Introduction
Clear cell renal cell carcinoma (ccRCC) is the most prevalent subtype of kidney cancer, accounting for more than 75% of all renal cell carcinoma (RCC) cases. Histologically, ccRCC is characterized by the appearance of thin-walled cells filled with abundant lipids and glycogen [
1,
2]. Notably, if ccRCC tumors exceed 7 cm in size or have metastasized, the 5-year survival rate is less than 10%. Although significant advancements have been made in the treatment of ccRCC through the development of targeted drugs and immunotherapy, many patients with advanced ccRCC do not exhibit favorable responses to these therapeutic approaches, resulting in poor long-term prognosis [
3‐
5]. The pathogenesis of ccRCC is unclear, although it may be related to genetic and chromosomal abnormalities, including gene fusion [
1,
3]. Therefore, further mechanistic dissection of ccRCC growth and metastasis is imperative to develop innovative therapeutic strategies.
Circular RNAs (circRNAs) are a subtype of noncoding RNAs characterized by covalently closed, single-stranded transcripts lacking 5’ to 3’ polarity or poly A tails. These circRNAs exist in various eukaryotes and exhibit higher stability than their linear counterpart mRNAs [
6]. Owing to their abundance and stability, circRNAs regulate a wide range of biological activities. They function as competing endogenous RNAs (ceRNAs), interacting with RNA-binding proteins or modulating transcription and splicing [
7‐
9]. CircRNAs have also been implicated in the tumorigenesis and development of ccRCC. For instance, circCHST15 regulates the miR-125-5p/EIF4EBP1 axis to promote ccRCC progression [
10]. Our previous research demonstrated that circTNPO3 suppresses ccRCC proliferation and migration by interacting with IGF2BP2 [
11]. Furthermore, emerging evidence suggests that circRNAs, despite being classified as noncoding RNAs, can generate functional peptides or proteins. The expression of open reading frames (ORFs) in circRNAs is facilitated by internal ribosomal entry site (IRES) or N6-methyladenosine (m6A) modifications. CircEIF6, circMAP3K4, and circNSUN2 have been identified as translatable circRNAs that play important roles in cancer development [
12‐
14]. However, it remains unclear whether protein coding circRNAs are involved in ccRCC tumorigenesis, and functional protein products have not yet been identified.
The majority of ccRCC cases exhibit inactivation or mutation of the Von Hippel‒Lindau (VHL) gene [
15‐
17]. VHL is a substrate-recognition component of the E3 ligase complex. VHL loss leads to aberrant accumulation of hypoxia-inducible factor (HIF) proteins by regulating the ubiquitin‒proteasome system. HIF-1A and HIF-2A are the two most important HIF subtypes [
18,
19] that play distinct roles in ccRCC. HIF-2A plays a crucial oncogenic role in ccRCC, whereas HIF-1A appears to be a tumor suppressor [
20]. The oncogenic effects of HIF-2A in ccRCC are multifaceted, with a key role in enhancing the transcriptional activity of target genes [
21,
22].
The phosphoprotein phosphatase (PPP) family members include PP1, PP2A (PP2), PP2B (PP3), and PP4 [
23]. PP1 and PP2A, as the major eukaryotic protein phosphatases in the PPP family that are reported to participate in over 90% of serine/threonine dephosphorylation and regulate various cellular processes through the dephosphorylation of different substrates [
24]. Protein phosphatase 1 (PP1) is a member of the major serine/threonine-specific protein phosphatase family that is responsible for dephosphorylating its target proteins and participating in the regulation of various cellular processes [
25]. PP1 consists of three catalytic subunits, PPP1CA, PPP1CB, and PPP1CC. Notably, PPP1CA played a significant role in several cellular pathways. For instance, PPP1CA dephosphorylates B-Raf at both ERK phosphorylation sites to trigger PPP1CA/B-Raf/ERK pathway activation and promote prostate cancer (CaP) cell invasiveness [
26]. Furthermore, PPP1CA directly dephosphorylates AKT, thereby controlling gene expression, cell survival, and differentiation [
27,
28].
In the present study, we identified an upregulated circRNA in ccRCC, circPDHK1 (hsa_circ_0057090), and its expression was positively correlated with tumor grade in patients with ccRCC. It has been reported that PDHK1 encoded by linearPDHK1 promotes the growth and metastasis of cancer by regulating the aerobic glycolysis of tumor cells [
29,
30]. Through in vitro and in vivo functional investigations, we demonstrated that circPDHK1 enhanced the proliferation and metastatic ability of cultured ccRCC cells. Additionally, we revealed that circPDHK1 encodes a novel peptide, PDHK1-241aa, which is responsible for the oncogenic function of circPDHK1. Mechanistically, PDHK1-241aa interacts with PPP1CA, resulting in the nuclear translocation of PPP1CA. This process leads to reduced dephosphorylation of AKT and subsequent activation of the AKT-mTOR signaling pathway, thereby promoting ccRCC proliferation and metastasis.
