Reducing In-Stent Restenosis

Background Drug-eluting stents reduce the incidence of in-stent restenosis, but they result in delayed arterial healing and are associated with a chronic inflammatory response and hypersensitivity reactions. Identifying novel interventions to enhance wound healing and reduce the inflammatory response may improve long-term clinical outcomes. Micro–ribonucleic acids (miRNAs) are noncoding small ribonucleic acids that play a prominent role in the initiation and resolution of inflammation after vascular injury. Objectives This study sought to identify miRNA regulation and function after implantation of bare-metal and drug-eluting stents. Methods Pig, mouse, and in vitro models were used to investigate the role of miRNA in in-stent restenosis. Results We documented a subset of inflammatory miRNAs activated after stenting in pigs, including the miR-21 stem loop miRNAs. Genetic ablation of the miR-21 stem loop attenuated neointimal formation in mice post-stenting. This occurred via enhanced levels of anti-inflammatory M2 macrophages coupled with an impaired sensitivity of smooth muscle cells to respond to vascular activation. Conclusions MiR-21 plays a prominent role in promoting vascular inflammation and remodeling after stent injury. MiRNA-mediated modulation of the inflammatory response post-stenting may have therapeutic potential to accelerate wound healing and enhance the clinical efficacy of stenting.

combination of subcutaneous Hypnorm (25 mg/kg, Bayer) and Hypnovel (25 mg/kg, Roche). The right common carotid artery was mobilized free from the thoracic inlet to its bifurcation, ligated, and divided between ties at its midpoint. Polyethylene cuffs (0.65 mm diameter, Portex LtD) were placed at both ends of the artery, were everted over the cuff and secured with suture and the stented aorta was interposition grafted between the two ends of the carotid artery by sleeving the ends over the cuffed artery cuff and secured with 8/0 suture. Mice were allowed to recover in heated chambers for 24 hours and returned to normal housing conditions and maintained on aspirin supplemented water and normal chow diet for a further 28 days. To reduce the risk of in-stent thrombosis (IST), one week prior to surgery each mouse was pre-dosed with aspirin supplemented drinking water at a concentration of 300mg/L and continued until the end of the experiment. At harvesting animals were euthanized and transcardial perfusion with heparinized saline performed. The stented graft was removed and fixed for 12 hours in 4% Paraformaldehyde (PFA).

Electrolysis
Following fixation in paraformaldehyde the stented murine arteries were immersed in 5% (w/v) citric saline and electrochemical dissolution performed by passing a small direct current through the stent as described previously (4).

Immunohistochemistry
For histological analysis, vessels were fixed in 4% paraformaldehyde for 24 hours. Murine stented aortas were electrolyzed to remove the stents, as described above, dehydrated in graded ethanol cleared in Histo-Clear and embedded in paraffin wax, before sectioning and subsequent staining. In brief serial paraffin sections were dewaxed and rehydrated. Tissue sections were subjected to citric acid antigen retrieval and endogenous peroxidase activity was inhibited by incubation with 3% hydrogen peroxide. After blocking sections with 20% (vol/vol) rabbit or goat serum in PBS, sections were incubated overnight at 4 0 C with purified rat mAb against mouse macrophages (mac-2; BD Biosciences, Oxford, UK) at 3.12µg/ml, mouse mAb against α-smooth muscle actin (Sigma) at 82µg/ml in 1% (wt/vol) BSA in PBS, proliferating cell nuclear antigen at 2µg/ml (PCNA, Abcam, Cambridge, UK), rat anti-mouse CD-31 (Dianova, Germany) at 5µg/ml or Rabbit anti-mouse YM-1 (Stem Cell technology) at 3.2µg/ml or mouse anti human CD68 (DAKO, UK) at 2µg/ml. Sections were then incubated with the appropriate biotinylated secondary antibody (DAKO) diluted 1:200 in 3% (vol/vol) goat serum in PBS, and then horseradish peroxidise labelled Extravidin (diluted 1:400 in 1% (wt/vol) BSA in PBS). Optimal visualization of staining was achieved using 3',3'-diaminobenzidine tetrahydrochloride dihydrate and hydrogen peroxide. Sections were counter-stained with hematoxylin. A negative control, where the primary antibody was replaced with mouse, rat or rabbit IgG at the same dilution, was routinely included. All immunohistochemistry staining was quantified as the area of positive staining expressed as a percentage of total neointimal area.

