The Effect of Oligonucleotides to c-myb on Vascular Smooth Muscle Cell Proliferation and Neointima Formation After Porcine Coronary Angioplasty
Abstract Proto-oncogenes, including c-myb, are expressed early after vascular injury. The application of antisense oligodeoxynucleotides (AS-ODNs) against these genes inhibits cell proliferation and neointima formation in small animals and in peripheral arteries. The aim of this study was to investigate the specificity of action of AS-ODN–c-myb in vitro and to assess its effect, when delivered locally, on neointima formation after percutaneous transluminal coronary angioplasty (PTCA) in porcine coronary arteries. AS-ODN–c-myb inhibited the proliferation of vascular smooth muscle cells (VSMCs) in vitro in a dose-dependent manner. There was a corresponding reduction in steady state levels of c-myb mRNA and protein. Expression of another early gene, c-fos, was unaffected. S1 nuclease analysis demonstrated intact full-length AS-ODN–c-myb retrieved from VSMCs in culture after 12 hours. A range of ODNs, related and unrelated to c-myb, with and without a GGGG sequence, inhibited VSMC proliferation. Phosphorothioated AS-ODN–c-myb was 30 times less potent than unphosphorothioated AS-ODN–c-myb. PTCA induced porcine coronary artery neointima formation. c-myb mRNA was maximally induced 18 hours after injury. Unmodified AS-ODN–c-myb, sense-ODN–c-myb, saline, or nothing was delivered immediately after balloon dilatation via a double-skinned porous balloon (Transport, SciMed). Fluorescence-labeled AS-ODN–c-myb was deposited throughout the vessel wall. Mean maximum intima/media cross-sectional area 4 weeks after PTCA was reduced with AS-ODN–c-myb by 79% compared with saline (P<.05), 82% compared with sense-ODN–c-myb, and 63% compared with nothing (P<.10). Conclusions are as follows: (1) c-myb is expressed in VSMCs after vascular injury. (2) AS-ODN–c-myb is retained intact in VSMCs, reducing their proliferation in vitro in dose-dependent fashion, with reduction in c-myb mRNA and protein, whereas sense-ODN–c-myb is not. (3) A range of ODNs can reduce VSMC proliferation by a non–sequence-specific mechanism. (4) Phosphorothioate protection of antisense molecules may reduce their efficacy. (5) Local delivery of unmodified AS-ODN–c-myb via the Transport catheter reduces neointima formation after porcine PTCA. (6) Local delivery of fluid may exacerbate neointimal thickening.
Restenosis remains the major limitation of PTCA, with a rate of between 20% and 50%, according to the definition used, in the first six months.1 Luminal encroachment in patients with restenosis is the result of not only vessel wall thickening resulting from excess accumulation of VSMCs, extracellular matrix, and collagen deposition2 but also vascular shrinkage and remodeling.3 Both proliferation and migration occur within the vessel wall and may be directed by the expression of cellular growth factors and cytokines, which signal via proto-oncogene expression.4 Mechanical splinting with stents shows promise at reducing vascular remodeling, but intimal thickening is increased with their use,5 and the stent restenosis rate is, at best, 18% to 30%.6 7 A number of pharmacological agents have been shown to inhibit VSMC proliferation in vitro, but only a few have proven efficacy, usually in specific clinical settings.8 9 10
Several proto-oncogenes have been shown to be upregulated in proliferating VSMCs, notably c-myc,11 c-fos,11 and c-myb.12 c-myb was first identified as a transforming gene from two retroviruses associated with avian myeloblastic leukemia,13 human myeloid leukemia,14 and human embryonal neural tissue.15 Human c-myb has a transcript of 3.8 kb with three conserved DNA binding domains.16 c-myb expression peaks in chicken embryo cells 4 hours after growth stimulation; c-myc, by comparison, peaks at 1 hour.17 c-myb expression is low in quiescence and increases in mid to late G1 phase.12 Inhibition of its expression prevents entry into S phase.12 c-myb has been implicated in the mechanism of heparin-mediated inhibition of VSMC proliferation: heparin blocks the expression of c-myb and the early entry into S phase, with little effect on c-fos and c-myc.18 c-myb has been identified in VSMCs from several species, eg, bovine18 and rat,19 and, as reported in the present study, in proliferating porcine and human VSMCs. c-myb is, therefore, a potential target in the prevention of VSMC proliferation leading to restenosis.
The antisense approach to therapy has been successfully used in vitro in the suppression of VSMC proliferation, using AS-ODNs directed against proliferating cell nuclear antigen,20 nonmuscle myosin heavy chain,21 c-myc,22 23 24 and c-myb,21 among others. More recently, in vivo models of restenosis have used AS-ODNs to reduce intimal hyperplasia. These have been directed against c-myc (rat carotid24 25 and pig coronary26 arteries), cdc-2 kinase (rat carotid),27 and c-myb (rabbit iliac28 and rat carotid25 29 30 arteries). However, the mechanism of action of AS-ODNs, originally thought to involve neutralization of specific mRNAs with complementary sequences, has recently become the subject of much debate.31
Systemic AS-ODN administration in the dose required to achieve VSMC growth suppression may be toxic. In particular, there is a potential for sequence-specific side effects on healthy proliferating tissues. There are also nonspecific predominantly cardiovascular side effects, eg, hypotension.32 Therefore, local delivery to the intended site of action has been pursued as a method of administering AS-ODNs with the minimum systemic dose. Application of pluronic gel as a carrier for AS-ODN–c-myb24 25 29 is not clinically practicable in the context of restenosis. Balloon catheters with small (25-μm) pores are capable of local vascular drug delivery, but at the expense of “jetting,” which may cause morphological disruption.33 Therefore, in the present study, a double-skinned balloon with large (250-μm) pores (the Transport, SciMed) was used. It allows intramural fluid delivery and primary dilatation to be performed independently, but with the same device. The Transport was used to perform over-sized balloon inflation, thus creating injury and subsequent neointima formation, as a model for restenosis. It was also used to perform a single local delivery of AS-ODN–c-myb immediately after PTCA to the site of trauma.
This aims of the present study were (1) to determine the specificity of the effect of AS-ODN–c-myb on VSMC proliferation in vitro and (2) to investigate its effect, when delivered locally by porous balloon, on neointimal hyperplasia at the site of porcine coronary angioplasty.
Materials and Methods
Short segments of human saphenous vein were obtained from patients undergoing coronary artery bypass grafting. Porcine aorta was obtained from animals undergoing terminal anesthesia after PTCA. Bovine VSMCs were obtained from a commercial source (NIA Aging Cell Culture Repository, Coriell Institute for Medical Research, Camden, NJ). All clinical specimens were obtained with local ethical committee approval, and animal experiments were performed under a UK Home Office license. VSMCs were cultured using an explant technique and maintained in DMEM supplemented with penicillin (100 μg/mL), streptomycin (100 U/mL), amphotericin (2.5 μg/mL), glutamine (2 mmol/L), nonessential amino acids (5 mL/500 mL) (all from GIBCO-BRL), and 20% (primary cultures) or 10% (passaged cells) FCS. They were equilibrated with 95% air/5% CO2 at 37°C. Cells were passaged using 0.25% trypsin (GIBCO-BRL) and 0.02% EDTA (Sigma Chemical Co). SMCs were used for experiments at passages 2 and 3. The identity of SMCs was confirmed by immunocytochemistry for SMC α-actin.