Materials and methods
Patients and tissue samples
We collected 152 paired human ccRCC and adjacent nontumor tissues (4 paired tissues for screening phase and 148 paired tissues for validation phase) from the Southwest Hospital of the Army Medical University (Chongqing, China, 2019–2022). Clinical characteristics are listed in Supplementary Table
S1-
2. The diagnosis of ccRCC was confirmed via histopathological examination. This study was approved by the Ethics Committee of Southwest Hospital of Army Medical University (approval number: KY2020121). Written informed consent was obtained from each participant prior to the surgery. The tissues were immediately preserved in RNAlater (Thermo Fisher, Shanghai, China) for storage.
Microarray analysis
We performed microarray to identify differentially expressed circRNAs in 4 pairs of ccRCC and adjacent noncancerous tissues. RNA extraction, quality identification, analysis of differentially expressed circRNAs were performed by Shanghai Genomics Corporation (Shanghai, China) as described previously [
11].
Cell lines and cell culture
Caki-1, 786-O, RCC-JF, RCC-23 (ccRCC cell lines), HK-2 (adult renal tubular epithelial cells) and 293 T (normal human embryonic kidney cell lines) were purchased from Meisen CTCC (Zhejiang, China). All cell lines were authenticated via STR profiling. 293 T and HK-2 cells were cultured in DMEM (HyClone, Logan, Utah, USA), Caki-1 cells were cultured in McCoy’s 5A (Basal Media, Shanghai, China), and 786-O, RCC-23 and RCC-JF cells were cultured in RPMI-1640 (HyClone). All media were supplemented with 10% FBS (PAN, Aidenbach, Germany), 1% penicillin/streptomycin (Biosharp, China) and 1‰ Mycoplasma Elimination Reagent (Yeasen, Shanghai, China) to remove mycoplasma. The cells were cultured in 5% CO2 at 37 °C.
Actinomycin D (ActD) assay
Caki-1 and 786-O cells were seeded into 12-well plates the day before the experiment. cells were treated with 10 μg / mL actinomycin D (Genview, Beijing, China) for 0, 6, 12 or 24 h to inhibit RNA transcription. Total RNA was extracted from Caki-1 and 786-O cells using TRIzol reagent (Takara, Shiga, Japan) according to the manufacturer’s protocol. Followed by reverse transcription using PrimeScript RT (Takara, Shiga, Japan). The cDNA was amplified by qPCR using 2 × SP qPCR Mix (BioGround, China) on a CFX96 Real-Time PCR System (BioRad, USA). GAPDH were used as internal references and calculated using the 2−ΔΔCt method.
RNase R treatment
Total RNA was extracted from Caki-1 and 786-O cells using TRIzol reagent (Takara, Shiga, Japan) according to the manufacturer’s protocol. 5 μg RNA of Caki-1 and 786-O cells were treated with RNase R (6 U) (Lucigen, Middleton, WI, USA) at 37 °C for 10 min and 85 °C for 5 s according to the manufacturer’s instructions. For controls, 5 µg of total RNA was mock treated under the same conditions without the enzyme. The treated RNAs were subjected to reverse transcription using PrimeScript RT reagent Kit (Takara, Shiga, Japan) following the manufacturer’s instructions. The primers for circPDHK1 and linear PDHK1 were synthesized by Tsingke (Tsingke, China). The expression levels of circPDHK1 and linear PDHK1 were assessed using RT-qPCR with the Bio-Rad CFX96™ Real-Time PCR System (BioRad, USA).
