Cytostatic gene therapy for occlusive vascular disease

José M. Gonzálezand Vicente Andrés

Laboratory of Vascular Biology, Department of Molecular and Cellular Pathology and Therapy, Instituto de Biomedicina de Valencia (IBV-CSIC), Jaume Roig 11,Valencia 46010, Spain.

Send correspondence to:

Vicente Andrés (Tel: +34-96-3391752; FAX: +34-96-3391751; E-mail: )

KEY WORDS: atherosclerosis, restenosis, bypass graft failure, cell cycle, gene therapy

1. Introduction

2. Antiproliferative gene therapyin animal modelsof occlusive vascular disease

2.1. Cell cycle regulatory genes

Inactivation of positive cell cycle regulatory genes

Overexpression of growth suppressors

2.2. Growth factors

2.3. Transcription factorsinvolved in cell cycle control

2.4. Miscellaneous

3. Antiproliferative gene therapy clinical trials for human occlusive vascular disease

3.1. E2F

3.2. c-myc

3.3. VEGF

4. Expert opinion

Abstract

The formation of occlusive vascular lesions during the course of atherosclerosis, in-stent restenosis, transplant vasculopathy, and vessel graft failure is a chronic inflammatory process characterized byexcessive cellular proliferation within the injured artery wall. Therefore, candidate targets for the treatment of vasculoproliferative disease include cell cycle regulatory factors, such as cyclin-dependent kinases (CDKs), cyclins, CDK inhibitory proteins (CKIs), tumor suppressors, growth factors and their receptors, and transcription factors involved in cell cycle control. Althoughseveral genetically-modified mouse models have conclusively demonstrated that increased cell proliferation aggravates atheroma development,the potential benefit of cytostatic strategies for the treatment of atherosclerosis in clinic is doubtful because humanatherosclerosis is often diagnosed at advanced stages when neointimal proliferation appears low or absent. In contrast, restenosis and graft atherosclerosis appear amenable for cytostatic strategies because neointimal lesionstypically develops over a short period of time after revascularization (e. g., 2-12 months) and is localized at the site of the intervention. Vascular interventions, both endovascular and open surgical, allow minimally invasive, easily monitored gene delivery. In this review, we will discuss preclinical studies and clinical trialsutilizingcytostatic gene therapy for occlusive vascular disease.

1. Introduction

Atherosclerosis and associated cardiovascular disease (e.g., myocardial infarction and stroke) are the main cause of mortality and morbidity in industrialized countries, and their incidence in developing countries is increasing at an alarming rate.Atherosclerosis is a chronic inflammatory disease of middle-sized and large-caliber arteries that normally progresses over several decades and may remain silent until fatal manifestations occur at advanced disease stages. Endothelial damage caused by several risk factors, such ashyperlipemia, hypertension, diabetes, and smoking, is considered the first manifestation of occlusive vascular disease (e. g., atherosclerosis, in-stent restenosis, transplant vasculopathy, and vessel graft failure)[1, 2].Both acellular (e.g., deposition of lipids and extracellular matrix components) and cellular components are involved in the growth of neointimal lesions. Accumulation of neointimal cells occurs via both transendothelial migration of circulating leukocytes and migration of vascular smooth muscle cells (VSMCs) from the tunica media towards the atheroma, and from excessive proliferation of nointimal monocyte/macrophages and VSMCs. Rupture or erosion of advanced atherosclerotic plaques can lead to thrombus formation and acute ischemic events (e.g., myocardial infarction and stroke).

Percutaneous transluminal angioplasty (PTA)and vessel by-pass grafting are widely used interventions for revascularization of obstructed vessels. Although both procedures have a high rate of initial success, their long-term efficacy is often limited by luminal narrowing, typicallywithin 3-12 months afterintervention. Among other factors, abundant neointimal cell proliferation contributes to restenosis after angioplasty and vessel graft failure [3-5].

