IL18 was highly expressed and associated with poor prognosis in RCC patients

IL18 is an anti-tumoral factor or an oncogenic factor in different type of cancers. To investigate the role of IL18 in RCC, we firstly analyzed the expression of IL18 in tumor tissues and paired normal adjacent tissues. We found IL18 was highly expressed in RCC tumor tissues compared to normal tissues (Fig. 1A, B). Furthermore, we also found that the expression of IL18 was differentially expressed in renal cells with different pathological grades and GRADE grades, and it was positively correlated with the grade (Fig. 1C–F). By analyzing the prognostic information of RCC patients in the TCGA database, we found that the higher expression of IL18 was associated with the shorter overall survival and disease-free survival of patients (Fig. 1G, H). These results demonstrated that IL18 was highly expressed and associated with poor prognosis in RCC patients.

Fig. 1

IL18 is highly expressed and associated with poor prognosis in RCC patients. A. Relative mRNA expression of IL18 between tumor and adjacent normal tissues using the data from TCGA cohort; B Relative expression of IL18 between tumor and adjacent normal tissues using the data from CPTAC cohort; C–F The expression of IL18 expression in different stages and pathological grades of patients from TCGA cohort; G, H Kaplan–Meier survival analysis of IL18 expression for overall survival and disease-free survival of patients from TCGA RCC cohort, respectively

IL18 promoter was abnormally methylated in RCC

More and more researches have suggested that DNA promoter methylation plays an important role in prognosis and grading of tumors [24]. Therefore we further explored the role of IL18 promoter methylation in RCC. First of all, we used the relevant website (http://www.bioinfo-zs.com/smartapp/#) to explore the methylation condition of the IL18, and further examined the six main methylation sites: cg04100971, cg04929355, cg05687149, cg09122223, cg11304234 and cg26534425 (Additional file 1: Fig. S1). To validate IL18 promoter methylation in tumor tissues, we analyzed these 6 methylation sites in IL18 promoter in TCGA database and anther two GEO datasets. The sites of cg04100971 and cg05687149 were significantly hypermethylated in tumor compared to normal tissue. (Fig. 2A–C). We also found that the other four methylation sites were hypomethylated in tumor tissue versus normal tissue in all three or some of the datasets (Fig. 2A–C). We further explored the correlation between the methylation of IL18 promoter and the clinical characteristics of renal cancer patients, we analyzed IL18 and the six methylation sites on IL18 promoter (cg04100971, cg04929355, cg05687149, cg09122223, cg11304234 and cg26534425) with the clinicopathological data of renal cancer patients by using the TCGA data. We found that the promoter methylation level of IL18 was different in different pathological grades and stages of RCC patients. The higher the grade was, the higher the methylation levels of cg04100971 and cg05687149 were, but the lower the methylation level of the cg04929355, cg09122223, cg11304234 and cg26534425 were (Fig. 3A–K). The promoter methylation level of IL18 was also related to the gender and ages (Additional file 2: Table S7), but not associated with smoking (Additional file 2: Table S8). These results indicated that IL18 promoter was abnormally methylated in RCC and differences in methylation levels were also related to the grading of RCC.

Fig. 2

IL18 promoter was abnormally methylated in tumor versus normal adjacent tissues in RCC. A–C Relative methylation of the six differentially methylated CpG sites (cg04100971, cg04929355, cg05687149, cg09122223, cg11304234 and cg26534425) located in IL18 promoter between tumor and adjacent normal tissues of TCGA cohort, GSE70303 and GSE105260, respectively

Fig. 3

Promoter methylation was associated with aggressive clinical phenotypes in RCC. A–L Relative expression of methylation of cg04100971, cg04929355, cg05687149, cg09122223, cg11304234 and cg26534425 in different stages and pathological grade of patients from TCGA cohort

IL18 promoter methylation correlated with IL18 expression in RCC

We further explored whether the expression of IL18 is related to the methylation level of IL18 promoter. To this end, we examined the IL18 expression and promoter methylation level in the TCGA database, and found that in addition to the cg05687149 site, the other five methylation sites were also related to the expression of IL18 (Fig. 4A–F). To further explore the role of IL18 promoter methylation in its expression, we used 5-Azacytidine experiments, a cytidine nucleoside analog that specifically inhibits DNA methylation, to inhibit DNA methylation in ACHN and 786-O cells. We found that IL18 expression was significantly increased after treatment with 5-Azacytidine for 72 h (Additional file 1: Fig. S2). These mean that abnormal methylation of IL18 promoter may cause the abnormal high expression of IL18 in RCC.

Fig. 4

The correlation of IL18 promoter methylation with IL18 in RCC. A–F The correlation of cg04100971, cg04929355, cg05687149, cg09122223, cg11304234 and cg26534425 methylation with IL18 expression in TCGA tumor tissues

The promoter methylation level of IL18 predicted prognosis in RCC

To explore the prognostic value of IL18 promoter methylation in RCC, we first analyzed the association of these 6 methylation sites with the overall survival in RCC patients. We found that the hypermethylated site cg04100971 was associated with a poor overall survival (Fig. 5A) and disease-free survival (Fig. 6A), while another hypermethylated site cg05687149 predicted favorable overall survival (Fig. 5C). Oddly, the methylation of cg05687149 was not related to IL18 expression (Fig. 4C). We also found that the other four hypomethylated sites were associated with favorable overall (Fig. 5B, D–F) and disease-free survival (Fig. 6B, D–F). These results showed that the promoter methylation level of IL18 may be a potential biomarker in predicting prognosis in RCC.

