ERAP1^+/tm1a mice were generated by the Korea Mouse Phenotyping Center (KMPC) using a clone of mouse embryonic stem cells carrying a mutant ERAP1 allele (ERAP1^{tm1a (EUCOMM)Wtsi}), as previously described (34). To obtain ERAP1⁺/⁻ mice, ERAP1^+/tm1a were crossed with a Cre recombinase-expressing strain, and the offspring were maintained on a C57BL/6 background. For the generation of ERAP1 knockout mice, a targeting vector was constructed with two loxP sites flanking exons 6 and 7 of the ERAP1 gene, enabling the specific deletion of these exons upon Cre recombinase expression.

Genotyping was performed using the tail tissue collected approximately two weeks after birth. Genomic DNA was extracted from tail tissue using the DNeasy Blood & Tissue kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The following primer sets were used for genotyping: ERAP1-F 5′- GTG GTC ATT AGC AGG AGG CA - 3′; neoF, 5′- GGG ATC TCA TGC TGG AGT TCT TCG - 3′, and ERAP1-ttR 5′-CTC TCT GTA TGT GGT CAG TCC C - 3′. The amplicon (bp) of the PCR products for the mutant alleles and wild-type (WT) were 959 and 808, respectively. PCR protocol included the following steps: 95 °C for 5 minutes (min); 40 cycles of 95 °C for 1 min, 60 °C for 1 min, and 72 °C for 1 min, this was finally followed by 72 °C for 7 min.

Germ-free ERAP1 WT and ERAP1⁺/⁻ mice (8 weeks old, 20 – 22 g) were used and housed in a specific pathogen-free (SPF) facility. All mice were on the C57BL/6 background. Environmental conditions were controlled: a 12-hour (h) light/dark cycle, a stable temperature of 22 °C, and a relative humidity of 22 – 30%. Mice had access to autoclaved water and food at all times. Mouse cages were changed weekly. Approval for animal work was carefully assessed following the guidelines set by the Ajou University Animal Care and Use Committee (IACUC - 2024 - 0003). Mice were continuously observed and monitored throughout the entire experiment.

Eight-week-old ERAP1 WT and ERAP1⁺/⁻ mice were assigned randomly to three groups each: WT, WT + DSS, WT + DSS + SSZ; and ERAP1⁺/⁻, ERAP1⁺/⁻ + DSS, ERAP1⁺/⁻ + DSS + SSZ. Each group consists of 4 to 6 mice. Group 1 served as the control and received only regular water for 28 days. Group 2 underwent a cyclic DSS treatment to induce colitis. Mice were administered 2.5% DSS (molecular weight: 36 kDa-50 kDa; MP Biomedicals, LLC, Canada) in their drinking water for 7 days, followed by 7 days of recovery with regular water. This cycle was repeated once more (7 days of 2.5% DSS followed by 7 days of regular water), resulting in a 28-day experimental period. This alternating DSS–recovery regimen was chosen based on previous studies to induce moderate, non-lethal colitis in C57BL/6J mice (35). It more closely replicates the episodic flare-recovery pattern of human ulcerative colitis (36) and provides a relevant model for studying chronic intestinal inflammation. Group 3 followed the same cyclic DSS protocol as Group 2; however, during the final 7-day recovery phase, sulfasalazine (SSZ, 40 mg/kg, Sigma, USA) was added to the drinking water to assess its therapeutic efficacy. On day 29, mice were euthanized via CO₂ inhalation in accordance with IACUC guidelines. CO₂ was introduced at a rate of 10%–30% of the chamber volume per minute until loss of consciousness, after which death was confirmed by the cessation of respiration and heartbeat. Following anesthesia, euthanasia was completed by cervical dislocation. Colon, spleen, and peripheral blood samples were then collected for further analysis.

