Gene transfer and cell transplant: an experimental approach to repair a 'broken heart'

Time for primary review 22 days. Despite significant progress in prevention and therapy of ischemic heart disease, treating patients with heart failure after myocardial infarction remains a major therapeutic challenge. Adult cardiomyocytes cannot regenerate after injury. Therefore, cardiomyocyte loss due to myocardial infarction is irreversible. Currently, congestive heart failure is the only major cardiovascular disorder that is increasing in incidence and mortality [1]. Thus, there is still a need to develop alternative therapeutic strategies to prevent, arrest or reverse congestive heart failure after myocardial infarction. Recent insights into the pathogenesis of myocardial disease and advances in molecular biology have opened up a new era of molecular and cellular therapies that target genes, molecules and peptides. Gene transfer as a therapeutic approach for the treatment of myocardial infarction and congestive heart failure has been suggested as a new treatment strategy for these serious disorders [2, 3]. The introduction, into injured myocardium, of recombinant transgenes that encode growth factors, could stimulate new vessel formation with increased collateral blood flow, accelerate healing and enhance myocardial performance. Today it is even possible to consider genetic manipulation leading to regeneration of myocytes within the infarcted myocardium. An alternative experimental strategy to increase viability and augment ventricular function after myocardial infarction is cell transplantation [4]. Engrafted fetal cells might increase the number of functional myocytes in the infarcted myocardium, could serve as a potential source of growth factors and could be programmed for myocyte-based gene transfer. The purpose of our review is to summarize recent advances in the attempts to develop molecular and cellular strategies for repairing a ‘broken heart’. The biological rationale for these new therapies and the potential limitations of gene therapy and cell transplant in treating myocardial infarction are discussed. Left ventricular remodeling following myocardial infarction refers to the process of early expansion and thinning of the infarcted segment that results in progressive dilation of the left ventricle (reviewed in [5, 6]). Left ventricular remodeling is also characterized by cardiac myocyte hypertrophy and matrix changes in the remote myocardium. This process is influenced by activation of local and systemic neurohormonal systems (e.g. the autonomic nervous system and renin-angiotensin system) and various cytokines. Initially, the remodeling process serves to restore stroke volume to normal but finally it leads to severe left ventricular dysfunction. Lengthening of the non-infarcted area also contributes to an increase in end-systolic volume and global dilatation. The heart cannot increase the number of myocytes. However, it can synthesize more sarcomeres and consequently increase the size of cardiac myocytes which leads to myocardial hypertrophy. The growth response of the heart is often accompanied by cardiac dilatation. Because prognosis after myocardial infarction is correlated with end-systolic volume, left ventricular remodeling is an important determinant of survival. Current therapies to prevent or arrest left ventricular dilatation and hypertrophy following myocardial infarction include limitation of infarct size by early reperfusion, administration of angiotensin-converting enzyme inhibitors, new β-adrenergic blockers (e.g. carvedilol), and perhaps administration of new agents such as endothelin antagonists or cytokine antagonists such as vesnarinone. These interventions, however, are limited since they are unable to reconstruct the injured myocardium. In addition, these interventions should be started relatively early after myocardial infarction and are applied to selected populations only. Therefore, there is an urgent necessity to find new therapeutic strategies to repair irreversible and progressive myocardial damage. The ability to perform gene transfer into the myocardium has only become feasible in recent years because of major advances in the field of molecular biology. The major advance has been the ability to isolate individual genes, or synthesize the analogous complementary DNA and to insert this DNA into plasmid expression vectors or recombinant viruses. The foreign DNA can then be introduced into the cell using a variety of delivery systems such as viruses or liposomes. Once the DNA enters the cell, it is taken to the nucleus where, depending on its mechanism of transport, it remains as an episome or becomes integrated into the host genome. Subsequently the foreign DNA is transcribed and the resulting mRNA is translated into the therapeutic protein. Several strategies have been used for gene transfer into the myocardium (reviewed in [2, 3]) including direct injection of naked plasmid DNA [7], implantation of ex vivo genetically engineered transplanted cells, application of liposome–DNA complexes, or injection of recombinant viruses. The short-term results