Four Gene Expression Database (GEO) datasets of patients with active UC and healthy controls were included in our study. The keywords utilized for the search include “Homo sapiens,” “ulcerative colitis,” “active’, “healthy control” and “mucosal biopsy.” Spain cohort GSE38713 (27), USA cohort GSE92415 (28) and Belgium cohorts GSE75214 (29) and GSE73661 (30) were used in this study (Table 1).

The clinical tissue samples of 10 UC patients diagnosed according to ECCO guidelines for IBD (31) and 10 controls were obtained from Linyi People’s Hospital. Participants in the control group were selected from patients undergoing colonoscopy for polyps and cancer surveillance. Subjects were excluded if they had undergone major abdominal surgery; or had a history of malignant neoplasms. Colonic tissue samples from patients were taken under colonoscopy. This study was approved by the Medical Ethics Committee of Linyi People’s Hospital (YX200563).

We retrieved the datasets GSE38713, GSE92415, GSE75214, and GSE73661 from the GEO database and the CIBERSORTx algorithm was utilized to assess M1 macrophage infiltration in UC patients from the four GEO cohorts (32, 33). The transcriptomic dataset was uploaded as a mixed file to the CIBERSORTx web portal,¹ and the LM22 signature matrix was utilized in the deconvolution analysis to define cell populations. The LM22 Signature Matrix, provided by CIBERSORT, is a reference database for analyzing the composition of immune cells in complex tissues (34). Combined with the CIBERSORT algorithm, it enables the deconvolution of bulk gene expression data to quantify immune cell subsets to support disease research and clinical prediction. The data output from CIBERSORTx was analyzed using the R software.

The extracted gene expression data from the four GEO cohorts were processed using the “limma” package in R software (version 4.4.1) with adjusted p < 0.05 and |logFC| >3 for identifying differentially expressed genes (DEGs) across varying conditions. Furthermore, we employed the “ggvenn” package to construct a Venn diagram for identifying the common DEGs across the four datasets. Additionally, we conducted a correlation analysis between these common DEGs and M1 macrophage proportion in each dataset using the “corrplot” package.

To gain insights into the biological functions of extracted gene expression data in the context of Gene Ontology (GO) analysis they are involved in, the gene expression data in GSE38713, GSE92415, GSE75214, and GSE73661 were subjected to GO-cellular component (CC) pathway analysis using the Gene Set Variation analysis (GSVA) package, which is a non-parametric and unsupervised method for evaluating gene enrichment in transcriptomes. By providing a comprehensive score for interesting gene sets, GSVA converts genetic variation into pathway-level changes and further determines the biological functions of samples. The GSVA data was then processed using the “limma” package to identify the top CC pathways.

Female C57BL/6 mice aged 6 to 8 weeks were purchased from JEKAIYER (Jinan, China). The experimental protocol was strictly in accordance with the ethical regulations of the Animal Care and Use Committee of Linyi People’s hospital. Animals were maintained at a temperature of 23 ± 2°C and a humidity of 55 ± 10% in a specific pathogen-free environment with a 12 h light/dark cycle and were provided free access to standard laboratory diet and water.

After acclimatization for at least 1 week, the mice were randomly divided into 5 groups (n = 5/each group): control group, DSS-Day2 group, DSS-Day7 group, DSS + 1-MT (2 mg/mL)-Day7 group and 1-MT (2 mg/mL) group. As previously described (35), colitis was induced by 3.5% DSS dissolved in drinking water for 7 consecutive days, while control mice drank the same volume of distilled water. 1-MT was dissolved in sweet-tasting weakly acidic drinking water at a dose of 2 mg/mL (36, 37) and given to mice daily, alone or with DSS, for 7 days. Body weights were recorded daily. Colitis severity was evaluated by the disease activity index (DAI) based on weight loss, stool bleeding and stool consistency (38, 39). Fecal specimens were systematically collected during the circadian morning active phase [Zeitgeber Time (ZT) 3–5]. Feces were immediately frozen after collection and stored in a vacuum sealed environment at −80°C, and unified detection was performed after the end of the experiment. After animals in each group were sacrificed, colons were measured and processed for histopathological studies (40), western blotting and enzyme-linked immunosorbent (ELISA) assay. This study was approved by the Science and Technology Ethics Committee of Linyi People’s Hospital (202410-A-002).

