Adenovirus-mediated delivery of the Gax transcription factor to rat carotid arteries inhibits smooth muscle proliferation and induces apoptosis

Article metrics


Adenovirus-mediated gene delivery in animal models of vascular injury has provided insights into the mechanisms underlying vessel wall pathologies. We have previously demonstrated that overexpression of the Gax transcription factor inhibits neointimal formation in rat and rabbit models of arterial injury. Here, we evaluate potential mechanisms for the reduction in stenotic lesion size due to Gax overexpression. At 3, 7 and 14 days after injury the Ad-Gax-infected arteries displayed a marked decrease in medial vascular smooth muscle cell number (3 days, 54% reduction P < 0.01; 7 days, 41% reduction p < 0.003; 14 days, 49% reduction p < 0.02). at 3 days after injury, pcna expression was attenuated in the ad-gax-treated vessels compared with control vessels (65% reduction p < 0.02), indicating a reduction in cellular proliferation. at 7 days and 14 days after injury ad-gax-infected arteries exhibited elevated number of tunel-positive medial vsmcs compared with control-treated arteries (7 days, 9.2-fold increase p < 0.03; 14 days, 17.2-fold increase p < 0.03), indicating an induction of apoptotic cell death. these data suggest that deregulated gax expression induces first cell cycle arrest and then apoptosis in the vascular smooth muscle cells that contribute to the neointimal layer. therefore, the efficacy of this therapeutic strategy appears to result from the ability of the gax transcriptional regulator to modulate multiple cellular responses.


The vascular smooth muscle cells (VSMCs) that line the normal vessel wall are quiescent, contractile, non-migratory and display a differentiated phenotype. Upon injury these VSMCs become activated, migrate and proliferate to form an intimal lesion.1 These lesions also exhibit high frequencies of VSMC apoptosis.2,3,4

In VSMCs, the decision to proliferate, differentiate, or undergo apoptosis may be linked through the actions of common nuclear factors that function as nodal point regulators. Recent investigations have begun to identify transcription factors that regulate VSMC phenotype including Gax,5 Ets-1,6 GATA-6,7,8 and MEF2A.9,10 These transcription factors are regulated, either positively or negatively, by mitogens suggesting a function in coordinating the expression of cell cycle and apoptosis regulatory genes during VSMC phenotypic modulation.11

Homeobox genes are transcription factors that function as pleiotropic regulators of cell fate, functioning in either a positive or negative fashion to regulate cell growth12,13 and viability.14,15,16,17 The muscle-specific homeobox gene gax, first identified in rat aorta smooth muscle, is highly expressed in quiescent VSMCs where it is down-regulated upon mitogen stimulation5 or vascular injury.18 Adenovirus-mediated expression of Gax (Ad-Gax) in culture inhibits cell cycle activity by arresting cells in the G1 phase of the cell cycle.19 In addition, Gax-expressing cells undergo apoptotic cell death under conditions of prolonged mitogen-activation while quiescent cells are refractive to Gax-induced apoptosis.20

When the effects of constitutive Gax expression were analyzed in vivo, neointimal hyperplasia was markedly attenuated in both rat carotid19 and rabbit iliac arteries21 following injury. Other investigators have reported similar results with adenovirus-mediated expression of cell cycle inhibitors or cytotoxic genes in the vessel wall.22,23,24,25 In most cases, the effect of adenovirus-mediated transgene overexpression at the cellular level was not analyzed. Here, we have elucidated a potential mechanism for the reduction in stenotic lesion size that occurs in the presence of constitutive Gax expression. At 3, 7 and 14 days after injury the Ad-Gax-infected arteries displayed a decrease in medial cell number which can be attributed to a reduction in cell proliferation and/or cell viability. At 3 days after injury PCNA expression was markedly attenuated in the Ad-Gax-treated vessels compared with the control vessels, suggesting that Gax overexpression inhibits cell cycle activity in vivo. At 7 and 14 days Ad-Gax-infected arteries displayed elevated numbers of TUNEL-positive medial VSMCs compared with saline- or Ad-β-gal-treated arteries, suggesting that Gax also induces apoptosis in vivo. Thus, Ad-Gax promotes vessel patency by inducing both cell cycle arrest and apoptosis in the VSMCs that contribute to the hyperproliferative lesion.


