Metabolic landscape of head and neck squamous cell carcinoma informs a novel kynurenine/Siglec-15 axis in immune escape

Abstract

Background: Metabolic reprograming and immune escape are two hallmarks of cancer. However, how metabolic disorders drive immune escape in head and neck squamous cell carcinoma (HNSCC) remains unclear. Therefore, the aim of the present study was to investigate the metabolic landscape of HNSCC and its mechanism of driving immune escape.

Methods: Analysis of paired tumor tissues and adjacent normal tissues from 69 HNSCC patients was performed using liquid/gas chromatography-mass spectrometry and RNA-sequencing. The tumor-promoting function of kynurenine (Kyn) was explored in vitro and in vivo. The downstream target of Kyn was investigated in CD8⁺ T cells. The regulation of CD8⁺ T cells was investigated after Siglec-15 overexpression in vivo. An engineering nanoparticle was established to deliver Siglec-15 small interfering RNA (siS15), and its association with immunotherapy response were investigated. The association between Siglec-15 and CD8⁺ programmed cell death 1 (PD-1)⁺ T cells was analyzed in a HNSCC patient cohort.

Results: A total of 178 metabolites showed significant dysregulation in HNSCC, including carbohydrates, lipids and lipid-like molecules, and amino acids. Among these, amino acid metabolism was the most significantly altered, especially Kyn, which promoted tumor proliferation and metastasis. In addition, most immune checkpoint molecules were upregulated in Kyn-high patients based on RNA-sequencing. Furthermore, tumor-derived Kyn was transferred into CD8⁺ T cells and induced T cell functional exhaustion, and blocking Kyn transporters restored its killing activity. Accroding to the results, mechanistically, Kyn transcriptionally regulated the expression of Siglec-15 via aryl hydrocarbon receptor (AhR), and overexpression of Siglec-15 promoted immune escape by suppressing T cell infiltration and activation. Targeting AhR in vivo reduced Kyn-mediated Siglec-15 expression and promoted intratumoral CD8⁺ T cell infiltration and killing capacity. Finally, a NH₂-modified mesoporous silica nanoparticle was designed to deliver siS15, which restored CD8⁺ T cell function status and enhanced anti-PD-1 efficacy in tumor-bearing immunocompetent mice. Clinically, Siglec-15 was positively correlated with AhR expression and CD8⁺PD-1⁺ T cell infiltration in HNSCC tissues.

Conclusions: The findings describe the metabolic landscape of HNSCC comprehensively and reveal that the Kyn/Siglec-15 axis may be a novel potential immunometabolism mechanism, providing a promising therapeutic strategy for cancers.

Keywords: Siglec‐15; T cell exhaustion; head and neck squamous cell carcinoma; kynurenine; untargeted metabolomics.

FIGURE 1

GC/LC‐MS analysis revealed metabolic reprograming in HNSCC and alterations of the Trp/Kyn pathway. (A) Flowchart of the study design. Tumor tissues and adjacent normal tissues from HNSCC patients (n = 69) were analyzed using multi‐omics profiling, including GC‐MS, LC‐MS, and RNA‐seq, to unveil the metabolic landscape of HNSCC. (B) Metabolomics heatmap based on LC‐MS results. Rows show the different metabolites (n = 178), and columns show the different tissues (n = 69). The log‐transformed metabolite intensities were Z scored/standardized. (C) The numbers of differential metabolites. (D) The expression of human amino acids and their metabolites were measured in HNSCC. Each point represents the fold change of metabolites in each patient (T/N). (E) Heatmap depicts the relative intensities of the metabolites (n = 138) in the Trp pathway, analyzed by GC‐MS. (F) Venn diagram shows the overlap of the metabolites in Trp pathways identified by GC‐MS and LC‐MS. (G) A representative chromatogram of Kyn in GC‐MS. (H) Schematic representation of key enzymes and metabolites involved in the Trp/Kyn pathway. Red and blue indicated high and low expression/levels, respectively. (I) LC‐MS targeted metabolomics analysis shows the Trp and Kyn contents in 5 HNSCC patients. (J) Levels of Kyn, as measured using GC‐MS, in HNSCC patients (n = 69). (K) Levels of Kyn and their association with TNM stage, pathological grade, and depth of invasion in HNSCC (n = 69). Data are represented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001. Abbreviations: HNSCC, head and neck squamous cell carcinoma; LC‐MS, liquid chromatography‐mass spectrometry; GC‐MS, gas chromatography‐mass spectrometry; FC, fold change; OS, overall survival; Kyn, kynurenine; Trp, tryptophan; 5‐HTP, 5‐hyroxy‐L‐tryptophan; RT, retention time; I3P, indole‐3‐pyruvate; IAA, indole‐3‐acetic acid; ILA, indole‐3‐lactic acid; I3A, indole‐3‐aldehyde; KYNA, kynurenic acid; XA, xanthurenic acid; IL4I1, interleukin‐4‐induced‐1; IDO1, indoleamine‐2,3‐dioxygenase 1; TDO2, tryptophan‐2,3‐dioxgenase; MAOA/B, monoamine oxidase A/B; KATs, kynurenine aminotransferases; KMO, kynurenine‐3‐monooxidase, KYNU, kynureninase; NAD⁺, nicotinamide adenine dinucleotide; SEM, tandard error of the mean.