Neointimal thickening is a complex process initiated and sustained by growth factors and their receptors, signal transduction pathways and transcription factors that ultimately control de activity of CDK/cyclins (see below). In the following sections, we will discuss gene therapy strategies designed to prevent cellular proliferation for the treatment of occlusive vascular disease, both in experimental animals and in the clinical setting.

2. Antiproliferative gene therapy in animal models of occlusive vascular disease

2.1. Cell cycle regulatory genes

Cell cycle progression in higher eukaryotes is positively regulated by several holoenzymes composed of a catalytic and a regulatorysubunit, named CDK and cyclin, respectively[6](Fig. 1).Activation of CDK/cyclins by mitogenic stimuli causes the hyperphosphorylation of the retinoblastoma protein (pRb) and the related pocket proteins p107 and p130 from mid G1 to mitosis. The complex interaction among E2F transcription factors and individual pocket proteins determines whether E2F proteins function as transcriptional activators or repressors[7].Proteins of the CKI family promote growth arrest by inhibiting CDK activity via interaction withCDK/cyclincomplexes. CKIs of the Cip/Kip family (p21Cip1, p27Kip1 and p57Kip2) bind to and inhibit a wide spectrum of CDK/cyclin holoenzymes, while members of the Ink4 family (p15Ink4b, p16Ink4a, p18Ink4c, and p19Ink4d) are specific for cyclin D-associated CDKs. Mitogenic and antimitogenic stimuli affect the rate of synthesis and degradation of CKIs, as well as their redistributionamong different CDK/cyclin pairs[6].

Inactivation of positive cell cycle regulatory genes

Arterial cell proliferation in response to balloon angioplasty in the rat carotid artery is associated with a temporally and spatially coordinated expression of CDKs and cyclins. [5][8].Importantly, augmented expression of these factors correlated with CDK2 and CDC2 activation, demonstrating the assembly of functional CDK/cyclin holoenzymes within the injured arterial wall. CDK2 and cyclin E expression has been detected in human VSMCs within atherosclerotic and restenotic tissue[9], suggesting that increased expression (and possibly activation) of positive regulators of cell cycle progression is a characteristic of vasculoproliferative disease in humans.

Neointimal thickening in animal models of balloon angioplasty is limited by antisense oligodeoxynucleotide (ODN) strategies targeting CDKs and cyclins, including cdk2[10][11], cdc2[10][11], and cyclin B1[11]. Cotransfection of antisense ODN against cdc2 and cyclin B1was more effective at reducing neointimal thickening than blockade of either gene target alone[11].Likewise, combined inactivation of cdc2 and proliferating cell nuclear antigen (pcna) by a single intraluminal delivery of antisense ODNs resulted in sustained inhibition of neointima formation in the rat carotid artery balloon-injury model[12]. However, this approach was ineffective in balloon-injured porcine coronary arteries[13]. The use of hammerhead ribozyme to pcnaalone has been successful in reducing in-stent restenosis in a porcine coronary model [14].Inactivation of cyclin G1 gene expression by retrovirus-mediated antisense gene transfer inhibited VSMC proliferation and neointima formation after balloon angioplasty[15]. Moreover, ODN against cdk2[16], and a combination of antisense ODN against pcna and cdc2[17], attenuated vessel graft failure in experimental animals.

Immusol Inc. has described methods of producing ribozymes especially targeted to cyclin B1, cdc2, and pcna to inhibit VSMC proliferation in vascular tissue, as well as ribozyme delivery systems for anti-restenosis gene therapy [301]. Another method based on siRNA inhibition of CDK4 activity has been claimed for preventing or treating cancer[302], which could also find application for treating vascular proliferative disease.

