Human specimens were collected after obtaining informed consent in accordance with the Declaration of Helsinki and was approved by Institutional Review Board of Chungbuk National University Hospital.

Total RNA and microRNA (miRNA) from patient samples were extracted by miRNeasy Mini Kit (Qiagen) according to the manufacturer’s instructions and was sent to Macrogen (South Korea) for sequencing and analysis.

The expression of genes in renal cancer stages and survival analyses of 20 candidate genes in The Cancer Genome Atlas Program - Kidney renal clear cell carcinoma dataset (TCGA-KIRC) were screened by UALCAN (https://ualcan.path.uab.edu) (33). The signaling pathways were examined by Geneset Enrichment Analysis (GSEA) version 4.3.2 (34), and gene enrichment was analyzed by ConsensusPathDB (http://cpdb.molgen.mpg.de) (35). The Pearson correlation between APOL1 and microRNA candidate in our sequencing data and TCGA-KIRC dataset were performed using GraphPad Prism 9 version 9.3.0 and ENCORI platform (https://rnasysu.com/encori/panMirCoExp.php), respectively (36). The survival curve of APOL1 and miR-30a were generated by using the Kaplan-Meier plot (https://kmplot.com/analysis/index.php?p=service&cancer=pancancer_rnaseq) (37).

HK-2, A-498 and Caki-1 cell lines were purchased from the Korea Cell Bank (South Korea). 786-O cell line was purchased from American Type Culture Collection (USA). All cell lines were cultured at 37°C and 5% CO₂ in RPMI 1640 medium (10-040-CV, Corning), except for A-498 cells which were cultured in DMEM (10-013-CV, Corning). All media were supplemented with 10% fetal bovine serum (35-015-CV, Corning) and Penicillin Streptomycin (15140-122, Gibco).

APOL1 shRNA plasmids (V3SH11240-230184401, V3SH11240-229063063, Dharmacon), APOL1 overexpression plasmid (RC217000L3, Origene) were co-transfected with lentiviral packaging vectors (pMD2G and pSPAX2) to HEK293T cells using Lipofector-pMAX (Aptabio, South Korea). The target cells were transduced by the harvested supernatants. The stable cell lines were generated by selection with puromycin.

Negative control miRNA and miR-30a (Bioneer, South Korea) were transfected to primary renal cancer cells using Lipofectamine RNAiMAX (Invitrogen). The cells were incubated for 3 days.

Total RNA from cell lysate were extracted by Ribospin II (314-150, Gene All) according to the manufacturer’s instructions. The total amount of RNA was measured by NanoDrop 2000 Spectrophotometer (Thermo Scientific). cDNAs were synthesized by ReverTra Ace qPCR RT Kit (FSQ-101, Toyobo) using the T100 Thermal Cycler (Bio-Rad). Real-time qPCR was performed using SYBR Green Realtime PCR Master Mix (QPK-201, Toyobo) with the CFX96 Real-Time System (Bio-Rad). The primers for APOL1 and β-actin used were APOL1-Forward: 5’-AAGTAAGCCCCTCGGTGACT-3’, APOL1-Reverse: 5’-GAGCTCATCTGCCTCATTCC-3’, β-actin-Forward: 5’-CTCCTTAATGTCACGCACGAT-3’, and β-actin-Reverse: 5’-CATGTACGTTGCTATCCAGGC-3’.

Total RNA and miRNA from tumors were extracted by miRNeasy Mini Kit (Qiagen) according to the manufacturer’s instructions. The total amount of RNA was measured by NanoDrop 2000 Spectrophotometer (Thermo Scientific). miRNAs were reverse transcripted by TaqMan MicroRNA Reverse Transcription Kit (4366597, Applied Biosystems) with miR-30a-3p-RT-primer: 5’-GTCGTATCCAGTCCAGGGACCGAGGACTGGATACGACGCTGC-3’ using the T100 Thermal Cycler (Bio-Rad). Real-time qPCR was performed using TaqMan Universal Master Mix II, no UNG (4440040, Applied Biosystems) with the CFX96 Real-Time System (Bio-Rad). The primers and probe for miR-30a-3p were miR-30a-3p-Forward: 5’-AGCCGCTTTCAGTCGGATGTTT-3’, miR-30a-3p-Reverse: 5’-TCCAGGGACCGAGGA-3’, miR-30a-3p-Probe: 5’-(FAM)-CTGGATACG ACGCTGC-(BHQ1)-3’ (38). The relative level of miR-30a-3p was normalized to RNU48 (assay ID: 001006, Applied Biosystems).

Total protein from cell lysates were extracted by RIPA Lysis and Extraction Buffer (89900, Thermo Scientific) supplemented with Halt Protease Inhibitor Cocktail (78410, Thermo Scientific) and Halt Phosphatase Inhibitor Cocktail (78420, Thermo Scientific). Protein concentrations were measured by Pierce BCA Protein Assay Kit (23225, Thermo Scientific).