Synthesis and Purification of ODNs
Antisense and sense unmodified ODNs to the human sequence of c-myb mRNA were made by using an automated DNA synthesizer (Applied Biosystems). After deprotection, ODNs were purified by HPLC (Oswel DNA), and concentrations were determined by spectrophotometry. ODNs were used in their unprotected unphosphorothioated form, unless otherwise stated. The sequences used for sense and antisense were as follows: 5′ GCC CGA AGA CCC CGG CAC 3′ and 5′ GTG CCG GGG TCT TCG GGC 3′, respectively.15 These correspond to the sequence immediately adjacent to the initiation codon in human VSMCs. Other sequences, synthesized in the same way, were as follows: murine AS-ODN–c-myb (5′ GTG TCG GGG TCT CCG GGC 3′), AS-ODN–c-myb 2-bp mismatch (5′ GTG CCG TGC TCT TCG GGC 3′), human scrambled AS-ODN–c-myb (5′ GCT GTG GGG CGG CTC CTG 3′), AS-ODN ICE (5′ GGC CGA CAA GGT CCT GAA 3′), and sense-ODN ICE (5′ CAG GAC CTT GTC GGC CAT 3′). FAM-ODNs were synthesized using the c-myb antisense, labeled with 5′-FAM, and subsequently purified by HPLC (Oswel DNA).
Cell Proliferation Studies
VSMCs were plated into 96-well plates (5×103 cells per well) in the presence of 10% FCS. At 24 hours, cells were washed and rendered quiescent with 0.5% FCS for a further 48 hours. After this period, proliferation was reinitiated with 10% FCS in the presence of 0.05 to 30 μmol/L ODNs. These comprised human sense-ODN–c-myb, human AS-ODN–c-myb, murine AS-ODN–c-myb, murine AS-ODN–c-myb 2-bp mismatch, human scrambled AS-ODN–c-myb, AS-ODN ICE, and sense-ODN ICE. After 24 hours, [3H]thymidine (1 μCi/mL; specific activity, 25 μCi/mmol; Amersham International) was added, and the cultures were maintained for a further 24 hours. [3H]Thymidine incorporation was measured in trichloroacetic acid–precipitable material by scintillation spectroscopy. Each experiment was performed in triplicate and repeated on VSMCs isolated from several separate donors (human, porcine, and bovine). In separate sets of experiments, cells were plated into 24-well plates, and the proliferation assay was performed as described above without the addition of [3H]thymidine. Cells were trypsinized after exposure to ODN for 48 hours, and cell counts were determined using a hemocytometer. Trypan blue exclusion was used to assess cell viability. Cell recovery experiments were carried out by performing proliferation assays as above, washing the cells, and allowing them to recover for 24 hours with the addition of [3H]thymidine. They were then harvested for scintillation spectroscopy.
c-myb and c-fos mRNA Analysis
c-myb and c-fos mRNA were analyzed by RT-PCR. Detection of c-fos mRNA was used to ensure that AS-ODNs to c-myb did not alter mRNA levels of a nontargeted transcription factor gene that is also involved in cellular proliferation. RT-PCR was used instead of Northern analysis because arterial segments yielded <80 μg total RNA and because the low abundance of the mRNA in question mandated amplification. Arterial segments or cultured VSMCs were harvested for isolation of RNA using the one-step phenol/chloroform method of Chomczynski and Sacchi.34 Briefly, tissue was harvested in RNAzol (Biogenesis Ltd), RNA-precipitated using ice-cold isopropanol (BDH Chemicals), and washed in ice-cold ethanol (75% [vol/vol]) (BDH Chemicals). The pellets were resuspended in sterile water and checked for purity and yield by spectrophotometry. c-myb RNA was identified using RT-PCR with the ubiquitous mRNA 7B6 (for human VSMCs)35 or β-actin (for porcine VSMCs) as a control gene transcript. Total RNA (2 μg) was reverse-transcribed in a 40-μL reaction containing AMV reverse transcriptase (1.25 U/mL), random hexanucleotide primers (2.5 μmol/L), MgCl2 (5 mmol/L), Tris-HCl (10 mmol/L) (pH 9.0), KCl (50 mmol/L), Triton X-100 (0.1%), dNTPs (1 mmol/L each), and RNAsin (1 U/mL). Incubation was at 42°C for 60 minutes. RNA-cDNA hybrids were used immediately as template for RT-PCR. For PCR, each 50-μL reaction contained 10 μL cDNA derived from 0.5 μg RNA, 7.5 pmol of each primer, 1.25 U Taq DNA polymerase, 10 mmol/L Tris-HCl (pH 9.0), 50 mmol/L KCl, 2.5 mmol/L MgCl2, and 0.1% Triton X-100. Reagents were from Promega Ltd. Samples were overlaid with light mineral oil (Sigma) to prevent evaporation. Thermal cycling consisted of 1-minute denaturing at 95°C, 1-minute annealing at 54°C (human c-myb, β2-microglobulin, and 7B6) or 58°C (porcine c-myb and β-actin), and 1-minute extension at 72°C. This was repeated for 40 cycles (human c-myb), 36 cycles (32 for the in vivo experiments) (porcine c-myb), 34 cycles (human c-fos), and 30 cycles (28 for the in vivo experiments) (7B6, β2-microglobulin, and β-actin), with a final extension period of 6 minutes at 72°C. Primer sequences were as follows: human c-myb, 5′ AAT TAA ATA CGG TCC CCT GAA 3′ (forward), 5′ TGC TCC TCC ATC TTT CCA CAG 3′ (reverse), 423-bp predicted product size; porcine c-myb, 5′ CAG CAC CGA TGG CAG AAA GTA 3′ (forward), 5′ GCT GGC TGA GGG ACA TTG ACT 3′ (reverse), 593-bp predicted product size; human c-fos, 5′ AAG GAG AAT CCG AAG GGA AAG GAA TAA GAT GGC T 3′ (forward), 5′ AGA CGA AGG AAG ACG TGT AAG CAG TGC AGCT 3′ (reverse), 612-bp predicted product size; 7B6, 5′ CTA AAA CAG CGG AAG AGG T 3′ (forward), 5′ AGC CGT AGA CGG AAC TTC GC 3′ (reverse), 434-bp predicted product size; and β-actin, 5′ CTC GGT CAG GAT CTT CAT GAG G 3′ (forward), 5′ TTC TAC AAT GAG CTG CGT GTG G 3′ (reverse), 324-bp predicted product size. PCR products were resolved by agarose gel electrophoresis and visualized by ethidium bromide staining under UV transillumination. The fragment amplified using human and porcine c-myb specific primers was cloned into the pCR II vector (Invitrogen). The identity of the c-myb cDNA PCR product was then confirmed by sequencing with a reverse primer, using an Applied Biosystems automated DNA sequencer (model 373A) or manually using forward and reverse primers and a Sequenase sequencing kit (US Biochemical Corp).