RAW 264.7 cells were cultured in Dulbecco Modified Eagle medium containing 10% (volume/volume) fetal bovine serum, 100 U/mL penicillin and 100 U/mL streptomycin (all from Gibco, United States) at 37°C and 5% CO₂ in a plastic disposable cell culture flask (Corning, United States). To investigate the effect of IDO1 on the polarization of M1 macrophages, cells were divided into eight groups: control group, 6 h model group (LPS 20 μg/mL + IFN-γ 4 μg/mL), 24 h model group (LPS 20 μg/mL + IFN-γ 4 μg/mL), 1-MT (100 μM) + LPS-IFN-γ group, 1-MT (100 μM) group, IDO1-overexpression group, IDO1-overexpression + LPS-IFN-γ group and IDO1-overexpression + 1-MT + LPS-IFN-γ group. Cells were seeded at a density of 2 × 10⁵ cells/mL in 6-well plates (2 mL/well) and allowed to adhere for 24 h in a 37°C, 5% CO₂ incubator until reaching 70–80% confluency. For polarization induction, culture medium was replaced with fresh complete medium containing 20 μg/mL LPS and 4 μg/mL IFN-γ. Prior to stimulation, cells were gently washed twice with prewarmed PBS to remove residual cytokines. The IDO1 inhibitor 1-MT was dissolved in dimethylsulfoxide (DMSO; Sigma-Aldrich) prior to administration and cells were pretreated with 1-MT for 6 h prior to induction with LPS + IFN-γ. Three independent replications were performed for all the experiments. Upon reaching the designated treatment time points, cells from each experimental group were immediately harvested. RNA and protein were subsequently isolated using TRIzol-based dual extraction protocols, aliquoted into sterile cryovials, and cryopreserved at −80°C under vacuum-sealed conditions to ensure biomolecular stability for downstream analyses.

Overexpression plasmids were transfected into cells with E-trans TM (Genechem) according to the manufacturer’s protocol. Plasmids were synthesized by Genechem (Shanghai, China). Vector name and element order is listed in Table 2.

RAW264.7 cells were plated in 6-cm culture dishes to achieve 80% confluence. Total RNA was extracted using Trizol reagent (Thermo Fisher Scientific). The mRNA levels of IDO1, inflammatory cytokines and iNOS were evaluated by reverse transcription (RT) with Moloney murine leukemia virus (M-MLV) reverse transcriptase (Thermo Fisher), followed by conventional quantitative PCR (qPCR) with SYBR Green Pro Taq HS (Agbio, Hunan, China). The qPCR was performed on an ABI 7500 real-time PCR system (Thermo Fisher). GAPDH was used as an internal control. The oligonucleotide sequences of primers used are listed in Table 2. The PCR amplification conditions were as follows: predenaturation at 95°C for 30 s, followed by 40 cycles at 95°C for 5 s and 60°C for 35 s, and then the primers were unchained at 60°C for 1 min and 95°C for 15 s. Relative mRNA expression was calculated using the 2 − ΔΔCT method.

The total RNA of the control group, LPS-IFN-γ group and IDO1-overexpression + LPS-IFN-γ group was extracted with Trizol reagent (Invitrogen, Carlsbad, United States). Total RNA concentration and purity were measured using an ND-1000 NanoDrop spectrophotometer (NanoDrop Technologies, Wilmington, United States). Quality control and deep sequencing are conducted by Major Bio (Shanghai, China). Utilizing the MajorBio Cloud platform (Shanghai, China), Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis were performed on the genes derived from sequencing with the top six pathways visualized using MajorBio Cloud platform.

Colon tissue of colitis mice was homogenized in PBS buffer (pH 7.4) and centrifuged at 14,000 g and 4°C for 30 min. Supernatants were collected and stored at −80°C for further analysis. The protein concentration was measured using a BCA protein assay kit (Beyotime, Jiangsu, China), and the level of iNOS, IL-6, TNF-α was detected by ELISA kit.

