For in-depth analysis, the GSE37013 dataset was accessed from the Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo/) repository, originally reported by Kip et al (16). This dataset comprises 28 samples, including seven non-ischemic surgical control jejunal samples collected intraoperatively and 21 I/R case samples. Differential expression analysis on the GSE37013 dataset was conducted using the limma package (version 3.48.3; http://bioconductor.org/packages/limma) implemented in R (version 4.1.2; R Foundation for Statistical Computing; http://www.r-project.org/). Genes exhibiting a fold change (FC) of >1.3 were categorized as upregulated, whereas those with an FC of <0.77 were deemed downregulated; P<0.05 was employed to denote a statistically significant difference. To visualize the DEGs, the ggplot2 package (https://cran.r-project.org/package=ggplot2) was utilized in R.

To elucidate the interactions among DEGs, a PPI network analysis was conducted using the Search Tool for the Retrieval of Interacting Genes (https://string-db.org/). The top 15 genes were identified using the Cytohubba plugin within Cytoscape (version 3.9.1; http://cytoscape.org), employing algorithms such as the maximum neighborhood component (MNC), maximal clique centrality (MCC) and degree. Subsequently, an intersection analysis was performed on the top 15 genes derived from these three network modules using the Bioinformatics and Evolutionary Genomics online tool (http://bioinformatics.psb.ugent.be/webtools/Venn/). The gene expression levels of the intersection genes in both the control and case groups of the GSE37013 dataset were then analyzed using the SangerBox platform (http://sangerbox.com/). P<0.05 was deemed statistically significant for the scores obtained in this analysis.

Caco-2 cells, from a human colorectal adenocarcinoma, were sourced from National Collection of Authenticated Cell Cultures. These cells were cultured in DMEM (Procell Life Science & Technology Co., Ltd.), supplemented with 10% fetal bovine serum (Procell Life Science & Technology Co., Ltd.) and 1% penicillin-streptomycin solution (Procell Life Science & Technology Co., Ltd.). Cell culture was performed at 37°C within a humidified incubator maintained at 5% CO₂.

An OGD/R model was established to simulate I/R injury. The cells were subjected to OGD/R by being incubated in glucose-free DMEM at 37°C within a microaerobic environment (comprising 1% O₂, 94% N₂, 5% CO₂) for varying durations of 0, 2, 4 and 6 h. Subsequently, the cells underwent reoxygenation in complete DMEM under normoxia at 37°C (21% O₂, 5% CO₂, 74% N₂) for 24 h. Control cells were maintained in complete DMEM (with glucose and 10% FBS) at 37°C under normoxia (21% O₂, 5% CO₂, balance N₂) and underwent the same medium-change schedule as the OGD/R groups. Additionally, to induce inflammation, Caco-2 cells (which had been treated with OGD for 4 h followed by reoxygenation for 24 h) were exposed to 1 µg/ml lipopolysaccharide (LPS; Beijing Solarbio Science & Technology Co., Ltd.) for 6 h at 37°C. Finally, adenosine triphosphate (ATP; Beyotime Institute of Biotechnology) was introduced at a concentration of 5 mM and the cells were incubated for 30 min at 37°C.

Transfection was performed after OGD/R. After seeding at a density of 2×10⁵ cells/well in 6-well plates, Caco-2 cells were cultured until they reached 60–70% confluence. To achieve knockdown of CTSB, the cells were transfected with specific small interfering RNA (siRNA; final concentration, 50 nM) designed to target CTSB, adhering strictly to the manufacturer's protocol. The transfection was carried out using 10 µl Lipofectamine™ 2000 reagent (Invitrogen; Thermo Fisher Scientific, Inc.) at 37°C for 6 h. As a negative control (NC), a non-targeting siRNA was employed. Following transfection, the cells were incubated for an additional 48 h before being harvested for use in further experiments. The siRNA sequences utilized for transfection were as follows: si-CTSB, sense 5′-CCUGUCGGAUGAGCUGGUCAACU-3′, antisense 5′-AUAGUUGACCAGCUCAUCGACAG-3′; si-NC, sense 5′-CCUAGCGAGUGGUCAACUGUAU-3′, antisense 5′-AUACAGAGUUGACCACUCGCCUAG-3′.

The Cell Counting Kit-8 (CCK-8) assay (Beyotime Institute of Biotechnology) was utilized to evaluate cell viability, according to the manufacturer's instructions. Initially, 96-well plates were seeded with 5×10³ Caco-2 cells/well in 100 µl complete growth medium. The plates were then incubated overnight at 37°C in a humidified incubator maintained at 5% CO₂, allowing the cells to adhere and achieve a stable growth state. Following this incubation period, each well was supplemented with 10 µl CCK-8 solution, and the plates were further incubated for 2 h at 37°C. Subsequently, cell viability was assessed by measuring the absorbance at 450 nm using a microplate reader (Bio-Rad Laboratories, Inc.). To ensure the robustness and reliability of the findings, data from a minimum of three independent experiments were subjected to statistical analysis.

