RGS16 promotes glioma progression and serves as a prognostic factor

Abstract Background RGS protein family members have recently became new potentially promising therapeutic targets in many cancers. However, as a key member of RGS family, RGS16 has seldom been studied in glioma. The present study was designed to investigate the prognostic value and biological function of RGS16 based on large‐scale databases and functional assays in vitro. Methods Here, we performed comprehensive analysis for the expression characteristic of RGS16 in Chinese Glioma Genome Atlas (CGGA) microarray database with 301 patients and validated in The Cancer Genome Atlas (TCGA) microarray and RNA sequencing database. Student's t‐test, one‐way ANOVA test and long‐rank test were used to assess differences between groups. Kaplan‐Meier survival, univariate and multivariate Cox analysis and ROC curve were used to estimate the survival distributions. Biological implication of abnormal expression of RGS16 in glioma was also explored. Functional analysis of RGS16 was performed in several glioblastoma (GBM) cell lines. R language and SPSS were used for statistical analysis and graphical work. Results We found that the expression of RGS16 was positively related to the grade of glioma. High level of RGS16 commonly gathered in glioma of mesenchymal subtype and wild‐type IDH1. Moreover, higher expression level of RGS16 was found to be significantly correlated with poor prognosis. The univariate and multivariate Cox regression analysis and ROC curve showed that RGS16 was an independent prognostic factor for glioma patients. Gene ontology analysis, gene set enrichment analysis, and gene set variation analysis suggested that the overexpression of RGS16 tightly related to cell proliferation, migration, epithelial‐mesenchymal transition (EMT), immune and inflammatory response of glioma. Knockdown of RGS16 in glioma cell lines also showed that RGS16 promoted the malignant progress of glioma cell lines. Conclusions RGS16 plays an important role in glioma progression and serves as an independent prognostic factor, especially in GBM patients.


| INTRODUC TI ON
Glioma is the most common and lethal type of primary brain tumor. 1,2 Despite the progress of neurosurgery, the prognosis of patients with glioma is still unsatisfactory. 3,4 Due to the characteristic of infiltrative growth, the tumor tissues is difficult to removed clearly which eventually lead to postoperative recurrence and resistance to both radiotherapy and chemotherapy. 3 Thus, it is vital to have an insight into the complex regulating mechanism of the progression of this disease, and figure out the critical driver factors which also could be adopted as indicator of survival and worth further investigation as potential therapeutic targets.
The regulators of G protein singling (RGS) gene family members are originally identified as signal transduction G protein-coupled receptor (GPCRs) inhibitors, due to their ability to increase the intrinsic GTPase activity of G protein. 5,6 This protein family plays a critical role in the regulation of G protein-mediated pathways and many other biological processes in various tissues. [7][8][9] To our knowledge, many RGS family members, including RGS16, act as oncogenes and promote malignancy progression of many human cancers. [10][11][12][13] Besides, studies focus on RGS protein family also found that RGS16 played central roles in immune and inflammatory responses. 14,15 However, only a few studies have reported that RGS16 was dysregulated in glioma tissues or cell lines. What's more, the expression pattern and prognostic value of RGS16 in glioma are still unclear.
In this study, we identified RGS16 as a novel prognostic factor in glioma. The Chinese Glioma Genome Atlas (CGGA) microarray dataset was used to evaluate the expression preference, prognostic value, and biological functions of RGS16. Interestingly, we found that RGS16 not only played an important role in malignant progression of glioma, but was also tightly related to the immune and inflammatory response in glioma. These results suggested that RGS16 might serve as a novel immune biomarker of glioma. The above observations were validated in the Cancer Genome Atlas (TCGA) RNA sequencing dataset. Besides, we further detected the roles of RGS16 in several glioma cell lines. All these results indicated that RGS16 was a novel independent prognostic factor and played a crucial role in the malignant progress of glioma. Besides, RGS16 may also serve as a potential therapeutic target in glioma.

| Patients and samples
The transcriptome data and clinical information of 301 samples from CGGA microarray database (http://www.cgga.org.cn) was used as discovery set. This database was generated by Agilent Whole Human Genome Array platform. TCGA RNA sequencing database and microarray (http://cance rgeno me.nih.gov) were used as validation set.
Before using TCGA database for further analysis, RNA sequencing data were log2-transformed. In these databases, only samples with definite WHO classification were used for expression analysis and the samples without survival data were excluded in survival analysis. was performed with rabbit anti-RGS16 polyclonal antibody (Abcam, 1:800). Goat anti-rabbit IgG-HRP (Abcam, 1:5000) was used as secondary antibodies and β-tubulin was used for loading control.