Plasmids, siRNAs and cell transfection
The circPDHK1 siRNA (si-circPDHK1#1) and PPP1CA siRNA (si-PPP1CA) utilized in the experimental procedures were synthesized by GenePharma (Shanghai, China). Additionally, circPDHK1 siRNA (si-circPDHK1#2) was synthesized by Tsingke (Beijing, China). The specific siRNA sequences are listed in Supplementary Table
S3. The IRES and its truncated mutants were synthesized and cloned and inserted into the Luc2-Report plasmid (Geneseed, Guangdong, China) using
EcoRI and
BamHI sites. Full-length circPDHK1 and circPDHK1-3 × flag was synthesized and cloned and inserted into the pLC5-ciR vector (Geneseed) using
EcoRI and
BglII sites. The 3 × flag sequence is located after the start codon ATG in the predicted coding sequence (CDS) of circPDHK1. The circPDHK1-3 × flag ATG mutation overexpression vector was constructed by QuickMutation™ Site-Directed Mutagenesis Kit (Beyotime, Shanghai, China) following the manufacturer’s instructions. The human PPP1CA overexpression vector pCDH-PPP1CA was synthesized by Tsingke, whereas the human HIF-2A overexpression vector pCDNA3.1-HIF-2A was obtained from Unibio (Chongqing, China). The plasmids were transfected using a commercial DNA transfection reagent (Neofect, Beijing, China) following the manufacturer’s instructions. siRNAs were transfected using Lipo8000 transfection reagent (Beyotime, Shanghai, China) following the manufacturer’s instructions.
Cell counting kit-8 assay
Caki-1 and 786-O cells were collected and seeded into 96-well plates at a density of 3000 cells per well. Subsequently, 90 µL complete medium and 10 µL CCK-8 solution (TargetMol, USA) was added to each well at 0 h, 24 h, 48 h, 72 h, 96 h. The timing was started after the cells were attached to the well. The cells were incubated at 37 °C for 2 h. The absorbance at 450 nm was measured with Multiskan SkyHigh microplate reader (Thermo, China). The obtained data were analyzed using GraphPad 7.0 software (GraphPad, San Diego, CA, USA).
Caki-1 and 786-O cells (3 × 103 cells per well) were seeded in 12-well plates to assess their proliferative ability and incubated in complete medium for two weeks. Then, the cells were washed with PBS, fixed with 4% paraformaldehyde solution (Biosharp, China), and stained with 1% crystal violet staining solution (Beyotime, Shanghai, China), and the images were obtained using a scanner (EPSON, Japan).
5-ethynyl-2’-deoxyuridine (EdU) assay
EdU assays were used to assess cell proliferation capacity using the Click-iT EdU kit (Bioscience, Shanghai, China). Briefly, 2 × 104 ccRCC cells were seeded into 48-well plates and incubated for 16 h, followed by incubation with 25 µM EdU reagent at 37 °C for 2 h. The cells were fixed with 4% paraformaldehyde solution (Biosharp, China) for 15 min. After fixation, cells were incubated with 0.5% Triton X-100 for 10 min, stained with YF®594Azide Click-iT EdU for 30 min in the dark, and incubated with Hoechst 33,342 for 15 min. Cells were visualized by fluorescence microscopy (Lecai, Germany). The proliferation rate of cells was determined by calculating the ratio of EdU-positive nuclei to total nuclei.
Migration and invasion assay
Transwell plates (Corning, NY, USA) were used for the migration and invasion assays. Briefly, for transwell migration assay, 3 × 10
4 cells were plated in medium without serum in the upper chamber of a transwell with an 8.0 μm pore size. The lower chamber was filled with 500 µL complete medium. Likewise, for invasion assay, prepare Matrigel working solution proportionally (Matrigel: serum-free medium = 1:8) was added to the upper chamber and incubated at 37 °C for 2 h. Then, 6 × 10
5 cells were seeded in Matrigel-coated chambers with serum-free medium. After incubation for 24 h, the Transwell filters were fixed with 4% paraformaldehyde solution (Biosharp, China), stained with 1% crystal violet staining solution (Beyotime, Shanghai, China), and photographed by same microscopy. The number of migrated cells was counted using ImageJ software. Additionally, a wound healing assay was performed to detect ccRCC migration following a previously described protocol [
11,
31]. ccRCC cells transfected in 6-well plates were scratched 48 h after transfection. The medium need replaced with serum-free medium the night before the wound healing assay. Then, a sterile 10 µL plastic pipette tip (Labselect, China) was used to scratch through the single-cell layer, and images were captured after 0 h, 12 h and 24 h at the same place with a microscope (Nexscope, China). The cell migration distances were counted using ImageJ software.
