IL-4-STAT6-induced high Siglec-G/10 expression aggravates the severe immune suppressive tumor microenvironment and impedes the efficacy of immunotherapy in head and neck squamous cell carcinoma

Abstract

Background: Immune checkpoint blockade therapy has shown limited efficacy in head and neck squamous cell carcinoma (HNSCC). Sialic acid binding immunoglobulin-like lectin (Siglec)-15 has been identified as a novel immune evasion biomarker, while the role of Siglec-10 in the specific immune suppressive tumor microenvironment remains largely unknown.

Methods: Immunohistochemical assays were employed to investigate the correlation of the expressions of Siglec-10 and Siglec-15 with the clinicopathological features as well as the prognosis of immunotherapy in patients with HNSCC. The Gene Expression Omnibus datasets were used to identify the upstream transcriptional regulators of SIGLEC10 in tumor-associated macrophages (TAMs) and the downstream biological functions it mediates. These findings were then validated through in vitro and in vivo experiments. The impact of Siglecg deficiency on the efficacy of immunotherapy and the activation of CD8+T cells was analyzed in mouse HNSCC tumor-bearing models.

Results: The expression of Siglec-G/10, rather than that of Siglec-15, was positively correlated with immune suppressive marker programmed death-ligand 1 (PD-L1) expression and was associated with cervical lymph node metastasis, poorer pathologic stage, and lower sensitivity to immunotherapy. Siglecg deficiency rescued the immune suppressive tumor microenvironment, as evidenced by decreased TAM-associated phenotype and increased CD8+T cell infiltration and activation, which inhibited tumor growth significantly. Single-cell sequence and transcription factor prediction revealed that signal transducer and activator of transcription 6 (STAT6) could induce Siglec-G/10 transcription. Interleukin (IL)-4 could upregulate Siglec-G/10 expression significantly via STAT6 activation, as proved by overexpression and inhibition of STAT6. Signal transduction mechanism revealed that Siglec-G/10 could promote TAM differentiation and activation via increasing HIF1α (hypoxia-inducible factor 1α) expression. Furthermore, Siglecg deficiency could enhance the efficacy of immune checkpoint inhibitor, and increase the infiltration and cytotoxic functions of CD8+T cells.

Conclusions: Our results suggest that high Siglec-G/10 expression aggravates the immune suppressive tumor microenvironment and impedes the immunotherapy efficacy in HNSCC, which indicates that targeting Siglec-G/10 may represent a promising therapeutic option for improving the immunotherapy efficacy in HNSCC.

Keywords: Head and Neck Cancer; Immune Checkpoint Inhibitor; Immunotherapy; JAK-STAT; Macrophage.

Figure 1. Siglec-10 expression is positively correlated with immune suppressive markers in HNSCC. (a) IHC analysis of Siglec-15, Siglec-10 and PD-L1 expression levels in 40 HNSCC tissues. Representative immunohistochemistry images of high expression and low expression of Siglec-15, Siglec-10 and PD-L1 staining were shown, magnification: ×50, ×200 and × 400. (b) The IHC scores of PD-L1 and Siglec-10 show a positive correlation. (c) The TCGA database indicates that in HNSCC, the mRNA expression level of CD274 is positively correlated with that of SIGLEC10. (d) The IHC scores of PD-L1 and Siglec-15 exhibit no associative relationship. (e) The TCGA database indicates that in HNSCC, the mRNA expression level of CD274 is not correlated with that of SIGLEC15. (f–i) Statistical analysis about cervical lymph node metastasis, pathologic stage, treatment response and progression-free survival rate with Siglec-10 stain intensity in HNSCC tissues. (j) The expression level of Siglec-10 protein in HNSCC tumor tissues and normal tissues was determined by immunoblotting. GAPDH was used as an internal loading control. (k) The correlation between SIGLEC10 expression and tumor purity was analyzed using the TIMER2.0 database. (l) Immunofluorescence staining of Siglec-10 distribution (green) and epithelial cell marker pan-ck (red) in tissues from patients with HNSCC. Scale bar: 200 µm. (m) The correlation of SIGLEC10 expression with immune infiltration level in HNSCC investigated in TCGA database based on six deconvolution algorithms. (n) Representative images and cumulative results of flow cytometry showed infiltration of Siglec-10+cells in total CD45+cells in tumor and peritumor tissues (n=10). Data were analyzed using the student’s t-test. *p<0.05, **p<0.01. Data are expressed as mean±SD. CR: complete response; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; HNSCC, head and neck squamous cell carcinoma; IHC, immunohistochemical; LN, lymph node; mRNA, messenger RNA; PD: progressive disease; PD-L1: programmed death-ligand 1; PR: partial response; SD: stable disease; SSC-A: side scatter area; Siglec, sialic acid binding immunoglobulin-like lectin; TCGA, The Cancer Genome Atlas.