| Cell scratch assays
U87 and LN229 cells were seeded in 6-well plates (1 × 10 5 cells per well) and incubated in a 37°C, 5% CO2 incubator. At 48 hours after transfected with RGS16 siRNA or a NC siRNA, the cell monolayer was scraped with a sterile 200-μL pipette tip. Then, fresh medium without serum was added to the plates and each well was photographed to mark the "zero point" of migration. After completion of 48 hours incubation, the samples were washed twice very gently with PBS (Gibco; Thermo Fisher Scientific). Each well was photographed using computer-assisted microscopy. Phase-contrast images were taken at the beginning (0 hours) and after 48 hours of incubation for the same scratch area.

| Clonogenic assays
LN229 and U251 cells were digested with 0.25% trypsin-EDTA (Gibco; Thermo Fisher Scientific) when they were in the log phase of growth.
Cells were then dissociated and seeded into a 6-well culture plate.
After adherence, cells were transfected with RGS16 siRNA or NC siRNA. Medium was replaced after 8 hours. Plates were maintained K E Y W O R D S EMT, glioma, immune response, prognosis, RGS16 in a 37°C, 5% CO2 incubator for 2 weeks. To maintain a low-level of RGS16 in the whole assay, transfection was repeated at the 7th day after the first transfection. Cell colonies were stained with crystal violet and photographed at the end of after the 2 weeks of incubation.

| Cell migration assays
The cell migration assay was performed using 24-well transwell chambers (Corning). After the transfection with RGS16 siRNA and NC siRNA, 1 × 10 5 Cells were seeded into upper chambers with 100 μL serum-free medium and medium with 10% FBS was added into the lower chamber. The LN229 and U251 cells at the bottom surface of filters were fixed and stained with 0.5% crystal violet after 2 hours and 6 hours, respectively.

| Bioinformatics analysis
The Pearson correlation analysis was performed to detect genes most closely related to RGS16 expression in CGGA and TCGA datasets.
Then, these genes were uploaded to DAVID website (http://david. abcc.ncifc rf.gov/ home.jsp); gene ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis were performed to detect the biological processes that correlated with RGS16 expression. Gene set enrichment analysis (GSEA) and gene set variation analysis (GSVA) were used to estimate various biological processes enrichment in each sample. 16 The immune checkpoints and inflammatory signatures have been described in our previous studies. 17

| Statistical analysis
Kolmogorov-Smirnov (K-S) test of normality was used to assess data distribution. Student's t-test and one-way ANOVA test were used to assess the significance of differences between groups. The prognostic value of RGS16 was estimated by Kaplan-Meier survival analysis.
Univariate and multivariate Cox regression analysis including age, gender, WHO grade, isocitrate dehydrogenase1 (IDH1) mutation status, O-6-methylguanine-DNA methyltransferase (MGMT) methylation status, chemotherapy, and radiotherapy were performed to F I G U R E 1 Expression pattern of RGS16 in different grades of gliomas. (A, B) In CGGA microarray and TCGA sequencing database, the mRNA expression level of RGS16 increased with tumor grade. (C, D) In CGGA microarray and TCGA sequencing database, the mRNA expression level of RGS16 was higher in IDH1 wild-type gliomas than gliomas with mutated IDH1 in each grade, though some groups have no statistically significant. K-S test of normality was used to assess the distribution of RGS16 expression in CGGA microarray and TCGA sequencing database (P = .479 and P = .085, respectively). * P < .05, ** P < .01, **** P < .0001, ns: no statistically significant test independent prognostic factors in SPSS 16.0. Several R packages including ggplot2, survival, pheatmap, pROC, circlize, and corrgram were used for pictures drawing and statistical analysis in R software 3.4.3. In all statistical analysis, a two-sided P value < .05 was considered statistically significant.

| RGS16 expression is associated with glioma grade and subtype
In consideration of heterogeneity across different grades of glioma, we compared expression levels of differentially expressed genes from the CGGA microarray dataset and found that RGS16 expression was positively correlated with tumor grade. These results were validated in TCGA RNA sequencing and microarray database ( Figure 1A,B). After dividing patients into two subgroups according to the IDH1 mutation status, we found that IDH1 mutant-type showed lower expression of RGS16 across different grades, though some groups have no statistically significant ( Figure 1C,D and Figure S1A).
We analyzed RGS16 expression level among different molecular subtypes defined by TCGA network; 18 the results showed that the mesenchymal subtype had the highest RGS16 expression in both CGGA and TCGA databases (Figure 2A,B and Figure S1B). These results also suggested that RGS16 had the potential to serve as a biomarker for mesenchymal subtype. To prove it, we performed ROC curves analysis for RGS16 expression and mesenchymal subtype, and the area under curve (AUC) was 79.8% and 81.5% in CGGA microarray and TCGA RNA sequencing database, respectively ( Figure 2C,D).

