Prospects for adenovirus-mediated gene therapy of inherited diseases of the myocardium

Time for primary review 30 days. Many genetic loci linked to a variety of diseases have been identified, and disease-causing mutations characterized. These have included diseases of the myocardium, such as X-linked dilated cardiomyopathy [1–3], hypertrophic cardiomyopathy [4–6], and Long QT syndrome [7–9]. The elucidation of gene defects has allowed different therapeutic strategies to be proposed, including the use of pharmaceutical agents to replace or to antagonize the mutated protein, and replacement of the defective gene with a functional one (gene therapy). There have been many publications describing the use of vectors to transduce target cells for the correction of gene defects or for anti-viral therapy (for reviews see [10–14]). Such vectors have included recombinant retroviruses, adenoviruses, adeno-associated viruses, and herpes viruses, as well as non-viral vectors. Each vector has inherent advantages and disadvantages. At this time the adenoviruses are most commonly used and represent the most likely vector for efficient transduction of the myocardium. While many groups have reported the use of recombinant adenoviruses to transduce cells in vitro and in vivo, and even their use in clinical trials for the correction of the cystic fibrosis gene defect [15, 16], there have been few reports describing the targeted expression of gene products for the correction of myocardial diseases. In this review we will discuss the potential of gene therapeutic approaches for the treatment of myocardial disease, as well as consider some of the limitations and risks associated with the use of the adenovirus-based vectors. Recombinant adenoviruses have been the most commonly utilized vectors for the transduction of cells both in vitro and in vivo. This has primarily been due to their ability to be propagated and purified to high titers, their ability to transduce non-dividing cells, and their broad spectrum of target tissues. Most studies have used so-called first-generation adenovirus vectors (Fig. 1). The functions of the adenoviral early (E) and late (L) proteins have been reviewed elsewhere [17]. In brief, the E1 region encodes proteins essential for cell transformation, as well as for transactivation of other viral genes, host cell shutoff, and control of the lytic cycle. The E2 region encodes the DNA polymerase, a DNA binding protein involved in the control of viral gene transcription, and a terminal protein involved in viral assembly. The E3 region is non-essential for in vitro replication and down-regulates major histocompatibility complex (MHC) at the cell surface, decreasing the target for recognition by cytotoxic T-cells. The E4 region encodes proteins involved in the regulation of L gene transcript splicing. Most of the L proteins are viral structural proteins. A representation of the adenovirus genome, indicating (with arrows) the positions of early (E1–4) and late (L1–6) primary transcripts. The early region encodes proteins involved in controlling virus replication, including the DNA polymerase (E2) while the late region encodes structural proteins, including the penton base (L1–2), hexon (L3) and fiber (L5) proteins. Note that each region gives rise to multiple protein products by alternative splicing of the primary transcripts. In first-generation recombinant adenoviral vectors the E1 region is deleted and can be replaced by the promoter/transgene construct. For the preparation of recombinant adenoviruses, the foreign gene, together with a suitable promoter, are cloned into a plasmid containing the left end of an adenovirus genome, replacing the E1 region [18]. Since E1 controls the expression of the other adenoviral proteins, deletion of this region essentially shuts off virus transcription in recombinant viruses. This plasmid is transfected into 293 cells, which carry a complementary copy of the E1 region, together with adenoviral genomic DNA lacking the E1 region. The recombinant genome, derived by homologous recombination between the plasmid and the adenoviral DNA, is packaged into virion and released into the culture media following cell lysis. The infectious virus is purified away from empty virion by density gradient centrifugation. One major problem associated with this preparation method is that a small amount of contaminating wild type virus is also generated, albeit usually less than 0.01% of the titer of the recombinant virus. Adenovirus enters cells by receptor-mediated endocytosis and since the receptor is widely expressed, a variety of tissues can be transduced. The viral genome usually persists as an episome, although integration into the host genome has been reported, especially for adenovirus type 12 [19]. Most commonly the virus is delivered systemically: it was shown by Huard et al. [20]that, depending upon the route of virus administration, different tissues could be transduced. There have been few reports describing the delivery of a transgene to the myocardium. Transduction by direct injection into the myocardium has been reported in rat [21–23], mouse [24], dog [25]and pig [26], as well as by intracoronary infusion in rabbits [27], perfusion of the donor heart [23], and systemic injection [19, 28]. Virus transduction mainly occurred in the region of the myocardium directly surrounding the site of injection [26], or the coronary artery injected [27], while the liver was also transduced in some cases [21, 27]. In hearts that were infected by perfusion, the distribution of transgene expression was more widespread within the myocardium than after direct injection. However, perfusion is only a viable option if the target organ is to be transplanted in the