Inhibiting SLAMF8 modulates tumor-associated macrophages and restores CD8+ T cell-mediated antitumor immunity in colorectal cancer.

Introduction
Colorectal cancer (CRC) ranks a leading cause of cancer mortality worldwide.1 Only approximately 5% of advanced CRC patients with the microsatellite instability-high (MSI-H) phenotype can benefit from immune checkpoint inhibitors (ICIs).2,3 Notably, the highly immunosuppressive tumour microenvironment (TME) is a major factor limiting the efficacy of immunotherapy.4 Therefore, exploring the immunoregulatory factors present in the TME and identifying new immunotherapy targets have become new strategies for research on CRC treatment.
Tumour-associated macrophages (TAMs) represent a critical component of the immunosuppressive TME, making them a focal point in the current research landscape of immunotherapy targets. TAMs are the most abundant infiltrating immune cells in the tumour immune microenvironment. It is generally believed that macrophage infiltration is associated with a poor prognosis for solid tumours.5,6 However, the conclusions regarding the association between the abundance of macrophages and the prognosis of CRC are inconsistent. One study encompassing 168 CRC patients found that progression-free and overall survival rates were significantly worse in the high-TAM density group compared to the low-TAM group.7 In contrast, another study demonstrated that a dense macrophage infiltration at the tumour front was a positive prognostic factor in a cohort of 446 CRC patients.8 These inconsistent conclusions may be attributable to the profound plasticity and heterogeneity of macrophages in the TME.9,10 The M1 and M2 phenotypes represent two well-recognised subtypes of macrophages: M2-like macrophages exhibit an immunosuppressive phenotype that promotes tumour progression and immunotherapy resistance, whereas M1-like macrophages have been demonstrated to have antitumor features.11–13 There is complicated crosstalk between tumour cells and immune cells within the TME, and the regulatory factors generated during this interaction facilitate the M2 polarisation of TAMs.14,15 Moreover, immunosuppressive TAMs have been shown to directly or indirectly inhibit the activity of CD8+ T cells.12,16 Therefore, identifying key immune factors that regulate the function of TAMs and reprogramming TAMs to exert antitumor functions have become new research hot spots in tumour immunotherapy.
SLAMF8 is a type I transmembrane protein and a member of the lymphocyte activation signalling molecule (SLAM) family;17 it is reported to be expressed predominantly in certain populations of professional antigen-presenting cells (APCs), activated monocytes, and dendritic cells (DCs).18 Research has revealed that SLAMF8 is involved in regulating macrophage activation and function19,20 and plays an important role in inflammatory and autoimmune conditions. Its roles extend from endotoxin-induced liver inflammation21 to autoimmune diseases such as rheumatoid arthritis22,23 and inflammatory bowel disease.24 New studies have shown that high SLAMF8 expression is associated with a poor prognosis and high expression of classic checkpoints in tumours,25,26 suggesting its potential as a novel immune checkpoint molecule. Our previous research revealed that SLAMF8 expression predicts the efficacy of anti-PD1 immunotherapy in gastrointestinal cancers, suggesting that SLAMF8 may play a role in modulating the malignant TME.26,27 However, the specific regulatory functions and precise mechanisms of SLAMF8 within the CRC TME remain to be fully elucidated, and the rationality of SLAMF8 as a novel target for CRC immunotherapy warrants further in-depth investigation.
Herein, we demonstrated that high macrophage-specific SLAMF8 expression is associated with poor prognosis in CRC patients. Additionally, SLAMF8 preserves the immunosuppressive phenotype of macrophages and T cell dysfunction through the PI3K/AKT and JAK/STAT3 pathways. Inhibition of SLAMF8 effectively promoted remodelling of the tumour immunosuppressive microenvironment and sensitised CRC to anti-PD1 therapy in vivo. In conclusion, targeting SLAMF8 represents a promising strategy for CRC immunotherapy, revealing a previously unrecognised role for SLAMF8 as a modulator of antitumor immunity and its implications in tumour immunotherapy.

Materials and methods

Patients and specimens
For the analysis of public databases, we downloaded the Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/) cohort with accession number GSE17538, which includes RNA sequencing and clinical data for 232 CRC patients. The data were normalised using the “limma” package in R software, and the infiltration of immune cells was assessed employing the “cibersort” package, The optimal cut-off values for SLAMF8 mRNA and macrophage infiltration abundance were determined using the “survminer” R package. Additionally, RNA sequencing data from 453 individuals diagnosed with colon adenocarcinoma (the TCGA-COAD cohort) were acquired from the UCSC Xena (https://xena.ucsc.edu/) database. We retrospectively analysed paraffin-embedded tumour tissue samples that were obtained from the department of pathology archives of Nanjing Drum Tower Hospital. The cohort included 110 patients with histologically confirmed CRC, including both colon and rectal carcinomas, at stage I to IV, who underwent surgery at Drum Tower Hospital between April 2017 and November 2021. Signed informed consents were obtained from all the patients prior to sample collection. Experiments involving human samples were conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Nanjing Drum Tower Hospital (2023-347-01).