Gax mediated inhibition of neointimal hyperplasia

Balloon injury to the rat carotid artery results in the formation of a neointimal layer composed of proliferating VSMCs.26,27 Within hours of injury, Gax mRNA18 and protein (HP, unpublished data) are down-regulated in the medial VSMCs of the vessel wall. Saline-treated vessels or vessels infected with a control virus that expresses β-galactosidase (Ad-β-gal), develop considerable neointimal mass by 2 weeks following injury. In contrast, the proliferative response following balloon injury in arteries infected with a replication-deficient adenovirus expressing Gax (Ad-Gax) were markedly attenuated.19

We have expanded on this earlier study by examining neointimal formation in response to various doses of Ad-Gax. Rat carotid arteries were denuded with a balloon catheter and immediately exposed to 1 × 107 p.f.u. to 1 × 1010 p.f.u. of Ad-Gax or 1 × 1010 p.f.u. Ad-β-gal virus. Following a 20-min infection period, the viral solution was withdrawn and the ligatures were removed to restore circulation. Rats were killed 2 weeks later and quantitative morphometric analyses were performed on cross sections of the treated vessels. The Ad-β-gal-transduced carotid arteries had developed considerable neointimal thickening with a intima:media ratio of 1.31 ± 0.13. In contrast, Ad-Gax-treatment markedly reduced the hyperproliferative response to balloon injury at doses of 1 × 108 p.f.u. or greater. All vessels treated with the high doses of Ad-Gax displayed a reduced intima:media ratios of 0.53 ± 0.12 (1 × 108 p.f.u.), 0.68 ± 0.15 (1 × 109 p.f.u.), and 0.62 ± 0.15 (1 × 1010 p.f.u.). Statistical analysis confirmed that Ad-Gax treatment significantly inhibited the development of intimal thickening relative to the Ad-β-gal controls (Figure 1). In contrast, balloon-injured rat carotid arteries infected with a low dose of Ad-Gax (1 × 107) displayed neointimal formation (intima:media ratio 1.25 ± 0.14) similar to Ad-β-gal-treated or saline-treated vessels (intima:media ratio 1.4 ± 0.14).

Figure 1

Ad-Gax treatment following injury inhibits neointimal formation. The common carotid artery of male Sprague–Dawley rats (n = 32) were de-endothelialized by distention with a balloon catheter to induce the hyperproliferation of vascular myocytes. Following injury the de-endothelialized region was treated with adenovirus for 20 min. The rats were killed at 14 days following treatment and the injured region of each artery was recovered at necropsy. Intimal and medial areas were measured by quantitative morphometric analysis of arterial cross sections and the results of that analysis are reported as the ratio of the intimal to medial area (I/M). Values represent the mean ± standard error and were compared for statistical significance by application of an unpaired two-tailed Student’s t test. An asterisk (*) indicates P < 0.01 relative to saline or Ad-β-gal.

Ad-Gax reduces medial cell number

Previous studies in rat carotid arteries have shown that balloon injury results in a 65% loss of medial VSMCs within 4 h.28 Under normal conditions the repopulation of the media is complete by 3 days.25 Thus, we assessed the inhibitory action of Ad-Gax on hyperplasia by examining medial VSMC number at 3, 7 and 14 days after injury in the Ad-Gax- and Ad-β-gal-treated arteries. At 3 days after injury the Ad-Gax-treated arteries displayed a 50% decrease in media cell number compared with the saline- and Ad-β-gal-treated vessels (Figure 2). At 7 days after injury, the Ad-Gax infected vessels displayed a 41% decrease in medial cell number compared with the saline-treated vessels. A decrease was also evident at 14 days after injury as arteries transduced with Ad-Gax contained only 37% and 59% of the medial cells found in Ad-β-gal- or saline-treated vessels, respectively.

Figure 2

Ad-Gax treatment following injury reduces medial cell number. Ad-Gax- or Ad-β-gal-treated (1 × 109 p.f.u.) balloon-injured carotid arteries (n = 42) were harvested at 3, 7 and 14 days after injury. Cell number was assessed in multiple cross sections by counting the number of nuclei in sections stained by hematoxylin and eosin. Values represent the mean ± standard error and were compared for statistical significance by application of an unpaired two-tailed Student’s t test. An asterisk (*) indicates P < 0.05 relative to saline.

