In this study, the Gene Expression Omnibus (GEO) microarray datasets GSE40435, GSE53000, GSE53757, GSE105261 and GSE15641 were included for further analysis (23-27). The online tool GEO2R (https://www.ncbi.nlm.nih.gov/geo/) was used to detect the differentially expressed genes between normal kidney samples and ccRCC samples with the cutoff criteria of adjusted P-value <0.05 and logFC >1. A Venn diagram was drawn to depict the extraction of commonly upregulated genes present in more than two datasets. Furthermore, the online survival analysis tool OncoLnc (http://www.oncolnc.org/) was utilized to screen candidate genes associated with patient survival.

The TMSB10 mRNA expression level and The Cancer Genome Atlas Kidney Renal Clear Cell Carcinoma (TCGA-KIRC) clinical data, including age, sex, grade, stage, T stage, N stage, M stage, overall survival (OS) time and disease-free survival (DFS) time were downloaded from the TCGA data portal (https://tcga-data.nci.nih.gov/tcga/). Patients with corresponding gene expression were included in the present study, while those with missing OS or DFS data were excluded. The Yusenko Renal, Lenburg Renal, Jones Renal, Gumz Renal and Beroukhim Renal datasets were obtained from the Oncomine database (https://www.onco-mine.org/). Furthermore, to elucidate the possible role of TMSB10 in ccRCC pathogenesis, the online STRING database (https://string-db.org/) was used to acquire biological processes, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways and Reactome data of TMSB10 (28). Furthermore, a gene set enrichment analysis (GSEA) was conducted using GSEA software (http://www.broadinstitute.org/gsea). A false discovery rate <25% and P-value <0.05 were considered to indicate statistical significance.

Between January 2016 and January 2019, 80 pairs of adjacent normal renal tissue and ccRCC tumor tissue were collected from patients who underwent partial or radical nephrectomy at the Wuhan Union Hospital (Wuhan, China) for subsequent western blotting and immunohistochemistry (IHC) experiments. All patients did not receive any adjuvant therapy and provided written informed consent before surgery (Table SI). As a prolonged follow-up was not performed, the DFS and OS of these patients was not known. In future, the present research team plan to conduct a strict and scheduled follow-up to obtain complete clinical characteristics of the patients. The Human Research Ethics Committee of Huazhong University of Science and Technology (Wuhan, China) approved the present study and experimental procedures. The study methodologies conformed to the standards set by the Declaration of Helsinki.

The cell lines ACHN, 786-O, Caki-1, A-498 and HK-2 were purchased from the American Type Culture Collection. Dulbecco's modified Eagle's medium (DMEM; Gibco; Thermo Fisher Scientific, Inc.) containing 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.) and 1% streptomycin-penicillin was used to culture cells. Cells were cultivated in a 5% CO₂ and 37˚C environment. Small interfering (si)RNA oligonucle-otide sequences specifically targeting TMSB10 (si-TMSB10) and negative control siRNA (si-NC) were constructed by Guangzhou RiboBio Co., Ltd. The si-TMSB10 sequence was as follows: 5′-GAG AAG CGG AGT GAA ATT T-3′ and the si-NC sequence was: 5′-TTC TCC GAA CGT GTC ACG TdT dT-3′. For transfection, ccRCC cells were seeded in 6-well plates at 50-70% confluence; cells were transfected with 100 pmol siRNA using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.). After 48 h transfection, subsequent assays were conducted.

Cells and tissues were lysed with RIPA buffer (Beyotime Institute of Biotechnology) and the protein concentration was determined using a bicinchoninic acid kit (Beyotime Institute of Biotechnology). Subsequently, 30 µg total proteins/lane were separated using 10% SDS-PAGE. Then, the gel system was transferred to a polyvinylidene fluoride (PVDF) membrane (EMD Millipore) for 90 min at 90 V, after which the PVDF membrane was blocked with 5% non-fat milk dissolved in PBS for 1 h at room temperature and then incubated with antibodies against TMSB10 (1:2,000; ab14338; Abcam), vascular endothelial growth factor (VEGF; 1:3,000; 66828-1-Ig; ProteinTech Group, Inc.), phosphorylated (P-)PI3K (1:1,000; AP0854; ABclonal Biotech Co., Ltd.), PI3K (1:1,000; ab32089; Abcam) and β-actin (1:3,000; ab8226; Abcam) at 4˚C overnight. The next day, the membrane was washed and incubated with secondary antibodies (1:3,000; GB23303; Servicebio, Inc.) for 2 h at room temperature. The proteins were then developed with WesternBright™ ECL (Advansta, Inc.) and visualized with ChemiDoc XRS⁺ (Bio-Rad Laboratories, Inc.).