host. Therefore, improved methods of delivery to the myocardium will need to be developed to achieve sufficient levels of transgene expression, particularly since for many diseases most, if not all, myocytes will be required to be transduced to achieve a therapeutic effect. Further, in most cases, transgene expression peaked approximately one week after virus injection and declined rapidly thereafter [21, 23, 26]. This has been observed in most in vivo studies using adenovirus vectors and is probably due to the host immune response against virus-encoded proteins [29], or in many model systems, against the transgene product being used as an indicator of virus transduction [22, 30]. For the first-generation recombinant adenovirus vectors the synthesis of virus proteins results either from leaky expression from the vector sequences or from the contamination of vector stocks with wild-type virus. Two approaches have been used to improve the persistence of transgene expression: (1) the development of the so-called second- and third-generation vectors, and (2) immunosuppression of the host. The first second-generation vectors had a temperature-sensitive mutation in the E2a region, further ablating virus replication [31]. However, the use of this vector produced only small increases in the duration of transgene expression in the cotton rat [31], mice and dogs [30], although Engelhardt et al. [31]reported a reduction in the cellular immune response. Other possibilities are to develop vectors constitutively expressing E3 [32], which could reduce viral antigen presentation and, thus, T-cell killing of transduced cells, and vectors lacking E4 as well as E1 [33]. Lee et al. [32]reported that transduction with an adenoviral vector expressing both β-galactosidase and E3 failed to stimulate the proliferation of anti-adenovirus or anti-transgene-specific antibodies. Gao et al. [33]reported that transduction of mouse liver with an E1/E4-deleted vector resulted in reduced virus protein expression and a blunted immune response against the virus, as well as a reduction in virus-induced apoptosis in the target organ. Recently recombinant adenovirus vectors lacking all virus-encoded genes have been described [34–36], but the ability of these vectors to achieve persistent transgene expression has not been reported. Barr et al. [37]reported that following systemic recombinant adenovirus administration into immunodeficient mice, transgene expression persisted indefinately within the liver. Further, tracheal administration in neonatal cotton rats, having a relatively naı̈ve immune system, resulted in persistent transgene expression, for at least 6 months [38]. Based upon the observations that, in the absence of an immune response against the viral-encoded proteins or the transgene product, persistent expression was obtained, host immunosuppression has been studied. Cyclosporin A administration in a canine hemophilia B model resulted in a prolongation of therapeutic levels of the blood clotting factor IX, from 3 weeks to 6 months, following adenovirus administration [39]. The co-administration of soluble CTLAIg (which blocks co-stimulatory signals between T-cells and antigen presenting cells) with recombinant adenovirus in mice, resulted in persistent transgene expression, but without long-term immunosuppression [40]. An important consideration for the correction of myocardial disease using adenoviral vectors is the limitation of the expression of the transgene to within the myocardium. The widespread expression of the adenovirus receptor probably means that such control will have to be exerted at the level of gene expression, by using a cardiac muscle-specific promoter. Lee et al. [41]used the myosin light chain-2 promoter to direct expression of the luciferase reporter gene in transgenic mice, while Rothmann et al. [42]obtained cardiac muscle-specific expression of a luciferase gene under the control of the same promoter, following direct injection of a recombinant adenovirus into the cardiac cavity. Virus was detected by polymerase chain reaction (PCR) in many tissues, but gene expression was limited, almost exclusively, to the heart [42]. Thus, it should be possible to direct the expression of a gene to a specific tissue but many problems remain in controlling the stoichiometry and timing of gene expression, although the ability to control gene expression by drug or hormone treatment has been proposed [43–45]. Dilated cardiomyopathy (DCM) is the most common form of cardiomyopathy, responsible for approximately 60% of cases. This disorder, a primary myocardial disease that causes ventricular dilation and dysfunction, primarily of the left ventricle, has many etiologies [46–49]. It is believed that approximately 30% of cases are familial in nature [50, 51]and of these, the transmission may be autosomal dominant or recessive, X-linked, or mitochondrial. The most common form of disease appears to be the autosomal dominant type, however. Patients with DCM are treated symptomatically with anticongestive measures, antiarrhythmic medications (when necessary) and, in some cases, β-blockers. Failure of medical therapy usually leads to consideration for cardiac transplantation. Two X-linked cardiomyopathies have been described, X-linked dilated cardiomyopathy (XLCM) and Barth syndrome. XLCM typically presents in teenage boys or males in their twenties and rapidly leads to severe symptoms of congestive heart failure with associated ventricular arrhythmias, usually resulting in death or transplantation within 1 year of presentation. Towbin et al. [52]first identified linkage in XLCM to the dystrophin gene on Xp21, the gene responsible for Duchenne and Becker muscular dystrophy. Muntoni et al. [1]described