FIGURE 2

Kyn promoted HNSCC aggressive progression both in vivo and in vitro. (A) CCK8 assay following indicated Kyn treatments in Cal27 and HN30 cells. (B) Colony formation assay was performed using Cal27 and HN30 cells following stimulation with Kyn. (C) EdU assay was performed to detect cell proliferation after Kyn stimulation for 24 h. (D) Wound healing assay was conducted to detect the migration ability of HNSCC cells after Kyn stimulation for 36 h. (E‐F) Migration and invasion capacities were detected using the transwell assay after Kyn stimulation for 24 h. (G) E‐cadherin, Vimentin, and Snail were detected after Kyn stimulation for 24 h. (H) Schematic diagram of the Kyn administration strategy in C3H/He tongue orthotopic transplant models. (I) The volume of tumor was compared between the two groups (n = 6 mice per group). (J) Neck lymph node metastasis analysis of the orthotopic transplant models. Data are represented as the mean ± SEM (A‐G and I‐J) based on three independent experiments (A–G). *P < 0.05, **P < 0.01, ***P < 0.001. Abbreviations: Kyn, kynurenine; HNSCC, head and neck squamous cell carcinoma, CCK8, Cell Counting Kit‐8; EdU, 5‐ethynyl‐2′‐deoxyuridine; SEM, tandard error of the mean.

FIGURE 3

Kyn promoted CD8⁺ T cell dysfunction and Siglec‐15 expression in tumor cells. (A) Heatmap analysis showing classical immune checkpoint gene expression in HNSCC, grouped by Kyn levels (n = 69). (B) Gene set enrichment analysis (GSEA) of mRNAs in the Kyn‐high group. (C) The percentage of EdU⁺ (48 h, upper) and CFSE⁺ (96 h, lower) cells among human primary CD8⁺ T cells, isolated from the peripheral blood of healthy controls using a human CD8 MicroBeads kit and measured using flow cytometry, after the indicated Kyn stimulation. (D) The percentage of PD‐1⁺ cells in primary CD8⁺ T (upper) and Jurkat cell line (lower), as measured using flow cytometry, after the indicated Kyn stimulation for 48 h. (E) Western blotting analysis of PD‐1 expression after the indicated Kyn stimulation for 48 h in Jurkat cell line with or without PHA stimulation and in primary CD8⁺ T cells. Representative images (left) and three experiment replicates (right) are displayed. (F) Comparative heatmap depicted differential gene expression after Kyn stimulation (200 µmol/L, 48 h) using RT‐qPCR data. (G) Kyn (100 µmol/L) and the system L inhibitor, BCH (5 mmol/L) were used as the indicated treatments to analyze the functions of Kyn on the dysfunction of primary CD8⁺ T cells and Jurkat cells (48 h). (H) Heatmap analysis shows classical common immune checkpoint ligand gene expression in HNSCC and adjacent normal tissues from 69 patients. (I) Analysis of SIGLEC15 expression in HNSCC stratified by high or low CD274 expression. (J) PD‐L1 and Siglec‐15 expression after treatment with the indicated Kyn concentrations for 48 h (upper) or 200 µmol/L Kyn for 0, 12, 24, and 48 h (lower) by Western blotting. (K) Nuclear expression of AhR was detected by immunofluorescence staining after 200 µmol/L Kyn stimulation for 1 h. (L) Cal27 and HN30 cell lines were treated with PBS or Kyn (200 µmol/L) and/or BAY‐218 (10 µmol/L) for 48 h, and PD‐L1 and Siglec‐15 expression was detected by Western blotting. (M) Cal27 and HN30 cell lines were treated with PBS or Kyn (200 µmol/L) and/or CH‐223191 (10 µmol/L) for 48 h, and PD‐L1 and Siglec‐15 expression was detected by Western blotting. (N) PD‐L1 and Siglec‐15 expression was detected using Western blotting after siAhR transfection for 48 h and 200 µmol/L Kyn treatment for 48 h. Data are represented as the mean ± SEM based on three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001, ns: not significant. Abbreviations: Kyn, kynurenine; HNSCC, head and neck squamous cell carcinoma, FC, fold change; GSEA, gene set enrichment analysis; ES, enrichment score; NES, normalized enrichment score; FDR, false discovery rate; EdU, 5‐ethynyl‐2′‐deoxyuridine; CFSE, carboxyfluorescein succinimidyl ester; PD‐1, programmed cell death protein 1; PHA, polyhydroxyalkanoate; IFN‐γ, interferon‐gamma; TNF‐α, tumor necrosis factor‐alpha; RT‐qPCR, real‐time quantitative polymerase chain reaction; BCH, 2‐Aminobicyclo‐(2,2,1)‐heptane‐2‐carboxylic acid; FC, fold change; PD‐L1, programmed death‐ligand 1; DAPI, 4',6‐diamidino‐2‐phenylindole; AhR, aryl hydrocarbon receptor; MFI, mean fluorescence intensity; PBS, phosphate buffered saline; SEM, tandard error of the mean; ns, not significant.