Peripheral blood leukocytes (PBLs), spleen, and intestinal epithelial cells (IELs) were isolated from each mouse after euthanasia. PBLs were collected by cardiac puncture. The whole spleen was carefully excised and gently mashed using a 70 µm filter to prepare a single-cell suspension. PBLs and spleen tissues were treated with ammonium-chloride-potassium buffer to lyse the red blood cells. IELs were collected using a procedure described previously (37). In brief, after euthanasia, the colon was carefully removed using scissors. Fat, connective tissue, and visible Peyer’s patches were surgically removed before further processing to ensure the purity of the IEL population. The intestines were then thoroughly rinsed three times using ice-cold phosphate-buffered saline (PBS) to eliminate gut contents. After longitudinal incision along the intestine, they were further sliced into 0.5 cm pieces and transferred into 50 mL tubes containing RPMI media enriched with 2% FBS, 5% DTT, and 0.5M EDTA. The samples were incubated at 37 °C for 1h and shaken vigorously every 15 seconds (s) to detach the epithelial cells. After repeating this process once, the cell suspension was filtered through a 70 µm filter to remove debris. Finally, the epithelial cells were collected by centrifugation at 3000 rpm for 5 min at 4 °C. The cell suspensions obtained from PBLs and IELs were washed and resuspended in appropriate buffers and used for subsequent analyses such as flow cytometry.

The frequencies of neutrophils, DCs, and lymphocytes in PBLs and IELs were analyzed using flow cytometry as previously described (34). Cell suspensions from all collected tissues were stained with fluorescently labeled antibodies targeting specific cell surface markers. The concentration of immune cells was adjusted to 1 × 10⁶ cells per sample and then incubated with the antibody mixture for 30 min on ice in the dark. After staining, cells were washed with PBS and resuspended in the appropriate buffer. Data acquisition and analysis were performed using Cytek Aurora flow cytometer (Cytek Biosciences, Fremont, CA, USA), identifying immune cell populations based on surface marker expressions. Anti-mouse antibodies used in this study are provided in Supplementary Table S1 and all purchased from eBioscience (San Diego, CA, USA). Neutrophils, NK cells, and CD8⁺ T cells were identified by flow cytometry using standard gating strategies, as depicted in Supplementary Figure S1 .

Colonic tissues were collected from mice, washed with cold PBS, and fixed in 4% paraformaldehyde solution at 4 °C for preservation. The fixed tissues were dehydrated through a graded series of alcohols and embedded in paraffin blocks for histopathological analysis. Paraffin-embedded tissue blocks were sectioned into 4 µm thick slices using a microtome (Reichert-Jung Biocut 2030, Ramsey, MN, USA). The tissue sections were deparaffinized in xylene, rehydrated through a graded ethanol series, and stained with hematoxylin and eosin (H&E) for the evaluation of histopathological changes under a light microscope. Histological inflammation was assessed using the simplified Geboes score as previously described (38). This scoring system comprises five categories: grade 0, structural changes without active inflammation; grade 1, chronic inflammatory infiltrates in the lamina propria; grade 2, presence of neutrophils in the lamina propria and/or epithelium; grade 3, epithelial damage with crypt destruction or erosion; and grade 4, ulceration. Higher grades correspond to greater histological disease activity.

Total RNA was extracted from the intestine tissue using Trizol reagent (Invitrogen). The quality of RNA was evaluated by Agilent 4200 TapeStation System (Agilent Technologies, Santa Clara, CA, USA), and RNA integrity was confirmed by agarose gel electrophoresis. RNA integrity number (RIN) was measured using the Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA). Only RNA samples with a RIN score exceeding 7 or 8 were selected for further library preparation and sequencing. Equal amounts of RNA were pooled from three individual samples per group to reduce interindividual variability. Library preparation and RNA sequencing were conducted using the QuantSeq 3’ mRNA-Seq Library Prep Kit FWD (Lexogen, Vienna, Austria) according to the manufacturer’s instructions. Briefly, the process consisted of preparing RNA, hybridizing with an oligo-dT primer with an Illumina-compatible sequence at the 5’ end, and performing reverse transcription. After the RNA template was degraded, second-strand synthesis was initiated using random primers with Illumina-compatible linker sequences at the 5’ end. The generated double-stranded library was purified using magnetic beads to remove unintegrated reaction components. The library was then amplified using the entire adapter sequence required for cluster formation. The completed libraries were purified after amplification to eliminate any remaining PCR components. Single-end 75 bp sequencing was performed using the NextSeq 500 platform (Illumina, Inc., USA). Raw sequencing data were quality controlled using FastQC prior to data processing (39). Sequenced reads were trimmed for adapter sequences, and low-quality reads were filtered using bbduk (40). The purified reads were aligned to the reference genome using STAR (41), and read quantification was performed using HTSeq-count (42). Read counts were processed based on TMM+CPM.

