SLAMF9 mRNA expression in CRC tissues
To assess the abnormal expression of SLAMF9 in cancers, the expression levels of SLAMF9 were analyzed based on TCGA data, between matched normal and malignant samples from 17 malignancies (Figure 1A). Surprisingly, SLAMF9 was significantly aberrantly upregulated in 11 among 17 cancers, including colorectal carcinoma (CRC). To verify the results for CRC, we used the TCGA database to evaluate the mRNA expression levels of SLAMF9 in CRC patients and compared these with the expression levels in adjacent tissues, the results showed SLAMF9 mRNA expression was increased in CRC tissues compared to adjacent normal tissues (Fig.1B, P< 0.05), the results were verified in CRC and paired paracarcinoma tissues (Figure 1C). The area under the curve (AUC) was 0.913 (95% CI = 0.83-0.944) for SLAMF9 expression in CRC (Fig. 1D). Overall, the results bear significance to the prognostic implication of SLAMF9 mRNA expression in CRC patients.
Examination of relationship between SLAMF9 expression level and prognosis
To concentrate on the correlation between prognosis and SLAMF9 expression level, we played out a survival association study for each tumor, including overall survival (OS), the progression free survival (PFI), disease-free survival (DFI) and disease-specific survival (DSS). Utilizing information from the TCGA database, we analyzed the connection between SLAMF9 expression levels and OS in different tumor types through single variate Cox-regressive analysis. The risk proportions for SLAMF9 were critical for Kidney Chromophobe (KICH) (HR =1.35, 95%CI = 1.03~1.77, P = 0.02), Adrenocortical carcinoma (ACC) (HR =1.25, 95% CI =1.14~1.41, P = 5.6e-5), Kidney renal clear cell carcinoma (KIRC) (HR = 1.08, 95% CI = 1.03~1.14, P = 1.0e-3), Brain Lower Grade Glioma (LGG) (HR = 1.28, 95% CI = 1.21~1.35, P = 2.2e-17), LUAD (HR = 1.10, 95% CI = 1.02~1.19, P = 0.02), Pancreatic adenocarcinoma (PAAD) (HR = 1.20, 95% CI = 1.07~1.33, P = 1.5e-3), Pan-kidney cohort (KIPAN) (HR = 1.05, 95% CI = 1.01~1.09, P = 0.02), and COAD (HR = 1.12, 95% CI = 1.01~1.25, P =0.04), among which SLAMF9 had the highest risk effect in KICH (Figure 2A). The ensuing analysis of survival, which utilized patient information dichotomized for ideal cut off value in every tumor type, showed that survival distinctions in OS-related tumor types were all huge, demonstrating that patients with high SLAMF9 expression had more unfortunate prognosis (Figure 2B). Additionally, Cox proportional hazards model study showed that SLAMF9 expression level was substantially related to PFI of patients with ACC (HR = 1.17, 95% CI = 1.07~1.29, P = 3.5e-4), PAAD (HR = 1.18, 95% CI = 1.06~1.30, P = 2.8e-3), KIPAN (HR =1.05, 95% CI = 1.01~1.10, P = 0.01), LGG (HR = 1.21, 95% CI = 1.15~1.27, P = 4.4e-15), Pheochromocytoma and Paraganglioma (PCPG) (HR = 1.15, 95% CI = 1.02~1.30, P = 0.02), Thyroid carcinoma (THCA) (HR = 1.10, 95% CI = 1.01~1.20, P = 0.03), Testicular Germ Cell Tumors (TGCT) (HR = 1.14, 95% CI = 1.01~1.28, P = 0.04), and KICH(HR = 1.24, 95% CI = 1.00~1.52, P = 0.04) (Figure 2C). Kaplan-Meier survival analysis suggested that patients with high levels of SLAMF9 had shorter PFI in patients with KIRC (HR = 1.18, 95% CI = 1.11~1.25, P <0.0001), LGG (HR = 1.3, 95% CI = 1.07~1.59, P<0.0001), ACC (HR = 10.83, 95% CI = 3.29~35.7, P <0.0001), BLCA (HR = 1.04, 95% CI = 1.01~1.08, P = 0.014), KICH (HR = 2.56, 95% CI = 1.44~4.55, P = 0.00034), LUAD (HR = 1.04, 95% CI = 1.02~1.07, P = 0.042), MESO (HR = 1.17, 95% CI = 1.04~1.31, P = 0.00016), and PAAD (HR = 1.04, 95% CI = 1.01~1.07, P = 0.0053) (Figure 2D).
