IDO1 inhibitor enhances the effectiveness of PD-1 blockade in microsatellite stable colorectal cancer by promoting macrophage pro-inflammatory phenotype polarization

Abstract

Microsatellite stable (MSS) colorectal cancer (CRC) is a subtype of CRC that generally exhibits resistance to immunotherapy, particularly immune checkpoint inhibitors such as PD-1 blockade. This study investigates the effects and underlying mechanisms of combining PD-1 blockade with IDO1 inhibition in MSS CRC. Bioinformatics analyses of TCGA-COAD and TCGA-READ cohorts revealed significantly elevated IDO1 expression in CRC tumors, correlating with tumor mutation burden across TCGA datasets. In vivo experiments demonstrated that the combination of IDO1 inhibition and PD-1 blockade significantly reduced tumor growth and increased immune cell infiltration, particularly pro-inflammatory macrophages and CD8+ T cells. IDO1 knockdown in CRC cell lines impaired tolerance to interferon-γ and increased apoptosis in vitro, which were rescued by the application of kynurenine, the end product of IDO1. IDO1 knockdown in MSS CRC enhanced the effectiveness of PD-1 blockade therapy in vivo. IDO1 knockdown cancer cells promoted pro-inflammatory macrophage polarization and enhanced phagocytic activity in vitro, associated with the upregulation of JAK2-STAT3-IL6 signaling pathway. These findings highlight the role of IDO1 in modulating the tumor immune microenvironment in MSS CRC and suggest that combining PD-1 blockade with IDO1 inhibition could enhance therapeutic efficacy by promoting macrophage pro-inflammatory polarization and infiltration through the JAK2-STAT3-IL6 pathway.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00262-024-03925-w.

Introduction

Colorectal cancer (CRC) is a major global health challenge, being one of the most prevalent cancers worldwide and imposing a significant burden on healthcare systems. Statistics from Chinese National Cancer Center demonstrate that the incidence of colorectal cancer has risen over the past decades, making it the second leading cause of cancer mortality in Chinese population [1]. Despite advances in treatment strategies, the efficacy of current therapies varies, particularly concerning immune checkpoint inhibitor (ICI) therapy.

Among ICIs, targeting programmed cell death protein 1 (PD-1) has demonstrated remarkable efficacy in patients with microsatellite instability-high (MSI-H) tumors [2]. However, this benefit is not universally observed, particularly in patients with microsatellite stable (MSS) tumors, who constitute the majority of CRC, show limited response to PD-1 blockade [3]. To enhance the response of MSS CRCs to ICI therapy, oncologists have explored combining ICIs with other treatments, such as radiotherapy, or the use of HDAC inhibitors and VEGF antibody to promote a pro-inflammatory tumor microenvironment conducive to immune activation [4, 5]. Despite these efforts, the application of immunotherapy in MSS CRC remains challenging.

Indoleamine 2,3-dioxygenase 1 (IDO1) functions as a suppressive immune checkpoint gene pivotal in regulating the tumor microenvironment [6]. IDO1 depletes tryptophan by catalyzing its conversion to kynurenine, leading to the activation of aryl hydrocarbon receptor (AhR) pathway. This activation leads to the nuclear translocation of AhR, fostering an immunosuppressive milieu that promotes tumor growth [7]. In most theories, IDO1 expression is upregulated by pro-inflammatory cytokines, notably interferon-γ (IFN-γ) and is considered a protective “brake” that prevent uncontrollable inflammatory response [8, 9].

Clinical studies targeting IDO1 inhibition have shown variable efficacy [10]. Most of these studies focus on MSI-H CRCs, which are already known to be sensitive to immunotherapy using anti-PD-1 antibodies. Some potential has been observed when IDO1 inhibitors are combined with other ICIs, such as Pembrolizumab (a widely used PD-1 monoclonal antibody); however, none of these studies had explored this combination’s potential in treating MSS CRCs [11].

This study aims to the potential role of IDO1 inhibitor as enhancers of PD-1 blockade therapy in MSS CRCs. By elucidating the interplay between IDO1 inhibition, macrophage behavior and immune cell infiltration, this research seeks to provide insights that could inform the development of more effective, personalized treatment strategies for MSS CRC patients.

