Author + information
- Received March 23, 2007
- Revision received August 14, 2007
- Accepted August 20, 2007
- Published online December 4, 2007.
- Jan Hinrich Bräsen, MD⁎,†,1,
- Olli Leppänen, MD⁎,‡,1,
- Matias Inkala, MD⁎,
- Tommi Heikura, MSc⁎,
- Max Levin, MD, PhD§,
- Fabian Ahrens, MSc∥,
- Juha Rutanen, MD, PhD⁎,
- Hubertus Pietsch, DVM¶,
- David Bergqvist, MD, PhD‡,
- Anna-Liisa Levonen, MD, PhD⁎,
- Samar Basu, PhD‡,
- Thomas Zeller, MD, FESC⁎⁎,
- Günter Klöppel, MD†,
- Mikko O. Laukkanen, PhD†† and
- Seppo Ylä-Herttuala, MD, PhD, FESC⁎,‡‡,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Seppo Ylä-Herttuala, A. I. Virtanen Institute, University of Kuopio, P.O. Box 1627, FIN-70211 Kuopio, Finland.
Objectives This study examined whether local gene therapy with extracellular superoxide dismutase (EC-SOD) could inhibit in-stent restenosis in atherosclerotic Watanabe heritable hyperlipidemic rabbits.
Background Stenting causes an acute increase in superoxide anion production and oxidative stress; EC-SOD is a major component of antioxidative defense in blood vessels and has powerful cardioprotective effects in ischemic myocardium.
Methods Endothelial denudation and stenting were done in 36 adult (15 to 18 months old) rabbits. Catheter-mediated intramural delivery of clinical good manufacturing practice-grade adenoviruses encoding rabbit EC-SOD were done simultaneously with stenting. Control animals received adenovirus-encoding nuclear-targeted β-galactosidase (AdLacZ). Circulating markers for oxidative stress (nonesterified 8-iso-prostaglandin F2alpha) were measured. Analysis of 6-day, 28-day, and 90-day vessel histology, radical production, oxidation-specific epitopes, and expression studies were performed.
Results The EC-SOD treatment reduced oxidant production in stented vessels compared with control vessels. Early systemic recovery of total SOD activity was observed in the treated rabbits. The EC-SOD significantly accelerated endothelial recovery (67.4% ± 10.8% vs. 24.2.1% ± 4.6% at 6 days, p < 0.05; 89.3% ± 3.7% vs. 45.1% ± 9.6% at 28 days, p < 0.05), and the beneficial effect involved increased proliferation of regenerating endothelium. The EC-SOD group showed a 61.3% lower (p < 0.05) neointimal formation at 28 days, with a similar, albeit nonsignificant trend at 90 days (1.20 ± 0.32 mm2vs. 1.88 ± 0.24 mm2, p = 0.06).
Conclusions The results suggest a central pathogenetic role of oxidation sensitive signaling processes in endothelial recovery and developing in-stent restenosis in atherosclerotic vessels. Local therapy against oxidative stress represents a promising therapeutic strategy in stent-induced vascular injury.
Currently available drug-eluting stents have significantly reduced target lesion revascularization rates but interfere with endothelial integrity (1), demanding further research for complementary therapies (2). Balloon angioplasty induces an acute increase in systemic oxidative stress (3–4), local superoxide anion (O2−) generation (5), and decreased vascular superoxide dismutase (SOD) activity (6–7). Vascular trauma imposed by struts and prolonged arterial strain initiates a repair process that is distinct from ballon injury (1), and may impose additional oxidative burden that can be chronic (8).
Extracellular superoxide dismutase (EC-SOD) is a major component of antioxidative defense in blood vessels (9), and exogenously delivered EC-SOD protects against balloon-induced neointima formation (6) and constrictive remodeling (7) and has powerful cardioprotective properties (10).
The aim of this study was to test whether local EC-SOD delivery has beneficial effects on stent-induced vascular trauma. We used an established Watanabe heritable hyperlipidemic (WHHL) rabbit model (11), in which atherosclerosis closely mimics human disease (12). It was found that EC-SOD reduced tissue oxidant production, attenuated developing in-stent restenosis (ISR), and accelerated endothelial regrowth.
The study was approved by the University of Kuopio Research Animal Ethics Committee and conforms to National Institutes of Health guidelines. Adult WHHL rabbits (n = 36) received daily aspirin and clopidogrel, were kept on standard diet, and were randomized to 6 groups (Table 1).
Three days after aortic denudation (13), the rabbits were heparinized and a bare-metal stent (approximately 1.0:1 to 1:1 stent-to-artery ratio, 3.0 × 18 mm, approximately 9 atm, Guidant Multi-Link, Guidant Corporation, Santa Clara, California) was implanted to an infrarenal aortic segment free of side branches. Infusion of adenoviruses encoding rabbit extracellular superoxide dismutase (AdEC-SOD) or control adenovirus-encoding nuclear-targeted β-galactosidase (AdLacZ) (1.15 × 1010plaque-forming units in 2.0 ml saline over 2 min) was performed to the stented segment via injection ports of Infiltrator drug delivery catheter (Boston Scientific, Natick, Massachusetts). Instrumentations and euthanasia were performed using fluanisone/fentanyl and midazolam anesthesia.
