Skip to main content
Download PDF

MicroRNA profiling identifies VHL/HIF-2α dependent miR-2355-5p as a key modulator of clear cell Renal cell carcinoma tumor growth

Cancer Cell International volume 25, Article number: 71 (2025) Cite this article

Abstract

Inactivation of the von Hippel-Lindau (VHL) tumor suppressor gene is one of the first truncal events in clear cell Renal Cell Carcinoma (ccRCC) tumorigenesis. The accumulation of Hypoxia Induced Factor (HIFα) resulting from VHL loss can promote ccRCC tumorigenesis by regulating microRNA (miRNA) expression. Here, we performed miRNA profiling and high-throughput analysis to identify a panel of VHL-dependent miRNAs in ccRCC. Validation of these miRNAs revealed the overexpression of miR-2355-5p in ccRCC cell models and primary tumors. Moreover, we showed a significant increase in circulating miR-2355-5p in plasma from patients with ccRCC. Mechanistically, miR-2355-5p overexpression was confirmed to be HIF-2α dependent. Targeting miR-2355-5p with the CRISPR/Cas9 system not only negatively disrupted the ability of ccRCC cells to stimulate angiogenesis but also decreased cell proliferation and drastically reduced tumor growth in mouse xenograft models. Finally, a miR-2355-5p pulldown assay identified five tumor suppressor genes, ACO1, BTG2, CMTM4, SLIT2, and WDFY2, as potential targets. All five genes were significantly downregulated in ccRCC tumors and mouse xenograft tumors. The results from this research demonstrate the oncogenic ability of miR-2355-5p and shed light on the possible mechanism by which this miRNA controls angiogenesis and tumor growth in VHL-deficient ccRCC.

Introduction

In 2020, more than 430,000 people were diagnosed with kidney cancer worldwide, accounting for 2.2% of all cancers [1]. Clear Cell Renal Cell Carcinoma (ccRCC) is the most prevalent subtype of kidney cancer and represents more than 70% of all diagnoses [2]. Currently, surgery is the first-line treatment for localized ccRCC and has a 5-year survival rate of over 80% [3]. Unfortunately, approximately 30% of ccRCC patients still progress to advanced metastatic disease, and one-third already present with metastases at the time of diagnosis [4]. Advances in targeted therapies (e.g. Sunitinib, Axitinib) and newly approved immunotherapies (e.g. Nivolumab, Ipilimumab) have improved the treatment of metastatic ccRCC, although the 5-year survival rate still rarely exceeds 20% [5]. Therefore, a better understanding of ccRCC carcinogenesis is crucial for enhancing patient treatment options.

Genomic alterations such as the loss of the short arm of chromosome 3 and the inactivation of the von Hippel-Lindau (VHL) tumor suppressor gene are characterized as pivotal truncal events of ccRCC tumorigenesis [6]. VHL is part of an E3 ubiquitin ligase recognition complex formed by Elongin B, C, and Cullin-2 (VBC-Cul2), which target specific proteins for proteasomal degradation. Indeed, VHL is mostly known for its ability to negatively regulate the Hypoxia Inducible Factor subunits alpha (HIF-1α and HIF-2α) in an oxygen-dependent manner [7]. In an oxygen-rich environment, HIFα is hydroxylated by prolyl hydroxylases (PHDs) to favor its binding with VHL and promote its polyubiquitination and subsequent proteasomal degradation [8, 9]. However, when oxygen is lacking, PHD inhibition prevents the recognition of HIFa by VHL, causing HIFα stabilization and dimerization with HIFβ to form a functional transcription factor that translocates to the nucleus. Therefore, VHL deficiency observed in ccRCC triggers chronic HIFα accumulation, creating a pseudohypoxic environment in cells. Consequently, HIF binds to Hypoxia Response Elements (HREs) located in the promoter regions of hundreds of genes involved in multiple cellular pathways, such as angiogenesis and glycolytic metabolism [10]. Additionally, the transcriptional impact of HIFα can also be amplified by modulating the expression of microRNAs (miRNAs, miRs) [11].

MiRNAs are a family of small noncoding regulatory RNAs that are approximately 20 nucleotides long. Mature miRNAs are bound to Argonaute (AGO) proteins, which form the RNA-Induced Silencing Complex (RISC) [12]. The main role of RISC is to negatively regulate the translation of specific mRNA transcripts through binding the 5’ SEED region (nucleotide 2–8) of the miRNA and a complementary sequence found in the 3’ untranslated region (UTR) of the mRNA [13]. Binding of the miRNA will cause translational repression and ultimately mRNA decay, resulting in decreased protein expression [14]. Alterations in miRNA expression have been linked to important hallmarks of cancer development and progression, including ccRCC [15]. For example, miR-210-3p is tightly linked to hypoxia and serves as a potential biomarker in ccRCC [16]. On the other hand, miR-224-5p and miR-497-5p have been shown to regulate PD-L1 expression in ccRCC, suggesting a potential combinatory effect with immunotherapies [17, 18].

Our research project aimed to identify and characterize dysregulated miRNAs following VHL inactivation in ccRCC. Our high throughput miRNA profiling analysis uncovered the miR-2355-5p in ccRCC. Previous studies have shown that miR-2355-5p is dysregulated in a few cancers, such as Gastric Cancer (GC), Cervical Cancer (CC), and Esophageal Squamous Cell Carcinoma (ESCC) [19,20,21]. However, little is known about the functions of miR-2355-5p, particularly in cancer cells. In addition to renal cancer cell models, we demonstrated that miR-2355-5p is overexpressed in tumor and plasma samples from patients with ccRCC compared to those from healthy individuals. Furthermore, the loss of miR-2355-5p slowed the cell proliferation rate and markedly inhibited tumor growth in mouse xenograft models. Finally, five tumor suppressor genes, ACO1, BTG2, CMTM4, SLIT2, and WDFY2, were identified as potential targets of miR-2355-5p. The results from this project open new routes by which VHL inactivation promotes miRNA-related carcinogenesis.

Materials and methods

Cell culture

Human ccRCC cell lines (RCC4, RCC10, and 786-0) and their isogenic counterparts stably expressing VHL were kindly provided by Amato J. Giaccia (Stanford University, CA). A498 cells were a gift from Réjean Lapointe (CRCHUM, Montreal, QC). All cell lines were tested for mycoplasma, and authentication was performed by short tandem repeat (STR) DNA profiling at Genetica DNA Laboratories (Burlington, NC, USA). HUVECs were gifted by Borhane Annabi (UQAM, Montreal). RCC4, RCC10, and 786-0 cells were maintained in DMEM/high glucose supplemented with 10% Fetal Bovine Serum (FBS), 2mM L-glutamine (all from Wisent Bioproducts, QC) and 1mM sodium pyruvate (Cytiva) while A498 cells were maintained in EMEM (Wisent Bioproducts) supplemented with 10% FBS, 1mM sodium pyruvate and MEM Non-Essential Amino Acids (ThermoScientific). HUVECs were maintained in Endothelial Cell Growth Base Media (R&D Systems, Toronto, ON). Cells were cultured at 37 °C in a humidified incubator with 5% CO2. For the hypoxia experiments, cells were incubated in a H35 hypoxystation (Don Whitley Scientific) under 5% CO2 and 1% O2.

miRNA profiling by next-generation sequencing

Small RNAs were isolated using mirPremier microRNA Isolation Kit (MilliporeSigma, Oakville, ON) according to the manufacturer’s instructions. Small RNAs were loaded on an Ion PI Chip v2 and sequenced with an Ion Proton Sequencer using the 260-flow parameter. Sequencing data were filtered by eliminating reads of < 16 and > 60 nucleotides. Reads with bases showing Q scores < 20 were eliminated. The data were normalized using the trimmed mean of M-values (TMM) method and analyzed with the EdgeR package from R (Version 4.0.4) for differential expression using the Generalized Linear Model likelihood ratio test (GLM-LRT). MiRNAs were considered significantly differentially expressed (DE) when the absolute Fold Change (FC) was ≥ 1.5 and the False Discovery Rate (FDR) was < 0.05.

The cancer genome atlas (TCGA) analysis

MiRNA/RNA sequencing data and clinical information from the TCGA-KIRC and CPTAC3 projects were retrieved from the Genomic Data Commons (GDC) Data portal (National Cancer Institute). Only patients whose primary tumor and paired normal adjacent tissue data were available were included in the analyses. Raw Read counts were analyzed with the EdgeR package in R (Version 4.0.4) as described above.