Materials and methods

Bioinformatic analysis

ESTIMATE (ver 1.0.13) and TIMER2.0 were used to predict the immune infiltration status in TCGA-COAD and TCGA-READ datasets [12, 13]. The simple nucleotide variation (SNV) datasets were downloaded from TCGA, and the tumor mutation burden (TMB) of each sample were calculated using R package maftools [14]. RNA-seq expression matrix was imported into the GSEA software (Broad, v4.3.2), and the HALLMARK 2023.2 gene sets of Homo sapiens were used for gene set enrichment analysis [15].

Cell culture

CT26, MC38, HCT116, HT29, THP-1 and RAW264.7 were purchased from ATCC. CT26 and THP-1 were cultured with RPMI-1640 with 10% fetal bovine serum (FBS), MC38 and RAW264.7 was cultured with DMEM with 10% FBS, HCT116, HT29 was cultured with Mcoy-5A medium with 10% FBS. All cells were detected mycoplasma free and were maintained with anti-bacterial 1 × penicillin and streptomycin solution (Solarbio Inc, Beijing).

Small interfering RNA transfection

Small interfering RNAs (siRNAs) were dissolved in RNase free water to a final concentration of 100 nmol/μL. Lipofectamine 3000 (Invitrogen) and siRNAs were mixed in Opti-MEM(Gibco) and incubated at room temperature for 20 min to form transfection complexes, which were then added to the cells. 8 h after transfection, the culture medium was removed and replaced with fresh culture medium with FBS. Total RNA or protein was extracted 48 h after transfection for downstream analyses. The RNAi sequences used in this study were listed in Supplementary Table 1.

IDO1 knockdown lentivirus transfection

LVRU6GP plasmids with RNAi sequences were constructed by GeneCopoeia Inc. Lentivirus was packaged in HEK293T cells via co-transfection of target plasmids, psPAX2 and pMD2G using Lipofectamine 3000. The lentivirus-containing supernatant was purified by centrifugation and filtration. Stable cells were selected with puromycin at predetermined concentrations.

Quantitative RT-PCR

Total RNA was extracted using TRIzol (Invitrogen) according to the standard procedure. The RNA was then reverse-transcribed into cDNA using reverse transcription kits (Yeason, Shanghai) following the manufacturer's instructions. RT-qPCR was performed using the SYBR Green mix (Yeason, Shanghai) and specific primers according to the recommended protocol in GTEx (BioRad). The primer sequences are listed in Supplementary Table 2.

Western blot

Cells were lysed in RIPA buffer with 1% PMSF on ice for 30 min. Protein concentration was determined using a BCA kit. Proteins were separated by SDS-PAGE and transferred to membranes. Following blocking with skimmed milk, membranes were incubated with primary antibodies, then HRP-conjugated secondary antibodies for chemiluminescence detection. The expression levels were compared using chemiluminescence blotting. For nuclear and cytoplasmic protein separation, the Nuclear and Cytoplasmic Protein Extraction Kit (Beyotime, P0027) was used according to the manufacturer’s instructions. Antibodies used are listed in Supplementary Table 3.

Animal experiment

All animal experiments were approved by the Ethics Committee of Sun Yat-Sen University Cancer Center (Approval ID: 025503202112028). C57BL/6 or Balb/c mice were purchased from GemPharmatech Co., Ltd (Guangdong) or the Experimental Animal Center of Sun Yat-Sen University Cancer Center. Each mouse was inoculated subcutaneously with 1 × 10.⁵ tumor cells into the left axillary area. Anti-mouse PD-1 (10 mg/kg, InVivoMAb, clone ID: RMP1-14, Cat # BE0146) was administered intraperitoneally every three days, and epacadostat (10 mg/kg, INCB 024360, MedChemExpress) was given daily [16, 17]. For macrophage depletion assay, clodronate liposome or PBS liposome were given intraperitoneally every 5 days. Mice were sacrificed 15 to 17 days after the final treatment. The individual time of drug exposure was determined by pilot studies and previously described methods [18].