For western blotting, the protein samples were mixed with 5X Laemmli loading buffer, and were heated at 95°C for 10 minutes. The ladder and the samples were loaded and resolved on polyacrylamide gel, and were transferred to a PVDF membrane (IPVH00010, Merck). Then the protein membranes were blocked in 5% skim milk for 1 hour at room temperature, and incubated with primary antibody at 4°C overnight. The following antibodies were used: Actin (A2103) from Sigma-Aldrich (USA); APOL1 (ab169952) and Kindlin 1 (ab68041) from abcam (USA); p-Akt (4060), Akt (3063), p-NFκB (3033), NFκB (8242), p-GSK3β (5558), GSK3β (4818), E-cadherin (3195), N-cadherin (4061), Snail (3879), Slug (9585), Integrin αV (4711), Integrin β3 (4702), p-FAK (8556), α-actinin (6487), p-Talin (5426), GAPDH (2118) from Cell Signaling (USA); Vinculin (sc-25336) from Santacruz (USA). Blots were washed with TBS/T three times, followed by an incubation with the secondary Ab (anti Rabbit or anti Mouse) (1:5000) for 1 hour at room temperature (RT). Blots were again washed with TBS/T and developed with the Clarity Max Western ECL Substrate (Bio-Rad). The protein bands were visualized by Chemidoc system (Bio-Rad).

Cells were stained with CellTrace CFSE reagent (C34554, Invitrogen) and were incubated for 0, 1, 3, 5 days. Each day, cells were harvested and fixed with 4% paraformaldehyde. Analysis was completed using an FACS Canto II Flow Cytometer (BD Life Sciences, USA) with 488−nm excitation and a 530/30-nm bandpass emission filter. The results were analyzed by FlowJo software version 10 (BD Life Sciences).

The cells were seeded on a 6-well plate at 80% confluency. After attaching overnight, a scratch was made by using the 200 µL tip. Images of four random microscopic fields under a light microscope (bright field) were taken at 0 hour and indicated hours with a 10X objective lens. The percentage of wound healing was calculated by TScratch program version 1.0 (ETH Zürich, Switzerland).

For transwell experiments, 100 µL of 0.4 mg/mL Matrigel (356230, Corning) and 750 µL of complete media (10% FBS) were respectively loaded into the upper chamber and lower chamber of Transwell with 8.0 µm Pore Polycarbonate Membrane Insert (3422, Corning), and incubated for 2 hours in the CO₂ incubator. After 2 hours, 1 x 10⁵ cells suspension in serum free media were added onto Matrigel-coated cell culture insert and incubate for indicated hours (16h, 18h or 24h). Then, the medium in upper chamber was removed and the insert was washed twice with PBS. Cells were fixed with 4% paraformaldehyde for 2 min at RT, and were washed twice with PBS. Cells were then permeablized with 100% methanol for 20 min at RT, and washed twice with PBS. After that, cells were stained with 0.4% crystal violet and incubated at room temperature for 15 min, and washed twice with PBS. Non-invaded cells were scrapped off with cotton swabs. Invasive cells were counted in four random microscopic fields under the Leica DMi8 microscope (bright field) with a 20X objective lens. For migration assay, the same protocol was performed without Matrigel-coated on the insert.

The mice experiment protocol was approved by the Institutional Animal Care and Use Committee of Soonchunhyang University (SCH20-0049) and performed in accordance with the Animal Facility of Soonchunhyang Institute of Medi-Bio Science guidelines. 12-week-old male BALC/c nude mice (Nara Biotech, South Korea) were housed in specific pathogen free conditions. The mice were intravenously injected via the tail vein with 2 x 10⁶ cells in 100 μL of 1X PBS of control or APOL1 overexpressing Caki-1 cells. After 8 weeks, these mice were euthanized by carbon dioxide inhalation, and their lungs were perfused and dissected for visual inspection of the metastatic nodules. The lungs were fixed in 4% paraformaldehyde for 24 hours and then were processed and embedded into a paraffin block. Paraffin sections (5 μM) were mounted on microscope slides and stained with Harris hematoxylin (YD Diagostics, South Korea) and eosin (BBC Bichemical, USA), and were imaged with Leica DMi8 microscope (bright field) with a 20X and 40X objective lens. 7-week-old male BALC/c nude mice (Nara Biotech, South Korea) were subcutaneously injected on the right flank with 10⁷ cells in 100 μL of 1X PBS of A-498, 786-O and Caki-1 cells. After 8 weeks, these mice were euthanized, and their tumors were collected for subsequent analysis. For immunohistochemistry, the tumors were fixed in 4% paraformaldehyde for 24 hours and then were processed and embedded into a paraffin block. Paraffin sections (5 μM) were mounted on microscope slides and proceeded with primary antibody APOL1 (sc-390440, Santa Cruz) and counterstained with Harris hematoxylin (YD Diagostics, South Korea). Images were taken by Leica DMi8 microscope (bright field) with a 20X objective lens.