Total protein extracts were obtained from colon tissues or RAW264.7 in ice-cold RIPA lysis buffer (Beyotime) with phosphatase inhibitor mixture and protease inhibitor (Thermo Fisher). Using the BCA protein assay kit (Beyotime) quantitative concentration. The extracted proteins were separated by SDS-PAGE and transferred onto 0.22 μm polyvinylidene difluoride membranes (Millipore, MA, United States). Membranes were blocked with 5% non-fat powdered milk for 2 h at room temperature (25–30°C), and then incubated separately with rabbit antibodies against β-tubulin (1:2,000; Cat No. 2128S; CST), XBP1s (1:1,000; Cat No. 12782S; CST), iNOS (1:500; Cat No. PA1-036; Thermo Fisher) and rat antibodies against GRP78 (1:200; Cat No. sc-13539; Santa Cruz Biotechnology), IDO1 (1:200; Cat No. sc-53978; Santa Cruz Biotechnology) in TBST (Tris-buffered saline, 0.1% Tween 20) buffer at 4°C overnight. Subsequently, membranes were incubated with HRP-conjugated goat anti-rabbit IgG or HRP-conjugated goat anti-mouse IgG secondary antibody (Thermo Fisher Scientific) for 1.5 h at room temperature. The protein was visualized by BeyoECL Star (Beyotime) using chemiluminescence and quantified with Image J software.

1-methyl-DL-tryptophan (1-MT) and Lipopolysaccharides (LPS) were purchased from Sigma Chemical Co (St. Louis, MO, United States). IFN-γ was purchased from Peprotech (Rocky Hill, NJ, United States) and the reconstitution was done in 1% BSA in phosphate buffer saline (PBS) at a concentration of 1,000 μg/mL and stored at −80°C. 4-Phenylbutyric acid (4-PBA) was purchased from APEXBIO (China) and dissolved in sterile, enzyme-free water and stored in a refrigerator at −80°C. Antibody name (ratio, product number, manufacturer): β-tubulin (1:2,000; Cat No. 2128S; CST), XBP1s (1:1,000; Cat No. 12782S; CST), iNOS (1:500; Cat No. PA1-036; Thermo Fisher) and rat antibodies against GRP78 (1:200; Cat No. sc-13539; Santa Cruz Biotechnology), IDO1 (1:200; Cat No. sc-53978; Santa Cruz Biotechnology).

Results are presented as mean ± standard error of mean (SEM). Statistical analysis was performed using one-way analysis of variance with Tukey’s multiple comparison test to analyze the significance among multiple groups, and an unpaired Student’s t-test was used to analyze the significance between the two groups. Differences were considered to be statistically significant if p < 0.05. No exclusion criteria were incorporated into the design of the experiments for this study.

Macrophages play a crucial role in chronic inflammation and pathological processes. Research has demonstrated that the number of M1 macrophages significantly increases during active UC, indicating their involvement in the pathogenesis of UC (7, 41). Consequently, we employed the CIBERSORTx web portal to assess the differences in M1 macrophage proportion between the UC group and healthy control group across four cohorts obtained from the GEO database. The results showed that the infiltration of M1 macrophages was significantly increased UC patients in the active phase, and the results of the four cohorts were consistent (Figure 1A). To further elucidate the upstream regulatory mechanisms governing macrophage polarization in UC, we performed DEG analysis on each selected cohort (adjusted p < 0.05, |logFC| >3) and conducted an intersection analysis of the DEGs identified across all cohorts, revealing that a total of 17 common DEGs exhibited significant differences in the four cohorts (Figure 1B). Subsequently, we employed the “corrplot” package to conduct a correlation analysis between the 17 selected common DEGs and M1 macrophage proportion across the four cohorts. The results indicated that IDO1 exhibited the strongest correlation with M1 macrophages in all cohorts (Figures 1C–F). In parallel, our analysis of publicly available single-cell RNA sequencing dataset (42) revealed that compared with healthy controls (HC), M1 macrophages predominantly localized in tissues from patients with acute-phase UC and Crohn’s disease (Supplementary Figure 1). This study further demonstrated that IDO1 overexpression was principally concentrated within these M1 macrophage populations (Supplementary Figure 1). Subsequent co-expression analysis additionally identified possible concurrent enrichment of both GRP78 and XBP1s in highly-IDO1-expressing M1 macrophages (Supplementary Figure 1).

To further investigate the potential mechanism of IDO1 in macrophage polarization in the progression of UC, we used the GSVA algorithm to perform differential analysis of the cellular component (CC) terms within the gene expression profiles across all cohorts. The results indicated that the endoplasmic reticulum chaperone complex exhibited significant differential expression in the four cohorts (Figures 2A–D). The Chord Diagram analysis using the “circlize” package indicated a strong significant positive correlation among GRP78, XBP1, IDO1, ER stress, and M1 macrophages (Figure 2E). The expression levels of IDO1, GRP78 and XBP1 were significantly elevated during the acute colitis phase of UC in the four GEO cohorts (Figure 2F). Furthermore, we confirmed the increased expression of IDO1 in UC by performing western blotting analysis in newly recruited UC patients (Figure 2G).