obtained by us and others have been encouraging. One of the most promising techniques for gene transfer employs an adenovirus vector [8]. Although this vector delivers genes into the rat myocardium efficiently, there was a sharp decline in gene expression after 1 week associated with an inflammatory reaction in our studies [9]. This suggested an immunological mechanism behind this sharp decline. When we compared long-term results of adenovirus-mediated gene transfer into the myocardium between normal, immunocompetent rats and immunocompromised athymic nude rats, we found that the efficiency of recombinant gene expression assayed by production of a recombinant protein was sustained and vigorous in the immune-compromised rats, and that there was less inflammatory reaction. Our findings suggested that a host immune reaction against the adenovirus affects the expression of genes delivered by adenoviral vectors. In view of these results, we raised concerns about the feasibility of gene transfer into an ischemic or infarcted myocardium [9]. To begin to address this issue, Prentice et al. [10]subjected rats to 15 min or 60 min of coronary occlusion and 7 days of reperfusion, Fifteen minutes of ischemia does not cause cell death while 60 min produces myocardial necrosis. A DNA containing the luciferase reporter gene was injected into the ischemic area. Seven days after injection, significant levels of luciferase expression were obtained in hearts subjected to either 15 min or 60 min of ischemia. Thus, ischemic myocardium is capable of taking up and expressing foreign DNA. In another series of experiments [10], we assessed the feasibility of using retroviruses to transduce myocardial scar tissue. Retroviruses are able to transduce only replicating cells. Thus, its efficacy for gene transfer into non-replicating adult cardiac myocytes is limited. However, granulation tissue after myocardial infarction is rich in proliferative fibroblasts, endothelial cells and macrophages. We hypothesized therefore that a retroviral vector could be used for in vivo gene transfer into granulation tissue. We used a canine model of myocardial infarction. Infarction was produced percutaneously by embolizing a helical coil into the left anterior descending coronary artery. After 6–11 days of coronary artery occlusion, retrovirus carrying the reporter gene β-galactosidase was injected directly into the infarcted area under direct visualization. After 7–12 days, hearts were examined for histology and the presence of the reporter protein by X-gal staining. Four out of five hearts with myocardial infarction that were injected with retrovirus stained positive for β-galactosidase. One dog that was not infarcted was negative for β-galactosidase. These experiments suggest that retroviral uptake, gene expression and recombinant protein expression are possible in the infarcted area and localize to the injection site in the vicinity of granulation tissue. Our findings suggest that ischemic or infarcted myocardium can be a target for gene transfer. These were the first studies showing that ischemic/reperfused myocardium is capable of taking up and expressing foreign genes. Because of its high infectivity, adenovirus remains a promising tool for recombinant gene delivery to the myocardium. Based on our earlier studies [9], however, we hypothesized that following myocardial infarction, the efficiency of adenovirus-mediated gene transfer might be compromised by the intense inflammatory reaction in the necrotic area. To assess and measure the efficiency of adenovirus-mediated gene transfer and expression in infarcted myocardium, we produced myocardial infarction by subjecting rats to 60 min of coronary artery occlusion followed by sustained reperfusion [11]. Gene transfer into the infarcted area was performed using direct injection of a replication-defective adenovirus vector encoding the bacterial reporter gene, β-galactosidase. A total of 5.0×109 plaque-forming units of virus were injected into the infarcted myocardium either immediately or at 7, 22 or 30 days after reperfusion of rat hearts. Control rats received either saline or were not subjected to infarction and also received adenovirus carrying the β-galactosidase gene. All hearts were processed 7 days after cardiac injection and stained for β-galactosidase activity. Relative β-galactosidase activity was expressed as the % of the maximal area of β-galactosidase staining relative to the total area of the section examined (%±S.E.M.). The area of transgene expression in the non-infarcted hearts (28±7%) was significantly higher (p = 0.02) than at any time point studied in infarcted tissues (3.4±1.2%, 1.4±1.0%, 2.8±0.8% and 3.4±0.9% at reperfusion and at 7, 22 and 30 days after myocardial infarction, respectively (Fig. 1). β-Galactosidase gene expression was limited mainly to viable myocytes at the border of the myocardial infarction (Fig. 1 and Fig. 2). Hearts injected 7 days after infarction had significantly less transgene activity with 3 of 5 samples displaying no macroscopically visible β-gal activity. Following viral injection, an inflammatory response consisting of