Ad-Gax reduces medial PCNA expression

Since Ad-Gax infection inhibits cell cycle progression in vitro,19 PCNA expression was analyzed by immunohistochemistry to determine if Ad-Gax inhibits VSMC proliferation in vivo. Previous investigations have demonstrated that peak medial cell proliferation occurs between 48 and 72 h after balloon injury.27,29 At 3 days after injury arteries treated with Ad-Gax (Figure 3a) displayed decreases of 67% and 65% in PCNA expression compared with saline- and Ad-β-gal-treated vessels, respectively (Figure 3b). At later time-points (7 and 14 days) little or no medial PCNA expression was detected under any condition analyzed (not shown), consistent with previous reports.27,29 Thus, Gax overexpression appears to decrease VSMC populations at early time-points through cell cycle inhibition.

Figure 3

Gax overexpression reduces PCNA expression. (a) A representative photomicrograph of rat carotid arteries 3 days after injury and treatment with either saline, Ad-β-gal or Ad-Gax. Sections obtained from Ad-Gax (n = 7), Ad-β-gal (n = 3) (1 × 109 p.f.u.) or saline-treated (n = 4) balloon-injured rat carotid arteries. Arteries were incubated with either mouse anti-PCNA antibody or control IgG followed by the application of a biotinylated horse anti-mouse secondary antibody. Alkaline phosphatase conjugated to streptavidin and the vector blue alkaline phosphatase substrate kit III were used to detect immune complexes (blue). (b) Quantitative analysis demonstrating that Ad-Gax inhibits medial VSMC proliferation. PCNA positive cells were assessed in multiple sections by counting the number of stained nuclei. Values represent the mean ± standard error and were compared for statistical significance by application of an unpaired two-tailed Student’s t test. An asterisk (*) indicates P < 0.05 relative to Ad-β-gal.

Gax induces VSMC apoptosis in vivo

To test whether enhanced VSMC apoptosis also contributed to a decrease in lesion size in Ad-Gax-treated vessels, DNA fragmentation analysis by TdT-mediated dUTP-fluorescein (TUNEL) labeling was performed on arterial sections that were harvested at 3, 7 and 14 days after injury. At 3 days after injury no significant difference in TUNEL-positive medial VSMCs was observed in the Ad-Gax transduced arteries compared with the saline- or Ad-β-gal-treated arteries. However, at 7 and 14 days after injury the Ad-Gax-infected vessels displayed a marked increase in TUNEL-positive medial VSMCs. Pyknotic nuclei, indicated by condensed chromatin, (Figure 4a) co-localized with TUNEL-positive cells in the Ad-Gax-treated arteries, providing further evidence that the TUNEL-positive cells are apoptotic. Quantitative analysis revealed that Ad-Gax infected arteries harvested at 7 and 14 days after injury displayed 22-fold (P < 0.05) and 44.3-fold (P < 0.01) increases, respectively, in the number of TUNEL-positive medial VSMCs compared with saline-treated or Ad-β-gal-infected arteries (Figure 4b). Thus, deregulated Gax expression induces apoptosis in vivo and may be responsible for a reduction in the hyperproliferative response of VSMCs to balloon injury at later time-points.

Figure 4

Ad-Gax treatment of balloon-injured rat carotid arteries induces medial VSMC apoptosis. (a) Representative micrographs of rat carotid arteries obtained at 14 days after injury and stained for TUNEL-positive medial VSMCs. Balloon-injured rat carotid arteries treated with 1 × 109 p.f.u. of adenovirus were examined for apoptotic cells by TUNEL analysis as described in methods at 3 (n = 14), 7 (n = 8) and 14 days (n = 26) following treatment and injury. Nuclei were stained with Hoechst 33258. TUNEL-positive cells exhibit green fluorescence while the position of the nuclei can be determined by the blue staining. The white arrows indicate the location of the internal elastic lamina and white arrowheads indicate condensed chromatin that correspond to TUNEL-positive nuclei. (b) Quantification of the TUNEL-positive medial VSMCs in infected arteries at 3, 7 and 14 days after injury. Approximately 100 Hoechst-positive nuclei were counted per section. The values represent the mean ± standard error and were compared for statistical significance by application of an unpaired two-tailed Student’s t test. An asterisk (*) indicates P < 0.03 relative to saline or β-gal.