To investigate the impact of IDO1 on UC development, we used DSS-induced colitis mice and inhibited IDO1 using 1-MT (43). After DSS administration, we observed severe colitis at day 7 characterized by greater weight loss, higher HE score, shorter colon length and fecal bleeding, with significantly higher DAI scores as compared to control group and day 2 after colitis (Figures 3A–E). However, administration of 1-MT significantly (2 mg/mL) ameliorated DSS-induced colitis in mice, reversed weight loss (Figure 3A), decreased HE score (Figure 3B) and decreased DAI score (Figure 3E). These findings collectively revealed that inhibition of IDO1 exhibited a strong therapeutic effect in ameliorating DSS-induced colitis.

Western blotting analysis indicated that IDO1 and iNOS, a molecular marker for M1-polarized macrophage, was significantly increased after induction of colitis (Figure 3F). Also, ELISA results showed that markers of M1 macrophages (iNOS, IL-6, TNF-α) in colon tissue increased significantly during the colitis stage (Figure 3G; Supplementary Figure 2). However, the expression of both M1 makers and IDO1 was significantly downregulated after treatment with 1-MT in DSS-induced colitis mice (Figures 3F,G; Supplementary Figure 2).

To further verify that IDO1 might cause M1 macrophage polarization via the GRP78-XBP1 pathway to promote colitis development, we detected the expression of GRP78 and XBP1s in colitis mice. The study findings suggested that the expression levels of GRP78 and XBP1s were significantly increased after colitis induction, which was consistent with the results in human datasets. However, when IDO1 was inhibited by 1-MT, the colitis severity was significantly ameliorated with a marked decrease in the expression levels of both GRP78 and XBP1s (Figure 3H).

To elucidate the role of IDO1 in the regulation of M1 macrophage polarization in vitro, we used LPS + IFN-γ to create the transition towards M1 phenotype in RAW264.7 cells. We observed a significant increase in the mRNA and protein levels of IDO1 and elevated mRNA levels of inflammatory factors after LPS + IFN-γ induction for 6 h and 24 h, with the most significant increase at 6 h (Figures 4A,B). Concurrently, a 6 h pretreatment with 1-MT (100 μM) effectively reversed the expression level of IDO1 and production of IL-6 and TNF-α induced by LPS + IFN-γ (Figures 4C,D). However, overexpression of IDO1 induced the production of IL-6 and TNF-α, and aggravated the production of those proinflammatory cytokines induced by LPS + IFN-γ (Figures 4E,F).

Subsequently, RNA sequencing was performed to investigate GO and KEGG enrichment pathways involved in overexpression of IDO1 in RAW264.7 cells. The results showed that the gene expression data were enriched in endoplasmic reticulum to cytosol transport pathway (GO analysis, Figure 4G) and protein processing in ER pathway (KEGG analysis, Figure 4H) in IDO1-overexpressing LPS + IFN-γ RAW264.7 cells as compared to LPS + IFN-γ RAW264.7 cells. Also, the mRNA level (Figure 5A) and protein level (Figure 5B) of GRP78 and XBP1s were significantly increased in IDO1-overexpression RAW264.7 cells and were more significantly upregulated in IDO1-overexpressing LPS + IFN-γ RAW264.7 cells; while, pretreatment with 1-MT reversed the increase of GRP78 and XBP1s (Figures 5A,B).

To further validate the role of IDO1 in macrophages, the expression level of iNOS was detected and it was revealed that overexpression of IDO1 increased the level of iNOS and more significantly upregulated iNOS expression in IDO1-overexpressing LPS + IFN-γ RAW264.7 cell. The results above demonstrated that IDO1 might induce the transition towards M1 macrophage phenotype via ER stress-associated GRP78-XBP1 pathways.

To further investigate whether IDO1 regulates the M1 polarization of RAW264.7 cells via the ER stress pathway, we employed 4-PBA to inhibit ER stress signaling in RAW264.7 cells (44, 45). The data indicated that although the mRNA level of IDO1 was not significantly decreased with 4-PBA treatment, the production of IL-6 and TNF-α as well as the mRNA level of GRP78 and XBP1s were significantly reduced (Figure 5C). In LPS + IFN-γ-induced RAW264.7 cells, administration of 4-PBA significantly down-regulated iNOS expression level and decreased the level of GRP78 and XBP1s (Figure 5D).