c-myb Protein Analysis
c-myb protein expression was analyzed by immunocytochemistry and Western blotting. For immunocytochemistry, VSMCs were cultured onto Lab-Tek tissue culture chamber slides (Life Technologies) before incubation with sense- or AS-ODN–c-myb for various time intervals. Cells were fixed in ice-cold methanol, quenched using bovine serum albumin, and incubated with an antibody to c-myb (Cambridge Research Biochemicals) for 2 hours. After two washes in Tris-buffered saline, biotinylated secondary antiserum was added for 60 minutes. Cells were then incubated with streptavidin complexed with biotinylated alkaline phosphatase (Dako Ltd), followed by a chromogenic substrate to visualize the staining and Mayer’s hematoxylin counterstain. Endogenous alkaline phosphatase was blocked with levamisole. Stained cells were examined using transmission microscopy by an independent observer blinded to the experimental protocol. For Western blotting, human VSMCs were seeded at 5×103 cells/cm2, incubated at 37°C for 24 hours in DMEM and 10% FCS, and then serum-starved for 48 hours. Cells were then incubated with 5 μmol/L AS-ODN–c-myb in DMEM and 10% FCS for 48 hours. The controls were incubated with DMEM and 10% FCS only. Cell lysates, equivalent to 5×105 cells, were loaded onto a 10% SDS gel and transferred to a nitrocellulose membrane. The membrane was blocked for 12 hours with 2% nonfat milk in Tris buffer. The c-myb protein was detected with a sheep polyclonal antibody (Cambridge Research Biochemicals), an anti-sheep secondary antibody, the APAAP technique (Dako Ltd), and new fuchsin substrate system (Dako).
Incubation of AS-ODN–c-myb With VSMCs in Culture
To assess the uptake of ODNs, human saphenous vein SMCs were plated onto Lab-Tek chamber slides and incubated with FAM-labeled ODNs in DMEM containing 10% FCS for various time intervals. Cells were then washed in PBS, fixed in methanol, and observed using fluorescence microscopy. For FACS analysis, saphenous vein SMCs were plated onto T25 flasks, quiesced in 0.5% FCS before incubation with 2.5 μmol/L FAM-ODN–c-myb in DMEM containing 10% FCS for various time intervals. Cells were then washed in PBS and incubated with 0.02% EDTA to allow a single cell suspension. Ten thousand cell events were analyzed using a FACScan flow cytometer and Consort 30 software (Becton Dickinson).
S1 Nuclease Analysis: Determination of Oligomer Stability and Intracellular Duplex Formation
To confirm uptake of intact ODNs into VSMCs and duplex formation with cellular RNA, 5′ end-labeled 18-base c-myb AS- and sense-ODNs (specific activity, 2×106 cpm) were added at 2 μmol/L to 6×105 exponentially growing porcine VSMCs in DMEM with 20% pig serum. After 6 to 12 hours of incubation, the cells were washed three times in PBS, recovered by trypsin treatment, and further washed before lysing in a 0.05% Nonidet P-40 lysis buffer (10 mmol/L Tris [pH 7.5], 10 mmol/L NaCl, 3 mmol/L MgCl2 and 0.05% Nonidet P-40) containing 0.5% SDS and 100 μg/mL proteinase K. After deproteination and ethanol precipitation of the nucleic acids, S1 nuclease digestion was performed at 37°C for 1 hour in 5 mmol/L sodium acetate, 28 mmol/L NaCl, and 4.5 mmol/L ZnSO4 in the presence of 400 U of S1 nuclease (Boehringer-Mannheim). After the addition of “S1 stop buffer” (3.7 mol/L ammonium acetate and 50 mmol/L EDTA [pH 8.0]) containing 100 mg/mL yeast tRNA, phenol/chloroform extraction, and precipitation with ethanol, the samples were resuspended in a formamide loading dye (80% [vol/vol] formamide, 10 mmol/L EDTA, 1 mg/mL bromophenol blue, and 1 mg/mL xylene cyanol FF), heated in a boiling water bath for 5 minutes, immediately placed on ice, and analyzed on a denaturing 12% polyacrylamide/7 mol/L urea–containing gel. Untreated samples and unbound ODNs were run on the same gel to serve as controls. The protected fragments were visualized by autoradiography.
Duplex formation to cellular RNA isolated from serum-stimulated porcine VSMCs was determined using the above method. The VSMCs were serum-stimulated for 16 hours in DMEM with 20% pig serum. Total RNA was isolated using RNAzol B (Biogenesis). Total RNA (200 μg) was concentrated by ethanol precipitation and resuspended in hybridization buffer (80% [vol/vol] deionized formamide, 40 mmol/L PIPES-KOH [pH 6.4], 1 mmol/L EDTA, and 40 mmol/L NaCl). Radioactive ODNs (specific activity, 4×105 cpm) were added to the reaction mixture, heated to 85°C for 10 minutes, and hybridized at 42°C for 24 hours. S1 nuclease analysis was performed as above, but on ice.
The Transport Balloon
The Transport catheter (Fig 1⇓) is a dual-purpose over-the-wire device. It consists of a conventional inner balloon capable of normal therapeutic high-pressure inflation surrounded by an outer balloon with 48 pores, 250 μm in diameter, and supplied by a separate infusion channel. This design permits liquid to be infused into the lumen (if the inner balloon is deflated) or the vessel wall (if it is inflated). It was used in these experiments both to administer the vascular injury, with the inner balloon inflated to 8 atm, and to support the delivery balloon against the arterial wall, with the inner balloon inflated to 1 atm.
Porcine Coronary Angioplasty Model
Thirty-kilogram domestic crossbred pigs (Sus scrofa) were used for this and subsequent experiments. Coronary artery intimal hyperplasia was induced by high-pressure marginally oversized balloon angioplasty of one coronary artery in each pig. Each pig was sedated by an intramuscular injection of azaperone (12 mg/kg, Janssen Animal Health). Anesthesia was induced by intravenous propofol (4 mg/kg, Zeneca Pharmaceuticals) and maintained by inhaled enflurane and oxygen via endotracheal tube. There was continuous ECG monitoring. An incision was made to expose the right carotid artery, and an angioplasty guide catheter was maneuvered under fluoroscopic guidance into either the right or left coronary ostium. An intravenous bolus of 2500 U sodium heparin was given. A 3.0-mm Transport balloon was positioned in a single coronary artery using conventional angioplasty techniques. The site of balloon inflation was selected such that the balloon-to-artery ratio was ≈1.3:1, judged by visual estimation from fluoroscopy. After PTCA±delivery, the catheters were removed, the carotid artery was ligated, and the animal was allowed to recover. Normal feed was given. At a selected interval, the pig was reanesthetized, a median sternotomy was performed, the great vessels were ligated, and a lethal injection of thiopentone was given. The heart was explanted, and the coronary arteries were washed with normal saline. The angioplastied vessel was identified and excised, preserving the adventitia, with the contralateral coronary artery being kept as an untreated control. Both arteries were serially cross-sectioned and suitably preserved for subsequent analysis.