mutations in the muscle promoter and muscle-expressed exon 1 in XLCM and more recently, other mutations have been described [3]. Barth syndrome, or X-linked cardio-skeletal myopathy with neutropenia, abnormal mitochondria and 3-methylglutaconic aciduria typically presents with severe life-threatening heart failure in infancy. The gene responsible, G4.5 [2], is located on Xq28. Multiple loci have been linked to autosomal dominant inherited DCM, including genes for pure DCM on 1q32 [53], 9q13–q22 [54], and 10q21–23 [55], and genes for DCM with conduction disease on 1p1–1q1 [56]and 3p25–p22 [57]. None of these genes have been identified at this time. While the correction of the dystrophin defects associated with XLCM has not been reported, a number of groups have reported approaches for the correction of Duchenne muscular dystrophy (DMD). This is a potentially lethal disorder of skeletal muscle resulting from mutations in the dystrophin gene [58], distinct from those identified in XLCM [1, 3]. The mdx mouse lacks dystrophin, and has been used by several groups as a model for therapeutic approaches for DMD [59]. Due to their limited capacity, the first-generation adenovirus vectors were unable to accommodate the 14kb dystrophin cDNA and promoter. Therefore, so-called dystrophin mini genes, obtained by cloning the truncated cDNA from Becker muscular dystrophy patients [60], have been used as the therapeutic gene. The transduction of skeletal muscle of mdx mice by recombinant adenoviruses encoding such dystrophin constructs, with transient correction of the dystrophin defect, has been reported [61–63]. Recently, two groups described the use of third-generation adenovirus vectors lacking all viral genes and capable of packaging the entire dystrophin cDNA under the control of either the muscle-specific creatine kinase promoter [64]or the Rous sarcoma virus long terminal repeat promoter [35]. In both cases, efficient transduction of mdx muscle fibers were obtained following intramuscular injection and in one of the studies restoration of the dystrophin-associated proteins to the muscle membrane and a decrease in centrally located nuclei, a characteristic of dystrophin-deficient muscle, were observed [64]. These vectors offer the ability to package up to 30 kb of promoter/gene construct, a size that should be sufficient for most genes, and the absence of virus genes should reduce the problems of the host immune response against the vector, as long as the vector is sufficiently purified from helper virus. Neither study reported on the persistence of transgene expression, although Haecker et al. [35]noted some decrease in dystrophin expression was observed between the second and could be due to an immune response against the transgene product A model of inherited dilated cardiomyopathy has been appears to of the disease, particularly with to in the The genetic defect the disease in this model is at and thus, to of the In for Barth syndrome or XLCM at the of study reported in the mdx However, such in gene delivery should to use gene therapy as an for the treatment of these diseases. hypertrophic cardiomyopathy is a cardiac disorder that is inherited in an autosomal dominant and causes It is as ventricular the and associated with myocardial and different genetic loci have been for including and and genes A number of are not linked to the indicating that other genes is primarily using either therapy or therapy These agents are at the of the disease and but not reduce the hypertrophic response. In patients with left ventricular are also these, is the myocardium is the size of the left ventricular and for as well as the to option in patients with may be This which is may decrease the level of and has also been reported to decrease the amount of in some patients In et al. that the gene for myosin chain is responsible for in the gene are responsible for the disease in 30% of of that the mutations a the mutated protein with the of the wild In et al. that mutations in the gene are proteins associated with the of and also that the cardiac gene is responsible for is a of the complex that is located on the in the cardiac gene of cases, while less than of cases had mutations in It is not by which mutations in the cardiac and In et al. et al. that mutations in the cardiac myosin binding gene is in and myosin chain in and in of appears to recently, et al. with due to mutations in the myosin light chain at and myosin essential light chain Since cardiac cardiac myosin and essential light chain mutations the same is a disease of the et al. reported in patients with and that mutations in early death in while other mutations were with to long-term studies have been reported these for as well as for mutations in of patients from in et al. described a model of resulting from a mutation of the gene. In the cardiac and the while for the mutation within a few of Such will the of the of gene therapy An adenovirus vector encoding the under the control of a promoter, has been described This cardiac myocytes but other have been reported. The and of gene therapy approaches will upon the of of the For mutations will that the expression of the gene be or by the of with the functional gene. and that transduction of the liver of transgenic mice that hormone with a recombinant adenovirus encoding a against resulted in than reduction in et al. shown that transduction with a recombinant adenovirus encoding a to the form of the resulted in of the in Long QT syndrome is a cardiac disorder that causes and death from ventricular arrhythmias, and is by QT on There are two of inherited an autosomal dominant and an autosomal as and syndrome, is associated with genetic has been for autosomal dominant as syndrome, is more