Siglec-G Suppresses CD8+ T Cells Responses through Metabolic Rewiring and Can be Targeted to Enhance Tumor Immunotherapy

Abstract

CD8⁺ T cells play a critical role in cancer immune-surveillance and pathogen elimination. However, their effector function can be severely impaired by inhibitory receptors such as programmed death-1 (PD-1) and T cell immunoglobulin domain and mucin domain-3 (Tim-3). Here Siglec-G is identified as a coinhibitory receptor that limits CD8⁺ T cell function. Siglec-G is highly expressed on tumor-infiltrating T cells and is enriched in the exhausted T cell subset. Ablation of Siglec-G enhances the efficacy of adoptively transferred T cells and chimeric antigen receptor (CAR) T cells in suppressing solid tumors growth. Mechanistically, sialoglycan ligands, such as CD24 on tumor cells, activate the Siglec-G-SHP2 axis in CD8⁺ T cells, impairing metabolic reprogramming from oxidative phosphorylation to glycolysis, which dampens cytotoxic T lymphocyte (CTL) activation, expansion, and cytotoxicity. These findings discover a critical role for Siglec-G in inhibiting CD8⁺ T cell responses, suggesting its potential therapeutic effect in adoptive T cell therapy and tumor immunotherapy.

Keywords: CD8+ T cells; Siglec‐G; metabolic rewiring; tumor immunotherapy.

Figure 1

Characteristic expression of Siglec‐G in T cells. A) Heatmap showing Siglecs family genes expression in four CD8⁺ T cells clusters from peripheral blood of healthy volunteers from GEO DataSet (GSE93776). B) Q‐PCR analysis of Siglec‐G expression in murine splenic immune cells sorted from WT mice (n = 3). C,D) Siglec‐G‐GFP expression on different murine splenic immune cells (n = 3). E) Murine naïve CD8+ T cells were activated with BMDCs supplemented with 100 µg ml⁻¹ OVA and Siglec‐G expression was detected over time. F) Siglec‐10 expression on different human T cells from healthy donors (n = 6). G) Human CD8⁺ T cells were activated with Human CD3/CD28 stimulation beads and Siglec‐10 expression was detected over time. Representative data are shown from three independent experiments. The error bar represents mean ± SD. Statistical significance was determined by one‐way ANOVA with multiple comparisons (B,C,E–G). **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Figure 2

Siglec‐G(Siglec‐10) is coordinately expressed with exhausting markers in tumor‐infiltrating lymphocytes. A) Siglec‐10 expression on CD8⁺ T lymphocytes from healthy donor peripheral blood, tumor‐matched peripheral blood, and human CRC tumors (n = 3–7). B) PD‐1 expression on Siglec‐10⁻ and Siglec‐10⁺ tumor‐infiltrating CD8⁺ cells (n = 4). C) Plots showing four clusters of intratumor CD8⁺ T cells and the rank‐normalized expression of Siglec‐10 and Pdcd1, Havcr2, Lag3 on tumor‐infiltrating single CD8⁺ T cells from GEO DataSet (GSE231559). D) Heatmap of mean expression of T cell function‐associated genes on each cell cluster. E) Siglec‐G expression on splenocytes, dLNs, and tumor‐infiltrating CD8⁺ T cells in Siglecg^Gfp+ reporter mice inoculated with MC38 (n = 6). F,G) PD‐1(F) and Tim‐3(G) expression in the Siglec‐G⁻ and Siglec‐G⁺ tumor‐infiltrating CD8⁺ cells (n = 4,5), H–K) Intracellular IFN‐γ and TNF‐α co‐expression (H), Granzyme B (I), Perforin (J), and fast proliferative proportion (K) in the Siglec‐G⁻ and Siglec‐G⁺ tumor‐infiltrating CD8⁺ cells (n = 4,5). Representative data are shown from three independent experiments. The error bar represents mean ± SD. Statistical significance was determined by one‐way ANOVA with multiple comparisons (A and E) and paired Student's t‐test (B and F–K). *p < 0.05, **p < 0.01, and ****p < 0.0001.