Though numerous cytostatic and cytotoxic agents have been delivered to the vessel wall to inhibit intimal hyperplasia, few studies have examined the mechanisms of transgene action at a cellular level in vivo. In carotid arteries, the expression of the transcription factor Gax is down-regulated upon injury.18 This down-regulation may be required for successful VSMC cell cycle progression or to ensure cell survival during the initial proliferative response. This appears to be the case in vivo since forced Gax expression inhibits injury-induced intimal hyperplasia.19,21 In this paper we addressed the cellular mechanisms that may be responsible for this inhibition.

We have evaluated the effects of Gax overexpression on intimal hyperplasia in balloon-injured rat carotid arteries at the cellular level. At 3, 7 and 14 days after injury, arteries infected with Ad-Gax displayed reduced medial cell populations compared with either saline- or Ad-β-gal-treated vessels. The medial cell decrease at 3 days after injury may be attributed to the growth arrest by Gax overexpression and the inhibition of VSMC repopulation of the injured vessels, as PCNA expression was decreased by 67% and 65% in the Ad-Gax-treated arteries compared with saline- or Ad-β-gal-transduced vessels, respectively. No significant difference in the apoptosis rate in the 3-day injured vessels was observed between the Ad-Gax and saline- or Ad-β-gal-treated arteries. However, by 7 or 14 days following injury, arteries treated with Ad-Gax exhibited a significant increase in medial-VSMC apoptosis compared with control arteries. Apoptosis induced by Gax overexpression was largely observed in the medial, but not the neointimal layer. Consistent with these observations, immunohistochemical analyses revealed positive transgene expression using anti-HA antibody in medial, but not neointimal VSMCs (not shown).

Since homeodomain proteins generally function as pleiotropic regulators through their abilities to activate and suppress batteries of downstream genes, it is possible that Gax overexpression following acute vascular injury might have multiple effects on medial VSMCs. Here, we have shown that Gax exhibits dual actions in VSMCs of balloon-injured rat carotid arteries, promoting both cell cycle arrest and apoptosis, consistent with previous in vitro observations.19,20 It is noteworthy that Gax is unable to induce apoptosis in quiescent VSMCs, apoptosis is observed only in proliferating cells,20 the subpopulation that forms the stenotic lesion in injured vessels.29 Such a lack of toxicity to quiescent VSMCs indicates that Ad-Gax should prove an advantageous therapeutic strategy for treating conditions characterized by VSMC hyperproliferation.

Materials and methods

Rat model of smooth muscle hyperplasia

The model of balloon injury was based on that described by Clowes et al.29 Male Sprague–Dawley rats weighing 400–500 g were anesthetized with an intraperitoneal injection of sodium pentobarbital (45 mg/kg, Abbot Laboratories, North Chicago, IL, USA). The bifurcation of the left common carotid artery was exposed through a midline incision and the left common, internal and external carotid arteries were temporarily ligated. A 2F embolectomy catheter (Baxter Edwards Healthcare, Irvine, CA, USA) was introduced into the external carotid and advanced to the distal ligation of the common carotid. The balloon was inflated with saline and drawn towards the arteriotomy site three times to produce a distending, de-endothelializing injury. A temporarily isolated segment of the artery was treated with saline or 1 × 107 p.f.u. to 1 × 1010 p.f.u. in a volume of 100 μl of either Ad-Gax19,20 or Ad-β-gal.19,20,30 The solution was injected via a cannula inserted just proximal to the carotid bifurcation and the vessel was incubated for 20 min, after which the solution was withdrawn and the cannula removed. All transgenes are expressed under control of the cytomegalovirus promoter. The proximal external carotid artery was ligated and blood flow was restored to the common carotid artery by release of the ligatures. The experimental protocol was approved by the Institutional Animal Care and Use Committee and complied with the ‘Guide for the Care and Use of Laboratory Animals’ (NIH Publication No. 86–23, revised 1985).

Rats were killed at 3, 7 and 14 days following treatment with an intraperitoneal injection of pentobarbital (100 mg/kg). The balloon injured segment of the left common carotid artery, from the proximal edge of the omohyoid muscle to the carotid bifurcation, was perfused with saline and dissected from the surrounding tissue. Tissue was fixed in 4% paraformaldehyde or 100% methanol until embedded in paraffin. Several 5-μm sections were cut from each tissue specimen. Separate sections from each specimen were stained with hematoxylin and eosin or with Richardson’s combination elastic-trichrome stain for conventional light microscopic analysis.