DEGs were identified in intestinal samples of ERAP1 WT and ERAP1⁺/⁻ mice. Sequencing reads were aligned to the GENCODE GRCm38 v23 transcript reference using Bowtie2, and gene expression levels were quantified using RSEM. Expression values were normalized using the trimmed mean of M values (TMM) method. DEG analysis was performed using the edgeR package. Genes with a false discovery rate (FDR)-adjusted P-value < 0.05 and an absolute log2 fold change (|logFC|) > 2 were considered differentially expressed. Genes with logFC < 0 were classified as downregulated, while those with logFC > 0 were classified as upregulated.

To quantify gene expressions, real-time quantitative PCR (qRT-PCR) was conducted using the 7500 Real-Time PCR System (Applied Biosystems). Total RNA was isolated from spleen tissue using TRIzol reagent (Thermo Fisher, Waltham, MA, USA), and the concentration and purity were measured using a spectrophotometer. The extracted RNA was reverse-transcribed into cDNA using the PrimeScript cDNA Synthesis Kit (Takara, Japan) according to the manufacturer’s protocol. SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA) was used for PCR amplification. Each qRT-PCR reaction was performed in a total volume of 20 µL, including 1 µL of synthesized cDNA template. The thermal cyclic program was set as follows: initial denaturation was performed at 94 °C for 2 min, followed by 40 cycles of denaturation at 94 °C for 3 s, annealing at 55 °C for 30 s, and extension at 72 °C for 30 s. The final extension step was performed at 72 °C for 10 min. Gene expression was normalized to β-actin as an internal control, and relative quantification was determined using the 2−ΔΔCt method. Specific primers for the target genes are presented in Table 1 .

GO analysis is a widely used method for determining the biological functions of genes and their RNA or protein products in high-throughput transcriptomic or genomic datasets (43). KEGG pathway enrichment analysis provides insights into molecular interaction and reaction networks by integrating information on genomes, biological pathways, diseases, drugs, and chemicals. The Database for Annotation, Visualization, and Integrated Discovery (DAVID; version 6.7; https://davidbioinformatics.nih.gov/) was used to perform functional annotation and enrichment analysis. DAVID offers a comprehensive suite of GO term annotations and KEGG pathway enrichment to help interpret the biological significance of large gene lists (44). Functional categories and pathways were considered statistically significant if the p-value was less than 0.05.

To assess significant differences among experimental groups, a one-way analysis of variance with Tukey’s post hoc test and a two-tailed t-test (95% confidence interval) were performed, and appropriate corrections were applied for multiple comparisons. All statistical analyses were performed using GraphPad Prism version 8.3.1 (GraphPad Software, La Jolla, CA, USA). Each experiment was conducted at least twice, and the data were expressed as the mean ± standard deviation. The results were considered significant at the significance level of p < 0.05.

DSS-induced colitis model was used to investigate the role of ERAP1 in colitis development. ERAP1 wild-type (WT) and ERAP1⁺/⁻ mice were given 2.5% DSS in their drinking water for two separate 7-day cycles, each followed by a 7-day recovery period with plain water, over a total duration of four weeks. During the final 7 days of the second DSS cycle, the treatment group received sulfasalazine in place of plain water ( Figure 1A ). Body weight was monitored every three days throughout the study. Mice treated with DSS exhibited significant weight loss, with WT mice losing approximately 20% and ERAP1⁺/⁻ mice losing around 30% of their initial body weight. Sulfasalazine treatment did not restore body weight to baseline levels ( Figure 1B ). Colon length, a key indicator of colitis severity, was significantly reduced in DSS-treated WT and ERAP1⁺/⁻ mice compared to controls. While sulfasalazine treatment restored colon length in DSS-induced WT mice, it had no significant effect in DSS-treated ERAP1⁺/⁻ mice ( Figure 1C ). Additionally, sulfasalazine treatment led to a reduction in spleen size in both WT and ERAP1⁺/⁻ colitis mice compared to those treated with DSS alone ( Figure 1D ). To further assess the impact of ERAP1 on colonic morphology during inflammation, H&E staining was performed. Control mice exhibited well-preserved colonic architecture characterized by intact crypts, organized villous structures, and no signs of ulceration or inflammatory infiltration. In contrast, DSS-treated WT and ERAP1⁺/⁻ mice showed severe colonic damage, including structural disorganization, villous atrophy, extensive inflammatory cell infiltration, epithelial injury, and increased apoptosis. Sulfasalazine treatment ameliorated some of these pathological alterations in WT colitis mice but failed to improve colonic damage in ERAP1⁺/⁻ mice ( Figure 1E ). Histological evaluation using the Geboes scoring system showed that both WT and ERAP1⁺/⁻ mice treated with DSS had significantly higher scores than the control group, indicating severe colonic inflammation (p < 0.001). In WT mice, sulfasalazine treatment significantly reduced Geboes scores compared with the DSS group (p < 0.05), indicating partial improvement in mucosal damage. In contrast, sulfasalazine had no effect in ERAP1⁺/⁻ mice ( Figure 1F ), suggesting that partial loss of ERAP1 may compromise key mechanisms involved in colonic protection and the resolution of inflammation.