Figure 3

Siglec‐G⁻ CD8+ cells exhibit stronger functional capacities and proliferation potential. A) Representative line chart showing Siglec‐G expression on transferred Siglecg^Gfp/+ reporter OT‐I cells after LM‐OVA infection (n = 3). B) Gene set variation analysis (GSVA) showing T cell‐associated function pathway from RNA‐seq data in isolated Siglec‐G⁻ and Siglec‐G⁺ OT‐I cells. C–G) Intracellular IFN‐γ and TNF‐α co‐expression (C), Granzyme B (D), Perforin (E), IL‐2 (F) production on Siglec‐G⁻ and Siglec‐G⁺ OT‐I cells and fast proliferative proportion (G) of these two groups (n = 3,4). H) Short‐term killing assay showing MC38‐OVA cell lysis after co‐cultured with sorted Siglec‐G⁻ and Siglec‐G⁺ OT‐I cells for 5 hours (E: T ratio, 10:1). I) Long‐term killing assay showing MC38‐OVA cells growth when co‐cultured with sorted Siglec‐G⁻ and Siglec‐G⁺ OT‐I cells over time (E: T ratio, 1:1), both H and I were measured by Incucyte S3 Live Cell Analysis Instrument. Representative data are shown from three independent experiments. The error bar represents mean ± SD. Statistical significance was determined by paired Student's t‐test (C–G), unpaired Student's t‐test (H), and two‐way ANOVA with multiple comparisons (I). *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 4

Siglec‐G deficiency enhances CD8⁺ T cell activation, cytokine production, and cytotoxicity in vitro. A) Naïve CD8⁺ OT‐I cells from Siglecg^−/− and WT littermate mice were cultured with BMDCs supplemented with 100 µg mL⁻¹ OVA. CD25 and CD69 levels determined by flow cytometry. B) CTV‐labelled naïve CD8⁺ T cells from Siglecg^−/− and WT littermate mice were cultured with BMDCs supplemented with 100 µg mL⁻¹ OVA and T cell proliferation (CTV dilution) determined by flow cytometry. C–F) IFN‐γ, IL‐2, Granzyme B and CD107a production on activated CD8⁺ T cells from Siglecg^−/− and WT littermate mice. G) MC38‐OVA cells were labeled by CTV previously and the percentage of Annexin V⁺ PI⁺ and Annexin V⁺ MC38‐OVA cells after co‐cultured with activated Siglecg^−/− and WT OT‐I CD8⁺ T cells for 8 hours. Representative data are shown from three independent experiments. The error bar represents mean ± SD. Statistical significance was determined by unpaired Student's t‐test (A–F) and one‐way ANOVA with multiple comparisons (G). *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Figure 5

Deletion of Siglec‐G promotes CD8⁺ T cell function and proliferation in vivo. A) The model of bone marrows chimera mice by reconstitution of Siglecg^−/− mice (CD45.2⁺) and WT mice (CD45.1⁺) into recipient mice. B–D) PD‐1 and Tim‐3 positive (B), IFN‐γ producing (C) and Edu positive cells (D) on tumor‐infiltrating CD8⁺ T cells from chimeric mice (n = 4–6). E‐G) OVA‐specific tetramer⁺ CD8⁺ T cells in peripheral blood, dLNs, and spleen (E,F), and IFN‐γ‐producing cells (G) from LM‐OVA infected chimera mice (n = 4–6). H) CFU in the spleen/liver was determined from Siglecg ^−/− or WT OT‐I transferred recipient mice which i.v. infected with virulent LM‐OVA (n = 3). Representative data are shown from three independent experiments. The error bar represents mean ± SD. Statistical significance was determined by one‐way ANOVA with multiple comparisons (H) and paired Student's t‐test (B–G). *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 6