Cell lines
Mouse CRC cell lines (CT26 and MC38), the mouse macrophage line RAW264.7, the human monocyte‒macrophage line THP-1, and the human CRC cell line SW480 were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai). CT26, MC38, THP-1, and SW480 cells were cultured in Roswell Park Memorial Institute (RPMI)-1640 medium (Corning), while RAW264.7 cells were cultured in Dulbecco's modified Eagle's medium (DMEM, Biosharp). Both media were supplemented with 10% foetal bovine serum (FBS, Gibco) and 1% penicillin‒streptomycin solution (Beyotime, #C0222). All cell lines used in this study were authenticated by short tandem repeat (STR) profiling and confirmed to be free of mycoplasma contamination prior to experimentation.

Mice
Six- to eight-week-old C57BL/6 and BALB/c female mice were purchased from GemPharmatech (Nanjing, China) and maintained in the specific pathogen-free (SPF) Laboratory of the Animal Centre of the Affiliated Nanjing Drum Tower Hospital of Nanjing University Medical School (Nanjing, China). The animal experiments were conducted from March 2024 to August 2025, in accordance with the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines, and approved by the Animal Ethics Committee of Nanjing Drum Tower Hospital (2024AE01067).

Primary macrophage isolation
C57BL/6 mice were euthanized, and their femurs and tibias were aseptically irrigated with PBS. The bone marrow cells were subsequently isolated via red blood cell lysis buffer and cultured in DMEM supplemented with 10% FBS and 20 ng/mL M-CSF (PeproTech, #315-02) for 5 days to induce the formation of marrow-derived macrophages (BMDMs).
Peripheral blood from healthy donors was diluted and carefully layered onto Ficoll-Hypaque solution (TBD, #LTS1077). After density gradient centrifugation at 800 × g for 20 min at room temperature, the peripheral blood mononuclear cell (PBMC)-enriched intermediate layer was collected. Isolated PBMCs were washed with PBS by centrifugation at 300 × g for 10 min to remove residual Ficoll. CD14⁺ monocytes were further purified from PBMCs using a human CD14⁺ cell separation kit (RWD Life Science, #K1204) following the manufacturer’s protocol. The isolated monocytes were then cultured in RPMI-1640 medium supplemented with 10% FBS and 20 ng/mL M-CSF (PeproTech, #300-25) for 6 days to differentiate into macrophages. Peripheral blood samples were collected from healthy donors between July and August 2025 with written informed consent prior to blood collection, following the Declaration of Helsinki and approved by the Ethics Committee of Nanjing Drum Tower Hospital (2025-0085-02).

Induction of M2 macrophages
For the RAW264.7 cell line, M2 macrophages were induced by treating the cells with 20 ng/mL mouse IL-4 (PeproTech, #214-14) for 48 hours. Isolated CD14+ monocytes were induced to differentiate with M-CSF for 6 days, for the THP-1 cell line, M0 macrophages were first induced by treating the cells with 100 ng/mL phorbol 12-myristate 13-acetate (PMA) (MCE, #HY-18739) for 48 hours, both cell populations were subsequently induced to polarise into M2 macrophages by stimulation with 20 ng/mL human IL-4 (PeproTech, #200-04) and human IL-13 (PeproTech, #200-13) for an additional 48 hours. To generate mouse BMDMs, differentiated bone marrow cells were collected and stimulated with mouse IL-4 (PeproTech, #214-14) for 48 hours to induce M2 macrophages.

Construction of tumour-conditioned medium (TCM) and tumour-associated macrophages (TAMs) in vitro
CRC tumour cells (CT26, MC38, and SW480) were cultured until they reached 60% confluence in the culture dish. After two washes with PBS, medium containing 1% serum was added, and the cells were further cultured for 48 hours. The supernatant was then collected, centrifuged, and filtered through a 0.22 µm filter to obtain TCM. Macrophages (RAW264.7, BMDMs, CD14+ monocytes and THP-1) were subsequently cultured in a 1:1 mixture of TCM and conventional medium for 24 hours to facilitate their differentiation into TAMs.

Small-interfering RNA (siRNA) transfection
To knockdown SLAMF8 expression, siRNA targeting SLAMF8 was synthesised with the sequence 5'-AUUCUCUUUCUGGUCUGAATT-3'. In accordance with the manufacturer's instructions, BMDMs were transfected with siSLAMF8 or a negative control siRNA using a transfection reagent (BIO-Generating, #11042). The following experiments were conducted 48 hours post-transfection.