SLAMF9 protein expression in CRC tissues
Studies have shown that mRNA expression does not always correlate with similar protein expression patterns due to post-transcriptional regulatory mechanisms. Next, the expression of SLAMF9 in the CRC, and surrounding tissues was examined initially. Results of western blotting suggested that SLAMF9 expression was increased in CRC tissues than para-cancerous tissues (Figure 3A). Immunohistochemistry was used to measure the amount of SLAMF9 protein expression in CRC tissues, and the results of this investigation indicated that SLAMF9 protein expression was upregulated in CRC tissues (Figure 3B). Subsequently, SLAMF9 mRNA expression in five CRC cell lines were examined by qRT-PCR and western blotting. As compared with the NCM460, the level of SLAMF9 was significantly increased in four CRC cells (Figure 3C). Consistent protein levels were observed in western blotting (Figure 3D).
Relationship between the expression of SLAMF9 and immune infiltrating level of tumors
To explore the correlation between the expression level of SLAMF9 and the tumor immune response, we used the TIMER database to investigate immune infiltration in human tumors with different SLAMF9 expression level. In general, the level of immune infiltration with a large number of infiltrations including B cells, CD4+ T cells, CD8+ T cells, and DCs was positively correlated. However, the presence of NK cells, mast cells, myeloid-derived suppressor cells (MDSC), and macrophages was positively linked with SLAMF9 expression. The profile revealed that SLAMF9 participated in immune infiltration to some extent and was crucial to immuno-oncological cooperation.
(Figure 4A).
In order to determine whether SLAMF9 contributes to the development of immune infiltration in pan-cancer, we evaluated the connection between SLAMF9 expression and cancer purity. As Figure 4B indicated, the expression of SLAMF9 was determined to be strongly associated with stromal score, immune score, and estimate score, and the most significant correlations were observed in COAD, READ, and STAD.
Correlation between immunotherapy, immune checkpoints, and SLAMF9
The pan-cancer associations between SLAMF9 and immunological checkpoints were shown in Figure 5A. In some tumors, such as Cholangiocarcinoma (CHOL), COAD, KICH, LGG, PCPG, READ, and Thymoma (THYM), significant relationships existed between SLAMF9 expression and expression levels of recognized immune checkpoints, including neuronilin-1 (NRP1), leukocyte-associated immunoglobulin-like receptor-1 (LAIR1), CD48, CD28, HAVCR2, CD276, CD80, CD70, CD274, CD86, and CD44. This implied that SLAMF9 might work in concert with established immunological checkpoints.
During DNA replication and genetic recombination, the mismatch repair mechanism is essential for detecting and correcting mismatched nucleotides [12]. Microsatellite instability (MSI), a hypermutator phenotype brought on by frequent polymorphism in short repetitive DNA sequences and single nucleotide substitution, results from a DNA mismatch repair deficiency and increases tumor mutation burden (TMB) by accumulating mutation loads in cancer-related genes [13, 14]. They are thought to be independent predictors of ICB effectiveness and are in charge of tumor initiation [15, 16]. Here, we focus on the relationships between the expression of SLAMF9 and several critical MMR characteristics. SLAMF9 expression was negatively correlated with MutL homolog 1 (MLH1) in CHOL, HNSC, LGG, OV, and THYM. In ACC, SLAMF9 expression was positively correlated with MutS homolog 2 (MSH2) and MutS homolog 6 (MSH6). However, there was a negative association between THYM and UCEC. In COAD, esophageal carcinoma (ESCA), glioblastoma multiforme (GBM), head and neck squamous cell carcinoma (HNSC), KIRP, LGG, and THYM, it was negatively associated with epithelial cell adhesion molecule (EPCAM) (Figure 5B). Additionally, in BRCA, Cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC), COAD, HNSC, LGG, Ovarian serous cystadenocarcinoma (OV), PAAD, Sarcoma (SARC), Skin Cutaneous Melanoma (SKCM), STAD, Uterine Corpus Endometrial Carcinoma (UCEC), and BLCA, SLAMF9 expression demonstrated a favorable connection with TMB (P < 0.05; Figure 5C). The majority of MSI-High tumors were found to express more SLAMF9 than genetically stable ones, but the LUAD and Lung squamous cell carcinoma (LUSC) cohorts revealed the opposite trend (P < 0.05, Figure 5D).