Flow cytometry

Tumor tissues were cut into 2 mm pieces and dissociated into single-cell suspension using the Tumor Dissociation Kit (Miltenyi, Cat# 130-096-730) and Single Cell Suspension Dissociator (RWD, China) following the manufacturer’s instructions [19]. Erythrocytes were lysed with BD Pharm Lyse™ solution (Cat # 555,899). After Fc receptor blockade with TruStain FcX™ (Biolegend, Cat# 101,319), surface markers were stained with antibodies listed in Supplementary Table 3. Cells were then permeabilized and fixed using True-Nuclear™ Transcription Factor Buffer Set (Biolegend, Cat# 424,401) for nuclear marker staining. Single fluorescence-stained Compensation Beads (Biolegend, Cat# 424,602) were used for compensation adjustment. The stained cell suspension was analyzed using a CytoFLEX LX Flow Cytometer (Beckman Coulter). Cell markers were determined by previously described articles overlapping the cell marker database CellMarker 2.0 [20]. In brief, Zombie dye was applied to identify viable cells, CD45 was selected as a pan-leukocyte marker. CD3 and CD49b/NK1.1 were selected as T cell marker or NK cell markers, respectively. CD4 and CD8 were used to differentiate cytotoxic T cells and helper T cells. FoxP3 and CD25 double positive cells were identified as regular T cells. In CD3 and CD49b/NK1.1 negative cells, F4/80 and CD11b double positive subset was identified as macrophage, and we used CD11c and CD206 to represent pro-inflammatory and anti-inflammatory macrophage respectively [21–23]. Despite CD11c is not a regular marker for pro-inflammatory macrophage, we analyzed its role in macrophage using single cell RNA sequencing dataset GSE236581 [24]. The result turned out that CD11c (ITGAX in gene symbol) is expressed in a proportion of macrophage that associated with immune response and is upregulated in the HALLMARK IL6, JAK-STAT3 pathway, which is more representative in antigen processing and presentation (Figure S1A-F). Fluorescence labeled antibody for FCS analysis were listed in Supplementary Table 4.

In vitro cell function assays

In vitro cell proliferation was assessed using the CCK-8 viability kit. Briefly, 1 × 10⁴ cells were cultured in a 96-well plate and incubated under standard conditions. Medium was replaced with 10% CCK-8 in serum-free medium and incubated at 37 °C for 45 min. Absorbance at 450 nm was measured using a microplate reader. For the colony formation assay, 500 cells per well were plated in 6-well plates and cultured with 10% FBS medium for 10–14 days to form colonies. For migration and invasion assays, 2 × 10⁵ cells in 250 μL serum-free medium were placed on the upper side of an 8 μm transwell membrane coated with Matrigel. Transwell inserts were placed in 24-well plates containing 10% FBS medium in the lower chamber. After 24 h of incubation, cells on the upper surface were removed. Cells from both assays were stained with crystal violet, photographed, and quantified using ImageJ (1.54 g).

Immune fluorescence staining

Cells were incubated on 15 mm diameter round coverslips and fixed with 4% paraformaldehyde. Plasma membranes were permeabilized using Triton X-100. Blocking was performed with 5% bovine serum albumin (BSA). Primary antibodies, diluted as listed in Supplementary Table 3, were applied overnight at 4 °C. Fluorescence-labeled secondary antibodies were then incubated for 1 h at room temperature in a light-protected environment. Finally, coverslips were mounted with an anti-fading agent, and fluorescence images were acquired using a laser confocal microscope (Olympus).

FITC-labeled dextran uptake assay

Phagocytic activity of macrophages was examined using FITC-labeled dextran (MW 4,000, Beyotime, Cat# ST2930). THP-1 cells were differentiated into M0 macrophages by treating with 100 ng/mL PMA (Beyotime, Cat# S1819) for 24 h. FITC-labeled dextran was dissolved in Hank’s balanced salt solution (HBSS) at a stock concentration of 100 mg/mL. Macrophages were incubated with serum-free RPMI 1640 medium containing 1 mg/mL FITC-labeled dextran at 37 °C with 5% CO2 for 2 h. Cells were then washed three times with PBS and resuspended in PBS. Fluorescence images were captured, and flow cytometry was used to analyze the mean fluorescence intensity in the FITC channel, quantifying the dextran uptake by macrophages [25].

Statistical analysis

GraphPad Prism9 was applied for the presentation of figures. ANOVA was used for multiple sample comparison and two-sided Dunnett's t-test was applied for one-to-one comparison. The p value was recorded in the figures with exact numbers.