Caco-2 cells were exposed to OGD for 2, 4 and 6 h, followed by 24 h of reoxygenation to evaluate the resultant impact on LDH levels. According to the manufacturer's guidelines, culture supernatants were carefully collected, and LDH levels were measured using an LDH cytotoxicity assay kit (Beyotime Institute of Biotechnology). To ensure the reliability of the data, the assay was conducted in triplicate for each time point. LDH activity was quantified by recording the absorbance at 490 nm using a microplate reader. Data obtained from a minimum of three independent experiments were subjected to rigorous statistical analysis to discern the significance of the differences in LDH levels across the various OGD/R treatment durations.

Total RNA was extracted from Caco-2 cells using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's detailed instructions, after the aforementioned OGD/R and/or LPS/ATP treatments. Subsequently, 1 µg isolated total RNA was reverse-transcribed into cDNA utilizing the PrimeScript RT reagent kit (Takara Biotechnology, Ltd.) according to the manufacturer's protocol. qPCR was performed on a Bio-Rad CFX96 Real-Time PCR Detection System (Bio-Rad Laboratories, Inc.), employing the SYBR Green kit (Takara Biotechnology, Ltd.). The expression levels of the target genes, as well as the internal control GAPDH, were quantified using gene-specific primers (Table I). The thermocycling conditions were as follows: 95°C for 30 sec, followed by 40 cycles at 95°C for 5 sec and 60°C for 30 sec; a melt-curve analysis was then performed from 65–95°C with 0.5°C increments. The relative expression of each candidate gene was determined using the 2^−ΔΔCq method (17). To ensure the robustness of the findings, data from a minimum of three independent experiments were subjected to statistical analysis.

To lyse Caco-2 cells, RIPA lysis buffer (Beyotime Institute of Biotechnology) was prepared with the addition of protease and phosphatase inhibitors. The total protein content in the lysates was then quantified using a BCA protein assay kit (Beyotime Institute of Biotechnology). Equal amounts of protein (20 µg) were separated by SDS-PAGE on 10% polyacrylamide gels and were subsequently transferred onto PVDF membranes (Beyotime Institute of Biotechnology). To prevent non-specific binding, the membranes were blocked for 1 h at room temperature with 5% skim milk dissolved in Tris-buffered saline containing 0.1% Tween-20 (TBST). To detect the target proteins, the antibodies for each target protein and the internal control GAPDH antibody were mixed separately, and the PVDF membranes were incubated with the primary antibody solutions at 4°C overnight. The following primary antibodies were used: CTSB (cat. no. 12216-1-AP; 1:1,000; Wuhan Sanying Biotechnology), active CTSB (cat. no. 12216-1-AP; 1:1,000; Wuhan Sanying Biotechnology), ZO-1 (cat. no. ab276131; 1:1,000; Abcam), claudin-1 (cat. no. ab211737; 1:2,000; Abcam), NLRP3 (cat. no. 30109-1-AP; 1:2,000; Wuhan Sanying Biotechnology), gasdermin D (GSDMD; cat. no. ab210070; 1:1,000; Abcam), cleaved N-terminal GSDMD (GSDMD-N; cat. no. ab215203; 1:2,000; Abcam), ASC (cat. no. 10500-1-AP; 1:5,000; Wuhan Sanying Biotechnology), cleaved caspase-1 (p20; cat no. AF4005; 1:2,000; Affinity Biosciences), caspase-1 (cat. no. ab207802; 1:1,000; Abcam), IL-1β (cat. no. ab283818; 1:1,000; Abcam), IL-18 (cat. no. 82696-15-RR; 1:2,000; Wuhan Sanying Biotechnology) and GAPDH (cat. no. 80570-1-RR; 1:2,500; Wuhan Sanying Biotechnology). After washing the membranes three times with TBST, they were incubated for 1 h at room temperature with HRP-conjugated goat anti-rabbit IgG (cat. no. A0208; 1:1,000; Beyotime Institute of Biotechnology). Protein bands were detected using an enhanced chemiluminescence detection system (Bio-Rad Laboratories, Inc.). Band intensities were semi-quantified using ImageJ software (version 2.0.0; National Institutes of Health). To ensure accurate normalization, the bands of both the proteins of interest and GAPDH were developed from the same membrane, subjected to the same blocking, incubation and washing steps, and visualized under identical conditions.