Each experiment was performed in triplicates. Statistical analyses were performed using GraphPad Prism 9 version 9.3.0. All tests were two-tailed unpaired t tests. The p-value <0.05 was considered to be statistically significant.

To characterize the molecular mechanisms related to malignant properties of ccRCC, we first performed RNA sequencing (RNA-seq, GSE252600) on 9 pairs of tumors and adjacent normal tissues, with 3 of them having metastatic sites ( Table 1 ). The hierarchical clustering heatmap and the multidimensional analysis of the RNA-seq revealed the distinguished expression profiles of tumor from the normal tissues. The tumors were grouped together rather than with their normal adjacent tissues ( Figures 1A, B ). There were 2390 differentially expressed genes (DEG) obtained from the RNA-seq analysis, and among them 459 candidate genes overlapped with the TCGA-KIRC dataset ( Supplementary Figure S1A ). Among the top 20 upregulated genes that were highly expressed in ccRCC compared to normal tissues ( Supplementary Figure S1B ), TGFBI, PLOD2, TMEM45A, FCGR1C, HIST1H2BH, CCL5 and APOL1 were closely correlated to poor survival rate in ccRCC patients. Furthermore, TGFBI, PLOD2 and APOL1 had clear tumor progression patterns in the TCGA-KIRC dataset ( Figure 1C , Supplementary Figures S1C, D ). In an effort to identify which of the three genes we should investigate in, we also sequenced for differential expressions in microRNA expression (GSE252629), to see which gene was the most regulated. We overlapped differentially expressed miRNAs of tumors and adjacent normal tissues with miRNAs targeting these 3 genes. APOL1 and TGFBI were targeted by 28 and 24 miRNAs, respectively, whereas PLOD2 were targeted by 13 miRNAs (data not shown). We focused on the novelty of APOL1 as a potential cancer marker whose low expression strongly associated to poor survival rate at metastatic stage of ccRCC ( Figure 1D ), considering that TGFBI had already been identified as oncogene in RCC (39).

After realizing the potential of APOL1 as a candidate marker, we then examined its expression in various renal cell lines. We utilized immortalized proximal tubule epithelial cell line from normal adult human kidney HK-2, clear cell renal cell carcinoma cell lines A-498 and 786-O, and the metastatic cell line Caki-1. Both APOL1 transcript and protein levels were not detected in HK-2 but were robustly expressed in primary cancer cells and retained less expression in the metastatic cells ( Figures 1E, F ). To mimic the condition of primary and metastatic stages of ccRCC, we subcutaneously injected the primary cancer A-498 and 786-O cells and metastatic Caki-1 cells to nude mice, respectively. As a result, APOL1 was highly expressed in tumors of primary cancer cells and lowly expressed in tumor of metastatic cells ( Supplementary Figures S1E, F ). Furthermore, moderate level of APOL1 were observed in ccRCC patients at late stages ( Supplementary Figure S1G ). These observations were consistent with the fluctuating expression of APOL1 protein in ccRCC from Clinical Proteomic Tumor Analysis Consortium (CPTAC) ( Supplementary Figure S1H ).

To further investigate the miRNA that could regulate APOL1 expression and ccRCC metastasis, we analyzed small RNA sequencing data (miRNA-seq) of 9 patient samples. Hierarchical clustering heatmap and the multidimensional scaling of miRNA-seq revealed the distinguished expression profiles of the tumor and normal tissues ( Figures 2A, B ). There were 103 differently expressed miRNAs, and among them, 28 miRNAs were predicted to target the 3’UTR region of APOL1 by TargetScan, where 16 were upregulated and 12 were downregulated miRNAs ( Supplementary Figure S2A ). Due to the high expression of APOL1 in ccRCC, we focused on the miRNAs that had attenuated expression in ccRCC ( Supplementary Figure S2B ), and therefore examined the correlation between APOL1 and these 12 downregulated miRNAs. We found that miR-30a-3p expression was negatively correlated with APOL1 expression, and this was statistically significant in both miRNA-seq and the TCGA-KIRC dataset ( Figures 2C, D , Supplementary Figures S2C, D, E, F ). In in vivo model, the expression of miR-30a-3p was low in tumors of primary cancer cells and high in tumor of metastatic cells, which negatively correlated with APOL1 expression ( Supplementary Figure S2G ). Moreover, the low level of miR-30a-3p in ccRCC ( Figure 2C , Supplementary Figures S2H, I ) was associated with poor survival rates ( Figure 2E ). To validate whether the miR-30a-3p targeting 3’UTR of APOL1 regulated its expression in vitro, we transfected this miRNA to the high APOL1 expression renal cells, A-498 and 786-O ( Supplementary Figure S2J ). As expected, APOL1 protein expression was reduced ( Figure 2F ). Furthermore, we treated A-498 cells with the top 3 significantly downregulated miRNAs. As a result, these miRNAs did not specifically suppress APOL1 expression in renal cancer cells as the effect of miR-30a-3p ( Supplementary Figure S2K ). In conclusion, we found that the miRNAs and RNA worked in cohort to actively regulate ccRCC, and miR-30a-3p regulated APOL1 expression, which is in turn a potential hallmark of ccRCC.

