2.1.1 Effect of Tryptophan Depletion and Its Metabolites on Tumor Immunity

The first pathway (effect of tryptophan depletion): Increased IDO expression facilitates the conversion of Trp to Kyn, depleting Trp in the tumor microenvironment, resulting in several effects: (1) Unbound tRNA accumulation activates the general control non-derepressible 2 (GCN2) kinase pathway, phosphorylating translation initiation factor eIF2α and inhibiting cap-dependent translation [72]. At the same time, eIF2α activates transcription factor 4 (ATF4), which promotes the expression of tryptophan transporter SLC7A5 [38, 39]. Since tumor cells and effector T cells exhibit different abilities to upregulate SLC1A5 in a tryptophan-depleted environment, tumor cells gain a competitive advantage in capturing Trp, helping to deplete Trp in the tumor microenvironment (TME) and suppressing effector T-cell activation, thus inhibiting their immune function [58]. (2) As a metabolic regulation pathway, mTOR can decide whether to engage in protein synthesis or autophagy depending on amino acid levels. In a Trp-deficient environment, mTOR activity is inhibited through the Glucokinase (GLK) pathway [35], leading to the conversion of CD4+Foxp3+Treg cells (iTreg) and suppression of effector T cells [36, 37]. Experiments suggest mTOR mediates ICOS expression on T cells, which interacts with B7h to promote immune infiltration of tumors. Reduced ICOS expression results in incomplete T-cell activation and localized immune suppression [40]. Similarly, PKC activity is affected, with low Trp concentrations inhibiting its function, thereby weakening its ability to activate effector T cells [35]. (3) Tryptophan depletion also increases AhR expression, a mechanism that is independent of GCN2, and further enhances the activation of AHR by kynurenine [41]. Recent molecular docking experiments by Badawy revealed a strong binding between Trp and AhR. Combined with previous reports of Trp's downregulation and inactivation of AhR [42], as well as Trp's antagonistic effect on AhR activation, these findings [41, 43] suggest that Trp is likely an antagonist of AhR, and a reduction in Trp could lead to overactivation of AhR.

Notably, Badawy's molecular docking experiments revealed that Kyn did not bind to AhR, but rather activated AhR through its metabolite KYNA [42]. This result is supported by two factors: first, most previous experiments showing Kyn activation of AhR were conducted under conditions where Kyn concentrations far exceeded physiological levels (this will be discussed in detail below); second, prior studies [45] indirectly demonstrated Kyn's activation of AhR through the induction of AhR target genes (such as cytochrome P450-dependent drug metabolism), without controlling for Kyn's potential metabolic transformation, where its downstream products like KYNA, AA, and 3-HAA may also be involved. Although Kyn is not a suitable AhR ligand from a chemical structure perspective, Seok et al. found that Kyn in DMSO could spontaneously convert into a higher-affinity aromatic condensation product [77]. Solvay's experiments also detected the same condensation derivatives in older Kyn solutions [41]. Therefore, it is possible that Kyn may activate AhR in a different form. However, this hypothesis requires further verification through cell and animal experiments.

The discovery of Trp as an AhR antagonist highlights its key role in maintaining bodily homeostasis and redefines the significance of Trp depletion—suggesting that this depletion may lead to a loss of protection against excessive AhR activation. This new concept not only overturns traditional understanding but also opens new pathways for future biological and clinical research. Future research should further explore the roles of Trp and Kyn in different cellular systems, particularly in the context of key enzyme expression or deficiency, to provide a solid scientific foundation for the development of novel therapeutic strategies.

2.1.2 Non-Enzymatic Functions of IDO1

In terms of tumor immune suppression, past studies have focused greatly on the enzymatic function of IDO1, but its non-enzymatic function should not be overlooked. Mouse IDO harbors two Immunoreceptor Tyrosine-based Inhibition Motifs (ITIMs), known as ITIM1 and ITIM2, which are highly conserved across mammals and positioned distantly from the catalytic domain [78].