To assess the impact of ERAP1 on neutrophil infiltration during colitis progression, Ly6G+ CD11b+ neutrophils were analyzed in peripheral blood leukocytes (PBLs) and intraepithelial lymphocytes (IELs) of DSS-treated ERAP1 WT and ERAP1⁺/⁻ mice using flow cytometry. In PBLs, ERAP1⁺/⁻ colitis mice exhibited a significant increase in the frequencies of Ly6G+, CD11b+, and Ly6G+CD11b+ cells compared to ERAP1⁺/⁻ healthy controls (p < 0.05) ( Figures 2A-C ). Similarly, these neutrophil subsets were significantly elevated in sulfasalazine-treated ERAP1⁺/⁻ colitis mice compared to ERAP1⁺/⁻ controls (Ly6G+, p < 0.01; CD11b+, p < 0.05; Ly6G+CD11b+, p < 0.01), indicating that sulfasalazine treatment failed to suppress neutrophilic inflammation under ERAP1 haploinsufficient conditions. Moreover, Ly6G+ and Ly6G+CD11b+ cells were markedly elevated in WT colitis mice compared to WT healthy controls (p < 0.05) ( Figures 2A, C ). In IELs, Ly6G+ neutrophils were significantly increased in WT colitis mice treated with sulfasalazine compared to untreated controls (p < 0.05) ( Figure 2D ). However, the frequencies of CD11b+ and Ly6G+CD11b+ cells in IELs remained unchanged across all experimental groups ( Figures 2E, F ). Moreover, ERAP1⁺/⁻ mice exhibited a 70% increase in Ly6G+CD11b+ neutrophil levels compared to WT mice under colitis conditions, indicating that partial ERAP1 deficiency affects neutrophil infiltration in this model. These findings indicate that ERAP1 may play a role in regulating neutrophil accumulation during colitis.

To observe the effect of ERAP1 on DC activation during colitis conditions, the expression of co-stimulatory molecules CD40, CD83, CD80, CD86, and CD11c was examined in PBLs and IELs of ERAP1 WT and ERAP1⁺/⁻ mice using flow cytometry. In PBLs, the frequencies of CD40+ and CD80+ cells were significantly increased in DSS-treated ERAP1⁺/⁻ mice (p < 0.01) and in sulfasalazine-treated ERAP1⁺/⁻ colitis mice (p < 0.05) compared to ERAP1⁺/⁻ healthy controls ( Figures 3A, B ). Additionally, DSS-treated ERAP1⁺/⁻ mice showed significantly higher frequencies of CD40+, CD80+, and CD83+ cells compared to DSS-treated WT mice (p < 0.05) ( Figures 3A-C ). Notably, CD40⁺ cell frequencies remained elevated in ERAP1⁺/⁻ colitis mice even after sulfasalazine treatment, compared to WT colitis mice (p < 0.05). In contrast, CD86+ cells were significantly reduced in ERAP1⁺/⁻ colitis mice (p < 0.05); this reduction was also observed following sulfasalazine treatment (p < 0.05) ( Figure 3D ). These findings suggest an altered immune response and potential dysregulation of antigen presentation under conditions of the ERAP1 haploinsufficiency. CD11c+ cell frequencies in PBLs remained unchanged across all groups ( Figure 3E ). In IELs, sulfasalazine-treated WT colitis mice exhibited a significant increase in CD40⁺ cell frequencies compared to untreated WT colitis mice (p < 0.05) ( Figure 3F ). Moreover, sulfasalazine-treated ERAP1⁺/⁻ mice showed significantly elevated CD86+ cell (p < 0.05) and CD11c+ cell (p < 0.05) frequencies compared to ERAP1⁺/⁻ healthy controls ( Figures 3I, J ). Further, ERAP1⁺/⁻ colitis mice showed a higher frequency of CD11c+ cells than WT colitis mice following sulfasalazine treatment. In addition, CD11c+ expression was significantly elevated in DSS-treated WT mice compared to WT healthy controls (p < 0.05) but was decreased by 55% upon sulfasalazine treatment, indicating a partial recovery effect of sulfasalazine in modulating immune responses during colitis ( Figure 3J ). The percentages of CD80+ and CD83+ cells in IELs remained unchanged across all experimental groups ( Figures 3G, H ). These results collectively indicate that ERAP1 is essential in regulating DC activation during experimental colitis by modulating the expression of co-stimulatory molecules.

