The expression levels of SYT4 in cancers
First, we used the Oncomine database to explore the mRNA expression levels of SYT4 in different tumours (Figure 1A). The results suggested that the mRNA expression of SYT4 was higher in the lung tumour tissues than in normal lung tissues. In contrast, the mRNA expression of SYT4 was lower in the brain and nervous system cancer, colorectal, oesophageal, gastric, and prostate cancer, as well as in sarcoma tissues, than in the respective normal tissues. Supplemental Table 1 explicitly presents the mRNA expression levels of SYT4 in tumours according to different studies.
Subsequently, we used the TIMER database to explore the expression levels of SYT4 in different cancers (Figure 1B). We found a significant difference in the expression levels of SYT4 between cancer tissues and normal tissues in the following cancers: bladder urothelial carcinoma (BLCA), breast invasive carcinoma (BRCA), colon adenocarcinoma (COAD), oesophageal carcinoma (ESCA), head and neck squamous cell carcinoma (HNSC), kidney chromophobe (KICH), kidney renal clear cell carcinoma (KIRC), kidney renal papillary cell carcinoma (KIRP), liver hepatocellular carcinoma (LIHC), lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), prostate adenocarcinoma (PRAD), rectum adenocarcinoma (READ), stomach adenocarcinoma (STAD), thyroid carcinoma (THCA), and uterine corpus endometrial carcinoma (UCEC).
To verify the above results at the tissue and cell levels, we analyzed SYT4 expression in HPA and CCLE. In the immunohistochemical staining data, we found that the expression level of SYT4 was not high in cancer tissues. Furthermore, compared to the levels in cell lines of neuroblastoma, small cell lung cancer and prostate cancer, the levels of SYT4 in other cancer cell lines were lower (Supplemental Figure 1A-P).
Prognostic significance of SYT4 in different cancers
We used gene chip data derived from the Kaplan-Meier plotter database to explore the association between SYT4 expression and the survival of breast, lung, gastric and ovarian cancer patients. The results are shown in Figure 2. For patients with gastric cancer, SYT4 was associated with an unfavourable prognosis [overall survival (OS): HR=1.39 (1.12-1.73), log-rank P = 0.0025; progression-free survival (PFS): HR=1.31 (1.03-1.67), log-rank P = 0.025]. For breast cancer patients, SYT4 was related to recurrence-free survival (RFS) [HR=0.76 (0.65-0.88), log-rank P = 0.00042] but had no significant impact on OS [HR=0.74 (0.54-1.02), log-rank P = 0.065]. SYT4 did not affect the OS or PFS of patients with lung cancer [OS: HR=1.02 (0.86-1.20), log-rank P = 0.84; PFS: HR=1.12 (0.86-1.47), log-rank P = 0.40] or ovarian cancer [OS: HR=1.01 (0.82-1.24), log-rank P = 0.93; PFS: HR=1.07(0.89-1.29), log-rank P = 0.46].
After the exploration of SYT4 in the Kaplan-Meier plotter database, we analysed the impact of SYT4 on prognosis of different cancers by analysing RNA sequencing data from the TCGA database through GEPIA2 (Supplemental Figure 2). The results that the impacts of SYT4 on both the OS and disease-free survival (DFS) of LGG and STAD were consistent. High expression levels of SYT4 were related to a poor prognosis in patients with STAD (OS: HR=1.60, log-rank P = 0.006; DFS: HR=1.80, log-rank P = 0.0041). In contrast, high expression levels of SYT4 were correlated with a good prognosis in LGG patients (OS: HR=0.53, log-rank P = 0.00047; DFS: HR=0.58, log-rank P = 6e-04).
To further explore the correlation between the expression levels of SYT4 and the survival of LGG patients, we verified it in the CGGA database (Supplemental Figure 3). In the mRNAseq_325 dataset, the result suggested a significant correlation between the expression level of SYT4 and survival in all primary glioma patients, including patients with WHO grade II and III tumours, but there was no correlation between SYT4 and prognosis in patients with grade IV tumours. Similarly, in the mRNAseq_693 dataset, the result indicated a correlation between the expression level of SYT4 and the survival of all primary glioma patients, including patients with WHO grade III tumours, but there was no correlation between SYT4 and prognosis in patients with grade II and IV tumours. The above results partially verify our hypothesis, suggesting the prognostic value of SYT4 in LGG and STAD. SYT4 had a beneficial effect on the prognosis of LGG patients. In contrast, SYT4 had a detrimental effect on the survival of STAD patients.
The expression of SYT4 has an impact on the survival of gastric patients with lymphatic metastasis
To explore the mechanism by which the expression level of SYT4 affected the survival of gastric cancer patients, we used the Kaplan-Meier plotter database to explore the association between the expression levels of SYT4 and clinical factors of gastric cancer patients (Table 1). We conducted a stratified analysis of clinical factors affecting OS and PFS, such as sex, AJCC stage, T stage, N stage, and M stage. In terms of sex, high expression of SYT4 was related to an unfavourable prognosis in terms of OS in both males [HR = 1.49 (1.11-2.01), log-rank P = 0.0082] and females [HR = 1.89 (1.22-2.93), log-rank P = 0.0039] and an unfavourable prognosis in terms of PFS in female patients [HR = 1.56 (1.02-2.38), log-rank P = 0.038]. In addition, a significant correlation was also shown between the expression level of SYT4 and the lymphatic metastasis of patients. We did not find a significant association between the expression level of SYT4 and the survival of gastric cancer patients without lymph node metastasis [OS: HR = 1.10 (0.47-2.57), log-rank P = 0.83; PFS: HR = 1.16 (0.50-2.70), log-rank P = 0.73]. However, there was a significant correlation between the expression level of SYT4 and the survival of gastric cancer patients with N1 stage tumours [OS: HR = 1.85 (1.21-2.81), log-rank P = 0.0036; PFS: HR = 1.90 (1.27-2.83), log-rank P = 0.0014]. Moreover, those patients with gastric cancer with lymph node metastasis had correlations of SYT4 expression with OS [HR = 1.31 (1.01-1.70), log-rank P = 0.045] and PFS [HR = 1.32 (1.02-1.70), log-rank P = 0.031]. Therefore, the expression level of SYT4 might further affect the prognosis of gastric cancer patients by affecting lymph node metastasis.
