 4B). Consistently, FITC-dextran permeability was markedly increased under combined ER stress and Poly(I:C) exposure compared with either single treatment condition (both P<0.001; Figure 4C), whereas partial restoration of barrier function was observed following 4-PBA or MitoTEMPO intervention. As revealed by MitoSOX staining, stronger mtROS fluorescence signals were detected in ER + Poly(I:C)-treated cells than in either ER or Poly(I:C) groups alone (all P<0.001; Figure 4D), indicating that mitochondrial oxidative stress was further aggravated under ER stress preconditioning. In parallel, JC-1 staining showed a substantial reduction in the JC-1 red/green fluorescence ratio in the ER + Poly(I:C) group, reflecting more severe mitochondrial membrane potential collapse than that observed after individual stimulation (P<0.01; Figure 4E). Both rescue interventions significantly alleviated mtROS accumulation and partially restored mitochondrial membrane potential compared with the ER + Poly(I:C) group (all P<0.001). Altogether, these findings suggested that pre-existing ER stress markedly sensitized airway epithelial cells to Poly(I:C)-induced inflammatory injury, thereby aggravating mitochondrial dysfunction and epithelial barrier disruption. Discussion The present study demonstrated that RRTIs in children were accompanied by a characteristic mucosal profile involving enhanced ER stress activation, impaired epithelial barrier-associated gene expression, mitochondrial homeostasis disturbance, and exaggerated inflammatory responses. By establishing a tunicamycin-induced 16HBE model, we further showed that persistent ER stress not only disrupted mitochondrial function, as reflected by mtROS accumulation and mitochondrial membrane potential collapse, but also directly impaired epithelial barrier integrity, which was evidenced by reduced TEER values and increased epithelial permeability. More importantly, once epithelial cells entered an ER stress “preconditioned” state, subsequent Poly(I:C) stimulation induced markedly amplified inflammatory responses together with aggravated mitochondrial dysfunction and barrier leakage. Both 4-PBA and MitoTEMPO partially reversed these pathological alterations, supporting the notion that mitochondrial oxidative injury functions as an important mediator linking ER stress to epithelial vulnerability. One particularly interesting observation was that even during relatively stable periods, children with RRTI already exhibited elevated expression of HSPA5 and DDIT3 accompanied by decreased TJP1, OCLN, and TFAM expression within nasal epithelial samples. This finding suggests that epithelial abnormalities in RRTI may not merely represent transient consequences of acute infection, but instead may reflect a chronically vulnerable mucosal state. Although recurrent infection has traditionally been attributed to immature adaptive immunity or repeated pathogen exposure, increasing evidence indicates that epithelial barrier dysfunction itself substantially influences respiratory susceptibility (24). Airway epithelial cells normally form a tightly regulated interface that limits pathogen invasion while coordinating innate immune signaling; however, once tight junction organization becomes destabilized, inflammatory mediators and pathogens can penetrate the epithelial layer more easily, thereby amplifying local immune activation (4,25). Similar reductions in tight junction-associated proteins have been reported in viral airway injury and chronic inflammatory airway diseases, where epithelial leakage often correlates with disease severity (26). In our study, reduced TFAM expression was simultaneously observed in RRTI children, which may imply that mitochondrial instability already exists before acute inflammatory exacerbation occurs. Since TFAM is critically involved in mitochondrial DNA maintenance and oxidative phosphorylation stability, its downregulation may indicate an underlying metabolic insufficiency that predisposes epithelial cells to stress intolerance. Nevertheless, we acknowledge that TFAM primarily reflects mitochondrial biogenesis rather than direct functional performance. In the current study, we provided direct functional readouts including mtROS levels (MitoSOX staining) and mitochondrial membrane potential (JC-1 staining), which are well-accepted indicators of mitochondrial integrity. Future investigations should incorporate comprehensive bioenergetic assays such as ATP quantification, oxygen consumption rate, and respiratory chain complex activities to fully characterize mitochondrial functional deterioration in this context. Another major finding of this study is that mitochondrial dysfunction appeared to occupy a central position within ER stress-mediated epithelial injury rather than acting as a secondary epiphenomenon. Although ER stress has long been associated with inflammatory signaling and apoptosis, the pathological importance of ER-mitochondria coupling in airway epithelial fragility has remained insufficiently explored. Mechanistically, sustained ER stress may enhance calcium flux from the ER to mitochondria through mitochondria-associated membranes, while excessive reactive oxygen species (ROS) propagation between these organelles further destabilizes cellular bioenergetics (11). In the current study, TM-induced ER stress caused substantial mtROS accumulation together with collapse of mitochondrial membrane potential, changes that occurred simultaneously with TEER reduction and increased FITC-dextran permeability. Importantly, rescue effects were observed not only after ER stress inhibition by 4-PBA but also following selective suppression of mitochondrial oxidative stress using MitoTEMPO. Such findings suggest that mitochondrial oxidative injury actively participates in epithelial barrier destabilization. Similar observations have been described in intestinal epithelial and alveolar barrier injury models, in which mtROS overproduction promoted cytoskeletal remodeling and tight