Despite the widespread use of intracoronary stents, in-stent restenosis remains a major clinical problem. Neointima formation with vascular smooth muscle cell (VSMC) hyperplasia is believed to play a critical role in in-stent restenosis.1 Drug-eluting stents (DESs) have been shown to be effective to prevent in-stent restenosis. DESs with sirolimus or paclitaxel have been used widely to prevent in-stent restenosis in humans.2-5 Sirolimus halts cell-cycle progression,6 and sirolimus-coated DESs can prevent in-stent restenosis by inducing complete inhibition of VSMC hyperplasia. However, complications such as subacute thrombosis or late thrombosis have been reported in patients implanted with a sirolimus-coated DES.7,8 Sirolimus prevents reendothelialization of the inner side of the metal stent, which may cause late thrombosis. These complications have led to the development of second- generation DESs that do not induce late thrombosis.
Platelet-derived growth factor (PDGF), a potent stimulator of VSMC proliferation, is a dimer composed of a disulfide-linked polypeptides A-chain and B-chain.9,10 There are 3 isoforms of PDGF (AA, AB, and BB). Nilsson et al11 showed that normal growth-arrested VSMCs do not express PDGF messenger ribonucleic acid (mRNA), whereas cultured VSMCs and VSMCs in atherosclerotic plaques express PDGF A-chain mRNA and secrete PDGF-AA protein. We have reported that spontaneously hypertensive rat-derived VSMCs, which show exaggerated proliferation and the synthetic phenotype, express increased levels of PDGF A-chain mRNA.
Antisense oligodeoxynucleotides (ODNs) have been used clinically to suppress gene expression. Antisense ODNs can be used to inhibit DNA transcription and the production of mature mRNA and to inhibit peptide synthesis. Antisense therapy is being developed for the treatment of in-stent restenosis. We have shown that an antisense ODN to the PDGF A-chain inhibits the exaggerated proliferation of VSMCs from spontaneously hypertensive rats.12-14 In a rat model of neointima formation in the carotid artery after balloon injury, we showed a 60% decrease in neointima in response to antisense therapy.
To aid in the development of safer DESs that do not inhibit reendothelialization, we examined the effects of stent-based delivery of an antisense ODN targeted to the PDGF A-chain on in-stent restenosis and reendothelialization of the coronary artery in pigs.
Synthetic Antisense Oligodeoxynucleotides to the Platelet-Derived Growth Factor (A-chain)
We used a 15-mer antisense ODN to the PDGF A-chain (5′-AGGTCCTCATCGCGT-3′) complementary to the region containing the initiation codon of human and rat PDGF A-chain complementary deoxyribonucleic acid (cDNA), as reported previously.12 A nonsense control ODN (5′-TGCCGT-CAGCTGCTA-3′) contained an identical proportion of bases but in random order. ODNs were synthesized with a DNA synthesizer (model 394; Applied Biosystems, Foster City, CA, USA) and purified on an OPC column (Applied Biosystems). ODNs were modified with a phosphorothioate linkage by oxidizing phosphate linkages with 3H-1,2-benzodithiol-3-one-1,1-dioxide instead of the standard iodine reagent.15
Animal Preparation and Stent Implantation
Animal care and handling were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee of Nihon University. Eighteen domestic male pigs weighing 22-28 kg each were used in the present study. Bare metal stents (n = 6), nonsense ODN-coated stents (n = 3), and antisense ODN-coated stents (n = 9) were implanted into the left anterior descending artery. Hydrogel-coated matrical stents (diameter 3.5 mm, length 20 mm, Interventional Radiology Co., Ltd., Tokyo, Japan) were dip-coated in 20 μg/mL antisense or nonsense ODN solution containing 100 μ/mL polyethylenimine reagent (MBI Fermentas) for 10 min and were air-dried for 10 min. Control stents were treated identically but were not coated with ODN.16,17
Aspirin (325 mg) was administered 1 day before stent implantation. After overnight fasting, intramuscular injection of 25 mg/kg sodium pentobarbital was administered and anesthesia was maintained with a continuous intravenous infusion of 1mg/kg per hour ketamine chloride. A 5000-IU bolus of heparin was administered intravenously. After endotracheal intubation, controlled mechanical ventilation was established at a tidal volume of 10-15 mL/kg with a volume-cycle ventilator (Servo 900-E; Siemens-Elema Inc, Stockholm, Sweden).