Siglec‐G enhances oxidative phosphorylation of CD8⁺ T cells via SHP‐2. A) KEGG pathway analysis of upregulated genes in Siglec‐G⁺ OT‐I cells compared to Siglec‐G⁻ OT‐I cells by RNA‐Seq (q‐value ≤ 0.05, fold change ≥ 2). B) GSEA analysis of OXPHOS‐related genes. C) Seahorse analysis of mitochondrial OCR of isolated Siglec‐G⁻ and Siglec‐G⁺ OT‐I cells, treated with 10 µm SHP2 inhibitor SHP099 for 2 h or not. D) Heatmap of downregulated genes associated with glycolysis in Siglec‐G⁻ CD8⁺ cells relative to that in Siglec‐G⁺ cells. E) Immunoblotting analysis of the indicated proteins in Siglecg^−/− and WT CD8⁺ T cells activated by BMDCs. F) LDH activity in Siglec‐G⁺ and Siglec‐G⁻ OT‐I cells, or treated with 10 µm SHP2 inhibitor SHP099 or AKT inhibitor MK‐2206. G) Seahorse analysis of glycolytic proton efflux rate (glycoPER) of isolated Siglec‐G⁻ and Siglec‐G⁺ OT‐I cells, treated with 10 µm SHP099 or not for 2 h. H) The percentage of Annexin V⁺ MC38‐OVA cells co‐cultured with activated Siglecg^−/− or WT OT‐I CD8⁺ T cells, treated with glycolysis inhibitor 2 mm 2‐DG or not for 4 h (E:T ratio, 10:1). I) Immunoblotting showing phosphorylated and total SHP1 and SHP2 in activated Siglecg^−/− and WT CD8⁺ T cells. J) The percentage of Annexin V⁺ MC38‐OVA cells co‐cultured with activated Siglecg^−/− or WT OT‐I CD8⁺ T cells, treated with SHP099 or not (E:T ratio 10:1). K,L) Immunoblotting analysis of the indicated proteins in Siglecg^−/− and WT CD8⁺ T cells activated as indicated. Representative data are shown from three independent experiments. The error bar represents mean ± SD. Statistical significance was determined by one‐way ANOVA with multiple comparisons. *p < 0.05, ***p < 0.001, and ****p < 0.0001.

Figure 7

Siglec‐G deletion in CD8⁺ T cells enhances anti‐tumor immunity. A) CD8⁺ T cells from the spleen of Siglecg ^−/− and WT OT‐I littermate mice (CD45.2⁺) were activated as indicated and transferred to tumor‐bearing mice (CD45.1⁺). Intratumor‐transferred OT‐I CD8⁺ T cells were analyzed 12 days later. B) Tumor growth of MC38‐OVA‐bearing mice were measured every 2–3 days and tumor were dissected at last day (n = 5). C) Survival analysis of MC38‐OVA‐bearing mice. D) The proportions of adoptively transferred Siglecg ^−/− and WT OT‐I CD8⁺ T cells in total intratumor CD8⁺ T cells (n = 4). E) Quantification of Edu positive cells in intratumor adoptively transferred Siglecg ^−/− and WT OT‐I CD8⁺ T cells (n = 4). F) Representative immunofluorescence staining of CD8 (red) or Tunel (green) in tumor. Scale bar, 100 µm (magnified: 33 µm). G,H) IFN‐γ and TNF‐α expression in adoptively transferred Siglecg ^−/− and WT OT‐I CD8⁺ T cells from tumors (G) and dLNs (H) (n = 4). I,J) MC38‐hCLDN18.2 tumor cells were injected s.c. into CD45.1⁺ mice followed by the adoptive transfer of 2 × 10⁶ CAR T cells generated from Siglecg ^−/− and WT splenocytes (CD45.2⁺). Tumor growth of MC38‐hCLDN18.2‐bearing mice were measured every 2–3 days and tumor were dissected at last day (n = 7). K) Survival analysis of MC38‐hCLDN18.2‐bearing mice. L) The proportions of adoptively transferred Siglecg ^−/− and WT CAR T cells in total intratumor CD8⁺ T cells (n = 4). M) IFN‐γ expression in adoptively transferred Siglecg ^−/− and WT CAR T cells tumor infiltrates (n = 4). N) The mechanism of Siglec‐G suppressing CTL responses through SHP2‐PI3K‐AKT‐mTOR mediated metabolic rewiring. Representative data are shown from three independent experiments. The error bar represents mean ± SD. Statistical significance was determined by two‐way ANOVA with multiple comparisons (B and I), a Log‐rank (Mantel–Cox) test (C and K), and unpaired Student's t‐test (D, E, G,H, J, L, and M). **p < 0.01, ***p < 0.001, and ****p < 0.0001.