Functional analysis by gene set enrichment analysis
In order to further analyze the relevant pathways involved with SLAMF9 in the process of tumor immunosuppression, we performed differential expression genes (DEGs) analysis on SLAMF9 high expression group and low expression group. The results manifested that immune-related pathways, including TNFA-signaling-via-NFKB, inflammatory response, IL6-JAK-STAT3 signaling, IL2-STAT5 signaling, EPITHELIAL-MESENCHYMAL-TRANSITION, were positively correlated with SLAMF9 expression in various tumors (Figure 6A). It is well known that tumor metastasis the principal cause of death has a considerable impact on EMT. The association between SLAMF9 expression and EMT score was examined to determine whether SLAMF9 is relevant to tumor metastasis. The findings suggest that the expression of SLAMF9 and EMT score are significantly positively correlated in the majority of malignancies (Figure 6B). Gene set enrichment analysis (GSEA) was furtherly utilized to determine the functional enrichment of DEGs. Generally, the top three positively enriched KEGG terms in high SLAMF9 subgroup were pathogenic escherichia coli infection, glycolysis gluconeogenesis, and pentose phosphate pathway, but negatively enriched terms basal cell carcinoma, and the top positively enriched HALLMARK terms including hypoxia, glycolysis, and epithelial mesenchymal transition. (Figure 6C). The above results revealed the immunosuppressive microenvironment created by SLAMF9 in pan-cancer was obviously guided by these pathways.
SLAMF9 promotes cell viability and motility in CRC cells
To investigate the role of SLAMF9 in colorectal carcinogenesis, we knocked down the expression of SLAMF9 in CRC cells by three siRNAs. The effectiveness of the knockdown transfection was performed by qRT-PCR and western blotting, and the findings indicated that SLAMF9 expression was significantly reduced in CRC cells after transfection with siRNA (Figure 7A). CCK‐8 assays demonstrated that the ability of cell proliferation was significantly decreased in siRNA transfected CRC cells than that of siNC group (Figure 7B). We employed transwell migration experiments to investigate how SLAMF9 affects CRC cell invasion and migration. The results indicated that SLAMF9 was positively associated with migration and invasive abilities of CRC cells (Figure 7C). We identified EMT indicators by western blotting to ascertain whether SLAMF9 increased the invasiveness of CRC through EMT mechanisms. We discovered that E-cadherin, a marker for epithelial cells, was significantly expressed in both siSLAMF9 transfected CRC cells. Vimentin and N-cadherin expression, however, decreased in CRC cell lines transfected with siSLAMF9, demonstrating a mesenchymal character. Collectively, our observations imply that SLAMF9 induces EMT to facilitate tumor invasion and metastasis (Figure 7D).
SLAMF9 knockdown inhibits tumor growth in vivo
According to the findings of SLAMF9 inhibiting CRC cell proliferation, migration and invasion in vitro, we next examined the effect of SLAMF9 on CRC growth in vivo. We firstly established a nude mice model by subcutaneously injecting the DLD1 cells into the ventral side, and then measured regularly (Figure 8A and 8B). Two weeks after the subcutaneous injection, we found that the SLAMF9 knockdown group’s tumor weight and size were considerably lower than those of the control group (Figure 8C and 8D).