Stable cell line generation
Lentivirus for SLAMF8 overexpression and control lentivirus for infecting RAW264.7 cells were designed by Vigene Biosciences. Stable cell lines were selected 48 hours post infection with 5 µg/mL puromycin (Beyotime, #ST551). Lentivirus for SLAMF8 overexpression and control lentivirus for infecting THP-1 cells and CD14+ monocytes were designed by GenePharma Co., Ltd. Stable cell lines were selected 72 hours post infection with 2 µg/mL puromycin.

Isolation of CD8+ T cells and co-culture assay
CD8+ T cells were isolated from the splenocytes of 6-week-old C57BL/6 mice following the manufacturer's instructions for the CD8a+ T Cell Isolation Kit (Miltenyi Biotec, #130-104-075). Isolated CD8+ T cells were subsequently stimulated with anti-CD3 antibodies (eBioscience, #16-0031-82) precoated in a 24-well plate. The culture medium consisted of RPMI 1640 supplemented with 20 ng/mL IL-2 (PeproTech, #212-12) and 5 µg/mL anti-mouse CD28 antibodies (eBioscience, #16-0281-82). CD8+ T cells were subsequently co-cultured with BMDMs that had undergone different treatments at a ratio of 2:1. After 48 hours of co-culture, the cells were harvested via centrifugation, and the activation and function of the CD8+ T cells were assessed using flow cytometry.

RNA extraction and quantitative real-time PCR (qRT‒PCR)
In accordance with the manufacturer's procedure, TRIzol reagent (Ambion) was used to extract total RNA from cells and tumour tissue. cDNA was synthesised using HiScript III RT SuperMix for qPCR (Vazyme, #R323-01). Each cDNA sample was amplified via Taq Pro Universal SYBR qPCR Master Mix (Vazyme, #Q712-02) on an ABI QuantStudio 7 Flex real-time PCR system (Applied Biosystems), and GAPDH served as an internal control for sample normalisation. The experiment was performed in triplicate. The fold changes in the mRNA expression of these genes were calculated using the 2−ΔΔCt method for relative expression quantification.The primers used in this study are listed in Supplemental Table S3.

RNA-seq analysis
RNA extraction was conducted on BMDM–TAMs samples collected from both the control group and the SLAMF8 overexpression group (n = 4 per group). Subsequently, RNA-Seq analysis was performed by Genewiz (Suzhou, China). In brief, total RNA of each sample was extracted using TRIzol Reagent following the manufacture’s protocol.RNA integrity was assessed using an Agilent 2100/2200 Bioanalyzer (Agilent Technologies). Libraries were constructed using 1 μg of total RNA and then pooled. Following the manufacturer's protocols, the pooled libraries were loaded onto an Illumina HiSeq/Novaseq instrument for 2×150 bp paired-end sequencing. Image analysis and base calling were performed using HiSeq Control Software (HCS) + OLB + gappipeline-1.6 (Illumina).Adaptor sequences, PCR primers, and low-quality bases (quality score < 20) were trimmed from the raw FASTQ files using Cutadapt (v1.9.1) to generate high-quality clean reads, which were then aligned to the mouse reference genome mm10. Differentially expressed genes were identified using the DESeq2 package in R software with the threshold criteria log2 (fold change) ≥1 and adjusted P value < 0.05. To elucidate the biological functions and pathways associated with the identified DEGs, we performed Kyoto Encyclopaedia of Genes and Genomes (KEGG) pathway enrichment analysis using the DAVID online database (https://davidbioinformatics.nih.gov/).

Western blotting
The cells were lysed in RIPA lysis buffer (Beyotime, P0013B) supplemented with protease and phosphatase inhibitors (Beyotime, #P1045). The protein concentration was quantified via the BCA protein assay. Proteins were separated via 10% sodium dodecyl sulphate‒polyacrylamide gel electrophoresis (SDS‒PAGE) and subsequently transferred to 0.2 µm polyvinylidene difluoride (PVDF) membranes (Millipore, Billerica). The membranes were blocked with QuickBlock Western sealing solution (Beyotime, P0252) for 40 minutes and then incubated with primary antibodies overnight at 4 °C. The membranes were incubated with the corresponding secondary antibodies for 1 hour at room temperature after washing. The protein bands were visualised and analysed with a gel image analysis system (Tanon).The antibodies used in this study are listed in Supplemental Table S2.

Flow cytometry (FACS)
The cells were incubated with an Fc receptor-blocking solution (BioLegend, #101302) at room temperature for 10 minutes to minimise nonspecific binding. Surface staining was conducted at 4 °C for 30 minutes. Intracellular staining was performed following membrane permeabilization when necessary. For intracellular cytokine staining, the cells were stimulated with a cell stimulation cocktail (eBioscience, #00−4975−03) for 4 hours. After surface staining, the cells were fixed with fixation buffer (eBioscience, #00−8222−49) and then subjected to membrane permeabilization (eBioscience, #00−8333−56). Antibody staining was performed at 4 °C for 30 minutes, followed by washing and flow cytometry analysis. The flow cytometry results were analysed with FlowJo software (version 10.8.1).The antibodies used in this study are listed in Supplemental Table S2.