To investigate the interaction between CTSB and NLRP3, co-IP was employed. Cells were harvested and lysed in RIPA buffer (Beyotime Institute of Biotechnology) supplemented with a protease-inhibitor cocktail. Lysates were clarified at 12,000 × g for 10 min at 4°C, and protein concentrations were determined by BCA assay. The resulting cell lysates were then mixed with 30 µl protein A/G agarose bead slurry (Santa Cruz Biotechnology, Inc.) per 1 mg total protein for 1 h at 4°C on a rotator to remove non-specific binding. Following centrifugation (2,500 × g, 3 min, 4°C), the agarose beads were discarded, and the supernatant, containing the pre-cleared lysates, was collected. Primary antibodies specific for either CTSB (cat. no. ab30443; Abcam) or NLRP3 (cat. no. 30109-1-AP; Wuhan Sanying Biotechnology), or an equal amount of species-matched normal IgG (negative control), were added at 2 µg per IP and mixtures were incubated for 6 h at 4°C on a rotator to allow antigen-antibody binding. Subsequently, an additional 30 µl protein A/G agarose beads was added, and the suspensions were rotated overnight at 4°C to capture the antigen-antibody complexes. Beads were collected by centrifugation (2,500 × g, 3 min, 4°C) and washed four times with ice-cold RIPA lysis buffer (1 ml/wash) to remove any unbound proteins. The antigen-antibody complexes were eluted from the beads using 2X Laemmli SDS sample buffer at 95°C for 5 min. The eluate was boiled and denatured to prepare it for subsequent WB analysis.

ELISA was carried out to quantify CTSB, TNF-α, IL-6 and IL-1β using the following commercially available kits: Human CTSB ELISA Kit (cat. no. ab119584; Abcam), Human TNF-αELISA Kit (cat. no. BMS2034; Invitrogen; Thermo Fisher Scientific, Inc.), Human IL-6 ELISA Kit (cat. no. EH2IL6; Invitrogen; Thermo Fisher Scientific, Inc.), and Human IL-1β Instant ELISA Kit (cat. no. BMS224INST; Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. For each assay, 100 µl Caco-2 cell culture supernatant/well was carefully transferred to pre-coated 96-well plates, and the plates were incubated according to the manufacturer's guidelines. The enzymatic reaction was developed using the tetramethylbenzidine substrate, and the absorbance at 450 nm was measured using a microplate reader. Protein concentrations were calculated based on standard curves generated with kit-supplied standards. To ensure the robustness and reliability of the results, each experiment was replicated at least three times. Data from these independent experiments were subjected to statistical analysis to evaluate the significance of differences in protein levels between experimental groups.

After OGD for 4 h followed by 24 h reoxygenation, cells were fixed for 15 min at room temperature in 4% paraformaldehyde solution. Subsequently, the cells were permeabilized for 10 min with 0.1% Triton X-100 to facilitate antibody penetration and then blocked for 1 h with 5% bovine serum albumin (BSA; Beyotime Institute of Biotechnology) dissolved in phosphate-buffered saline (PBS) to mitigate non-specific binding. Primary antibodies targeting ZO-1 (cat. no. ab276131; 1:500) and claudin-1 (cat. no. ab211737; 1:1,000) (both from Abcam) were diluted in 1% BSA and incubated with the cells overnight at 4°C to allow for specific antigen-antibody interactions. The following day, the cells were incubated with Alexa Fluor^® 488-labeled goat anti-rabbit IgG secondary antibodies (cat. no. A-11008; 1:1,000; Invitrogen; Thermo Fisher Scientific, Inc.) for 1 h at room temperature in the dark to preserve the fluorescence signal. The nuclei were counterstained DAPI (Beyotime Institute of Biotechnology) for 5 min at room temperature after washing with PBS. Fluorescence microscopy images were captured using a fluorescence microscope (Nikon Eclipse Ti-U; Nikon Corporation). To ensure the reproducibility and reliability of our findings, the entire immunofluorescence staining protocol was replicated at least three times, with three technical replicates per condition in each experiment.

After 4 h of OGD followed by 24 h of reoxygenation, cells were lysed on ice using the kit-supplied sample lysis buffer (Beyotime Institute of Biotechnology). The lysates were subsequently analyzed using commercially available detection kits for reactive oxygen species (ROS; cat. no. S0033S), malondialdehyde (MDA; cat. no. S0131S), superoxide dismutase (SOD; cat. no. S101S) and glutathione peroxidase (GSH-Px; cat. no. S0059S) (all from Beyotime Institute of Biotechnology). These assays were conducted according to the manufacturer's instructions. Absorbance values were recorded using a microplate reader at the following specific wavelengths: 488 nm (excitation) and 525 nm (emission) for ROS, 412 nm for GSH-Px, 450 nm for SOD and 532 nm for MDA. To ensure the reliability and reproducibility of the data, each test was performed in triplicate.