Local Delivery of Fluorescence-Labeled AS-ODN–c-myb to Porcine Coronary Arteries
Initial experiments were conducted ex vivo to optimize the conditions for local drug delivery. First, the inner balloon of a 3.0-mm Transport catheter was inflated to 1-atm pressure within each coronary artery, and one of three 2 mL solutions of FAM-AS-ODN–c-myb, containing 5, 50, and 500 μg FAM-AS-ODN–c-myb, was infused via the delivery channel in saline at 2 atm, at which pressure fluid exited the balloon over ≈120 seconds. There was no prior high-pressure inflation, so that the effect of local delivery alone could be ascertained. Immediately after drug delivery, the coronary arteries were serially sectioned at 2-mm intervals. Alternate sections were immersion-fixed in buffered formalin (10%) for 24 hours, processed, and embedded in paraffin wax, and 4-μm transverse histological sections were cut. Uptake of AS-ODN–c-myb was assessed by fluorescence microscopy (Leica) and photographed. Alternate 2-mm sections were snap-frozen and stored in liquid nitrogen. Frozen sections were cut from these specimens, which were used to assess uptake of AS-ODN by fluorescence microscopy without the potential for its elution from extracellular material during either fixation or processing of tissue to paraffin sections. The optimal concentration of ODN was found to be 500 μg in 2 mL. This concentration and volume of FAM-AS-ODN–c-myb was then given at incremental balloon support pressures of 1, 2, 3, 4, and 5 atm, with corresponding delivery pressures of 2, 3, 4, 5, and 6 atm to five different porcine coronary arteries. Analysis of the sections from these vessels, as just described, showed that 1-atm support pressure and 2-atm delivery pressure gave a density and pattern of deposition not discernibly different from higher pressures. Although there was no evidence of trauma in any of the five pressures, these last minimum practical pressures were adopted for the in vivo studies. In one further artery, high-pressure balloon inflation (8 atm for 30 seconds twice) preceded delivery of FAM-AS-ODN–c-myb to ascertain any difference in distribution of the ODN after injury.
Local Delivery of AS-ODN–c-myb, Sense-ODN–c-myb, and Saline to Porcine Coronary Arteries After PTCA
There were four groups of animals in this experiment. All underwent high-pressure (8 atm), marginally oversized (1.3:1 balloon-to-artery ratio, estimated visually from the fluoroscopic image) PTCA with two 30-second inflations of the inner balloon of a 3.0-mm Transport catheter to a single coronary artery. Local drug delivery followed. Delivery conditions consisted of low support and delivery pressures (1 and 2 atm, respectively). Delivery was of 500 μg AS-ODN–c-myb dissolved in 2 mL saline in the first group, 500 μg sense-ODN–c-myb in 2 mL saline in the second, and 2 mL saline in the third; in the fourth, no delivery was performed. The contralateral coronary artery of each animal in the fourth group made up the control arteries for the whole experiment. Four weeks later, a sternotomy was performed, a lethal injection was given, and the heart was explanted. The coronary arteries were washed, dissected free, and serially sectioned at 5-mm intervals. Each segment was then divided into three. One was preserved by immersion fixation in buffered formalin (10%) for light microscopy; the next two were snap-frozen in liquid nitrogen.
Time Course of c-myb Expression In Vivo
An additional group of pigs underwent PTCA alone, without drug delivery. They were killed immediately or at 1, 6, 12, 18, or 24 hours after the procedure. Untreated and angioplastied coronary arteries were dissected and snap-frozen in liquid nitrogen before analysis of c-myb mRNA expression, as described above.
Tissue Fixation, Histology, and Quantification
The histological tissue blocks from angioplastied and control porcine coronary arteries were immersion-fixed in formalin for 24 hours. Perfusion fixation was not used for three reasons: (1) to prevent artifactual stimulation of c-myb expression, (2) to avoid delay in obtaining the snap-frozen tissue for assay of mRNA, and (3) to prevent washout of fluorescent label, where used. Comparison of the effects of perfusion and immersion fixation on quantitative histology was also performed in two animals of equal weight with no arterial injury. Perfusion fixation of one animal was with 500 mL formalin in saline (4% [vol/vol]) via an aortic root cannula at 80 mm Hg for 10 minutes, followed by 24-hour immersion fixation as previously described. The second was immersion-fixed only. The tissue from all arteries was then processed, embedded in paraffin wax, cut into 4-μm transverse sections, and stained with either hematoxylin and eosin or Alcian blue, Miller’s elastin, and van Gieson’s stain. Quantitative histology was performed by a single investigator (J.G.) blinded to the identity of each section. Equipment used was an Olympus BH-2 light microscope with a Sony video camera XC-711P attachment and output to a computerized image analysis system (Seescan plc) and television monitor. This enabled boundary definition of the intima, media, and dense (adherent) adventitia by hand-held mouse and yielded semiautomated calculated CSA. Radial thicknesses and luminal measurements were not measured because of the distortion present in immersion-fixed vessels. For each artery, only sections with a breach in the IEL, as evidence of injury, were analyzed. The CSAs of intima and media in these sections was recorded. The ratio of intimal-to-medial CSA was calculated, thereby correcting for vessel size.36 The injury score in each section was also recorded as percent breach of the IEL. A direct correlation between FL of the IEL and neointimal area has previously been demonstrated in our model.37 Therefore, correction for the variation in level of trauma was possible by dividing the intimal-to-medial ratio by the % FL-IEL. Histological measurements were independently validated by a second observer (M.G.) blinded to the experimental protocol.
The inhibition of VSMC proliferation was determined for each concentration of sense- and AS-ODN–c-myb by comparison with basal proliferation occurring in the presence of serum alone. Significant differences between sense- and AS-ODN–c-myb treatments were assessed by general factorial ANOVA. The concentrations of ODNs required for 50% inhibition of cell proliferation were assessed by curve estimation (polynomial regression). The Wilcoxon-Mann-Whitney test was used to compare CSAs of the coronary arteries in the four groups. Interobserver variation in the accuracy of histological measurement was sought using a panel of 26 sections, composed of 13 PTCA and 13 control sections. No significant difference was found between the areas of intima and media as measured by the two observers (Wilcoxon signed-rank test).
Expression of c-myb mRNA in Porcine Coronary Arteries After Angioplasty
RT-PCR revealed very low levels of c-myb expression in control and angioplastied arteries harvested immediately after the procedure by comparison with β-actin–loading control. c-myb expression then increased, reaching a peak ≈18 hours after PTCA (Fig 2⇓).
Cellular Uptake of AS-ODN–c-myb In Vitro
After incubation for 24 hours with FAM-AS-ODN–c-myb, virtually all VSMCs demonstrated some degree of uptake by fluorescence microscopy. This was confirmed by FACs analysis. AS-ODN–c-myb was present in the nucleus within 10 minutes and remained in the nucleus and cytoplasm for up to 48 hours.
Effect of AS-ODN–c-myb on VSMC Proliferation
As assessed by incorporation of [3H]thymidine, AS-ODN–c-myb significantly inhibited the proliferation of porcine aortic SMCs in a dose-dependent manner, whereas minimal inhibition, not rising to significance, was seen after incubation with sense-ODN–c-myb (Fig 3⇓). A 50% decrease in VSMC proliferation was seen after incubation with AS-ODN–c-myb (0.13 μmol/L), with no effect on cell viability, as assessed by trypan blue exclusion. VSMCs were exposed to AS-ODN–c-myb (5 μmol/L) for 6, 24, and 48 hours, then washed, and refed with medium containing serum. The cells were harvested 48 hours later. Steady state growth characteristics, as determined by [3H]thymidine incorporation, then resumed at the previous rate.
