**1. Introduction**

Clear cell renal cell carcinoma (ccRCC), the most common form of sporadic and inherited kidney cancer, is highly associated with mutations in the von Hippel-Lindau (*VHL*) gene [1,2]. The protein product of the *VHL* gene (pVHL) is an E3 ubiquitin ligase involved in the degradation of hypoxia-inducible transcription factor subunits (HIF1α). Under normal oxygen tension, hydroxylated HIF1α can be recognized by the ubiquitin ligase complex containing pVHL and rapidly degraded. Upon hypoxia or loss of functional pVHL, HIF1α-subunits can no longer be hydroxylated and begin to accumulate. Stabilized HIF1α activates the expression of a large suite of downstream target genes (Erythropoietin (*EPO)*, vascular endothelial growth factor (*VEGF*), etc), the actions of which are vital to promote angiogenesis. However, many of the changes initiated by the stabilization of HIF1<sup>α</sup>, such as increased angiogenesis, an upregulation of antiapoptotic signaling, and a shift to anaerobic glycolosis, can contribute to tumor growth and survival. People born with a mutation in one *VHL* allele may acquire somatic mutations in the second allele, resulting in consequent angiogenic symptoms and a variety of tumors, including ccRCC [3]. Another hallmark of ccRCC is the activated phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3k)/AKT pathway signaling, higher levels of which is significantly correlated with a worse survival rate [2], although the mechanism by which this occurs is still not fully understood.

MicroRNAs (miRNAs) are small noncoding RNAs that posttranscriptionally regulate the expression of groups of target genes by inhibition of the translation of their targeting messenger RNAs (mRNAs) or marking these mRNAs for degradation. miRNAs are key regulators in many physiological and pathological processes [4], including the dynamic regulation of ccRCC during tumor progression [2]. By promoting the expression of vascular endothelial growth factor (*VEGF*), VHL/HIF1 signaling increases Cyclic adenosine monophosphate (cAMP) response element binding protein (CREB) levels, a transcription factor which upregulates the expression of pro-angiogenic miR-212/132 [5]. This implies that pVHL loss-of-function would stimulate miR-212/132 expression and therefore contribute to excessive angiogenesis. In this study, using a combination of cellular models, patient ccRCC material with biallelic loss of VHL and a previously described vhl−/− mutant zebrafish model, we show that miR-212/132 is upregulated after VHL knockdown or mutation and that this upregulation is at least partially responsible for pro-angiogenic effects. In order to grow, cancer tissues such as ccRCC have developed various strategies to provide sufficient blood supply by promoting angiogenesis [6]. Many of the common pharmaceutical treatments for ccRCC, such as sunitinib, use the strategy of reducing pathophysiological angiogenesis [7]. Conversely, many different strategies have been tested to improve perfusion of certain ischemic tissues or engineered tissue constructs by promoting neovascularization, which is essential for functional recovery of the organ after ischemic events or survival of transplanted engineered tissue constructs [8]. A scarcity of functional pVHL induces excessive vascular outgrowth, which further is enhanced by miR-212/132 expression, providing an exciting target for the modulation of angiogenesis.

#### **2. Materials and Methods**

#### *2.1. MicroRNA In Situ Hybridization*

microRNA in situ hybridization was performed with a modified microRNA in situ hybridization method as described [9]. Formalin-fixed paraffin-embedded tumor tissue from two ccRCCs and one healthy donor kidney dating from September 2006 were collected from the pathology archives of the University Medical Centre Utrecht (UMCU) after authorization of the UMCU institutional review board in accordance with Dutch medical ethical guidelines. Sequencing results of *VHL* identified no variants in the normal healthy kidney. However, in ccRCC #1, in addition to the already known germline deletion of *VHL* exons 1 and 2, an additional somatic mutation was found in the tumor (c.277delG/p.Gly93Ala\_fs\_x158). ccRCC #2 has a germline mutation c.266T> p.Leu89Pro and a somatic mutation of c.419-420delTC/p.Leu140Gln\_fs\_x142. Mutation analysis of these tumors has been previously published [10].

