Reduction of Restenosis After Angioplasty in an Atheromatous Rabbit Model by Suicide Gene Therapy
Background Gene delivery of the thymidine kinase (tk) gene combined with ganciclovir (GCV) limits intimal hyperplasia after abrasion of normal arteries. However, the low efficiency of adenoviral-mediated gene transfer to atherosclerotic arteries has raised concerns about the applicability of this strategy to the prevention of restenosis.
Methods and Results A replication-defective adenoviral vector expressing tk (Ad-RSVtk) demonstrated selective toxicity toward GCV-treated arterial smooth muscle cells, with oligonucleolytic cleavage suggesting apoptosis. In vivo, after demonstration of tk expression after Ad-RSVtk delivery, the combination of Ad-RSVtk followed by GCV was tested in a rabbit model of angioplasty of atheromatous iliac arteries. Angioplasty (8 atm, 20 minutes) was performed by use of a hydrogel balloon coated with Ad-RSVtk (4×109 plaque forming units). GCV was infused (25 mg.kg−1 IV BID) from days 2 through 7 after angioplasty in 8 of 12 rabbits. Four weeks later, morphometric analysis demonstrated a reduced intima-to-media ratio in the group receiving combination therapy compared with Ad-RSVtk alone (3.0±1.2 versus 5.2±0.5, P<.018). GCV per se had no effect on intimal hyperplasia after arterial injury.
Conclusions In vitro, Ad-RSVtk demonstrates selective toxicity toward GCV-treated arterial smooth muscle cells involving apoptosis. In vivo, GCV conditions reduction of neointimal formation after percutaneous delivery of Ad-RSVtk during angioplasty of atheromatous arteries.
Restenosis is, at least in part, related to intimal hyperplasia,1 especially after stent implantation.2 Adenoviral-mediated but not retroviral-mediated3 transduction of the HSV-tk gene, followed by treatment with the nucleoside analog GCV, the “suicide gene” strategy, has given promising results in the prevention of intimal hyperplasia after arterial injury in normal4 5 or atheromatous arteries.6 The efficacy of adenoviral-mediated gene transfer, however, is markedly reduced in atheromatous as opposed to normal arteries when strict percutaneous delivery is used,7 thereby raising concerns about the applicability of adenoviral-based gene therapy to the prevention of restenosis.8
Transfer of the HSV-tk gene into arterial SMCs has no cytotoxic effect.9 After treatment with GCV, the HSV-tk gene product phosphorylates GCV to yield a toxic compound,10 which selectively kills proliferating cells.11 Neighboring untransduced proliferating cells may also be affected by the cytotoxic effect.10 11 12 13 The precise mechanism of this “ bystander effect ” is still unclear but may involve diffusion of toxic metabolites through gap junctions or phagocytosis of apoptotic vesicles. The present study investigated the effects of suicide gene therapy in vitro on arterial SMCs and the effect of percutaneous gene therapy on intimal thickening after angioplasty in the atherosclerotic rabbit.
Replication-Defective Recombinant Adenoviral Vectors
Ad-RSVβgal was constructed from adenovirus serotype 5.14 Ad-RSVtk was generated in transformed human kidney 293 cells by homologous recombination between pAd.RSVtk and the Cla I fragment of Ad5 del324 viral DNA as previously described.15
In Vitro Experiments
Rabbit aortic SMCs were plated in DMEM supplemented with 20% FCS. After 24 hours, different viral concentrations were added in DMEM/0.5% FCS for 60 minutes. Cells were then incubated in a low-serum medium (0.5% FCS) for 24 hours and then with 20% FCS in the absence or presence of GCV at increasing doses. Cell viability was assessed by trypan blue exclusion after 4 days. To assess the apoptotic nature of cell death, DNA fragmentation was also visualized.
In Vivo Experiments
All experiments were performed by use of vectors from a single viral stock to avoid interstock variability.