Total RNA extraction and quantification by RT-qPCR

Total RNA was isolated using TRIzol reagent (Invitrogen) and purified using Monarch Total RNA Miniprep Kit (New England BioLabs (NEB), Whitby, ON) according to the manufacturer’s instructions. For miRNA expression, 100ng of total RNA was subjected to reverse transcription with MultiScribe reverse transcriptase and specific TaqMan probes (Applied Biosystems, Supplementary Table S6). For mRNA quantification, 1 µg of total RNA was subjected to reverse transcription with SuperScript™ III Reverse Transcriptase (Invitrogen) using oligo dT (20-mer) primers (Integrated DNA Technology). SYBR™ Green PCR Master Mix (Applied Biosystem) and specific primers (PrimerBank, Supplementary Table S7) were used for qPCR. Fold change was calculated with the delta-delta Cq method following normalization with RNU44 or RNU48 (miRNA), RPLPO and/or HPRT1 (mRNA).

miRNA/RNA extraction from clinical samples

Consenting ccRCC patients were recruited at the Centre Hospitalier Universitaire Sherbrooke (#2021–4061). For plasma samples, blood was collected into K2EDTA blood collection tubes and centrifuged at 1,500 × g for 15 min. The clear upper phase was centrifuged at 2,500 × g for 15 min to remove any remaining cells and stored at -80 °C. Plasma from healthy individuals was purchased from Innovative Research (IPLASK2E10). To compare with clinical distribution of ccRCC, plasma from 8 men and 4 women (Ratio 2:1) over 50 years old was selected. miRNeasy Serum/Plasma Advanced Kit (Qiagen) was used to isolate RNA. Cel-miR-39-3p Spike-in control (Qiagen) was added during the extraction. MiRNA levels were quantified as described above and normalized to those of Cel-miR-39-3p. For tissue samples, total RNA was isolated using TRIzol reagent. Samples were vortexed, and tissues were further disrupted using a homogenizer (SCILOGEX D160). RNA was purified using Monarch Total RNA Miniprep Kit. RNU48 and RPLPO + HPRT1 were used for normalization of miRNA and mRNA expression, respectively.

Western blot analysis

Total proteins were extracted from cells with M-PER lysis buffer (50mM TRIS (HCl) pH 7.5, 200mM NaCl, 0.25% Triton X-100, and 10% Glycerol) supplemented with protease inhibitor (Sigma-Aldrich) and phosphatase inhibitor cocktail (Selleckchem). Proteins were separated on SDS-PAGE gels and transferred onto a 0.45-micron PVDF membrane as previously described [22]. Immunoblots were incubated with specific primary antibodies against VHL (#68547), HIF-1α (#14179) or HIF-2α (#59973) (Cell Signaling) or against β-actin (Santa Cruz Biotechnologies). Immunoblots were washed and incubated with HRP-conjugated secondary antibodies (Jackson ImmunoResearch). Proteins were imaged on a ChemiDoc MP Imaging system (Bio-Rad) with ECL Prime Detection Reagent (Cytiva).

Short-Hairpin, CRISPR/Cas9 and overexpression models

HIF-1α and HIF-2α stable knockdown constructs were generated using the Human pLKO.1 lentiviral shRNA target gene sets (Open Biosystem #TRCN3808 and #TRCN3804). For CRISPR constructs, guide RNAs (gRNAs) were designed using Benchling, CHOPCHOP, and CRISPR design tools (https://zlab.bio/guide-design-resources). Designed gRNA primer pairs were inserted into the lentiCRISPRv2 plasmid, a gift from Feng Zhang (Addgene plasmid, #52961). Overexpression of miR 2355-5p was accomplished using a miR 2355-5p precursor plasmid (Genecopoeia, #HmiR0909-MR03) and a control plasmid (Genecopoeia, #CmiR0001-MR03). Lentiviruses were produced by co-transfection of the target plasmid and third-generation lentiviral packaging mix in 293T cells. Serial dilutions were used for clonal isolation.

T7 endonuclease

Genomic DNA was extracted using Monarch Genomic DNA Purification Kit (NEB). Approximately 1 kb flanking the gRNAs target site (Forward Primer: 5’-TGATTGTGACTCTTCCTGAGCC-3’, Reverse Primer: 5’-CGCCTCAGCTTACCCTTTCTT-3’) was amplified using Q5 DNA Polymerase (NEB). Amplicons were subjected to T7 Endonuclease I (NEB) and separated on a 1% agarose gel. The gel was photographed using a ChemiDoc MP Imaging system (Bio-Rad).

Clonogenic assay, cell proliferation assay, and cell cycle analysis

For the clonogenic assay, 300 cells were seeded in triplicate and incubated at 37 °C for 7 (786-0) or 10 (A498) days. Cells were fixed and stained with 0.5% Crystal Violet in methanol/Acidic Acid (200:1). Colonies were manually counted from technical and biological replicates of 4 independent experiments. Cell proliferation was assessed by seeding 20,000 cells into 12-well plates in duplicate. Cells were trypsinized and manually counted with Trypan blue stain at different time points. For cell cycle analysis, cells were fixed in 70% ethanol and stained with FxCycle™ PI/RNase Staining Solution (Life technologies) according to the manufacturer’s instructions. Cells were processed and analyzed on an Attune NxT flow cytometer (ThermoScientific) with Kaluza analysis software.

Migration assay

For the migration assay, 40,000 cells were seeded into a Transwell chamber (Greiner Bio-one, Monroe, NC) pre-treated with 0.1% gelatin (MilliporeSigma) and DMEM 0.5% FBS. DMEM supplemented with 10% FBS was used as a chemoattractant. After 16 h, nonmigrated cells were removed with a cotton swab, and migrated cells were fixed with 4% formaldehyde, stained with Crystal Violet, and photographed on a Nikon TMS inverted microscope. Cells were manually counted from at least 5 images per technical replicate.

Tube formation and angiogenesis profiling

Conditioned Media (CM) was acquired by starving ccRCC cells in 0.5% FBS medium for 48 h. For the tube formation assay, 20,000 HUVECs were seeded with 100µL of CM in triplicate on a 96-well plate covered with 50 µL of Reduced Growth Factor Basement Membrane Extract (R&D Systems). After 10 h, cells were photographed on a Nikon TMS inverted microscope, and the number of tubes (Meshes) and total tube size (Total meshes area) were calculated with the Angiogenesis Analyzer plugin in ImageJ [23]. Angiogenic profiling was performed with the Proteome Profiler Human Angiogenesis Array Kit (R&D Systems) following the manufacturer’s recommendation. The membranes were photographed on a ChemiDoc MP Imaging system at low and high exposure. Dot intensities were measured with ImageLab (Bio-Rad).

miRNA pulldown, total RNA sequencing and functional analysis

The miR-2355-5p pulldown assay was performed as described by Wani and Cloonan [24]. Briefly, a synthetic miR-2355-5p was designed with a biotin molecule at the 3’ end of the miR-2355-5p strand with a C6 linker (Forward Primer: 5’-AUCCCCAGAUACAAUGGACAA-C6-Biotin-3’, Reverse Primer: 5’-GUCCAUUGUAUCUGGGGUUAU-3’). The miR-2355-5p-biotin was transfected into 786-0 cells using Dharmafect 1. After capture of the miR-2355-5p-biotin and linked-transcripts with streptavidin-coated magnetic beads, total RNA was isolated using TRIzol reagent and purified with Monarch Total RNA Miniprep Kit. Samples were further purified and concentrated with RNA clean & concentrator-5 (Zymo Research). The RNA concentration and integrity were assessed with a High Sensitivity RNA kit (Agilent, ON) on a Fragment Analyzer (Agilent). RNA library construction was performed using a Takara SMARTer kit (TaKaRa, CA) and sequencing was conducted on a NovaSeq 6000 (Illumina) at the McGill Genome Centre (Montreal, QC, GSE246950). Raw Reads counts were analyzed with the DESeq2 package in R (Version 4.0.4). RNA samples extracted from beads were compared to RNA retrieved from leftover cell lysate samples. Significantly enriched transcripts were identified using a one-tailed Wald test with a cutoff set at an FC ≥ 1.5 and a p adjusted value (padj) < 0.05. Functional analysis was performed with the GOseq package in R.

Luciferase target reporter assay

A perfect complementary sequence to miR-2355-5p (FP: 5’-AATTCTGCTGAGGGTTGTCCATTGTATCTGGGGATCATCCACCGCACGA-3’, RP: 5’-CTAGTCGTGCGGTGGATGATCCCCAGATACAATGGACAACCCTCAGCAG-3’) was inserted into the 3’UTR of a firefly luciferase expressing plasmid (MT06 Target Reporter plasmid, Genecopeia) using restriction enzymes. The synthetic miR-2355-5p-Biotin or negative mimic control (Ambion) was transfected into 786-0 cells using Dharmafect 1 (Horizon Discovery) at different concentrations. The luciferase plasmid was transfected with Dharmafect kb 24 h later. The next day, the concentration of firefly Luciferase was measured using the Luc-Pair Duo-Luciferase Assay Kit 2.0 (Genecopoeia). The concentration of Renilla luciferase was simultaneously measured for normalization.

Mouse xenograft

All animal experiments were approved by the Canadian Council on Animal Care from our institution (#19 − 01). Nude (NU/J, #002019) and NSG (NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ) immunodeficient mice aged 5–6 weeks were purchased from The Jackson Laboratory. Cells were subcutaneously injected into the dorsal flank at a concentration of 5 × 106 cells in 150 µl of PBS. Tumors were measured every other day with a caliper 2 weeks postinjection, until the tumor volume exceeded 1cm3. Tumor volume was calculated as \(\:(Lenght\:\times\:Width\:\times\:Height\:\times\:0.5)\). Body weight was measured weekly, and the mice were monitored for signs of distress. The tumor tissues were isolated for RNA extraction as described above.

Statistical analysis

Statistical analysis was performed using GraphPad Prism 9 or R. Data are expressed as the mean ± standard error of the mean (SEM). Statistical tests are indicated in the figure legends. All experiments were performed at least three times, and all samples were analyzed in triplicate, unless specified otherwise. A p value < 0.05 indicated statistical significance.