common and is not associated with other In et al. the first gene for autosomal dominant to Towbin et al. genetic and et al. the second to and the to Recently, et al. the to for and have been In et al. the of the gene for as a cardiac gene with and et al. the that is the cardiac gene, and of in that it encodes the rapidly increases in were shown to were also in and it was that either a or a Thus, that are likely to be for has a that of homologous each of which mutations in a late of by either or Thus, mutations a but with of that the persistent associated with mutations could potentially be in et al. the cloning of a gene for is in the heart and encodes a protein homologous to with a by is a protein with a with only one the in cardiac to mutations probably either a or a different therapeutic for including left cardiac or of a None of these the or ventricular in all and of several genes and their disease-causing therapeutic strategies have a has been shown to the of patients and to have only a on and patients the was in the for patients with It is important to that, although the described are in on it is these can ventricular and associated clinical including An treatment for patients with is The in vitro transduction of canine myocytes by a recombinant encoding an B the control of the promoter, has been reported. were from dog as well as hearts from dogs with congestive heart resulting from and transduced with the recombinant In myocytes there is a of resulting in of the gene into the myocytes the potential prolongation and the observed was upon the level of transgene The of gene therapy for the treatment of will upon the number of defective identified in these the ability to between these and the ability to transgene Since some of the mutations may or it is likely that a gene will at least in of the expression of the mutated protein, as described for At the development of as a means of controlling this appears to be the more viable option typically presents in as an disease and is associated with high while more commonly with the less form of the a number of studies have for persistent viral of the myocardium in patients with or dilated cardiomyopathy by the of viral genomic sequences These the that is a of a viral Most of the studies viral genomic sequences in the myocardium of patients with or upon the and as However, is a direct for these viruses, the of other and disease in is associated with persistent viral et al. et al. myocardial of presenting with either or by to and as well as virus virus and of viral genome was in of cases of and in 6 of cases of but in of the controls In virus was culture from only one cardiac and in all cases from patients with In of the virus cases by adenovirus was (Fig. these included all of the cases of In 12 cases were for and each for and Since that time we have such which have these These are in specific for the hexon region of adenovirus within the L were DNA of a number of the hexon that in most cases the was infected by adenovirus type to the that the adenoviruses and are the most commonly detected in the myocardium, it is to that adenovirus and and B virus a common receptor using to the adenovirus hexon region and the gene is shown The products were detected by of a A 1 kb size is shown in the first Adenovirus control is as a in the second the control is in the Note that 3 is of a The results obtained after of two ventricular and a are The from was for adenovirus DNA while the that from and the were is used as a control for of the of DNA from tissue that either or with from the of the The of virus by in myocardial from patients with or or from control cases The of virus by in myocardial from patients with or or from control cases While there are the of the response to recombinant adenovirus vectors and of the ability of these vectors to in the transduced tissue due to by wild type virus, it appears that the of adenoviruses, and potentially adenoviral vectors, in diseases of has not been The of adenoviruses in and is although studies in cotton that systemic delivery of adenovirus results in with a of and is being to develop adenoviral vectors lacking of the genome in to virus gene expression and replication in the target the of wild type virion from recombinant adenoviral and the use of drug therapy to the immune response at the time of However, it should be that, the of adenovirus sequences within the myocardium of and virus is and are not detected by that the virus persists in a defective as has been shown for of muscle Further, the detected in the myocardium is less than that detected in indicating that the absence of not the development of these that the associated with and may be upon the expression of a limited number of adenoviral genes in the myocardium. that the of wild type virus from recombinant adenovirus should be a for the use of these vectors in clinical While there have been in the development of adenoviral vectors with to the problems associated with persistence of transgene expression and the of the immune response against virus-encoded proteins, many remain to be gene therapy will be a viable option for the treatment of myocardial These (1) the ability to transduce sufficient of myocytes to a therapeutic response. (2) there is between the gene mutation and disease and there will be a for and early and mutation The regulation of gene expression, particularly in diseases such as the stoichiometry of the expression of the is likely to be In many diseases only a mutated protein is it is likely that the transgene product will an immune response. of the of of the mutated gene will the of gene therapy since dominant and mutations will that the gene is the of adenovirus in diseases of the myocardium to be further in to the