Overexpression of growth suppressors

CKIs

The efficacy of CKIs as cell cycle suppressors has been widely documented in a variety of normal and tumor cells in vitro. Evidence suggesting that the CKIs p21Cip1 and p27Kip1play important roles incardiovascular pathophysiology includes the following: 1)p21Cip1 and p27Kip1expression is upregulatedat late time points after balloon angioplasty in rat and porcine arteries coinciding with the reestablishment of the quiescent phenotype after the initial proliferative response [18][19]; 2) p27Kip1may function as a molecular switch that regulates the phenotypic response of VSMCs to both hyperplastic and hypertrophic stimuli[20, 21]; 3) p27Kip1is a negative regulator of endothelial cell (EC) proliferation and migration in vitro[22], and adenovirus-mediated overexpression of p27Kip1 inhibited angiogenesis in vivo[23]; 4) p21Cip1 and p27Kip1 may contribute to integrin-mediated control of VSMC proliferation[24]; 5) p27Kip1 limits cardiomyocyte proliferation during early postnatal development and after injury in adult mice[25][26]; 5) expression of p27Kip1 and p21Cip1 is more frequent within regions of human coronary atheromas not undergoing proliferation [19]; 6) intrinsic differences in the regulation of p27Kip1 may contribute to establishing regional variability in atherogenicity via distinct regulation of VSMC proliferation and migration [27].

Analysis of genetically-modified mice further supports the notion that CKIs are key regulators of neointimal lesion development. Geneticp27Kip1ablation, either global or selectively in hematopoietic precursors, acceleratesarterial cell proliferation and aggravates atherosclerosis in apolipoprotein E (apoE)-deficient mice[28][29]. Surprisingly, both global and hematopoietic cell-specific disruption of p21Cip1 in apoE-null mice protects from atherosclerosis, possibly due to effects of this CKI on macrophage function and inflammatory responsesthat appear independent of its cell cycle regulatory function[30].

Global genetic inactivation of either p16Ink4a[31]or p21Cip1[32]exagerates neointimal thickening induced in the mouse by mechanical vessel denudation.Regardingthe effect of p27Kip1 inactivation on neointimal lesion formation after mechanical arterial injury, Roque et al. foundsimilar lesion size in wild-type and p27Kip1-null mice [33]; however, Boehm et al. reported a marked increase in mechanically-induced neointimal tickening inp27Kip1-null mice [34].

Intraluminal delivery of replication-defective adenoviral vectors encoding p21Cip1 and p27Kip1reduced neointimal thickening in rat, porcine and murine models of balloon angioplasty[18, 35-39]. Neointimal lesion formation in a rabbit model of vein grafting was also attenuated by ectopic overexpression of p21Cip1[40], and adenovirus-mediated transfer of ap27 Kip1-p16 Ink4a fusion gene inhibited neointimal hyperplasia in balloon-injured porcine coronary [41] and cholesterol-fed rabbit carotid arteries [42].

Methods of preparation and use of recombinant adenoviral vectors capable of expressing human p21Cip1, p27Kip1, p16Ink4a and other growth suppressors have been described for inducing growth arrest of proliferating cells, as well as methods for the eradication of cancer and diseased cells [303]. Likewise, methods of inhibiting cell proliferation using purified p18Ink4c or p19Ink4d proteins, and methods of gene therapy using nucleic acids that encode these genes have been described [304, 305]. The University of Texas System has claimed the use of non-viral and viral expression vectors encoding formutant p21Cip1proteins as a therapy for the treatment of proliferative cell disorders, including cancer, restenosis, neurogenerative disease and angiogenesis-related conditions[306]. The mutations consist of Thr145Ala or Thr145Asp substitutions, which result in nuclear retention or cytoplasmic translocation of p21Cip1, respectively, which result in preferential suppression of cell growth (p21Cip1Thr145Ala), or enhanced cell survival (p21Cip1Thr145Asp).

Reagents and methods for identifying genes whose expression is modulated by induction of CKI gene expression have been presented [307]. This invention also provides reagents and methods for identifying compounds that inhibit or augment the effects of CKIs, such as p21Cip1 and p16Ink4a, on cellular gene expression, as a first step in rational drug design for preventing cellular senescence, carcinogenesis and age-related diseases (such as atherosclerosisand related disease), or for increasing the efficacy of anticancer therapies.