Immunohistochemistry and immunofluorescence
Immunohistochemistry (IHC) and immunofluorescence analysis were performed on paraffin-embedded sections of human CRC tissue. The slides were deparaffinized in xylene and rehydrated using a graded series of alcohols. The slides were treated with citrate buffer at pH 6.0 for antigen retrieval. Endogenous peroxidase activity was quenched, and the samples were subsequently blocked with 5% bovine serum albumin (BSA). The membranes were incubated with primary antibodies overnight at 4 °C. The slides for IHC were then incubated with the corresponding secondary antibody followed by visualisation with diaminobenzidine (DAB) and counterstaining with hematoxylin. For immunofluorescence, after incubation with primary and secondary antibodies, the slides were stained with DAPI and mounted for imaging under a fluorescence microscope (Leica). The mean density was quantified utilising ImageJ software. Statistical analysis was conducted with GraphPad Prism software (version 8.0.2).

Tumour growth and treatment
CRC cells (MC38 cells or CT26 cells) were used to establish subcutaneous tumour xenograft models, and single-cell suspension containing 0.5–1.0 ∗ 106 CRC tumour cells (CT26 or MC38 cells) was administered to the mice via subcutaneous (s.c.) injection to induce the formation of solid tumours. The tumour dimensions were measured with calipers, and the tumour volume was calculated using the formula V = L ∗ W2/2. Once the tumours reached a predetermined size, the mice were randomised into groups according to the experimental design for subsequent treatments. Euthanasia was performed by cervical dislocation when the tumour volume reached 1500 mm³ or the maximum tumour diameter reached 20 mm.
To overcome the poor stability and short half-life of unmodified siRNA in vivo, cholesterol-modified siRNAs targeting SLAMF8 (siSLAMF8) or control siRNA (siNC) (Hanyi Biotechnology Co., Ltd) were administered via intratumoral (i.t.) injection at a dose of 2.5 nmol every three days to inhibit SLAMF8 expression. For depletion of CD8+ T cells, the mice were administered intraperitoneal (i.p.) injections of an anti-CD8 monoclonal antibody (BioXcell, #BE0061) every three days starting from the initiation of treatment. For macrophage depletion, anti-CSF1R (BioXcell, #BE0213) or clodronate liposomes (Yeasen, 40337ES10) were used by intraperitoneal (i.p.) injection to deplete TAMs within TME. Balb/c mice were injected i.p. with anti-CSF1R at 300 μg per mouse one day before and every three days after tumour cell inoculation, or inject i.p.with clodronate liposomes or PBS liposomes (Yeasen, 40338ES10) two days before tumour cell inoculation and then every three days until the endpoint. Anti-PD1 monoclonal antibody treatment involved the i.p. administration of an anti-mouse PD1 monoclonal antibody (BioXcell, #BE0146) or an IgG monoclonal antibody (BioXcell, #BE0089) at a dose of 200 μg every three days. At the indicated times, mice were euthanized using cervical dislocation method. To detect changes in immune cell populations within the TME, solid tumours were collected and examined via flow cytometry and qRT‒PCR. Additionally, the samples were fixed with paraformaldehyde and embedded in paraffin for IHC and IF analysis.

Statistical analysis
Statistical analyses were conducted using RStudio software (version 4.3.1), GraphPad Prism (version 8.0.2), and SPSS Statistics (version 26.0). Two-tailed Student’s t tests were used for comparisons between two groups, whereas one-way or two-way ANOVA was used for comparisons among multiple groups. Survival curves were constructed using the Kaplan‒Meier method, and statistical significance was assessed via the log-rank test. Independent prognostic factors were identified using Cox proportional hazards regression models. The data are shown as the means with SEMs; ns, not significant; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Results