Results

VHL-regulated miRNA profiling

We performed small RNA sequencing in the VHL-deficient ccRCC cell line, RCC4, and its isogenic counterparts expressing a functioning VHL protein, RCC4 VHL. Of the 352 miRNAs quantified, 75 were differentially expressed (DE), including 47 upregulated and 28 downregulated miRNAs in the absence of VHL (Fig. 1A, Supplementary Table S1). To test the clinical implications of these results, we used miRNA sequencing data from ccRCC patients (TCGA-KIRC and CPTAC3) from The Cancer Genome Atlas (TCGA). Only patients whose miRNAseq was performed on both primary tumor and normal adjacent tissue were selected (TCGA-KIRC N = 71, CPTAC3 N = 148) (Supplementary Table S2). Paired differential expression analysis comparing tumor samples to normal adjacent tissue was used, which considered tissue origin (tumors vs. normal tissue) and patient origin (Patient ID) in the design to account for patient’s variability. We identified 167 DE miRNAs in tumor samples from the TCGA-KIRC cohort (78 upregulated, 89 downregulated) and 347 DE miRNAs in the CPTAC3 cohort (157 upregulated, 190 downregulated) (Fig. 1B-C). Matching all three sequencing analyses revealed 45 DE miRNAs identified in RCC4 cells and in at least 1 TCGA cohort (Fig. 1D, Supplementary Table S1). Of these 45 miRNAs, 11 were selected for further analysis. Alternatively, 5 miRNAs identified only in our cell-based miRNAseq data and 5 unique to patient-based sequencing data were also selected (Fig. 1D).

Fig. 1
figure 1

Profiling of VHL-regulated miRNAs in ccRCC. A Heatmap showing significant DE miRNAs in RCC4 cells based on VHL status. MiRNAs with an absolute FC ≥ 1.5 and an FDR ≤ 0.05 were considered significantly DE. Bold miRNAs are also significantly DE in ccRCC tumors (B-C). B-C Volcano plot summarizing the miRNA expression profiles of ccRCC patients obtained from the TCGA-KIRC (B) and CPTAC3 (C) databases. Only patients whose miRNA sequencing was performed on both paired primary tumors and normal adjacent tissues were considered (TCGA-KIRC: N = 71, CPTAC3: N = 148). Significant miRNAs (red) were chosen as described in A. D Schematic representation of the similar miRNAs identified between analyses. The blue circles represent the number of miRNAs selected for further analysis. E Immunoblot analysis of VHL, HIF-1α, HIF-2α, and β-actin in RCC4, RCC10, and 786-0 cell lines with or without VHL. β-actin and total protein were used as loading controls. F RT-qPCR quantification of selected miRNAs in the cell lines described in E (N = 3–4). G Expression of selected miRNAs in ccRCC patients (CPTAC3) based on tumor VHL status. H-I Paired expression of miR-2355-5p in ccRCC patients from the TCGA-KIRC (H) and CPTAC3 (I) databases. MiR-2355-5p was deemed upregulated when \(\:\frac{Tumor\:CPM}{Paired\:Normal\:Tissue\:CPM}>1.5\). J RT-qPCR quantification of miR-2355-5p in ccRCC primary tissue compared to normal adjacent tissue (N = 13). K RT-qPCR quantification of miR-2355-5p in plasma from ccRCC patients (N = 43) compared to plasma from healthy individuals (N = 12). MiRNA expression was normalized with RNU44 (F), RNU48 (J) or Cel-miR-39-3p spike-in (K). The data are presented as the Mean ± SEM. Statistical analysis was performed using two-tailed unpaired Student’s t-test (F), Wilcoxon signed-rank tests (G, H, I), or Mann-Whitney U tests (J, K). (*P < 0.05, **P < 0.01, ***P < 0.001).

Full size image

Thus, to further validate the VHL-miRNA relationship, we quantified the expression of the selected miRNAs in three VHL-deficient ccRCC cell lines (RCC4, RCC10, and 786-0) and their matched cells expressing a functional wild-type VHL (RCC4 VHL, RCC10 VHL, and 786-0 VHL) (Fig. 1E). Most miRNAs showed similar alterations following the reintroduction of VHL in at least 2 ccRCC cell lines, which supported our miRNAseq analysis, although, few miRNAs seemed to be less affected (miR-9-5p, miR-99a-5p, miR-218-5p, miR-130b-3p, and miR-181a-3p), and 3 (miR-34a-5p, miR-138-5p, and miR-1301-3p) even exhibited cell-specific expression patterns based on their VHL status (Fig. 1F, Fig. S1A). In parallel, we also evaluated the expression of the selected miRNAs in ccRCC patient tumors based on their VHL status. Of the 11 matched miRNAs, 7 (miR-9-5p, miR-495-3p, miR-381-3p, miR-154-3p, miR-210-3p, miR-2355-5p, and miR-138-5p) were significantly DE between VHL-mutated tumors and non-mutated tumors (Fig. 1G). Of these 7 miRNAs, only 2 (miR-210-3p and miR-2355-5p) showed a consistent upregulation pattern in both cell and patient analyses based on their VHL status. Since previous studies have already shown the VHL-miR-210-3p relationship in ccRCC, we further focused on miR-2355-5p [25, 26]. Additionally, miR-2355-5p was overexpressed in almost all patients in both cohorts (TCGA-KIRC: 91.55% and CPTAC3: 85.96%) and was among the most consistently upregulated miRNAs (Fig. 1H-I, Supplementary Tables S3 and S4). Using tissues from our cohort of consented patients with ccRCC, we confirmed the overexpression of miR-2355-5p in primary tumors, which corroborated with all previous analyses (Supplementary Table S5, Fig. 1J). Furthermore, greater significant levels of miR-2355-5p were detected in the plasma of ccRCC patients than in healthy individuals (Fig. 1K). However, no further difference was observed between plasma from individuals with primary ccRCC and plasma from individuals with more advanced metastatic disease, which mirrored TCGA tissue data (Fig. S1B-D). Nonetheless, these results demonstrate a VHL-dependent overexpression of miR-2355-5p in ccRCC, which could potentially be used as a new biomarker.

miR-2355-5p overexpression is HIF-2α dependent

To determine the relationship between VHL and miR-2355-5p, we knocked down VHL in 786-0 VHL cells using the CRISPR/Cas9 system (Fig. 2A) [22]. As anticipated, miR-2355-5p expression significantly increased in Cr.VHL cells (Fig. 2B), which mirrored the increase previously observed between 786 and 0 and 786-0 VHL cells (Fig. 1F). In parallel, we also observed an increase in the expression of miR-210-3p (Fig. 2B). To investigate whether miR-2355-5p overexpression was HIFα dependent, we used short hairpin RNA (shRNA) to decrease HIF-1α or HIF-2α expression in RCC4 cells (Fig. 2C). Strikingly, targeting HIF-2α decreased miR-2355-5p expression to RCC4 VHL levels, while targeting HIF-1α did not alter its expression (Fig. 2D). In contrast, miR-210-3p was downregulated following the decrease in either HIF-1α or HIF-2α compared to RCC4. We also performed this experiment in 786-0 cells, which only express HIF-2α. Using shRNA or CRISPR against HIF-2α significantly reduced miR-2355-5p expression (Fig. 2E-H). Finally, we measured the expression of miR-2355-5p in 786-0 and RCC4 cells with functional VHL under hypoxic conditions. Hypoxia was confirmed by the expression of HIF-α proteins in wild-type VHL cells (Fig. 2I). Our results demonstrated a significant increase of miR-2355-5p in 786-0 VHL cells under hypoxia, while a similar but less noticeable increase was observed in RCC4 VHL cells, which could be reflected by the presence of HIF-1a (Fig. 2J).

Fig. 2
figure 2

miR-2355-5p overexpression is HIF-2α dependent. A-B Immunoblot analysis of VHL, HIF-2α, and β-actin and RT-qPCR quantification of miR-210-3p and miR-2355-5p (N = 3) in 786-0, 786-0 VHL, 786-0 VHL Cr.(Crispr)VHL A, and 786-0 VHL Cr.VHL B cells, respectively. C-D Immunoblot analysis of VHL, HIF-1α, HIF-2α, and β-actin and RT-qPCR quantification of miR-210-3p and miR-2355-5p (N = 3) in RCC4, RCC4 VHL, RCC4 shHIF-1α, and RCC4 shHIF-2α cells, respectively. E-F Immunoblot analysis of VHL, HIF-2α, and β-actin and RT-qPCR quantification of miR-2355-5p (N = 4) in 786-0, 786-0 VHL, and 786-0 shHIF-2α cells, respectively. G-H. Immunoblot analysis of VHL, and HIF-2α and RT-qPCR quantification of miR-210-3p and miR-2355-5p (N = 4) in 786-0, 786-0 VHL, and 786-0 Cr.HIF-2α cells, respectively. I-J Immunoblot analysis of VHL, HIF-1α (arrow), HIF-2α, and β-actin and RT-qPCR quantification of miR-2355-5p (N = 3) in 786-0, 786-0 VHL, RCC4, and RCC4 VHL cells cultured for 48 h under normal (21% O2) or hypoxic (1% O2) conditions, respectively. K Illustration of the genomic location of miR-2355-5p (GRCh38). L. Expression of KLF7 in ccRCC patient tissues extracted from the TCGA-KIRC (left) and CPTAC3 (right) databases. M Spearman correlation between miR-2355-5p and KLF7 expression in ccRCC patient tissues from the TCGA-KIRC (left) and CPTAC3 (right) cohorts. N Expression of KLF7 in ccRCC patients (CPTAC3) based on tumor VHL status. O RT-qPCR quantification of KLF7 (N = 4) in the cells mentioned in C. P RT-qPCR quantification of KLF7 (N = 3) in the cells mentioned in E. Q RT-qPCR quantification of KLF7 and GLUT1 (N = 3) in the cells mentioned in I. For immunoblot analysis, β-actin and total protein were used as loading controls. The miRNA and gene expression were normalized with RNU44 (B, D,F, H,J) or RPLPO (O, P,Q). The data are presented as the Mean ± SEM. Statistical analysis was performed using two-tailed unpaired Students t-tests (B, D,H, O), one-way ANOVA with Tukey’s multiple comparison test (F, P), two-way ANOVA with Tukey’s multiple comparison test (J) or Dunnett’s multiple comparison test compared to 786-0 VHL in 21% O2 (Q), Wilcoxon signed-rank test (L, N), and Spearman correlation test (M). (ns = not significant, *P < 0.05, **P < 0.01, ***P < 0.001)