Histological images of cross sections of hematoxylin and eosin or elastic-trichrome-stained arterial sections were projected on to a digitizing board (Summagraphics, Fairfield, CT, USA) and the intimal, medial and luminal areas were measured by quantitative morphometric analysis using a computerized sketching program (MACMEASURE, version 1.9, National Institute of Mental Health, Bethesda, MD, USA).

Statistical analysis

Results were expressed as the mean ± s.e.m. Differences between groups were analyzed using an unpaired two-tailed Student’s t test.

TUNEL labeling and nuclear condensation

Paraformaldehyde fixed sections (5 μm) were deparaffinized and rehydrated. The tissue was permeabilized with 20 mg/ml of proteinase K for 30 min after which fluorescein-conjugated dUTP and TdT enzyme were added according to the manufacturer’s specifications (Boehringer Mannheim, Indianapolis, IN, USA; in situ death detection kit). Nuclei were counterstained with Hoechst 33258 (Sigma, St Louis, MO, USA), and mounted for examination using mounting media for fluorescence (Kirkegaard and Perry Laboratories, Gaithersburg, MD, USA). Specimens were examined and photographed on a Nikon (Tokyo, Japan) Diaphot microscope equipped with a phase-contrast and epifluorescence optics.


Five-micron sections from uninjured and injured rat arterial tissue fixed in methanol were deparaffinized and blocked in 10% horse serum. Sections were incubated with mouse anti-PCNA antibody (Signet, Dedham, MA, USA) or mouse IgG (Sigma). Prostates from rats 3 days after castration were used as a positive control.31 A biotinylated horse-anti-mouse secondary antibody (Vector Laboratories, Burlingame, CA, USA) followed by alkaline phosphatase (BioGenex, San Ramon, CA, USA) conjugated to streptavidin were used to detect primary antibody complexes. Visualization was accomplished using the Vector Blue alkaline phosphatase substrate kit III (Vector Laboratories).


This research was supported by National Institutes of Health grants HL50692 and AR40197 awarded to KW. KW was an established investigator of the American Heart Association during the course of this study. We thank Linda Whittaker for assistance in preparation of this manuscript.


  1. 1

    Campbell JH, Campbell GR . Biology of the vessel wall and atherosclerosis Clin Exp Hypertens 1989 11: 901–913

  2. 2

    Geng JY, Libby P . Evidence for apoptosis in advanced human atheroma: colocalization with interleukin-1β converting enzyme Am J Pathol 1995 147: 251–266

  3. 3

    Han DKM et al. Evidence for apoptosis in human atherogenesis and in a rat vascular injury model Am J Pathol 1995 147: 267–277

  4. 4

    Isner JM, Kearney M, Bortman S, Passeri J . Apoptosis in human atherosclerosis and restenosis Circulation 1995 91: 2703–2711

  5. 5

    Gorski DH et al. Molecular cloning of a diverged homeobox gene that is rapidly down-regulated during the G0/G1 transition in vascular smooth muscle cells Mol Cell Biol 1993 13: 3722–3733

  6. 6

    Hultgardh-Nilsson A et al. Regulated expression of the Ets-1 transcription factor in vascular smooth muscle cells in vivo and in vitro Circ Res 1996 78: 589–595

  7. 7

    Morrisey EE, Ip HS, Parmacek MS . GATA-6: a zinc finger transcription factor that is expressed in multiple cell lineages derived from lateral mesoderm Dev Biol 1996 177: 309–323

  8. 8

    Suzuki E et al. The human GATA-6 gene: structure, chromosomal location and regulation of expression by tissue-specific and mitogen-responsive signals Genomics 1996 38: 283–290

  9. 9

    Firulli AB et al. Myocyte enhancer binding factor-2 expression and activity in vascular smooth muscle cells Circ Res 1996 78: 196–204

  10. 10

    Suzuki E et al. Serum-induction of MEF2/RSRF expression in vascular myocytes is mediated at the level of translation Mol Cell Biol 1995 15: 3415–3423