Results

IDO1 upregulation is associated with increased immune infiltration in colorectal cancer

We analyzed the role of IDO1 in colorectal cancer using the TCGA database, including its correlation with immune checkpoints, expression in cancer vs. normal tissues, mutation levels, and impact on tumor mutation burden and immune cell infiltration. IDO1 showed positive correlations with PDCD1 (PD-1), CD274 (PD-L1), CTLA4, and other immune checkpoints like TIGIT, TGFB1, CXCL9, CXCL10, and IFNG (Fig. 1A, Figure S2A). IDO1 expression was higher in cancer tissues compared to normal epithelium in a wide range of adenocarcinomas, including colon cancer and rectal cancer (Fig. 1B). The mutation rate of IDO1 was low, at 1.8% in COAD and 1.1% in READ (Fig. 1C). Pan-cancer analysis revealed a strong correlation between IDO1 expression and tumor mutation burden (Fig. 1D). Using ESTIMATE and TIMER algorithms, we found positive correlations between IDO1 expression and immune infiltration scores in CRC (Fig. 1E). Increased IDO1 expression in CRC was associated with heightened interferon responses and activation of the HALLMARK IL-6, JAK-STAT signaling pathway, and inflammatory response (Fig. 1F). CRC tissue samples showed concurrent elevation of IDO1 and immune cell markers (CD4, CD8, CD68, CD206) (Fig. 1G). Upregulation of IDO1 in colorectal cancer cell lines and paired cancer-normal tissue samples was confirmed (Fig. 1H, I). Higher IDO1 expression was also linked to a greater overall mutation frequency (Figure S2B). We also explore the correlation between IDO1 expression and survival of patients, no statistically significant finding was explored (Figure S2C and S2D).

Fig. 1.

IDO1 upregulation is associated with increased immune infiltration in colorectal cancer. A Scatterplots illustrating the correlation between the immune checkpoint genes PDCD1 (encoding PD-1), CD274 (encoding PD-L1), and CTLA4, and the IDO1 gene within the TCGA-COAD transcriptomic datasets. B Expression of IDO1 in tumor versus adjacent normal tissues in TCGA adenocarcinoma datasets, highlighted COAD and READ. C The mutation sites, types, and frequencies of the IDO1 gene in TCGA-COAD and TCGA-READ. D Correlation of IDO1 expression and tumor mutation burden (TMB) in TCGA pan-cancer database. E Correlation between immune infiltration and IDO1 in TCGA-COAD and TCGA-READ. Overall immune infiltration status was predicted by ESTIMATE, the infiltration of CD8 + T cell, CD4 + T cell and Macrophages were predicted by IPS. F Gene set enrichment analysis (GSEA) of TCGA-COAD and TCGA-READ, presenting upregulation of inflammatory associated pathways in IDO1^high cancer datasets. G Immunohistochemistry staining of tumor infiltrating leukocytes: CD4, CD8a, CD68 and CD206 in tumor versus normal epithelium. H Expression of IDO1 in clinically collected colon cancer and adjacent normal tissue (n = 4). I Expression of IDO1 expression in multiple human derived colorectal cancer cell lines versus colon epithelial cell (CCD841con)

IDO1 inhibition sensitized PD-1 blockade therapy in MSS colorectal cancer

We evaluated whether IDO1 inhibition enhances the effectiveness of PD-1 blockade in colorectal cancer using an animal model (Fig. 2A). Two mouse derived cell lines were used: CT26 (MSS, Balb/c mice) initially resistant to PD-1 blockade, and MC38 (MSI-H, C57 mice) (Fig. 2A) [26, 27]. Tumor cells were inoculated superficially in mice, followed by treatments with PD-1 blockade, IDO1 inhibition (epacadostat), or their combination. Results showed that epacadostat was effective in CT26, especially when combined with PD-1 blockade (Fig. 2B, C). IDO1 expression and macrophage markers (F4/80, CD206, IL-6) were examined, revealing IDO1 upregulation with PD-1 treatment. CD206 increased in the PD-1 group, while IL-6 was elevated with epacadostat, with or without PD-1 blockade (Fig. 2D). Leukocyte infiltration analysis indicated increased CD8 + T cells and a higher pro-inflammatory/anti-inflammatory ratio with epacadostat (Fig. 2E, F). Epacadostat also showed combined antitumor effects with PD-1 blockade in MC38 cell lines (Figure S3A-C).

Fig. 2.