Full size image

Intronic miRNAs are often coexpressed with their host gene. Hsa-miR-2355 is found in the second intron of the Kruppel-Like Transcription Factor 7 (KLF7) gene (Fig. 2K). Using TCGA data, we observed that KLF7 expression was also significantly increased in ccRCC tumors and was positively correlated with miR-2355-5p expression (Fig. 2L-M). Furthermore, KLF7 expression was higher in tumors harboring VHL mutations than in VHL-intact tumors (Fig. 2N). Moreover, our results indicated a significant decrease in KLF7 following the reintroduction of VHL in RCC4 and 786-0 cells (Fig. 2O-P). Additionally, only the loss of HIF-2α significantly decreased KLF7 expression (Fig. 2O-P). However, KLF7 expression was not significantly affected in 786-0 VHL cells under hypoxia (p = 0.104), while Glucose Transporter 1 (GLUT1) overexpression confirmed HIFα transcriptional activity under hypoxia (Fig. 2Q). Taken together, these results demonstrate a direct VHL/HIF-2α/miR-2355-5p axis in ccRCC, which could be attributed to a VHL/HIF-2a-dependent increase in KLF7.

miR-2355-5p pulldown

To investigate which cellular pathways involved in ccRCC tumorigenesis could be influenced by miR-2355-5p, we performed a miRNA pulldown assay using a designed synthetic biotinylated miR-2355-5p (Fig. 3A-B). We first tested whether the added biotin molecule interfered with the functionality of the miRNA. We cloned a perfect complementary sequence in the 3’UTR of a firefly luciferase expressing plasmid (Fig. 3C). The presence of miR-2355-5p-Biotin significantly reduced the concentration of firefly luciferase compared to the plasmid alone, while transfection with a miRNA mimic did not alter its expression (Fig. 3C). Therefore, the synthetic miR-2355-5p-Biotin is functionally active.

Fig. 3
figure 3

MiR 2355-5p pulldown. (A) Schematic representation of the miRNA pulldown protocol (created in BioRender. Page, P. (2024) https://BioRender.com/o69m105). (B) Sequence of the synthetic miR-2355-5p-Biotin duplex used for the miRNA pulldown. (C) Luciferase assay measuring the capability of the miR-2355-5p-Biotin to bind a perfect 3’UTR complementary sequence compared to that of a negative mimic control (N = 3). (D) PCA plot of sequenced miR-2355-5p pulldown RNA enriched samples (miR2355) compared with leftover RNA control samples (CTRL). (E) Volcano plot summarizing enriched potential targets (red). Transcripts with an FC > 1.5 and a padj < 0.05 were considered enriched. (F) Pie chart of the gene biotype of each enriched RNA transcript (Other = Pseudogenes (99), rRNA (19), miRNA (5), miscRNA (5), snoRNA (5), snRNA (3), TEC (3), and artifact (2)). (G) Functional analysis of enriched protein-coding genes. Representation of a portion of significantly enriched Gene Ontology Biological Process (GO: BP) terms. The data are presented as the Mean ± SEM. Statistical analysis was performed using one-way ANOVA with Dunnett’s multiple comparison test compared to untreated sample (0nM) (C) (ns = not significant, ****P < 0.0001)

Full size image

After sequencing both the captured RNA fractions and the control leftover RNA fractions, more than 1044 RNA transcripts were identified as being significantly enriched in the miR-2355-5p pulldown fraction (Fig. 3D-E, supplementary data). The presence of miR-2355 as the most enriched transcript supports the efficacy of the experiment (Fig. 3E). Of the 1044 transcripts, 862 encoded functional proteins, and 41 encoded long noncoding RNAs (lncRNAs) (Fig. 3F). Next, we subjected the list of identified protein transcripts to a Gene Ontology (GO) functional analysis. The analysis revealed enrichment of important biological processes involved in ccRCC cancer hallmarks such as Gene Regulation, Metabolic Processes, Lipid and Protein Post-Translational Modification, Immune System, Cell Growth, and Hypoxia/Angiogenesis (Fig. 3G, supplementary data).

Loss of miR 2355-5p reduces cell survival and proliferation

To investigate the impact of miR-2355-5p on cell growth, we generated a knockout (KO) miR-2355-5p cell line using CRISPR/Cas9. We designed 3 gRNA sequences targeting the genomic region encoding hsa-miR-2355, which were subsequently introduced into 786-0 cells (Fig. 4A). The second (gRNA.2) and third (gRNA.3) sequences were able to significantly reduce miR-2355-5p expression in the heterogeneous population compared to parental 786-0 cells and 786-0 cells expressing an empty GFP-LentiCRISPRv2 plasmid (786-0 Cr.CTRL) (Fig. S2A). Furthermore, 3 out of 4 clones generated from g.RNA.2 showed complete KO of miR-2355-5p (Fig. 4B, Fig. S2A). Fragmentation of localized DNA following T7 endonuclease treatment in Cr.2355 models but not in controls confirmed the presence of CRISPR/Cas9-stimulated mutations in the hsa-miR-2355 genomic region (Fig. S2B). Additionally, KLF7 expression was not altered in Cr.2355 cells (Fig. S2C). In parallel, using the same gRNA.2, miR-2355-5p KO was also achieved in A498 cells, another VHL-deficient ccRCC cell line (Fig. 4C).

Fig. 4
figure 4

Loss of miR 2355-5p reduces cell survival and proliferation. (A) Schematic representation of the genomic location targeted by three CRISPR/Cas9 gRNAs to repress miR-2355-5p. (B) RT-qPCR quantification of miR-2355-5p in 786-0, 786-0 Cr.Control (CTRL), 786-0 Cr.2355–2 A, 786-0 Cr.2355-2B, and 786-0 Cr.2355–2 C cells (N = 3). (C) RT-qPCR quantification of miR-2355-5p in A498, A498 Cr.CTRL, A498 Cr.2355–2 A, and A498 Cr.2355-2B cells (N = 3). E-G Image illustration (D-E) and relative number of colonies (F-G) measured by clonogenic assays in 786-0 Cr.2355 and A498 Cr.2355 clones compared to appropriate controls (N = 4). H-I Cell proliferation measured by daily cell counts in 786-0 Cr.2355 (H) and A498 Cr.2355 (I) clones compared to appropriate control cells (N = 4). J-K Flow cytometry analysis and quantification of 786 Cr.2355 models and control cells following PI staining. MiRNA expression was normalized with RNU44. The data are presented as the Mean ± SEM. Statistical analysis was performed using two-tailed Student’s t-tests (B, C,F, G) or two-way ANOVA with Dunnett’s multiple comparison test compared to 786 Cr.CTRL (H, J) or A498 Cr.CTRL (I). (ns = not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001)

Full size image

Using our developed miR-2355-5p KO models, we found that the loss of the miRNA caused a significant decrease in the number of colonies in both 786-0 and A498 cells compared to the appropriate controls (Fig. 4D-G). Additionally, the colony size of 786.0 Cr.2355 appeared to be slightly smaller compared to the controls. Indeed, we measured a significant reduction in cell proliferation in the absence of miR-2355-5p. On day 5, we observed a 29.4% and 40.4% cell growth reduction in the 786 Cr.2355 clones and a 39.4% and 48.4% reduction in the growth of the A498 Cr.2355 clones, respectively (Fig. 4H-I). Furthermore, cell cycle analysis revealed that loss of miR-2355-5p blocked cells in the DNA synthesis phase (S phase). In fact, a significant increase in the number of cells present in the S phase associated with a decrease in the G0/G1 phase was observed in the Cr.2355 models compared to both controls (Fig. 4J-K). Finally, we generated stable RCC cell lines overexpressing miR-2355-5p (786-0 + + 2355 and A498 + + 2355). Although miR-2355-5p was significantly increased, our results showed that the number of colonies and cell proliferation were not altered compared to those of controls in vitro, which could be attributed to the already high endogenous level of miR-2355-5p (Fig. S2D-G). Together, these results suggest that targeting miR-2355-5p reduces ccRCC cell growth.