Plott's previous experiments demonstrated that under the intervention of the immunosuppressive cytokine transforming growth factor β (TGF-β), plasmacytoid DCs (pDCs) can phosphorylate IDO's ITIM1 via the Frk kinase [55], then recruit SHP-1 and SHP-2 (upregulated by TGF-β) to induce downstream pathways. SHP-1 inhibits the kinase IRAK1, thus promoting noncanonical NF-κB pathways, enabling pDCs to induce a higher frequency of Foxp3+ cells in CD4+ populations, leading to an immunosuppressive effect [52, 53]. Furthermore, the TGF-β-IDO-SHP axis boosts the expression of TGF-β and type I interferons, the latter of which binds to IDO1 promoters via IFN-stimulated elements (ISREs) to regulate gene transcription, ultimately increasing IDO1 expression [54]. This upregulation creates a positive feedback loop to sustain the long-term immunosuppressive phenotype of pDCs. These effects cannot be blocked by the inhibitor 1-methyltryptophan (1-MT) (Figure 4).

Curiously, ITIM2 has an opposite function [55]. When cytokines like IL-6 are present, the SH2 binding domain of upregulated Suppressor of Cytokine Signaling 3 (SOCS3) binds to the phosphorylated site of IDO1's ITIM2, recruiting the E3 ubiquitin ligase complex, leading to IDO1 degradation via ubiquitination, and thereby activating immunity [56]. Although the phosphorylation site is far from the catalytic domain, Albini's experiment simulated the tyrosine phosphorylation state by replacing tyrosine residues with glutamic acid, revealing that IDO1 lost its enzymatic activity [55]. This indicates that IDO1's enzymatic and non-enzymatic activities may not co-occur, and phosphorylated IDO1 likely undergoes a conformational change at the catalytic site.

Recent studies have found that, in addition to ITIMs, IDO1 contains a binding site for the regulatory subunit p85 of PI3K, known as YENM, which also contains a tyrosine residue [79]. When this residue is phosphorylated, it binds to p85, recruiting the catalytic subunit p110 to regulate immune and anti-inflammatory responses. Additionally, PI3K affects the transport of IDO1 proteins [80], anchoring them to early endosomes (EEs), which are rich in various effector proteins and receptors that facilitate signal transduction, including TGF-β [81]. This promotes the TGF-β-IDO-SHP signaling pathway. Epacadostat, a potent IDO1 catalytic inhibitor, demonstrated good efficacy in preclinical and early clinical trials [82] but did not show extra benefit when combined with pembrolizumab in Phase III randomized trials for metastatic melanoma [83]. One explanation proposed is that Epacadostat strengthens the interaction between IDO1 and SHP family tyrosine phosphatases while also facilitating IDO1's intracellular trafficking. Through binding with PI3K, IDO1 is anchored to EEs [84], which supports its immunosuppressive function—simultaneously inhibiting its catalytic activity and enhancing immune tolerance signaling.

The conflicting roles of ITIM1 and ITIM2 in IDO1 make sense in the context of responding to inflammatory signals, as pDCs, positioned at the forefront of immune responses, must quickly react to fluctuating pro-inflammatory and anti-inflammatory cues. In the tumor microenvironment, however, our understanding of how phosphorylated IDO1 influences immune signaling pathways is limited, and the response of other IDO1-expressing cells, apart from pDCs, to their microenvironment remains to be studied. Previous research [33] has indicated that IL-6 upregulates and sustains IDO1 expression in tumor cells, implying that IDO1 may act differently in response to identical signaling stimuli across different cell types. Considering both the catalytic and non-catalytic roles of IDO1, the development of IDO1 inhibitors as anticancer therapies should account for both the inhibition of IDO1's catalytic activity and its impact on IDO1 signaling to enhance therapeutic efficacy.

2.1.3 The Role of IDO1 Isoenzymes and Other Enzymes Associated With the Kynurenine Pathway on Tumor Immunity

TDO2 is expressed in various cancers [85], and Perez-Castro's analysis of tumor mRNA in TCGA revealed a downregulation of IDO1 along with an upregulation of TDO in lung cancer [58]. Both TDO2 and IDO1 play key roles in the metabolism of Trp, and therefore, it is hypothesized that they play a role in tumor immunity. After IDO1 silencing, TDO2 may function as a compensatory mechanism to sustain tumor growth. Experiments [45, 85] Reversal of tumoral immune resistance by inhibition of tryptophan 2,3-dioxygenase in vivo and in vitro have demonstrated that TDO2 influences T-lymphocyte function by inhibiting their proliferation. TDO2 promotes M2 macrophage activation by upregulating IL-8, triggering the AKT/GSK3β pathway, thereby supporting the growth of esophageal squamous cell carcinoma [86]. TDO2 functions similarly to IDO1, as the metabolism of Trp can reach levels sufficient to activate AhR [87]. Moreover, TDO-positive cells suppress IFN-γ production through TDO2-mediated antibacterial mechanisms [88], potentially linking this effect to immune suppression in the tumor environment.