Effect of a Panel of ODNs on VSMC Proliferation
The effect of sense-ODN–, murine AS-ODN–, 2-bp mismatch ODN–, scrambled sequence ODN–, and phosphorothioated AS-ODN–c-myb, as well as unrelated sense- and AS-ODNs (to ICE) on bovine VSMC proliferation, was assessed by [3H]thymidine incorporation. Bovine VSMCs were used for reasons of availability, with the bovine sequence for c-myb in the region of the ODN being 100 percent homologous to the human sequence. A dose-dependent reduction in cell proliferation was seen with all these ODNs apart from sense-ODN–c-myb, in which no such relationship existed. The concentration at which half-maximal inhibition occurred for each ODN was <0.7 μmol/L for all the ODNs apart from phosphorothioated AS-ODN–c-myb, for which it was 16 μmol/L. The concentration at which 90% maximal inhibition occurred was comparable (range, 2.5 to 15 μmol/L), apart for phosphorothioated AS-ODN–c-myb, for which it was 28 μmol/L; for AS-ODN–c-myb, it was 12 μmol/L.
Effect of AS-ODN–c-myb on Steady State c-myb and c-fos mRNA Expression
c-myb mRNA levels as assessed by RT-PCR were similar in VSMCs treated with sense- or AS-ODN–c-myb or FCS 10% alone for time intervals up to 24 hours. At 48 hours, a decrease in c-myb mRNA was observed in cells treated with AS-ODN–c-myb compared with cells treated with sense-ODN–c-myb or FCS alone (Fig 4A⇓). To investigate if other genes were affected by AS-ODN–c-myb, we performed RT-PCR for c-fos mRNA. No significant changes were observed on c-fos mRNA expression (Fig 4B⇓).
Effect of AS-ODN–c-myb on Steady State c-myb Protein Production
Control saphenous vein VSMCs and cells treated with sense-ODN–c-myb showed dense perinuclear staining for c-myb protein by immunocytochemistry. Weak perinuclear staining was seen in cells incubated with AS-ODN–c-myb for 48 hours. Western blotting confirmed suppression of c-myb protein in the AS-ODN–c-myb–treated group.
S1 Nuclease Analysis: Determination of Specificity of Binding of AS-ODN–c-myb to Target Nucleic Acid and Recovery of Intact ODN
Duplexes resistant to S1 nuclease activity were formed when sense- and AS-ODN–c-myb were added to cultured porcine aortic VSMCs for 6 hours (Fig 5A⇓). Duplex formation with cellular RNA was confirmed by adding sense- and AS-ODNs separately to total RNA isolated from porcine VSMCs stimulated for 16 hours with porcine serum. Hybridization of both sense- and AS-ODN to RNA was observed (Fig 5B⇓ and 5C⇓). The size of the duplex observed after polyacrylamide gel electrophoresis was equivalent to an 18-base ODN, as determined by the separation rate of 32P-labeled ODNs of differing sizes. This indicated that recovery of intact ODNs from intact VSMCs had been achieved.
Effect of Oversized-Balloon Inflation on Porcine Coronary Arteries
Histological examination of the dilated segments excised immediately after oversized-balloon angioplasty (n=4 coronary arteries) revealed a pattern of disruption varying from none to extensive breach of intima, IEL, and media (Fig 6B⇓), compared with uninjured artery (Fig 6A⇓). Twenty animals underwent PTCA only and were killed at 4 weeks, and 15 were eligible for analysis. PTCA in these animals had been performed in five RCA, six LAD, and four Cx arteries. Deployment in each artery was primarily determined by accessibility, visibility, and sizing. There was a variability seen in the quantity of vessel wall thickening 4 weeks after angioplasty, corresponding to the extent of injury, with little of either occurring in larger vessels. For this and all subsequent treatment groups, only vessels with evidence of injury (breached IEL) were eligible for analysis. There was a positive correlation between trauma score (% FL-IEL) and intimal-to-medial thickness ratio (Spearman’s rank correlation coefficient, ρ=+0.67, P<.01). The range of disruption of the IEL was 1% to 50% in the dilated vessels. Mean±SEM maximum CSA ratio was 0.79±0.27 in PTCA vessels and approximately zero in control vessels (no neointima). Mean±SEM % FL-IEL was 22±3%. Mean±SEM intimal-to-medial CSA ratio divided by FL was 0.030±.005 in the PTCA group. This parameter was not applicable in the control vessels because of the absence of trauma. The effect of pressure versus immersion fixation (in paired arterial segments at equivalent levels in the LAD and RCA of two pigs, one processed each way, yielding n=15 pairs of arterial sections for analysis) was as follows: cross-sectional profile, more circular in 15 of 15; total vessel CSA, 3.90±0.67 versus 3.67±0.45 mm2, 6% larger (P=NS); media CSA, 1.25±0.29 versus 1.65±0.20 mm2, 24% smaller (P=NS); and lumen CSA, 2.65±0.41 versus 2.02±0.29 mm2, 31% larger (P=NS).
Delivery of AS-ODN–c-myb to the Vessel Wall
No complications occurred during or after AS-ODN–c-myb delivery. Balloon inflation (high-pressure dilatation or low-pressure delivery) was associated with ECG ST-segment and T-wave changes. When ECG changes persisted, vasospasm was noted angiographically and resolved with intracoronary glyceryl trinitrate. Intramural deposition of FAM-AS-ODN–c-myb was seen, with intense deposition at 500 μg/2 mL, this dose being used in subsequent deliveries. FAM-AS-ODN–c-myb deposition was greatest in the damaged intima, inner media, and adventitia (Fig 7A⇓ and 7B⇓). Where prior high-pressure inflation was performed, there was a similar pattern of uptake, with intense fluorescence in the remnants of the damaged endothelium and around the medial breach. Within the adventitia, uptake was concentrated in the vasa vasorum (Fig 7C⇓), confirmed with von Willebrand factor staining (Fig 7D⇓). There was sparing of the outer media. Frozen sections showed similar results, indicating that fixation and tissue processing did not result in loss of FAM-AS-ODN–c-myb.
Effect of Sense- and AS-ODN–c-myb and Saline on Neointima 4 Weeks After PTCA
The following animals were eligible for analysis: 14 animals in the PTCA plus AS-ODN–c-myb group, yielding four RCA, nine LAD, and 1 Cx vessel (mean±SEM % FL-IEL, 26±6%); 13 animals in the PTCA plus sense-ODN–c-myb group, yielding four RCA, six LAD, and three Cx vessels (% FL-IEL, 32±2%); and 13 animals in the PTCA plus saline group, yielding four RCA, seven LAD, and two Cx vessels (% FL-IEL, 28±7%). The mean±SEM maximum intimal-to-medial area ratio was 0.30±0.14 in the PTCA plus AS-ODN–c-myb group, 1.90±0.40 in the PTCA plus sense-ODN–c-myb group, and 1.87±0.61 in the PTCA plus saline group. Mean±SEM maximum intimal-to-medial FL was 0.011±0.004 in the PTCA plus AS-ODN–c-myb group, 0.061±0.014 in the PTCA plus sense-ODN–c-myb group, and 0.053±0.011 in the PTCA plus saline group. Therefore, there was a reduction of 79% with AS-ODN–c-myb compared with saline (P<.05), 82% compared with sense-ODN–c-myb, and 63% compared with PTCA (P<.10) (Fig 8⇓).
The present study has shown that c-myb is expressed in proliferating human and porcine VSMCs. Unmodified AS-ODN–c-myb is incorporated rapidly into VSMCs in vitro, inhibits their proliferation, and decreases c-myb mRNA and protein without affecting c-fos expression, effects not seen with sense-ODN–c-myb. c-myb is expressed 18 hours after porcine PTCA, intracellular AS-mRNA duplexes form, and intact full-length AS-ODN–c-myb may be retrieved from cells after 12 hours of incubation. Compared with angioplasty of vessels without local delivery of AS-ODN–c-myb, the Transport catheter can deliver AS-ODN–c-myb deeply into the arterial wall, independent of vessel trauma, and reduce neointimal hyperplasia at 4 weeks. The action of AS-ODN–c-myb is at least partly nonspecific, partly non–sequence specific, and not solely related to the presence of a GGGG sequence or to a phosphorothioate moiety. Unprotected ODNs are able to retain biological activity, and phosphorothioation may reduce their efficacy.