Loss of miR 2355-5p alters the ability of cells to stimulate angiogenesis

Increased mobility and tumor vascularization are both important steps for tumor growth and metastasis. Using a Transwell assay, we observed a significant reduction of 30.5% and 35.3% in the number of migrated 786-0 Cr.2355 cells compared to the number of migrated control cells (Fig. S3A-B). Next, we collected conditioned media (CM) from our Cr.2355 models and measured their angiogenic capability through a tube formation assay. Our results showed the formation of tube-like structures when HUVECs were incubated with CM from control cells, while CM from cells deprived of miR-2355-5p was not able to induce these capillary structures (Fig. 5A-D). In fact, a significant 64.6% and 84.5% decrease in the number of tubes and a 93.5% and 95.4% reduction in the total tube size were measured for 786-0 Cr.2355 clones compared to 786-0 Cr.CTRL cells, respectively (Fig. 5B). Similarly, a 44.9% and 53.4% reduction in number of tubes and a 56.4% and 80.2% decrease in tube size were measured in A498 Cr.2355 clones compared to A498 Cr.CTRL cells, respectively (Fig. 5D).

Fig. 5
figure 5

Loss of miR 2355-5p alters the ability of cells to stimulate angiogenesis. A-B Tube formation assays performed with CM from 786-0, 786-0 Cr.CTRL, 786-0 Cr.2355-2B, and 786-0 Cr.2355–2 C cells (N = 3). Image illustration (A) and quantification (B) of the number of tube-like structures (left) and total tube size (right) under each experimental condition. C-D Tube formation assays performed with CM from A498, A498 Cr.CTRL, A498 Cr.2355–2 A, and A498 Cr.2355-2B cells (N = 3). Image illustration (C) and quantification (D) under each experimental condition. Positive (10% FBS) and negative (0.5% FBS) controls were used. (Scale bars = 200µM) E-F Proteome Profiler Human Angiogenesis Array Kit used with CM from 786-0 Cr.CTRL and 786-0 Cr.2355–2 C cells. Image illustration from low-exposure (E) and high-exposure (F) and quantification of altered observable proteins. G-H Angiogenesis Array used with protein cell lysate from 786-0 Cr.CTRL and 786-0 Cr.2355–2 C cells. Image illustration from low-exposure (G) and high-exposure (H) and quantification of altered observable proteins. Each protein is represented by two dots and quantified as Mean. I-J RT-qPCR quantification of selected angiogenic factors in 786-0 Cr.2355 models (I) and A498 Cr.2355 models (J) compared to appropriate control cells (N = 3). Gene expression was normalized with RPLPO. The data are presented as the Mean ± SEM. Statistical analysis was performed using two-tailed unpaired Student’s t-tests (*P < 0.05, **P < 0.01)

Full size image

Angiogenesis is a complex process involving several pro- and antiangiogenic factors. We used a Proteome Profiler Human Angiogenesis Array to profile 55 angiogenic proteins in CM from 786 to 0 Cr.CTRL and 786-0 Cr.2355–2 C cells. Differences in the concentration of multiple excreted proteins were observed (Fig. 5E-F). Some proteins were decreased in CM from 786 to 0 Cr.2355–2 C (ANG, AREG, CSF2, CXCL8, EDN1, F3, IL1B, PTX3, THBS1, and VEGF), while others were increased (COL18A1, DPP4, MMP9, PDGFA, PLAU, and TIMP4) compared to those in CM from 786 to 0 Cr.CTRL cells. In parallel, we analyzed cell lysates from both cell models. Similarly, the expression of some proteins (AREG and IGFBP3) was decreased in 786-0 Cr.2355–2 C cells (AREG and IGFBP3), while others (FGF1, FGF2, IGFBP1, PLAU, SERPINE1, TIMP1, THBS1, and VEGF) were increased when compared to 786-0 Cr.CTRL cells (Fig. 5G-H). Some proteins showed similar patterns between CM and cell lysate (PLAU and AREG), while some showed an inverse correlation (THBS1 and VEGF). We next investigated whether the decrease in the concentration of these factors could be observed at the transcript level. Of the 7 selected genes, only AREG showed a significant decrease in mRNA expression in most Cr.2355 models (Fig. 5I-J). Accordingly, most downregulated secreted proteins were overexpressed in ccRCC tumors (Fig. S3C). Together, we have shown that multiple angiogenic pathways are affected by the loss of miR-2355-5p and we postulate that the increase in miR-2355-5p positively contributes to ccRCC tumor vascularization.

miR 2355-5p is important for ccRCC tumor growth

To better understand the molecular changes induced by miR-2355-5p in ccRCC, we investigated its direct protein targets through our pulldown assay (Fig. 3). We briefly explored the interaction between miR-2355-5p and the 41 identified lncRNAs. Among them, 11 lncRNAs were downregulated in ccRCC tumors, while 8 were overexpressed (Fig. 6A, Fig. S4A). Using the starBase V2.0 database (ENCORI) we found that 5 lncRNAs were predicted to bind to miR-2355-5p, of which 2 were downregulated (KCNQ1OT1 and MIR600HG), 2 were unmodified (NEAT1 and LINC00205) and 1 was overexpressed (PVT1) in ccRCC tumors (Fig. S4A). Subsequently, we further studied the 862 protein-coding genes and found that 70 were downregulated in both TCGA-KIRC and CPTAC3 cohorts (Fig. 6B). Using 3 miRNA-mRNA prediction algorithm software, we discovered that out of the 70 downregulated potential targets, 33 were predicted by at least two programs (Fig. 6B). From this list, 10 were selected based on published literature for further analysis. Interestingly, the mRNA expression of 5 out of the 10 (ACO1, BTG2, CMTM4, SLIT2, and WDFY2) was significantly higher in most Cr.2355 models compared to appropriate controls (Fig. 6C-D). Furthermore, all five genes were significantly downregulated in our cohort of ccRCC patients, which mirrored the findings in the TCGA cohorts (Fig. 6E, Fig. S4B).

Fig. 6
figure 6

Clinical implications of miR-2355-5p. (A) Volcano plot summarizing the RNA expression profiles of ccRCC patients obtained from the TCGA-KIRC (top) and CPTAC3 (bottom) databases (TCGA-KIRC: N = 72, CPTAC3: N = 149). RNAseq analysis was performed similarly to miRNAseq previously described. (B) Left: Venn diagram comparing enriched protein transcripts and significantly downregulated genes in ccRCC tumors from TCGA-KIRC and CPTAC3 cohorts. Right: Venn diagram comparing selected protein transcripts from the left diagram (marked in green) and predicted miR-2355-5p targets from prediction software (miRDB, TargetScan, miRWalk). C-D RT-qPCR quantification of selected potential target genes in 786-0 (B) and A498 (C) Cr.2355 models compared to appropriate controls (N = 4–6). E. RT-qPCR quantification of selected genes in ccRCC primary tissue compared to normal tissue (N = 13). F Tumor growth of the 786-0 Cr.2355 xenograft mouse model (N = 4–8). G Image illustration of extracted xenograft tumors H-I RT-qPCR quantification of miR-2355-5p (H) and selected targets (I) in 786-0 Cr.2355 mouse tumors (N = 4–8). J Tumor growth of the 786-0 + + 2355 xenograft mouse model (N = 6–7). K-L. RT-qPCR quantification of miR-2355-5p (K) and selected targets (L) in 786-0 + + 2355 mouse tumors (N = 6–7). MiRNA and gene expression was normalized with RNU44 (H, J), RPLPO (C, D,K, L), or RPLPO + HPRT1 (E) expression. The data are presented as the Mean ± SEM. Statistical analysis was performed using two-tailed unpaired (C, D, H, J, K, L) or paired (E) Student’s t-tests or two-way ANOVA with Dunnett’s multiple comparison test compared to 786-0 Cr.CTRL (F) and 786-0 + + CTR: (I). (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001)

Full size image

Finally, to further evaluate the importance of miR-2355-5p in ccRCC development, we subcutaneously injected 786-0 Cr.2355 clones and 786-0 Cr.CTRL into the dorsal flank of NSG immunodeficient mice. The growth of 786-0 Cr.2355 xenograft tumors was drastically reduced, supporting, and strengthening, our in vitro results (Fig. 6F-G). Analysis of the extracted tumors showed that miR-2355-5p was still inhibited (Fig. 6H). We also injected our overexpression model (786-0 + + 2355) and related control (786-0 + + CTRL) into nude mice. We noticed that tumor growth was slower in nude mice compared to NSG mice, even though the injection protocol was similar. However, although we did not observe a significant difference in vitro, overexpression of miR-2355-5p further increased tumor growth in mice (Fig. 6I, Fig. S5C). The overexpression of miR-2355-5p was confirmed in the extracted tumors (Fig. 6J). Interestingly, the expression of the 5 potential miR-2355-5p targets was also significantly increased in Cr.2355 xenograft tumors (Fig. 6K). Furthermore, WDFY2 was significantly reduced in + + 2355 tumors (Fig. 6L). Taken together, these results obtained in this figure demonstrate the implication of miR-2355-5p in ccRCC tumorigenesis in vivo in mice, and identify 5 potential targets that are strongly correlated with miR-2355-5p in mice, in the TCGA cohort and in our cohort of ccRCC patients.