  11. 11

    Walsh K, Perlman H . Molecular strategies to inhibit restenosis: modulation of the vascular myocyte phenotype Semin Intervent Cardiol 1996 1: 173–179

  12. 12

    Song K, Wang Y, Sassoon D . Expression of Hox-7.1 in myoblasts inhibits terminal differentiation and induces cell transformation Nature (Lond) 1992 360: 477–481

  13. 13

    Castrillo JL, Theill LE, Karin M . Function of the homeodomain protein GHF1 in pituitary cell proliferation Science 1991 253: 197–199

  14. 14

    Dear TN et al. The Hox11 gene is essential for cell survival during spleen development Development 1995 121: 2909–2915

  15. 15

    Xiong WC, Montell C . Defective glia induce apoptosis in the repo visual system of drosophila Neuron 1995 14: 581–590

  16. 16

    Quaggin SE, Yeger H, Igarashi P . Antisense oligonucleotides to Cux-1, a cut-related homeobox gene, cause increased apoptosis in mouse embryonic kidney cultures J Clin Invest 1997 99: 718–724

  17. 17

    Marazzi G, Wang Y, Sassoon D . Msx2 is a transcriptional regulator in the BMP-4 mediated programmed cell death pathway Dev Biol 1997 186: 127–138

  18. 18

    Weir L et al. Expression of GAX, a growth-arrest homeobox gene, is rapidly down-regulated in the rat carotid artery during the proliferative response to balloon injury J Biol Chem 1995 270: 5457–5461

  19. 19

    Smith RC et al. p21CIP1-mediated inhibition of cell proliferation by overexpression of the gax homeodomain gene Genes Dev 1997 11: 1674–1689

  20. 20

    Perlman H et al. Bax-mediated cell death by the Gax homeoprotein requires mitogen-activation but is independent of cell cycle activity EMBO J 1998 17: 3576–3586

  21. 21

    Maillard L et al. Percutaneous delivery of the gax gene inhibits vessel stenosis in a rabbit model of balloon angioplasty Cardiovasc Res 1997 35: 536–546

  22. 22

    Guzman RJ et al. 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 USA 1994 91: 10732–10736

  23. 23

    Ohno T et al. Gene therapy for vascular smooth muscle cell proliferation after arterial injury Science 1994 265: 781–784

  24. 24

    Simari RD et al. Regulation of cellular proliferation and intimal formation following balloon injury in atherosclerotic rabbit arteries J Clin Invest 1996 98: 225–235

  25. 25

    Sata M et al. Fas ligand gene transfer to the vessel wall inhibits neointima formation and overrides the adenovirus-mediated T cell response Proc Natl Acad Sci USA 1998 95: 1213–1217

  26. 26

    Clowes AW, Reidy MA, Clowes MM . Mechanisms of stenosis after arterial injury Lab Invest 1983 49: 208–215

  27. 27

    Wei GL et al. Temporally and spatially coordinated expression of cell cycle regulatory factors after angioplasty Circ Res 1997 80: 418–426

  28. 28

    Perlman H, Maillard L, Krasinski K, Walsh K . Evidence for the rapid onset of apoptosis in medial smooth muscle cells following balloon injury Circulation 1997 95: 981–987

  29. 29

    Clowes AW, Reidy MA, Clowes MM . Kinetics of cellular proliferation after arterial injury. I: Smooth muscle cell growth in the absence of endothelium Lab Invest 1983 49: 327–333

  30. 30

    Stratford-Perricaudet LD, Makeh I, Perricaudet M, Briand P . Widespread long-term gene transfer to mouse skeletal muscles and heart J Clin Invest 1992 90: 626–630

  31. 31

    Columbel M, Olsson CA, Ng P, Buttyan R . Hormone-regulated apoptosis results from re-entry of differentiated prostate cells on to a defective cell cycle Cancer Res 1992 52: 4313–4319

Download references

Author information

Correspondence to K Walsh.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Perlman, H., Luo, Z., Krasinski, K. et al. Adenovirus-mediated delivery of the Gax transcription factor to rat carotid arteries inhibits smooth muscle proliferation and induces apoptosis. Gene Ther 6, 758–763 (1999) doi:10.1038/

Download citation


  • smooth muscle
  • growth arrest
  • apoptosis
  • gene therapy

Further reading