In situ hybridization
In situ hybridization was carried out on routinely processed tissue sections which were rehydrated as described above. These tissue sections were then treated with 0.5U/ml proteinase K (Sigma) at 37 degrees for 15 minutes, then fixed with 4% paraformaldehyde for a further 10 minutes. After washing with phosphate buffered saline (PBS), slides were incubated with hybridization buffer for 1h at 52 or 58 degrees for miR-21 and scramble probe, respectively. Slides were then hybridized with 40nM double DIG-labelled miR-21 or double DIG-labelled scramble probe (Exiqon) at 52 or 58 degree overnight. After washing and blocking, slides were incubated with anti-DIG-AP Fab fragments (Roche) in blocking buffer at 4 degrees overnight, and washed with PBST (PBS plus 0.1% Tween 20, Sigma) and AP buffer (0.1M Tris-HCL pH9.5, 150mM NaCL, 5mM MgCL2, 0.1% Tween20). MiR-21 was visualized with BM purple solution (Roche) for 16-24 hrs at room temperature until the staining was visible under microscope.

Vessel storage and RNA isolation
Harvested porcine coronary arteries were placed in RNAlater ® -ice (Invitrogen, Paisley, UK Paisley, UK) and stored at -80 0 C until the day of isolation. RNA from arteries was isolated following disruption of the vessels under liquid nitrogen using a pestle and mortar, and these vessel fragments were placed in pre-cooled Qiazol (Invitrogen, Paisley, UK) and homogenized using a tissue homogenizer (Polytron, Switzerland). The RNA was processed through miRNeasy kit (Qiagen, Hilden, Germany) following the manufacturer's instructions, treated with DNAse 1 (amplification grade; Sigma, St. Louis, MO, USA) in order to eliminate genomic DNA contamination. The yield and purity of RNA were measured using a NanoDrop ND-1000 Spectrophotometer (Nano-Drop Technologies, Wilmington, DE, USA), and RNA integrity was assessed using the RNA 6000 Nano LabChip kit (Agilent Technologies).

Global MicroRNA Expression Profiling
The global profiling for miRNAs in control un-stented arteries and stented porcine coronary arteries was performed using the TaqMan Low-Density Array Human MicroRNA Panel v2.3 (Applied Biosystems, CA, USA), which included Cards A and B in a 384 well format. The A cards contained 377 human miRNAs and 3 endogenous controls, B cards contained 287 miRNA and 6 endogenous control miRNAs, these cards were experimentally processed following manufacturer's instructions. In brief, total RNA (100 ng) was first reverse-transcribed with the Multiplex RT pool set A or B (Applied Biosystems, CA, USA) through a reverse transcription (RT) step using the High-Capacity cDNA Archive kit (Applied Biosystems, CA, USA), wherein a stem-loop RT primer specifically binds to its corresponding miRNA and initiates its reverse-transcription. The RT mix included 50 nM stem-loop RT primers, 1 x RT buffer, 0.25 mM each dNTPs, 10 U/µl MultiScribe reverse transcriptase, and 0.25 U/µl RNase inhibitor. The 7.5 µl reaction mixture was then incubated for 30 min at 16 0 C, 30 min at 42 0 C, 5 min at 85 0 C, and then held at 4 0 C. The RT products were subsequently amplified with sequence-specific primers; using the Applied Biosystems 7900 HT Real-Time PCR system. Six µl of RT product were added to 444 µl nuclease-free water and mixed with 450 µl Taqman Universal Master Mix II, No UNG, then dispensed into a 384 well plate by centrifugation. The reactions were incubated in the plate at 95 0 C for 10 min followed by 40 cycles of 95 0 C for 15 sec and 60 0 C for 1 min.
For analysis, fold-changes for each miRNA were normalized to U6 since this miRNA was the most suitable endogenous miRNA in porcine tissue. The relative expression levels between samples were calculated using the comparative delta Ct (threshold cycle number) method with a control sample (normal) as the reference point. Data analysis was performed by using the SDS software version 2.3 (Applied Biosystems, CA, USA) and the baseline and threshold were automatically set. Data were normalized and then analyzed to identify miRNAs that are differentially expressed between the control (un-stented arteries) and arteries subjected to stenting for a period of 7 or 28 days. Data were analyzed using Data Assist analysis software, version 3 (Applied Biosystems, CA, USA).

miRNA quantification by real-time polymerase chain reaction (qRT-PCR)
cDNA was synthesized from RNA using stem-loop reverse transcription primers (applied biosystems, Foster City, CA, USA). qRT-PCR was performed using TaqMan® universal master mixII with Taqman microRNA expression probes according to the manufacturer's instructions. Expression was normalized to U6 or Tbp for microRNA or target expression data, respectively and expressed as relative expression levels of miRNAs or mRNAs of interest as described in (5). Results are shown relative to the experimental control using the -2∆∆Ct method described by Livak (6). Assay information available on request.