Three main new points arise from the present study. First, a reduction in neointima in the porcine coronary artery with AS-ODN–c-myb has been shown; previous studies have used small mammals and peripheral arteries, usually without deep injury. Second, transluminal local delivery with a porous balloon was used to make a single application of the drug, immediately after PTCA, to the damaged segment of artery; previous studies have used direct surgical application, negating the advantages of the percutaneous approach. Third, unmodified ODNs were shown to be efficacious in vitro and in vivo, more so, indeed, than modified ODNs.
The mechanism of action of AS-ODNs and the conclusions drawn from experiments using them are under debate.31 There are a number of potential mechanisms by which AS-ODNs may work at the cellular level. The initial theory was hybridization with a specific mRNA of complementary sequence, interfering with translation or promoting degradation by RNase H. Early results, eg, those of Simons and Rosenberg,21 29 were consistent with this theory. Its validity has, however, been undermined by a number of observations. First, total hybridization is not necessary to induce destruction of mRNA by ODNs.38 Second, Villa et al30 noted inconsistency in the effect of antisense and a positive sense effect. Third, the same group, and others, have raised the possibility that a GGGG sequence is responsible, by an aptameric mechanism, for the nonspecific effect of AS-ODN–c-myb.28 30 39 Fourth, ODNs with mismatched sequences, including, in the case of Wang et al,40 a cytidine homopolymer, have been shown to have biological effects. Fifth, the applicability of observations made in certain experimental models, notably, the rat carotid model, has been questioned. Whereas specific hybridization may be important, additional less-specific mechanisms have been proposed.
It was essential, then, that we should use extensive controls in vitro and carefully selected controls in vivo. We demonstrated dose-dependent inhibition of proliferation for a wide variety of ODNs in vitro. In descending order of potency, they were as follows: a 2-bp mismatch to AS-ODN–c-myb, AS-ODN-ICE (13 mismatches), murine AS-ODN–c-myb (2 mismatches), ICE sense-ODN (17 mismatches), human AS-ODN–c-myb (no mismatches), scrambled AS-ODN–c-myb (13 mismatches), and phosphorothioated AS-ODN–c-myb (no mismatches). There was no specific sequence (of length ≥3 bp, and in particular not GGGG) common to the ODNs studied. Nor was there a relationship between the number of mismatches and potency. Our results show that a wide range of ODNs can act as inhibitors of VSMC proliferation. This is in general agreement with other studies,22 30 although the early Simons and Rosenberg study21 used a 2-bp mismatch, which failed to show any effect. Indeed, the nonspecific mechanism is probably independent of base sequence.40
The groups used in the in vivo experiments were PTCA alone, PTCA plus saline, PTCA plus AS-ODN–c-myb, and PTCA plus sense-ODN–c-myb. The reduction in neointima with PTCA plus AS-ODN–c-myb compared with both PTCA plus saline and PTCA plus sense-ODN–c-myb suggested some specificity of antisense. Compared with PTCA alone, PTCA plus sense-ODN–c-myb, like PTCA plus saline, showed a trend toward exacerbating neointima formation. This is likely to be either a statistical quirk or the result of the physical effect of 2-mL fluid entering the vessel wall. The latter seems more plausible.
A GGGG sequence has been propounded as being responsible for the inhibitory effects of ODNs in VSMCs.28 30 39 Our data, although not refuting the possibility that GGGG may be responsible for some of the nonspecific effect, clearly demonstrate that it cannot be responsible for all of it, since ODNs lacking this motif also cause inhibition of VSMC proliferation in our hands. There is support for this from other studies: even the murine interleukin-1β AS-ODN (lacking GGGG), for example, used by Villa et al,30 inhibited cell proliferation. A further possibility is that GGGG plus the flanking sequences are responsible,41 but this also cannot be the only nonspecific mechanism, for the same reason.
A weak antiproliferative effect of sense-ODN observed in our in vitro study, especially compared with the stronger effect of completely unrelated ODNs, remains to be explained. Sense-ODN hybrids may form with unidentified genes important in the control of cell proliferation. This effect may be potentiated if small fragments of the original ODNs used in these experiments remain active. Such fragments might have a wide range of potential target nucleic acid sequences. This hypothesis is supported by the observation that RNase H–induced destruction of mRNA may be caused by ODNs composed of only 4 bases.42 Not only can ODNs bind to nucleic acid sequences and promote their destruction, but it is feasible that ODNs, and especially sense-ODNs, may bind to endogenous antisense strands of template DNA exposed during transcription and cause inhibition of RNA synthesis. Our finding of FAM-labeled ODN within the nucleus is consistent with this explanation. Also, naturally occurring c-myb antisense-RNA may be present within cells as a feedback loop.43 Sense ODN may bind to this, causing an antiproliferative effect. Our S1 nuclease protection data suggest that both sense- and AS-ODNs bind to target nucleic acid, lending some support to the latter theory. Finally, and relevant to the in vivo work, there is a potential for the unmethylated CpG motifs contained within sense-ODN–c-myb to upregulate the immune response and have an adverse effect on cell proliferation.44
From our data, it is still not possible to define a single mechanism of action for the inhibition of neointima formation with AS-ODN–c-myb. Some degree of specificity is supported by the observations that (1) the inhibition of VSMC proliferation by AS-ODN–c-myb in our experiments was accompanied by corresponding decreases in c-myb mRNA and protein, (2) these effects were reversible and not present to the same degree with sense, and (3) no effect of AS-ODN–c-myb was seen on c-fos expression. These data do not support the hypothesis that AS-ODNs act by a GGGG mechanism only. The in vivo results, comparing the effect of AS-ODN–c-myb with the lack of it with sense after PTCA, confirm the in vitro findings.