Experiment 1 verified the presence of the HSV-tk protein after percutaneous delivery of Ad-RSVtk through a 2.5-mm hydrogel-coated angioplasty balloon (8 atm, 20 minutes) in iliac arteries of normal rabbits (n=3). The hydrogel balloon catheter has previously been used to achieve simultaneous angioplasty and gene delivery to medial SMCs by use of adenoviral vectors.16 The viral solution was pipetted onto the balloon surface and allowed to dry at room temperature. To prevent loss of the hydrogel content into the bloodstream,16 the catheter was encased in a protective sheath for advancement to the delivery site. Three days later, the rabbits were killed, and the presence of the HSV-tk protein was assessed by immunohistochemistry.
Experiment 2 tested the combination of Ad-RSVtk and GCV on neointimal formation in a double-injury model. Twelve rabbits were fed a 1% cholesterol diet. Two weeks after the rabbits started the diet, endothelial abrasion (five passes of a 4F Fogarty balloon catheter) was performed. Hydrogel balloon angioplasty and simultaneous gene delivery were performed 4 weeks later, followed by angiography. Eight rabbits received GCV (25 mg · kg−1 IV BID) from days 2 through 7 after angioplasty. Four weeks later, after angiography, the rabbits were killed, and the arteries were pressure fixed. Samples of brain, liver, lung, heart, testes, and kidney were taken for analysis. After orcein-hematoxylin-safranin staining, 12 evenly spaced sections from the arterial segments subjected to angioplasty and gene delivery were analyzed in a blind fashion by morphometry by use of the Biocom 200 image analysis system with a Dialux 20EB microscope (Leitz), a charge-coupled device (CCD) camera, and the Histo software (Biocom). Arteries with thrombotic occlusions were excluded from the analysis (n=3), as were sections in which rupture of the internal elastic lamina prevented identification of the contours (n=2).
Experiment 3 verified the absence of effect of GCV on intimal hyperplasia in a single-injury model. Nine rabbits underwent balloon angioplasty 2 weeks after starting the atherogenic diet, following the protocol described above, and received either GCV (n=4) or PBS (n=5) intravenously from days 2 through 9 after angioplasty. Immediately after angioplasty, the atherogenic diet was replaced with normal rabbit chow. Rabbits were killed 4 weeks later.
In vitro and in vivo, the presence of the HSV-tk protein was assessed using a specific mouse monoclonal antibody (provided by M. Janicot, Gencell).
All values are expressed as mean±SD. Comparisons between groups used Mann-Whitney U tests.
Ad-RSVtk was selectively toxic toward GCV-treated arterial SMCs transduced at increasing MOIs. In the presence of 25 μmol · L−1 GCV, an MOI of 1000 achieved 95% cell killing (Fig 1A⇓). Immunohistochemistry demonstrated the presence of tk protein in the cytoplasm and nuclei of transduced cells, and 75% of the SMCs were consistently transduced at this MOI. The transduction efficiency of Ad-RSVβgal at the same MOI was similar (data not shown). Therefore, with conditions that allowed efficient transduction, the IC50 for GCV was 0.3 μmol · L−1. At this high MOI, a slight decrease in cell number was seen after incubation with either Ad-RSVtk or Ad-RSVβgal in the absence of GCV. Trypan blue exclusion established that this was due to a cytostatic rather than toxic effect. The DNA fragmenting activity of the Ad-RSVtk/GCV combination was visualized by agarose gel electrophoresis. Four days after addition of GCV, a ladderlike pattern of DNA degradation was selectively observed in cells first transduced with Ad-RSVtk and then treated with GCV (Fig 1B⇓). Neither GCV alone nor recombinant adenovirus per se (Ad-RSVtk, Ad-RSVβgal) induced detectable DNA cleavage.
The transduction efficiency of Ad-RSVtk decreased markedly with reduced MOI. However, even at the lowest tested MOI of 10, efficient cell killing was also seen (Fig 1A⇑). The IC50 of GCV was 4 μmol · L−1. At this MOI, transduction efficiency was <5%, supporting the hypothesis of a bystander effect. The fact that this was observed in nonconfluent cell cultures argues against diffusion of toxic metabolites through gap junctions as its mechanism. Similar results were observed on primary cultures of human aortic SMCs (data not shown).