Continuous hemodynamic and surface electrocardiography monitoring were maintained throughout the procedure. The right carotid artery was inserted with a 6F sheath. After baseline coronary angiography, mechanical intravascular ultrasound (IVUS) (40 MHz, Atlantis™; Boston Scientific Corp., Natick, MA, USA) of the left anterior descending artery was performed. IVUS catheters were automatically pulled back at 0.5 mm per second. IVUS images were recorded continuously on Super VHS videotape for analysis. Based on the IVUS images, a segment of diameter of 3.0 mm in diameter was selected for stent implantation. Stents of 3.5 mm × 20 mm were used with a 12-atm inflation pressure to reach a 1.3:1 stent-to-artery ratio. Immediately after implantation, IVUS was performed to confirm the stent-to-artery ratio. If the ratio was not adequate, the stent balloon was reinflated with higher pressure in the stented segment. Follow-up coronary angiography and IVUS were performed 28 days after implantation, and the stented coronary arteries were harvested for angioscopic evaluation and pathologic analysis.
To assess the distribution of antisense ODN targeted to the PDGF A-chain, stents coated with fluorescein-isothiocyanate (FITC)-labeled antisense ODN were implanted. Twenty-four hours after implantation, the heart was removed, and the stented coronary artery segment was harvested and fixed in 4% paraformaldehyde. Sections were examined by fluorescence microscopy. In addition, 3.5 mm × 23 mm sirolimus-coated stents (Cypher™; Cordis Corp., Miami Lakes, FL, USA) were implanted by the same method. After 28 days, the stented coronary artery segment was harvested for angioscopic evaluation and pathologic analysis.
Intravascular Ultrasound Measurements
Follow-up IVUS images were recorded on Super VHS tape and analyzed with a computer-based contour detection program for 3-dimensional reconstruction and volumetric measurement (NetraIVUS™ software package for Windows NT®, ScImage Corp, Los Altos, CA, USA). Cross grids recorded on the IVUS images were used for calibration. The interface between the neointima and lumen and the outer border of the external elastic membrane were traced manually. On the basis of the manual traces, the computer software determined measurements for lumen volume (LV) and vessel volume (VV). Volumetric in-stent restenosis (%AS) was calculated as (VV-LV)/VV.
Twenty-eight days after stent implantation, the stented segment of coronary artery was harvested after perfusion with saline at physiologic pressure. Angioscopic observations were made while the blood was perfused away. We used an MC-800E angioscope (Nihon Kohden, Tokyo, Japan) and an AS-003 optic fiber (Nihon Kohden). We examined the entire stented segment from the distal to the proximal end and evaluated the inner surface for neointima and thrombi.
Three cross sections were cut for pathologic analysis, 1 each from the proximal, central, and distal portions along each stent. Sections were ground to a thickness of approximately 50 μm and stained with hematoxylin and eosin. A computerized imaging system (ImageJ 1.30, NIH) was used for histomorphometric measurements of the following parameters: lumen area, defined as the area circumscribed by the neointima-lumen border, and neointima area, defined as the area between the lumen and the internal elastic lamina (IEL). In-stent restenosis was estimated as the neointima area/IEL area.
Data are expressed as mean ± SEM. Mean values of variables were compared by a 2-sided unpaired t-test. A value of P < 0.05 was considered statistically significant.
Distribution of Antisense Oligodeoxynucleotide Targeted to the PDGF A-Chain
The distribution of FITC-labeled antisense ODN 24 hours after implantation is shown in Figure 1. FITC-labeled antisense ODN was distributed predominantly in the endothelial layer of the stented coronary artery.
IVUS Findings of In-Stent Restenosis
The stent/artery ratio was 1.3 ± 0.1, 1.3 ± 0.1, and 1.4 ± 0.1 in coronary arteries implanted with bare metal stents, nonsense ODN-coated stents, and antisense ODN-coated stents, respectively. There were no significant differences between the 3 groups.
Occlusive neointima filled the lumen and encircled the ultrasound catheter in arteries implanted with control stents (Fig. 2A, B), whereas the lumen was clear in coronary arteries implanted with antisense ODN-coated stents (Fig. 2C, D). Volumetric in-stent restenosis ratios were 64 ± 11.9, 44 ± 3.4, and 26 ± 3.8% in coronary arteries implanted with bare metal stents, nonsense ODN-coated stents, and antisense ODN-coated stents, respectively. The in-stent restenosis ratio was significantly less in the antisense group than in the control and nonsense groups (Fig. 2E).