IDO2, the isoenzyme of IDO1, is overexpressed in non-small cell lung cancer, pancreatic cancer, and cervical cancer [89]. However, because of its low affinity for tryptophan [90], its IDO1-like effects are less noticeable, and its role in immune suppression remains controversial. Research has shown that IDO2 induces the activation of autoreactive T and B cells, facilitating autoantibody production, indicating its pro-inflammatory role. Further studies [91] have revealed that when IDO1 and IDO2 are co-expressed, competition for heme binding occurs, thereby negatively regulating IDO1. Whether IDO2 functions similarly in a tumor context requires further investigation.

Downstream metabolic enzymes of Kyn are also crucial for the function of the KP, as Kyn's metabolic products play corresponding roles in tumor immunity. On the other hand, Kyn metabolism implies a reduction in its concentration, which will inevitably impact its associated functions. The three main downstream metabolic enzymes of Kyn mentioned above are KMO, KUNY, and KAT.

KMO is largely downregulated in most tumors, but it is upregulated in inflammatory environments under stimulation from factors like IFN-γ [92, 93]. This indicates that in immune-stimulated hot tumors, KMO upregulation causes increased 3-HK expression. As previously mentioned, 3-HK suppresses effector T cells. The upregulation of IDO1 in an inflammatory environment also compensates for the Kyn consumed by elevated KMO.

KAT shows different responses to the same stimuli across tumor types; for instance, IFN-γ induces KAT expression in head and neck squamous cell carcinomas, but it decreases in glioblastoma multiforme [94]. Given KAT's higher Km, its impact on the Kyn:Trp ratio is less pronounced than that of the other two enzymes [95]. However, when Kyn accumulates to a certain level in the microenvironment, this enzyme's catalytic reaction can still proceed [94]. As a metabolic product of KAT, KYNA has multifaceted effects on tumors, promoting antitumor immunity while also inhibiting tumor cell cycle progression and invasiveness. In breast cancer [96, 97], KYNA binds to AhR to promote IL-6 expression (enhanced binding of AhR with RelB leads to increased IL-6 transcription). IL-6 regulates immunity in a bidirectional manner. Fisher's review summarized IL-6's antitumor immune effect, which is achieved by influencing lymph node activation and metastasis, as well as immune infiltration in the tumor microenvironment [98]. On the other hand, IL-6 partially relies on autocrine IL-10 mechanisms and inhibits NF-κB activity and alters NF-κB subunit composition, thereby synergistically inhibiting CCR7 expression in DCs, which affects their migration ability and prevents effective activation of Teff [99]. In breast tumor cells, co-treatment with KA and IL-1β significantly induces IL-6 expression, and this induction depends on the expression of AhR [97]. In lung cancer [33], IL-6 activates STAT3 to induce IDO1, which in turn produces Kyn and KYNA to activate AhR and induce IL-6 expression. This establishes an AhR-IL6-STAT3 loop, which maintains the continuous activation of the KP pathway and suppresses antitumor immunity. KYNA, acting as an AhR ligand, has been shown to have a higher affinity and stability than Kyn [97]. Badawy recently suggested through molecular docking experiments that Kyn might activate AhR via KYNA [42]. Regardless of the situation, it is certain that KYNA can promote tumor immune evasion via AhR [100]. Additionally, KYNA may inhibit tumor cell proliferation upon receptor activation and influence tumor metastasis. This will be further discussed below.