Western blot
Patient tissues or cultured cells were lysed using RIPA buffer supplemented with a protease inhibitor cocktail and 1% PMSF. To isolate cytoplasmic and nuclear proteins, we used a Nuclear and Cytoplasmic Protein Extraction Kit (Beyotime, Shanghai, China) following the manufacturer’s instructions. Protein samples were separated by sodium dodecyl sulfate‒polyacrylamide gel electrophoresis (SDS‒PAGE) and subsequently transferred to polyvinylidene fluoride (PVDF) membranes (0.2 μm) by electroblotting. After they were blocked in 5% nonfat milk for 1 h, the membranes were incubated overnight at 4 ℃ with the indicated primary antibodies. Next, they were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies for 1 h at room temperature and visualized with the enhanced chemiluminescence (ECL) detection reagent (Bioscience, China). The antibodies used in the present study were as follows: rabbit anti-PDHK1/PDHK1-241aa (Novus, Centennial, CO, USA, NBP1-85955, 1:1000 dilution), rabbit anti-Flag (CST; Cell Signaling Technology, Danvers, MA, USA, #14793S, 1:1000 dilution), mouse anti-Flag (Beyotime, AF519-1, 1:1000 dilution), rabbit anti-PPP1CA (ABclonal, Wuhan, China, A12468, 1:1000 dilution), rabbit anti-HIF-2A (CST, #59973S, 1:1000 dilution), rabbit anti-pan-AKT (CST, #4691S, 1:1000 dilution), rabbit anti-phospho-AKT Thr308 (CST, #13038S, 1:1000 dilution), rabbit anti-mTOR (CST, #2983S, 1:1000 dilution), rabbit anti-phospho-mTOR Ser2448 (CST, #5536S, 1:1000 dilution), rabbit anti-β-actin (ZSGB-BIO, China, ZM-0001, 1:1000 dilution), mouse anti-GAPDH(Beyotime, AF0006, 1:1000 dilution), rabbit anti-α-tubulin (Beyotime, AF5012, 1:1000 dilution), mouse anti-histone H3 (Beyotime, AF0009, 1:1000 dilution), rabbit anti-PI3K (CST, Cell Signaling Technology, Danvers, MA, USA #4257S, 1:1000 dilution), rabbit anti-p-PI3K (CST; Cell Signaling Technology, Danvers, MA, USA, #4288 T, 1:1000 dilution), rabbit anti-PTEN (CST; Cell Signaling Technology, Danvers, MA, USA, #9559 T, 1:1000 dilution), rabbit anti-PDK1 (Proteintech, Wuhan, China, 17086–1-AP, 1:1000 dilution).
Immunofluorescence (IF) assay
IF assays were employed to investigate the subcellular localization of PDHK-241aa and PPP1CA following previously described methods [
32]. Briefly, the transfected cells were incubated in u-Slide 8 well plates (Ibidi, Germany) with 2 × 10
4 cells per well one day in advance. The cells were fixed with 4% paraformaldehyde solution (Biosharp, China) for 15 min. After fixation, the cells were incubated with 0.5% Triton X-100 for 10 min at room temperature. Cells were incubated with 10% goat serum for 60 min at room temperature and incubated with the primary antibody (1:200 dilution) at 4 ℃ for overnight. After incubation with secondary antibody (1:100 dilution) combined with fluorescent dye for 1 h in the dark. Next, the cells were incubated with DAPI for 10 min. Images of the slides were obtained using a fluorescence microscope (Leica SP8, Wetzlar, Germany).
Immunohistochemistry (IHC) assay
IHC assays of mouse xenograft tumor tissues were performed by Servicebio (Wuhan, China). Proliferative markers were detected using anti-Ki-67 (Bioss, Beijing, China). Images were obtained using a fluorescence microscopy (Lecai, Germany).
RNA fluorescence in situ hybridization (FISH) assay
FISH assays were employed to determine the subcellular localization of circPDHK1 following previously described protocols [
11]. Briefly, the Cy3-labeled circPDHK1 probe and FISH probe kit were synthesized by GenePharma (Shanghai, China. Supplementary Table
S4). The cells were seeded in u-Slide 8 well plates (Ibidi, Germany) at a density of 2 × 10
4 cells/well. Cells were grown to 50% confluence and then fixed with 4% paraformaldehyde. The hybridization experiments were performed using a fluorescence in situ hybridization kit (GenePharma, China) according to the manufacturer’s protocol. 18 s ribosomal RNAs (rRNAs) were used as positive controls. Then, cell nuclei were stained with DAPI and observed under a fluorescence confocal microscope (Leica SP8, Wetzlar, Germany).