The lumen surface was smooth and lacked protrusions in the stented segments of coronary arteries in all groups (Fig. 3A-C). Struts in the antisense ODN-coated stent group were visibly embedded in the neointima (Fig. 3C), whereas embedding was invisible in the nonsense ODN-coated stent and control stent groups, indicating decreased neointimal hyperplasia in the antisense group. An irregular surface with protrusions into the lumen were observed in sirolimus-coated stented segments, and some ragged red thrombi were observed adhering to the lumen surface (Fig. 3D).
Microscopic images obtained 28 days after implantations are shown in Figure 4. Neointima formation was less in coronary arteries implanted with antisense ODN-coated stents than in arteries implanted with control stents or nonsense ODN-coated stents. In-stent stenosis in distal, central, and proximal areas of stented artery are shown in Figure 5. In-stent stenosis of control, antisense, and nonsense groups was 70 ± 9.7, 58 ± 4.2, and 41 ± 5.5% in the proximal end; 77 ± 5.8, 68 ± 12.2, and 38 ± 5.3% in the central segment; and 65 ± 9.1, 70 ± 4.9, and 38 ± 5.9% in the distal end. These findings indicate that the antisense ODN targeted to the PDGF A-chain significantly inhibited in-stent neointima formation.
Differences in endothelialization differed between sirolimus-coated stents and antisense ODN-coated stents (Fig. 6). Patchy and interrupted endothelial cells in the artery lumen were observed in arteries implanted with sirolimus-coated stents (Fig. 6A), whereas a continuous lining of endothelial cells was observed in arteries implanted with antisense ODN-coated stents (Fig. 6B). There is no difference in appearance of reendothelialization among antisense ODN-coated stent, nonsense ODN-coated stent, and bare metal stent. These data suggest that the antisense ODN does not interfere with in-stent endothelialization.
Various classes of agents are available and are under investigation for the use of DESs to prevent in-stent restenosis of the coronary artery and include immunosuppressive (sirolimus), antiinflammatory (corticoid and tranilast), antiproliferative (paclitaxel, angiopeptin, and actinomycin), and antithrombotic agents (hirudin and iloprost).18 Sirolimus halts cell-cycle progression at the G1 phase, resulting in inhibition of replication and proliferation.6 Sirolimus halts the cell cycle nonspecifically; therefore, sirolimus-coated DESs prevent not only VSMC hyperplasia but also endothelialization, which is required for the prevention of thrombosis. Patients implanted with a sirolimus-coated DES require the administration of antiplatelet agents for at least 3 months after stent implantation. The timing of reendothelialization differs between patients, and late thrombosis is a major problem in patients implanted with sirolimus-coated DESs. Thus, development of DESs that preserve endothelialization is necessary.
In the present study, stent-based delivery of an antisense ODN targeted to the PDGF A-chain showed a potent inhibitory effect on VSMC proliferation, showing a 60% decrease in percent volume stenosis and preservation of endothelialization. DESs with sirolimus did not preserve endothelialization.
In 2001 it was that VSMCs in the neointima originate from circulating bone marrow cells in addition to media VSMCs.19 Intimal VSMCs, which show the synthetic phenotype, abundantly express several cytokines and growth factors, including PDGF, transforming growth factor-beta, basic fibroblast growth factor, endothelin, and angiotensin II, which are involved in neointima formation.20 Neointima are composed of VSMCs and extracellular matrix.21 It is likely that antisense ODN targeted to the PDGF A-chain inhibits neointima formation at least in part.
The PDGF A-chain contributes to VSMC proliferation in arterial proliferative disease. The PDGF B-chain is expressed constitutively in the vasculature and in platelets, whereas the PDGF A-chain is produced only in VSMCs of the synthetic phenotype. Kruppel-like zinc-finger transcription factor 5 (KLF5) has been established as a transcription factor that alters VSMCs from the contractile to the synthetic phenotype, inducing expression of the PDGF A-chain.22 These findings suggest that the PDGF A-chain is a critical target for the inhibition of VSMC hyperplasia in neointima. Our results suggest that the antisense ODN targeted to the PDGF A-chain is a specific inhibitor of neointimal VSMC proliferation that does not affect endothelialization. Stent-based delivery of this antisense ODN inhibited neointima formation and preserved endothelialization in the absence of antiplatelet agents.