FIGURE 4

Kyn induced Siglec‐15‐mediated immune escape. (A) The chromatin immunoprecipitation (ChIP)‐PCR assay was performed using an IgG or AhR antibody after treatment with 200 µmol/L Kyn for 1 h. Two primers targeting the promoter region of SIGLEC15 and CD274 mRNA were used for RT‐qPCR analysis. (B) 293T cells were co‐transfected with SIGLEC15/CD274 promoter‐luciferase reporter PGL3 for 24 h and treated with the indicated concentration of Kyn for another 6 h, followed by an analysis of luciferase activity. (C) 293T cells were co‐transfected with SIGLEC15/CD274 promoter‐luciferase reporter PGL3 for 24 h and treated with 200 µmol/L Kyn for the indicated time, followed by an analysis of luciferase activity. (D) 293T cells were co‐transfected with promoter‐luciferase reporter plasmids and siAhR /siScr for 24 h and treated with 200 µmol/L Kyn for 6 h. (E) SCCVII cells were transfected with lentivirus‐SIGLEC15, and transfection was confirmed using Western blotting analysis. (F) Vector or SIGLEC15‐overexpressing SCCVII cells were subcutaneously injected into C3H/He mice (n = 5 per group). Tumor volumes were measured once every two days. (G) Tumor weights were measured after mice were euthanized. (H‐J) H&E (H), TUNEL (I), and Ki‐67 (J) staining analyses of tumor tissues in each group. (K‐P) Tumor‐infiltrating lymphocytes (TIL) harvested from xenograft tumors, and the percentages of CD3⁺CD8⁺ (K), Ki‐67⁺ (L), PD‐1⁺ (M), IFN‐γ ⁺ (N), granzyme B⁺ (O), and perforin⁺ cells (P) were analyzed using flow cytometry. (Q) The percentages of CD8⁺ in the indicated tumors were analyzed by multiplex immunofluorescence staining. (R) The volume and Ki‐67 expression of tumor were compared in tongue orthotopic transplant models established by vector or SIGLEC15‐overexpressing SCCVII cells (n = 6 mice per group). (S) PD‐1⁺CD8⁺ T cells were compared in tongue orthotopic transplant models (n = 6 mice per group), based on multiplex immunofluorescence staining. Three regions of interest (ROIs) in each tumor were analyzed and measured. Data are represented as mean ± SEM (A‐D, F‐G and I‐S) based on three independent experiments (A‐D). *P < 0.05, **P < 0.01, ***P < 0.001, ns: not significant. Abbreviations: Kyn, kynurenine; ChIP, chromatin immunoprecipitation; RT‐qPCR, real‐time quantitative polymerase chain reaction; AhR, aryl hydrocarbon receptor; siScr, siScramble; H&E, hematoxylin and eosin; TUNEL, TdT‐mediated dUTP‐biotin nick end labeling; PD‐1, programmed cell death protein 1; TIL, Tumor‐infiltrating lymphocyte; IFN‐γ, interferon‐gamma; ROI, region of interest; SEM, tandard error of the mean; ns, not significant.

FIGURE 5

Targeting AhR reversed Kyn‐Siglec‐15‐mediated immune escape in vivo. (A) Mice with SCCVII were treated with Kyn (100 mg/kg) or/and CH‐223191 (10 µmol/L) once every two days for three times (n = 5 per group). Tumor volumes were measured once every two days. (B) Tumor weights were measured after mice euthanized on day 20 (n = 5 per group). (C) Western blotting analysis of Siglec‐15 expression after Kyn stimulation in mice. (D‐F) H&E (D), TUNEL (E and F upper), and Ki‐67 (E and F lower) staining analyses of tumor tissues in each group. (G) Concentration of serum TNF‐α and IFN‐γ in mice were measured by ELISA. (H‐M) The percentage of CD3⁺CD8⁺ (H), Ki‐67⁺ (I), PD‐1⁺ (J), IFN‐γ⁺ (K), granzyme B⁺ (L), and perforin⁺ cells (M) in cytotoxic T lymphocytes (CTLs) isolated from the indicated tumors were analyzed. (N) Representative images of multiplex immunofluorescence staining of CD3⁺ (green) and CD8⁺ (red) are shown, and quantification analysis were performed in tumor tissues (n = 10 fields of 5 mice per group). Data are represented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ns: not significant. Abbreviations: AhR, aryl hydrocarbon receptor; Kyn, kynurenine; H&E, hematoxylin and eosin; TUNEL, TdT‐mediated dUTP‐biotin nick end labeling; TNF‐α, tumor necrosis factor‐alpha; IFN‐γ, interferon‐gamma; ELISA, enzyme‐linked immunosorbent assay; PD‐1, programmed cell death protein 1; CTL, cytotoxic T lymphocyte; DAPI, 4',6‐diamidino‐2‐phenylindole; SEM, tandard error of the mean; ns, not significant.