Macrophage-specific SLAMF8 expression is significantly associated with the prognosis of CRC patients
The Kaplan‒Meier Plotter online tool28 was used to evaluate the relationship between SLAMF8 expression and the prognosis of various tumours. The results indicated that the upregulation of SLAMF8 is associated with tumour progression in several cancer types including gastric cancer and lung cancer (Supplemental Figure S1). To assess the correlation between SLAMF8 expression level and the clinical outcomes of CRC patients, we first used data from the GSE17538 cohort. Compared with those in the SLAMF8 low-expression group, the disease-free survival (DFS) and overall survival (OS) of patients in the SLAMF8 high-expression group were markedly shorter (Figure 1A-B). In contrast, the abundance of macrophages did not show such a correlation with OS (Figure 1C-D). Notably, patients with high SLAMF8 expression and enrichment of TAMs presented the poorest clinical outcomes in terms of OS and DFS (Figure 1E-F).
We subsequently evaluated the expression of SLAMF8 and macrophage infiltration in human CRC tissues from the Drum Tower Hospital cohort using IHC staining., and then divided patients with CRC into SLAMF8-high or SLAMF8-low group based on the median value of SLAMF8 mean optical density (0.0096). Likewise, patients were classified into CD68-high and CD68-low group according to the median number of CD68 + cells (47 cells/high-power field) in the whole cohort (Figure 1G-H). The characteristics of CRC patients from Drum Tower Hospital Cohort are shown in Supplementary Table S1. Briefly, the median age at diagnosis was 59.8 years, both colon (63.6%) and rectal (36.4%) carcinomas were represented, the median follow-up time was 41 months (IQR:30.0−57.0). Compared with that of patients in the SLAMF8 low-expression group, the OS of patients in the SLAMF8 high-expression group was significantly shorter (Figure 1I). In contrast, no significant difference in survival rates was observed between the high CD68 expression group and the low CD68 expression group (Figure 1J). Furthermore, our analysis suggested that CRC patients with high-co-expression of SLAMF8 and CD68 had the poorest prognosis (Figure 1K).
To identify prognosis risk factors in CRC patients, we initially screened and assessed the impact of individual factors using univariate Cox regression. Variables with statistical significance (P < 0.05) in the univariate analysis, including TNM stage, CEA, CA199, and macrophage-specific SLAMF8 expression, were further included in the multivariate Cox regression analysis. The results confirmed that high-co-expression levels of SLAMF8 and CD68 could serve as independent risk factors for CRC prognosis (Figure 1L). In conclusion, high expression of macrophage-specific SLAMF8 is a crucial adverse prognostic factor for CRC patients.

SLAMF8 is predominantly expressed on macrophages and facilitates their polarisation toward the M2 phenotype.
Using the TNMplot web tool,29 we observed that, compared with that in normal tissues, SLAMF8 was upregulated in multiple tumours, including CRC (Supplemental Figure S2A). To clarify the expression of SLAMF8 in the TME, we utilised the Tumour Immune Single-Cell Hub (TISCH) database and discovered that SLAMF8 was predominantly expressed on TAMs in both human and mouse CRC tumours as well as various other tumours (Figure 2A, Supplemental Figure S2B). We further verified this conclusion by performing immunofluorescence co-staining of SLAMF8 and markers of macrophages, DCs, T cells, and B cells in human CRC tissues from our centre (Figure 2B).
To further explore the effect of SLAMF8 on macrophage polarisation, we induced the M2 polarisation of RAW264.7 cells, the human monocytic cell line THP-1, CD14+ monocytes and BMDMs via the addition of different cytokines (Figure 2C). We used lentiviruses to construct THP-1, CD14+ monocytes and RAW264.7 cell lines overexpressing SLAMF8 and added cytokines to induce their polarisation into M2 macrophages. The results of qRT‒PCR demonstrated that the expression of M2 markers (ARG1, TGFB, and IL10) in the SLAMF8 overexpression group was significantly increased (Figure 2D, Supplemental Figure S3A, S3C), and flow cytometry (FACS) indicated that the expression of CD206 and CD163 was elevated (Figure 2E, Supplemental Figure S3B, S3D). We used SLAMF8-specific siRNA (siSLAMF8) to knockdown the expression of SLAMF8 in BMDMs and polarised them to the M2 phenotype by adding IL4. The results of qRT-PCR and FACS showed that the expression of M2 markers decreased at both the mRNA and protein levels after SLAMF8 knockdown and M2 polarisation was significantly rescued (Figure 2F-G). Our results indicate that SLAMF8 is a significant regulator of M2 macrophage polarisation.