One problem of exogenously administered nucleic acid-based medicines, including antisense DNA, is that they are readily degraded by nucleases in vivo. However, locally delivered nucleic acids from a DES show less degradation than do systemically administered nuclei acids. In the present experiments, we chemically modified antisense ODN to the phosphorothioate type, dissolved it in hydrogel, and applied it to both the balloon and the stent, resulting in efficient delivery of antisense ODN to the coronary artery. Nonsense ODN with a random sequence also showed a tendency to inhibit the neointima formation to a lesser extent. Similar nonspecific effects of ODN have confounded the interpretation of previous antisense studies. Several studies report that charged ODN behaves like polyanions such as heparin and heparan, which bind and sequester heparin-binding growth factors, such as basic fibroblast growth factor or PDGF, at the basement membrane.23 Another report described the nonspecific cellular activation of the transcription factor Sp1 by phosphorothioate-linked ODN.24
Recent applications of antisense ODN targeted to c-myc have been reported for the prevention of in-stent stenosis of the coronary artery in pigs25 and humans.26 An advanced antisense ODN (AVI-4126) to c-myc-eluting stent inhibited c-myc expression and significantly inhibited neointimal hyperplasia by 40% in a pig model. C-myc is also required in the progression of cell cycle. Therefore, a DES with antisense ODN targeted to c-myc may inhibit reendothelialization in a manner similar to a sirolimus-coated DES.
There are some limitations in the present study. An important point in this animal model is that the arteries are devoid of atherosclerotic lesions, whereas in patients, most stent struts are in contact with atheromatous plaque and not with the media. In the present study, degree of in-stent restenosis was examined only at 1 month after stent implantation; longer follow-up should be performed in future experiments. The coating system in the present study is hydrogel, which is different with sirolimus-coated stents (Cypher stent). The impaired reendothelialization with the Cypher stent might be mainly related to the polymer-coated stent surface and not to the drug.27
Based on the results in the present study, we plan to repeat the experiments with longer follow-up and repeat the experiments in an animal model of coronary atherosclerosis. After animal experiments, we plan to apply stents coated with antisense ODN targeted to the PDGF A-chain for treatment of human coronary diseases. Clinical trials of antisense therapies have been performed for human diseases including cancer.28,29 Even high doses of infused antisense ODN for long periods show little or no side effects or toxicity in clinical trials, indicating the safety of antisense therapy in humans. Although antisense ODNs can be applied to DESs, certain technical problems need to be addressed. In the present study, we used polyethylenimine as the delivery regent. Procedures providing more effective and safer delivery of antisense ODN from DESs should be assessed.
In the present study, we applied hydrogel to the balloon and the stent. This was particularly helpful for coating the stent with antisense ODN. We are developing a coated stent that more effectively absorbs and delivers agents to the coronary artery. At physiologic pH, antisense ODNs are negatively charged. Thus, the metal surface of the stent can be coated with a positively charged substance, enabling antisense ODNs to adhere strongly to the stent.
In conclusion, we showed that stents coated with an antisense ODN targeted to the PDGF A-chain effectively inhibited in-stent stenosis and preserved reendothelialization. These results indicate that stent-based delivery of antisense ODN targeted to the PDGF A-chain is a safe and feasible therapy for coronary arterial disease.
We gratefully acknowledge the expert technical assistance of Yoshiki Taniguchi.
1. Virmani R, Farb A. Pathology of in-stent restenosis. Curr Opin Lipidol
2. Cohen DJ, Bakhai A, Shi C, et al. Cost-effectiveness of sirolimus-eluting stents for treatment of complex coronary stenoses: results from the Sirolimus-Eluting Balloon Expandable Stent in the Treatment of Patients With De Novo Native Coronary Artery Lesions (SIRIUS) trial. Circulation
3. Regar E, Serruys PW, Bode C, et al. Angiographic findings of the multicenter Randomized Study With the Sirolimus-Eluting Bx Velocity Balloon-Expandable Stent (RAVEL): sirolimus-eluting stents inhibit restenosis irrespective of the vessel size. Circulation
4. Colombo A, Drzewiecki J, Banning A, et al. Randomized study to assess the effectiveness of slow- and moderate-release polymer-based paclitaxel-eluting stents for coronary artery lesions. Circulation