Ten paired ccRCC tissues and adjacent normal tissues were fixed in 4% formalin at room temperature for 12 h, dehydrated and embedded in paraffin. Blocking was performed with xylene and paraffin (1:1) for 2 h at room temperature. The tissue sections (4 µm) were then incubated in primary rabbit TMSB10 polyclonal antibody (1:100; ab14338; Abcam) at 4˚C overnight. After rinsing three times with PBS, the sections were incubated at room temperature for 2 h with goat anti-rabbit secondary antibodies (K5007; Dako; Agilent Technologies, Inc.). The sections were then viewed using a light microscope.

ACHN cells were transfected with si-NC or si-TMSB10 at 48 h before the experiment as described by Cao et al (29). Subsequently, the cells were inoculated on 96-well plates at a cell density of 1×10³ cells/well. The cell proliferation rate was determined using a Cell Counting kit-8 (CCK-8; Dojindo Molecular Technologies, Inc.) every 24 h for a total of 96 h based on the manufacturer's instructions. In brief, 10 µl CCK-8 solution was added to each well. After incubation for 3 h at 37˚C, the optical density of each well was measured at 450 nm to evaluate the quantity of living cells. Finally, the number of cells were plotted over 4 days to reflect the rate of cell proliferation using GraphPad Prism 7.0 (GraphPad Software, Inc.).

As described above, ACHN cells was transfected with si-NC or si-TMSB10 48 h prior to the experiment. In addition, before the migration and invasion assays, cells were cultivated in DMEM without serum for 6-8 h to starve the cells. Boyden Transwell chambers (Corning Inc.) containing 8-µm membrane filters were utilized. Cells (1×10⁴) in serum-free medium were seeded to the upper chamber, whereas the bottom chamber was filled with DMEM containing 10% FBS (BD Biosciences). Following incubation at 37˚C for 24 h, the cells on the lower chamber were fixed with 100% methanol for 10 min at room temperature, and then stained with 0.05% crystal violet for 30 min at room temperature. Finally, five random fields were counted under a light microscope (Olympus CX41-32C02; Olympus Corporation) at ×100 magnification. Three independent experiments were conducted. With regard to invasion assays, Matrigel (BD Biosciences) was precoated into the upper chamber for 6-8 h. Then, cells (2×10⁴) were inoculated into the upper chamber in serum-free medium. The remaining procedure was the same as that of the migration assay.

All statistical analyses were performed using GraphPad Prism and SPSS Statistics (version 22.0; IBM Corp.). Numerical data are presented as the mean ± SD. Tukey's test was used to detect the differences between groups. A paired sample t-test was used for the analysis of differences between paired samples. The associations between TMSB10 expression and various clinicopathological characteristics in patients with ccRCC were evaluated using Pearson's χ² test. Receiver operator characteristic (ROC) curves and areas under the curve (AUC) were used to assess the diagnostic value of TMSB10 expression in patients with ccRCC. The association between TMSB10 expression level and either OS or DFS was analyzed using Kaplan-Meier curves with log-rank tests. P<0.05 was considered to indicate a statistically significant result.

As shown by the Venn diagram in Fig. 1, 19 genes were commonly upregulated in three of the five GEO microarray datasets. Survival analysis of these genes based on the OncoLnc online database revealed that high expression of TMSB10, ENO2 and NNMT indicated poor prognosis, while high expression of ST8SIA4 indicated a favorable outcome (Fig. 2A-D). However, the expression levels of the other 15 genes, namely CAV1, ANGPTL4, BHLHE41, CA9, CAV2, EGLN3, HILPDA, IGFBP3, NDUFA4L2, PRKCDBP, SAP30, SCARB1, SLC15A4, SLC16A3 and SPAG4 were not associated with OS in ccRCC (Fig. 2E-S).

To fully investigate the role of TMSB10 in ccRCC development, the association between the mRNA expression level of TMSB10 and various clinicopathological factors was explored. The TMSB10 expression level was found to be significantly higher in tumor tissues than in normal tissues from the TCGA database (Fig. 3A and B). To further verify these findings, TMSB10 expression levels in patients with ccRCC were retrieved from the Oncomine database. Studies by Yusenko et al (30), Lenburg et al (31), Jones et al (32), Gumz et al (33) and Beroukhim et al (34) showed consistent results (Fig. 3C-G). Subsequently, the present study analyzed the association of the TMSB10 expression levels of 530 cases from the TCGA database with different clinicopathological parameters. Pearson's χ² test revealed that the expression level of TMSB10 was significantly associated with patients' sex, histological grade, TNM stage, T stage, N stage, M stage and vital status (Table I). Furthermore, subgroup analysis confirmed that higher expression levels of TMSB10 were significantly associated with higher pathological TNM stage, higher tumor T stage, distant metastasis, lymph node metastasis, and higher histological grade in ccRCC (Fig. 4). However, no association was observed between TMSB10 expression levels and patients' age (data not shown). These results demonstrated that TMSB10 was overex-pressed and closely associated with sex, G grade, TNM stage, T stage, N stage, M stage and vital status in ccRCC. To further support the association between TMSB10 expression level and clinicopathological factors, verification was conducted in the GEO database. In GSE53757 (23), the expression of TMSB10 increased with increasing tumor stage, and in GSE40435 (25), the expression of TMSB10 increased with increasing histological grade (Fig. S1). However, future multi-center studies with a larger sample size are required to verify the results.