Figure 2. Siglec-G/10 promotes HNSCC transplant-tumor growth by reducing CD8+T cells infiltration. (a) Schematic illustration of Siglec-G-Fc and PD-1/PD-L1 inhibitory BMS-1 therapy. WT mice were subcutaneously inoculated with MTCQ1 HNSCC cells and administered with Siglec-G-Fc and BMS-1 or left untreated on days 3, 7, 10, 14 and 17. The tumor volume was monitored. (b, c) Volumes and weights of MTCQ1 subcutaneous tumorigenesis (n=5 mice per group), Scale bar: 1 cm. (d, e) Representative images of multiplex immunofluorescence staining of CD8+ (green) and Siglec-G+ (red) are shown, and quantification analysis was performed in tumor tissues (n=5 fields of five mice per group), Scale bar: 100 µm. (f) The exhibition of subcutaneous tumorigenesis in Siglecg^+/+ mice and Siglecg^−/− mice. (g) The exhibition of isolated tumors. (h, i) Tumor growth curve and tumor weight of MTCQ1 cells derived subcutaneous tumorigenesis model in two groups for 31 days (n=5). (j) Flow cytometry analysis for percentage of Siglec-G+cells in CD45+and CD45- subsets in tumors; (k) Flow cytometry analysis for infiltration of macrophages, monocytes, neutrophils, dendritic cells, T cells, B cells and NK cells in tumors. (l) Flow cytometry analysis for percentage of CD8+T cells in tumors. (m, n) Representative images of multiplex immunofluorescence staining of CD8+ (green) and F4/80+ (red) are shown, and quantification analysis was performed in tumor tissues (n=6 fields of six mice per group), Scale bar: 50 µm. (o) Representative images of multiplex immunofluorescence staining of F4/80+ (red), Siglec-G+ (green) and PD-L1+ (gold) are shown, and quantification analysis was performed in tumor tissues (n=6 fields of six mice per group), Scale bar: 100 µm. ns: not significant (p>0.05), *p<0.05, **p<0.01. Data are expressed as mean±SD. BMS-1: PD-1/PD-L1 inhibitor 1; DCs: dendritic cells; HNSCC: head and neck squamous cell carcinoma; i.p.: intraperitoneal; NK: natural killer; PD-1: programmed cell death protein 1; PD-L1: programmed cell death ligand 1; Siglec: sialic acid binding immunoglobulin-like lectin; WT: wild type.

Figure 3. Siglec-G/10 induces an immune-suppressive tumor microenvironment by promoting TAM differentiation in HNSCC. (a) Comparison of SIGLEC10 expression in different subpopulations in HNSCC from single-cell dataset GSE139324. (b) Representative flow cytometry figures and cumulative results showing subpopulations gated on Siglec-10+CD45+ leucocytes from fresh human HNSCC samples (n=10). (c) Immunofluorescence staining of Siglec-10 distribution (green) and macrophage marker CD68 (red) in tissues from patients with HNSCC. Scale bar: 200 µm. (d) Statistical analysis of the number of Siglec-10+TAMs per square millimeter in tumor samples from immunotherapy sensitive and immunotherapy resistant groups of patients with HNSCC. (e) Kaplan-Meier analysis of the progression-free survival for 40 patients with HNSCC in the two paired groups (Siglec-10+TAMs^high and Siglec-10+TAMs^low groups). (f) Flow cytometry analysis of the mean fluorescence intensity (MFI) of M1/M2 macrophage markers in HNSCC tumors of Siglecg^+/+ or Siglecg^−/− mice. (g) The exhibition of isolated tumors in four different groups. (h, i) Tumor growth curve and tumor weight of MTCQ1 cells derived subcutaneous tumorigenesis model in four groups for 31 days (n=5). (j) Flow cytometry analysis for percentage of CD45+CD3+ cells in tumors. (k) Flow cytometry analysis for infiltration of CD8+T cells and percentage of IFN-γ+ CD8+ T cells, GZMB+CD8+ T cells and IFN-γ+ GZMB+ CD8+ T cells in tumors. ns: not significant (p>0.05), *p<0.05, **p<0.01. Data are expressed as mean±SD. Arg 1: arginase 1; BMDM: bone marrow-derived macrophage; CR: complete response; DC: dendritic cell; DAPI: 4',6-diamidino-2-phenylindole; FPS: free progression survival; GZMB: granzyme B; HNSCC: head and neck squamous cell carcinoma; IFN: interferon; IL: interleukin; MHC-II: major histocompatibility complex class II; NK: natural killer; PD: progressive disease; PD-L1: programmed death-ligand 1; PFS: progression-free survival; PR: partial response; SD: stable disease; Siglec: sialic acid binding immunoglobulin-like lectin; SSC-A: side scatter area; TAMs: tumor-associated macrophages.