KYUN is elevated in certain autoimmune diseases in humans, and in the process of Kyn degradation, it reduces Foxp3 expression, thereby enhancing local immune responses [101]. Unlike the immune suppression observed in tumor microenvironments, IDO1 upregulation in the pro-inflammatory environment of these inflammatory diseases does not significantly reduce inflammation, possibly due to KYUN expression. Harden's analysis of cancers found that IDO1 and TDO2 are preferentially upregulated over KYUN, whereas the opposite occurs in pro-inflammatory environments. Thus, compared to inflammatory diseases, cancers show a significantly higher IDO1-TDO2 ratio. Upregulating KYNU and degrading kynurenine to weaken kyn-AhR signaling may be a reliable way to control tumor immunosuppression. For instance, overexpressing KYNU in chimeric antigen receptor (CAR)-T cells reduces immunosuppressive Kyn in the tumor microenvironment, enhancing CAR-T cell efficacy [102]. In mouse models of melanoma, breast cancer, and colon cancer, PEGKYNase (modified with polyethylene glycol (PEG) to enhance enzyme stability and half-life) can reduce serum Kyn, increase AA levels, and simultaneously increase the number of CD8+ T cells infiltrating the tumor microenvironment, reducing tumor volume [103].

Interestingly, León-Letelier and colleagues discovered that higher levels of KYNU expression in lung adenocarcinoma (LUAD) and pancreatic cancer tissues are linked to an immunosuppressive tumor microenvironment [104, 105]. Upregulation of KYNU can increase the number of Tregs and the expression of immune checkpoint blockade genes PD-1 and PD-L1. These studies suggest that the immunosuppressive role of KYNU is tumor-specific [104].

Heng's research may provide an explanation for this contradiction [106]. KYNU breaks down Kyn and 3-HK into AA and 3-HAA. Studies [107-111] have shown that 3-HAA can suppress nitric oxide (NO) production in macrophages, thereby reducing their tumor-killing function, inducing apoptosis in cytotoxic T cells and Th1 cells, weakening immune-mediated tumor destruction, and enhancing tumor immunosuppression by modulating HLA-G expression on dendritic cells and promoting Treg cell differentiation. AA, however, has not been observed to impact immune cell activity or function in existing research. Heng found that KYNU's substrate affinity changes under inflammatory stimulation; for example, 3-HAA synthesis is significantly increased in breast cancer cells under IFN-γ stimulation. The production of 3-HAA also depends on environmental pH and iron ion availability, factors that may limit the conversion from AA to 3-HAA [106]. Thus, variations in the tumor microenvironment result in different substrate preferences for KYNU, thereby affecting tumor immunosuppression.

Badawy's review concluded that most tumors prefer to upregulate enzymes upstream of Kyn and downregulate downstream enzymes, leading to higher Kyn levels and enhanced immunosuppressive effects [7]. Analysis of key enzymes in the Kyn metabolic pathway revealed their roles in immunosuppression, with KYNU appearing to serve as an “on–off” switch for immune suppression. Tumors undoubtedly favor Kyn, but different tumor types, owing to unique disease patterns, affect tryptophan metabolic enzymes in varied ways. These differences could result in opposite roles for the same enzyme in distinct tumors, making it necessary to examine enzyme alterations and metabolite levels in different tumors.

2.1.4 Tryptophan Acquisition by Tumor Cells

Tumor cells' requirement for essential amino acids causes upregulation of essential amino acid transporters, a characteristic shared by many cancers. Large neutral amino acid transporters (LAT), as Trp transporters, are also upregulated in most tumors. Studies have identified SLC7A5 (LAT1), SLC1A5, SLC7A11, and SLC6A14 as Trp transporters [7, 112]. Perez-Castro et al. [58] found that SLC7A5 and SLC1A5 are upregulated in 13 types of tumors, while SLC6A14, distinct from the other two transporters, is ideal for tumor growth and development due to its strong concentration capacity and broad-spectrum amino acid transport ability [113]. Its expression is upregulated in multiple epithelial-derived cancers [114, 115]. Additionally, another Trp transporter, distinct from the L-type system and expressed by IDO-positive tumor cells, significantly increases Trp uptake into the tumor under low Trp concentrations [116]. Thus, tumor cells hold an advantage in Trp acquisition over other cells in the same microenvironment due to these transporters.