All results are presented as the mean ± standard deviation. To determine statistical significance, one-way analysis of variance was utilized, followed by Tukey's post hoc test for multiple comparisons between groups. For pairwise comparisons between two specific groups, a two-tailed unpaired Student's t-test was employed. P<0.05 was considered to indicate a statistically significant difference. All statistical analyses were conducted using the R programming language. Additionally, graphs were generated exclusively for visualization purposes using GraphPad Prism software (version 8.0; Dotmatics) to facilitate interpretation of the results.

First, a differential expression analysis was conducted using the GSE37013 dataset, comparing I/R jejunal samples (n=21) with non-ischemic surgical controls (n=7), which yielded 181 downregulated and 224 upregulated DEGs (Fig. 1A). Subsequently, a PPI network analysis was performed on the DEGs. This analysis revealed the top 15 genes as identified by three different algorithms: Degree (15 nodes and 44 edges; Fig. 1B), MCC (15 nodes and 50 edges; Fig. 1C) and MNC (15 nodes and 52 edges; Fig. 1D). Following this, a cross-analysis of the top 15 genes from each of the three algorithms was conducted, resulting in the identification of 13 key intersection genes (Fig. 1E). Expression analysis was then performed to examine the expression levels of these genes in both the control and case samples from the GSE37013 dataset (Fig. 1F). The results demonstrated that in the case samples, BIRC3, CTSB, EZH2, GSK3B, LAMP1, PTPRC and TLR3 were significantly upregulated, whereas EGFR, NCAM1, PLCG1, SQSTM1, STAT5B and ZAP70 were significantly downregulated.

The current study assessed the effects of OGD/R on the expression of tight junction proteins and CTSB, LDH release and cell survival in Caco-2 cells. Compared with in the control group, as the duration of OGD/R treatment was extended, the results from the CCK-8 assay revealed a progressive decline in cell viability (Fig. 2A). Concurrently, a significant increase in LDH levels was observed with prolonged OGD/R exposure (Fig. 2B). Furthermore, following OGD/R treatment, there was a gradual decrease in the mRNA expression levels of ZO-1 and CLDN1, with their levels declining progressively as the duration of OGD/R exposure increased (Fig. 2C and D). Based on preliminary experiments (Fig. 2A and B), which demonstrated significant but sub-lethal injury at the 4-h OGD mark, consistent with established in vitro I/R models (18), subsequent experiments used the same 4-h OGD treatment to further examine its effect on CTSB activity. Subsequently, an ELISA was employed to examine CTSB activity in Caco-2 cells. In the OGD/R model, CTSB activity was markedly elevated compared with that in the control group (Fig. 2E). WB analysis further confirmed that both total CTSB protein and active CTSB protein levels were increased following OGD/R treatment (Fig. 2F and G).

RT-qPCR and WB analysis confirmed the efficient knockdown of CTSB, resulting in a significant reduction in CTSB expression following transfection (Figs. S1 and 3A-C). To evaluate the impact of CTSB knockdown on the viability of Caco-2 cells exposed to OGD/R, the CCK-8 assay was employed. OGD/R resulted in a marked reduction in cell viability compared with that in the control group, whereas CTSB knockdown effectively reversed this reduction, thereby preserving cell viability (Fig. 3D). Furthermore, LDH levels, which serve as an indicator of cell damage, were markedly elevated in the OGD/R-treated cells; however, CTSB knockdown notably reduced LDH release, suggesting a decrease in cellular injury (Fig. 3E). WB analysis results further revealed that the protein levels of ZO-1 and claudin-1, both of which were reduced in the OGD/R model, were restored following CTSB knockdown (Fig. 3F and G). Immunofluorescence staining provided additional confirmation, demonstrating the recovery of ZO-1 and claudin-1 expression after CTSB knockdown (Fig. 3H and I). Collectively, these findings suggested that CTSB knockdown may mitigate I/R-induced cellular injury and preserve epithelial barrier integrity in Caco-2 cells.