IDO1 inhibition sensitized ICI therapy in MSS colorectal cancer. A Schema of in vivo analysis of the effect to combined therapy of PD-1 blockage and IDO1 inhibition in superficial inoculated MSS or MSI-H colon cancer. B Snapshot of tumor gross morphologies at day 16 post-inoculation. C Tumor size statistics of CT26 and the increase of tumor volume with time post-inoculation. (n = 5, data was presented as mean ± SEM). D Immunohistochemistry staining of IDO1, the mouse monocyte/macrophage biomarker F4/80, the anti-inflammatory marker CD206, and the pro-inflammatory marker IL-6 in different treatment groups. E Scatterplots of flow cytometry analysis of CT26 tumor receiving different treatments. F Distribution of different tumor infiltrated leukocytes (total leukocytes, total T cells, CTLs, helper Ts, Tregs, Macrophages and pro-inflammatory/anti-inflammatory macrophage in CT26 tumor-bearing mice receiving different treatments. (n = 5, data was presented as mean ± SEM)

Knockdown of IDO1 impairs resistance to interferon-γ in CRC cell lines via AHR translocation

We used RNAi to knockdown IDO1 in MSS cell lines CT26 and HT29 to examine its impact on cell growth and invasion (Fig. 3A). No significant changes were observed in cell duplication, migration, or invasiveness after IDO1 suppression (Fig. 3 B-F). As IDO1 is IFN-inducible, we treated HT29 cells with 50 ng/mL human recombinant interferon gamma (rhIFN-γ, R&D, Cat# 285-IF-100) for 6 h to assess whether IDO1 inhibition affects tumor resistance to IFN stimulation. IFN-γ significantly increased apoptosis in IDO1 knockdown HT29 cells (Fig. 3G). We investigated the impact of IDO1 knockdown on AHR activation by performing nuclear-cytoplasmic protein separation on IFN-γ treated cancer cells, finding a significant reduction in AHR nuclear translocation in IDO1 knockdown cells (Fig. 3H). The reduced AhR nuclear translocation could be restored by the application of kynurenine (100 μM) (Fig. 3H). Kynurenine could also restore the tolerance of IFN-γ in CRC cells by reducing IFN-γ induced apoptosis (Fig. 3I). Additionally, IDO1 knockdown resulted in decreased overall AHR expression and increased levels of cleaved caspase 3 and PD-L1, which could be restored by the application of kynurenine (Fig. 3J).

Fig. 3.

IDO1 knockdown suppresses tumor resistance to IFN-γ. A Western blot showing the interfering efficiency in knockdown of IDO1 in CT26 with siRNA and HT29 using lentiviral infection. B & C Wound healing assay of CT26 and HT29 after knockdown of IDO1 expression in vitro. (n = 3, data was presented as mean ± SEM). D & E Cell invasiveness assay with 8 μm transwell in IDO1 knockdown cell lines, with statistical analysis. (n = 3, data was presented as mean ± SEM). F In vitro proliferation assay by CCK-8 in CT29 and HT29 with modified expression of IDO1. (n = 6, data was presented as mean ± SEM). G Apoptotic and dead cells detected by Annexin V/PI stain in HT29 after IFN-γ stimulation (50 ng/mL) for 12 h. H Nuclear and cytoplasmic protein presenting the decrease of AHR nuclear translocation in IDO1-knockdown HT29 after IFN-γ treatment, which could be restored by kynurenine (100 μM). I Expression of caspase 3, cleaved caspase 3, PD-L1 in rhIFN-γ treated HT29 cells, with or without kynurenine(100 μM). J Apoptotic and dead cells detected by Annexin V/PI stain in HT29 cells after co-treatment of IFN-γ (50 ng/mL) and kynurenine(100 μM)

Knockdown of IDO1 in MSS CRC improves sensitivity to PD-1 blockade therapy in vivo

We used IDO1 knockdown CT26 cells to assess the effectiveness of PD-1 blockade in vivo (Fig. 4A). To achieve long-term IDO1 inhibition at the transcription level, we constructed shIDO1 plasmids and generated stable knockdown CT26 cells via lentivirus infection (Fig. 4B). The positive fluorescence signal of GFP representing successful infection with lentivirus (Fig. 4C). As expected, IDO1 knockdown suppressed tumor growth under PD-1 blockade therapy (Fig. 4D & E). We also assessed macrophage marker expression, finding elevated pro-inflammatory IL-6 and reduced anti-inflammatory CD206 infiltration in the sh2 CT26 tumor, especially when treated with αPD-1 (Fig. 4F). Flow cytometry revealed increased leukocyte infiltration, particularly T cells, with a notable decrease in regulatory T cells (Fig. 4G, H). Additionally, IDO1 knockdown in MC38 cells enhanced response to PD-1 blockade therapy (Figure S4A-C).