5. Park SJ, Shim WH, Ho DS, et al. A paclitaxel-eluting stent for the prevention of coronary restenosis. N Engl J Med
6. Murakami M, Ichisaka T, Maeda M, et al. mTOR is essential for growth and proliferation in early mouse embryos and embryonic stem cells. Mol Cell Biol
7. Iakovou I, Sangiorgi GM, Stankovic G, et al. Effectiveness of sirolimus-eluting stent implantation for treatment of in-stent restenosis after brachytherapy failure. Am J Cardiol
8. Werner GS, Emig U, Krack A, et al. Sirolimus-eluting stents for the prevention of restenosis in a worst-case scenario of diffuse and recurrent in-stent restenosis. Cath Cardiovasc Interv
9. Hart CE, Forstrom JW, Kelly JD, et al. Two classes of PDGF receptor recognize different isoforms of PDGF. Science
10. Johnsson A, Heldin CH, Westermark B, et al. Platelet-derived growth factor: identification of constituent polypeptide chains. Biochem Biophys Res Commun
11. Nilsson J, Sjolund M, Palmberg L, et al. Arterial smooth muscle cells in primary culture produce a platelet-derived growth factor-like protein. Proc Natl Acad Sci U S A
12. Fukuda N, Kubo A, Watanabe Y, et al. Antisense oligodeoxynucleotide complementary to platelet-derived growth factor A-chain messenger RNA inhibits the arterial proliferation in spontaneously hypertensive rats without altering their blood pressures. J Hypertens
13. Kishioka H, Fukuda N, Nakayama M, et al. Effect of methylene methylimino linkage of antisense oligonucleotide to the platelet-derived growth factor A-chain on growth of vascular smooth muscle cells from spontaneously hypertensive rats. Eur J Pharmacol
14. Kishioka H, Fukuda N, Wen-Yang H, et al. Effects of PDGF A-chain antisense oligodeoxynucleotides on growth of cardiovascular organs in stroke-prone spontaneously hypertensive rats. Am J Hypertens
15. Iyer RP, Phillips LR, Egan W, et al. The automated synthesis of sulfur-containing oligodeoxyribonucleotides using 3H-1,2-benzodithiol-3-one 1,1-dioxide as a sulfur-transfer reagent. J Org Chem
16. Boussif O, Lezoualc'h F, Zanta MA, et al. A versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: polyethylenimine. Proc Natl Acad Sci U S A
17. Jaschke B, Michaelis C, Milz S, et al. Local statin therapy differentially interferes with smooth muscle and endothelial cell proliferation and reduces neointima on a drug-eluting stent platform. Cardiovasc Res
18. Wong A, Chan C. Drug-eluting stents: the end of restenosis? Ann Acad Med Singapore
19. Han CI, Campbell GR, Campbell JH. Circulating bone marrow cells can contribute to neointimal formation. J Vasc Res
20. Carmeliet P, Moons L, Dewerchin M, et al. Insights in vessel development and vascular disorders using targeted inactivation and transfer of vascular endothelial growth factor, the tissue factor receptor, and the plasminogen system. Ann N Y Acad Sci
21. Majesky MW. Neointima formation after acute vascular injury. Role of counteradhesive extracellular matrix proteins. Tex Heart Inst J
22. Shindo T, Manabe I, Fukushima Y, et al. Kruppel-like zinc-finger transcription factor KLF5/BTEB2 is a target for angiotensin II signaling and an essential regulator of cardiovascular remodeling. Nat Med
23. Wellstein A, Zugmaier G, Califano JA 3rd, et al. Tumor growth dependent on Kaposi's sarcoma-derived fibroblast growth factor inhibited by pentosan polysulfate. J Natl Cancer Inst
24. Perez JR, Li Y, Stein CA, et al. Sequence-independent induction of Sp1 transcription factor activity by phosphorothioate oligodeoxynucleotides. Proc Natl Acad Sci U S A
25. Kipshidze NN, Iversen P, Kim HS, et al. Advanced c-myc antisense (AVI-4126)-eluting phosphorylcholine-coated stent implantation is associated with complete vascular healing and reduced neointimal formation in the porcine coronary restenosis model. Cath Cardiovasc Interv
26. Kutryk MJ, Foley DP, van den Brand M, et al. Local intracoronary administration of antisense oligonucleotide against c-myc for the prevention of in-stent restenosis: results of the randomized investigation by the Thoraxcenter of antisense DNA using local delivery and IVUS after coronary stenting (ITALICS) trial. J Am Coll Cardiol
27. Hausleiter J, Kastrati A, Wessely R, et al. Prevention of restenosis by a novel drug-eluting stent system with a dose-adjustable, polymer-free, on-site stent coating. Eur Heart J
28. Advani R, Peethambaram P, Lum BL, et al. A Phase II trial of aprinocarsen, an antisense oligonucleotide inhibitor of protein kinase C alpha, administered as a 21-day infusion to patients with advanced ovarian carcinoma. Cancer
29. Marshall J, Chen H, Yang D, et al. A phase I trial of a Bcl-2 antisense (G3139) and weekly docetaxel in patients with advanced breast cancer and other solid tumors. Ann Oncol