Long noncoding RNA LEENE promotes angiogenesis and ischemic recovery in diabetes models

Impaired angiogenesis in diabetes is a key process contributing to ischemic diseases such as peripheral arterial disease. Epigenetic mechanisms, including those mediated by long noncoding RNAs (lncRNAs), are crucial links connecting diabetes and the related chronic tissue ischemia. Here we identify the lncRNA that enhances endothelial nitric oxide synthase (eNOS) expression (LEENE) as a regulator of angiogenesis and ischemic response. LEENE expression was decreased in diabetic conditions in cultured endothelial cells (ECs), mouse hind limb muscles, and human arteries. Inhibition of LEENE in human microvascular ECs reduced their angiogenic capacity with a dysregulated angiogenic gene program. Diabetic mice deficient in Leene demonstrated impaired angiogenesis and perfusion following hind limb ischemia. Importantly, overexpression of human LEENE rescued the impaired ischemic response in Leene-knockout mice at tissue functional and single-cell transcriptomic levels. Mechanistically, LEENE RNA promoted transcription of proangiogenic genes in ECs, such as KDR (encoding VEGFR2) and NOS3 (encoding eNOS), potentially by interacting with LEO1, a key component of the RNA polymerase II–associated factor complex and MYC, a crucial transcription factor for angiogenesis. Taken together, our findings demonstrate an essential role for LEENE in the regulation of angiogenesis and tissue perfusion. Functional enhancement of LEENE to restore angiogenesis for tissue repair and regeneration may represent a potential strategy to tackle ischemic vascular diseases.

intimal RNA was isolated from human mesenteric artery by flushing once the inner lumen of mesenteric arteries with TRIzol as published (2).

Cell culture, transfection, and treatment
HMVECs and HUVECs were purchased from and verified for negativity for mycoplasma contamination by Cell Applications, Inc. EC identity was authenticated by using immunostaining, flow cytometry, and the expression of CD144, CD31, and eNOS mRNA. HMVECs at passages 4-6 were cultured in HMVEC growth medium and HUVECs at passages 5-7 were cultured in M199 medium supplemented with growth factors, heparin sodium under standard cell culture conditions (humidified atmosphere, 5% CO2, 37°C) (1). HG condition was generated by adding D-glucose into the culture media to a final concentration of 25 mM. As normal glucose/osmolarity control, mannitol was added at 20 mM to the cells cultured in medium with 5 mM glucose. TNFα was added to the culture media to a final concentration of 100 ng/mL or 5 ng/mL for a combined HG and TNFa treatment (HT) for H3K27ac ChIP. The normoxic cells were kept at 37°C ventilated with 5% CO2 and atmospheric 21% oxygen. The hypoxic cells were maintained in an incubator infused with 2% O2, 5% CO2, and 93% nitrogen. To knock down LEENE or LEO1, cells were transfected with scrambled control LNA or siRNA or LNA targeting LEENE or LEO1 or MYC siRNA (Qiagen) using Lipofectamine-RNAiMAX transfection reagent (Thermo Fisher Scientific) in Opti-MEM (Thermo Fisher Scientific) according to manufacturer's protocol and as published (3). To overexpress LEENE, an adenoviral vector-driven expression of the predominant transcript of human LEENE (NR_026797.1) as previously described (3) was used to infect ECs or inject into leene-KO mice. Ad-GFP was used as a vector control.