SLAMF8 remodels the immunophenotype of TAM-like macrophages
In solid tumours, macrophages are referred to as TAMs; TAMs are polarised by intricate signals within the TME and acquire an immune phenotype distinct from that of macrophages in healthy tissues.30 To better simulate the influence of tumour cells on macrophages in the TME, we used BMDMs to confirm that suppressing SLAMF8 expression can reverse the immunosuppressive phenotype of TAM-like macrophages. We prepared TCM from the supernatant of MC38 murine CRC cells (MC38-TCM) and treated macrophages with varing SLAMF8 expression levels with the TCM to induce their polarisation in a TAM-like direction (Figure 3A). The expression of SLAMF8 was increased in these TAM-like macrophages. Interestingly, compared with those in the control group, the mRNA levels of M1 macrophages (IL6 and TNFα) in the knockdown group were elevated, whereas the mRNA expression of M2 macrophages (ARG1 and TGFB) was decreased after TCM induction (Figure 3B). The FACS results demonstrated that the induction of the M2 phenotype (CD206, CD163) in macrophages by TCM treatment was significantly attenuated by SLAMF8 inhibition, whereas the induction of the M1 phenotype (CD80, CD86) was further augmented (Figure 3C, Supplemental Figure S4C), revealing the crucial role of SLAMF8 in maintaining the immunosuppressive phenotype of TAMs. Consistent results were obtained when SLAMF8-overexpressing RAW264.7, CD14 + monocytes or THP-1 macrophages were treated with CT26-TCM or SW480-TCM, respectively (Figure 3D-E, Supplemental Figure S4A-B, Supplemental Figure S5A-B).
A considerable amount of evidence consistently indicates that TAMs promote tumour progression by suppressing CD8+ T cell-mediated tumour immunity.12,31,32 We performed an in vitro co-culture assay using TAMs and CD8+ T cells isolated from splenocytes (Figure 3F). The results revealed that interfering with SLAMF8 expression reduced PD1 expression and increased GZMB expression in co-cultured CD8+ T cells, suggesting that SLAMF8 deficiency decreased the proportion of exhausted CD8+ T cells (Figure 3G).
In conclusion, these results indicate that macrophages exhibiting elevated SLAMF8 expression represent a subtype characterised by a highly immunosuppressive phenotype within the TME of CRC and that knocking down SLAMF8 expression remodels TAMs into an M1-like and immunostimulatory phenotype that supports CD8+ T cell function.

SLAMF8 induces the immunosuppressive phenotype of macrophages via the PI3K/AKT and JAK/STAT3 pathways
To elucidate the molecular mechanism by which SLAMF8 remodels the phenotype of TAMs, we infected BMDMs with lentiviruses overexpressing SLAMF8 or control lentiviruses and treated them with MC38-TCM to induce TAM-like phenotypes (Figure 4A). RNA-seq analysis revealed that the majority of genes associated with M2 markers were upregulated in the overexpression group, whereas M1 markers and T cell activation-related genes were downregulated in this group (Figure 4B-C).
Next, enrichment analysis of differentially expressed genes in the RNA-seq and COAD TCGA datasets was conducted according to the KEGG pathway database via DAVID online, and we found that the PI3K/AKT and JAK/STAT pathways were enriched in SLAMF8-overexpressing macrophages and CRC patients with high SLAMF8 expression (Figure 4D-E). PI3K/AKT and JAK/STAT3 are major targets for TAM repolarization and cancer immunotherapy.33,34 Therefore, we hypothesise that SLAMF8 preserves the immunosuppressive phenotype of TAMs by activating the PI3K/AKT and STAT3 pathways. Western blotting experiments demonstrated that the phosphorylation levels of key molecules in the PI3K/AKT and JAK/STAT3 signalling pathways were significantly increased in SLAMF8-overexpressing in both BMDMs and RAW264.7 macrophages (Figure 4F).
A PI3K inhibitor (LY294002) or STAT3 inhibitor (Stattic) was subsequently used to pretreat SLAMF8-overexpressing BMDMs, which were then induced via TCM treatment for flow cytometry detection and in vitro co-culture experiments. Both inhibitors reversed the immunosuppressive phenotype of TAMs (Figure 4G, Supplemental Figure S6A-B) and dysfunction of CD8+ T cells (Figure 4H). Taken together, our data suggest that SLAMF8 achieves TAM phenotype conversion by activating the PI3K/AKT and JAK/STAT3 pathways within macrophages.

Inhibition of SLAMF8 suppressed CRC growth and remodelled the TME
According to our results, SLAMF8 reprograms the TME by facilitating the transformation of TAMs into an immunosuppressive phenotype and subsequently inhibiting CD8+ T cells. We sought to determine whether inhibiting SLAMF8 can alleviate the immunosuppressive TME. We used in vivo-optimised siRNAs targeting SLAMF8 (siSLAMF8) or control siRNA (siNC) to treat subcutaneous MC38 or CT26 tumours in mice (Figure 5A,C). The results showed that siSLAMF8 treatment significantly delayed CRC progression (Figure 5B,D). Flow cytometry analysis of TAMs and CD8+ T cells within the tumours disclosed that the infiltration of M1-TAMs (CD86+ F4/80 +) and cytotoxic CD8+ T cells (IFNγ + GZMB + CD8+ T cells) increased in tumours treated with siSLAMF8 (Figure 5E-F, Supplemental Figure S7, S8A-B), whereas the infiltration of M2-TAMs (CD206 + F4/80 + ) and exhausted CD8+ T cells (PD1 + LAG3 + CD8+ T cells) decreased (Figure 5G-H, Supplemental Figure S8C-D). qRT‒PCR and immunohistochemical staining further validated the role of siSLAMF8 in eliminating immunosuppressive TAMs and exhausted CD8+ T cells (Figure 5I-J).
To ascertain whether the antitumor efficacy of SLAMF8 inhibition in CRC relies on immune cells in the TME, we first depleted CD8+ T cells in mice using CD8 antibodies (Supplemental Figure S8G, Figure 5K). We discovered that the therapeutic effect of siSLAMF8 was attenuated after CD8+ T cell depletion; however, tumour growth was still decelerated in the depletion group after siSLAMF8 treatment (Figure 5L). FACS analysis revealed an increase in the infiltration of M1-TAMs with an immune-stimulatory phenotype, whereas the abundance of M2-TAMs with an immunosuppressive phenotype decreased (Figure 5M-N). Next, we used anti-CSF1R (Figure 5O) or clodronate liposomes (Supplemental Figure S8E) to deplete macrophages in mice, and the results indicated that the growth-inhibitory effect of SLAMF8 on tumours vanished after macrophage depletion (Figure 5P, Supplemental Figure S8F, S8H). These results indicate that the in vivo inhibition of SLAMF8 in delaying CRC tumour growth is dependent on the presence of macrophages.