A method for treating vascular proliferative diseases by administering in vivo a gene encoding p27Kip1 has been patented [308]. Proteasome degradation of p27Kip1 is thought to play a major role in the regulation of p27Kip1 expression. KyushuUniversity has patented the nucleic acid and amino acid sequence of a new p27Kip1 molecular variant exhibiting resistance to proteasome degradation, as well as expression vectors encoding this derivative of p27Kip1 for gene therapy applications targeting cell propagating lesions, such as tumours and atherosclerotic plaques [309].

p53

The tumor suppressor p53displays both antiproliferative and proapoptotic actions (Fig. 2). These effects result from transcriptional activation of antiproliferative and proapoptotic genes (e. g., p21Cip1 and Bax, respectively), transcriptional repression of proproliferative and antiapoptotic genes (e. g., IGF-II and bcl-2, respectively), and direct protein-protein interactions (e. g., with helicases and caspases). Antisense p53 ODN transfection [43, 44] or p53 gene transfer [45] increasesor decreases VSMC proliferation, respectively, and VSMCs isolated from p53-deficient mice exhibit more proliferative and migratory activity than their wild-type counterparts [46].

p53 is overexpressed but not mutated in human atherosclerotic tissue [47]. Becauselack of proliferation markers in vascular cells within advanced human atherosclerotic lesions coincided with p53 and p21Cip1coexpression, Ihling et al. suggested that p53-dependent transcriptional activation of p21Cip1 might protect against excessive vascular cell growth [48]. However, it is noteworthy that p21Cip1 expression aggravates atherosclerosis in apoE-null mice[49].Both global andhematopoietic cell-specificp53 genetic ablation results in increased atherosclerosis in several murine models, including apoE-null, apoE*3-Leiden transgenic, and LDL receptor-null mice, although the relative contribution of cellular proliferation and apoptosis in these animal models remains unclear[50-53].Genetic disruption of p53 in the mouse also accelerated neointimal lesion induced by vein grafting[46]andexternal vascular cuff placement [54].

Both animal and human studies suggest that p53 plays an important role in the pathogenesis of restenosis. Intraluminal transfection of antisense p53 ODN into rat carotid artery or p53 genetic inactivation in mice accelerated neointimal hyperplasia after vascular injury[44, 55]. It has been suggested that increased VSMC proliferation and migration via inactivation of p53 in response to human cytomegalovirus infection contributes to the development of atherosclerosis and restenosis [56-59].Compared to primary cultures of human VSMCs isolated from normal vessels, VSMCs from restenosis or in-stent stenosis sites exhibit normal or enhanced responses to p53 [60]. Moreover, p53 gene transfer effectively inhibited neointimal hyperplasia after experimental angioplasty [45]and in organ cultures of human saphenous vein [61].

Introgen Therapeutics Inc has claimed a method of inhibiting the growth of human papillomavirus-transformed keratinocytes and to prevent or suppress papillomavirus-mediated cell transformation by topically administering a p53 expression cassette carried in either a viral or non-viral vector, which could be administered with a secondary anti-hyperplastic therapy [310].

pRb

The complex interplay between pRb and transcription factors of the E2F family plays a critical role in the control of cell growth [7] (Fig. 1). In quiescent cells, E2F-dependent transactivation of genes required for cell cycle progression is prevented,at least in part,by the accumulation of hypophosphorylated pRb. In contrast, mitogen-induced pRb hyperphosphorylation leads to E2F activation and cell proliferation. Inactivation of pRb by antisense ODN resulted in the induction of the proapoptotic factors bax and p53, increased number of apoptotic cells and a higher rate of DNA synthesis in human VSMCs [43]. Growth arrest of cultured VSMCs and attenuation of neointima formation after balloon angioplasty can be achieved by adenovirus-mediated transfer of several forms of pRb, including full-length constitutively active (nonphosphorylatable) and phosphorylation-competent pRb, and truncated versions of pRb[62, 63]. Similarly, adenoviral transfer of the pRb related protein RB2/p130 inhibited VSMC proliferation in vitro and prevented neointimal hyperplasia after experimental angioplasty[64]. Adenoviral transfer of a constitutively active mutant pRb protein also inhibited neointima formation in a human explant model of vein graft disease[65]. Likewise, targeting overexpression of an anti-proliferative pRb/E2F hybrid transgene to VSMCs using the human smooth muscle α-actin promoter suppresses cell proliferation and neointima formation[66].