Tube formation, scratch, and spheroid sprouting assays
The tube formation assay was performed as previously reported (4). Briefly, HMVECs were plated on a Matrigel (BD Pharmingen)-coated 24-well plate, incubated for 8 hours in 5% CO2 at 37 °C, and examined for capillary tube formation under an inverted microscope and photographed. Three randomly selected views were captured, and the formed tubes were counted. For the scratch assay, HMVEC were seeded onto 24-well plates and grown to confluence. Cell monolayers were carefully wounded with a 200-μl pipette tip to generate a cut of ~1 mm in width. After two washing steps, cells were incubated for 24 hours and area lacking cells determined.
The 3D spheroid sprouting assay was performed as previous described (5). HMVECs cultured to confluency were trypsinized using 0.25% Trypsin/0.53 mM EDTA and counted. Enzyme activity was neutralized with culture media and centrifuged at 200 rcf for 5 minutes. The media was aspirated, and the cells were resuspended at 10 6 cells/mL in fresh media. To form homogeneous aggregates of 500 cell per aggregate, 125 µl of the resuspended cells were added to 3.875 mL of HMVEC media. One mL of 0.3% (W/V) methylcellulose (Sigma M0512-100G) in HMVEC media was then added to the suspension to bring the total volume to 5 mL, resulting in a final density of 500 cells per 20 µL. The cell suspension was then distributed onto the inside lid of a petri dish using a multichannel pipette to form rows of 20 µL droplets. The dish was then inverted, and 5 mL of PBS was added to the bottom of the dish. Cells were incubated overnight. On the next day, cell aggregates were collected, washed, and centrifuged at 100 rcf for 2.5 minutes. The aggregates were resuspended in pre-chilled Matrigel (Corning 356234) to allow for two aggregates per 20 µL.
Aggregates were distributed in 20 µL droplets onto the bottom of the plate. The plate was then turned upside down to form hanging drops and placed into a larger petri dish (to maintain sterility), which was then placed in incubator for 30 minutes. After the Matrigel has begun to gel, the plate was removed from the larger dish and turned upright to allow another hour to fully set. The gelled aggregates were then overlaid with HMVEC media containing 50 ng/mL of VEGF (Sigma V7259-10UG) and incubated for 3 days, with monitoring for sprouting.
Brightfield images of the sprouts were taken using an Amscope MU1000 camera and an Olympus IX50 microscope at 10X magnification. The images were then analyzed in FIJI (ImageJ) using the Sprout Morphology analysis tool (6). The images were first converted to 8-bit binary masks.
Threshold values were then globally adjusted across all images to uniformly darken the background and highlight the aggregates and their sprouts. The images were then manually assessed for bubbles or other artifacts in the gel that the software could misconstrue as an aggregate.
Pixel scale was determined using a hemocytometer, and this was applied globally to all images.
Then the images were batch-run through the analysis package, where the software measured the aggregates and the sprouts.

Measurement of blood pressure and echocardiography
Blood pressure was measured using a noninvasive computerized tail-cuff system (Visitech, Apex, NC) as previously described (7). After the mice were placed in a plastic holder, the occlusion and sensor cuff were positioned on the base of the tail. All the mice were given at least 1 week to adapt to the system prior to blood pressure measurement. Blood pressure was measured at least 20 times in each mouse. Echocardiography was performed as previously described (8). Briefly, mice under conscious condition were used with a Vevo 3100 Ultrasound Imaging System (FUJIFILM VisualSonics). Multiple parameters including heart rate, left ventricular internal dimensions at end of diastole and systole (LVIDd and LVIDs), end-diastolic interventricular septal thickness (IVSd), and LV posterior wall thickness (LVPWd) were determined from the ventricular M-mode tracing.
Percentage fractional shortening (%FS) and ejection fraction (EF) were used as indicators of systolic cardiac function.

Histology, immunostaining, and immunoblotting
Histological examinations were mainly processed by the Solid Tumor Pathology Core at City of Hope. Skeletal muscle from mice was collected and fixed in 4% paraformaldehyde over-night. The Invitrogen). Images were taken using a ZEISS Axio Observer. For immunoblotting of LEO1, antibody against LEO1 (Rabbit polyclonal #A300-175A, BETHYL, 1:1000 dilution) was used as primary, and anti-rabbit (7074S, Cell Signaling Technology, 1:5000 dilution) was used as the secondary antibody.