Fig. 4.

IDO1 knockdown in tumor cell promote immunotherapy sensitivity in MSS CRC. A Schema of in vivo analysis of the effect on knocking down IDO1 in tumor cell combined with PD-1 blockage in superficial inoculated MSS colon cancer; B Expression of IDO1 in lentivirus infected CT26 cell lines presented by Western Blot or qRT-CPR; C FITC channel presenting the GFP expression of shIDO1 or shNC cells. D Snapshot of tumor gross morphologies at day 17 post-inoculation; E Tumor size statistics of CT26 and the increase of tumor volume with time post-inoculation; F Immunohistochemistry staining of IDO1, the mouse monocyte/macrophage biomarker F4/80, the anti-inflammatory marker CD206, and the pro-inflammatory marker IL-6 in different treatment groups. G Scatterplots of flow cytometry analysis of Ido1-sh2 and Ido1-shNC mice receiving different treatments. H Distribution of different tumor infiltrated leukocytes (total leukocytes, total T cells, CTLs, Th Ts, Tregs, Macrophages and pro-inflammatory/anti-inflammatory macrophage in CT26 tumor-bearing mice receiving different treatments. (n = 5, data was presented as mean ± SEM)

IDO1 KD in cancer cell promotes macrophage pro-inflammatory phenotype polarization and activates the JAK2-STAT3-IL6 pathway

We used an in vivo coculture method to examine interactions between cancer cells and mouse derived macrophages (RAW264.7). When cocultured with CT26 cells, macrophages exhibited high CD163 and low IL-6 levels in the presence of shNC tumor cells. Knocking down IDO1 in tumor cells increased IL-6 and decreased CD163 in macrophages (Fig. 5A). We used FITC-labeled dextran to assess the phagocytic activity in THP-1 cells. The dextran uptake was higher in THP-1 when stimulated with rhIFN-γ loaded shIDO1 HT29 cell supernatant (Fig. 5B, C). Flow cytometry confirmed higher IL-6 expression in THP-1 cells stimulated by shIDO1 cell supernatant (Fig. 5D). JAK2 and STAT3 mRNA expression increased in THP-1 cells stimulated by IDO1-knockdown HT29 cells (Fig. 5E). Protein analysis showed increased JAK2 and STAT3 phosphorylation in IDO1 knockdown HT29 cells (Fig. 5F).

Fig. 5.

IDO1 knockdown in tumor cell potentiate pro-inflammatory macrophage polarization with elevated phagocytic activity. A Co-culture of RAW264.7 cells with CT26-shIDO1 cells shows increased IL-6 expression and decreased CD163 expression in RAW264.7 cells, treated by rmIFN-γ. B FITC channel of THP-1 stimulated by rhIFN-γ treated culture medium of HT29-shNC and HT29-shIDO1. C Fluorescence intensity distribution and mean fluorescence intensity (MFI) of FITC-dextran in THP-1, stimulated by IFN-γ treated culture medium of HT29-shNC and HT29-shIDO1. (n = 3, data was presented as mean ± SEM). D Expression of IL-6 in THP-1 after stimulated by IFN-γ treated culture medium of HT29-shNC and HT29-shIDO1 with MFI intensity barplot (n = 3, data was presented as mean ± SEM). E mRNA abundance of JAK2 and STAT3 in THP-1 after stimulated by IFN-γ treated culture medium of HT29-shNC and HT29-shIDO1. (n = ,4 data was presented as mean ± SEM). F Expression of JAK2, STAT3, IL6 in THP-1 stimulated by IFN-γ treated culture medium of HT29-shNC and HT29-shIDO1. G Experiment pipeline of macrophage depletion assay, abbreviations: L: clodronate liposome/PBS liposome; α: αPD-1; Epac: epacadostat. H Snapshot of tumor in macrophage depletion assay, each tumor was inoculated subcutaneously and was treated according to the experiment protocol (n = 5). I & J Tumor weight and volume of different treatment groups. K IHC staining of F4/80 presenting the satisfying depletion of macrophage and the associated infiltration of CD8 positive lymphocytes