Inhibition of SLAMF8 enhances the anti-PD1 effect in the treatment of CRC
Our previous study suggested that SLAMF8 expression is significantly correlated with PD-L1 expression in CRC patients26 and successfully established tumour models resistant and sensitive to anti-PD1 treatment.27 In this study, we used the CT26 tumour model, which is relatively resistant to anti-PD1 therapy, and the MC38 tumour model, which is sensitive to anti-PD1 therapy, to explore whether the inhibition of SLAMF8 and anti-PD1 therapy has synergistic effects. First, we treated MC38 tumour-bearing mice with SLAMF8-specific siRNA combined with an anti-PD−1 monoclonal antibody (mAb) and found that both the PD−1 mAb and siSLAMF8 alone inhibited tumour growth and that the combination of the two had the most significant inhibitory effect on CRC growth (Figure 6A-B). The above combination therapy also achieved significant efficacy in CT26 models that were relatively resistant to anti-PD1 therapy (Figure 6C‒D). Compared with monotherapy, combination therapy significantly increased the proportion of cytotoxic T cells (Figure 6E) and decreased the co-expression rate of CD8+ T-cell exhaustion markers (Figure 6F-G). IF analysis showed that the combination therapy markedly promoted the infiltration of iNOS + M1 macrophages while decreasing the infiltration of Arg1 + M2 macrophages (Figure 6H, Supplemental Figure S9). Taken together, our results highlight that targeting SLAMF8 enhances the efficacy of anti-PD1 therapy, providing a new combination treatment strategy for immunotherapy-resistant populations.