RNA isolation, RT-qPCR analyses
RNA was extracted from cells and tissues using TRIzol (Thermo Fisher Scientific). The total RNA was reverse transcribed using PrimeScript RT Master Mix (Takara), and cDNAs were used for qPCR analyses using the primers listed in Supplemental Table 5. Samples were subjected to qPCR using iTaq Universal SYBR Green Supermix on a CFX Connect system (BioRad). b-actin was used as the internal control in human and 36B4 in mouse samples.

Single-molecule RNA fluorescent in-situ hybridization (smFISH)
smFISH was performed on human mesenteric arteries using the RNAscope™ Multiplex Fluorescent V2 Assay (ACDBio). Cells grown and treated on coverslips were fixed with 4% paraformaldehyde (PFA) for 30 minutes at room temperature, ethanol dehydrated, pre-treated with hydrogen peroxide for 10 minutes at room temperature, and permeabilized with Protease III (1:10 dilution) for 30 minutes at room temperature prior to probe hybridization. RNAscope® Probe -Hs-linc00520 (502321, ACS) was used to detect human LEENE. Following probe hybridization, the RNAscope assay was developed following the recommended protocol.
The supernatant containing cytoplasmic component was quickly added to TRIzol LS for RNA extraction. The nuclear pellet was gently suspended into 200 μl cold glycerol buffer (20mM Tris pH 7.9, 75mM NaCl, 0.5mM EDTA, 50% glycerol, 0.85mM DTT). Cold nuclei lysis buffer (20mM HEPES pH 7.6, 7.5mM MgCl2, 0.2mM EDTA, 0.3 M NaCl, 1M urea, 1% NP-40, 1 mM DTT) was added and the mixture vortexed and centrifuged. The supernatant containing the nucleoplasmic fraction was mixed with TRIzol LS (Thermo Fisher Scientific) for RNA extraction. Cold PBS (50 μl) was added to the remaining pellet and gently pipetted. After vigorous vortexing to resuspend the chromatin, chromatin-associated RNA was extracted by adding 100 μl chloroform and TRIzol reagent. RNA samples from three different fractions were dissolved with same amount of RNase-free water, and same volume of RNA was used for reverse-transcript and qPCR.

ChIRP-seq
ChIRP was performed as described (3,9). HUVECs were fixed with 1% glutaraldehyde for 10 minutes at room temperature. The pelleted cells were lysed and sonicated for 10 minutes using a "30s ON, 30s OFF" program. The sonicated samples were then centrifuged and 1% of the supernatant was taken as input of ChIRP-DNA-seq and another 1% of the supernatant was taken as the input. About 100 pmol of the probes were hybridized with the residual supernatant at 37 °C for 4 hours, followed by incubation with streptavidin-conjugated magnetic beads for another 30 minutes. Following several rounds of washing, DNA was isolated from the ChIRP precipitates and subsequently used for sequencing. Subsequently, ChIRP-seq libraries were constructed using the KAPA HyperPrep Kit (Roche Diagnostics) following the manufacturer's manual. The DNA was quantified using Qubit double-stranded DNA High sensitivity assay (Thermo Fisher Scientific).
Peptides were purified with Oasis HLB columns. Mass spectrometry was performed on an orbitrap Fusion Tribrid instrument (Thermo) equipped with an Easy-nLC 1000 HPLC system, a 75 μm by 2 cm PepMap C18 trapping column, a 75 μm by 50 cm PepMap RSLC C18 analytical column, and an Easy-Spray ion source (Thermo). Peptides were separated by a 1 h gradient from 0.1% formic acid, 3% acetonitrile to 0.1% formic acid, 30% acetonitrile. Precursor ion scans were acquired in the orbitrap and CID fragments were acquired in the linear ion trap in rapid mode. Data analysis was performed using Proteome Discoverer with the Sequest search engine (Thermo) and Scaffold (Proteome Software).