Discussion

VHL loss of function is one of the major first steps in ccRCC carcinogenesis. Understanding the molecular changes surrounding VHL inactivation could help steer research toward better biomarker identification or targeted therapies. As a matter of fact, we previously demonstrated the possibility of directly targeting VHL deficiency as a means of treating ccRCC cells [27]. Here, we focused on further elucidating the effect of the VHL mutational status on miRNA expression in ccRCC. Our miRNA sequencing and bioinformatic analysis revealed 45 VHL-regulated miRNAs in both clinical samples and ccRCC cell lines. Interestingly, most of the selected miRNAs were validated in at least two or three ccRCC cell lines tested. Additionally, we observed that 3 miRNAs (miR-34a-5p, miR-138-5p, and miR-1301-3p) showed different expression patterns following the reintroduction of VHL in different cell lines. Although VHL is the most mutated gene, ccRCC tumors are highly heterogeneous, and other tumor suppressor genes found on chromosome 3p, such as PBRM1, SETD2, BAP1, and PTEN, are also frequently lost or mutated [28]. The mutational status of these genes could explain why these miRNAs respond differently to VHL in different ccRCC cell lines because they are not genetically identical. For instance, 786-0 cells are mutated for PTEN and do not express HIF-1α, while PBRM1 is mutated in RCC4 and both HIF-a isoforms are expressed [7, 29, 30]. Nonetheless, our results demonstrate that VHL can significantly impact the expression of multiple miRNAs in ccRCC.

To the best of our knowledge, we are the first to explicitly demonstrate the overexpression of miR-2355-5p in ccRCC. Our results indicated that miR-2355-5p overexpression occurs through the canonical VHL/HIF-2α axis. It is known that HIF-2α plays a more oncogenic role in ccRCC, while HIF-1α has preferential tumor suppressive activity [31]. Therefore, the fact that miR-2355-5p is promoted only by HIF-2α encourages its potential oncogenic function. Indeed, we demonstrated its importance for cell proliferation, migration, angiogenesis and tumor growth by directly targeting its expression using CRISPR/Cas9 in two VHL-deficient ccRCC cell lines. Furthermore, our analyses indicated that miR-2355-5p overexpression could be indirectly attributed to an increase in KLF7. This affirmation suggests the presence of an HRE binding site near its promoter region, although we cannot exclude the possibility that miR-2355-5p could possess its own promoter and HRE binding site [32]. Further work is needed to better understand how HIF-2α mechanistically controls the expression of both KLF7 and miR 2355-5p. Previous work has shown that KLF members can affect the progression of multiple cancers including ccRCC [33]. For instance, the targeting of KLF6 by miR-543 can promote the proliferation and invasion of ccRCC. Interestingly, miR-2355-5p has also been shown to directly target KLF6 in Human Umbilical-cord-blood-derived erythroid progenitor-2 (HUDEP-2) cells [34], revealing a potential route by which miR-2355-5p could promote ccRCC [35]. Additionally, KLF11 and KLF13 were both enriched during miR-2355-5p pulldown, and other KLF members, such as KLF-2, -5, and − 8, are closely related to HIFα [36,37,38]. Altogether, these observations further suggest an interesting dynamic between miR-2355-5p and KLF members in ccRCC carcinogenesis.

Previous studies have identified miR-2355-5p as a tumor suppressor in different cancer types, such as Diffuse Large B-Cell Lymphoma, Chondrosarcoma, Triple Negative Breast Cancer, Cervical Cancer, Non-Small Cell Lung Cancer, and Gastric Cancer, although it plays a more oncogenic role in Esophageal Squamous Cell Carcinoma (ESCC) [19,20,21, 39,40,41,42]. Our study further supports this carcinogenic role, as miR-2355-5p overexpression in VHL-mutated ccRCC promoted tumor growth. Moreover, we showed that the loss of miR-2355-5p significantly reduced cell migration and disrupted the ability of ccRCC cells to stimulate angiogenesis. Other studies previously labeled miR 2355-5p as an antiangiogenic miRNA by directly targeting VEGFR2 in Endothelial Colony Forming Cells (ECFC) and HUVECs [40, 43]. However, VEGFR2 mRNA is often increased in ccRCC tumors compared to normal tissue [44]. These contradictory observations could be explained by the cell type since HUVECs and ECFCs are noncancerous endothelial cells. Here, VEGFR2 was not identified as a potential target in our analyses. Additionally, only a small decrease of VEGF in CM and a slight increase in cell lysate were detected in Cr.2355 cells compared to control cells, suggesting that the angiogenic influence of miR-2355-5p does not occur through the VEGF axis. However, inactivation of miR-2355-5p decreased the secretion of multiple other angiogenic factors such as ANG, AREG, CSF2, CXCL8, EDN1, F3, IL1B, PTX3, and THBS1. The concentration of AREG also decreased in the protein cell lysate and significantly decreased at the transcript level in most Cr.2355 models. Previous studies have shown that AREG expression is often increased in tumors [45]. A greater presence of AREG in the tumor microenvironment reduces antitumor immunity, which in turn increases tumor growth and invasiveness [46,47,48]. Moreover, AREG, as well as other identified angiogenic factors (ANG, CXCL8, IL1B, VEGF, and EDN1) were significantly overexpressed in ccRCC tumors, further suggesting multiple routes by which miR-2355-5p could promote angiogenesis in ccRCC tumors. Even so, miR-2355-5p expression does not further increase with tumor stage. Therefore, although the angiogenic influence of miR-2355-5p could still be significant, more research is needed to characterize the underlying molecular mechanisms involved.

Interestingly, at least five potential targets of miR-2355-5p, ACO1, BTG2, CMTM4, SLIT2, and WDFY2, were identified in this study, all of which are tumor suppressors in different cancers including ccRCC. Zhu et al. showed that ACO1 was significantly downregulated in ccRCC and correlated with sex, tumor stage, and overall survival [49]. ACO1 can also regulate HIF-2α expression in an oxygen- and iron-dependent manner [50]. A HIF-2α/miR-2355-5p/ACO1 axis could thus improve HIF-2α stability and further increase the influence of HIF-2α in ccRCC. CMTM4 is involved in the immune system as a regulator of PD-L1 [51]. Previous work demonstrated that CMTM4 controls ccRCC migration and regulates tumor growth both in vitro and in vivo by causing G2/M cell cycle arrest [52]. Xue et al. later showed similar G2/M blockage by CMTM4 in Colorectal Cancer [53]. Moreover, SLIT2 was shown to be downregulated in RCC and decreased under hypoxia [54, 55]. Shen et al. also showed that SLIT2 plays an important tumor suppressive role in Cervical Cancer by mitigating migration/invasion and causing G0/G1 cell cycle arrest [56]. WDFY2 is a novel protein involved in endocytosis and has been shown to be downregulated in ccRCC patients [57]. Our group and others have shown that targeting endolysosomes and endocytic processes could be a promising avenue for treating VHL-deficient ccRCC [22, 58]. Finally, BTG2 has been shown to be downregulated in ccRCC [59]. Most recent work showed that BTG2 can reduce migration/invasion and inhibit the cell cycle (G0/G1 phase) in ccRCC cells [60]. Ultimately, our research revealed a tight correlation between miR-2355-5p and these tumor suppressor genes not only in cell models but also in mouse xenograft tumors and primary ccRCC tumors. Furthermore, the decrease in tumor growth through the loss of miR-2355-5p observed in this study could be related to the cell cycle arrest induced by the upregulation of one or more of these potential targets. Additionally, most potential targets were also previously linked with promoting angiogenesis in cancer [61,62,63,64].

Collectively, our research helps support previous work investigating the influence of VHL on miRNA expression, and we are the first to show direct VHL/HIF-2a-dependent overexpression of miR-2355-5p in ccRCC tumors. The results presented here demonstrate the oncogenic impact of miR-2355-5p on ccRCC through stimulating both angiogenesis and tumor growth. In the last decade, interest in miRNA-based therapeutics has increased, and clinical trials are currently ongoing for the treatment of different diseases including cancer [65]. Additionally, clinical trials evaluating panels of circulating miRNAs as tools for cancer detection are being conducted [66]. The increase in circulating miR-2355-5p in plasma samples suggests the inclusion of miR-2355-5p in future panels dedicated to ccRCC. For therapeutic applications, we have shown that directly targeting miR-2355-5p can significantly reduce ccRCC tumor growth and potentially tumor angiogenesis. Future research could uncover new routes to improve the treatment of ccRCC in patients. Altogether, we provide evidence that an important novel VHL/HIF-2a/miR-2355-5p axis is involved in VHL-deficient ccRCC tumorigenesis.

Data availability

Sequencing data that support the findings of miRNA pulldown have been deposited to GEO database with primary accession code GSE246950.

Abbreviations

ACO1:

Aconitase 1

ANG:

Angiogenin

AREG:

Amphiregulin

BTG2:

BTG Anti-Proliferation Factor 2

ccRCC:

clear cell Renal Cell Carcinoma

CM:

Conditioned Media

CMTM4:

CKLF Like MARVEL Transmembrane Domain Containing 4

CRISPR:

Clustered Regularly Interspaced Short Palindromic Repeats

DE:

Differentially Expressed

HIF:

Hypoxia Induced Factors

HRE:

Hypoxia Response Element

HUVEC:

Human Umbilical Vein Endothelial Cell

KLF:

Kruppel Like Factors

lncRNA:

Long Non-Coding RNA

miRNA/miR:

microRNA

mRNA:

messenger RNA

NSG:

Non-Obese Diabetic Severe Combined Immunodeficiency Gamma

PFKFB2:

6-Phosphofructo-2-Kinase/Fructose-2,6-Biphosphatase2

RISC:

RNA Induced Silencing Complex

shRNA:

short Hairpin RNA

TCGA:

The Cancer Genome Atlas

UTR:

Untranslated Region

VEGFA:

Vascular Endothelial Growth Factor A

VHL:

Von Hippel-Lindau

WDFY2:

WD Repeat And FYVE Domain Containing 2

References

  1. Sung H, et al. Global Cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. Cancer J Clin. 2021;71(3):209–49.

    Article  Google Scholar 

  2. Rini BI, Campbell SC, Escudier B. Ren Cell Carcinoma Lancet. 2009;373(9669):1119–32.

  3. Fisher R, Gore M, Larkin J. Current and future systemic treatments for renal cell carcinoma. Sem Cancer Biol. 2013;23(1):38–45.