Research has also determined that Kyn is transported through the system L transporter in activated T cells, with SLC7A5 recognized as a crucial transporter [117, 118]. SLC7A5 not only transports Kyn but also large neutral amino acids like leucine. Therefore, Kyn competes with essential amino acids for SLC7A5, influencing leucine availability, which in turn affects mTORC1 pathway activation and c-Myc expression. Myc, a critical metabolic reprogramming factor [119], coordinates metabolic gene transcription during T-cell activation to support growth and proliferation, also inducing SLC7A5 and SLC1A5 in multiple cell lines [120, 121]. Consequently, this competition ultimately inhibits T-cell activation, proliferation, and metabolism, producing an immunosuppressive effect.

As previously mentioned, Trp depletion and its decomposition products have synergistic effects. For instance, tumors expressing IDO1 have an advantage in acquiring Trp. After IDO1 consumes Trp, the subsequent reduction of Trp and Kyn production in the microenvironment enhances Trp transporter activity, further depleting Trp in immune cells and inducing apoptosis. Despite the upregulation of SLC7A5 in effector T cells under inflammatory conditions stimulated by the inflammatory factor IFN-γ, the large amount of Kyn produced by tumors competes for SLC7A5 binding (partially explaining the reduced Trp acquisition in effector T cells within the tumor microenvironment), affecting effector T-cell growth and proliferation [118]. Trp depletion increases the sensitivity of AhR to Trp metabolites like Kyn, potentially linked to upregulation of AhR and the LAT1 transporter for Trp and Kyn under Trp deprivation [41]. However, other studies suggest that under IDO1 overexpression, Trp levels do not reach the GCN2 activation threshold, with the primary effect arising from IDO1 metabolites [122]. Sinclair et al. observed that plasma kynurenine (Kyn) concentrations seldom surpass 3 μM, while concentrations of other system L substrates in plasma (such as leucine, tryptophan, and methionine) range from 20 to 100 μM, with Km values roughly one-tenth of Kyn's [118, 123]. This suggests that Kyn would require concentrations significantly above physiological levels to effectively compete with other high-affinity substrates. Additionally, in vitro experiments demonstrating AhR activation used Kyn concentrations (50–100 μM) exceeding normal physiological plasma Kyn levels [45, 124, 125], suggesting it may be difficult for Kyn to act in vivo; however, the differences between tumor microenvironments and normal plasma composition may make it possible. For instance, Ho et al.'s study showed that glucose concentrations measured in tumors were 10 times lower than in matched spleen or blood samples [126]. Earlier, we mentioned efficient Trp transporters that promote Trp accumulation in tumor cells, establishing the foundation for Trp depletion in tumors. Opitz et al. measured a Kyn concentration of 37 μM in a U87 xenograft tumor model [45]. If essential amino acids drop tenfold similar to glucose, IDO1 in the tumor environment could elevate Kyn tenfold, reaching the competitive level for SLC7A5 and effectively activating AhR. In the previous text, we mentioned that Badawy's molecular docking experiment found that Kyn did not bind to AhR, and related studies suggest that Kyn activates AhR indirectly through KA [42]. KAT-mediated metabolism of Kyn is more dependent on Kyn concentration. Previous research observed that when healthy individuals ingested small amounts of tryptophan (1.15 g, approximately 16 mg/kg), their plasma Kyn levels rose by 47% [127]. This change led to a significant 209% increase in KA concentration. This indicates that the change in KA concentration is amplified based on the change in Kyn concentration. In the tumor microenvironment, an increase in kyn concentration may lead to a more substantial rise in KYNA, which has a higher affinity and stability than Kyn, and its resulting AhR activation effect should not be overlooked. Thus, more detailed quantitative metabolite analysis in the tumor microenvironment may clarify these mechanisms in future research (Figure 4).

2.4.1 IDO1 Serves as a Crucial Interface Between IFN-γ and IL6 in Tumor Angiogenesis, Steering the Inflammatory Milieu Toward Promoting Vascular Development