Fig. 5

IL18 promoter methylation predicted overall survival in RCC. A–F. Kaplan–Meier survival analysis of J cg04100971, cg04929355, cg05687149, cg09122223, cg11304234 and cg26534425 for overall survival of patients from TCGA RCC cohort, respectively

Fig. 6

IL18 promoter methylation predicted disease-free survival in RCC. A–F Kaplan–Meier survival analysis of cg04100971, cg04929355, cg05687149, cg09122223, cg11304234 and cg26534425 for disease-free survival of patients from TCGA RCC cohort, respectively

IL18 expression and promoter methylation were correlated with immune cell infiltration in RCC

As it was reported that DNA methylation was closely related to tumor microenvironment, we further investigated the potential of IL18 and promoter methylation and expression in immune regulation. We analyzed the correlation of IL18 expression and promoter methylation in 23 immune cell lines including B cells, as well as CD4 and CD8 T cells. We found that high IL18 expression was associated with higher infiltration of immune cells (Fig. 7A). We also found that the methylation levels of the IL18 promoter were clearly associated with immune infiltration of immune cells. Among all the examined sites, we found cg04100971 hypermethylation and cg0912223 hypomethylation were significantly correlated with infiltration of immune cells (Fig. 7A). To further analyze the key immune-related pathways, we chosen 17 biological pathways, analyzed the correlation between them and IL18 expression and promoter methylation. We found that lower expression of IL18 was correlated with a higher angiogenesis score (Fig. 7B), meaning that patients with low IL18 expression were probably sensitive to anti-angiogenesis therapy. Besides, lower IL18 expression was also related to lower scores of processing machinery, CD8 T effector, antigen cytolytic activity, and MHC-HLA, which indicates the infiltration of lower immune cells in the tumor microenvironment. Also, higher IL18 expression was significantly related to higher co-inhibition APC, co-stimulation APC, co-inhibition T cell signature immune checkpoint activity and EMT3 (Fig. 7B). All these showed that tumors with high IL18 expression were abundant in the infiltration of the immune cells, and they also had stronger immune checkpoint signature resulting in an immunosuppressive microenvironment. Next, we analyzed the relation of IL18 promoter methylation with these 17 immune-related pathways (Fig. 7C–H). The cg11304234 and cg26534425 hypomethylation were associated with lower angiogenesis, but higher co-inhibition T cell signature, immune checkpoint activity, co-stimulation APC (Fig. 7C, D), which were consistent with our results in IL18 expression. These results indicated that high IL18 expression and promoter hypomethylation were positively correlated with immune cell infiltration, that contains high enrichment of immune checkpoint molecules, which may sensitively respond to immune checkpoint inhibitors. Therefore, IL18 expression and promoter methylation were the potential molecular biomarkers to predict the responses of RCC patients to immune checkpoint inhibitors therapy.

Fig. 7

IL18 expression and promoter methylation were correlated with immune cell infiltration in RCC. A. The correlation heatmap of IL18 expression and its differentially methylated sites with 23 types of immune cells in the TCGA RCC cohort, only statistically significant (P < 0.05) were shown in correlation coefficients. B–H. Relative enrichment scores of 17 immune-related pathways in the high and low groups according to the median of IL18 expression and the methylation of cg04100971, cg04929355, cg05687149, cg09122223, cg11304234 and cg26534425 respectively

IL18 expression and promoter methylation were associated with the expression of key immunomodulators in RCC

It was reported that tumor immune microenvironment could be regulated by various membrane proteins and cytokines. To clarify the relationship of IL18 expression and promoter methylation with the key molecules participating in the anti-tumor immune responses, we explored the relationship between IL18 expression and promoter methylation and 74 key immunomodulators using the TCGA cohort. We found that high IL18 expression and hypomethylation of cg04100971 and cg0912223 were correlated with the expression of most immunomodulators (Fig. 8A). Based on the different therapeutic effect of immune checkpoint inhibitors on various tumors, we analyzed the connection of IL18 expression and cg04100971, cg05687149, cg04929355, cg09122223, cg11304234 and cg26534425 methylation with three immune checkpoint molecules (CTLA4, LAG3, and PDCD1), using expression data from the TCGA dataset. We found that IL18 expression was significantly positive related to the expression of CTLA4, LAG3, and PDCD1 (Fig. 8B), whereas cg05687149, cg04929355, cg11304234 and cg26534425 methylation were negatively associated with expression of CTLA4, LAG3, and PDCD1 (Fig. 8C–F). These results showed that IL18 expression and its promoter methylation were related to the immune suppressive microenvironment in RCC.

Fig. 8

IL18 expression and its promoter methylation were associated with the expression of the key immunomodulators in RCC. A. The correlation heatmap of IL18 expression and the six methylated CpG sites with the key immunomodulators in the TCGA RCC cohort, only statistically significant (P < 0.05) are shown correlation coefficients. B–D. The correlation of JAK3 expression, cg04100971, cg04929355, cg05687149, cg09122223, cg11304234 and cg26534425 methylation with immune checkpoint molecules CTLA4, LAG3, PACD1, respectively