In all the studies reported here, unmodified ODNs were used. This is in contrast with the standard practice of using a stabilizing modification such as phosphorothioation. These are normally used because unmodified ODNs have been assumed to have a short half-life in biological systems. Survival up to 24 hours by phosphorothioation, however, has been documented.45 Nevertheless, unmodified ODNs appear to be taken up, to be retained intact, and to have the desired effect in our in vitro and in vivo studies. Our data also suggest that phosphorothioated ODNs are less potent than unmodified ODNs, although equally efficacious in higher dose. This reduction in potency may be due to the decreased hybridization to target sequences46 47 or inhibition of RNase H,42 thereby reducing the efficacy of the ODN. It is conceivable, therefore, that under the conditions of these experiments, enhanced cellular penetration is more important for AS-ODN–c-myb than protection against intracellular degradation. Very recently, Wang et al40 postulated that it was the phosphorothioate moiety itself that might account for some of the nonspecific effects of ODNs, perhaps by binding to intracellular proteins and growth factors. Another relevant finding, of Guvakova et al,48 is that phosphorothioate ODNs interact with fibroblast growth factor receptors and other heparin-binding growth factors, thereby potentially blocking the effect of fibroblast growth factor and other growth factors, perhaps explaining some of the nonspecific effects of AS-ODNs. Our findings, however, exclude phosphorothioation as being the only nonspecific mechanism on two grounds: first, all the ODNs we used were unprotected and yet retained biological activity; second, the phosphorothioated ODN exhibited less potency, not more, than its unprotected equivalent. Recently, ODNs with sulfur-modified linkages (S-chimeric ODNs) have been shown to be stable in the presence of human VSMCs for 36 hours and yet retain biological activity.49 This modification may prove to be superior to phosphorothioation and may avoid the nonspecific inhibition of VSMC proliferation seen in some studies.30
The porcine coronary artery model of restenosis used in the present study offers advantages over other experimental models. The cardiovascular system of the pig bears some anatomic and physiological similarity to that of the human,50 and the “restenotic” lesion induced appears similar to that seen in human restenosis.51 Furthermore, the stimulus of balloon inflation used in our model is more analogous to clinical practice than seen in other models; the balloon-to-artery ratio was ≈1.3:1, whereas, in other models, this can be as high as 1.5:1. Also coronary, rather than peripheral, arteries were used. There are important differences between them: eg, the aorta and carotid have a higher elastin content than the coronary artery; they are larger structures, where restenosis gives rise to a less functionally significant stenosis and less hemodynamic disturbance; and their blood flow is predominantly systolic rather than diastolic. The disadvantages of the porcine model as used in this study are as follows: First, the lesion produced is de novo, rather than being truly restenotic. Second, the use of juvenile animals, chosen for practical and economic reasons, may expose proliferating cells to a different growth factor milieu than in an adult. Third, atheroma is absent. Fourth, marginal oversizing resulted in a highly variable amount of neointima, probably explaining the marked average reduction in neointima accompanied by a relatively large standard error. Fifth, immersion fixation, used for the reasons outlined above, may, as we have shown, subtly alter arterial dimensions. This is unimportant, however, for comparison of results between our groups, as each group was treated in exactly the same way. Overall, we believe that the model described here is superior to a number of small mammal and peripheral arterial models.
AS-ODNs have been delivered effectively by the adventitial application of pluronic gel in the experimental setting.24 25 29 This approach is, however, impractical for the prevention of restenosis after PTCA. Percutaneous transluminal local drug delivery retains a number of advantages. Restenosis is a localized phenomenon, and potent antiproliferative therapy may be directed to it. The Transport offers both high-pressure therapeutic balloon inflation and low-pressure local delivery, thereby avoiding catheter exchange and ensuring precise deposition of drug. Its ease of use, low profile, and lack of jetting make it clinically viable. However, the kinetics of antisense delivery with this device are not yet known, and substantial washout may occur.
The most remarkable observation made is that a single application of AS-ODN–c-myb at the time of injury is associated with a reduction of neointima at 28 days. This reflects the importance of the activation of early genes, such as c-myb, and the entry of VSMCs into the cell cycle. We have proved that c-myb activation occurs early after PTCA, with maximal expression at ≈18 hours, and that AS-ODN–c-myb is retained intact in VSMCs for at least 12 hours. Therefore, it does not seem unreasonable to assume that sufficient ODN remained in the vessel wall throughout this important early period. This is a peculiarly favorable set of circumstances for the use of local drug delivery, with its advantages being to deliver high concentrations of therapeutic agent immediately after injury and precisely to the region of tissue damage. This has clear therapeutic advantages over systemic or continuous therapy.
The present study represents a comprehensive examination of AS-ODN–c-myb and related ODNs upon VSMC proliferation and neointima formation in a porcine model of restenosis after PTCA. AS-ODN–c-myb inhibits VSMC proliferation and, delivered locally at the time of PTCA via the Transport catheter, the formation of neointima. Control experiments suggest that this is partly a specific effect from the AS-ODN and partly non–sequence-specific effects, unrelated to a GGGG sequence or to phosphorothioation. Sense-ODN is without therapeutic effect. Local delivery seems suitable for such a strategy, but caution should be exercised with the delivery of relatively large volumes of even innocuous compounds such as physiological saline.
Selected Abbreviations and Acronyms
|% FL-IEL||=||percent FL of the IEL|
|FACS||=||fluorescence automated cell sorting|
|FAM||=||fluorescein addition monomer|
|ICE||=||interleukin-1β converting enzyme|
|IEL||=||internal elastic lamina|
|LAD||=||left anterior descending coronary artery|
|PCR||=||polymerase chain reaction|
|PTCA||=||percutaneous transluminal coronary angioplasty|
|RCA||=||right coronary artery|
|SMC||=||smooth muscle cell|
This study was supported by grants from the British Heart Foundation and the Northern General Hospital Trust Research Fund, Sheffield.
- Received July 16, 1996.
- Accepted January 13, 1997.
- © 1997 American Heart Association, Inc.
Popma JJ, Califf RM, Topol EJ. Clinical trials of restenosis after coronary angioplasty. Circulation. 1991;84:1426-1436.
Macleod DC, Strauss BH, De Jong M, Escaned J, Umans VA, van Suylen RJ, Verkerk A, de Feyter PJ, Serruys PW. Proliferation and extracellular matrix synthesis of smooth muscle cells cultured from human coronary atherosclerotic and restenotic lesions. J Am Coll Cardiol. 1994;23:59-65.
Glagov S. Intimal hyperplasia, vascular modeling, and the restenosis problem. Circulation. 1994;89:2888-2891.
Mintz G, Pichard A, Kent K, Satler L, Popma J, Wong S, Painter J, DeForty D, Leon M. Endovascular stents reduce restenosis by eliminating geometric arterial remodelling: a serial intravascular ultrasound study. J Am Coll Cardiol. 1995;(Feb suppl):36A. Abstract.
Serruys PW, de Jaegere P, Kiemeneij F, Macaya C, Rutsch W, Heyndrickx G, Emanuelsson H, Marco J, Legrand V, Materne P, Belardi J, Sigwart U, Colombo A, Goy J, van den Heuvel P, Delcan J, Morel M-A. A comparison of balloon-expandable stent implantation with balloon angioplasty in patients with coronary artery disease. N Engl J Med. 1994;331:489-495.
Fischman D, Leon M, Baim D, Schatz R, Savage M, Penn I, Detre K, Veltri L, Ricci D, Nobuyoshi M, Cleman M, Heuser R, Almond D, Teirstein P, Fish D, Colombo A, Brinker J, Moses J, Shaknovich A, Hirshfeld J, Bailey S, Ellis S, Rake R, Goldberg S. A randomised comparison of coronary stent placement and balloon angioplasty in the treatment of coronary artery disease. N Engl J Med. 1994;331:496-501.
Camenzind E, Kint P-P, DiMario C, Ligthart J, van der Giessen W, Boersma E, Serruys PW. Intracoronary heparin delivery in humans. Circulation. 1995;92:2463-2472.
Topol EJ, Califf RM, Weisman HF, Ellis SG, Teheng JE, Worley S, Ivanhoe R, George BS, Fintel D, Weston M, Sigmon K, Anderson KM, Lee KL, Willerson JT. Randomised trial of coronary intervention with antibody against platelet IIb/IIIa integrin for reduction of clinical restenosis: results at six months. Lancet. 1994;343:881-886.
Emanuelson H, Beatt KJ, Bagger J-P, Balcon R, Heikkila J, Piessens J, Schaeffer M, Suryapranata H, Foegh M. Long-term effects of angiopeptin treatment in coronary angioplasty. Circulation. 1995;91:1689-1696.