mononuclear cell infiltration was much less intense 7 days following injection in non-infarcted control rat hearts than at any of the time points examined for infarcted hearts. We concluded that gene transfer into infarcted myocardium, while feasible, was limited by low transfection efficiency when compared to non-infarcted normal myocardium. Still, the ability to introduce genes into viable peripheral cells might be a useful approach for enhancing neovascularization, collateral flow and healing [11]if expression of the transgene for a limited number of days might still lead to a therapeutic benefit. Hematoxylin–eosin stained section showing edge of rat myocardial infarct injected with adenovirus vector. β-Galactosidase-positive cells are shown in the border of the infarct. A: Frozen section of noninfarcted rat heart, 7 days after injection of adenovirus vector, showing diffuse β-galactosidase activity of the free wall. B: Frozen section of 30-day infarct in rat, 7 days after injection of adenovirus vector, showing β-galactosidase activity at the edges of the infarct, only. One of the more fascinating goals in the field of gene therapy for cardiovascular disease would be genetic manipulation leading to regeneration of myocytes after myocardial infarction. In skeletal muscle, the muscle-specific MyoD family of basic helix-loop-helix transcription factors functions as master genes that can induce the skeletal muscle differentiation program [12]. Most remarkably, members of the MyoD family can induce this program in a wide variety of cell types including fibroblasts. Conversion of cardiac fibroblasts populating the scar of myocardial infarction into functional skeletal myocytes has the potential to contribute to cardiac contraction. Based on clinical experience with cardiomyoplasty in which the latissimus dorsi skeletal muscle was used to improve hemodynamic function in patients with heart failure [13], we hypothesized that MyoD-converted cells could be adapted to function as cardiac myocytes. Prentice et al. [14]investigated the feasibility of converting cardiac fibroblasts into skeletal muscle cells by forced expression of the MyoD gene. Fibroblasts isolated from rat hearts were infected with retrovirus – carrying the MyoD gene. Within a few days after transfection, fibroblasts became elongated and formed multinucleated myotubes morphologically resembling striated skeletal muscle myocytes (Fig. 3A,B). Immunofluorescent staining identified the presence of skeletal muscle-specific skeletal fast myosin heavy chain, protein markers of skeletal muscle differentiation (Fig. 3C,D). In similar experiments, Tam et al. [15]reported spontaneous in a few of of in skeletal muscle by primary rat cardiac fibroblasts with MyoD expressing A: of fibroblasts to elongated B: rat cardiac with a MyoD retrovirus showing and points to of 5 of cardiac fibroblasts. of the cells as in using heavy After the feasibility of forced expression of MyoD to fibroblasts into cells with a skeletal muscle in we used our canine model of myocardial infarction to the feasibility of this strategy in We produced myocardial infarction by of a helical coil into the left anterior descending coronary artery of Four were injected with β-galactosidase expressing carrying the gene or days after Seven to days were with X-gal identified transfection in hearts. In of positive β-galactosidase against skeletal fast myosin heavy identified cells that were stained positive for the myosin heavy (Fig. These cells were also stained positive for the muscle (Fig. Our findings suggest that transfer of the MyoD gene can in of cardiac fibroblasts into cells which skeletal fast myosin heavy However, myotubes were not The of transfection and gene expression was limited and raised a about the efficiency of this approach Frozen from myocardial infarct in dog injected with MyoD retrovirus on 7 and with to skeletal fast myosin heavy chain, or of cells within the infarct positive for skeletal fast myosin heavy and for the muscle these cells expressed skeletal fast myosin heavy chain, was not as had been the for in In an to an alternative approach for MyoD gene transfer into myocardial infarction, we used the replication-defective adenovirus carrying another of the MyoD the gene, to our studies with we were able to fibroblasts to skeletal muscle cell in In the we injected adenovirus vector carrying either gene or reporter gene β-galactosidase into myocardial infarction of normal or athymic nude Four days after injection, we of positive X-gal staining transfection with β-galactosidase in immunocompetent and nude Immunofluorescent and staining for skeletal myosin heavy in was negative skeletal myotubes were not Based on these findings and our experiments on adenovirus-mediated gene transfer into myocardial infarction concluded that gene transfer is feasible at the border of infarction but limited in the infarcted granulation tissue. et al. in vivo with MyoD gene transfer into myocardial granulation tissue of used a rat model of myocardial injury. One week they injected adenovirus carrying the MyoD gene or β-galactosidase. were with to the adenoviral immune One