Proliferation and migration assays
Murine aortas were harvested from male mice aged 8 -12 weeks Vascular SMCs were isolated and cultured as described by Johnstone et al (7). Briefly aortic SMC were plated at sub-confluence in serum depleted medium before cells were subjected to scratch wounds with a 200ul pipette tip and incubation with 15% fetal calf serum of PDGF-BB for a period of 24hrs. Proliferation of VSMCs was assessed using a commercially available 5-Bromo-2'-deoxyuridine (BrdU) assay as per manufacturer's protocol (Millipore, Livingstone, UK). Murine VSMCs, from miR-21 WT and KO mice, passage 4 to 6 were seeded into 96 well plates at a density of 7x10 3 cells/well in DMEM with 10% FCS and incubated overnight at 37ºC in the presence of 5% CO 2 . VSMCs were quiesced for 24 hours in DMEM with 0.2% FCS. Following quiescence cell proliferation was stimulated by the replacement of media containing one of four experimental conditions: 10% FCS, Platelet Derived Growth Factor-ββ (PDGF-ββ) (R&D Systems, Abingdon, UK) 10ng/mL, PDGFββ 20 ng/mL or PDGF-ββ 50 ng/mL. 0.2% FCS was used as control. All experimental conditions were conducted in biological triplicate and technical duplicate. Experiments were terminated at 48 hours.
Scratch assay was utilized to assess cell migration as described previously (8). Briefly, VSMCs isolated from miR-21 WT and KO mice were seeded into 12 well plates at a density of 1x10 5 cells per well and grown in DMEM with 10% FCS, at 37ºC in the presence of 5% CO 2 until fully confluent. The cells were quiesced as described previously. Horizontal lines were drawn on the under surface of each well to act as a reference for measurements. Three straight, scratches perpendicular to the reference lines were incised in the cell monolayer of each well using a sterile 200 μl pipette tip. Cellular debris was removed by gently washing cells once with 1 mL of sterile PBS. Cells were incubated with DMEM containing 0.2% FCS, 10% FCS or PDGFββ 20 ng/mL.

Generation of Bone Marrow-Derived Macrophages
Bone Marrow Derived Macrophages (BMDM) were generated as described previously (9). Briefly, bone marrow cells were isolated from femurs and tibiae of WT and miR-21 KO mice. Cells were filtered through a 100 μm filter and then spun at 300 g for 5 minutes. Following treatment with Red Cell Lysis Buffer (Sigma) cells were grown for 6 days in non-bacteriological petri dishes (2x10 6 per plate) in 10mls complete media (RPMI-1640/10% heat-inactivated fetal bovine serum/2mM L-glutamine, 100 units/ml penicillin and 100 µg/ml streptomycin) containing 50 ng/mL recombinant murine macrophage colony stimulating factor (M-CSF, Peprotech). Media was refreshed on day 3 and on day 6 cells were scraped off using a cell scraper, counted and plated out at 0.5x10 6 cells/well on a 24 well plate prior to activation with IL-4 (2 ng/ml, Peprotech) or Lipopolysaccharide (LPS) (100 ng/ml, Bacterial LPS from E. coli, Serotype O111:B4 TLRgrade™, EnzoLifeSciences) for 20 hours.

In vitro invasion studies on BMDM from WT and miR-21 KO mice
Migration of BMDM was measured by the invasion of cells through Matrigel-coated Corning TM transwell inserts (Sigma). Transwell inserts containing 8µm pores were coated with Matrigel (40µL/well; BD Biosciences, UK). Cell suspensions (200µL; 1x10 5 cells) in RPMI 1640medium (supplemented with 100IU/ml penicillin, 100µg/mL streptomycin and 2mmol/L L-glutamine) was added to the upper chamber. The same medium (600µL) supplemented wit 1 or 10ng/mL of murine recombinant monocyte chemoattractant protein-1 (MCP-1) (R&D Systems, Abington, UK) was placed in the lower wells. Cultures were incubated for 6 hours and the cells on the under surface of the transwell membrane were fixed with 4% paraformaldehyde, stained with haematoxylin, counted in 10 random fields.

Luminex Assay
Macrophage culture supernatants were collected and analyzed on a Cytokine Mouse 20-Plex Panel (Life Technologies) according to the manufacturer's instructions.

Statistics
All data are mean ± standard error of mean (SEM). For the comparison of mean values a Bartlett's test for equal variances was performed -there was no evidence of heterogeneous variances between groups for any of the comparisons. Visual assessment was used to check for any lack of normality; as there was no evidence of this, one way ANOVA followed by a Tukey's multiple comparison test (for comparison of more than two groups) or Students t-test (for comparison of two groups) were carried out. For all the q-PCR experiments, values are expressed as fold change. All statistical analysis using Graph Pad Prism v4 (GraphPad Software ® ). The microRNA array data were analyzed in DataAssist TM software (Life Technologies). Comparisons of in vitro SMC proliferation and migration were performed by 2-way ANOVA and Bonferonni's post-hoc test.

S
Upstream regulatory and causal network analysis on predicted gene targets using Ingenuity Pathway Analysis. TNF-α was identified as a master regulator for predicted gene targets of (A) mir-21-5p and (B) mir21-3p. Known connections of both miRNAs to target genes are indicated in light blue. The letters on the relationship lines represent functional affects with the number of references to support the connections.