Paraffin samples were first deparaffinized with tissue clear (Cat# 1426, SAKURA) followed with 10 min of proteinase K treatment (5μg/ml, Cat# 03115828001, Roche). Hybridization was performed with 10 nM DIG-labeled miRCURY LNA miRNA detection probes in hybridization buffer (Urea (2 M), 2.5× SSC, 1× Denhardt's, 200 μg/ml yeas<sup>t</sup> tRNA, 0.1% CHAPS, 0.1% Tween, and 50mg/ml heparin) for miR-132 (Cat# 38031-15, Exiqon). Sections were subsequently incubated with anti-DIG alkaline phosphatase antibody (1:1,500, Cat# 1093274, Roche). To block endogenous alkaline phosphatase activity, sections were incubated with levamisole solution (Cat. X3021, DAKO), followed by NBT/BCIP (Cat# K0598 DAKO) incubation for visualization. A light Eosin counter staining was performed to

visualize histology of the tissue. Images were taken with an Olympus microscope (BX53) under bright field.

#### *2.2. Cell Culture and Transfection*

Human umbilical vascular endothelial cells (HUVECs) were cultured in EGM2 (Lonza, cat# cc-3156) according to manufacturer's instructions, and all experiments were performed before passage 7. HUVECS were either transfected with validated siVHL (ID: s14790), siPTEN (ID: s61222), silencer select negative control #1 (cat# 4390843), mirVana miRNA mimic negative control (cat# 4464085), hsa-miR-132-3p mimics (ID: MC10166), hsa-miR-212-3p mimics (ID: MC10340), mirVana miRNA inhibitor control1 (cat# 4464077), hsa-miR-132-3p inhibitor (ID: AM10166), or hsa-miR-212-3p inhibitor (ID: AM10340) (all from Life Technologies) using Lipofectamine 2000 (Life Technologies). The transfection was performed with a final concentration of 20 nM in opti-MEM reduced-serum medium (Cat# 31985062, Life Technologies) and replaced with fresh EGM2 after 6 hours. Cells were harvested 72 hours after transfection for protein or RNA analysis.

#### *2.3. RNA Isolation and RT-PCR*

Total RNA was isolated with Tripure Isolation Reagent following manufactory's instructions (Roche Applied Science) and treated with Dnase to remove potential genome DNA contaminations. cDNA was synthesized using the iScriptTM cDNA Synthesis Kit (Bio-Rad). Quantitative real-time polymerase chain reaction (qRT-PCR) was performed with iQ SYBR Green Supermix (Bio-Rad). The following primers were used for detection of human genes: *GAPDH* forward 5--GGCATGGACTGTGGTCAT GA-3- and reverse 5--TTCACCACCATGGAGAAGGC-3- ; *PTEN* forward 5--TGGATTCGACTTAGACT TGACCT-3- and reverse 5--GGTGGGTTATGGTCTTCAAAAGG-3- ; *VHL* forward 5--CAGCTACCGA GGTCACCTTT-3- and reverse 5--CCGTCAACATTGAGAGATGG-3-; the following primers are used for *zebrafish*: *ptena* forward 5--CCAGCCAGCGCAGGTATGTGTA-3- and reverse 5--GCGGCTGAGGAAAC TCGAAGATC-3- ; *ptenb* forward 5--GCTACCTTCTGAGGAATAAGCTGG-3- and reverse 5--CTTGATG TCCCCACACACAGGC-3- ; *rpl13*α forward 5--TCTGGAGGACTGTAAGAGGTATGC-3- and reverse 5--AGACGCACAATCTTGAGAGCAG-3- (All primers from Integrated DNA Technologies, Coralville, IA, USA).

#### *2.4. In Vitro Angiogenesis Assay*

HUVECs (Lonza) and human brain vascular pericytes (Cat#1200, Sciencell) were cultured on gelatin-coated plates in EGM2 medium (Lonza cat# cc-3156) and DMEM (10% FCS; Lonza), respectively, in 5% CO2 at 37 ◦C. Lentiviral transduced HUVECs expressing green fluorescent protein (GFP) and pericytes expressing red fluorescent protein (RFP) were used between passage 6–8. HUVECs were transfected either with siRNA or anti-miRs as described above. In order to monitor the e ffects of miR-132 and miR-212 in angiogenesis, transfected HUVEC-GFP and Pericytes-dsRed were suspended in a 2.5 mg/ml collagen type I (BD Biosciences) as described by Stratman [11]. Cocultures were imaged after 48 h and 120 h incubation in 5% CO2 at 37 ◦C by fluorescence microscopy, followed by automated thresholding and skeletonization of the images using a commercial analysis system (Angiosys, Buckingham, UK). These images were used for automated tubule length measurement, junction measurement, and other analyses according to manufacturer's instructions.