Discussion
The high incidence and mortality rates of CRC indicate an urgent need to identify novel and effective therapeutic targets in clinical practice. TAMs are characterised by tumour-derived regulators within TME, exhibit immunosuppressive pro-tumour phenotypes that facilitate tumour progression and hinder antitumor response by secreting immunosuppressive cytokines and inhibiting T cell function;32,35 therefore, they are considered critical targets for antitumor immunotherapy. This study revealed that SLAMF8 is predominantly expressed in macrophages rather than in tumour cells or other immune cells, as evidenced by the TISCH database and immunofluorescence detection. Moreover, we found that CRC with high macrophage-specific SLAMF8 expression had a poor prognosis. Further investigation confirmed that SLAMF8 significantly contributes to the M2 polarisation of macrophages and the immunosuppressive phenotype of TAMs; therefore, we speculate that SLAMF8 represents a promising target for CRC antitumor therapy focused on TAMs.
SLAMs, a group of glycoproteins expressed on activated lymphocytes and antigen-presenting cells, has been demonstrated to co-regulate antigen-driven T-cell responses.36 Emerging evidence suggests that the SLAM family plays a critical role in modulating immune cell activation and function within the TME, acting as inhibitory checkpoints that promote tumour progression. Agresta et al. discovered that the upregulation of SLAMF4 is correlated with the immunosuppressive function of immune cells such as CD8+ T cells and DCs, suggesting that SLAMF4 is a novel therapeutic target for head and neck squamous cell carcinoma (HNSCC).37 SLAMF3 and SLAMF4 have been identified as immune checkpoints that act as “don't eat me” receptors, inhibiting macrophage phagocytosis during phagocytosis.38 Yigit et al. reported that anti-SLAMF6 treatment reversed CD8+ T cell exhaustion and slowed the progression of melanoma and haematological malignancies.39 The SLAMF7 monoclonal antibody elotuzumab has been used to treat haematological malignancies by promoting the antitumor effects of NK cells, cytotoxic T cells and macrophages in the TME.40,41 In our previous studies, we reported a significant association between the expression of SLAMF8 and that of classic immune checkpoint molecules, including PD-L1, CTLA4, and TIM3. Herein, in vitro co-culture experiments demonstrated that the inhibition of SLAMF8 reversed the M2 phenotype of TAMs and exhaustion of CD8+ T cells. Employing murine subcutaneous tumour models, we proved that SLAMF8 inhibition effectively attenuated CRC tumour growth and increased the proportions of antitumor macrophages and cytotoxic CD8+ T cells. Collectively, our findings indicate that SLAMF8 inhibition improves the intratumor immunosuppressive TME and restores CD8+ T-cell function.
The PI3K/AKT signalling pathway plays a crucial role in tumour progression and immune cell regulation.42 Studies have confirmed that PI3K orchestrates the balance between immune activation and repression during inflammation and cancer. Activating the PI3K/AKT pathway promotes the expression of Arg1, TGF-β, and IL-10 while inhibiting the expression of IL-12, IFN-γ, and iNOS in macrophages, thereby suppressing T cell activation and fostering an immunosuppressive microenvironment. PI3Kγ inhibitors have been utilised to repolarize TAMs, leading to enhanced T cell activation and inhibition of tumour progression.33 Another target for macrophage repolarization is the JAK/STAT3 pathway.43 Pathria et al. reported that depletion of STAT3 in myeloid cells results in a reduced tumour burden, repolarization of TAMs, and increased infiltration of cytotoxic T lymphocytes.34 Our study revealed that SLAMF8 induces an immunosuppressive phenotype in TAMs via activating the PI3K/AKT and JAK/STAT3 pathways. PI3K or STAT3 inhibitor effectively polarises macrophages toward the M1 phenotype, thereby restoring CD8+ T-cell function. In conclusion, our results validate the pivotal roles of the PI3K/AKT and JAK/STAT3 pathways in SLAMF8-induced immunosuppression, providing valuable insights into the regulatory mechanisms of SLAMF8. It should be noted that the genomic analyses underpinning these mechanistic insights were derived from the TCGA-COAD cohort. Future studies including rectal adenocarcinoma cohorts will be important to confirm the broader applicability of this mechanism across all colorectal cancer subtypes.
Although immune checkpoint inhibitors, including anti-PD1 agents, confer clinical benefit in a specific subpopulation of CRC patients, T-cell exhaustion resulting from multiple elevated inhibitory checkpoints limits their long-term control of cancer.2 Therefore, durable therapeutic responses are not achieved in approximately 80% of cancer patients when these agents are used as monotherapies.44 In addition, a significant proportion of CRC patients exhibit primary resistance to checkpoint inhibitors. Combination regimens including immune checkpoint modulators are needed to broaden the patient population that responds to immunotherapy and to improve clinical outcomes. We used siSLAMF8 in combination with anti-PD1 therapy to augment antitumor efficacy and enhance sensitivity to immune checkpoint blockade in both PD1-sensitive and PD1-resistant mouse subcutaneous CRC tumour models.27 Our study revealed that targeting SLAMF8 represents a promising strategy for combined immunotherapy by modulating immunosuppressive TAMs and synergising with T-cell-activating agents, thereby achieving robust and durable immunotherapy efficacy.
This study was designed to address a key mechanistic gap in the understanding of SLAMF8 biology in CRC TME. Although previous studies, including our own, established a strong correlation between SLAMF8 expression and an immunosuppressive microenvironment, its underlying mechanisms and precise effects on immune cell function remained unknown. Here, We provide evidence that SLAMF8 expression in TAMs directly activates the PI3K/AKT and JAK/STAT3 signalling pathways, thereby elucidating its role in promoting macrophage M2 polarisation and mediating immunosuppressive effects in TAM-T cell interactions.
In conclusion, our investigation demonstrated the important regulatory role of SLAMF8 in modulating the immunophenotype of TAMs (Figure 7). Specifically, the results of RNA-seq pathway enrichment and protein phosphorylation analyses indicated that the upregulation of SLAMF8 in TAMs led to the activation of the PI3K/AKT and JAK/STAT3 signalling pathways, subsequently promoting the M2 polarisation of macrophages and impairing the activation of CD8+ T cells. Inhibition of SLAMF8 reduces the tumour burden and increases sensitivity to anti-PD1 therapy in CRC, underscoring its potential as a therapeutic target for immunotherapy focused on TAMs. These findings provide a promising new approach for combining targeted macrophage modulation with anti-PD1 therapy.

Supplementary Material

Supplementary Material
Table S1.Clinicopathological characteristics of CRC patients from Drum Tower Hospital cohort.Table S2.Antibody list in this study.Table S3.Primer list in this study.

Supplementary Material

Supplementary Material

Supplementary Material

Supplementary Material

Supplementary Material

Supplementary Material

Supplementary Material

Supplementary Material

Supplementary Material