FIGURE 6

Siglec‐15 specific siRNA delivered by NH₂‐MSN nanoparticles enhances immunotherapy efficacy in vivo. (A) Synthesis routes of NH₂‐MSN nanoparticles loaded with siRNA. (B) Macroscopy characterization of NH₂‐MSNs with siS15. (C) Transmission electron microscopy images of NH₂‐MSNs and NP‐siS15 with indicated ratios. (D‐E) The morphology and Zeta potential of NH₂‐MSNs and NP‐siS15. (F) Scanning microscopy analysis of the cellular uptake of NH₂‐MSNs with or without the indicated siS15‐Cy5 by SCCVII cells after in vitro treatment for 24 h. (G) Western blotting analysis confirming Siglec‐15 gene‐silencing effect by siRNA released from nanoparticles in SCCVII cells. (H) Schematic diagram of the treatment strategy in SCCVII mouse model. (I) Probability of survival analysis for each group was performed. (J) Analysis of TUNEL and Ki‐67 staining of tumor tissues in each group. (K‐L) The percentage of CD3⁺CD8⁺ (K) and PD‐1⁺cells in CTLs (L) of the indicated treatment groups. (M) Analysis of multiplex immunofluorescence staining of CD3⁺ (green) and CD8⁺ (red) were shown and quantification analysis were performed in tumor tissues (n = 10 fields of five mice per group). (N) Schematic diagram of the anti‐PD‐L1 and NP‐siS15 combination treatment strategy in C3H/He subcutaneous tumorigenesis models. (O‐P) Tumor volume and weights were measured and analyzed in the indicated groups. Data are represented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ns: not significant. Abbreviations: siS15, Siglec‐15 small interfering RNA; DAPI, 4',6‐diamidino‐2‐phenylindole; AhR, aryl hydrocarbon receptor; NP‐siS15, NH₂‐MSN‐siSIGLEC15; DAPI, 4',6‐diamidino‐2‐phenylindole; siScr, siScramble; anti‐PD‐1, anti‐programmed cell death protein 1 antibody; TUNEL, TdT‐mediated dUTP‐biotin nick end labeling; CTL, cytotoxic T lymphocyte; anti‐PD‐L1, anti‐programmed death‐ligand 1 antibody; SEM, tandard error of the mean; ns, not significant.

FIGURE 7

Siglec‐15 expression positively correlates with CD8⁺ T cell exhaustion in HNSCC patients. (A) Representative images of immunohistochemical staining for AhR, Siglec‐15, and PD‐L1 on a TMA of 70 HNSCC patients. (B) The correlation between Siglec‐15 expression and AhR expression using immunoreactive score (IRS) (n = 70). (C) Comparison of positive percentage of Siglec‐15 and PD‐L1 expression in HNSCC (n = 70). (D) GEPIA HNSCC dataset shows SIGLEC15 expression with the corresponding survival rates (n = 50) (E) Representative images show CD8⁺PD‐1⁺ T cells using multiplex immunofluorescence staining in TMA (n = 70). (F) A schematic showing the mechanism via which Kyn/AhR/Siglec‐15 signaling promotes CD8⁺ T cell exhaustion. Data are represented as mean ± SEM. *P < 0.05, ***P < 0.001. Abbreviations: HNSCC, head and neck squamous cell carcinoma; AhR, aryl hydrocarbon receptor; PD‐L1, programmed death‐ligand 1; TMA, tissue microarray; IRS, immunoreactive score; GEPIA, Gene Expression Profiling Interactive Analysis; TPM, transcripts per million; PD‐1, programmed cell death protein 1; DAPI, 4',6‐diamidino‐2‐phenylindole; Trp, tryptophan; IDO, indoleamine 2,3‐dioxygenase 1; Kyn, kynurenine; BCH, 2‐Aminobicyclo‐(2,2,1)‐heptane‐2‐carboxylic acid; SEM, tandard error of the mean.