SYT4 expression is related to the level of immune infiltration in gastric cancer and lower-grade brain glioma
Among the various factors affecting the survival and lymph node metastasis of cancer patients, lymphocyte infiltration is a significant independent predictor[25]. Therefore, we analysed the association between the expression levels of SYT4 and the levels of immune infiltration in 39 types of cancer in the TIMER database (Supplemental Figure 4). According to the results, we found that tumour purity and the expression levels of SYT4 had significant correlations in 12 cancer types. Similarly, the infiltration levels of B lymphocytes, CD4+ T lymphocytes, CD8+ T lymphocytes, macrophages, neutrophils, and DCs had significant correlations with the expression levels of SYT4 in 11, 14, 10, 19, 11 and 13 types of cancer, respectively.
Based on the findings in GEPIA2, Kaplan-Meier plotter and CGGA, we focused on cancer types in which the expression level of SYT4 was negatively correlated with tumour purity in TIMER and had a significant correlation with the prognosis of patients in GEPIA2 and Kaplan-Meier plotter: STAD and LGG (Figure 3). BRCA was selected as a control. It is worth noting that the expression level of SYT4 was negatively related to the prognosis of STAD but had a positive association with the infiltration of immune cells. We found that the expression level of SYT4 in STAD patients was negatively related to tumour purity (cor = −0.172, P=7.75e-04) but was positively associated with infiltration of the following immune cells: B cells (cor = 0.192, P=2.09e-04), CD4+ T cells (cor = 0.385, P=2.38e-14), CD8+ T cells (cor = 0.122, P=1.82e-02), macrophages (cor = 0.385, P=1.72e-14), and dendritic cells (cor = 0.207, P=6.01e-05). However, the expression level of SYT4 was positively related to the prognosis of LGG patients but negatively associated with infiltration of the following immune cells: B cells (cor = −0.385, P=6.54e-12), CD4+ T cells (cor = −0.577, P=1.34e-43), neutrophils (cor = − 0.375, P=2.49e-17), macrophages (cor = −0.505, P=6.45e-32), and dendritic cells (cor = −0.478, P=1.37e-28). We did not find similar correlations in BRCA. These above results suggested that the reasons why the expression level of SYT4 impacted the prognosis of STAD and LGG patients in the different ways was probably the different relationships between the expression level of SYT4 and the level of immune infiltration in STAD and LGG.
Correlations between the expression levels of SYT4 and markers of immune cells
To further explore the potential mechanisms of interaction between SYT4 and various immune infiltrating cells, such as CD8+ T cells, T cells (general), B cells, monocytes, TAMs, M1 and M2 macrophages, neutrophils, natural killer cells, and dendritic cells, in STAD and LGG, we analysed the correlations between the expression levels of SYT4 and immune markers of multiple immune cells in STAD and LGG based on the TIMER and GEPIA databases, In addition, we also performed correlation analysis with the immune markers of the following types of functional T cells: T helper cells, follicular helper T cells, regulatory T cells, and exhausted T cells. The correlation coefficient was adjusted based on tumour purity[26] (Table 2). According to the correlation analysis between the expression levels of 56 immune cell markers and the expression levels of SYT4, we found that the purity-adjusted correlation coefficients of 35 markers were statistically significant in STAD patients, and they were all positive. The purity-adjusted correlation coefficients of 46 markers were statistically significant in LGG patients, however, most of them were negatively. Only ten markers had statistically significant purity-adjusted correlations with the expression levels of SYT4 in BRCA patients. In addition, we also found significant correlations between the expression levels of markers of monocytes, TAMs and M2 macrophages and the expression levels of SYT4 in patients with STAD and LGG but not in BRCA (Table 2, Figure 4). In detail, these markers, such as CD115, CCL2, IL10, VSIG4, and MS4A4A, had significantly positive correlations with SYT4 expression levels in STAD (P <0.0001, Figure 4A-D). For LGG, these markers, such as CD163 of M2 macrophages, NOS2, IRF5 and COX2 of M1 macrophages, and CD86 of monocytes, also showed a significant correlation with SYT4 expression levels but the markers that were correlated with SYT4 expression in STAD did not show a correlation in LGG (P <0.0001, Figure 4I-L). However, the expression of SYT4 in BRCA did not show significant correlations with the above markers (Figure 4E-H). Then, to verify the results, we analysed the correlation between monocytes, TAMs, and M1 and M2 macrophages immune markers and the expression levels of SYT4 in STAD, LGG, and BRCA with the GEPIA2 database. The results were similar to those achieved in the TIMER analysis (Table 3). Moreover, it is worth noting that the correlations between the expression level of SYT4 and the levels of immune markers were positive and negative in STAD and LGG, respectively. Hence, we concluded that SYT4 likely interactes with various immune cells in STAD and LGG in the opposite ways