Saga's experiment [157] found that IDO1 promotes angiogenesis and enhances peritoneal dissemination of ovarian cancer. Smith's study [66] discovered that silencing IDO1 in a KRAS-induced lung adenocarcinoma model reduced the number of blood vessels in both tumors and non-tumor tissues. IL6 expression also significantly decreased with IDO1 silencing, but angiogenesis was restored after IL-6 supplementation. Their team confirmed this finding in follow-up studies [68]. In the OIR model, IL-6 silencing reduced angiogenesis, demonstrating that IDO1 is a crucial regulatory node between IFN-γ and IL6, promoting IL6 production through the GCN2-mediated Integrated Stress Response (ISR) pathway [67]. This balances IFN-γ's anti-angiogenic effects, supporting angiogenesis under inflammatory conditions. The direct pro-angiogenic effect of IL-6 cannot be overlooked. IL6 promotes angiogenesis via VEGF-independent mechanisms [158] and is associated with matrix metalloproteinases (MMPs), which facilitate endothelial cell sprouting by degrading the vascular extracellular matrix [159]. In patient-derived nasopharyngeal carcinoma, IL-6/NOS2 signaling regulates MMP-9 and MMP-2 dependent metastasis [160].

The specifics of IDO1-induced angiogenesis still require further study. Blaschitz et al. have identified that endothelial cells express IDO1, which may affect blood vessels through downstream tryptophan metabolites or local tryptophan depletion [161]. The tryptophan depletion effect has been confirmed in Dey's experiment, but their study did not indicate whether tryptophan metabolites affect angiogenesis [67]. Previous studies have found that endothelial cells expressing IDO1 in inflammation relax blood vessels through kynurenine. Zhang's recent experiment demonstrated that in a neonatal mouse cardiac deficiency model, Kyn produced by cardiomyocytes was transported to endothelial cells, promoting aryl hydrocarbon receptor translocation, which enhanced the expression of angiogenic genes like VEGF-A [162]. Previous studies have shown that AhR regulates angiogenesis through VEGF activation in endothelial cells and TGF-β inactivation in the stroma [45]. This provides a reference for studying IDO1-related angiogenesis in the tumor context, and exploring its underlying mechanisms could offer a basis for new drug development or more rational and effective combination therapies. IDO1's pro-angiogenic activity in both tumor and non-tumor tissues helps create a favorable microenvironment for tumor initiation, later promoting tumor growth and metastasis. Further research may examine whether differential IDO1 expression in normal tissues correlates with metastatic potential.

As an isoenzyme, IDO2 has no confirmed pro-angiogenic role, whereas TDO has been observed in the neovasculature of liver cancer, glioblastoma, melanoma, and bladder cancer [163]. It may work synergistically with IDO1, but the extent of its pro-angiogenic effect is yet to be clarified. Furthermore, studies [45, 164, 165] have shown that TDO expression increases under the influence of IL-1β, which upregulates IL-6 expression and promotes angiogenesis in endometrial stromal cells and glioma cells.

2.4.2 Overexpression of TDO and IDO1 Affects the Tumor Extracellular Matrix (ECM) and Enhances Epithelial-Mesenchymal Transition (EMT), Increasing Tumor Cell Motility, Promoting Distant Metastasis, and Resistance to Anoikis

The zinc finger transcription factor Slug is a target gene of AhR. Slug plays a role in epithelial–mesenchymal transition (EMT) by increasing the expression of vimentin and inhibiting the mRNA expression of E-cadherin [166]. TCDD (2,3,7,8-tetrachlorodibenzo-p-dioxin) is a widely recognized AhR agonist with high affinity for AhR [167]. TCDD affects Slug expression via AhR, thereby participating in the regulation of EMT in breast cancer cells. However, another AhR agonist, MEHP, competitively antagonizes TCDD, reducing its ability to promote tumor EMT [168]. Serum Response Factor (SRF), as a transcription factor, promotes epithelial cell migration and invasion in prostate cancer, gastric cancer, cervical cancer, and hepatocellular carcinoma and is associated with poor prognosis [169-171]. Xu's experiment revealed that SRF promotes EMT and increases tumor migration and invasion in OSCC cells by activating the IDO1 gene promoter and regulating the IDO1/Kyn-AhR signaling pathway [69]. This effect can be blocked by IDO1 and AhR inhibitors. Similarly, in bladder cancer, IDO1 expression is related to cell migration ability, and silencing IDO1 reduces EMT through the IL-6/STAT3/PD-L1 pathway [9]. MMPs, a family of zinc-dependent metalloproteinases, degrade ECM components, facilitating tumor cell migration to nearby tissues. In Lewis lung carcinoma cells [70], IDO1 overexpression upregulates MMP-2 and MMP-9 via the JAK2/STAT3 pathway. COL12A1 encodes a collagen protein that interacts with collagen types I and III to form homotrimers [51, 172, 173], activating the Collagen I/integrin β1 pathway [174]. COL12A1 is strongly associated with migration, invasion, and metastasis in breast and colorectal cancer. Xiang's study confirmed a positive regulatory loop between IDO1 and COL12A1, both promoting gastric cancer metastasis via the MAPK pathway [51, 175].