To elucidate the role of CTSB in oxidative stress and inflammatory responses within I/R-injured Caco-2 cells, a series of experiments were conducted. The levels of ROS in the OGD/R model, as assessed using an assay kit, were significantly higher than those in the control group, indicating an increase in oxidative stress (Fig. 4A). Notably, CTSB knockdown attenuated this elevation in ROS levels, suggesting that CTSB serves an important role in regulating oxidative stress. Furthermore, MDA levels, evaluated using an MDA detection kit, were markedly elevated in the OGD/R model group compared with those in the control group, whereas CTSB knockdown mitigated this increase (Fig. 4B), indicating a reduction in lipid peroxidation. SOD activity, which was markedly decreased in the OGD/R model group, was partially restored following CTSB knockdown (Fig. 4C), suggesting an enhancement in antioxidative capacity. Similarly, GSH-Px activity, which was significantly decreased in the OGD/R model group, was reversed by CTSB knockdown (Fig. 4D), further supporting its role in oxidative stress regulation. The results of ELISA revealed that the OGD/R model group exhibited higher levels of pro-inflammatory cytokines, including IL-6, IL-1β and TNF-α; by contrast, CTSB knockdown significantly reduced the levels of these inflammatory markers (Fig. 4E-G). Collectively, these findings suggested that CTSB knockdown may alleviate oxidative stress and inflammatory responses in Caco-2 cells following I/R injury, thereby highlighting the potential therapeutic benefits of targeting CTSB in this context.

To investigate the role of CTSB in the inflammatory responses induced by OGD/R in Caco-2 cells, the expression of key inflammasome components and pro-inflammatory cytokines was examined. WB analysis revealed that OGD/R treatment significantly elevated the protein expression levels of GSDMD-N, NLRP3, ASC, cleaved caspase-1, IL-1β and IL-18 compared with those in the control group (Fig. 5A-D). However, CTSB knockdown effectively reduced the protein levels of these inflammatory mediators. Given the pivotal role of CTSB in the inflammatory process, the current study aimed to explore whether CTSB directly interacts with NLRP3, a key component of the inflammasome. Notably, a previous study has suggested an interaction exists between CTSB and NLRP3 (11). To further investigate this, co-IP experiments were conducted. The results confirmed that CTSB interacts with NLRP3 in the intestinal I/R model (Fig. 5E and F). When CTSB was immunoprecipitated (Fig. 5E), NLRP3 was detected only in the OGD/R + si-NC group. This is consistent with the strong induction of NLRP3 by OGD/R and its marked reduction after CTSB silencing, rendering the co-precipitated NLRP3 in the control and OGD/R + si-CTSB groups below the detection limit. Conversely, when NLRP3 was immunoprecipitated (Fig. 5F), CTSB was recovered in all groups because CTSB is constitutively expressed, with the signal highest in the OGD/R + si-NC groups and decreased after CTSB knockdown. These reciprocal co-IP findings support a specific CTSB-NLRP3 interaction in this model. Furthermore, OGD/R treatment significantly elevated the mRNA expression levels of pro-inflammatory cytokines (Fig. 5G-5I), whereas CTSB knockdown markedly decreased the expression levels of these cytokines. Notably, when cells were further treated with LPS and ATP following CTSB knockdown, cytokine expression was partially restored, suggesting a CTSB-independent pathway of inflammatory activation.

Compared with those in the control group, WB analysis demonstrated that OGD/R treatment significantly reduced the expression levels of tight junction proteins, specifically ZO-1 and claudin-1 (Fig. 6A and B). However, CTSB knockdown significantly restored these protein levels. Notably, further treatment with LPS and ATP following CTSB knockdown partially reversed this recovery, suggesting a regulatory role of CTSB in maintaining tight junction integrity under inflammatory conditions. This suggests that LPS and ATP may counteract the benefits of CTSB knockdown. The results of the cell viability experiment indicated a substantial decrease in cell viability following OGD/R treatment compared with that in the control group (Fig. 6C). By contrast, CTSB knockdown significantly enhanced cell viability under OGD/R conditions, indicating its protective effect; however, the subsequent addition of LPS and ATP attenuated this protective effect, suggesting that LPS and ATP may counteract the benefits of CTSB knockdown. Subsequently, it was revealed that MDA levels in OGD/R-treated cells were significantly elevated, indicating an increase in oxidative stress (Fig. 6D). By contrast, CTSB knockdown significantly reduced MDA levels, demonstrating its antioxidant potential; however, treatment with LPS and ATP partially reversed this decrease, suggesting that LPS and ATP may exacerbate oxidative stress despite CTSB knockdown. Furthermore, OGD/R-treated cells exhibited a substantial reduction in SOD activity, which was restored by CTSB knockdown (Fig. 6E). Notably, the protective impact of CTSB knockdown on SOD activity was, however, diminished by subsequent LPS and ATP treatment, further emphasizing the complex interplay between CTSB, inflammation and oxidative stress.