Experiment 1: Presence of HSV-tk Protein After Percutaneous Delivery of Ad-RSVtk
There were only sparse cells positive on immunostaining that were located in the superficial layers of the media and represented <1% of the total number of cells of the arterial wall (Fig 2A⇓).
Experiment 2: Effect of the Combination of Ad-RSVtk and GCV on Neointimal Formation in a Double-Injury Model
There were massive and similar increases in plasma cholesterol levels in both groups. Body weight, serum creatinine, alanine aminotransferase, platelets, hemoglobin, and white blood cell count did not differ between groups. Angiography showed two occlusions in each group (P=NS). After angioplasty, the minimal luminal diameter was comparable between groups (1.7±0.3 and 1.4±0.4 mm in the tk and tk+GCV groups, respectively; P=NS). However, at the time of death, the angiographic minimal luminal diameter was 0.15±0.26 mm in the tk group and 0.46±0.29 mm in the tk-GCV group (P=.16), corresponding to a 46% reduction in the loss in minimal lumen diameter in the tk+GCV group (1.55±0.3 versus 0.8±0.5 mm, respectively; P=.05). On morphometry, there was a 42% reduction of the intima-to-media ratio in the tk+GCV group compared with the tk group (3.0±1.2 versus 5.2±0.5; P=.018; Fig 2B⇑). There was no difference between groups in terms of external elastic lamina–enclosed areas at the angioplasty site (4.2±1 versus 3.8±0.8 mm2 in the tk and tk+GCV groups, respectively; P=NS) or in the ratios of external elastic lamina–enclosed areas at the angioplasty site and in adjacent proximal sections (0.88±0.3 versus 0.83±0.2 in the tk and tk+GCV groups, respectively; P=NS), suggesting no difference in vessel remodeling.
There was no evidence of toxicity at sites remote from the angioplasty. The only exception was testicular atrophy, a classically reversible phenomenon after GCV administration. There were no lymphocytic or polymorphonuclear infiltrates at the gene delivery sites 3 (Fig 2A⇑) or 28 days after delivery.
Experiment 3: Absence of Effect of GCV on Intimal Hyperplasia After Injury
The intima-to-media ratios were similar (1.1±0.5 versus 1.0±0.6 in PBS- and GCV-treated groups, respectively; P=NS). Likewise, there was no difference in intimal or medial areas between groups (0.7±0.4 versus 0.6±0.3 mm2, P=NS, and 0.6±0.1 versus 0.6±0.5 mm2, P=NS, respectively).
This study shows in vitro selective toxicity of Ad-RSVtk toward GCV-treated arterial SMCs and, in vivo, a GCV-conditioned reduction of neointimal formation after angioplasty of atheromatous arteries and HSV-tk gene delivery. The demonstration of selective oligonucleolytic cleavage by the combination of tk and GCV argues that apoptosis is involved in the cytotoxic effect of suicide gene therapy. Apoptosis was further substantiated by specific TUNEL labeling in cells treated with the Ad-RSVtk/GCV combination (not shown).
Because the efficacy of adenoviral-mediated gene delivery is markedly reduced in atherosclerotic arteries,7 the clinical relevance of previous studies demonstrating reduction of intimal hyperplasia after balloon injury of normal arteries remained to be demonstrated.8 In addition, the response to endothelial abrasion achieved in these studies with latex balloon injury differs qualitatively and quantitatively from that produced by actual balloon angioplasty.17 Finally, in previous studies, surgical ligation of collateral vessels has been used to increase transduction efficiency.6 The present study used a model of angioplasty of atheromatous rabbit arteries, causing histopathological injury comparable to that of human angioplasty,17 gene delivery was performed in atheromatous arteries by use of clinically applicable methods, and the GCV concentrations and IC50 are compatible with clinical use.18
Selected Abbreviations and Acronyms
|FCS||=||fetal calf serum|
|HSV-tk||=||herpes simplex virus 1 thymidine kinase|
|MOI||=||multiplicity of infection|
|SMC||=||smooth muscle cell|
This work was supported by the Fédération Française de Cardiologie and the BioAvenir program (Rhône-Poulenc, Ministère de la Recherche and Ministère de l’Industrie). We are indebted to J.J. Barry and Boston Scientific for providing the catheters and to N. Ratet, C. Le Feuvre, L. Maillard, M. Brami, and A. Mahfoudi for help with the experiments and the study design.