We performed an in vivo assay to investigate whether macrophages play a key role in sensitizing immunotherapy mediated by IDO1 blockade by depleting macrophage using clodronate liposome (Fig. 5G). The result showed that the combined effect of epacadostat and αPD-1 was abolished in a macrophage-depleted tumor microenvironment (Fig. 5H to 5J). IHC analysis further revealed that macrophage depletion led to reduced infiltration of T cells, and IDO1 inhibition failed to elevate the overall T cell infiltration in tumor (Fig. 5K).

Discussion

Despite the overall effectiveness of immunotherapy in various type of cancer, it often fails in MSS CRCs due to the suppressive tumor immune microenvironment (TIME). The infiltration of suppressive immune cells such as regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs), along with reduced levels of stimulatory cytokines like CXCL9 and CXCL10 often contribute to an immunosuppressive milieu that hinders antigen recognition and impedes the generation of cancer-targeting cytotoxic T cells, thereby diminishing the efficacy of immune-regulated treatments [28].

IDO1 is a known suppressive immune checkpoint. IDO1 depletes tryptophan and produces kynurenine, which subsequently activates AhR. This activity creates a local immunosuppressive microenvironment conducive to tumor escaping from immune surveillance. The upregulation of IDO1 has been associated with the suppression of T cell responses and the promotion of Treg development, further contributing to immune evasion by tumors [29, 30].

Inhibiting IDO1 activity is hypothesized to restore local and systemic immune responses against tumor cells by reversing tryptophan depletion, reducing kynurenine levels, and promoting a more immunogenic tumor microenvironment. Clinical trials have investigated IDO1 inhibitors, both as monotherapies and in combination with other immunotherapeutic agents, such as PD-1/PD-L1 inhibitors, to enhance antitumor immunity. While results from these trials have been mixed, notably, none have focused on the potential of IDO1 inhibitor in MSS CRCs. Therefore, the synergic effect of IDO1 inhibitors and PD-1 blockade required more evidence.

In this study, we discovered that IDO1 is upregulated in tumor tissues compared to normal tissues. Although IDO1 is not a frequently mutated gene, its expression was closely correlated with the overall mutation burden and the pro-inflammatory status of the TIME. Collectively, these results suggested that IDO1 plays a pivotal role in modulating the tumor microenvironment in CRCs, playing an internal agent against inflammation, which is hijacked by tumor tissue to defend against immune surveillance.

To evaluate whether IDO1 inhibition could enhance PD-1 blockade immunotherapy, we conducted an in vivo assay. The results were highly encouraging, demonstrating a synergistic effect of IDO1 inhibitor epacadostat and PD-1 blockade in MSS CRCs. Notably, epacadostat significantly regulated the distribution of immune cells within the tumor microenvironment leading to a marked increase in the infiltration of cytotoxic T cells and pro-inflammatory macrophages, which are known to play key roles in antitumor immunity.

Our studies also uncovered a significant role for IDO1 in mediating cancer cell resistance to interferon-gamma (IFN-γ). The abnormal upregulation of IDO1 suggested a complex interplay, where cancer cells utilize IDO1 to counteract the immune-activating effects of IFN-γ. This mechanism not only illustrates IDO1's role in creating an immunosuppressive environment but also underscores its potential as a therapeutic target to enhance the efficacy of cancer immunotherapy. While IDO1 expression may not directly influence the proliferative or metastatic capabilities of cancer cells in vitro, its role in mediating IFN-γ resistance positions its importance in immune resistance.

Furthermore, our study revealed that inhibition of IDO1 presents a promising strategy not only by impacting the metabolic functions of cancer cells but also by modulating the behavior of immune cells within the tumor microenvironment. While IDO1 knockdown did not significantly alter the growth rate of CT26 tumors in vivo, it notably enhanced the effectiveness of PD-1 blockade therapy. This suggests that the primary role of IDO1 in these tumors may not be in directly promoting tumor cell proliferation but in modulating the immune microenvironment to favor tumor survival. Further investigation into the immune profiles of these tumors showed a marked increase in the infiltration of cytotoxic CD8 + T cells in the IDO1-knockdown models treated with PD-1 inhibitors, suggesting that IDO1 activity in the tumor cells contributes to an immune-exclusion phenotype.