To elucidate the function of ERAP1 in immune response, the percentages of CD4+, CD8+, NK1.1+, and CD8+NK1.1+ cells in PBLs and IELs of ERAP1 WT and ERAP1⁺/⁻ mice were analyzed using flow cytometry. In PBLs, CD4+ T cells were significantly decreased in DSS-treated ERAP1⁺/⁻ mice than those from healthy ERAP1⁺/⁻ mice (p < 0.01) ( Figure 4A ). Similarly, CD4+ and CD8+ T cells were reduced after sulfasalazine treatment in WT colitis mice, whereas this was not the case in WT colitis mice (p < 0.05) ( Figures 4A, B ). Moreover, CD8+ T cells were considerably decreased in ERAP1⁺/⁻ colitis mice in contrast to WT colitis mice (p < 0.05) ( Figure 4B ). The percentages of NK1.1+ and CD8+NK1.1+ cells in PBLs remained unchanged across all experimental groups ( Figures 4C, D ). In IELs, WT colitis mice exhibited a significant reduction in CD4+, CD8+, NK1.1+, and CD8+NK1.1+ cells compared to WT control mice. However, sulfasalazine treatment led to an increase in these cell populations, with a significant rise in CD4⁺ T cells, indicating a potential immunoregulatory effect that promotes recovery during colitis ( Figures 4E-H ). NK cell depletion worsens colitis by promoting leukocyte infiltration, colonic damage, and pro-inflammatory profiles, as previously reported (18). In sulfasalazine-treated ERAP1⁺/⁻ colitis mice, the percentages of NK1.1+ (p < 0.05) and CD8+NK1.1+ (p < 0.05) cells were reduced than those in ERAP1⁺/⁻ healthy controls ( Figures 4G, H ). Additionally, NK1.1+ cell levels were markedly lower in ERAP1⁺/⁻ colitis mice compared to WT colitis mice (p < 0.05) after sulfasalazine treatment ( Figure 4G ). These results suggest that ERAP1 deficiency may impair immune responses, worsen colitis symptoms, and reduce the efficacy of sulfasalazine.

RNA- sequencing analysis (RNA-seq) was performed to identify DEGs between ERAP1 WT and ERAP1⁺/⁻ mice. Total RNA was extracted from intestinal tissues of both groups for the analysis. As shown in the heatmap and scatter plot, there were 428 genes differentially expressed with fold change (FC) > 2 and p < 0.05. Among them, 111 genes (26%) were upregulated and 317 genes (74%) were downregulated in ERAP1⁺/⁻ samples compared to ERAP1 WT ( Figures 5A, B ).

Based on the RNA-seq results, 28 representative genes were selected based on a FC greater than 4 and p < 0.05, and reported relevance to colitis, colorectal cancer, or immune regulation, as well as their functional roles in inflammation, epithelial integrity, and immune cell modulation (45–47). These included Ang4, Aqp4, Anxa9, Atp2a1, Cdsn, Dsc1, ERAP1, Fabp2, Flg, Hepacam2, Igha, Igkc, Insl5, Ivl, Jchain, Krt15, Krt17, Lor, Mzb1, Neb, Reg3b, Reg3g, Sbsn, Scd3, Serpina3j, Spink5, Trim29, Ttn. Among them, 11 genes (Ang4, Aqp4, Fabp2, Hepacam2, Igha, Igkc, Insl5, Jchain, Mzb1, Reg3b, and Reg3g) were significantly upregulated in ERAP1⁺/⁻ mice, while 17 genes (Anxa9, Atp2a1, Cdsn, Dsc1, ERAP1, Flg, Ivl, Krt15, Krt17, Lor, Neb, Sbsn, Scd3, Serpina3j, Spink5, Trim29, Ttn) were downregulated relative to ERAP1 WT ( Figure 5C ).