To assess the role of APOL1 in tumor progression, we generated a stable knockdown of APOL1 in the primary cancer A-498 and 786-O cells and overexpressed APOL1 in the metastatic Caki-1 cells by lentiviral transfection. The renal cancer cells were efficiently knocked down and overexpressed at both transcript and protein levels ( Figures 3A–C , Supplementary Figure S3A ). The knockdown and overexpression of APOL1 did not affect the cell proliferation rate ( Supplementary Figure S3B ). Interestingly, the results from the wound healing and transwell assays showed that APOL1 knockdown promoted cell migration and invasion ( Figures 3D–G ). In coherence with these findings, APOL1 overexpression obstructed cell motility ( Figures 3H, I ).

To verify the function of APOL1 expression in metastatic progression in vivo, we intravenously injected the APOL1 overexpressing Caki-1 cells into the tail vein of nude mice. In line with the in vitro results, the number of lung metastatic nodules in the mice injected with APOL1 overexpressing Caki-1 cells were reduced compared to its control plasmids ( Figures 3J–L ). There was no difference in the weight between control and experimental mouse groups ( Supplementary Figure S3C ). Taken together, we found that APOL1 was critical in the progression of ccRCC, but was shown to block metastasis at later stages of cancer.

To gain insights into the mechanism by which APOL1 ameliorated renal malignant progression, we investigated the tumorigenic pathways by GSEA and gene enrichment analyses of 212 upregulated genes in ccRCC. As a result, Akt signaling pathway was shown to be the most significantly activated pathway by analysis with Wikipathways, Reactome, PID, KEGG database, among several other pathways in ccRCC. Namely, focal adhesion-Akt signaling pathway, negative regulation of the PI3K/Akt network, NF-kappa B signaling pathway were shown to be regulated ( Figures 4A, B ). The top terms for GO analysis were the intracellular functions of APOL1 rather than its secreted function, and was shown to be involved in regulation of cellular process and binding functions ( Supplementary Figure S4 ).

To determine whether APOL1 dependent Akt pathway could be involved in cell migration of tumor cells, we checked the Akt pathway in the APOL1 knockdown or overexpressing cells. Akt phosphorylation and the downstream targets including nuclear factor kappa light chain enhancer of activated B cells (NFκB) and glycogen synthase kinase 3 (GSK3β) were enhanced in the APOL1 knockdown A-498 and 786-O cells and downregulated in the APOL1 overexpressing Caki-1 cells ( Figure 4C ). Consistent with our findings, APOL1 suppression by miR-30a activated the Akt signaling pathway ( Figure 4D ).

Elevation of many integrin molecules may trigger metastasis in renal cell carcinoma (40, 41), moreover, integrins also activate Akt via FAK, paxillin, and integrin-linked kinase (42). Because APOL1 is known to interact with integrin αVβ3 to activate integrins in podocytes to dysregulate of actin cytoskeleton (43), we checked the expression of several focal adhesion proteins in APOL1 knocked down and overexpressing cells. In agreement with the increased cell migration in APOL1 knockdown cells, focal adhesion proteins integrin αVβ3, p-FAK, kindlin 1, vinculin, α-actinin, and p-Talin was upregulated. Consistent with these findings, APOL1 overexpression downregulated these focal adhesion proteins ( Figure 4E ).

We hypothesized that the EMT could also influence the migration and invasion of tumor cells. Thus, we examined the EMT related protein expression in APOL1 knockdown or overexpressing cells. Mesenchymal marker N-cadherin and transcriptional repressors of E-cadherin (Snail and Slug) were upregulated while epithelial marker E-cadherin was downregulated in APOL1 knockdown cells. On the other hand, APOL1 overexpression inhibited EMT and showed increased E-cadherin ( Figure 4F ).

In summary, suppression of APOL1 activated tumorigenic pathways, especially that of Akt, in which FAK contributed to increase tumor aggressiveness through the expression of adhesion molecules and EMT processes.