Replication-deficient E1-partially E3-deleted clinical good manufacturing practice-grade adenoviruses encoding nuclear-targeted β-galactosidase (LacZ) or rabbit EC-SOD were produced (6) and tested to be free from contaminants (14).
Blood sampling and autopsy
Blood was collected at intervals for nonesterified 8-iso-prostaglandin F2alpha (PGF2α), a biomarker of oxidative stress (15); circulating total SOD (Dojindo Laboratories, Kumamoto, Japan); and basic clinical chemistry parameters (13). Eighteen animals (3 per group) underwent veterinary autopsy (13).
The 28- and 90-day vessels, the proximal two-thirds of the stented vessel from 6-day animals, and all proximal and distal (≥2 cm from stent edges) reference segments were immersion-fixed in formalin (8). The distal one-third (6-day animals) was snap-frozen in liquid nitrogen and further processed for reverse transcriptase–polymerase chain reaction of transduced EC-SODmRNA and dihydroethidium (DHE) staining (6) after removal of struts.
The formalin-fixed samples were sectioned (<5-μm sections, 4 to 6 per animal, 4 mm apart) (Fig. 1A)and stained (8). Immunohistochemistry was performed with monoclonal antibodies (mAbs) against endothelium (platelet endothelial cell adhesion molecule-1 [PECAM-1], 1:50, Santa Cruz Biotechnology, Santa Cruz, California), rabbit macrophages (RAM 11 mAb, 1:50, DAKO A/S, Glostrup, Denmark), muscle-specific actin (HHF35 mAb, 1:50, DAKO A/S), β-galactosidase (anti-β Gal mAb, 1:50, Promega, Madison, Wisconsin), proliferating cell expression Ki-67 nuclear protein (Mib-1 mAb, 1:50, DAKO A/S), peroxynitrite (ONOO−)-modified epitopes (a polyclonal antibody against 3-nitrotyrosine, 1:50, Upstate Biotechnology, Lake Placid, New York), and hypochlorite (HOCl)-modified proteins (mAb clones 2D10G9 and 6E10E11, 1:20, a generous gift from Dr. E. Malle, Innsbruck, Austria). Apoptotic nuclei were assessed with a TdT-mediated dUTP-biotin nick end-labelling (TUNEL) kit (R&D, Minneapolis, Minnesota).
Digitized histomorphometry, macrophage and endothelial cells counts, and a modified (0 to 4) Schwartz injury score were determined. The mean value from all serial segments was calculated and used for statistical analysis. Stent edges were not analyzed. All measurements were performed blinded (F.A., J.H.B.).
Analysis of variance or nonparametric Kruskal-Wallis tests and ttest were used as appropriate. The Bonferroni correction was used when necessary. All data represent mean ± standard error. Differences in the results were considered statistically significant at p < 0.05.
EC-SOD inhibits ISR
The lesions had structural features similar to human plaques subjected to percutaneous coronary intervention (8) (Figs. 1B and 1C). At 6 days no differences in ISR were detected. As compared with 6-day control animals, the neointima size in 28-day control animals was 2.6-fold (p < 0.05) larger. At this principal 28-day end point (16), the EC-SOD rabbits had a 61.3% lower neointima than control animals (p < 0.05), with a nonsignificant trend at 90 days (1.20 ± 0.32 mm2vs. 1.88 ± 0.24 mm2, p = 0.06) (Fig. 1D). At 6 days, the apoptosis rate was higher in neointimal SMCs in EC-SOD animals versus 6-day control animals (6.8 ± 0.9% vs. 3.7 ± 0.4% in control animals, p < 0.01). No differences between the treatments were observed in SMC proliferation or apoptosis rates at other time points or vessel layers (data not shown).
No differences were observed in injury score, body weights, blood lipids (Table 1), the other wall areas (Fig. 1D, Table 1), clinical chemistry, or lesion formation in de-endothelialized but unstented segments (data not shown). No signs of toxicity were detected (data not shown).
EC-SOD accelerates re-endothelialization
The EC-SOD animals showed higher re-endothelialization rates at 6 days and 28 days. At 90 days, both groups were >90% endothelialized (Fig. 2A).At 6 days, endothelial cell proliferation was higher in EC-SOD rabbits versus 6-day control animals (12.6 ± 1.3% vs. 4.9 ± 0.7%, p < 0.01). No effects were seen at 28 days (4.5% ± 1.4% vs. 4.1% ± 1.2% in control animals, p = NS), and at 90 days only a few proliferating (Ki-67 immunopositive) nuclei were detected (p = NS, data not shown). Macrophages (Fig. 2B) or matrix accumulation (6 days: 47% ± 12% of vessel wall area within external elastic lamina in EC-SOD rabbits vs. 43% ± 9% in control animals, p = NS; 28 days: 65 ± 12% vs. 76 ± 16%, p = NS; 90 days: 74 ± 17% vs. 82 ± 21%, p = NS) did not differ among the groups.