To investigate the functional relevance of DEGs between ERAP1 WT and ERAP1⁺/⁻ mice, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses were conducted. GO enrichment was categorized into three domains: biological process (BP), cellular component, and molecular function (MF). The top 10 significantly enriched GO terms for each category are presented in Figure 6 . In the BP category, genes were significantly enriched in terms including keratinization, intermediate filament organization, epithelial cell differentiation, peptide cross-linking, keratinocyte differentiation, response to bacterium, establishment of skin barrier, lipid metabolic process, and regulation of striated muscle contraction ( Figure 6A ). For the CC category, enriched GO terms included cornified envelope, intermediate and keratin filaments, extracellular space, cytoplasm, extracellular region, myosin II complex, side of membrane, desmosome, and myosin complex ( Figure 6B ). In the MF category, genes were enriched in structural constituent of skin epidermis, structural molecule activity, structural constituent of muscle, calcium ion binding, transition metal ion binding, actin binding, microfilament motor activity, symporter activity, and metal ion binding ( Figure 6C ). KEGG pathway analysis revealed significant enrichment in pathways related to muscle contraction, estrogen signaling, Staphylococcus aureus infection, motor proteins, gastric acid secretion, peroxisome proliferator-activated receptor (PPAR) signaling, pancreatic secretion, p53 signaling, cAMP signaling, and cardiac muscle contraction ( Figure 6D ).

To further evaluate the impact of ERAP1 expression on DSS-induced colitis and sulfasalazine treatment in ERAP1 WT and ERAP1⁺/⁻ mice, three genes (Anxa9, Atp2a1, and Hepacam2) were selected from the RNA-seq data and validated using RT-qPCR in spleen tissue. This approach allowed us to assess systemic immune responses in addition to local colonic changes. Anxa9, a novel marker associated with poor prognosis, was significantly upregulated in ERAP1⁺/⁻ colitis mice compared to ERAP1⁺/⁻ controls (p < 0.05), with a further significant increase observed following sulfasalazine treatment (p < 0.05) ( Figure 7A ). Atp2a1, a gene associated with increased risk of IBD (48), also showed a significant increase in expression in sulfasalazine-treated ERAP1⁺/⁻ colitis mice compared to ERAP1⁺/⁻ controls (p < 0.001) ( Figure 7B ). Hepacam2, a potential biomarker for distinguishing UC (49), was upregulated by 53% in ERAP1⁺/⁻ mice relative to ERAP1 WT mice in the control group, consistent with RNA-seq results ( Figure 7C ). Notably, mRNA expression levels of Anxa9 (p < 0.05), Atp2a1 (p < 0.001), and Hepacam2 (p < 0.01) were all significantly upregulated in ERAP1⁺/⁻ colitis mice relative to WT colitis mice after sulfasalazine treatment ( Figures 7A-C ), indicating its potential role in colitis pathogenesis and therapeutic response. Additionally, three other DEGs (Gadd45b, Reg3β, Spink5) were examined for their reported relevance to IBD (50, 51), but none showed significant expression differences between groups ( Supplementary Figure S2 ). Beyond RNA-seq validation, we also assessed Foxp3 and IL - 17, two immune regulatory markers associated with Treg and Th17 responses, respectively. RT-qPCR analysis of splenic tissue revealed a modest reduction in Foxp3 expression in ERAP1⁺/⁻ colitis mice compared with ERAP1⁺/⁻ controls (p < 0.05), which persisted even after sulfasalazine treatment. Furthermore, under colitis conditions, ERAP1⁺/⁻ mice exhibited a 64% decrease in Foxp3 expression compared with WT mice. In contrast, IL - 17 levels remained unchanged across all experimental groups ( Supplementary Figure S3 ).

These results highlight changes in ERAP1 expression and differential regulation of these genes in response to sulfasalazine treatment, suggesting their potential as biomarkers for colitis progression and therapeutic efficacy.