    Article  Google Scholar 

  4. Gupta K, et al. Epidemiologic and socioeconomic burden of metastatic renal cell carcinoma (mRCC): A literature review. Cancer Treat Rev. 2008;34(3):193–205.

    Article  PubMed  Google Scholar 

  5. Hsieh JJ, et al. Renal cell carcinoma. Nat Reviews Disease Primers. 2017;3(1):17009.

    Article  PubMed  Google Scholar 

  6. Creighton CJ, et al. Comprehensive molecular characterization of clear cell renal cell carcinoma. Nature. 2013;499(7456):43–9.

    Article  CAS  Google Scholar 

  7. Maxwell PH, et al. The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature. 1999;399(6733):271–5.

    Article  CAS  PubMed  Google Scholar 

  8. Ivan, M., et al., HIFα Targeted for VHL-Mediated Destruction by Proline Hydroxylation: Implications for O < sub > 2 Sensing. Science, 2001. 292(5516): pp. 464–468.

  9. Jaakkola P, et al. Targeting of HIF-α to the von Hippel-Lindau ubiquitylation complex by O < sub > 2-Regulated Prolyl hydroxylation. Science. 2001;292(5516):468–72.

    Article  CAS  PubMed  Google Scholar 

  10. Semenza GL. Oxygen sensing, homeostasis, and disease. N Engl J Med. 2011;365(6):537–47.

    Article  CAS  PubMed  Google Scholar 

  11. Crosby ME, et al. Emerging roles of MicroRNAs in the molecular responses to hypoxia. Curr Pharm Des. 2009;15(33):3861–6.

    Article  CAS  PubMed  Google Scholar 

  12. Bartel DP. Metazoan MicroRNAs. Cell. 2018;173(1):20–51.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Bartel DP. MicroRNAs: target recognition and regulatory functions. Cell. 2009;136(2):215–33.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Jonas S, Izaurralde E. Towards a molecular Understanding of microRNA-mediated gene Silencing. Nat Rev Genet. 2015;16(7):421–33.

    Article  CAS  PubMed  Google Scholar 

  15. Peng Y, Croce CM. The role of MicroRNAs in human cancer. Signal Transduct Target Therapy. 2016;1(1):15004.

    Article  Google Scholar 

  16. Zhang J, et al. Global and targeted MiRNA expression profiling in clear cell renal cell carcinoma tissues potentially links miR-155-5p and miR-210-3p to both tumorigenesis and recurrence. Am J Pathol. 2018;188(11):2487–96.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Qin Z, et al. miR-224-5p contained in urinary extracellular vesicles regulates PD-L1 expression by inhibiting Cyclin D1 in renal cell carcinoma cells. Cancers. 2021;13(4):618.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Qu F, et al. MicroRNA-497-5p down-regulation increases PD-L1 expression in clear cell renal cell carcinoma. J Drug Target. 2019;27(1):67–74.

    Article  CAS  PubMed  Google Scholar 

  19. Zhou M, et al. Chitosan-Gelatin-EGCG Nanoparticle-Meditated LncRNA TMEM44-AS1 Silencing to activate the P53 signaling pathway for the synergistic reversal of 5-FU resistance in gastric Cancer. Adv Sci (Weinh). 2022;9(22):e2105077.

    Article  PubMed  Google Scholar 

  20. Li J, Wang H. K27ac-activated EGFR-AS1 promotes cell growth in cervical cancer through ACTN4-mediated WNT pathway. Biol Direct. 2022;3(1):3.

    Article  Google Scholar 

  21. Zhang Q, et al. LncRNA WDFY3-AS2 suppresses proliferation and invasion in oesophageal squamous cell carcinoma by regulating miR-2355-5p/SOCS2 axis. J Cell Mol Med. 2020;24(14):8206–20.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Bouhamdani N, et al. STF-62247 accumulates in lysosomes and blocks late stages of autophagy to selectively target von Hippel-Lindau-inactivated cells. Am J Physiol Cell Physiol. 2019;316(5):C605–20.

    Article  CAS  PubMed  Google Scholar 

  23. Carpentier G, et al. Angiogenesis analyzer for ImageJ - A comparative morphometric analysis of endothelial tube formation assay and fibrin bead assay. Sci Rep. 2020;10(1):11568.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Wani S, Cloonan N. Profiling direct mRNA-microRNA interactions using synthetic biotinylated microRNA-duplexes. bioRxiv. 2014: p. 005439.

  25. Neal CS, et al. The VHL-dependent regulation of MicroRNAs in renal cancer. BMC Med. 2010;8(1):64.

    Article  PubMed  PubMed Central  Google Scholar 

  26. McCormick RI, et al. miR-210 is a target of hypoxia-inducible factors 1 and 2 in renal cancer, regulates ISCU and correlates with good prognosis. Br J Cancer. 2013;108(5):1133–42.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Turcotte S, et al. A molecule targeting VHL-deficient renal cell carcinoma that induces autophagy. Cancer Cell. 2008;14(1):90–102.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  28. Turajlic S, et al. Deterministic evolutionary trajectories influence primary tumor growth: tracerx renal. Cell. 2018;173(3):595–e61011.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Gao W, et al. Inactivation of the PBRM1 tumor suppressor gene amplifies the HIF-response in VHL-/- clear cell renal carcinoma. Proc Natl Acad Sci U S A. 2017;114(5):1027–32.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Schneider E, et al. Migration of renal tumor cells depends on dephosphorylation of Shc by PTEN. Int J Oncol. 2011;38(3):823–31.

    CAS  PubMed  Google Scholar 

  31. Keith B, Johnson RS, Simon MC. HIF1α and HIF2α: sibling rivalry in hypoxic tumour growth and progression. Nat Rev Cancer. 2012;12(1):9–22.

    Article  CAS  Google Scholar 

  32. Monteys AM, et al. Structure and activity of putative intronic MiRNA promoters. RNA. 2010;16(3):495–505.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Li ZY, et al. The role of KLF transcription factor in the regulation of cancer progression. Biomed Pharmacother. 2023;162:114661.

    Article  CAS  PubMed  Google Scholar 

  34. Cheng Y, et al. MicroRNA-2355-5p regulates γ-globin expression in human erythroid cells by inhibiting KLF6. Br J Haematol. 2021;193(2):401–5.

    Article  CAS  PubMed  Google Scholar 

  35. Yang F, et al. MicroRNA-543 promotes the proliferation and invasion of clear cell renal cell carcinoma cells by targeting Krüppel-like factor 6. Volume 97. Biomedicine & Pharmacotherapy. 2018. pp. 616–23.

  36. Gong T, et al. Knockdown of KLF5 suppresses hypoxia-induced resistance to cisplatin in NSCLC cells by regulating HIF-1α-dependent Glycolysis through inactivation of the PI3K/Akt/mTOR pathway. J Translational Med. 2018;16(1):164.

    Article  CAS  Google Scholar 

  37. Liu N, et al. Krüppel-like factor 8 involved in hypoxia promotes the invasion and metastasis of gastric cancer via epithelial to mesenchymal transition. Oncol Rep. 2014;32(6):2397–404.

    Article  CAS  PubMed  Google Scholar 

  38. Kawanami D, et al. Kruppel-like factor 2 inhibits Hypoxia-inducible factor 1α expression and function in the Endothelium *. J Biol Chem. 2009;284(31):20522–30.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Xu H, et al. PAX5-activated LncRNA ARRDC1-AS1 accelerates the autophagy and progression of DLBCL through sponging miR-2355-5p to regulate ATG5. Life Sci. 2021;286:119932.

    Article  CAS  PubMed  Google Scholar 

  40. Cheng C, et al. Exosomal LncRNA RAMP2-AS1 derived from chondrosarcoma cells promotes angiogenesis through miR-2355-5p/VEGFR2 Axis. Onco Targets Ther. 2020;13:3291–301.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Yu L, et al. LncRNA SNHG11 aggravates cell proliferation and migration in triple-negative breast cancer via sponging miR-2355-5p and targeting CBX5. Exp Ther Med. 2021;22(2):892.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Lv C, et al. CircRNA SOD2 motivates non-small cell lungs cancer advancement with EMT via acting as microRNA-2355-5p’s competing endogenous RNA to mediate calmodulin regulated spectrin associated proteins-2. Bioengineered. 2022;13(3):5756–68.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Su SH, et al. Dysregulation of vascular endothelial growth factor Receptor-2 by multiple MiRNAs in endothelial Colony-Forming cells of coronary artery disease. J Vasc Res. 2017;54(1):22–32.

    Article  CAS  PubMed  Google Scholar 

  44. Ljungberg BJ, et al. Different vascular endothelial growth factor (VEGF), VEGF-receptor 1 and– 2 mRNA expression profiles between clear cell and papillary renal cell carcinoma. BJU Int. 2006;98(3):661–7.