We also sought to uncover the mechanism by which IDO1 in cancer cells influences of macrophage polarization. We found that IDO1 plays a pivotal role in maintaining the immunosuppressive characteristics of tumor associated macrophages (TAMs). By silencing IDO1 expression in cancer cells, the contacted macrophages tended to adopt a pro-inflammatory phenotype, evidenced by the increase of IL-6 expression upon the coculture of TAMs and IDO1-knockdown cancer cells. Additionally, macrophages exhibited enhanced phagocytic activity when indirectly stimulated by IDO1-knockdown cancer cells, indicating activation of antigen presentation. Collectively, these results illustrated that, IDO1 not only confers resistance to inflammatory stimuli in cancer cells but also facilitates the secretion of specific agents that regulate macrophage polarization. And the major effect of IDO1 inhibition to macrophage polarization is through the activation of the IL6, JAK2-STAT3 pathway. In our study, it’s a shame we didn’t verify that the effect of IDO1 inhibitor compromise by precisely inhibiting JAK-STAT3 pathway in macrophage. But a previous study implied that inhibition of IDO1 with selective inhibitor B37 induces the activation of JAK-STAT3 pathway, and inducing the elevation of IL6 expression [31].

Collectively, our findings highlight the dual role of IDO1 in cancer progression— serving both as a shield against immune attack directly on cancer cells and as a regulator of the immune landscape via macrophage polarization. Our findings underscore the potential therapeutic benefits of targeting IDO1 in cancer treatment, not only to inhibit tumors’ intrinsic pathways but also to reprogram the immunological milieu of the tumor to enhance the efficacy of existing and emerging therapies. These insights pave the way for novel therapeutic strategies that aim at disrupting the IDO1-mediated immunosuppressive network within tumors, potentially leading to more effective immunotherapy outcomes.

However, this study also has certain limitations, such as not thoroughly exploring whether changes in the levels of downstream metabolites of IDO1 affect the local conditions of the tumor immune microenvironment, and whether sensitizing the effect of IDO1 blockade in MSS CRC is achieved by reshaping the changes in metabolite levels caused by IDO1. Additionally, the regulatory effect of tumor cells on stromal cells and the mediators inducing this regulation were not fully clarified in this study. In future research, the sensitizing mechanisms of IDO1 inhibitors on ICIs require further elucidation.

Conclusions

Our study introduces a novel approach to enhance the efficacy of PD-1 blockade therapy in MSS colorectal cancer. By combining an IDO1 inhibitor with PD-1 blockade, we demonstrated significant therapeutic benefits in a pre-clinical MSS cancer model. Specifically, inhibition of IDO1 in CRC cells led to a significant alteration in the distribution of tumor-infiltrating lymphocytes (TILs) and remodeled the immune microenvironment into a pro-inflammatory state. This transformation effectively "lit up" the tumors, rendering them more responsive to PD-1 blockade therapy. This strategy not only highlights the potential of IDO1 inhibitors in modifying the tumor microenvironment but also underscores their role in improving the outcomes of existing immunotherapies (Fig. 6).

Fig. 6.

Mechanism of IDO1 inhibitors in enhancing PD-1 blockade therapy in MSS CRCs. (Generated by Figdraw)

Supplementary Information

Below is the link to the electronic supplementary material.

Author contribution

LG, WX and WM contributed equally to this work by performing the in vitro and in vivo experiments. MW was responsible for constructing the IDO1 gene modification plasmid and primers. LR assisted with the study design and provided statistical analysis support. PZ supervised the colorectal surgery experimental platform. ZR and CG are the principal supervisors of the study, overseeing the research design, data analysis, and manuscript preparation. All authors contributed to the drafting and revision of the manuscript and approved the final version for publication.

Data availability

The single-cell RNA sequencing (scRNA-seq) data from human colorectal cancer (CRC) used in this study were generated by Chen Y, Wang D, Li Y, Qi L, et al. and are publicly available in the Gene Expression Omnibus (GEO) database under accession number GSE236581 [24].

Declarations

Conflict of interest

The authors declare no competing interests.