| High level of RGS16 predicts poor prognosis in glioma patients
To evaluated the association between RGS16 mRNA expression and glioma patients' clinical outcomes, we performed Kaplan-Meier (K-M) survival curve analysis with the data of 299 and 631 patients from the CGGA microarray database and TCGA RNA F I G U R E 2 Expression preference of RGS16 in different subtypes of gliomas. (A, B) In CGGA microarray and TCGA sequencing database, RGS16 expression was highest in mesenchymal subtype glioma samples. (C, D) In CGGA microarray and TCGA sequencing database, ROC curve analysis showed that RGS16 has a larger area under the curve in mesenchymal gliomas (AUC is 79.8% and 81.5%, respectively). K-S test of normality was used to assess the distribution of RGS16 expression in CGGA microarray and TCGA sequencing database (P = .479 and P = .085, respectively). * P < .05, **** P < .0001  Figure 3C,D). Similar results were also observed in TCGA microarray database ( Figure S2). Moreover, in ROC curve analysis, the AUC for RGS16 expression level in predicting the 3 years of survival in CGGA microarray and TCGA RNA sequencing databases were 0.787 and 0.794, respectively ( Figure 3E,F). Compared with "age" and "grade," the AUCs for RGS16 expression level were slightly smaller than those of "grade" in both databases. Those founding mentioned above revealed that RGS16 was a negative prognostic factor in glioma patients.

| RGS16 is associated with the cell proliferation, cell migration and EMT
To   results, we also performed gene set variation analysis (GSVA) to further evaluate the enrichment score of biological process and pathways in each sample ( Figure 4G and Figure S3D). These analyses indicated that RGS16 might play a key role in the malignant progression of glioma.

| RGS16 is tightly correlated with T-cell immunity in glioma
According to the results of the GO analysis, genes that tightly cor-

| RGS16 is associated with inflammatory activities in gliomas
To explore the role of RGS16 in inflammatory activation in gliomas, we chose seven inflammatory metagenes described in previous studies to represent different inflammation response. 17 As showed by the results, most of these signatures were positively related to RGS16, while IgG was quite the opposite ( Figure 6A,B). Perhaps it meant that RGS16 was tightly correlated with macrophages and T cell-related inflammatory response rather than B lymphocyte. Correlogram analysis was performed to further validate what we found in cluster analysis. The results also indicated that RGS16 was not involved in B lineage-related inflammatory responses ( Figure 6C,D). Similar findings were also reported in previous researches on PD1 and TIM-3. 17,19

| RGS16 promotes glioma progression in vitro
As the results showed in GO analysis, the high level of RGS16 expression was positively correlated with cell proliferation, cell migration, and epithelial to mesenchymal transition (EMT). To further confirm these functional roles of RGS16 in glioma, we performed experiments in vitro. Firstly, we detected RGS16 protein levels in H4, LN229, U87, U251, U118 glioma cell lines and human astrocytes (HA) cell line; the Western blot analysis indicated that LN229 had the highest expression level while HA had the lowest ( Figure 7A).
Then, RGS16 small interference RNA (siRNA) was transfected to knockdown RGS16 expression in LN229, U251, and U87 cell lines ( Figure 7B,C). After that, we detected the protein levels of several EMT signaling pathway biomarkers; Western blot analysis showed that the expression level of TCF8, N-cadherin, and β-catenin protein in three cell lines has dropped in varying degrees ( Figure 7C). These results suggested that RGS16 has a significant impact on the EMT signaling pathway. Furthermore, to evaluate the function of RGS16 in glioma cell lines, transwell and cell scratch were performed. The results indicated that the silence of RGS16 expression suppressed the migration ability of glioma cell lines ( Figure 7D,E,F, and G), Besides, the clonogenic assay showed that the proliferative capacity and clonogenicity of glioma cell lines were inhibited after transfection with RGS16 siRNA (Figure 7H,I).

| D ISCUSS I ON
Due to the rapidly proliferation and intensive invasion behaviors, gliomas became the deadliest intracranial tumor. Although, in this study, the functions of RGS16 in gliomas were deeply analyzed and discussed, the relationship between RGS16 expression and the biological characteristics of glioma remains unclear.
The in vitro experiments conducted in this study still need to be improved. Further understanding of the function role and expression pattern of RGS16 in gliomas may provide more accurate diagnosis and prognosis predictions of patients, as well as the improvement of personalized therapeutics of glioma.

| CON CLUS ION
In conclusion, we found that RGS16 plays an important role in the malignant progression of glioma. Bioinformatics analyses showed that RGS16 was involved in cell proliferation, migration, EMT, and immune and inflammatory response of glioma. Experiments in vitro suggested that RGS16 was overexpressed in glioma cell lines and promoted cell proliferation and migration via EMT process. Thus, we believe that RGS16 had the potential to serve as a novel prognostic biomarker and therapy target.

ACK N OWLED G M ENTS
Not applicable.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interests.

E TH I C A L A PPROVA L
This study was approved by the institutional review board (IRB) of Beijing Tiantan Hospital, and informed consent was obtained from all individual participants for CGGA project in this study.