P.G.S. and O.T. are primary coauthors.
- Received March 7, 1997.
- Revision received May 19, 1997.
- Accepted May 20, 1997.
- Copyright © 1997 by American Heart Association
Ohno T, Gordon D, San H, Pompili VJ, Imperiale MJ, Nabel GJ, Nabel EG. Gene therapy for vascular smooth muscle cell proliferation after arterial injury. Science. 1994;265:781-784.
Guzman RJ, Hirschowitz EA, Brody SL, Crystal RG, Epstein SE, Finkel T. In vivo suppression of injury-induced vascular smooth muscle cell accumulation using adenovirus-mediated transfer of the herpes simplex virus thymidine kinase gene. Proc Natl Acad Sci U S A. 1994;91:10732-10736.
Feldman LJ, Steg PG, Zheng LP, Chen D, Kearney M, McGarr SE, Barry JJ, Dedieu J-F, Perricaudet M, Isner JM. Low-efficiency of percutaneous adenovirus-mediated arterial gene transfer in the atherosclerotic rabbit. J Clin Invest. 1995;95:2662-2671.
Cascells W, Willerson JT. Amphotropic but not atherotropic: another caveat for adenoviral gene therapy. J Clin Invest. 1995;95:2425-2426.
Culver KW, Ram Z, Wallbridge S, Ishii H, Oldfield EH, Blaese RL. In vivo gene transfer with retroviral vector producer cells for treatment of experimental brain tumors. Science. 1992;256:1550-1552.
Smythe R, Hwang H, Amin K, Eck S, Davidson B, Wilson J, Kaiser L, Adbelda S. Use of recombinant adenovirus to transfer the herpes simplex virus thymidine kinase (HSVtk) gene to thoracic neoplasms: an effective in vitro drug sensitization system. Cancer Res. 1994;54:2055-2059.
Freeman SM, Abboud CN, Whartenby KA, Packman CH, Koeplin DS, Moolten FL, Abraham GN. The “bystander effect’: tumor regression when a fraction of the tumor mass is genetically modified. Cancer Res. 1993;53:5274-5283.
Quantin B, Stratford-Perricaudet LD, Tajbakhsh S, Mandel J-L. Adenovirus as an expression vector in muscle cells in vivo. Proc Natl Acad Sci U S A. 1992;89:2581-2584.
Maron A, Gustin T, Le Roux A, Mottet I, Dedieu JF, Brion JP, Demeure R, Perricaudet M, Octave JN. Gene therapy of rat C6 glioma using adenovirus mediated transfer of the herpes simplex virus thymidine kinase gene: long term follow-up by magnetic resonance imaging. Gene Ther. 1996;3:315-22.
Steg PG, Feldman LJ, Scoazec J-Y, Tahlil O, Barry JJ, Boulechfar S, Ragot T, Isner JM, Perricaudet M. Arterial gene transfer to rabbit endothelial and smooth muscle cells using percutaneous delivery of an adenoviral vector. Circulation. 1994;90:1648-1656.
Schwartz RS. Animal models of human coronary restenosis. In: Topol EJ, ed. Textbook of Interventional Cardiology. Philadelphia, Pa: WB Saunders; 1994:365-381.
Paul S, Dummer S. Topics in clinical pharmacology: Ganciclovir. Am J Med Sci. 1992;4:272-277. Review.