TDO2 serves as the main enzyme metabolizing tryptophan in cancer-associated fibroblasts (CAFs), with the produced kynurenine inducing EMT and cancer cell migration via the AKT/WNK1 axis [176]. In glioma, TDO2 is upregulated and plays a decisive role in tryptophan metabolism, and its expression is related to glioma cell motility [45]. Anoikis is a type of apoptosis caused by the loss of cell attachment to the extracellular matrix. In triple-negative breast cancer (TNBC) [156], TDO2 expression is upregulated and strengthens anoikis resistance and metastasis through the TDO2-AhR signaling axis. These results indicate the important role of TDO in tumor migration.

Interestingly, recent studies have found that KYNA inhibits cell migration in glioblastoma [135] and renal cell carcinoma [137], but the specific mechanisms are still under investigation. Previous studies have found that nicotine activates α7nAChR, leading to ERK pathway activation, promoting cell proliferation and neovascularization. Tu et al. also reported the key role of this receptor in gastric cancer progression and metastasis [177]. As an α7nAChR antagonist, Kyna may act through this pathway. We previously discussed the tumor specificity of KYUN in tumor immunity. In gastric cancer, GPx2 knockdown-induced ROS accumulation promotes KYNU expression, which subsequently degrades Kyn, thereby inhibiting AhR-mediated tumor progression and EMT [178].

Recently, a new tryptophan metabolism-related enzyme, interleukin-4-induced-1 (IL4I1), was discovered [8]. IL4I1 converts tryptophan into I3P, which can further metabolize to produce KynA and I3A. All three metabolites activate AhR, enhancing tumor cell motility and invasiveness. Additionally, IL4I1 is an AHR target gene, suggesting that it can amplify its own expression through positive feedback. Unlike IDO1 and TDO2, IL4I1 is a secretory protein [179]. Sadik et al. found that IL4I1 levels increased in the serum of CLL mice, indicating that IL4I1 could extend AhR activation effects from a local to a systemic level, better facilitating the creation of a supportive environment for tumor growth in metastatic sites [8].

3.1.1 Protecting the Intestinal Barrier

In the intestine, the Pregnane X Receptor (PXR) is a crucial regulator of innate immune responses and responds to xenobiotic injury [190]. Venkatesh's experiments showed that PXR in intestinal epithelial cells is vital for maintaining intestinal barrier integrity and regulating intestinal inflammation [191]. The loss of PXR leads to disrupted tight junctions between intestinal epithelial cells, increasing intestinal permeability. Tryptophan metabolites, such as indole-3-propionic acid (IPA) and indole acetic acid (IA), protect the intestinal barrier through PXR. In patients with ulcerative colitis, IPA levels are reduced, but dietary enhancement of IPA increases the expression of tight junction proteins, IL-22, and AhR, and helps restore the balance of the gut microbiome [192, 193]. In a DSS (dextran sulfate sodium)-induced colitis mouse model, IPA promotes Muc2 gene expression through PXR, enhancing mucin secretion [20]. Tight junction proteins reduce intestinal permeability, and IL-22 supports mucosal immunity and integrity by inducing epithelial cells to produce antimicrobial peptides and mucins [194]. AhR also plays an important role in gut homeostasis (as explained below). In cancer associated with colitis, Toll-like receptor 4 (TLR4) acts as a key regulator of intestinal barrier function and inflammation. By regulating the expression of genes related to intestinal barrier function, TLR4 increases intestinal permeability, worsening the inflammatory environment. IPA binds to PXR and inhibits TLR4, thus protecting the intestinal mucosa [195].