2.2. Growth factors

Growth factors that have been implicated in the regulation of VSMC proliferation and migration include platelet-derived growth factor (PDGF), fibroblast growth factor (FGF), tumor necrosis factor- (TNF-epidermal growth factor (EGF), insulin-like growth factor-1 (IGF-1), heparin-binding epidermal growth factor-like growth factor, interleukin-1 and transforming growth factor-TGF. Expression of thesefactors is increased during atherogenesis[67]and restenosis after angioplasty[68].

TGF-

TGF1 regulatesextracellular matrix synthesis, cell chemotaxis, and proliferation in VSMCs,ECsand fibroblasts. Adenovirus-mediated antisense TGF-1 treatment reduces vein graft intimal hyperplasia[69]. Likewise, ribozymes targeted to human TGF-1 significantly inhibited angiotensin II-stimulated human VSMCs growth[70] and neointimal formation after balloon injury in the rat carotid artery [71][72].However, adenovirus-dependent gene transfer of TGF-1does not affectluminal loss after porcine coronary angioplasty, whileadenovirus-mediated TGF-3 delivery increased adventitial collagen content and the external elastic lamina area, and reduced luminal loss in this model [73].

IGF-1

IGF-1 has been implicated in the control of VSMC proliferation and locomotion [74]. Adenovirus-dependent transfer of a dominant negative truncated mutant of IGF-1 receptor in cultured VSMCs attenuated serum- or IGF-1-dependent phosphorylation of Akt and ERK1/2, inhibited migration and proliferation and increased apoptosis[74]. This strategy also reduced neointima formation after balloon angioplasty in the rat carotid artery[74].

EGF

The EGFR family consists of four receptor tyrosine kinases, EGFR (ERBB1), ERBB2 (HER2), ERBB3 (HER3) and ERBB4 (HER4), which specifically interact with approximately a dozen of EGF-like growth factors [75]. Genentech Inc has patented a method based on the administration of an antagonist of a native ErbB4 receptor for preventing excessive VSMC proliferation or migration and for the treatment of stenosis or restenosis [316].

FGF

Recombinant FGF-1 promoted neointimal hyperplasia[76]. Conversely, neutralizing antibodies directed against basic FGF (bFGF or FGF-2) inhibited neointimal VSMC accumulation after angioplasty[77], and gene transfer of a soluble FGF receptor 1molecule capable of sequestering circulating FGF-1 and FGF-2 reduced the development of accelerated graft arteriosclerosis in a rat aortic transplant model [78]. Inhibition of human FGF receptor 2 expression by antisense ODN has been claimed for treating hyperproliferative disorders [317].

PDGF

PDGF promotes proliferation of a wide range of cell types, including fibroblasts, VSMCs, and connective tissue cells. PDGF can be present as either homodimer or heterodimer of A and B chains, which bind to its dimeric receptor composed of all three combinations of  and  subunits [79]. The cleavage of the PDGF A-chain mRNA by hammerhead ribozyme attenuated human and rat VSMC growth in vitro [80][81]. This same strategy, as well as nanospheres containing antisense ODN against PDGF receptor, inhibited neointimal thickening and thrombus formation in the rat carotid artery model of balloon angioplasty [82][83].

Platelet derived endothelial cell growth factor (PD-ECGF)

PD-ECGF (also known as thymidine phosphorylase) stimulates chemotaxis of ECs and angiogenesis [84, 85]. Gene transfer of PD-ECGF increasedthe expression of heme oxygenase and p27kip1 in cultured rat VSMCs, and inhibited neointima formation in balloon-injured rat carotid arteries [86].

Vascular endothelial growth factor (VEGF)