Co‒immunoprecipitation (Co‒IP) assay
A Pierce Crosslink Magnetic IP/Co-IP Kit (Thermo Fisher, Shanghai, China) was used for immunoprecipitation experiments. Briefly, 1 × 107 cells were lysed with 1 mL Pierce™ IP lysis buffer (Thermo Scientific, USA) for 20 min on ice. Then, the supernatants were collected after centrifugation (13,000 g × 15 min). 25 μL protein A/G magnetic beads and antibodies or control IgG (5 µg antibody) were added to the centrifuge tubes and incubated 15 min at room temperature on a rotator. Then, using antibody-crosslinked magnetic beads by 20 µM Dextran sulfate sodium (DSS) at room temperature for 45 min on a rotator. The lysates were incubated with magnetic beads at 4 °C overnight on a rotator. 100 µL eluent buffer was added to retain the supernatant containing the target antigen for 5 min on a rotator, which was then neutralized and subjected to SDS‒PAGE and mass spectrometry (MS) analysis.
Liquid chromatography-mass spectrometry (LC–MS) analysis
The LC–MS analysis was performed by Sangon Biotech (Shanghai, China). Briefly, proteins were separated using SDS‒PAGE, followed by overnight trypsin digestion and subsequent desalting. The digested peptides were analyzed using a Q Exactive Plus LC–MS instrument (Thermo Fisher, Pittsburg, PA, USA), and the spectra were analyzed using Protein Pilot Ver. 4.5 supplied by the instrument.
Total RNA was extracted from cell lines and tissues using TRIzol reagent (Takara Bio, Shiga, Japan) according to the manufacturer’s protocol. In brief, tissues or cells were lysed in 1 ml of TRIzol reagent. Subsequently, 200 µl chloroform was added to the lysis buffer for 10 min, followed by centrifugation for 15 min at 13,000 g, 4 °C. After centrifugation, the aqueous phase containing the RNA was transferred to a new tube and isopropanol was added to the aqueous phase at an equal volume for 10 min at 4 °C, followed by centrifugation at 13,000 g for 10 min at 4 °C. Next, the pellet containing the RNA was retained, and 1 ml of 75% ethanol was added to the tube, followed by centrifugation at 13,000 g for 10 min at 4 °C. The pellet was retained and mixed with 30–50 µL RNase-free water. Then, the RNA was stored at − 80℃ until further use. The genomic DNA (gDNA) was extracted using the Universal Genomic DNA Kit (CWBIO, China) according to the manufacturer’s protocol. For analysis of circRNA or mRNA, cDNA was randomly or oligo(dT) primed from 500 ng of total RNA using PrimeScript RT reagent Kit (Takara, Shiga, Japan) following the manufacturer’s instructions. Real-time qPCR was performed using commercial 2 × SP qPCR Mix (Bioground, China). GAPDH or U6 was used as an endogenous control. The relative expression levels of target genes were calculated following the formula of 2
−ΔΔCt. All primers were synthesized by Tsingke (Beijing, China) and are listed in Supplementary Table
S5.
RNA-seq analysis
To determine downstream targets and pathways of circPDHK1, total RNA extracted from Caki-1 cells transfected with circPDHK1 siRNA, circPDHK1 overexpression vector, circPDHK1 ATG mutant vector or control siRNA/vector. Then, RNA-seq and KEGG analysis were performed by Shanghai Genomics Corporation (Shanghai, China) including library construction and computational analysis.
Subcellular fractionation assay
A PARIS™ Kit (Invitrogen, USA) was used for the subcellular fractionation assay. 5 × 106 cells were lysed with 300 µL cell separation buffer and incubated for 10 min on ice. The supernatant (cytoplasmic part) was added into RNase-free EP tube, and the precipitate (nuclear part) was added with 300 µL cell disruption buffer on ice. Then, the same amount of 2 × Lysis/Binding Solution and absolute ethanol were added to both the nuclear and cytoplasmic parts. After mixing, the mixture was added to a centrifuge column, centrifuged at 10,000 g for 1 min with the liquid was discarded. Finally, the eluate was added to eluting the RNA. The extracted RNA was subjected to reverse transcription and RT-qPCR to detect the expression of circRNA in the nucleus and cytoplasm. The U6 internal control was tested as follows: The expression of U6 was detected using the hairpin it™ U6 snRNA Quantitative detection kit (GenePharma, China). RNA reverse transcription system was as follows: 4 μL 5 × MMLV RT Buffer, 0.75 μL dNTP, 1.2 μL U6 snRNA RT primer mix, 0.2 μL MMLV Reverse Transcriptase, 1–3 μg RNA, RNase Free H2O To 20 μL. The reverse transcription reaction procedure was as follows: 25℃ for 30 min, 42℃ for 30 min, 85℃ for 5 min, 4℃. RT-qPCR was performed using the following system: 5 μL 2 × real-time PCR mix, 0.2 μL MiRNA/U6 snRNA specific primer, 0.1 μL Taq DNA polymerase, 1 μL RT product, RNase-Free H2O To 10 μL. The reaction procedure was as follows: 95°C for 3 min, (40 cycles of the following steps) 95℃ for 12 s, 62℃ for 40 s. The relative subcellular localization of circPDHK1 were calculated following the formula of 2−ΔΔCt.