week after gene cells were identified that expressed and skeletal myosin heavy MyoD mRNA in the injured of hearts the MyoD but not in control hearts β-galactosidase. In hearts they multinucleated to our studies et al. intense inflammatory reaction in the of high of adenovirus The suggested that gene transfer can induce skeletal muscle differentiation in healing heart The not the of this approach on the of myocardial or myocardial results need and In view of the significant in genes into infarcted myocardium alternative are A experimental strategy to increase viability and augment ventricular function after myocardial infarction is cell transplant This approach a to increase the number of functional cardiomyocytes in the infarcted myocardium In and that fetal cardiomyocytes can be and integrated within the normal myocardium of this the formation of fetal cardiomyocyte as as after in the myocardium of and this approach can be applied to the infarcted and myocardium is In to the feasibility of fetal myocardial tissue transplantation into myocardial infarction and the cardiomyocytes can we subjected rats after myocardial infarction to 3 of therapy In the first tissue of fetal were injected into the scar days after infarction. The rats were with of of In the of fetal rat were injected into the scar days after infarction. A of myocardial infarction was with injection of saline into the scar In tissue was in the infarcted myocardium, either by or in of rats after 7 days and in of rats after days. In the or staining the presence of fetal cardiac tissue of of of and days of after the in this was The of transplantation was of the after myocardial infarction. The fetal tissues stained for (Fig. which is for the adult rat myocardium. In infarcted myocardium staining was only positive within the A: cardiac myocytes days after implantation into noninfarcted rat myocardium with B: A of fetal cardiac tissue cells injected into a myocardial infarction, shown 1 week after transplantation with We were not able to that the cardiomyocytes and We not assess the of the transplanted tissue on myocardial Thus, we concluded that fetal cardiomyocyte tissue can be and in the infarcted myocardium. This experimental approach a therapeutic strategy for gene transfer for of therapeutic into myocardial infarction The of et al. our showing the feasibility of cardiac myocyte into the border of myocardial infarction. In this cells were injected immediately after coronary artery myocytes were identified in the border of myocardial infarction after In another et al. that fetal cardiac myocytes injected into the border of myocardial infarction, by min of coronary artery occlusion, can improve myocardial function as assessed by 1 after myocardial infarction. The that compared with cell transplant was associated with Our and studies however, suggested that cell transfer into the necrotic myocardium is limited immediately after myocardial infarction, perhaps because of the intense inflammatory response following infarction. et al. that fetal rat cardiac myocytes transplanted into the tissue of adult rat formed cardiac tissue that for at 3 This for increasing the of tissue in infarcted myocardium. In another et al. a rat model of myocardial Four after fetal rat cardiac myocytes or were injected into the scar tissue. Four after scar size and heart function were The found that scar and tissue in hearts subjected to cell transfer were than The transplanted cardiac myocytes had formed cardiac tissue that of the function as assessed with was in the transplantation than The suggested that cardiac myocytes formed cardiac tissue in myocardial scar and was associated with infarct of infarct expansion and heart function compared with findings in the control hearts However, the cells were directly to was not from either et studies that of et al. In a et al. that or delivery of cardiac myocytes is feasible in a canine of basic growth increased the efficiency of cell This approach up new in the field of gene transfer into the myocardium Several have been for heart function following cardiac myocytes direct of the transplanted myocytes to of infarct expansion by the of cardiac by factors from the cells resulting in collateral flow, and of by growth from the fetal cells. Based on our we that of infarct expansion and of therapeutic growth factors are the most mechanism than direct of the transplanted cells to cellular has as a potential experimental approach for delivery of therapeutic peptides. The of genetically cells for the delivery of recombinant molecules is a approach for ex vivo gene transfer the fetal could be engineered ex vivo to growth factors, as a useful strategy for enhancing to myocardial healing and cellular An alternative strategy to increase the of tissue in the infarcted myocardium is transplantation of skeletal muscle cells. are of skeletal muscle cells which to cardiac myocytes for infarct myocardium, skeletal muscle a skeletal a few cells as cells. These cells in an are following skeletal the and with and with injured and function of skeletal In addition, compared to cardiac skeletal muscle cells are more to