Activation of AhR in intestinal epithelial cells (IECs) is also crucial for maintaining the integrity and function of the intestinal barrier [196]. In Wlodarska's colitis mouse model, commensal bacteria, particularly Peptostreptococcus russellii (P. russellii), activate AhR through the metabolite indole acetic acid (IA), promoting Muc2 gene expression, enhancing Goblet cell differentiation and function, and supporting both gut barrier function and inflammation suppression [20]. In intestinal stem cells, RNF43 and ZNRF3 serve as negative feedback regulators of the Wnt signaling pathway, suppressing excessive stem cell proliferation by modulating the ubiquitination and degradation of Frizzled receptors [197]. When the intestinal barrier is damaged, the lack of AhR downregulates tumor suppressor genes Znrf3 and Rnf43 [197], resulting in abnormal activation of the Wnt-β-catenin signaling pathway, which disrupts epithelial cell differentiation from crypt stem cells and worsens gut barrier repair [196]. The indole derivative indole-3-carbinole (I3C), metabolized from glucobrassicin in cruciferous vegetables by gut microbiota [196], restores regulation of the Wnt-β-catenin pathway, controls epithelial cell differentiation, and reduces inflammation-induced tumor formation. Additionally, I3C promotes the differentiation of ILC3 cells, which produce IL-22, thereby maintaining mucosal immunity and integrity [198].

Besides IPA, IA, and I3C, indole-3-aldehyde and indole-3-acetic acid also affect intestinal barrier integrity and immune cells in mice by activating PXR and AhR [21, 191, 199].

3.1.2 The Indole Pathway Regulates Gut Inflammation

In colitis-associated colorectal cancer (CRC), AhR not only exerts anti-inflammatory protective effects by protecting the intestinal mucosa but also suppresses inflammation by promoting the secretion of anti-inflammatory cytokines [200]. Exposure to physiological concentrations of indole and its metabolites leads to increased expression of anti-inflammatory cytokines IL-10 and IL-22 [20, 201, 202]. IL-10 secretion further maintains mucin production by goblet cells, while IA activates the NRF2 (Nuclear Factor Erythroid 2-related Factor 2) antioxidant pathway, promoting antioxidant gene expression and reducing oxidative stress associated with intestinal inflammation [203]. In an inflammation-related colorectal cancer model, interruption of Trp-indole-AhR signaling resulted in significantly elevated mRNA levels of TNF-α, IL-1β, and IL-6 [185]. IPA reduced the expression of TNF-α, IFN-γ, and IL-1β through AhR, lowering the severity of intestinal inflammation in mice [192]. In rheumatoid arthritis [204], intervention to improve gut microbiota and elevate tryptophan metabolites (IA, IAA, IPA) activates AhR, reducing pro-inflammatory Th17 cells and increasing anti-inflammatory Treg cells, thereby regulating the Th17/Treg cell balance. Busbee PB et al. found that indole-3-carbinol (I3C) activates AhR in colitis mice, reducing pro-inflammatory Th17 cells associated with colitis and increasing Treg cells involved in maintaining gut homeostasis, thus alleviating inflammation [199]. During inflammation, polymorphonuclear leukocytes (PMNs) are recruited, and their myeloperoxidase (MPO) defends against invading microorganisms and maintains epithelial barrier function by producing hypochlorous acid (HOCl) [205]. However, MPO itself induces PMN degranulation, releasing MPO into the external environment, damaging tissues, and inducing inflammation [206]. Experiments found that indole and its derivatives inhibit MPO by directly binding to its active site or competing with MPO substrates (e.g., H₂O₂ and Cl⁻), interrupting the MPO-driven inflammatory feedback loop, thereby reducing inflammation and promoting tissue repair [207].

In addition to protecting the intestinal barrier and alleviating gut inflammation, studies have found that the indole metabolite IPA reshapes the gut microbiota and maintains gut homeostasis through AhR activation. Previous studies [208] found that IPA inhibits the biosynthesis of tryptophan in Mycobacterium tuberculosis and non-tuberculous mycobacteria by binding to TrpE, the first key enzyme in tryptophan synthesis. Additionally, some commensal bacteria that play a protective role in colitis utilize intestinal mucins as a carbon source [20]. In patients with inflammatory bowel disease (IBD), the intestinal mucus layer becomes thinner, affecting the structure and function of the microbiota. As mentioned above, IPA may restore the structure and function of these microbiota by promoting mucin expression (Figure 6).