    Article  CAS  PubMed  Google Scholar 

  45. Lofgren KA, et al. Pan-cancer distribution of cleaved cell-surface Amphiregulin, the target of the GMF-1A3 antibody drug conjugate. Antib Ther. 2022;5(3):226–31.

    CAS  PubMed  PubMed Central  Google Scholar 

  46. Xu Q, et al. Targeting Amphiregulin (AREG) derived from senescent stromal cells diminishes cancer resistance and averts programmed cell death 1 ligand (PD-L1)-mediated immunosuppression. Aging Cell. 2019;18(6):e13027.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Sun R, et al. Amphiregulin couples IL1RL1(+) regulatory T cells and cancer-associated fibroblasts to impede antitumor immunity. Sci Adv. 2023;9(34):eadd7399.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  48. Jeong BY, et al. Lysophosphatidic acid-induced Amphiregulin secretion by cancer-associated fibroblasts augments cancer cell invasion. Cancer Lett. 2022;551:215946.

    Article  CAS  PubMed  Google Scholar 

  49. Zhu T, et al. ACO1 and IREB2 downregulation confer poor prognosis and correlate with autophagy-related ferroptosis and immune infiltration in KIRC. Front Oncol. 2022;12:929838.

  50. Sanchez M, et al. Iron-regulatory proteins limit hypoxia-inducible factor-2alpha expression in iron deficiency. Nat Struct Mol Biol. 2007;14(5):420–6.

    Article  CAS  PubMed  Google Scholar 

  51. Mezzadra R, et al. Identification of CMTM6 and CMTM4 as PD-L1 protein regulators. Nature. 2017;549(7670):106–10.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  52. Li T, et al. CMTM4 is frequently downregulated and functions as a tumour suppressor in clear cell renal cell carcinoma. J Exp Clin Cancer Res. 2015;34:122.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  53. Xue H, et al. CMTM4 inhibits cell proliferation and migration via AKT, ERK1/2, and STAT3 pathway in colorectal cancer. Acta Biochim Biophys Sin (Shanghai). 2019;51(9):915–24.

    Article  CAS  PubMed  Google Scholar 

  54. Ma WJ, et al. Reduced expression of Slit2 in renal cell carcinoma. Med Oncol. 2014;31(1):768.

    Article  PubMed  Google Scholar 

  55. Zhou X, et al. Slit2 ameliorates renal inflammation and fibrosis after hypoxia-and lipopolysaccharide-induced epithelial cells injury in vitro. Exp Cell Res. 2017;352(1):123–9.

    Article  CAS  PubMed  Google Scholar 

  56. Shen X, et al. Raddeanin A inhibits proliferation, invasion, migration and promotes apoptosis of cervical cancer cells via regulating miR-224-3p/Slit2/Robo1 signaling pathway. Aging. 2021;13(5):7166–79.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  57. Ding G, et al. A pan-cancer analysis of the role of WDFY2 in human tumors. Biotechnol Genet Eng Rev. 2024;40(3):1456–71.

  58. Wang Y, et al. Regulation of endocytosis via the oxygen-sensing pathway. Nat Med. 2009;15(3):319–24.

    Article  CAS  PubMed  Google Scholar 

  59. Struckmann K, et al. Impaired expression of the cell cycle regulator BTG2 is common in clear cell renal cell carcinoma. Cancer Res. 2004;64(5):1632–8.

    Article  CAS  PubMed  Google Scholar 

  60. Sima J, et al. Overexpression of BTG2 suppresses growth, migration, and invasion of human renal carcinoma cells in vitro. Neoplasma. 2016;63(3):385–93.

    Article  CAS  PubMed  Google Scholar 

  61. Shang D, et al. Pancreatic cancer cell-derived Exosomal microRNA-27a promotes angiogenesis of human microvascular endothelial cells in pancreatic cancer via BTG2. J Cell Mol Med. 2020;24(1):588–604.

    Article  CAS  PubMed  Google Scholar 

  62. Li B, et al. MicroRNA-934 facilitates cell proliferation, migration, invasion and angiogenesis in colorectal cancer by targeting B-cell translocation gene 2. Bioengineered. 2021;12(2):9507–19.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  63. Chrifi I, et al. CMTM4 regulates angiogenesis by promoting cell surface recycling of VE-cadherin to endothelial adherens junctions. Angiogenesis. 2019;22(1):75–93.

    Article  CAS  PubMed  Google Scholar 

  64. Wang LJ, et al. Targeting Slit-Roundabout signaling inhibits tumor angiogenesis in chemical-induced squamous cell carcinogenesis. Cancer Sci. 2008;99(3):510–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  65. Rupaimoole R, Slack FJ. MicroRNA therapeutics: towards a new era for the management of cancer and other diseases. Nat Rev Drug Discov. 2017;16(3):203–22.

    Article  CAS  PubMed  Google Scholar 

  66. Ho PTB, Clark IM, Le LTT. MicroRNA-Based Diagnosis Therapy Int J Mol Sci, 2022. 23(13).

Download references

Acknowledgements

The authors would like to thank the support of the Atlantic Cancer Research Institute for access to installations and for providing cell culture reagents. We also thank Dr. Eric Allain for his insight into the miRNA pulldown sequencing analysis. This work was supported by the Kidney Foundation of Canada and the New Brunswick Health Research Foundation (180017). S.T. is supported by a Canadian Cancer Society Research Chair (#706199). P.M.P. was supported by the Cancer Research Training Program from the Beatrice Hunter Cancer Research Institute and a Doctoral Studentship Award from the New Brunswick Health Research Foundation.

Funding

This work was supported by the Kidney Foundation of Canada and the New Brunswick Health Research Foundation (180017). S.T. is supported by a Canadian Cancer Society Research Chair (#706199). P.M.P. was supported by the Cancer Research Training Program from the Beatrice Hunter Cancer Research Institute and a Doctoral Studentship Award from the New Brunswick Health Research Foundation.

Author information

Authors and Affiliations

  1. Department of Chemistry and Biochemistry, Université de Moncton, 18 Antonine-Maillet, Moncton, NB, E1A 3E9, Canada

    Patric M. Page, Sonia A. Dastous & Sandra Turcotte

  2. Atlantic Cancer Research Institute, Moncton, NB, Canada

    Patric M. Page, Sonia A. Dastous & Sandra Turcotte

  3. Department of Urology, Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, Canada

    Patrick O. Richard

  4. Institut de recherche sur le cancer de l’Université de Sherbrooke, Sherbrooke, Canada

    Patrick O. Richard & Michel Pavic

  5. Department of Hemato-Oncology, Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, Canada

    Michel Pavic

  6. Department of Human Genetics, McGill University, Montréal, Québec, Canada

    Tamiko Nishimura & Yasser Riazalhosseini

  7. McGill Genome Centre, McGill University, Montréal, Québec, Canada

    Tamiko Nishimura & Yasser Riazalhosseini

  8. Laboratoire de Génétique Moléculaire, Vitalité Health Network, Moncton, Canada

    Nicolas Crapoulet

  9. Centre de formation médicale du Nouveau-Brunswick, Moncton, NB, Canada

    Mykella Martin

Authors
  1. Patric M. Page
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Sonia A. Dastous
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Patrick O. Richard
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Michel Pavic
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Tamiko Nishimura
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Yasser Riazalhosseini
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Nicolas Crapoulet
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Mykella Martin
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Sandra Turcotte
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

S.T. and P.M.P. conceived and designed the research and wrote the manuscript; P.M.P. performed the experiments, analyzed the data, interpreted the results, and prepared the figures. S.A.D. performed the experiments shown in Fig. 1 (A, F), S1 (A), and 2 (B, D, F-H). P.O.R. and M.P. are both clinicians involved in the recruitment of patient samples, as presented in Fig. 1 (J, K), S1 (B-D), and 6 (E). T.N. and Y.R. were involved in the sequencing of the miRNA pulldown products shown in Fig. 3 (D, E). N.C. performed the sequencing shown in Fig. 1 (A). M.M. performed the experiments shown in Fig. 1 (K).

Corresponding author

Correspondence to Sandra Turcotte.

Ethics declarations

Ethics approval and consent to participate

Informed consent and approval were obtained from all the patients and the Ethics Committee of the Centre Hospitalier Universitaire de Sherbrooke, and the Université de Moncton (Approval #2021–4061). All animal experiments were performed following the Canadian Council on Animal Care from our institution and were approved by the Ethics Committee at the Université de Moncton (#19 − 01).

Consent for publication

All authors approved the final version of the manuscript and the submission to this journal.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Novelty and impact: miR-2355-5p is overexpressed in VHL-inactivated clear cell Renal Cell Carcinoma tumors and circulating plasma. MiR-2355-5p expression is controlled through the canonical VHL/HIF-2α axis, and its inhibition reduces angiogenic processes and suppresses tumor growth. The quantification of miR-2355-5p could be indicative of the disease.

Electronic supplementary material

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Page, P.M., Dastous, S.A., Richard, P.O. et al. MicroRNA profiling identifies VHL/HIF-2α dependent miR-2355-5p as a key modulator of clear cell Renal cell carcinoma tumor growth. Cancer Cell Int 25, 71 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12935-025-03711-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12935-025-03711-3

Keywords

Cancer Cell International

ISSN: 1475-2867

Contact us