junction disassembly (27). Nevertheless, compared with previous reports that mainly focused on inflammatory regulation, our study further extends this mechanism into the context of pediatric recurrent respiratory infections and emphasizes the possibility that mitochondrial homeostasis disruption represents a core pathological node underlying epithelial susceptibility. In interpreting our findings, it is important to distinguish between direct inflammatory signals emanating from ER stress and secondary inflammation propagated by mitochondrial damage-associated molecular patterns (DAMPs). In our model, ER stress induction alone (TM treatment) caused only moderate elevation of IL-6, IL-8, and CXCL10, suggesting that direct UPR-mediated NF-κB activation contributes to a relatively mild baseline inflammatory state. However, when ER stress-preconditioned cells were exposed to Poly(I:C), cytokine production surged to levels far exceeding those induced by either stimulus alone. Notably, this amplification was markedly blunted by MitoTEMPO, a mitochondrial-targeted antioxidant, indicating that mitochondrial oxidative stress acts predominantly as an inflammatory amplifier rather than the primary initiator. This distinction carries therapeutic implications: while alleviating ER stress (e.g., with 4-PBA) may address the upstream trigger, directly quenching mitochondrial oxidative stress (e.g., with MitoTEMPO) could be particularly effective in breaking the amplification loop even when the primary ER stress persists. Future mechanistic studies employing UPR-specific inhibitors or mitochondrial DAMP neutralization strategies will be required to quantitatively dissect the relative contributions of these two inflammatory layers. Perhaps the most clinically meaningful finding was that Poly(I:C)-induced inflammatory injury became markedly amplified under ER stress preconditioning. Poly(I:C) alone triggered moderate increases in IL-6, IL-8, and CXCL10 production, whereas dramatically higher cytokine levels were detected once stimulation occurred in ER-stressed epithelial cells. At the same time, barrier leakage and mitochondrial dysfunction became substantially aggravated. These results imply that ER stress may establish a metabolically “primed” epithelial state in which subsequent viral-like stimulation triggers disproportionate inflammatory activation. A comparable phenomenon has been reported in chronic inflammatory airway diseases, where stressed epithelial cells respond excessively to secondary environmental or viral insults (14). However, unlike studies primarily emphasizing NF-κB activation or interferon signaling, our results suggest that mitochondrial instability itself may function as an inflammatory amplifier during this process. Because mitochondria are deeply involved in antiviral signaling, ROS generation, and innate immune regulation, excessive mitochondrial stress may convert adaptive antiviral responses into persistent inflammatory injury (28). From this perspective, airway epithelium in RRTI children may exist in a chronic low-threshold stress state, thereby rendering otherwise common viral exposure capable of repeatedly inducing exaggerated mucosal damage and inflammatory amplification. Several limitations should nevertheless be acknowledged. First, nasal swab samples primarily reflect upper airway mucosal status, whereas RRTIs frequently involve the lower respiratory tract; future studies incorporating bronchial brushings or sputum samples will be necessary to validate our findings in the lower airway. Second, while our in vitro rescue experiments strongly support the proposed axis, the current study lacks in vivo validation using animal models of respiratory infection, which should be addressed in future investigations. Third, the clinical cohort size was relatively limited, and the precise molecular pathways linking ER stress to mitochondrial instability—such as calcium signaling, mitochondrial dynamics, or ER-mitochondria contact structures—were not fully dissected. Fourth, tunicamycin and Poly(I:C) are pharmacological and virus-like tools rather than physiological triggers; future studies employing clinically relevant stimuli (e.g., respiratory syncytial virus, allergens, or pollutants) and comprehensive bioenergetic assays (e.g., ATP levels, respiratory chain complex activities) will further corroborate the clinical relevance of this mechanism. Despite these limitations, the present study provides evidence that ER stress-mediated mitochondrial homeostasis disruption contributes to airway epithelial fragility and inflammatory susceptibility in children with RRTIs, and raises the possibility that interventions targeting mitochondrial oxidative injury may represent a promising strategy for reducing recurrent airway inflammation in susceptible pediatric populations. Conclusions The present study demonstrated that persistent ER stress, through disrupting mitochondrial homeostasis and promoting mitochondrial oxidative injury, weakened airway epithelial barrier integrity and ultimately enhanced inflammatory susceptibility in children with RRTIs. Under ER stress conditions, increased mtROS accumulation together with mitochondrial membrane potential collapse contributed to epithelial barrier leakage, which was accompanied by exaggerated inflammatory responses after Poly(I:C) stimulation. Moreover, once airway epithelial cells entered an ER stress-preconditioned state, virus-like stimulation induced more severe inflammatory amplification and barrier dysfunction, thereby suggesting the existence of a chronically vulnerable mucosal background in RRTI children. Importantly, both 4-PBA and MitoTEMPO partially restored mitochondrial function and epithelial barrier stability, indicating that modulation of ER stress or mitochondrial oxidative stress may exert protective effects against recurrent airway inflammatory injury. Supplementary The article’s supplementary files as Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0508/prf