ischemia. Several have raised the that transplanted cells obtained from skeletal muscle can and become cardiac muscle cells. This has been cardiomyoplasty in which the skeletal muscle latissimus dorsi is with and from fast to The skeletal muscle is the heart and serves to augment cardiac contraction. et al. from cell into hearts of normal that the can into The of that the cell from the cell This also that cells with a plasmid encoding could induce the was no between the host cardiac myocytes and the skeletal cells et al. that cells can be isolated from skeletal muscle of can be into the myocardium of the dog and can into myocytes and to by resembling or The cells were identified in the injured myocardium as as et al. similar results in a Most et al. that skeletal can be into rat The and after 3 days to multinucleated The suggested that the myotubes into and to expressing high of fast myosin heavy the suggested that the tissue could when ex vivo and perform The concluded that skeletal can new muscle tissue when into injured and this new muscle be to a cardiac muscle cells, however, have that from Thus, new muscle cells might an This might also be a with cardiac myocytes. is still the fetal cells can into functional cardiac myocytes that with cells and enhance cardiac by directly to contraction. formed skeletal muscle within an infarcted scar should and and to viable cardiac myocytes to This is a significant since skeletal myocytes not synthesize is possible that in the these cells could be engineered to these strategies to the heart include attempts to reverse the differentiation of adult cardiac myocytes and to induce growth in cardiac myocytes approach is to to induce of cardiac myocytes by genes that cell has been shown that transfection with genes such as the induce cardiac myocyte in and Several studies suggested that expression of such genes can in induce of fetal or cells. experiments using adult cardiac however, to induce is to a cardiac myocyte cell that can be used as a of cell et al. of of cardiac myocytes from cells. into cells a gene consisting of myosin heavy and a encoding for suggested that genetic manipulation can be used to cardiac myocytes from cells. that these cardiac myocytes were for cardiac transplantation and formation of of and of the β-adrenergic are the of the β-adrenergic associated with congestive heart Thus, of this to improve cardiac can be a new therapeutic approach to congestive heart and raised and the that genetic of the β-adrenergic can enhance cardiac function that with cardiac of either the an of β-adrenergic an enzyme that and have increased myocardial Most they that gene transfer with adenovirus encoding either the or a β-adrenergic β-adrenergic in cardiac myocytes recent in growth factors as an alternative therapeutic strategy to the of cardiomyocytes and to improve myocardial function and viability new in cardiovascular therapy (reviewed in factors can therapy with direct myocardial and increase the viability and function of the ischemic myocardium. The ability to introduce growth genes into vessel be a useful strategy for therapeutic and prevention of as as to enhance healing and viability after injury. Although growth factors have not become of clinical they have a of and are in clinical need to be MyoD gene therapy or cell transplant can be an therapeutic tool for myocardial repair and The efficacy of gene transfer into myocardial infarction are to the associated with injection of high of adenovirus and the immune reaction against the virus or the gene. The of MyoD-converted cells in vivo is low and in our studies we were unable to the formation of myotubes in the in vivo vectors or more master genes are We experimental cardiac repair genetic and cell These experimental strategies are and limitations of therapies for heart failure after myocardial infarction. experimental strategies for myocardial regeneration should be These include attempts to growth of cardiac myocytes and to a cardiac myocyte cell as a source of cells for transplant approach growth therapy for of myocardial collateral and of the β-adrenergic is another experimental strategy to enhance cardiac the advances in our of the of cardiac myosin of and heavy genes and new gene transfer techniques it become possible to cardiac myosin and to low activity myosin with the high activity of genes under control such as or of the might improve of the myocardium Most loss of cardiac myocytes due to programmed cell death has been as an important process that loss of cardiac myocytes and contributes to of cardiac failure of become a new target for prevention of cardiac myocyte loss and progressive myocardial In genetic and cell transplant a experimental approach for cellular cardiomyoplasty and tissue repair after myocardial injury. Current from suggest that it be possible to heart failure by genetic or cardiac myocytes into injured myocardium. These however, still significant they can become therapeutic The progress of molecular the of systems for in vivo gene transfer and of new genes, that gene therapy and cell transfer as a therapeutic tool to repair a ‘broken heart’. This was in by from the