Figure 4. IL-4-STAT6 signaling pathway transcriptionally upregulates Siglec-G/10 expression in macrophages. (a) UMAP plot from GSE139324 analyzed by ST. Each dot corresponds to a single spot, color coded by individual sections or regions. (b) Heat map of differentially expressed genes in SIGLEC10+ TAMs and SIGLEC10neg TAMs. Rows represent genes and columns represent individual cells. (c) Venn diagram showing the differentially expressed gene (DEG) of SIGLEC10+ TAMs and the predicted SIGLEC10 transcription factors in the Jaspar database and the Knock TFs database, the numbers of shared and exclusive genes were exhibited. (d) THP1-derived macrophages and Raw264.7 were transfected with STAT6 overexpression plasmids for 48 hours. STAT6, Siglec-G/10 and β-actin expressions were detected by western blotting. (e) Macrophages were treated at indicated concentrations of IL-4 for 24 hours and Siglec-G/10 was detected by western blotting. (f) THP1-derived macrophages were treated with 20 ng/mL IL-4 for 24 hours. Immunofluorescence assay was applied for assessing the location and expression of Siglec-10 in macrophages. (g, h) Cells were treated at indicated concentrations of STAT6 inhibitor AS1810722 for 24 hours and treated with IL-4 (20 ng/mL). STAT6, p-STAT6, and Siglec-G/10 expression were detected by western blotting after 24 hours. (i) Predicted sequence of STAT6 binding site. (j) ChIP-qRT-PCR analysis of p-STAT6 binding on the SIGLEC10/siglecg promoter region after 24 hours of 40 ng/mL IL-4 treatment. (k) ChIP-qRT-PCR analysis of p-STAT6 binding on the SIGLEC10/siglecg promoter region after 24 hours of 40 ng/mL IL-4 treatment with 200 nM AS1018722 or not. (l) WT mice were subcutaneously inoculated with MTCQ1 HNSCC cells and administered with AS1810722 and BMS-1 or left untreated. Mice were sacrificed at 21 days and the tumors were harvested, and those from the four different groups were presented. (m, n) The tumor growth curves of the subcutaneous tumorigenesis models established using MTCQ1 cells in four groups over 21 days and the tumor volumes on the 21st day (n=5). (o, p) Flow cytometry analysis for the percentage of Siglec-G+PD-L1+ macrophages in mouse tumor-bearing models. *p<0.05, **p<0.01. Data are expressed as mean±SD. BMS-1: PD-1/PD-L1 inhibitor 1; ChIP-qRT-PCR: chromatin immunoprecipitation followed by quantitative real-time polymerase chain reaction; DAPI: 4',6-diamidino-2-phenylindole; HNSCC: head and neck squamous cell carcinoma; IL: interleukin; KO: knockout; neg: negative; OE: overexpression; PD-L1: programmed death-ligand 1; PBS: phosphate buffered saline; Siglec: sialic acid binding immunoglobulin-like lectin; STAT6: signal transducer and activator of transcription 6; ST: spatial transcriptomics; TAMs: tumor-associated macrophages; TFs: transcription factors; UMAP: uniform manifold approximation and projection; WT: wild type.