EC-SOD reduces oxidative stress
In EC-SOD rabbits, a reduction in oxidant production was seen in areas where the transgene expression was detected primarily in SMCs, by X-gal staining (a chromogen [5-brom-4-chlor-3-indoxyl-β-D-galactopyranoside] substrate for β-galactosidase) and monoclonal antibody against β-galactosidase. The EC-SODtransgene expression was demonstrated at the mRNA level (Fig. 2C). Staining for ONOO−modified epitopes was found to be colocalized with DHE labeling, whereas HOCl-modified epitopes were primarily detected more deeply, in macrophage-rich areas (Fig. 2B). Morphometrically determined immunostained areas of ONOO−and HOCl epitopes did not differ among the groups (data not shown).
The intravascular manipulations caused transient approximately 2-fold increases in 8-iso-PGF2α(p < 0.05 vs. baseline in both groups) (Fig. 3).In EC-SOD animals, faster normalization of the values (Fig. 3) and recovery in the circulating SOD activity were detected (Table 1).
The EC-SOD had significant beneficial effects on endothelial recovery and attenuation of ISR. Treated rabbits had a 61.3% lower neointimal area than control animals at the principal 28-day end point (16). Although no intervention against the acute (<24 h) increase in stenting-associated oxidative stress (3) could be obtained using the current study design (simultaneous delivery of gene therapy vectors and stenting), the treatment had important beneficial effects on normalization of circulating F2-isoprostanes and total SOD activity, suppressed further generation of tissue oxidants, and attenuated lesion growth. Our observations with stents are well in line with those of Szöcs et al. (5) regarding the importance of SMC-derived O2−during the first 15 days after injury.
Although a trend (p = 0.06) toward a 36.2% lower neointima at 90 days in the EC-SOD group was seen, long-term efficacy remains unknown. The failure of several systemic antioxidants in clinical studies, the relatively narrow therapeutic index of adenoviruses, and the potential accumulation to nontarget organs (in case of delivery failure) are important concerns for human applications. These issues warrant further study and assessment before initiation of clinical trials.
The most striking feature of this study was the rapid re-endothelialization in the EC-SOD rabbits, which apparently contributed to the beneficial effects. The fact that ONOO−-modified epitopes and the area of reduced oxidant production were colocalized suggests an increased bioactivity of nitric oxide near the lumen after EC-SOD delivery. More importantly, the absence of differences in arterial oxidation epitopes in relation to neointima area, despite reduced lesion formation, suggests an important pathophysiological role for intracellular oxidation-sensitive signaling pathways in developing neointima and ISR.
In conclusion, our study shows that local EC-SOD delivery by clinical-grade adenoviruses accelerates re-endothelialization and attenuates developing ISR in WHHL rabbits. Local therapy against oxidative stress represents a promising therapeutic strategy against stent-induced vascular injury in atherosclerotic arteries.
The authors thank Dr. B. Sipos, Ms. G. Philipp, Mrs. B. Kuhlmann, Ms. M. Nieminen, and the staff at the National Laboratory Animal Center, Finland, for technical assistance; Dr. P. Syrjälä for veterinary autopsy; Dr. V. Kiviniemi for expert comments; and Mrs. M. Poikolainen for preparation of the manuscript.
↵1 Drs. Bräsen and Leppänen contributed equally to this work.
Supported by the Finnish Academy, Finnish Foundation for Cardiovascular Research, FIT Biotech, Klinisch Pharmakologischer Verbund Berlin-Brandenburg, and the Swedish Society for Nephrology.
- Abbreviations and Acronyms
- extracellular superoxide dismutase
- in-stent restenosis
- prostaglandin F2alpha
- smooth muscle cell
- superoxide dismutase
- Watanabe heritable hyperlipidemic
- Received March 23, 2007.
- Revision received August 14, 2007.
- Accepted August 20, 2007.
- American College of Cardiology Foundation
- Joner M.,
- Finn A.V.,
- Farb A.,
- et al.
- Tsimikas S.,
- Lau H.K.,
- Han K.R.,
- et al.
- Szöcs K.,
- Lassegue B.,
- Sorescu D.,
- et al.
- Laukkanen M.O.,
- Kivela A.,
- Rissanen T.,
- et al.
- Leite P.F.,
- Danilovic A.,
- Moriel P.,
- et al.
- Bräsen J.H.,
- Kivela A.,
- Roser K.,
- et al.
- Agrawal R.S.,
- Muangman S.,
- Layne M.D.,
- et al.
- Rosenfeld M.E.,
- Tsukada T.,
- Chait A.,
- et al.
- Leppänen O.,
- Rutanen J.,
- Hiltunen M.O.,
- et al.
- Schwartz R.S.,
- Edelman E.R.,
- Carter A.,
- et al.