Bauters C, de Groote P, Adamantidis M, Delcayre C, Hamon M, Lablanche J-M, Bertrand ME, Dupuis B, Swynghedauw B. Proto-oncogene expression in rabbit aorta after wall injury: first marker of the cellular process leading to restenosis after angioplasty? Eur Heart J. 1992;13:556-559.
Brown KE, Kindy MS, Sonenshein GE. Expression of the c-myb proto-oncogene in bovine vascular smooth muscle cells. J Biol Chem. 1992;267:4625-4630.
Thiele CJ, Cohen PS, Israel MA. Regulation of c-myb expression in human neuroblastoma cells during retinoic acid-induced differentiation. Mol Cell Biol. 1988;8:1677-1683.
Majello B, Kenyon LC, Dalla-Favera R. Human c-myb proto-oncogene: nucleotide sequence of cDNA and organization of the genomic locus. Proc Natl Acad Sci U S A. 1986;83:9636-9640.
Reilly CF, Kindy MS, Brown KE, Rosenberg RD, Sonenshein GE. Heparin prevents vascular smooth muscle cell progression through the G1 phase of the cell cycle. J Biol Chem. 1989;264:6990-6995.
Morishita R, Gibbons GH, Ellison KE, Nakajima M, Zhang L, Kaneda Y, Ogihara T, Dzau VJ. Single intramural delivery of antisense cdc2 kinase and proliferating cell nuclear antigen ODNs results in chronic inhibition of neointimal hyperplasia. Proc Natl Acad Sci U S A. 1993;90:8474-8478.
Speir E, Epstein SE. Inhibition of smooth muscle cell proliferation by an antisense oligodeoxynucleotide targeting the messenger RNA encoding proliferating cell nuclear antigen. Circulation. 1992;86:538-547.
Simons M, Rosenberg RD. Antisense nonmuscle myosin heavy chain and c-myb oligonucleotides suppress smooth muscle cell proliferation in vitro. Circ Res. 1992;70:835-843.
Biro S, Fu YM, Yu ZX, Epstein SE. Inhibitory effects of antisense oligodeoxynucleotides targeting c-myc mRNA on smooth muscle cell proliferation and migration. Proc Natl Acad Sci U S A. 1993;90:654-658.
Shi Y, Hutchinson HG, Hall DJ, Zalewski A. Downregulation of c-myc expression by antisense oligonucleotides inhibits proliferation of human smooth muscle cells. Circulation. 1993;88:1190-1195.
Bennett MR, Anglin S, McEwan JR, Jagoe R, Newby AC, Evan GI. Inhibition of vascular smooth muscle cell proliferation in vitro and in vivo by c-myc antisense oligodeoxynucleotides. J Clin Invest. 1994;93:820-828.
Edelman ER, Simons M, Sirois MG, Rosenberg RD. c-myc in vasculoproliferative disease. Circ Res. 1995;76:176-182.
Shi Y, Fard A, Galeo A, Hutchinson HG, Vermani P, Dodge GR, Hall DJ, Shaheen F, Zalewski A. Transcatheter delivery of c-myc antisense oligomers reduces neotimal formation in a porcine model of coronary artery balloon injury. Circulation. 1994;90:944-951.
Morishita R, Gibbons GH, Ellison KE, Nakajima M, von der Leyen H, Zhang L, Kaneda Y, Ogihara T, Dzau VJ. Intimal hyperplasia after vascular injury is inhibited by antisense cdc 2 kinase ODNs. J Clin Invest. 1994;93:1458-1464.
Riessen R, Hogan J, Pastore C, Karsch K, Isner J. Local delivery of c-myb antisense ODNs for the prevention of restenosis in atherosclerotic rabbits. Eur Heart J. 1994;15(suppl):380. Abstract.
Villa AE, Guzman LA, Poptic EJ, Labhasetwar V, D’Souza S, Farrell CL, Plow EF, Levy RJ, DiCorleto PE, Topol EJ. Effects of antisense c-myb oligonucleotides on vascular smooth muscle cell proliferation and response to vessel wall injury. Circ Res. 1995;76:505-513.
Bennett MR, Schwartz SM. Antisense therapy for angioplasty restenosis: some critical considerations. Circulation. 1995;92:1981-1993.
Francis SE, Duff GW. p7B6, a ubiquitous RNA with homology to the ribosomal RNA L41. Nucleic Acids Res. 1993;21:2944-2945.
Gunn J, Holt C, Francis S, Shepherd L, Grohmann M, Crossman D, Cumberland D. Restenosis after baloon angioplasty: intimal hyperplasia or vessel remodeling? Heart. 1996;75:P27. Abstract.
Woolf TM, Melton DA, Jennings CJB. Specificity of antisense oligonucleotides in vitro. Proc Natl Acad Sci U S A. 1992;89:7305-7309.
Burgess TL, Fisher EF, Ross SL, Ready JV, Qian Y-X, Bayewitch LA, Cohen AM, Herrera CJ, Hu SS-F, Kramer TB, Lott FD, Martin FH, Pierce GF, Simonet L, Farrell CL. The antiproliferative activity of c-myb and c-myc antisense oligonucleotides in smooth muscle cells is caused by a non-antisense mechanism. Proc Natl Acad Sci U S A. 1995;92:4051-4055.
Maltese J, Sharma H, Vassilev L, Narayanan R. Sequence context of antisense RelA/NF-κB phosphorothioates determines specificity. Nucleic Acids Res. 1995;23:1146-1151.
Donis-Keller H. Site-specific cleavage of RNA. Nucleic Acids Res. 1979;7:179-192.
Mizuno T, Chou M-Y, Inouye M. A unique mechanism regulating gene expression: translational inhibition by a complementary RNA transcript (micRNA). Proc Natl Acad Sci U S A. 1984;81:1966-1970.
Wickstrom E. Antisense DNA therapeutics: neutral analogs and their stereochemistry. In: Erickson R, Izant J, eds. Gene Regulation: Biology of Antisense RNA and DNA. New York, NY: Raven Press Publishers; 1992:261-272.
Milligan JF, Matteucci MD, Martin JC. Current concepts in antisense drug design. J Modern Chem. 1993;36:1924-1937.
Hoke GD, Draper K, Freier SM, Gonzalez C, Driver VB, Zounes MC, Ecker DJ. Effects of phosphorothioate capping on antisense oligonucleotide stability, hybridization and antiviral efficacy versus herpes simplex virus infection. Nucleic Acids Res. 1991;19:5743-5748.
Guvakova MA, Yakubov LA, Vlodavsky I, Tonkinson JL, Stein CA. Phosphorothioate oligodeoxynucleotides bind to basic fibroblast growth factor, inhibit its binding to cell surface receptors, and remove it from low affinity binding sites on extracellular matrix. J Biol Chem. 1995;270:2620-2627.
Pickering JG, Isner JM, Ford CM, Weir L, Lazarovits A, Rocnik E, Chow L. Processing of chimeric antisense oligonucleotides by human vascular smooth muscle cells and human atherosclerotic plaque. Circulation. 1996;93:772-870.
Schwartz RS, Murphy JG, Edwards WD, Camrud AR, Vliestra RE, Holmes DR. Restenosis after balloon angioplasty: a practical proliferative model in porcine coronary arteries. Circulation. 1990;82:2190-2200.