Dual‑luciferase reporter assay
For IRES activity analysis, 293 T cells were transfected with the IRES and its truncated mutant vectors and Renilla expression plasmids per well using Neofect DNA transfection reagent in 96-well plates. After 24 h of transfection, the cells were lysed with passive lysis buffer (Promega, USA), and Firefly and Renilla luciferase activities were measured using the Dual-Glo luciferase assay kit (Promega, Madison, WI, USA). To verify whether HIF-2A binds to the PDHK1 promoter region. The sequence of PDHK1 promoter was subcloned into the Luc2-report Plamsid using BamHI and EcoRI sites. Caki-1 and 786-O cells were transfected with the Luc2-PDHK1 promoter vectors and Renilla expression plasmids per well using Neofect DNA transfection reagent in 96-well plates. Firefly and Renilla luciferase activities were measured using the Dual-Glo luciferase assay kit (Promega, Madison, WI, USA). The fluorescence intensity was determined using a fluorescence microplate reader (Thermo, USA). The ratio of Firefly/Renilla luminescence was calculated to determine the relative luciferase activity.
Chromatin immunoprecipitation(ChIP) assay
To investigate the interaction between HIF-2A and the promoter region of PDHK1, ChIP assays were performed using a Magnetic ChIP Kit (Merck, Darmstadt, Germany) according to the manufacturer’s procedure in our study. Briefly, Caki-1 and 786-O cells were incubated with 1% formaldehyde and neutralized with glycine solution. DNA fragments within the range of 200–1000 bp were obtained by ultrasonication. The nuclear lysate was immunoprecipitated using anti-HIF-2A or anti-IgG antibodies. Transcription factor-binding sites were predicted using Contra Ver. 3
http://bioit2.irc.ugent.be/contra/v3. The purified DNA fragments were further analyzed by RT-qPCR using specific primers (Supplementary Table
S5).
RNA immunoprecipitation (RIP) assay
RIP experiments were conducted using the Magna RNA-binding protein immunoprecipitation kit (Millipore, Burlington, MA, USA). The cells were processed at 1 × 10
7 cells/reaction density and lysed by adding lysis buffer containing protease and RNase inhibitors for 5 min. The magnetic beads were combined with 5 μg antibody and control IgG at room temperature, and the corresponding cell lysates were added; the samples were incubated overnight at 4 °C. The samples were subjected subsequently to RNA purification and protein extraction. The immunoprecipitated RNA obtained from Caki-1 cells was validated by RT-qPCR with specific primers (Supplementary Table
S5).
786-O and Caki-1 subcutaneous tumor-bearing male NSG mice were purchased from the Shanghai Model Organisms Center and randomly assigned to groups based on tumor size. In the siRNA xenograft tumor model group (n = 6, each group), cholesterol-modified circPDHK1-siRNA (10 nmol, GenePharma) was intratumorally injected once every two days for five consecutive weeks. Tumor volumes were measured every 2 d. Tumor tissues were excised and weighed after mice were euthanized, fixed, and stained for immunohistochemistry. In the xenograft tumor model overexpression group (n = 5, each group), a lentiviral vector containing cloned inserts (Oligobio, Beijing) was injected intratumorally for 18 days. Tumor volume measurements and processing were carried out as described above. For the lung metastasis model, stably transfected Caki-1 cells (6 × 106 cells/0.2 mL PBS) were injected into the tail vein of 4-week-old male mice. After 48 days, the mice were assessed using the Aura Spectral Instruments in vivo Imaging System and sacrificed. Lung tissues were excised, photographed, and stained with hematoxylin and eosin (H&E), and the lung metastatic nodes were observed. All the animal experiments were conducted according to the protocol approved by the Ethics Committee of Chongqing Medical University (approval number: 2022013).
Statistical analysis
One-way analysis of variance, Student’s t test, and chi-squared test were used to analyze the means between different groups. Statistical differences were calculated using Prism 7.0 (GraphPad, San Diego, CA, USA). The results are presented as the mean ± standard deviation (SD). In silico expression analyses were conducted with a cutoff of │log2FC│ ≥ 1 and weighted gene correlation network analysis (WGCNA) using per kilobase of exon per million mapped fragments (FPKM) > 1 and expressed in at least three samples. Statistical significance was set at P < 0.05.