Figure 5. Siglec-G/10 augments the angiogenic activities of macrophages in HNSCC. (a) GO analysis based on the differentially expressed genes in SIGLEC10+ TAMs from GSE139324. (b) Pathway analysis based on the differentially expressed genes in SIGLEC10+ TAMs that are different from SIGLEC10neg TAMs in GSE139324. (c) Indicated cytokines expression in THP1 derived macrophages with different Siglec-10 expression were detected via ELISA. (d) Representative images of tube formation experiments of HUVECs in the THP1-derived macrophages and BMDMs with different Siglec-G/10 expression groups. Quantitative statistics of cell junctions and meshes. (e) The correlation between SIGLEC10 and CD31 expression in HNSCC was analyzed in the TCGA database. (f) The effect of conditioned medium derived from macrophages with different Siglec-G/10 expression levels on the CD31 expression of HUVECs was detected by western blotting. (g) HIF1α and Siglec-G/10 expression of THP1 derived macrophages and BMDMs with different Siglec-G/10 expression were detected by western blotting. Membranes were probed with a β-actin antibody as a loading control. ImageJ densitometric analysis of the HIF1α/β-actin ratio was shown. (h, i) Representative images of multiplex immunofluorescence staining of F4/80+ (red), Siglec-G+ (green) and HIF1α+ (purple) are shown, and quantification analysis was performed in tumor tissues (n=6 fields of six mice per group), Scale bar: 100 µm. (j, k) THP1_{SIGLEC10 OE} derived macrophages and BMDM_{Siglecg OE} were treated with HIF1α inhibitors BAY and LW6 for 24 hours under hypoxic conditions. HIF1α expression was detected by western blotting (j) and angiogenesis-related cytokines were detected by ELISA (k). (l) The effect of conditioned medium derived from macrophages treated with HIF1α inhibitors on angiogenesis capacity of HUVECs was detected by tube formation experiments. (m–o) The effect of conditioned medium derived from macrophages treated with HIF1α inhibitors on CD31 expression of HUVECs was detected by immunofluorescence (m), flow cytometry (n) and western blotting (o). *p<0.05, **p<0.01. Data are expressed as mean±SD. BMDM: bone marrow-derived macrophage; FGF: fibroblast growth factor; GO: gene ontology; HIF1α: hypoxia-inducible factor 1 alpha; HNSCC: head and neck squamous cell carcinoma; HUVECs: human umbilical vein endothelial cells; KO: knockout; MHC: major histocompatibility complex; NC: negative control; OE: overexpression; PDGF: platelet-derived growth factor; Siglec: sialic acid binding immunoglobulin-like lectin; TCGA: The Cancer Genome Atlas; TAMs: tumor-associated macrophages; TGF-β: transforming growth factor-beta; THP1: a human acute monocytic leukemia cell line; VEGF: vascular endothelial growth factor.

Figure 6. Targeting Siglec-G/10 enhances the therapeutic effect of ICI in vivo in HNSCC. (a) Schematic illustration of PBS and PD-1/PD-L1 inhibitory BMS-1 therapy. MTCQ1 cells were injected into the backs of Siglecg^+/+ mice or Siglecg^−/− mice. Mice were sacrificed at 28 days and the tumors were harvested for further analysis after BMS-1 treatment. (b) The exhibition of isolated tumor in four different groups. (c) Tumor growth curve MTCQ1 cells derived subcutaneous tumorigenesis model in four groups for 28 days (n=5). (d) Flow cytometry analysis for MFI of Siglec-G and PD-L1 in CD45+cells in tumors. (e) Representative images of multiplex immunofluorescence staining of CD8+ (green), Siglec-G+ (red) and F4/80+ (purple) are shown, and quantification analysis were performed in tumor tissues (n=5 fields of five mice per group). (f) Flow cytometry analysis for infiltration of macrophages and percentage of CD86+macrophages, MHC-II+macrophages and CD206+macrophages in tumors. (g) Flow cytometry analysis for infiltration of CD8+T cells and percentage of IFN-γ+ CD8+ T cells, GZMB+CD8+ T cells and IFN-γ+ GZMB+ CD8+ T cells in tumors. (h) Representative images of multiplex immunofluorescence staining of CD31+ (red) and HIF1α+ (gold) are shown, and quantification analysis was performed in tumor tissues (n=5 fields of five mice per group). ns: not significant (p>0.05), *p<0.05, **p<0.01. Data are expressed as mean±SD. BMS-1: PD-1/PD-L1 inhibitor 1; GZMB: granzyme B; HIF1α: hypoxia-inducible factor 1 alpha; HNSCC: head and neck squamous cell carcinoma; ICI: immune checkpoint inhibitor; IFN: interferon; i.p.: intraperitoneal; KO: knockout; MHC: major histocompatibility complex; MFI: mean fluorescence intensity; PD-1: programmed cell death protein 1; PD-L1: programmed death-ligand 1; PBS: phosphate buffered saline; Siglec: sialic acid binding immunoglobulin-like lectin; WT: wild type.

Figure 7. IL-4-STAT6-induced high Siglec-G/10 expression aggravates the severe immune suppressive TME and impedes the efficacy of immunotherapy in HNSCC via increasing HIF1α expression and TAM function. Siglecg deficiency could enhance ICI efficacy of HNSCC by rescuing the TAM function and enhancing the activation of CD8+T cells. FGF: fibroblast growth factor; HNSCC: head and neck squamous cell carcinoma; HIF: hypoxia-inducible factor; ICI: immune checkpoint inhibitor; IL: interleukin; PDGF: platelet-derived growth factor; p-STAT6: phosphorylated signal transducer and activator of transcription 6; Siglec: sialic acid binding immunoglobulin-like lectin; TAMs: tumor-associated macrophages; TME: tumor microenvironment; TGF-β: transforming growth factor-beta; VEGF: vascular endothelial growth factor.