Kaplan-Meier survival analysis and a log-rank test were applied to determine the OS and DFS in different subgroups of patients according to their TMSB10 expression level. The results demonstrated that TMSB10 may be a potential prognostic biomarker for patients with the following characteristics: Female (Fig. 5A), male (Fig. 5B), age ≥60 years (Fig. 5C), M0 stage (Fig. 5D) and N0 stage (Fig. 5E). Furthermore, the present study analyzed the relationship between TMSB10 expression and DFS. The results for low and high TMSB10 expression relative to median expression indicated that patients with low TMSB10 expression tended to have a more favorable DFS compared with those with high TMSB10 expression (Fig. 6A). Moreover, DFS analysis in subgroups of patients with ccRCC revealed that high expression of TMSB10 was a useful prognostic indicator for ccRCC patients with the following features: Age <60 or ≥60 years (Fig. 6B and C), female (Fig. 6D) or male (Fig. 6E), G1+G2 grade (Fig. 6F), G3+G4 grade (Fig. 6G), N0 stage (Fig. 6H), M0 stage (Fig. 6I), T1+T2 stage (Fig. 6J) and stage I+II (Fig. 6K).

To investigate the association between TMSB10 expression and the diagnosis of patients with ccRCC, ROC curve analysis of various clinicopathological factors was performed. The results indicated that TMSB10 could adequately distinguish patients with ccRCC with an AUC of 0.9543 (P<0.0001; Fig. 7A). Additionally, the TMSB10 expression level also exhibited diagnostic value for subgroups of patients with ccRCC as follows: Alive vs. dead (AUC=0.6176, P<0.0001; Fig. 7B), recurrence vs. disease-free (AUC=0.6733, P<0.0001; Fig. 7C), M0 vs. M1 stage (AUC=0.6146, P=0.00768; Fig. 7D), N0 vs. N1 stage (AUC=0.6943, P=0.0197; Fig. 7E), stage I+II vs. stage III+IV (AUC=0.6577, P<0.0001; Fig. 7F), T1+T2 vs. T3+T4 stage (AUC=0.6251, P<0.0001; Fig. 7G) and G1+G2 vs. G3+G4 grade (AUC=0.6642, P<0.0001; Fig. 7H).

Western blotting analysis and IHC were conducted to verify the results of public databases. Western blotting assays revealed that the protein level of TMSB10 in tumor tissues was upregulated compared with that in normal tissues (Fig. 8A), and the expression level in ACHN, 786-O, Caki-1 and A-498 cells was elevated compared with that in HK-2 cells (Fig. 8B). Moreover, IHC analysis showed that TMSB10 was principally located at the membranes of cancer cells, cytoplasm and renal tubular epithelial cells, and the protein expression was distinctly detected in tumor tissues (Fig. 8C).

To elucidate how TMSB10 participates in ccRCC pathogenesis, biological processes of TMSB10 were retrieved from the STRING database and GSEA analysis was performed in the TCGA database. The main Gene Ontology terms included 'vascular endothelial growth factor signaling pathway', 'regulation of cell population proliferation' and 'regulation of cell migration' (Fig. 9A; Table II). The KEGG pathways primarily consisted of 'HIF-1 signaling pathway', 'FoxO signaling pathway', 'focal adhesion', 'Ras signaling pathway', 'MAPK signaling pathway', 'regulation of actin cytoskeleton', 'PI3K-Akt signaling pathway' and 'pathways in cancer' (Fig. 9B; Table III). The Reactome terms principally included 'regulation of gene expression by hypoxia-inducible factor' and 'signaling by VEGF' (Fig. 9C; Table IV). The protein-protein interaction network in Fig. 9D indicated that many proteins interact with TMSB10. Moreover, the GSEA results demonstrated that the activated gene sets 'DNA replication' and 'P53 signaling pathway' were associated with patients having a higher TMSB10 expression level (Fig. 9E and F).

To elucidate the functional role of TMSB10 in ccRCC, TMSB10 was knocked down in ACHN cells through transfection with si-TMSB10. After transfection with si-TMSB10, decreased TMSB10 protein expression was observed in ACHN cells (Fig. 10A). CCK-8 assays demonstrated that transfection with si-TMSB10 significantly reduced the proliferation of the cells (Fig. 10B). Furthermore, Transwell assays revealed that downregulation of TMSB10 significantly attenuated the migration and invasion ability of the cells compared with that in the si-NC group (Fig. 10C and D). Furthermore, knockdown of TMSB10 down-regulated the P-PI3K/total PI3K ratio and VEGF expression level (Fig. S2).