Discussion
Accumulating evidence suggests that circRNAs play crucial roles in biological regulatory networks and are closely associated with various diseases, including liver, breast, and esophageal cancers and other carcinomas [
38‐
40]. However, only a few circRNAs have been implicated in ccRCC. For example, circME1 promotes aerobic glycolysis in ccRCC cells by enhancing the expression of its parental gene ME1, which results in ccRCC progression and the development of sunitinib resistance [
41]. Additionally, circPPP6R3 serves as a miR-1238-3p sponge, resulting in the upregulation of CD44 expression, thereby modulating the proliferation, migration, and invasion of ccRCC [
42]. Our previous study also found that circTNPO3 suppressed the proliferation and migration of ccRCC cells by directly interacting with IGF2BP2 and synergistically destabilizing SERPINH1 mRNA [
11]. Exploration of circRNA-based targeted therapeutic strategies has provided new insights into the treatment of ccRCC. Therefore, it is necessary to identify additional functional circRNAs in ccRCC.
Herein, we identified, for the first time, that circPDHK1 derived from exons 2 to 8 of PDHK1 exhibited high expression levels in ccRCC. Importantly, its expression levels were positively correlated with the WHO/ISUP stage, T stage, and M stage in ccRCC patients. Our functional investigations revealed that circPDHK1 promotes the proliferation and metastasis of ccRCC cells. CircRNAs derived from the PDHK1 host gene are implicated in tumorigenesis and metastasis. For instance, hsa_circ_0057104, an alternative splicing isoform of PDHK1, modulates the miR-628-3p/BPTF axis and degrades BIN1, thereby enhancing the growth, metastasis, and glycolysis of pancreatic cancer cells [
36]. CircPDHK1 (hsa_circ_0118104) is significantly correlated with tumor metastasis through the circPDHK1-miR-377-3P-NOTCH1 axis [
43]. These findings collectively demonstrate that circPDHK1 plays a pivotal role in ccRCC progression and is a potential biomarker for the diagnosis and prognosis of ccRCC.
Multiple studies have highlighted the significant roles of circRNAs in various functional mechanisms, including the miRNA sponge effect, interaction with RNA-binding proteins, formation of R-loops, and translation of functional proteins [
44,
45]. Although circRNAs have been defined as noncoding RNAs, some possess IRES or m
6A modifications and are no longer noncoding RNAs in the traditional sense but are special circular RNAs capable of encoding peptides through multiple mechanisms [
46‐
49]. We conducted a study based on the prediction results obtained using the circRNADb and Transcirc databases. Our findings revealed that circPDHK1 is translatable via an IRES and encodes a novel isoform, PDHK1-241aa. We confirmed that PDHK1-241aa was highly expressed in ccRCC tissues and predominantly present in the cytoplasm. We also showed that the PDHK1-241aa protein is the regulatory molecule involved in ccRCC cell proliferation and metastasis, since the nontranslatable ATG mut mutant was inactive. Thus, circPDHK1 can improve the proliferation and metastasis of ccRCC cells by encoding PDHK1-241aa, but further validation of the underlying mechanism is needed.
Emerging studies have identified the role of peptides encoded by circRNAs in tumorigenesis and cancer development. For instance, circHER2 encodes a novel protein, HER2-103aa, which forms a dimer with EGFR to phosphorylate AKT (Thr308) in certain triple-negative breast cancers (TNBCs). HER2-103aa promotes TNBC cell proliferation, invasion and tumorigenesis [
50]. CircSEMA4B encodes SEMA4B-211aa, which is expressed at a low level in breast cancer and exerts a tumor suppressor function in vivo and in vitro by complexing with free p85 to decrease the p85/p110 PI3-kinase complex, leading to the inhibition of AKT (Thr308) phosphorylation and the generation of PIP3 [
51]. In the present study, we observed a positive correlation between PDHK1-241aa levels and ccRCC cell proliferation and metastasis. Furthermore, these levels were associated with positive regulation of phosphorylation within the AKT-mTOR signaling pathway.