ChIP assay
ChIP assays were performed as previously described (10). Briefly, HUVECs were treated with 0.75% formaldehyde for 20 minutes at room temperature. Fixation was stopped by adding 125 mM glycine and the cells were collected. The pelleted cells were lysed and sonicated for 4 minutes using a "30s ON, 30s OFF" program at 4 °C. The sonicated samples were then centrifuged and 1% of the supernatant was taken as input. After sonication, the chromatin was incubated with rabbit anti-human H3K4me3 (Rabbit polyclonal #39159, Active Motif) or anti-human H3K27ac (Rabbit polyclonal #39133, Active Motif) conjugated to prewashed Protein A Dynabeads (Thermo Fisher Scientific). Protein and RNA were digested by proteinase K and RNase A, respectively. The purified chromatin DNA was then used as the template for qPCR.

RNA IP (RIP) and Co-IP
RIP was performed as previously described (3)

Nascent RNA pulldown
To capture nascent RNA, newly synthesized mRNA was isolated using the Click-iT Nascent RNA Capture Kit (Thermo Fisher Scientific). HUVECs were synchronized with 2% FBS in M199 medium for 8 hours, followed by incubation in 0.2mM of 5-ethymyluridine (EU, an alkynemodified uridine analog which is incorporated into the nascent RNA) for another 24 hours, and total RNA was isolated using TRIzol. A copper-catalyzed click reaction was performed using 5 μg RNA with 0.5mM azide-modified biotin. The mixture was incubated at room temperature for 30 minutes following RNA precipitation. Biotin-labeled EU-RNA was then pulled down by mixing with Streptavidin T1 magnetic beads at room temperature for 30 minutes and the unbound RNA was washed away. The cDNA synthesis was performed directly on the beads using the Superscript VILO cDNA synthesis kit (Thermo Fisher Scientific), followed by qPCR analysis.  Figure 5E.

Supplemental Figure 10. Enriched pathways with LEENE rescued genes. (A)
Top 6 Biological Pathway terms from pathway enrichment analysis with GO term among LEENErescued genes, namely the overlap between downregulated by leene KO (WT+GFP vs KO+GFP) and upregulated by LEENE overexpression in KO (KO+GFP vs KO+LEENE), plotted with Pvalue and gene count. (B) WT and KO mice were subjected to HLI fed a HFHS diet and received Ad-GFP or Ad-LEENE as in Figure 5. qPCR of KDR mRNA in ischemic tissues. n=3/group. Bar graphs represent mean ± SEM. *P =0.003 and 0.05 between indicated groups based on ANOVA followed by Tukey's test.

Supplemental Figure 11. Enriched pathways in VSMCs from ischemic muscles with LEENE OE.
Leene-KO mice were subjected to HLI and Ad-GFP/LEENE injection. The hindlimb muscles underwent scRNA-seq analysis as presented in Figure 7. DEGs in VSMCs identified from scRNAseq were subjected to pathway enrichment analysis. The top 30 enriched biological pathways are shown. Figure 7, with gene names showing for all involved ligands and receptors.

Supplemental Figure 13. qPCR of LEENE in different subcellular compartments.
HUVECs were infected with Ad-GFP or Ad-LEENE for 72 hours, followed by subcellular fractionation to obtain cytoplasm (Cyt), nucleus (Nuc), and chromatin-bound fractions (Chr) and qPCR (n=3/group). Data are represented as mean ± SEM. *P =0.05, 0.02, 0.0001 based on t test. Figure 14. ChIRP-qPCR of LEENE-bound DNA. qPCR was performed with chromatin pulldown using probes specific for LEENE or LacZ RNAs. eNOS and KDR promoters were detected in the DNA extracted from precipitates (n=2/group). Data are represented as mean ± SEM. HUVEC total protein lysates were used for LEO1 IP and LEO1 protein was detected using immunoblotting in the IP beads and flow-through. (B) ChIRP was performed with ECs infected by Ad-GFP/Ad-LEENE in biological replicates. All 10 probes were used. LEO1 was detected using western blotting. (C) Odd and even probes were used in ChIRP followed by LEO1 detection using immunoblotting. In another ChIRP sample with all 10 probes, RNase was added to degrade RNA.