Multiple studies have shown that AKT-mTOR signaling is a promising target for cancer therapy [
33,
52,
53]. This pathway regulates cancer cell survival and growth, and mTOR is a direct substrate of AKT kinase. Augmented activation or amplification of the AKT-mTOR pathway inhibits cell apoptosis, accelerates cell proliferation, and induces metastasis. The hyperactivation of AKT has been implicated in various pathophysiological conditions [
54,
55]. Generally, AKT undergoes dual phosphorylation, and its PH domain binds to PIP3 and translocates to the cell membrane. Thr308 and Ser473 are two important phosphorylation sites for AKT [
56,
57]. Given that kinases, such as 3-phosphoinositide-dependent protein kinase 1 (PDPK1), are also recruited to the membrane via their PH domain and phosphorylate AKT at Thr308 in its active loop [
56,
58], we wondered whether PDHK1-241aa has a PH-like domain to phosphorylate AKT. Interestingly, our findings confirmed that PDHK1-241aa increases the phosphorylation of p-AKT (Thr308) and p-mTOR (Ser2448). However, the co-IP and LC‒MS data indicated that PDHK1-241aa did not bind to AKT or p-AKT (Thr308). Therefore, PDHK1-241aa indirectly regulates AKT phosphorylation.
We further explored the mechanism of PDHK1-241aa function and screened for potential interacting proteins using PDHK1-241aa-flag IP assays and LC‒MS. We focused on the serine/threonine-protein phosphatase PPP1CA (PP1-α catalytic subunit), which is closely associated with phosphorylation patterns and plays a significant role in regulating cellular processes [
28,
59,
60]. For instance, PPP1CA physically interacts with the NACA protein, dephosphorylating NACA to enhance c-Jun activity in osteoblasts [
60]. PPP1CA is a major phosphatase that can directly dephosphorylate AKT to regulate cell survival and differentiation [
27]. Through co-IP experiments, we confirmed that PDHK1-241aa directly binds to PPP1CA and inhibits AKT dephosphorylation. Interestingly, the interaction between PDHK1-241aa and PPP1CA promoted nuclear aggregation of PPP1CA, suggesting that PDHK1-241aa translocates PPP1CA from the cytoplasm to the nucleus. PPP1CA has been shown to shuttle between the nucleus and cytosol [
61]. Similarly, the small nucleolar RNA (snoRNA) SNORD12B can bind to PPP1CA and relocate it from the cytosol to the nucleus by disrupting the interactions between 14–3-3ζ and PPP1CA. This relocation of PPP1CA enhances AKT phosphorylation in the cytosol, activates the AKT-mTOR signaling pathway, and contributes to esophageal cancer tumorigenesis [
28]. Thus, our findings provide a novel mechanism by which PDHK1-241aa inhibits the inactivation of the AKT-mTOR signaling pathway. Furthermore, multiple studies have shown that abnormal activation of the AKT-mTOR axis is a vital cause of tyrosine kinase inhibitors (TKI) resistance in ccRCC [
62]. Due to the regulatory effect of circPDHK1 on AKT-mTOR signaling pathway, we discovered that silencing circPDHK1 can partially enhance the anti-tumor efficacy and sensitivity of TKI drugs or AKT-mTOR inhibitors for ccRCC in vitro.
CircRNA biogenesis regulation is complex [
63‐
65]. In addition to back-splicing factors, transcription regulatory factors also affect the abundance of circRNAs [
66]. It has been reported that circRNA expression can be modulated through activation of host gene transcription. For example, Twist1 promotes Cul2 transcription by binding to the Cul2 promoter. This process upregulates circCul2 expression to accelerate endothelial-mesenchymal transition (EMT) in hepatocellular carcinoma [
67]. Sp1 directly binds to the B3 region of the Hipk2 promoter and upregulates circHipk2 and linear Hipk2 expression in myoblasts [
68]. Interestingly, hsa_circ_0057104, which is also derived from PDHK1, was activated by HIF-1A at the transcriptional level. Furthermore, HIF activity is necessary for increased exon inclusion and the transcription of linear PDHK1 [
69]. Mutational inactivation of VHL is considered the initial genetic event in most ccRCC cases, which leads to the accumulation of HIF transcription factors. HIF-2A has been generally implicated as an oncoprotein, whereas HIF-1A is an inhibitor of ccRCC tumorigenesis and development [
22]. HIF-2A is a major player in ccRCC pathogenesis and represents a potential therapeutic target [
70]. We investigated whether HIF-2A upregulates circPDHK1 expression by interacting with the genomic PDHK1 promoter. ChIP assays indicated specific interactive regions of HIF-2A at the PDHK1 promoter, and Pearson’s correlation analysis confirmed the association between circPDHK1 and HIF-2A expression in ccRCC tissue samples. Our study provides new insights into the mechanism underlying circPDHK1 generation and the etiology of ccRCC.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.